university of groningen orthostatic hypotension in elderly ... · author(s) and/or copyright...

University of Groningen

Orthostatic hypotension in elderly patients.Hartog, Laura

IMPORTANT NOTE: You are advised to consult the publisher's version (publisher's PDF) if you wish to cite fromit. Please check the document version below.

Document VersionPublisher's PDF, also known as Version of record

Publication date:2017

Link to publication in University of Groningen/UMCG research database

Citation for published version (APA):Hartog, L. (2017). Orthostatic hypotension in elderly patients. [Groningen]: Rijksuniversiteit Groningen.

CopyrightOther than for strictly personal use, it is not permitted to download or to forward/distribute the text or part of it without the consent of theauthor(s) and/or copyright holder(s), unless the work is under an open content license (like Creative Commons).

Take-down policyIf you believe that this document breaches copyright please contact us providing details, and we will remove access to the work immediatelyand investigate your claim.

Downloaded from the University of Groningen/UMCG research database (Pure): http://www.rug.nl/research/portal. For technical reasons thenumber of authors shown on this cover page is limited to 10 maximum.

Download date: 22-04-2020

https://www.rug.nl/research/portal/en/publications/orthostatic-hypotension-in-elderly-patients(11414df2-12f6-4536-88a1-59cec5688309).html

https://www.rug.nl/research/portal/en/publications/orthostatic-hypotension-in-elderly-patients(11414df2-12f6-4536-88a1-59cec5688309).html

Orthostatic hypotension in elderly patients

L.C. Hartog

© L.C. Hartog, 2016

All rights are reserved. No part of this publication may be reproduced, stored in a retrievel system, or transmitted in any form or by any other means without the written permission of the author.

Financial support for printing this thesis was kindly provided by University of Groningen, University Medical Center Groningen, Stichting Zwols Wetenschapsfonds Isala Klinieken, Diabetes Centre, and Trivium Meulenbelt Zorg.

Lay-Out: Gildeprint, EnschedeCover: © Can Stock Photo / iLexx and Gildeprint, EnschedePrinting: Gildeprint, Enschede

ISBN 978-90-367-9280-6ISBN Electronic version 978-90-367-9279-0


Proefschrift

ter verkrijging van de graad van doctor aan de Rijksuniversiteit Groningen

op gezag van derector magnificus prof. dr. E. Sterken

en volgens besluit van het College voor Promoties.

De openbare verdediging zal plaatsvinden op

woensdag 8 februari 2017 om 12:45 uur

door

Laura Caroline Hartog

geboren op 22 december 1980te Rotterdam

Promotor

Prof. dr. H.J.G. Bilo

CopromotoresDr. K.J.J. van Hateren Dr. N. Kleefstra

BeoordelingscommissieProf. dr. J.P.J. Slaets Prof. dr. S.U. Zuidema Prof. dr. H.E. van der Horst

ParanimfenDhr. E.H. JutteDhr. L.J. Hartog

Voor oma

TABLE OF CONTENTS

Chapter 1 Introduction 11

Chapter 2 Measuring orthostatic blood pressure during different postural changes: standing versus sitting. 25

Chapter 3 Diagnosing orthostatic hypotension with continuous and interval blood pressure measurement devices. 41

Chapter 4 Orthostatic hypotension does not predict recurrent falling in a nursing home population. 53

Chapter 5 Is orthostatic hypotension related to falling? A meta-analysis of individual patient data of prospective observational studies. 65

Chapter 6 The association between orthostatic hypotension, falling and successful rehabilitation in a nursing home population. 85

Chapter 7 The association between orthostatic hypotension and handgrip strength with successful rehabilitation in elderly hip fracture patients. 101

Chapter 8 Orthostatic changes in blood pressure and mortality in a nursing home population. 117

Chapter 9 The clinical relevance of orthostatic hypotension in elderly patients. 135

Chapter 10 Health-related quality of life, rehabilitation and mortality in a nursing home population. 145

Chapter 11 Discussion 161

Chapter 12 Summary 183

Chapter 13 Nederlandse samenvatting (summary in Dutch) 189

Dankwoord 201Curriculum Vitae 205List of publications 207Previous dissertations 209

CHAPTER 1Introduction

R1

R2

R3

R4

R5

R6

R7

R8

R9

R10

R11

R12

R13

R14

R15

R16

R17

R18

R19

R20

R21

R22

R23

R24

R25

R26

R27

R28

R29

R30

R31

R32

R33

R34

R35

R36

R37

R38

R39

Chapter 1

12

R1

R2

R3

R4

R5

R6

R7

R8

R9

R10

R11

R12

R13

R14

R15

R16

R17

R18

R19

R20

R21

R22

R23

R24

R25

R26

R27

R28

R29

R30

R31

R32

R33

R34

R35

R36

R37

R38

R39

ntroduction

13

1ORTHOSTATIC HYPOTENSION

Orthostatic hypotension (OH) occurs frequently in the elderly population, and its prevalence increases with advancing age [1, 2]. OH is a decrease in blood pressure upon standing that reflects an impaired hemodynamic homeostasis. OH is defined by the International consensus as a decrease in systolic blood pressure (SBP) by at least 20 mmHg or a decrease in diastolic blood pressure (DBP) by at least 10 mmHg within 3 minutes after changing from supine to a standing position, ideally measured with a continuous blood pressure device [1]. In the few studies that reported the prevalence of OH in frail elderly nursing home residents, the prevalence varied between 18% and 50% [3-5]. In hospitalized elderly patients the prevalence of OH ranged from 8% to 67% [6, 7].In healthy individuals, an active postural change triggers a baroreceptor-mediated response, which leads to an increase in heart rate, myocardial contractility, and peripheral vascular resistance in response to the shifting of a considerable amount of blood (300-800 ml) to the venous system below the diaphragm, mainly in the splanchnic venous system to the pelvis [7-12]. Stabilization of blood pressure is normally achieved within one minute [12]. OH occurs when baroreceptor-mediated autonomic responses are inadequate to maintain blood pressure on standing or because the blood volume is insufficient to support ventricular filling [7, 10]. The etiology of OH is multifactorial; decrease of baroreceptor sensitivity, pure autonomic failure, the use of different medications, hypovolemic disorders, and bed rest are all considered being possible causes of OH [6-8, 13, 14]. The normal age-related impairment of the baroreflex sensitivity, the higher prevalence of comorbidities, and the use of different medications are the main reasons for the higher prevalence of OH among elderly patients [7, 12]. OH can be accompanied by symptoms of cerebral hypoperfusion like light-headedness, dizziness, or syncope [1, 10, 12]. The amount of decrease in blood pressure is related to whether the patient experiences orthostatic complaints [12]. Baroreflex failure could also contribute to orthostatic hypertension (OHT) [15]. Next to OH, also OHT occurs more frequently in the elderly population [2]. OHT is an increase in blood pressure after orthostatic change. Although there is no official definition of OHT, many studies use an increase in SBP of 20 mmHg after postural change as a cut-off point. OHT can be considered as a form of prehypertension and is associated with hypertension-associated target-organ damage [15, 16].

R1

R2

R3

R4

R5

R6

R7

R8

R9

R10

R11

R12

R13

R14

R15

R16

R17

R18

R19

R20

R21

R22

R23

R24

R25

R26

R27

R28

R29

R30

R31

R32

R33

R34

R35

R36

R37

R38

R39

Chapter 1

14

MEASURING ORTHOSTATIC HYPOTENSION

According to the International consensus definition, active standing or the passive head-up tilt testing (HUT) and ideally continuous blood pressure measurements, instead of interval blood pressure measurements, are recommended to measure OH [1]. The HUT is a noninvasive tool for diagnosing syncope or orthostatic intolerance by rapidly moving the patient from a supine to an upright position [17]. Although interval blood pressure measurements to determine orthostatic hypotension are less accurate, automated sphygmomanometers are commonly used for this purpose in daily practice [18]. A previous study reported a lower prevalence of OH with the interval compared to the continuous blood pressure measurement device [19]. Due to these different results, concerns are raised against the threshold in the diagnostic criteria of OH and it is hard to apply the criteria on different blood pressure measurement devices in clinical practice.Besides, orthostatic blood pressure is advised to be measured in standing position after 5 minutes of rest in supine position [1]. However, as many elderly patients are not able to stand during several minutes, sitting orthostatic blood pressure measurements are sometimes used as an alternative [20]. Only a few studies published results regarding OH in sitting position [21-23]; the prevalence of OH in these studies varied from 8% within community-dwelling individuals to 56% within elderly hospitalized patients [23]. It can be questioned whether orthostatic blood pressure measurements in the standing position can be replaced by measuring orthostatic blood pressure in the sitting position.

CLINICAL IMPLICATIONS OF ORTHOSTATIC HYPOTENSION

Some studies report that the presence of OH is related to an increased risk of cardiovascular disease and all-cause mortality in elderly people [24-27]. Besides, some studies suggest OH to be associated with falls, which in turn can lead to serious morbidity [28-30]. As the prevalence of OH in elderly is high, and associations with fall incidents were observed in these populations, OH may also be an important prognostic factor for chances of rehabilitation in nursing home patients. In the paragraphs below, the role of OH and hypertension in relation to different endpoints will be discussed.

rthostatic h potension and allingLike the aetiology of OH, fall risk is a complex and multifactorial phenomenon and OH is one out of many risk factors believed to contribute to an increased fall risk in elderly people [30-32]. The combination of OH and other disabilities or medication usage may further impair orthostatic control and cause orthostatic complaints like syncope [30, 33, 34]. Especially the combination of OH and complaints has been identified as a cause of falls [30].

R1

R2

R3

R4

R5

R6

R7

R8

R9

R10

R11

R12

R13

R14

R15

R16

R17

R18

R19

R20

R21

R22

R23

R24

R25

R26

R27

R28

R29

R30

R31

R32

R33

R34

R35

R36

R37

R38

R39

ntroduction

15

1Two recent systematic reviews confirmed that OH is a risk factor for falling in elderly patients. However, the absolute attributive risk was not established due to the lack of a meta-analysis [30, 31]. Furthermore, studies that described the relationship between OH and falling have important limitations that need to be addressed; the lack of prospective fall data [35-37], not adjusting for important confounders [38], or not using the international consensus definition of OH [4]. These previous studies reported different results regarding the association with falling. Only the results regarding the relationship between recurrent falling and OH seem consistent; a twofold fall risk for patients with OH was described [4, 33].Although most clinicians regard OH as a causal factor for falls, it thus remains highly uncertain whether and to what extent OH contributes to falling [31, 34].

rthostatic h potension and rehabilitationThe outcome of rehabilitation reflects the condition of the elderly patient and is a summation of many factors, including both physical and mental parameters [20, 39-46]. Also, OH and muscle strength are amongst the factors that have been found to influence rehabilitation in elderly patients [20, 43, 45]. As the prevalence of OH and low muscle strength is high in elderly patients and are considered as important risk factors for falling and frailty, these variables are likely to influence successful rehabilitation [29-31, 47-50]. As muscle strength and OH are probably related to successful rehabilitation, and possibly also interrelated, these factors should be combined (and adjusted for) in analysing their association with rehabilitation. If OH appears to be a prognostic factor for successful rehabilitation, clinicians should take OH into account, and should consider applying interventions aimed at reducing OH or orthostatic complaints.

rthostatic h potension and ortalitPrevious studies reported that the presence of OH increases the risk of cardiovascular disease, cardiovascular mortality or all-cause mortality in elderly people [2, 14, 24-27, 31, 51-53]. Two meta-analyses were published regarding the relationship between OH and mortality [25, 54]. Both studies described that OH increased the risk of all-cause mortality, with risk ratios (RR) of 1.40 (95%CI 1.20-1.63) [25] and 1.50 (95%CI 1.24-1.81) [54]. However, an independent relationship can be questioned, since many included studies did not include important confounders in the relationship between OH and mortality [25]. In addition, it is questionable whether OH is a component of cardiovascular disease or an independent factor which increases mortality risk [54].

R1

R2

R3

R4

R5

R6

R7

R8

R9

R10

R11

R12

R13

R14

R15

R16

R17

R18

R19

R20

R21

R22

R23

R24

R25

R26

R27

R28

R29

R30

R31

R32

R33

R34

R35

R36

R37

R38

R39

Chapter 1

16

HYPERTENSION IN OLD AGE

Hypertension and using antihypertensive drugs are frequently mentioned as causal factors for OH in elderly patients [1, 12, 55, 56]. Although hypertension is a well-known risk factor for cardiovascular morbidity and mortality, it remains unclear whether hypertension is related to complications and death, and whether antihypertensive treatment is beneficial in old age. The Hypertension in the Very Elderly Trial (HYVET) was performed to investigate intensive blood pressure targets in elderly patients [57]. This study showed that elderly patients benefit from intensive antihypertensive treatment with a reduction in cardiovascular and all-cause mortality. The results of the HYVET study suggest strict treatment targets in elderly patients. However, the included patients in the HYVET study represent a highly selected population, and therefore the results cannot be extrapolated to the general elderly population but only to a very select group of vital elderly patients. On the other hand, the relationship between blood pressure itself and mortality in old age has been investigated in many observational studies, and mostly inverse associations were reported [58, 59]. The observational PARTAGE study and the study of Askari et al. are the only studies that specifically investigated this relationship in nursing home populations. An inverse relationship was observed between SBP and mortality in the PARTAGE study [60, 61], while Askari et al. found no association between blood pressure and cardiovascular outcome [62]. At this moment, the intensity of antihypertensive treatment in individual elderly patients will have to be judged taking into account the degree of frailty, the estimated survival time, and the chances or presence of side effects of medication. Applying strict treatment targets to an elderly and frail population may lead to an increased prevalence of OH, and subsequently in an increased fall risk without any well-known benefits.

HEALTH-RELATED QUALITY OF LIFE

Comorbidities, depression, cognitive impairment and other geriatric syndromes are highly prevalent in old age and can greatly impact health-related quality of life (HRQOL) [63-68]. As a consequence, HRQOL is generally low in nursing home residents [64-66, 69-71]. Low HRQOL corresponds to substantial limitations in physical, emotional and social well-being due to a medical condition or its treatment [72].The evaluation of HRQOL in individual patients can be used to measure disease-related distress and overall perception of health. Next to the evaluation of HRQOL as a separate outcome measure, HRQOL also has prognostic value in non-nursing home settings: a lower HRQOL has been associated with increased mortality risk, also in elderly patients [73].

R1

R2

R3

R4

R5

R6

R7

R8

R9

R10

R11

R12

R13

R14

R15

R16

R17

R18

R19

R20

R21

R22

R23

R24

R25

R26

R27

R28

R29

R30

R31

R32

R33

R34

R35

R36

R37

R38

R39

ntroduction

17

1HRQOL is also used to evaluate therapeutic interventions. Therefore, HRQOL measurement outcomes could have a variety of implications in decision-making processes regarding patients and medical interventions.Measuring HRQOL in nursing home patients could lead to an increased understanding on factors that negatively impact HRQOL, ultimately aiming to improve HRQOL of this patient group. As mortality risk is already very high in old age, other clinical outcomes, like HRQOL may be more relevant.

AIM AND OUTLINE THESIS

In summary, OH is a decrease in blood pressure upon standing that reflects an impaired hemodynamic homeostasis. There is limited evidence about which methods to use to accurately determine OH, about the consequences when OH is established, and how to deal with OH in specifically the elderly population. It is advised to perform a continuous BP measurement, while automated sphygmomanometers with interval measurements are commonly used for this purpose in daily practice. Clinicians assume OH is causally related to falling or cardiovascular complications, despite the sparse evidence. The lack of evidence regarding OH underlines the need for gaining more knowledge about the implications of OH. Besides, information regarding mortality risk prediction of blood pressure in a nursing home population is lacking, while the need for strict antihypertensive treatment in this population is questionable. Finally, the role of HRQOL in this specific population seems important. It is unknown however, in what way it is related to cardiovascular morbidity and mortality.

The objectives of this thesis are to study the OH measurements itself, to study the clinical implications of OH in elderly patients, and to study factors (including OH) that are related to mortality and successful rehabilitation in nursing home residents.

In Chapters 2 and 3 various methods to measure OH are described. Firstly, a postural change from supine to standing position is compared with a change to sitting position. Secondly, continuous and interval blood pressure measurements are compared with each other. The hypothesis of this study is that measuring the blood pressure from supine to sitting instead of supine to standing leads to an underestimation of the prevalence of OH. This is possibly due to a difference in hemodynamic change in the standing versus the sitting position. When postural change is performed from supine to sitting, it is reasonable that less venous pooling in the legs will occur and a blood pressure drop will be observed less often, resulting in a lower prevalence of OH. The second hypothesis is that the prevalence of OH with the continuous blood pressure measurements would be higher compared to the interval blood pressure measurements.

R1

R2

R3

R4

R5

R6

R7

R8

R9

R10

R11

R12

R13

R14

R15

R16

R17

R18

R19

R20

R21

R22

R23

R24

R25

R26

R27

R28

R29

R30

R31

R32

R33

R34

R35

R36

R37

R38

R39

Chapter 1

18

In chapters 4 and 5 the relationship between OH and falling is described. In chapter 4, the

relationship between OH, first fall incident and recurrent falling in nursing home residents is prospectively investigated. A meta-analysis of prospective observational studies regarding this relationship is presented in chapter 5. The aim of the study is to investigate whether and to what extent orthostatic hypotension contributes to fall risk.

In chapters 6 and 7 the relationship between OH and rehabilitation in elderly patients is described. In chapter 6 the association between orthostatic hypotension, falling and successful rehabilitation in a nursing home population is investigated. The purpose of this study is to identify the prevalence of OH in frail elderly nursing home residents, and assess its association with falling and chances of successful rehabilitation. The hypothesis is that the presence of OH will negatively influence the time to successful rehabilitation. In chapter 7 a study that aims to investigate the influence of OH and muscle strength on the chances of successful rehabilitation within elderly hospitalized hip fracture patients is described. Hip fractures are relatively common in frail elderly patients and associated with a high risk of morbidity and mortality [39, 43]. It is expected that OH is highly prevalent in hip fracture patients. This is caused by hip fracture-related factors like bed rest, surgery, inadequate water intake, and blood loss. The aim is to investigate the relation between OH, muscle strength and the chances of successful rehabilitation within elderly hospitalized patients.

The associations in the very elderly and frail patients between blood pressure, especially orthostatic changes in blood pressure, and mortality, are unclear. The relationship between orthostatic changes in blood pressure and mortality in a nursing home population is assessed in chapter 8. The aim is to investigate the association between blood pressure levels, an impaired hemodynamic homeostasis (OH and OHT) and mortality in nursing home patients. Furthermore, the additional value of OH and blood pressure in mortality prediction is assessed.

A cross-sectional study that investigates if the International consensus definition of OH is the best definition in relation with orthostatic complaints and falling is described in chapter 9. The relationship between different definitions of OH and orthostatic complaints and falling is explored.

In chapter 10 the associations between HRQOL, rehabilitation and mortality in a nursing home population are described. The purpose of this study is to investigate the associations between HRQOL and two clinical relevant outcome measures; all-cause mortality and successful rehabilitation, in a nursing home population. Furthermore, the predictive capability of HRQOL on mortality is assessed.

R1

R2

R3

R4

R5

R6

R7

R8

R9

R10

R11

R12

R13

R14

R15

R16

R17

R18

R19

R20

R21

R22

R23

R24

R25

R26

R27

R28

R29

R30

R31

R32

R33

R34

R35

R36

R37

R38

R39

ntroduction

19

1Finally, in chapter 11 I will discuss the different studies in this thesis. After the general discussion of this thesis, recommendations for daily practice and suggestions for future research are presented.

R1

R2

R3

R4

R5

R6

R7

R8

R9

R10

R11

R12

R13

R14

R15

R16

R17

R18

R19

R20

R21

R22

R23

R24

R25

R26

R27

R28

R29

R30

R31

R32

R33

R34

R35

R36

R37

R38

R39

Chapter 1

20

REFERENCES

1. Lahrmann H, Cortelli P, Hilz M, et al. EFNS guidelines on the diagnosis and management of orthostatic hypotension. European journal of neurology : the official journal of the European Federation of Neurological Societies. 2006 Sep;13(9):930-6.

2. Veronese N, De Rui M, Bolzetta F, et al. Orthostatic Changes in Blood Pressure and Mortality in the Elderly: The Pro.V.A Study. American journal of hypertension. 2015 Mar 11.

3. Ooi WL, Barrett S, Hossain M, Kelley-Gagnon M, Lipsitz LA. Patterns of orthostatic blood pressure change and their clinical correlates in a frail, elderly population. Jama. 1997 Apr 23-30;277(16):1299-304.

4. Graafmans WC, Ooms ME, Hofstee HM, et al. Falls in the elderly: a prospective study of risk factors and risk profiles. American journal of epidemiology. 1996 Jun 1;143(11):1129-36.

5. Valbusa F, Labat C, Salvi P, et al. Orthostatic hypotension in very old individuals living in nursing homes: the PARTAGE study. Journal of hypertension. 2012 Jan;30(1):53-60.

6. Weiss A, Beloosesky Y, Kornowski R, et al. Influence of orthostatic hypotension on mortality among patients discharged from an acute geriatric ward. J Gen Intern Med. 2006 Jun;21(6):602-6.

7. Feldstein C, Weder AB. Orthostatic hypotension: a common, serious and underrecognized problem in hospitalized patients. Journal of the American Society of Hypertension : JASH. 2012 Jan-Feb;6(1):27-39.

8. Medow MS, Stewart JM, Sanyal S, et al. Pathophysiology, diagnosis, and treatment of orthostatic hypotension and vasovagal syncope. Cardiology in review. 2008 Jan-Feb;16(1):4-20.

9. Wieling W, Krediet CT, van Dijk N, Linzer M, Tschakovsky ME. Initial orthostatic hypotension: review of a forgotten condition. Clinical science (London, England : 1979). 2007 Feb;112(3):157-65.

10. Freeman R, Wieling W, Axelrod FB, et al. Consensus statement on the definition of orthostatic hypotension, neurally mediated syncope and the postural tachycardia syndrome. Clinical autonomic research : official journal of the Clinical Autonomic Research Society. 2011 Apr;21(2):69-72.

11. Thijs RD, Kamper Am Fau - van Dijk AD, van Dijk Ad Fau - van Dijk JG, van Dijk JG. Are the orthostatic fluid shifts to the calves augmented in autonomic failure? 20100215 DCOM- 20100430(1619-1560 (Electronic)).

12. Ricci F, De Caterina R, Fedorowski A. Orthostatic Hypotension: Epidemiology, Prognosis, and Treatment. J Am Coll Cardiol. 2015 Aug 18;66(7):848-60.

13. Woolcott JC, Richardson KJ, Wiens MO, et al. Meta-analysis of the impact of 9 medication classes on falls in elderly persons. Archives of internal medicine. 2009 Nov 23;169(21):1952-60.

14. Luukinen H, Koski K, Laippala P, Kivela SL. Prognosis of diastolic and systolic orthostatic hypotension in older persons. Archives of internal medicine. 1999 Feb 8;159(3):273-80.

15. Kario K. Orthostatic hypertension-a new haemodynamic cardiovascular risk factor. Nature reviews Nephrology. 2013 Dec;9(12):726-38.

16. Kario K. Orthostatic hypertension: a measure of blood pressure variation for predicting cardiovascular risk. Circulation journal : official journal of the Japanese Circulation Society. 2009 Jun;73(6):1002-7.

17. Teodorovich N, Swissa M. Tilt table test today - state of the art. 20160329 DCOM- 20160329(1949-8462 (Electronic)).

18. Caine Se Fau - Alsop K, Alsop K Fau - Mac Mahon M, Mac Mahon M. Overlooking orthostatic hypotension with routine blood-pressure equipment. 19980827 DCOM- 19980827(0140-6736 (Print)).

19. Pasma JH, Bijlsma AY, Klip JM, et al. Blood pressure associates with standing balance in elderly outpatients. PloS one. 2014;9(9):e106808.

20. Hartog LC, Cizmar-Sweelssen M, Knipscheer A, et al. The association between orthostatic hypotension, falling and successful rehabilitation in a nursing home population. LID - S0167-4943(15)00074-6 [pii] LID - 10.1016/j.archger.2015.05.005 [doi]. 20150531(1872-6976 (Electronic)).

R1

R2

R3

R4

R5

R6

R7

R8

R9

R10

R11

R12

R13

R14

R15

R16

R17

R18

R19

R20

R21

R22

R23

R24

R25

R26

R27

R28

R29

R30

R31

R32

R33

R34

R35

R36

R37

R38

R39

ntroduction

21

1

21. Aronow WS, Lee Nh Fau - Sales FF, Sales Ff Fau - Etienne F, Etienne F. Prevalence of postural hypotension in elderly patients in a long-term health care facility. 19880826 DCOM- 19880826(0002-9149 (Print)).

22. Frith J. Diagnosing orthostatic hypotension: a narrative review of the evidence. 20150909(1471-8391 (Electronic)).

23. Gorelik O, Cohen N. Seated postural hypotension. 20151221(1878-7436 (Electronic)).

24. Verwoert GC, Mattace-Raso FU, Hofman A, et al. Orthostatic hypotension and risk of cardiovascular disease in elderly people: the Rotterdam study. Journal of the American Geriatrics Society. 2008 Oct;56(10):1816-20.

25. Xin W, Lin Z, Mi S. Orthostatic hypotension and mortality risk: a meta-analysis of cohort studies. Heart. 2014 Mar;100(5):406-13.

26. Masaki KH, Schatz IJ, Burchfiel CM, et al. Orthostatic hypotension predicts mortality in elderly men: the Honolulu Heart Program. Circulation. 1998 Nov 24;98(21):2290-5.

27. Rose KM, Tyroler HA, Nardo CJ, et al. Orthostatic hypotension and the incidence of coronary heart disease: the Atherosclerosis Risk in Communities study. American journal of hypertension. 2000 Jun;13(6 Pt 1):571-8.

28. Benvenuto LJ, Krakoff LR. Morbidity and mortality of orthostatic hypotension: implications for management of cardiovascular disease. American journal of hypertension. 2011 Feb;24(2):135-44.

29. Scheffer AC, van Hensbroek PB, van Dijk N, et al. Risk factors associated with visiting or not visiting the accident & emergency department after a fall. BMC health services research. 2013;13:286.

30. Shaw BH, Claydon VE. The relationship between orthostatic hypotension and falling in older adults. Clinical autonomic research : official journal of the Clinical Autonomic Research Society. 2014 Feb;24(1):3-13.

31. Angelousi A, Girerd N, Benetos A, et al. Association between orthostatic hypotension and cardiovascular risk, cerebrovascular risk, cognitive decline and falls as well as overall mortality: a systematic review and meta-analysis. Journal of hypertension. 2014 Aug;32(8):1562-71; discussion 71.

32. Ambrose AF, Paul G, Hausdorff JM. Risk factors for falls among older adults: A review of the literature. Maturitas. 2013;75(1):51-61.

33. Ooi WL, Hossain M, Lipsitz LA. The association between orthostatic hypotension and recurrent falls in nursing home residents. The American journal of medicine. 2000 Feb;108(2):106-11.

34. Maurer MS, Cohen S, Cheng H. The degree and timing of orthostatic blood pressure changes in relation to falls in nursing home residents. Journal of the American Medical Directors Association. 2004 Jul-Aug;5(4):233-8.

35. Blumenthal MD, Davie JW. Dizziness and falling in elderly psychiatric outpatients. The American journal of psychiatry. 1980 Feb;137(2):203-6.

36. Romero-Ortuno R, Cogan L, Foran T, Kenny RA, Fan CW. Continuous noninvasive orthostatic blood pressure measurements and their relationship with orthostatic intolerance, falls, and frailty in older people. Journal of the American Geriatrics Society. 2011 Apr;59(4):655-65.

37. van der Velde N, van den Meiracker AH, Stricker BH, van der Cammen TJ. Measuring orthostatic hypotension with the Finometer device: is a blood pressure drop of one heartbeat clinically relevant? Blood pressure monitoring. 2007 Jun;12(3):167-71.

38. Heitterachi E, Lord SR, Meyerkort P, McCloskey I, Fitzpatrick R. Blood pressure changes on upright tilting predict falls in older people. Age and ageing. 2002 May;31(3):181-6.

39. Lieberman D, Friger M, Lieberman D. Inpatient rehabilitation outcome after hip fracture surgery in elderly patients: a prospective cohort study of 946 patients. Archives of physical medicine and rehabilitation. 2006 Feb;87(2):167-71.

40. Koot VC, Peeters PH, de Jong JR, Clevers GJ, van der Werken C. Functional results after treatment of hip fracture: a multicentre, prospective study in 215 patients. The European journal of surgery = Acta chirurgica. 2000 Jun;166(6):480-5.

R1

R2

R3

R4

R5

R6

R7

R8

R9

R10

R11

R12

R13

R14

R15

R16

R17

R18

R19

R20

R21

R22

R23

R24

R25

R26

R27

R28

R29

R30

R31

R32

R33

R34

R35

R36

R37

R38

R39

Chapter 1

22

41. Clague JE, Craddock E, Andrew G, Horan MA, Pendleton N. Predictors of outcome following hip fracture. Admission time predicts length of stay and in-hospital mortality. Injury. 2002 Jan;33(1):1-6.

42. Muir SW, Yohannes AM. The impact of cognitive impairment on rehabilitation outcomes in elderly patients admitted with a femoral neck fracture: a systematic review. Journal of geriatric physical therapy (2001). 2009;32(1):24-32.

43. Di Monaco M, Castiglioni C Fau - De Toma E, De Toma E Fau - Gardin L, et al. Handgrip strength is an independent predictor of functional outcome in hip-fracture women: a prospective study with 6-month follow-up. 20150214 DCOM- 20150423(1536-5964 (Electronic)).

44. Oude Voshaar RC, Banerjee S Fau - Horan M, Horan M Fau - Baldwin R, et al. Fear of falling more important than pain and depression for functional recovery after surgery for hip fracture in older people. 20061016 DCOM- 20070206(0033-2917 (Print)).

45. Di Monaco M, Castiglioni C, De Toma E, et al. Handgrip strength but not appendicular lean mass is an independent predictor of functional outcome in hip-fracture women: a short-term prospective study. 20140830 DCOM- 20141021(1532-821X (Electronic)).

46. Visschedijk JH, Caljouw MA, Bakkers E, van Balen R, Achterberg WP. Longitudinal follow-up study on fear of falling during and after rehabilitation in skilled nursing facilities. 20151205(1471-2318 (Electronic)).

47. Fried LP, Tangen Cm Fau - Walston J, Walston J Fau - Newman AB, et al. Frailty in older adults: evidence for a phenotype. 20010316 DCOM- 20010329(1079-5006 (Print)).

48. O’Connell MD, Savva GM, Fan CW, Kenny RA. Orthostatic hypotension, orthostatic intolerance and frailty: The Irish Longitudinal Study on Aging-TILDA. 20150413 DCOM- 20151117(1872-6976 (Electronic)).

49. Rockwood K, Abeysundera Mj Fau - Mitnitski A, Mitnitski A. How should we grade frailty in nursing home patients? 20071113 DCOM- 20071213(1538-9375 (Electronic)).

50. Moreland JD, Richardson Ja Fau - Goldsmith CH, Goldsmith Ch Fau - Clase CM, Clase CM. Muscle weakness and falls in older adults: a systematic review and meta-analysis. 20040622 DCOM- 20050405(0002-8614 (Print)).

51. Rutan GH, Hermanson B, Bild DE, et al. Orthostatic hypotension in older adults. The Cardiovascular Health Study. CHS Collaborative Research Group. Hypertension. 1992 Jun;19(6 Pt 1):508-19.

52. Weiss A, Beloosesky Y Fau - Kornowski R, Kornowski R Fau - Yalov A, et al. Influence of orthostatic hypotension on mortality among patients discharged from an acute geriatric ward. 20060630 DCOM- 20060911(1525-1497 (Electronic)).

53. Casiglia E, Tikhonoff V Fau - Caffi S, Caffi S Fau - Boschetti G, et al. Orthostatic hypotension does not increase cardiovascular risk in the elderly at a population level. 20131204 DCOM- 20140728(1941-7225 (Electronic)).

54. Ricci F, Fedorowski A, Radico F, et al. Cardiovascular morbidity and mortality related to orthostatic hypotension: a meta-analysis of prospective observational studies. European heart journal. 2015 Jul 1;36(25):1609-17.

55. Fedorowski A, Burri P Fau - Melander O, Melander O. Orthostatic hypotension in genetically related hypertensive and normotensive individuals. 20090428 DCOM- 20090723(1473-5598 (Electronic)).

56. Moonen JE, Foster-Dingley JC, de Ruijter W, et al. Effect of discontinuation of antihypertensive medication on orthostatic hypotension in older persons with mild cognitive impairment: the DANTE Study Leiden. Age and ageing. 2016 Jan 11.

57. Beckett NS, Peters R, Fletcher AE, et al. Treatment of hypertension in patients 80 years of age or older. The New England journal of medicine. 2008 May 1;358(18):1887-98.

58. Rastas S, Pirttila T, Viramo P, et al. Association between blood pressure and survival over 9 years in a general population aged 85 and older. Journal of the American Geriatrics Society. 2006 Jun;54(6):912-8.

R1

R2

R3

R4

R5

R6

R7

R8

R9

R10

R11

R12

R13

R14

R15

R16

R17

R18

R19

R20

R21

R22

R23

R24

R25

R26

R27

R28

R29

R30

R31

R32

R33

R34

R35

R36

R37

R38

R39

ntroduction

23

1

59. van Bemmel T, Gussekloo J, Westendorp RG, Blauw GJ. In a population-based prospective study, no association between high blood pressure and mortality after age 85 years. Journal of hypertension. 2006 Feb;24(2):287-92.

60. Benetos A, Labat C, Rossignol P, et al. Treatment With Multiple Blood Pressure Medications, Achieved Blood Pressure, and Mortality in Older Nursing Home Residents: The PARTAGE Study. JAMA internal medicine. 2015 Feb 16.

61. Benetos A, Gautier S, Labat C, et al. Mortality and cardiovascular events are best predicted by low central/peripheral pulse pressure amplification but not by high blood pressure levels in elderly nursing home subjects: the PARTAGE (Predictive Values of Blood Pressure and Arterial Stiffness in Institutionalized Very Aged Population) study. J Am Coll Cardiol. 2012 Oct 16;60(16):1503-11.

62. Askari M, Kiely DK, Lipsitz LA. Is pulse pressure a predictor of cardiovascular complications in a frail elderly nursing home population? Aging clinical and experimental research. 2004 Jun;16(3):206-11.

63. Dev MK, Paudel N, Joshi ND, Shah DN, Subba S. Psycho-social impact of visual impairment on health-related quality of life among nursing home residents. BMC health services research. 2014;14:345.

64. Kanwar A, Singh M, Lennon R, et al. Frailty and health-related quality of life among residents of long-term care facilities. Journal of aging and health. 2013 Aug;25(5):792-802.

65. Almomani FM, McDowd Jm Fau - Bani-Issa W, Bani-Issa W Fau - Almomani M, Almomani M. Health-related quality of life and physical, mental, and cognitive disabilities among nursing home residents in Jordan. 20140124 DCOM- 20140912(1573-2649 (Electronic)).

66. Drageset J, Natvig GK, Eide GE, et al. Differences in health-related quality of life between older nursing home residents without cognitive impairment and the general population of Norway. Journal of clinical nursing. 2008 May;17(9):1227-36.

67. Ferrer A, Formiga F Fau - Cunillera O, Cunillera O Fau - Megido MJ, et al. Predicting factors of health-related quality of life in octogenarians: a 3-year follow-up longitudinal study. 20150524(1573-2649 (Electronic)).

68. Sitoh YY, Lau Tc Fau - Zochling J, Zochling J Fau - Schwarz J, et al. Determinants of health-related quality of life in institutionalised older persons in northern Sydney. 20050211 DCOM- 20050610(1444-0903 (Print)).

69. Missotten P, Squelard G, Ylieff M, et al. Quality of life in older Belgian people: comparison between people with dementia, mild cognitive impairment, and controls. International journal of geriatric psychiatry. 2008 Nov;23(11):1103-9.

70. Wetzels RB, Zuidema SU, de Jonghe JF, Verhey FR, Koopmans RT. Determinants of quality of life in nursing home residents with dementia. Dement Geriatr Cogn Disord. 2010;29(3):189-97.

71. Tabali M, Ostermann T, Jeschke E, Dassen T, Heinze C. Does the care dependency of nursing home residents influence their health-related quality of life?-A cross-sectional study. Health and quality of life outcomes. 2013;11:41.

72. van der Zee KI SR. Het meten van de algemene gezondheidstoestand met de Rand-36, een handleiding. Tweede herziene druk. UMCG / Rijksuniversiteit Groningen, Research Institute SHARE.; 2012 [December 2014]; Available from: http://www.rug.nl/share.

73. Tsai SY, Chi LY, Lee CH, Chou P. Health-related quality of life as a predictor of mortality among community-dwelling older persons. European journal of epidemiology. 2007;22(1):19-26.

CHAPTER 2Measuring orthostatic blood pressure during different

postural changes: standing versus sitting

Submitted as: Breeuwsma AC, Hartog LC, Kamper AM, Groenier KH, Bilo HJG, Kleefstra N,

Van Hateren KJJ. Measuring orthostatic blood pressure during different

postural changes: standing versus sitting.

R1

R2

R3

R4

R5

R6

R7

R8

R9

R10

R11

R12

R13

R14

R15

R16

R17

R18

R19

R20

R21

R22

R23

R24

R25

R26

R27

R28

R29

R30

R31

R32

R33

R34

R35

R36

R37

R38

R39

Chapter 2

26

ABSTRACT

ac ground: Orthostatic hypotension (OH) is defined as a drop in systolic blood pressure (SBP) of ≥20 mmHg and/or diastolic blood pressure (DBP) of ≥10 mmHg within 3 minutes of standing. Sitting blood pressure (BP) measurements can be an alternative in patients unable to stand. We aimed to investigate the difference in BP response and OH prevalence between the standard postural change to the sitting and the standing position.

Methods: A cross-sectional observational study. Inclusion criteria: adults >50 years and a medical history of cardiovascular disease, diabetes mellitus, and/or hypertension. BP was measured with a continuous BP measurement device during two postural changes, from supine to the sitting, and from supine to the standing position, in each patient. The lowest SBP and DBP were recorded in eight different timeframes after postural change. Linear mixed models were used to investigate the differences in changes (D) of SBP and DBP between the two postural changes. Prevalence, positive and negative proportions of agreement were calculated to observe the agreement of diagnosis OH between the two postural changes.

esults: 104 patients with a mean age of 69 years were included. DSBP was significantly larger in the standing position compared to the sitting with -11.5 [95% Confidence Interval (CI) -17.0, -5.9] and -8.7 [95% CI -14.2, -3.2] mmHg between 0-44 s.DDBP was significantly larger in the sitting position compared to the standing 75-224 s after postural change; with 4.1 (95%CI 1.4,6.9), 3.3 (95%CI 0.6,6.0), 4.46 (95%CI 1.8,7.2), 4.3 (95%CI 1.6,7.0), and 4.4 (95%CI 1.7,7.1) mmHg.The prevalence of OH was 66.3% (95% Confidence Interval (CI) 57.2; 75.4) in standing position and 67.3% (95%CI 58.3; 76.3) in sitting position. The positive proportion of agreement was 74.8% and the negative proportion of agreement was 49.3%.

onclusions: A clear difference was seen in blood pressure response between the two postural changes. Although no significant difference in prevalence of OH was observed, the positive and negative proportion of agreement of the prevalence of OH were poor to moderate which indicates a different outcome between both postural changes.

R1

R2

R3

R4

R5

R6

R7

R8

R9

R10

R11

R12

R13

R14

R15

R16

R17

R18

R19

R20

R21

R22

R23

R24

R25

R26

R27

R28

R29

R30

R31

R32

R33

R34

R35

R36

R37

R38

R39

Measuring orthostatic blood pressure during different postural changes: standing versus sitting

27

2

INTRODUCTION

Orthostatic hypotension (OH) is a key manifestation of haemodynamic dysfunction observed when adaptive mechanisms fail to compensate for a sudden reduction in venous return during active postural change [1, 2]. The postural change leads to pooling of blood in the pelvis and lower extremities caused by gravity. Counteracting circulatory mechanisms lead to an increase in heart rate (HR) and peripheral vasoconstriction and result, in combination with the skeletal muscle pump, in an increase of venous return [3-5]. The prevalence of OH rises with age [6, 7], varying from 12% to 18% in community-dwelling elderly [8, 9], and from 37% to more than 50% in nursing home residents [10, 11].

The international consensus definition recommends continuous beat-to-beat blood pressure (BP) measurements to diagnose OH [12], and orthostatic BP is advised to be measured in standing position after 5 minutes of rest in supine position [12]. However, as many elderly patients are not able to stand during several minutes, sitting orthostatic BP measurements are sometimes used as an alternative [10]. The prevalence of seated OH was described in the review of Gorelik and Cohen and varied from 8% within community-dwelling individuals to 56% within elderly hospitalized patients [13]. They conclude that seated OH should be assessed in patients unable to stand. None of the studies described in the review compared seated versus standing OH measurement.

Differences in prevalence of OH measured either in the standing or the sitting position are unknown in the elderly population. Therefore, we aimed to investigate the difference in BP response and prevalence of OH between two different postural changes: standing versus sitting.

MATERIAL AND METHODS

tud populationFor this cross-sectional observational study patients were recruited from the outpatient clinic of internal medicine (Isala hospital, Zwolle, the Netherlands). Inclusion criteria were adults over 50 years of age combined with a medical history of one or more of the following diseases; cardiovascular disease (CVD), diabetes mellitus (DM), and hypertension. Exclusion criteria were the inability to perform BP measurements at one arm, inability to stand without assistance, known peripheral vessel disease in one or both arms, needing a large (≥42 cm) or small (≤28 cm) upper arm cuff, and incapability of giving consent. We aimed to include at least 100 elderly participants. Patients were non-blinded randomized for both the

R1

R2

R3

R4

R5

R6

R7

R8

R9

R10

R11

R12

R13

R14

R15

R16

R17

R18

R19

R20

R21

R22

R23

R24

R25

R26

R27

R28

R29

R30

R31

R32

R33

R34

R35

R36

R37

R38

R39

Chapter 2

28

sequence of the postural changes and the side of the BP measurements. The randomization was performed in blocks of four (Figure 1).

104 patients

52 patientsFirst postural change:

Supine to standing

52 patientsFirst postural change:

Supine to sitting

26 patientsBP measurement rightb

26 patientsBP measurement lefta

26 patientsBP measurement lefta

26 patientsBP measurement rightb

Figure 1. Randomization. a: Continuous BP measurement device on the left arm. b: Continuous BP measurement device on the right arm.

ata collectionBaseline data included demographic characteristics, a full medical history including a history of CVD, DM, hypertension, polyneuropathy, Parkinson’s disease, pacemaker implantation, falls in the previous year, and medication use. A history of CVD was defined as a history of angina pectoris, myocardial infarction, percutaneous transluminal coronary angioplasty, coronary artery bypass grafting, stroke, and/or transient ischemic attack. All measurements were performed by one of the authors (ACB). Each participant performed both postural changes in a cross-over design (supine to standing and supine to sitting) and remained in those positions for at least 4 minutes. Prior to the postural change, baseline BP was measured in the supine position after 5 minutes of rest. BP was measured with the Finometer Pro (Finapres Medical Systems BV), a continuous non-invasive beat-to-beat BP measurement device, which has previously been validated compared to invasive BP recordings [14, 15]. Finger circumference was measured to apply the proper sized finger cuff of the Finometer Pro [16]. In addition, height differences were corrected by a height nulling procedure and by supporting the finger cuff at heart level during the whole procedure [16, 17]. During the measurements, correct positioning of the arm was repeatedly checked. The Finometer Pro was calibrated approximately 3 minutes before each postural change using the return-to-flow (RTF) calibration system, which monitors the finger pressure distal of an occluding upper arm cuff to align the finger BP to brachial BP [16]. The presence of characteristic symptoms of OH like dizziness, blurred vision, or light-headedness was asked.BP measurement data of the Finometer Pro were exported with the BeatScope software (Finapres Medical Systems BV). By measuring the arterial finger pressure, cardiac output (CO) was calculated with the use of the model flow method [18]. Baseline mean supine

R1

R2

R3

R4

R5

R6

R7

R8

R9

R10

R11

R12

R13

R14

R15

R16

R17

R18

R19

R20

R21

R22

R23

R24

R25

R26

R27

R28

R29

R30

R31

R32

R33

R34

R35

R36

R37

R38

R39


29

2

systolic blood pressure (SBP), diastolic blood pressure (DBP), HR, and CO were calculated over the last minute prior to each postural change. After postural change, the lowest SBP and DBP values were recorded, and the mean HR and CO values were calculated over eight different timeframes (0-14s, 15-44s, 45-74s, 75-104s, 105-134s, 135-164s, 165-194s, and 195-224s). Records with poor quality signals (e.g. artefacts) were excluded by visual inspection of the graphics in the BeatScope output files. OH was defined as a drop in SBP of ≥20 mmHg or a drop in DBP of ≥10 mmHg within 3 minutes after postural change [19], excluding the first 15 seconds. Initial OH (IOH) was defined as a drop in SBP of ≥40 mmHg and/or a drop in DBP of ≥20 mmHg within the first 15 seconds after postural change accompanied by orthostatic complaints [1, 20].

ndpointsThe primary endpoints were the differences in change of SBP and DBP between the two postural changes (supine to sitting versus supine to standing). Secondary endpoints were the difference in change of CO and HR, and the difference in prevalence and proportions of agreement of OH, IOH, and orthostatic complaints between the two postural changes.

tatistical anal sisContinuous variables were presented as mean and standard deviation (SD) for normally distributed variables, or as median and interquartile range (IQR) for non-normally distributed variables. Categorical variables were presented as proportions. Q-Q plots and histograms were constructed to examine deviations of normality. The difference in SBP, DBP, HR, and CO between supine and sitting or standing position at each timeframe was defined as change (D). Linear mixed models (with timeframe nested within posture) were performed to investigate the differences in DSBP, DDBP, DHR, and DCO at each particular timeframe between the two postural changes for the patients and healthy participants. The differences between both postural changes at each timeframe were tested using the Bonferroni correction for multiple testing. In addition, linear mixed models adjusted for sequence (sitting-standing or standing-sitting) and period (first or second measurement) were used to compare the Area Under the Curve (AUC) of the sitting and standing SBP, DBP, HR, and CO curves. Evaluating the AUC is a better method compared to single clinical BP measurements to determine the haemodynamic state in hypertensive subjects [21]. Additionally, the differences in prevalence of OH, IOH, and orthostatic complaints according to the postural change were analysed with McNemar tests. The positive and the negative proportions of agreement were calculated [22]. The positive proportion of agreement is the number of both postural changes diagnosed OH divided by the total number of OH diagnosed for each of the postural changes. The negative proportion of agreement is the number of both postural changes excluded OH divided by the total number of excluded OH

R1

R2

R3

R4

R5

R6

R7

R8

R9

R10

R11

R12

R13

R14

R15

R16

R17

R18

R19

R20

R21

R22

R23

R24

R25

R26

R27

R28

R29

R30

R31

R32

R33

R34

R35

R36

R37

R38

R39

Chapter 2

30

for each of the postural changes. Both positive and negative proportions of agreement are reported as percentages.All tests were two-sided and P-values were considered to be significant at p <0.05. Statistical analyses were performed using the SPSS software (version 23; IBM, Armonk, New York, USA). The corresponding author had full access to all the data in the study and assumes responsibility for the accuracy and completeness of the data and all the analyses.

thical approval and clinical trial registrationThis study was approved by the local medical ethics committee of Isala (number 15.06.95) and was performed in accordance with the declaration of Helsinki. Written informed consent was obtained from each participant during their scheduled appointment. All data was analysed anonymously. The ‘Strengthening the Reporting of Observational Studies in Epidemiology’ (STROBE) statement was used to describe this observational cohort study [23]. The study was registered at www.trialregister.nl (NTR5525).

RESULTS

The inclusion and all study procedures were performed in January and February 2016. A total of 104 patients were included in the present study (Figure 2).

402 patients

127 patients

104 patients

275 excluded by their specialist or unwillingness to

participate

23 excluded due to measurement problems

Figure 2. Patient selection.

R1

R2

R3

R4

R5

R6

R7

R8

R9

R10

R11

R12

R13

R14

R15

R16

R17

R18

R19

R20

R21

R22

R23

R24

R25

R26

R27

R28

R29

R30

R31

R32

R33

R34

R35

R36

R37

R38

R39


31

2

Reasons to exclude patients due to measurement problems consisted of failing to find a HR on the finger cuff (n=11) and unavailability of a proper sized cuff (n=12). Baseline characteristics of the study population are presented in Table 1. In this cohort, 104 patients (59 men, 45 women) with a mean age of 68.8 years (SD 8.5) were included. Baseline characteristics of patients with the first postural change to sitting and patients with the first postural change to standing, and patients with the Finometer on the left arm and patients with the Finometer on the right arm were presented in Appendix Table A1.

Table 1. Baseline characteristics.

Patients (n=104)CharacteristicsAge (years) 68.8 (8.5)*Female 45 (43)BMI (kg/m2) 27.0 (24.7-31.7)†

Medical HistoryHypertension 82 (79)DM 64 (62)History of CVD 38 (37)Polyneuropathy 33 (32)Parkinson’s disease 0 (0)Pacemaker implantation 8 (8)History of falls 31 (30)

MedicationAnti-hypertensive medication 86 (83)Anti-arrhythmic medication 6 (6)Nitrates 13 (13)Oral glucose lowering therapy 32 (31)Insulin 52 (50)Psychiatric medication 13 (13)Anti-parkinsonian medication 1 (1)

MeasurementsLying SBP (mmHg) 150.3 (137.3-162.9)†

Lying DBP (mmHg) 76.7 (9.2)*Lying HR (beats/min) 67.6 (60.4-73.8)†

Lying CO (L/min) 6.2 (1.2)* Drinking and/or eating 63 (61)

Values are presented as n (%), unless indicated otherwise. *: Mean (±SD). †: Median (p25-75).

R1

R2

R3

R4

R5

R6

R7

R8

R9

R10

R11

R12

R13

R14

R15

R16

R17

R18

R19

R20

R21

R22

R23

R24

R25

R26

R27

R28

R29

R30

R31

R32

R33

R34

R35

R36

R37

R38

R39

Chapter 2

32

ostural change and hae od na ic changes The results of the linear mixed models are presented in table 2, and illustrated in figure 3. ∆SBP was significantly larger in the standing position compared to the sitting with -11.5 [95% Confidence Interval (CI) -17.0, -5.9] and -8.7 [95% CI -14.2, -3.2] mmHg between 0-44 s (p<.001). Beside, ∆DBP was significantly larger in the sitting position compared to the standing 75-224 s after postural change; with 4.1 (95%CI 1.4,6.9), 3.3 (95%CI 0.6,6.0), 4.5 (95%CI 1.8,7.2), 4.3 (95%CI 1.6,7.0), and 4.4 (95%CI 1.7,7.1) mmHg (p<.05). In sitting position, ∆HR was larger compared to the standing position at all timeframes after postural change (p<.001), difference in ∆ ranged from 4.5-8.1/min. Sitting ∆CO was smaller compared to standing ∆CO for all timeframes (difference in ∆ ranged from minus 0.4-0.7 L/min), except during first 14 seconds, in which sitting ∆CO was higher than standing ∆CO (difference in ∆ -0.7 L/min (95%CI -1.1,-0.4)(p<.001)) (Table 2).

Both the AUC of SBP (p=.023) and CO (p=.001) were larger, while the AUC of DBP (p=.002) and HR (p<.001) were smaller in sitting position, all compared to the standing position (Figure 3).

revalence o and orthostatic co plaints The prevalence of OH was 66.3% (95% Confidence Interval (CI) 57.2; 75.4) in standing position and 67.3% (95%CI 58.3; 76.3) in sitting position. In 52 out of all 104 patients OH was diagnosed in both postural changes. The positive proportion of agreement was 74.8% and the negative proportion of agreement was 49.3%. IOH was present in 5.8% (95%CI 1.3; 10.3) and 16.3% (95%CI 9.2; 23.4) in sitting and standing position, respectively (p-value for difference = .013). The positive proportion of agreement was 26.1% and the negative proportion of agreement was 90.8%. Orthostatic complaints were reported in 13 patients (12.5%, 95%CI 6.1; 18.9) in the sitting position and in 23 patients (22.1%, 95%CI 14.1; 30.1) in the standing position (p-value for the difference = .021). The positive proportion of agreement was 55.6% and the negative proportion of agreement was 90.7%.

R1

R2

R3

R4

R5

R6

R7

R8

R9

R10

R11

R12

R13

R14

R15

R16

R17

R18

R19

R20

R21

R22

R23

R24

R25

R26

R27

R28

R29

R30

R31

R32

R33

R34

R35

R36

R37

R38

R39


33

2

Tabl

e 2.

Mea

n si

tting

and

sta

ndin

g ∆S

BP, ∆

DBP

, ∆H

R, a

nd ∆

CO p

er ti

mef

ram

e in

all

patie

nts.

Syst

olic

Blo

od P

ress

ure

(mm

Hg)

*D

iast

olic

Blo

od P

ress

ure

(mm

Hg)

*Ti

mef

ram

eSi

tting

∆†

[95%

CI]

Stan

ding

∆‡

[95%

CI]

Diff

eren

ce in

∆Si

tting

∆†

[95%

CI]

Stan

ding

∆‡

[95%

CI]

Diff

eren

ce in

∆

0-14

s30

.3 [2

7.4,

33.

2]41

.8 [3

8.9,

44.

7]-1

1.49

[-17

.01,

-5.9

6]§

19.4

[18.

1, 2

0.8]

20.1

[18.

7, 2

1.4]

-0.6

5 [-

3.36

, 2.0

6]15

-44s

17.7

[14.

8, 2

0.6]

26.4

[23.

5, 2

9.3]

-8.7

2 [-

14.2

4, -3

.19]

§9.

8 [8

.4, 1

1.2]

7.8

[6.4

, 9.1

]2.

02 [-

0.69

, 4.7

3]45

-74s

15.2

[12.

3, 1

8.1]

19.5

[16.

6, 2

3.4]

-4.3

1 [-

9.84

, 1.2

1]6.

6 [5

.2, 7

.9]

4.0

[2.7

, 5.4

]2.

55 [-

0.16

, 5.2

6]75

-104

s12

.9 [1

0.0,

15.

7]12

.3 [9

.4, 1

5.2]

0.53

[-5.

00, 6

.05]

5.2

[3.9

, 6.6

]1.

1 [-

0.3,

2.5

]4.

14 [1

.43,

6.8

5]§

105-

134s

13.2

[10.

3, 1

6.1]

12.0

[9.1

, 14.

9]1.

22 [-

4.31

, 6.7

4]5.

0 [3

.7, 6

.4]

1.8

[0.4

, 3.1

]3.

29 [0

.58,

6.0

0]|

|

135-

164s

13.1

[10.

2, 1

6.0]

9.1

[6.2

, 12.

0]4.

02 [-

1.51

, 9.5

4]6.

1 [4

.7, 7

.4]

1.6

[0.2

, 3.0

]4.

46 [1

.75,

7.1

7]§

165-

194s

13.2

[10.

3, 1

6.1]

9.4

[6.5

, 12.

3]3.

76 [-

1.77

, 9.2

8]5.

8 [4

.5, 7

.2]

1.5

[0.1

, 2.9

]4.

33 [1

.62,

7.0

4]§

195-

224s

11.7

[8.9

, 14.

6]7.

6 [4

.7, 1

0.5]

4.14

[-1.

38, 9

.66]

4.9

[3.6

, 6.3

]0.

5 [-

0.9,

1.9

]4.

43 [1

.72,

7.1

4]§

Hea

rt R

ate

(bea

ts/m

in)*

Card

iac

Out

put (

L/m

in)*

Tim

efra

me

Sitti

ng ∆

†

[95%

CI]

Stan

ding

∆‡

[95%

CI]

Diff

eren

ce in

∆Si

tting

∆†

[95%

CI]

Stan

ding

∆‡

[95%

CI]

Diff

eren

ce in

∆

0-14

s-7

.2 [-

8.2,

-6.3

]-1

1.7

[-12

.6, -

10.7

]4.

45 [2

.70,

6.2

1]§

-1.0

0 [-

1.18

, -0.

83]

-0.2

8 [-

0.45

, -0.

10]

-0.7

3 [-

1.06

, -0.

39]§

15-4

4s-2

.1 [-

3.0,

-1.1

]-9

.3 [-

10.2

, -8.

3]7.

23 [5

.48,

8.9

8]§

-0.0

9 [-

0.27

-0.0

9]0.

56 [0

.38,

0.7

4]-0

.65

[-0.

98, -

0.31

]§

45-7

4s-1

.3 [-

2.3,

-0.4

]-9

.4 [-

10.4

, -8.

5]8.

10 [6

.35,

9.8

6]§

0.24

[0.0

6, 0

.42]

0.71

[0.5

3, 0

.89]

-0.4

7 [-

0.83

, -0.

16]§

75-1

04s

-1.1

[-2.

0, -0

.1]

-8.7

[-9.

7, -7

.8]

7.67

[5.9

2, 9

.43]

§0.

27 [0

.10,

0.4

5]0.

73 [0

.56,

0.9

1]-0

.46

[-0.

79, -

0.13

]§

105-

134s

-1.6

[-2.

6, -0

.7]

-8.9

[-9.

9, -8

.0]

7.29

[5.5

4, 9

.05]

§0.

27 [0

.09,

0.4

4]0.

65 [0

.48,

0.8

3]-0

.39

[-0.

72, -

0.05

]||

135-

164s

-1.6

[-2.

6, -0

.7]

-9.1

[-10

.0, -

8.1]

7.47

[5.7

2, 9

.22]

§0.

19 [0

.10,

0.3

6]0.

57 [0

.40,

0.7

5]-0

.39

[-0.

72, -

0.05

]||

165-

194s

-1.2

[-2.

1, -0

.2]

-8.1

[-9.

1, -7

.2]

6.97

[5.2

2, 8

.72]

§0.

20 [0

.02,

0.3

7]0.

61 [0

.43,

0.7

9]-0

.41

[-0.

75, -

0.08

]||

195-

224s

-0.8

[-1.

7, 0

.2]

-8.0

[-8.

9, -7

.0]

7.21

[5.4

5, 8

.96]

§0.

24 [0

.06,

0.4

1]0.

59 [0

.41,

0.7

7]-0

.35

[-0.

69, -

0.02

]||

*: T

este

d w

ith li

near

mix

ed m

odel

s. †

: Mea

n ly

ing

BP m

inus

BP

at s

ingl

e tim

efra

mes

in s

itting

pos

ition

. ‡

: M

ean

lyin

g BP

min

us B

P at

sin

gle

timef

ram

es i

n st

andi

ng p

ositi

on.

§: S

igni

fican

t at

p <

.001

(Bo

nfer

roni

cor

rect

ed).

||:

Sign

ifica

nt a

t p

<.05

(B

onfe

rron

i cor

rect

ed).

R1

R2

R3

R4

R5

R6

R7

R8

R9

R10

R11

R12

R13

R14

R15

R16

R17

R18

R19

R20

R21

R22

R23

R24

R25

R26

R27

R28

R29

R30

R31

R32

R33

R34

R35

R36

R37

R38

R39

Chapter 2

34

Figure .

Figure 3. SBP (A), DBP (B), HR (C), and CO (D) classified in postural change as predicted by linear mixed models.

DISCUSSION

Standing resulted in a greater SBP decrease compared to sitting, whereas the opposite was observed for DBP. Although no significant difference in the prevalence of OH was observed, the positive and negative proportions of agreement of the prevalence of OH were at best moderate, indicating that a diagnosis of OH is highly dependent on the method that was used.

ostural change and hae od na ic changesIt is known that by changing positions from supine to the sitting or the standing position, haemodynamic adaptive mechanisms are activated due to the sudden decrease in BP [3, 4,

R1

R2

R3

R4

R5

R6

R7

R8

R9

R10

R11

R12

R13

R14

R15

R16

R17

R18

R19

R20

R21

R22

R23

R24

R25

R26

R27

R28

R29

R30

R31

R32

R33

R34

R35

R36

R37

R38

R39


35

2

24-26]. As seen in the sitting and standing curves, SBP changed differently between the two positions at several timeframes, although the shape of both curves appears fairly similar. The decrease in SBP was higher in the standing position compared to the sitting at the first two timeframes, which could be explained by the larger hydrostatic effects in the standing position. Due to the loss of elastin fibres and consequently less compliance and elasticity in patients with atherosclerosis and thence increased arterial stiffness caused by for instance hypertension or diabetes [27, 28], the compensation for the larger hydrostatic effects could be delayed [29]. Although the DBP curves showed a similar trend, the decrease in DBP was larger in the sitting position compared to the standing at the last five timeframes after postural change. As muscle activity in the sitting position is almost 2.5 times lower compared to the standing [30], the lower DBP could be a physiological effect from reduced activation of the skeletal muscle pump. Subsequently, a lower muscle activity can result in a reduced peripheral vascular resistance and a fall in DBP [31].Postural change resulted in an increased HR and an increase in CO, which subsequently was followed by a decrease in CO. HR was higher and CO was lower in standing position compared to sitting after postural change. This reaction could be explained by the response to the larger hydrostatic effects in the standing position compared to the sitting and thereby a decreased venous return [32]. This postural response of HR was previously described in elderly [26].

revalence o and orthostatic co plaintsAll the above mentioned BP differences did not result in an overall difference in the prevalence of OH between the sitting and the standing postural change. Nevertheless, although prevalence was similar in both postural changes, the positive proportion of agreement of the prevalence of OH was only 75% in the present study. This indicates that 75% of the subjects with OH were diagnosed with OH in both postural changes. The negative proportion of agreement of the prevalence of OH was 49%. Although the proportion of agreement is highly useful in clinical practice, no standard references for high or low proportion of agreement are described [22]. In our opinion, a positive proportion of 75% is moderate and a negative proportion of 49% is low, indicates that the different outcome between both postural changes is relevant.

The prevalence of IOH and orthostatic complaints were significantly higher after postural change to standing position. The higher prevalence of IOH and orthostatic complaints in the standing position compared to the sitting could be explained by the larger decrease in standing SBP in the first timeframe after postural change. The prevalence of standing IOH in present study of 16.3% was low compared to a previous published study, in which a

R1

R2

R3

R4

R5

R6

R7

R8

R9

R10

R11

R12

R13

R14

R15

R16

R17

R18

R19

R20

R21

R22

R23

R24

R25

R26

R27

R28

R29

R30

R31

R32

R33

R34

R35

R36

R37

R38

R39

Chapter 2

36

prevalence of 58% was reported [33]. The difference in prevalence was probably caused by the higher age in the previous study (80.6 versus 68.8 years). In a previous study concerning a group of patients with OH, the prevalence of orthostatic complaints was comparable to the results in the present study [8]. The positive proportion of agreement for IOH and orthostatic complaints were both poor, which indicates that different patients were diagnosed with IOH or orthostatic complaints between the two postural changes.

trengths and li itationsIn the present study, several strengths can be mentioned. As far as we know, this is the first study investigating the difference in haemodynamic response between the sitting and the standing postural change. Linear mixed models are highly reliable in comparing haemodynamic parameters over multiple timeframes [34] and all measurements were performed by the same individual. Finally, all patients were non-blinded randomized for both the sequence of the postural changes and the side of the BP measurements.Limitations of our study are the small study sample and the possibility of selection bias. Due to the fact that the patients included in this study had to be able to stand for five minutes without assistance, the study group was slightly biased compared to the more vital visitors of the outpatient department and the results are, of course, only useful in patients who are able to stand. Also, two types of measurement problems can be mentioned. Firstly, estimating CO with the model flow method in the Finometer Pro has previously been questioned and is therefore not completely reliable [35]. Secondly, delayed standing or sitting in patients with mobility problems subsequently affected the first period of continuous BP recordings after postural change and thereby the prevalence of IOH and the overall curves. Finally, the visual inspection of the graphics was performed by one author.

CONCLUSION

A clear difference was seen in blood pressure response between the two postural changes. Standing resulted in a greater SBP decrease compared to sitting, whereas the opposite was observed for DBP. Although no difference in prevalence of OH was observed, the positive and negative proportion of agreement of the prevalence of OH were poor to moderate which indicates relevant differences in the diagnosis of OH depending on the method of OH measurement. It is advisable to perform OH measurements only in accordance with the consensus statement to standing position.

R1

R2

R3

R4

R5

R6

R7

R8

R9

R10

R11

R12

R13

R14

R15

R16

R17

R18

R19

R20

R21

R22

R23

R24

R25

R26

R27

R28

R29

R30

R31

R32

R33

R34

R35

R36

R37

R38

R39


37

2

REFERENCES


2. Gupta V, Lipsitz LA. Orthostatic hypotension in the elderly: diagnosis and treatment. The American journal of medicine. 2007;120 (10)(1555-7162 (Electronic)):841-847.

3. Guyton AC HJ. Textbook of Medical Physiology. 11th ed. Philadelphia, Pennsylvania: Elsevier Saunders; 2006.

4. Schatz IJ. Orthostatic hypotension. I. Functional and neurogenic causes. 19840504 DCOM- 19840504(0003-9926 (Print)).

5. Thijs RD, Kamper Am Fau - van Dijk AD, van Dijk Ad Fau - van Dijk JG, van Dijk JG. Are the orthostatic fluid shifts to the calves augmented in autonomic failure? Clinical autonomic research : official journal of the Clinical Autonomic Research Society. 2010;20 (1)(1619-1560 (Electronic)):19-25.

6. Davis Br Fau - Langford HG, Langford Hg Fau - Blaufox MD, Blaufox Md Fau - Curb JD, et al. The association of postural changes in systolic blood pressure and mortality in persons with hypertension: the Hypertension Detection and Follow-up Program experience. Circulation. 1987;75(2)(0009-7322 (Print)):340-346.

7. Masaki KH, Schatz Ij Fau - Burchfiel CM, Burchfiel Cm Fau - Sharp DS, et al. Orthostatic hypotension predicts mortality in elderly men: the Honolulu Heart Program. Circulation. 1998;98(21):2290-2295.

8. Alagiakrishnan K, Patel K Fau - Desai RV, Desai Rv Fau - Ahmed MB, et al. Orthostatic hypotension and incident heart failure in community-dwelling older adults. The journals of gerontology. Series A, Biological sciences and medical sciences. 2014;69(2):223-230.

9. McJunkin B, Rose B, Amin O, et al. Detecting initial orthostatic hypotension: a novel approach. Journal of the American Society of Hypertension : JASH. 2015;9(5):365-369.

10. Hartog LC, Cizmar-Sweelssen M, Knipscheer A, et al. The association between orthostatic hypotension, falling and successful rehabilitation in a nursing home population. Archives of gerontology and geriatrics. 2015;61(2):190-196.



13. Gorelik O, Cohen N. Seated postural hypotension. Journal of the American Society of Hypertension : JASH. 2015;9(12):985-992.

14. Imholz BP, Settels Jj Fau - van der Meiracker AH, van der Meiracker Ah Fau - Wesseling KH, Wesseling Kh Fau - Wieling W, Wieling W. Non-invasive continuous finger blood pressure measurement during orthostatic stress compared to intra-arterial pressure. Cardiovascular research. 1990;24(3):214-221.

15. Schutte AE, Huisman Hw Fau - van Rooyen JM, van Rooyen Jm Fau - Malan NT, Malan Nt Fau - Schutte R, Schutte R. Validation of the Finometer device for measurement of blood pressure in black women. Journal of human hypertension. 2004;18(2):79-84.

16. KH W. FinometerTM User’s Guide. 1.10th ed: Hasselt: Pragma Ade; 2002.

17. Netea RT, Lenders Jw Fau - Smits P, Smits P Fau - Thien T, Thien T. Influence of body and arm position on blood pressure readings: an overview. Journal of hypertension. 2003;21(2):237-241.

18. Wesseling KH, Jansen Jr Fau - Settels JJ, Settels Jj Fau - Schreuder JJ, Schreuder JJ. Computation of aortic flow from pressure in humans using a nonlinear, three-element model. Journal of applied physiology (Bethesda, Md. : 1985). 1993;74(5):2566-2573.

R1

R2

R3

R4

R5

R6

R7

R8

R9

R10

R11

R12

R13

R14

R15

R16

R17

R18

R19

R20

R21

R22

R23

R24

R25

R26

R27

R28

R29

R30

R31

R32

R33

R34

R35

R36

R37

R38

R39

Chapter 2

38

19. Consensus statement on the definition of orthostatic hypotension, pure autonomic failure, and multiple system atrophy. The Consensus Committee of the American Autonomic Society and the American Academy of Neurology. Neurology. 1996 May;46(5):1470.

20. Wieling W, Krediet Ct Fau - van Dijk N, van Dijk N Fau - Linzer M, Linzer M Fau - Tschakovsky ME, Tschakovsky ME. Initial orthostatic hypotension: review of a forgotten condition. Clinical science (London, England : 1979). 2007;112(3):157-165.

21. White WB, Lund-Johansen P Fau - Weiss S, Weiss S Fau - Omvik P, Omvik P Fau - Indurkhya N, Indurkhya N. The relationships between casual and ambulatory blood pressure measurements and central hemodynamics in essential human hypertension. Journal of hypertension. 1994;12(9):1075-1081.

22. de Vet HC, Mokkink Lb Fau - Terwee CB, Terwee Cb Fau - Hoekstra OS, Hoekstra Os Fau - Knol DL, Knol DL. Clinicians are right not to like Cohen’s kappa. BMJ. 2013;12;346:f2125(1756-1833 (Electronic)).

23. von Elm E, Altman DG, Egger M, et al. The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) Statement: Guidelines for reporting observational studies. International journal of surgery. 2014;12(1743-9159 (Electronic)):1495-1499.

24. Medow MS, Stewart Jm Fau - Sanyal S, Sanyal S Fau - Mumtaz A, et al. Pathophysiology, diagnosis, and treatment of orthostatic hypotension and vasovagal syncope. Cardiology in review. 2008;16(1):4-20.

25. Perlmuter LC, Sarda G Fau - Casavant V, Casavant V Fau - Mosnaim AD, Mosnaim AD. A review of the etiology, associated comorbidities, and treatment of orthostatic hypotension. American journal of therapeutics. 2013;20(3):279-291.

26. Smith JJ, Porth Cm Fau - Erickson M, Erickson M. Hemodynamic response to the upright posture. J Clin Pharmacol. 1994;34(5):375-386.

27. Kozakova M, Palombo C. Diabetes Mellitus, Arterial Wall, and Cardiovascular Risk Assessment. International journal of environmental research and public health. 2016;13(2):201.

28. Mitchell GF. Arterial stiffness and hypertension: chicken or egg? Hypertension. 2014;64(2):210-214.

29. Takahashi M, Miyai N, Nagano S, et al. Orthostatic Blood Pressure Changes and Subclinical Markers of Atherosclerosis. American journal of hypertension. 2015;28(9):1134-1140.

30. Tikkanen O, Haakana P, Pesola AJ, et al. Muscle activity and inactivity periods during normal daily life. PloS one. 2013;8(1):e52228.

31. Rickards CA, Newman DG. A comparative assessment of two techniques for investigating initial cardiovascular reflexes under acute orthostatic stress. European journal of applied physiology. 2003 Nov;90(5-6):449-57.

32. Medow MS, Stewart JM. The postural tachycardia syndrome. Cardiology in review. 2007;15(2):67-75.


34. Detry MA, Ma Y. Analyzing Repeated Measurements Using Mixed Models. Jama. 2016;315(4):407-408.

35. Azabji Kenfack M, Lador F Fau - Licker M, Licker M Fau - Moia C, et al. Cardiac output by Modelflow method from intra-arterial and fingertip pulse pressure profiles. Clinical science (London, England : 1979). 2004;106(4):365-369.

R1

R2

R3

R4

R5

R6

R7

R8

R9

R10

R11

R12

R13

R14

R15

R16

R17

R18

R19

R20

R21

R22

R23

R24

R25

R26

R27

R28

R29

R30

R31

R32

R33

R34

R35

R36

R37

R38

R39


39

2

SUPP

LEM

ENTA

L M

ATER

IAL

Tabl

e A

1. D

iffer

ence

s in

bas

elin

e ch

arac

teri

stics

div

ided

by

the

side

of t

he B

P m

easu

rem

ents

and

the

first

pos

tura

l cha

nge.

Fino

met

er o

n th

e le

ft a

rmFi

nom

eter

on

the

righ

t ar

mFi

rst p

ostu

ral c

hang

e to

si

tting

pos

ition

Firs

t pos

tura

l cha

nge

to

stan

ding

pos

ition

Char

acte

risti

csA

ge (y

ears

)69

.0 (6

4.3-

74.0

)*69

(60.

3-76

.8)*

69.0

(62.

0-75

.8)*

69.5

(64.

3-73

.8)*

Fem

ale

25 (4

8)20

(39)

24 (4

6)21

(40)

BMI (

kg/m

2)

26.6

(23.

8-32

.0)*

27.3

(24.

7-31

.6)*

26.9

(24.

3-30

.9)*

27.4

(25.

4-32

.4)*

Dri

nkin

g an

d/or

eati

ng

33 (6

4)30

(58)

35 (6

7)28

(54)

Med

ical

His

tory

Hyp

erte

nsio

n 42

(80)

40 (7

7)37

(71)

45 (8

7)D

M

31 (6

0)33

(64)

29 (4

5)35

(67)

His

tory

of C

VD18

(35)

20 (3

9)18

(35)

20 (3

9)Po

lyne

urop

athy

17 (3

3)16

(31)

16 (3

1)17

(32)

Pace

mak

er im

plan

tatio

n4

(8)

4 (8

)3

(6)

5 (1

0)H

isto

ry o

f fal

ls18

(35)

13 (2

5)17

(33)

14 (2

7)M

edic

ation

Anti

-hyp

erte

nsiv

e m

edic

ation

44

(85)

42 (8

1)39

(75)

47 (9

0)A

nti-a

rrhy

thm

ic m

edic

ation

1 (2

)5

(10)

1 (2

)5

(10)

Nit

ra

te

s6

(12)

7 (1

4)4

(8)

9 (1

7)O

ral g

luco

se lo

wer

ing

ther

apy

14 (2

7)18

(35)

11 (2

1)21

(40)

Insu

lin

25 (4

8)27

(52)

23 (4

4)29

(56)

Psyc

hiat

ric

med

icati

on

5 (1

0)8

(15)

7 (1

4)6

(12)

Mea

sure

men

tsBa

selin

e SB

P (m

mH

g)15

9.6

(139

.3-1

67.1

)*14

6.8

(135

.3-1

62.0

)*14

9.2

(137

.5-1

64.6

)*15

1.0

(136

.3-1

62.7

)*Ba

selin

e D

BP (m

mH

g)77

.5 (7

0.9-

80.2

)*76

.7 (7

2.4-

83.5

)*77

.4 (7

1.7-

81.0

)*76

.0 (7

1.5-

82.6

)*Ba

selin

e H

R (b

eats

/min

)67

.6 (6

0.3-

74.3

)*67

.9 (6

1.3-

74.0

)*64

.9 (5

7.0-

72.8

)*70

.4 (6

2.9-

75.7

)*Ba

selin

e CO

(L/m

in)

6.3

(4.6

-7.2

)*6.

6 (5

.0-8

.1)*

6.2

(4.7

-7.0

)*6.

7 (4

.9-8

.4)*

Valu

es a

re p

rese

nted

as

n (%

), un

less

indi

cate

d ot

herw

ise.

*: M

edia

n (p

25-7

5).

CHAPTER 3Diagnosing orthostatic hypotension with continuous

and interval blood pressure measurement devices

Submitted as: Breeuwsma AC, Hartog LC, Kamper AM, Groenier KH, Bilo HJG, Kleefstra N,

Van Hateren KJJ. Diagnosing orthostatic hypotension with continuous and

interval blood pressure measurement devices.

R1

R2

R3

R4

R5

R6

R7

R8

R9

R10

R11

R12

R13

R14

R15

R16

R17

R18

R19

R20

R21

R22

R23

R24

R25

R26

R27

R28

R29

R30

R31

R32

R33

R34

R35

R36

R37

R38

R39

Chapter 3

42

ABSTRACT

ac ground: Orthostatic hypotension (OH) is defined as a drop in systolic blood pressure (SBP) of ≥20 mmHg and/or a drop in diastolic blood pressure (DBP) of ≥10 mm Hg within 3 minutes of standing. The international guidelines recommend diagnosing OH with a continuous blood pressure (BP) measurement device, although in daily practice interval BP measurement devices are used more often. We aimed to investigate the difference in observed prevalence of OH between an interval and a continuous BP measurement device.

Methods: A cross-sectional observational study. BP was measured with both an interval and a continuous BP measurement device during postural change from supine to the standing position. The differences in prevalence were tested with the McNemar test. Positive and negative proportions of agreement were calculated to observe the agreement of diagnosing OH between the two devices.

esults: A total of 104 patients with a mean age of 69 years were included. The prevalence of OH was 35.6% (95% CI: 26.4-44.8) with the interval BP measurement and 45.2% (95% CI: 35.6-54.8) with the continuous BP measurement device (p-value for the difference = .121). The positive proportion of agreement was 59.5% and the negative proportion of agreement was 72.5%.

onclusions: Although the prevalence of OH was not significantly different between the continuous and the interval BP measurement devices, the positive and negative proportions of agreement were low. We conclude that continuous BP measurement cannot be substituted by an interval BP measurement to diagnose OH.

R1

R2

R3

R4

R5

R6

R7

R8

R9

R10

R11

R12

R13

R14

R15

R16

R17

R18

R19

R20

R21

R22

R23

R24

R25

R26

R27

R28

R29

R30

R31

R32

R33

R34

R35

R36

R37

R38

R39

iagnosing orthostatic h potension ith continuous and interval blood pressure

43

3

INTRODUCTION

Orthostatic hypotension (OH) is a clinical condition frequently affecting the elderly population and its prevalence rises with age [1, 2]. The prevalence varies from 7% to 55% in the elderly population [2-6].

OH is defined as a drop in systolic blood pressure (SBP) of at least 20 mmHg and/or a drop in diastolic blood pressure (DBP) of at least 10 mmHg within 3 minutes (min) of standing after 5 min of rest in supine position [7, 8]. Guidelines recommend to diagnose OH with continuous instead of interval blood pressure (BP) measurement devices [8], although the study of Romero-Ortuno suggested a lack of specificity for diagnosing OH [9]. On the contrary, automated sphygmomanometers are commonly used for this purpose in daily practice, but they underreport OH, compared to continuous measurement, due to the delay in time [10, 11]. Due to these different results, concerns are raised against the threshold in the diagnostic criteria and it is hard to apply the criteria on different BP measurement devices in clinical practice.

This study aimed to investigate the difference in the prevalence of OH when OH is measured using two BP measurement devices (continuous versus interval) during postural change from supine to standing position in patients of at least 50 years, and to investigate to what extend the results of these methods agree with each other. We hypothesized that the prevalence of OH with the continuous BP measurements would be higher compared to the interval BP measurements.

MATERIAL AND METHODS

tud populationFor this cross-sectional study, the inclusion and study period was from January to February 2016. Patients > 50 years of age combined with a medical history of cardiovascular disease (CVD), diabetes mellitus (DM), and/or hypertension with an appointment at the outpatient clinic of internal medicine (Isala hospital, Zwolle, the Netherlands) were included. Patients with a contraindication of BP measurements in one arm, inability to stand without assistance, known peripheral vessel disease in one or both arms, needing a large (≥42 cm) or small (≤28 cm) cuff, and incapability of giving consent were excluded. A non-blinded randomization procedure was performed for the side of the specific BP measurement device (Figure 1).

R1

R2

R3

R4

R5

R6

R7

R8

R9

R10

R11

R12

R13

R14

R15

R16

R17

R18

R19

R20

R21

R22

R23

R24

R25

R26

R27

R28

R29

R30

R31

R32

R33

R34

R35

R36

R37

R38

R39

Chapter 3

44

104 patients

52 patientsContinuous lefta

Interval rightb

52 patientsContinuous rightc

Interval leftd

Figure 1. Randomization. a: Continuous BP measurement device on the left arm. b: Interval BP measurement device on the right arm. c: Continuous BP measurement device on the right arm. d: Interval BP measurement device on the left arm.

ata collectionDemographic characteristics, a full medical history including a history of CVD, DM, hypertension, polyneuropathy, Parkinson’s disease, pacemaker implantation, falls in the previous year, and medication use were collected. History of CVD was defined as a history of angina pectoris, myocardial infarction, percutaneous transluminal coronary angioplasty, coronary artery bypass grafting, stroke, and/or transient ischemic attack. The BP measurement procedure was performed by a trained medical student. BP was measured using an interval and continuous BP measurement device simultaneously during postural change from supine to standing position. Interval BP was measured with an automated sphygmomanometer, the validated A&D UA-767 plus [12]. Interval BP was measured twice in supine position after 5 minutes of rest, and twice at 1 min and twice at 3 min after postural change to standing position, resulting in a total of 6 measurements. Continuous BP was measured with the Finometer Pro (Finapres Medical Systems BV), a continuous non-invasive beat-to-beat BP measurement device, which has been validated compared to invasive BP recordings [12, 13]. Finger circumference was measured to apply the proper sized finger cuff of the Finometer Pro [14]. Height differences were corrected by a height nulling procedure, by supporting both arms at heart level in supine and standing position, and by repeatedly checking the position of both arms [14, 15]. By using the return-to-flow (RTF) calibration system, the measured BP in the finger was reconstructed to the upper arm BP [14]. The presence or absence of characteristic symptoms of OH during postural change (dizziness, light-headedness, and blurred vision) was reported and defined as orthostatic complaints. BP measurement data of the Finometer Pro were exported with the BeatScope software (Finapres Medical Systems BV). Baseline mean supine SBP and DBP were calculated over the last minute prior to postural change. After postural change, lowest SBP and DBP were calculated for several timeframes (15-44s, 45-74s, 75-104s, 105-134s, 135-164s, 165-194s, and 195-224s). The first 14 s of the measurements were excluded. To

R1

R2

R3

R4

R5

R6

R7

R8

R9

R10

R11

R12

R13

R14

R15

R16

R17

R18

R19

R20

R21

R22

R23

R24

R25

R26

R27

R28

R29

R30

R31

R32

R33

R34

R35

R36

R37

R38

R39


45

3

compare the prevalence of OH between the interval and the continuous BP measurement device, only the data of the continuous BP measurement device from the four timeframes around 1 and 3 minutes after postural change (45-74s, 75-104s, 165-194s, and 195-224s) were used. Records with poor quality signals (e.g. artefacts) were excluded by visual inspection of the graphics in the BeatScope output files. OH was defined as a drop in SBP of at least 20 mmHg or a drop in DBP of at least 10 mmHg within 3 min after postural change [7].

tatistical anal sisMean and standard deviation (SD) were used to present normally distributed continuous variables, and median and interquartile range (IQR) were used for non-normally distributed continuous variables. Proportions were used to present categorical variables. Normality tests were performed by inspection of the Q-Q plots and histograms. The difference in prevalence of OH according to the BP measurement device was analysed with a two-sided McNemar test. The positive and the negative proportions of agreement were calculated [16]. The positive proportion of agreement is the number of both postural changes that diagnosed OH divided by the total number of OH diagnosed for each of the postural changes. The negative proportion of agreement is the number of both postural changes that excluded OH divided by the total number of excluded OH for each of the postural changes. Both positive and negative proportions of agreement were reported as percentages.P-values below .05 were considered to be statistically significant. Statistical analyses were performed using the SPSS software (version 23; IBM, Armonk, New York, USA).

thical approval and clinical trial registrationThe present study was registered at www.trialregister.nl (NTR5525) and was approved by the medical ethics committee (number 15.06.95). This study was performed in accordance with the declaration of Helsinki. Written informed consent was obtained during the scheduled medical appointment. The data was recorded and analysed anonymously. The ‘Strengthening the Reporting of Observational Studies in Epidemiology’ (STROBE) statement was used to describe this observational cohort study [17].

RESULTS

In the present study, 104 patients (59 men, 45 women) were included (Figure 2). Baseline characteristics of the study population are presented in Table 1. Baseline characteristics of patients with the Finometer on the left arm and patients with the Finometer on the right arm were presented in Appendix Table A1.

R1

R2

R3

R4

R5

R6

R7

R8

R9

R10

R11

R12

R13

R14

R15

R16

R17

R18

R19

R20

R21

R22

R23

R24

R25

R26

R27

R28

R29

R30

R31

R32

R33

R34

R35

R36

R37

R38

R39

Chapter 3

46

402 patients

127 patients

104 patients

275 excluded by their specialist or unwillingness to

participate

23 excluded due to measurement problems

Figure 2. Patient selection.

No significant difference in the prevalence of OH was observed between the interval BP and continuous BP measurement devices. The prevalence of OH with the interval BP measurement device was 35.6% (95% CI: 26.4-44.8) compared to 45.2% (95% CI: 35.6-54.8) when measured with the continuous BP measurement device. The positive percentage of agreement was 59.5% and the negative percentage of agreement was 72.5%. Orthostatic complaints were reported by 22.1% (95% CI: 14.1-30.1) of the study population.

The prevalence of OH increased to 66.3% (95% CI: 57.2-75.4) when all seven timeframes of the continuous BP measurement device were included, which is significantly higher compared to the prevalence with the interval BP measurement device (p-value <.001). In this case the positive percentage of agreement was 52.9% and the negative percentage of agreement was 54.7%.

R1

R2

R3

R4

R5

R6

R7

R8

R9

R10

R11

R12

R13

R14

R15

R16

R17

R18

R19

R20

R21

R22

R23

R24

R25

R26

R27

R28

R29

R30

R31

R32

R33

R34

R35

R36

R37

R38

R39


47

3


Patients (n=104)CharacteristicsAge (years) 68.8 (8.5)a

Female 45 (43)BMI (kg/m2) 27.0 (24.7-31.7)b

Drinking and/or eating before measurements 63 (61)Medical HistoryHypertension 82 (79)DM 64 (62)History of CVD 38 (37)Polyneuropathy 33 (32)Parkinson’s disease 0 (0)Pacemaker implantation 8 (8)History of falls 31 (30)MedicationAnti-hypertensive medication 86 (83)Anti-arrhythmic medication 6 (6)Nitrates 13 (13)Oral glucose lowering therapy 32 (31)Insulin 52 (50)Psychiatric medication 13 (13)Anti-parkinsonian medication 1 (1)Measurements continuous BPLying SBP (mmHg) 150.3 (137.3-162.9)b

Lying DBP (mmHg) 76.7 (9.2)a

Lying HR (beats/min) 67.6 (60.4-73.8)b

Lying CO (L/min) 6.21 (1.96)a

Measurements interval BPLying SBP (mmHg) 132.3 (124.0-152.8)b

Lying DBP (mmHg) 75.8 (67.1-82.4)b

Lying HR (beats/min) 65.5 (9.4)a

Values are presented as n (%), unless indicated otherwise. a: Mean (±SD). b: Median (IQR). c: % [95% Confidence Interval (CI)].

DISCUSSION

No significant difference in the prevalence of OH was observed between the continuous and the interval BP measurement device, 45.2% versus 35.6% respectively. The percentage of positive proportion of agreement was 59.5% and the percentage of negative proportion of agreement was 72.5%, which is considered to be poor.When all timeframes of the continuous BP measurement device were included, it lead to a 21% absolute increase in the prevalence of OH compared to the interval BP measurement device and even lower proportions of agreement.

R1

R2

R3

R4

R5

R6

R7

R8

R9

R10

R11

R12

R13

R14

R15

R16

R17

R18

R19

R20

R21

R22

R23

R24

R25

R26

R27

R28

R29

R30

R31

R32

R33

R34

R35

R36

R37

R38

R39

Chapter 3

48

Previous studies reported a prevalence of OH ranging from 6% to 18% in community-dwelling elderly and 37% to 50% in nursing homes, as measured with interval BP measurement devices [4, 11, 18-21]. The observed prevalence of OH with the automated sphygmomanometer in the present study lies within the range of these studies. On the other hand, Romero-Ortuno presented a prevalence of OH of 94% in community-dwelling elderly measured with a continuous BP measurement device [9] which is twice the prevalence described in the present study. Concerns about using continuous BP measurement devices for diagnosing OH are previously reported [9, 22, 23]. A lack of specificity is suggested and the clinical interpretation and relevance is questioned [9, 22, 23]. In response to the increasing use of these devices the updated consensus added the word ‘sustained’ to the criteria in 2011 [24]. However, the duration of sustained is not defined [25], which makes it difficult to apply these criteria. Therefore, diagnosing OH with a continuous BP measurement device combined with the current diagnostic criteria might not be reliable and could lead to diagnose clinically irrelevant OH.

The present study observed no significant difference in the prevalence of OH between the continuous and the interval BP measurement device when the same amount of measurements was compared. However, in clinical practice the continuous BP measurement device provides more than four measurements and the present study observed a significant higher prevalence of OH with the continuous BP measurement device when all seven timeframes were used. This implies that clinicians, who use the continuous BP measurement device, would diagnose OH more often than clinicians who use the interval BP measurement device. Due to the lack of a golden standard, no sensitivity and specificity tests were performed. However, the positive and negative proportions of agreement were poor, 59.5% and 72.5%, respectively. A positive proportion of agreement of 59.5% indicates that only a little more than half of the subjects diagnosed with OH, are diagnosed with OH with both devices.

trengths and li itationsThe present study had several strengths. All measurements were performed and evaluated by the same individual to overcome inter-observer bias. All patients were non-blinded randomized for both the sequence of the postural changes and the side of the BP measurements. Generalizability is limited to elderly patients visiting the internal outpatient clinic. Due to the fact that the patients included in this study had to be able to stand for five minutes without assistance, the study group was slightly biased compared to the population visiting the internal outpatient clinic, and the results are, of course, only useful in patients who are able to stand. Other limitations are the small study sample and the fact that the curves in BeatScope were judged by only one individual.

R1

R2

R3

R4

R5

R6

R7

R8

R9

R10

R11

R12

R13

R14

R15

R16

R17

R18

R19

R20

R21

R22

R23

R24

R25

R26

R27

R28

R29

R30

R31

R32

R33

R34

R35

R36

R37

R38

R39


49

3

CONCLUSION

Although no significant difference in the prevalence of OH was observed between the continuous and the interval BP measurement devices, the positive and negative proportions of agreement were low. We conclude that continuous and interval BP measurements cannot be replaced by each other for establishing OH.

R1

R2

R3

R4

R5

R6

R7

R8

R9

R10

R11

R12

R13

R14

R15

R16

R17

R18

R19

R20

R21

R22

R23

R24

R25

R26

R27

R28

R29

R30

R31

R32

R33

R34

R35

R36

R37

R38

R39

Chapter 3

50

REFERENCES

1. Davis Br Fau - Langford HG, Langford Hg Fau - Blaufox MD, Blaufox Md Fau - Curb JD, et al. The association of postural changes in systolic blood pressure and mortality in persons with hypertension: the Hypertension Detection and Follow-up Program experience. Circulation. 1987;75(2)(0009-7322 (Print)):340-346.

2. Masaki KH, Schatz Ij Fau - Burchfiel CM, Burchfiel Cm Fau - Sharp DS, et al. Orthostatic hypotension predicts mortality in elderly men: the Honolulu Heart Program. Circulation. 1998;98(21):2290-2295.

3. Aronow WS, Lee Nh Fau - Sales FF, Sales Ff Fau - Etienne F, Etienne F. Prevalence of postural hypotension in elderly patients in a long-term health care facility. A J Cardiol. 1988; 62 (4): 336.

4. Hartog LC, Cizmar-Sweelssen M, Knipscheer A, et al. The association between orthostatic hypotension, falling and successful rehabilitation in a nursing home population. Archives of gerontology and geriatrics. 2015;61(2):190-196.


6. Poon IO, Braun U. High prevalence of orthostatic hypotension and its correlation with potentially causative medications among elderly veterans. J Clin Pharm Ther. 2005; 30(2):173-8

7. Consensus statement on the definition of orthostatic hypotension, pure autonomic failure, and multiple system atrophy. The Consensus Committee of the American Autonomic Society and the American Academy of Neurology. Neurology. 1996 May;46(5):1470.



10. Caine Se Fau - Alsop K, Alsop K Fau - Mac Mahon M, Mac Mahon M. Overlooking orthostatic hypotension with routine blood-pressure equipment. Lancet 1998 8;352 (9126):458


12. Rogoza AN, Pavlova Ts Fau - Sergeeva MV, Sergeeva MV. Validation of A&D UA-767 device for the self-measurement of blood pressure. Blood press monit. 2000; 5(4): 227-31

13. Imholz BP, Settels Jj Fau - van der Meiracker AH, van der Meiracker Ah Fau - Wesseling KH, Wesseling Kh Fau - Wieling W, Wieling W. Non-invasive continuous finger blood pressure measurement during orthostatic stress compared to intra-arterial pressure. Cardiovascular research. 1990;24(3):214-221.

14. KH W. FinometerTM User’s Guide. 1.10th ed: Hasselt: Pragma Ade; 2002.

15. Netea RT, Lenders Jw Fau - Smits P, Smits P Fau - Thien T, Thien T. Influence of body and arm position on blood pressure readings: an overview. Journal of hypertension. 2003;21(2):237-241.

16. de Vet HC, Mokkink Lb Fau - Terwee CB, Terwee Cb Fau - Hoekstra OS, Hoekstra Os Fau - Knol DL, Knol DL. Clinicians are right not to like Cohen’s kappa. BMJ. 2013;12;346:f2125(1756-1833 (Electronic)).

17. von Elm E, Altman DG, Egger M, et al. The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) Statement: Guidelines for reporting observational studies. International journal of surgery. 2014;12(1743-9159 (Electronic)):1495-1499.

18. Alagiakrishnan K, Patel K Fau - Desai RV, Desai Rv Fau - Ahmed MB, et al. Orthostatic hypotension and incident heart failure in community-dwelling older adults. The journals of gerontology. Series A, Biological sciences and medical sciences. 2014;69(2):223-230.

19. McJunkin B, Rose B, Amin O, et al. Detecting initial orthostatic hypotension: a novel approach. Journal of the American Society of Hypertension : JASH. 2015;9(5):365-369.

R1

R2

R3

R4

R5

R6

R7

R8

R9

R10

R11

R12

R13

R14

R15

R16

R17

R18

R19

R20

R21

R22

R23

R24

R25

R26

R27

R28

R29

R30

R31

R32

R33

R34

R35

R36

R37

R38

R39


51

3


21. O’Connell MD, Savva GM, Fan CW, Kenny RA. Orthostatic hypotension, orthostatic intolerance and frailty: The Irish Longitudinal Study on Aging-TILDA. Arch Gerontol Geriatr 2015; 60(3):507-13

22. Wieling W, Schatz IJ. The consensus statement on the definition of orthostatic hypotension: a revisit after 13 years. Journal of hypertension. 2009 May;27(5):935-8.

23. Braam Ea Fau - Verbakel D, Verbakel D Fau - Adiyaman A, Adiyaman A Fau - Thien T, Thien T. Orthostatic hypotension: revision of the definition is needed. J Hypertens 2009; 27(10): 2119-20


25. Frith J. Diagnosing orthostatic hypotension: a narrative review of the evidence. Br Med Bull. 2015; 115 (1):123-34

R1

R2

R3

R4

R5

R6

R7

R8

R9

R10

R11

R12

R13

R14

R15

R16

R17

R18

R19

R20

R21

R22

R23

R24

R25

R26

R27

R28

R29

R30

R31

R32

R33

R34

R35

R36

R37

R38

R39

Chapter 3

52

SUPPLEMENTAL DATA

Appendix Table A1. Differences in baseline characteristics divided by the side of the continuous BP measurement device.

Finometer on the left arm

Finometer on the right arm

CharacteristicsAge (years) 69.0 (64.3-74.0)a 69.0 (60.3-76.8)a

Female 25 (48) 20 (39)BMI (kg/m2) 26.6 (23.8-32.0)a 27.3 (24.7-31.6)a

Drinking and/or eating 33 (64) 30 (58)Medical HistoryHypertension 42 (80) 40 (77)DM 31 (60) 33 (64)History of CVD 18 (35) 20 (39)Polyneuropathy 17 (33) 16 (31)Pacemaker implantation 4 (8) 4 (8)History of falls 18 (35) 13 (25)MedicationAnti-hypertensive medication 44 (85) 42 (81)Anti-arrhythmic medication 1 (2) 5 (10)Nitrates 6 (12) 7 (14)Oral glucose lowering therapy 14 (27) 18 (35)Insulin 25 (48) 27 (52)Psychiatric medication 5 (10) 8 (15)Baseline SBP (mmHg) 136.5 (125.0-159.0) 132.0 (120.1-153.4)Baseline DBP (mmHg) 74.3 (67.5-83.0) 77.0 (67.0-84.5)Baseline HR (beats/min) 65.5 (58.0-70.5) 63.5 (58.1-71.4)

Values are presented as n (%), unless indicated otherwise. a: Median (p25-75).

CHAPTER 4Orthostatic hypotension does not predict recurrent

falling in a nursing home population

Published as: Hartog LC, Cimzar-Sweelssen M, Knipscheer A, Groenier KH, Kleefstra N, Bilo

HJG, Van Hateren KJJ. Orthostatic hypotension does not predict recurrent

falling in a nursing home population. Archives of Gerontology and Geriatrics

2016 Sep 4; 68: 39-43.

R1

R2

R3

R4

R5

R6

R7

R8

R9

R10

R11

R12

R13

R14

R15

R16

R17

R18

R19

R20

R21

R22

R23

R24

R25

R26

R27

R28

R29

R30

R31

R32

R33

R34

R35

R36

R37

R38

R39

Chapter 4

54

ABSTRACT

b ective: Most studies regard orthostatic hypotension (OH) as a causal factor for falls. However, the evidence is lacking for this assumption. We aimed to investigate the relationship between orthostatic hypotension and fall incidents in nursing home residents.

Methods: A total of 249 patients was included in a prospective observational cohort study of nursing home residents. Falls were prospectively registered. Cox proportional hazard modelling and the conditional frailty model were used to analyse the relationship between OH and (recurrent) falling.

esults: Among the 249 patients, 450 falls were recorded during follow-up and OH was present in 93 out of 249 patients. No significant associations were found between OH and the first fall incident (Hazard Ratio (HR) 1.01 (95% Confidence Interval (CI) 0.60-1.69) and recurrent falling (HR 1.21 (95%CI 0.65-2.24)).

onclusions: Although falling and OH were both highly prevalent in nursing home residents, no relationship between OH and falling was found.

R1

R2

R3

R4

R5

R6

R7

R8

R9

R10

R11

R12

R13

R14

R15

R16

R17

R18

R19

R20

R21

R22

R23

R24

R25

R26

R27

R28

R29

R30

R31

R32

R33

R34

R35

R36

R37

R38

R39

rthostatic h potension does not predict recurrent alling in a nursing ho e population

55

4

INTRODUCTION

Orthostatic hypotension (OH) is frequently seen in the elderly population, and its prevalence increases with advancing age [1]. OH is regarded as a risk factor for falling and could potentially lead to severe morbidity [2-7]. Many studies reporting a positive association between falls and OH, were performed in nursing homes [5, 8]. However, the lack of prospective fall data [9-11], adjusting for important confounders [7], or using the international consensus definition of OH [5], limit the generalizability of these studies.Besides, various studies only described the relation between OH and falling in subgroups instead of within the total study population [8, 12, 13].Two recent systematic reviews described the relationship between OH and falling in elderly. However, the absolute attributive risk was not established due to the lack of a meta-analysis [3, 4]. Although OH is widely accepted as an important risk factor for falls, it remains to be established whether and to what extent OH contributes to falling [4, 12]. Despite the lack of evidence for a causal relationship between OH and fall risk, a variety of interventions are currently being investigated and already deployed that target OH, such as discontinuation of antihypertensive medication [14]. Because in particular nursing home residents are frail elderly, who are most likely to have serious injuries after falling, we aimed to prospectively investigate the relationship between orthostatic hypotension and fall incidents in nursing home residents.

MATERIALS AND METHODS

tud populationFor this prospective observational cohort study, patients were recruited from a nursing home facility in the north-eastern region of the Netherlands (TriviumMeulenbeltZorg, Hengelo). The design and details of the study population have been presented elsewhere [15]. For this study, all patients had to be at risk for falling; therefore bedridden patients were excluded.

ata collectionBaseline data included demographic characteristics, a full medical history including a history of cardiovascular disease, diabetes mellitus, hypertension, fall history, and medication use. During admission, every individual fall was registered with date and type of injury caused by falling. A fall was defined as an unintentionally coming to the ground. In case of a recurrent faller, only information about the first 5 fall incidents was collected.

R1

R2

R3

R4

R5

R6

R7

R8

R9

R10

R11

R12

R13

R14

R15

R16

R17

R18

R19

R20

R21

R22

R23

R24

R25

R26

R27

R28

R29

R30

R31

R32

R33

R34

R35

R36

R37

R38

R39

Chapter 4

56

Patients were considered to have cardiovascular disease when they had a history of angina pectoris, myocardial infarction, percutaneous transluminal coronary angioplasty, coronary artery bypass grafting, stroke or transient ischaemic attack. The activities of daily living were measured with the Barthel-Index [16]; the score ranges from 0-20. Blood pressure was measured following a standardized protocol, using an automated sphygmomanometer (Omron M6) [17]. If the automated sphygmomanometer displayed an error message, blood pressure was manually measured with a sphygmomanometer Heine Gamma XXL-T [18].Trained medical staff performed all tests. Blood pressure was measured twice in supine position after 5 minutes of rest, and twice each at 1 and 3 minutes after postural change. The forearm of the patient was supported at heart level during the measurements in upright position [19]. The postural change was from supine to standing position, with the exception of patients who were unable to stand during the orthostatic blood pressure measurement. For these patients the postural change was from lying to sitting position. All patients were asked whether they had consumed a meal or drink within two and one hours, respectively, prior to the measurements. Besides, the time of the day of the blood pressure measurement was registered. Mean supine blood pressure was compared to the four individual sitting or standing blood pressure measurements to diagnose OH. OH was defined as a drop in SBP of > 20 mmHg or diastolic blood pressure (DBP) of > 10 mmHg after postural change compared to the mean value of the baseline measurements in supine position within 3 minutes [1]. The presence or absence of characteristic symptoms of OH like light-headedness, syncope, or dizziness after postural change was scored, and the combination of OH and orthostatic complaints was described as symptomatic OH.

tatistical anal ses Continuous variables are presented as mean and standard deviation for normally distributed variables, or as median and interquartile range for non-normally distributed variables. Cox proportional hazard modelling was used to investigate the relationship between OH and first fall incidents. In the multivariate Cox regression models we adjusted for age, gender, body mass index (BMI), history of hypertension, previous fall (within last 12 months), the score on the Barthel questionnaire, type of department, and number of medications. The Schoenfeld residual plots were inspected for each predictor variable to check the assumption of proportional hazards. P values less than 0.05 were considered statistically significant. To analyse the relationship between OH and recurrent fall incidents, the conditional frailty model was used [20, 21]. The conditional frailty model was based on the Cox proportional hazards model. By using the conditional frailty model for analysing recurrent events, an individual was considered to be at risk for the n-th event only if the patient has experienced the (n-1)th event.

R1

R2

R3

R4

R5

R6

R7

R8

R9

R10

R11

R12

R13

R14

R15

R16

R17

R18

R19

R20

R21

R22

R23

R24

R25

R26

R27

R28

R29

R30

R31

R32

R33

R34

R35

R36

R37

R38

R39


57

4

Since the results of a previous study suggested that not OH but orthostatic complaints or symptomatic OH are a risk factor for falling [22], post-hoc analyses to investigate the relationship between orthostatic complaints or symptomatic OH and (recurrent) falling were performed. The post-hoc analyses were comparable to primary analyses.Since timing of the OH measurement and consumption of a meal or drink could have influenced the results a second post-hoc analyses was performed in which we additionally adjusted for timing of the OH measurement and consumption of a meal or drink [8, 23, 24].When necessary, interaction was tested between different variables. Interaction was considered to be significant, with a p-value less than 0.05.All statistical analyses were performed using SPSS version 22 software and with SAS 9.3 software. The ‘Strengthening the Reporting of Observational studies in Epidemiology’ (STROBE) statement was used to describe this observational cohort study [25].

thical approval and linical rial registrationThis study was performed in accordance with the Declaration of Helsinki. According to Dutch guidelines this study did not fall under the scope of the Medical Research Involving Human Subjects Act, and therefore this study did not need a formal approval of an accredited medical ethics committee. Written informed consent was obtained for all patients by the participating medical doctor or nurse. In case of a mentally incompetent patient, informed consent was obtained from the legally authorized representative. As in the present study only blood pressure measurements were performed, this study entailed no additional risk or burden. All data were analysed anonymously. The study was registered on ClinicalTrials.gov (NCT01362751).

RESULTS

A total of 249 patients were included in this cohort, 80 patients at the psychogeriatric department, 42 patients at the somatic department, and 127 patients at the rehabilitation department. During a median follow-up period of 1.3 years, 450 falls were recorded during the follow-up; 181 (73%) patients reported no falls and 68 (27%) patients reported one or more falls. Fifty-three of the fallers (78%) were recurrent fallers. When stratified according to department; 9 (21%) of the somatic, 38 (48%) of the psychogeriatric, and 6 (5%) of the rehabilitation patients recorded > 1 falls.

The baseline characteristics are presented in table 1. Median age of the total study population was 82 (IQR (interquartile range 76-87) years. OH was present in 93 out of 249 patients, resulting in a prevalence of 37% (95%CI (confidence interval) 31-43%).

R1

R2

R3

R4

R5

R6

R7

R8

R9

R10

R11

R12

R13

R14

R15

R16

R17

R18

R19

R20

R21

R22

R23

R24

R25

R26

R27

R28

R29

R30

R31

R32

R33

R34

R35

R36

R37

R38

R39

Chapter 4

58

Tabl

e 1.

Bas

elin

e ch

arac

teri

stics

tota

l pop

ulati

on a

nd d

iffer

ent g

roup

s of

pati

ents

.To

tal

Psyc

hoge

riat

ric

depa

rtm

ent

Som

atic

depa

rtm

ent

Reha

bilit

ation

dep

artm

ent

Char

acte

risti

cN

=249

N=8

0N

=42

N=1

27D

emog

raph

ics

Ag

e

82 (7

6-87

)84

(78-

89)

76 (6

1-84

)81

(77-

86)

Fem

ale

Gen

der

173

(70)

57 (7

1)25

(60)

91 (7

2)M

ean

body

mas

s in

dex,

kg/

m2

25 (2

2-29

)25

(22-

29)

28 (2

3-32

)25

(23-

28)

Hyp

erte

nsio

n 19

1 (7

7)58

(73)

29 (6

9)10

4 (8

2)H

isto

ry o

f CVD

110

(44)

35 (4

4)25

(60)

50 (3

9)D

iabe

tes

mel

litus

92 (3

7)21

(26)

17 (4

1)54

(43)

Dem

entia

89 (3

6)75

(94)

5 (1

2)9

(7)

Curr

ent s

mok

er

26 (1

0)5

(6)

9 (2

1)12

(9)

Mea

sure

men

tsCo

nsum

ption

mea

l or

drin

k 17

9 (7

2)64

(80)

29 (6

9)86

(68)

Mea

n SB

P ly

ing,

mm

Hg

144

(130

-162

)14

6 (1

30-1

62)

141

(128

-169

)14

4 (1

32-1

60)

Mea

n D

BP ly

ing,

mm

Hg

74 (6

6-81

)74

(68-

83)

75 (6

9-84

)73

(66-

79)

Mea

n pu

lse

freq

uenc

y, b

eats

/min

75

(13)

73 (1

4)75

(13)

75 (1

2)Fa

lls la

st y

ear

57 (2

3)7

(9)

4 (1

0)46

(36)

Pros

pecti

ve fa

lls68

(27)

45 (5

6)13

(31)

10 (8

)Re

curr

ent f

alls

53 (2

1)38

(48)

9 (2

1)6

(5)

Ort

host

atic

hypo

tens

ion

93 (3

7)33

(41)

13 (3

1)47

(37)

Ort

host

atic

com

plai

nts

62 (2

5)18

(23)

10 (2

4)34

(27)

Scor

e Ba

rthe

l que

stion

naire

11

(6-1

5)9

(4-1

4)7

(3-1

3)13

(9-1

6)

Med

icati

onM

ean

num

ber

of a

gent

s 9

(6-1

1)8

(6-1

1)9

(7-1

2)9

(6-1

1)A

ntihy

pert

ensi

ve m

edic

ation

- D

iure

tics

- Be

ta b

lock

ers

- Ca

lciu

m c

hann

el b

lock

ers

- AC

E in

hibi

tors

160

(64)

106

(43)

71 (2

9)35

(14)

81 (3

3)

48 (6

0)30

(38)

15 (1

9)10

(13)

30 (3

8)

22 (5

2)14

(33)

11 (2

6)4

(10)

9 (2

1)

90 (7

1)62

(49)

45 (3

5)21

(17)

42 (3

3)

Benz

odia

zepi

nes

131

(53)

38 (4

8)30

(71)

63 (5

0)A

ntips

ycho

tics

48 (1

9)26

(33)

8 (1

9)14

(11)

Anti

depr

essa

nts

59 (2

4)26

(33)

14 (3

3)19

(15)

Opi

oids

28 (1

1)7

(9)

8 (1

9)13

(10)

Dat

a ar

e m

eans

(±

SD),

med

ians

(in

terq

uarti

le r

ange

) or

n (

%).

CVD

= C

ardi

o va

scul

ar d

isea

se. S

BP =

sys

tolic

blo

od p

ress

ure.

DBP

= d

iast

olic

blo

od

pre

ssu

re

.

R1

R2

R3

R4

R5

R6

R7

R8

R9

R10

R11

R12

R13

R14

R15

R16

R17

R18

R19

R20

R21

R22

R23

R24

R25

R26

R27

R28

R29

R30

R31

R32

R33

R34

R35

R36

R37

R38

R39


59

4

Tabl

e 1.

Bas

elin

e ch

arac

teri

stics

tota

l pop

ulati

on a

nd d

iffer

ent g

roup

s of

pati

ents

.To

tal

Psyc

hoge

riat

ric

depa

rtm

ent

Som

atic

depa

rtm

ent

Reha

bilit

ation

dep

artm

ent

Char

acte

risti

cN

=249

N=8

0N

=42

N=1

27D

emog

raph

ics

Ag

e

82 (7

6-87

)84

(78-

89)

76 (6

1-84

)81

(77-

86)

Fem

ale

Gen

der

173

(70)

57 (7

1)25

(60)

91 (7

2)M

ean

body

mas

s in

dex,

kg/

m2

25 (2

2-29

)25

(22-

29)

28 (2

3-32

)25

(23-

28)

Hyp

erte

nsio

n 19

1 (7

7)58

(73)

29 (6

9)10

4 (8

2)H

isto

ry o

f CVD

110

(44)

35 (4

4)25

(60)

50 (3

9)D

iabe

tes

mel

litus

92 (3

7)21

(26)

17 (4

1)54

(43)

Dem

entia

89 (3

6)75

(94)

5 (1

2)9

(7)

Curr

ent s

mok

er

26 (1

0)5

(6)

9 (2

1)12

(9)

Mea

sure

men

tsCo

nsum

ption

mea

l or

drin

k 17

9 (7

2)64

(80)

29 (6

9)86

(68)

Mea

n SB

P ly

ing,

mm

Hg

144

(130

-162

)14

6 (1

30-1

62)

141

(128

-169

)14

4 (1

32-1

60)

Mea

n D

BP ly

ing,

mm

Hg

74 (6

6-81

)74

(68-

83)

75 (6

9-84

)73

(66-

79)

Mea

n pu

lse

freq

uenc

y, b

eats

/min

75

(13)

73 (1

4)75

(13)

75 (1

2)Fa

lls la

st y

ear

57 (2

3)7

(9)

4 (1

0)46

(36)

Pros

pecti

ve fa

lls68

(27)

45 (5

6)13

(31)

10 (8

)Re

curr

ent f

alls

53 (2

1)38

(48)

9 (2

1)6

(5)

Ort

host

atic

hypo

tens

ion

93 (3

7)33

(41)

13 (3

1)47

(37)

Ort

host

atic

com

plai

nts

62 (2

5)18

(23)

10 (2

4)34

(27)

Scor

e Ba

rthe

l que

stion

naire

11

(6-1

5)9

(4-1

4)7

(3-1

3)13

(9-1

6)

Med

icati

onM

ean

num

ber

of a

gent

s 9

(6-1

1)8

(6-1

1)9

(7-1

2)9

(6-1

1)A

ntihy

pert

ensi

ve m

edic

ation

- D

iure

tics

- Be

ta b

lock

ers

- Ca

lciu

m c

hann

el b

lock

ers

- AC

E in

hibi

tors

160

(64)

106

(43)

71 (2

9)35

(14)

81 (3

3)

48 (6

0)30

(38)

15 (1

9)10

(13)

30 (3

8)

22 (5

2)14

(33)

11 (2

6)4

(10)

9 (2

1)

90 (7

1)62

(49)

45 (3

5)21

(17)

42 (3

3)

Benz

odia

zepi

nes

131

(53)

38 (4

8)30

(71)

63 (5

0)A

ntips

ycho

tics

48 (1

9)26

(33)

8 (1

9)14

(11)

Anti

depr

essa

nts

59 (2

4)26

(33)

14 (3

3)19

(15)

Opi

oids

28 (1

1)7

(9)

8 (1

9)13

(10)

Dat

a ar

e m

eans

(±

SD),

med

ians

(in

terq

uarti

le r

ange

) or

n (

%).

CVD

= C

ardi

o va

scul

ar d

isea

se. S

BP =

sys

tolic

blo

od p

ress

ure.

DBP

= d

iast

olic

blo

od

pre

ssu

re

.

The intended postural change could be performed from lying to standing in 186 (75%) of the patients. The remaining patients performed the postural change from lying to sitting.

and allingTables 2 and 3 present the hazard ratios of OH with regard to the first fall incident and recurrent falls. No significant associations were found between OH and the first fall incident and recurrent falling, in both the unadjusted and adjusted models. In the fully adjusted models, the Barthel index and type of department were significantly related to first fall incident and recurrent falls. A 1-point higher score on the Barthel index, increased the risk of a first fall incident with 7% (95%CI 2-11%) and the risk of recurrent falling with 6% (95%CI 1-12%). Interaction was tested between hypertension and number of medication; no interaction was observed.In tables 2 and 3 also the HR’s are presented regarding the type of department and falling. The risk of first fall incident and recurrent falling was the highest in the psychogeriatric department, followed by the somatic, and rehabilitation departments. Because the intended postural change could be performed from lying to standing in 75% of the patients, we performed post-hoc analyses with only these patients. The hazard ratios regarding to first fall incident and recurrent falling did not relevantly change (Table A.1). Additional adjustment of timing of OH measurement or consumption of meals in the post-hoc analyses did not change the results (Table A.2). In the post-hoc analyses no associations were seen between orthostatic complaints and falling or recurrent falling, with HRs in the fully adjusted models of 0.91 (95%CI 0.51-1.61) and 0.92 (95%CI 0.46-1.83), respectively. Also, no associations were seen between symptomatic OH and falling or recurrent falling, with HRs in the fully adjusted models of 0.65 (95%CI 0.26-1.65) and 0.92 (95%CI 0.33-2.54), respectively.

Table 2. Cox regression analyses; hazard ratios for first fall incident (n=249).

Model 1HR (95%CI)

Model 2HR (95%CI)

Model 3HR (95%CI)

OH 1.04 (0.64-1.71) 0.99 (0.60-1.64) 1.01 (0.60-1.69)Age - 1.01 (0.98-1.04) 1.02 (0.99-1.05)Gender - 1.22 (0.71-2.07) 1.25 (0.72-2.16)BMI - - 1.00 (0.95-1.05)Hypertension - - 0.93 (0.52-1.66)Previous Fall - - 1.19 (0.59-2.38)Barthel Index - - 1.07 (1.02-1.11)Number of medication - - 1.00 (0.92-1.08)Type of Department- somatic vs psychogeriatric- somatic vs rehabilitation- psychogeriatric vs rehabilitation

---

---

0.50 (0.26-0.98)3.15 (1.31-7.56)6.29 (3.12-12.68)

R1

R2

R3

R4

R5

R6

R7

R8

R9

R10

R11

R12

R13

R14

R15

R16

R17

R18

R19

R20

R21

R22

R23

R24

R25

R26

R27

R28

R29

R30

R31

R32

R33

R34

R35

R36

R37

R38

R39

Chapter 4

60

Table 3. Cox regression analyses; hazard ratios for recurrent falling (n=249).

Model 1HR (95%CI)

Model 2HR (95%CI)

Model 3HR (95%CI)

OH 1.31 (0.74-2.31) 1.25 (0.69-2.24) 1.21 (0.65-2.24)Age - 1.01 (0.98-1.04) 1.01 (0.97-1.04)Gender - 1.34 (0.73-2.45) 1.32 (0.70-2.51)BMI - - 1.06 (1.01-1.12)Hypertension - - 0.76 (0.36-1.61)Previous Fall - - 1.38 (0.60-3.20)Barthel Index - - 1.06 (1.01-1.12)Number of medication - - 1.01 (0.92-1.10)Type of Department- somatic vs psychogeriatric- somatic vs rehabilitation- psychogeriatric vs rehabilitation

---

---

0.32 (0.14-0.72)2.74 (1.01-7.44)8.67 (3.97-18.93)

DISCUSSION

In the current study, we observed no relationship between OH and falling in nursing home residents. Both falling and OH were frequently seen in the current study; 27% and 37%, respectively. Almost eighty percent of the fallers were recurrent fallers.

Previous studies that described the relationship between OH and falling, have important limitations that need to be addressed; the lack of prospective fall data [9-11], not adjusting for important confounders [7], or not using the international consensus definition of OH [5]. Besides, different studies only described the relation between OH and falling in subgroups instead of within the total study population [8, 12, 13].Although different studies investigated this relationship with prospective fall data, the design, study population, and setting were quite different [6-8, 12, 26-29]. As in the current study, none of the studies showed a clear relationship between OH, considering the consensus definition of OH, and falling. Although there is not one study describing a significant relationship between OH and falling, different subgroups are mentioned with a higher risk of falling. One study described OH as an independent risk factor for recurrent falls [8, 27], other studies showed that a decrease in systolic blood pressure > 20 mmHg was related to falling [6, 7], and another study showed that only symptomatic OH was a predictor for falling [26].

The results of a previous study suggested that not OH but orthostatic complaints or symptomatic OH are a risk factor for falling [22]. In present study, no associations were seen between orthostatic complaints, symptomatic OH, and (recurrent) falling.

R1

R2

R3

R4

R5

R6

R7

R8

R9

R10

R11

R12

R13

R14

R15

R16

R17

R18

R19

R20

R21

R22

R23

R24

R25

R26

R27

R28

R29

R30

R31

R32

R33

R34

R35

R36

R37

R38

R39


61

4

The current study used prospective fall incidents whereas the previous study used a high risk of falling that was based on previous fall incidents and risk factors for falling [22]; this may explain the difference in results.

Important confounders regarding the relation between OH and falling were the Barthel index and the type of department, both significantly related to falling and recurrent falling. The risk of first fall incident and recurrent falling were the highest on the psychogeriatric department, which is not surprising [26, 30]. Higher scores on the Barthel index were related to an increased risk of falling. Patients with higher scores on the Barthel index are more independent but also seem to be more likely to fall. The more vital patients within this very frail population may receive less help with walking and mobilizing and may therefore have a greater risk of falling compared to the frailest patients.

trength and i itationsMain strengths of the present study are the prospective fall registration and use of the conditional frailty model for recurrent events. Our study has several important limitations that need to be addressed. Because the medical staff registered falls, some falls could be missed when the medical staff did not notice a fall. Also, the awareness of having OH or orthostatic complaints could play a preventive role in falling. Secondly, the information regarding previous falls was based on questioning patients or participants. It is very likely that the actual number of patients with previous fall incidents was higher. Thirdly, OH blood pressure measurement was performed only at one moment at baseline, which could have influenced accurate classification of patients as having OH or not. It is known that OH is variable over the course of the day [8, 24, 31].Furthermore, because of the observational design, establishing causality is not possible. Finally, the sample size was relatively small and therefore the results lack precision.

CONCLUSIONS

Although recent literature assumes orthostatic hypotension influences falling, the present study showed there is no relationship between orthostatic hypotension and (recurrent) falling in a nursing home population.

R1

R2

R3

R4

R5

R6

R7

R8

R9

R10

R11

R12

R13

R14

R15

R16

R17

R18

R19

R20

R21

R22

R23

R24

R25

R26

R27

R28

R29

R30

R31

R32

R33

R34

R35

R36

R37

R38

R39

Chapter 4

62

REFERENCES



3. Shaw BH, Claydon VE. The relationship between orthostatic hypotension and falling in older adults. Clinical autonomic research: official journal of the Clinical Autonomic Research Society. 2014 Feb;24(1):3-13.



6. Gangavati A, Hajjar I, Quach L, et al. Hypertension, orthostatic hypotension, and the risk of falls in a community-dwelling elderly population: the maintenance of balance, independent living, intellect, and zest in the elderly of Boston study. Journal of the American Geriatrics Society. 2011 Mar;59(3):383-9.







13. Luukinen H, Koski K, Laippala P, Kivela SL. Risk factors for recurrent falls in the elderly in long-term institutional care. Public health. 1995 Jan;109(1):57-65.

14. Moonen JE, Foster-Dingley JC, de Ruijter W, et al. Effect of discontinuation of antihypertensive medication on orthostatic hypotension in older persons with mild cognitive impairment: the DANTE Study Leiden. LID - afv199 [pii]. 20160203(1468-2834 (Electronic)).


16. Sainsbury A, Seebass G, Bansal A, Young JB. Reliability of the Barthel Index when used with older people. Age and ageing. 2005 May;34(3):228-32.

17. Altunkan S, Iliman N, Altunkan E. Validation of the Omron M6 (HEM-7001-E) upper arm blood pressure measuring device according to the International Protocol in elderly patients. Blood pressure monitoring. 2008 Apr;13(2):117-22.

R1

R2

R3

R4

R5

R6

R7

R8

R9

R10

R11

R12

R13

R14

R15

R16

R17

R18

R19

R20

R21

R22

R23

R24

R25

R26

R27

R28

R29

R30

R31

R32

R33

R34

R35

R36

R37

R38

R39


63

4

18. Dorigatti F, Bonso E, Zanier A, Palatini P. Validation of Heine Gamma G7 (G5) and XXL-LF aneroid devices for blood pressure measurement. Blood pressure monitoring. 2007 Feb;12(1):29-33.

19. Mariotti G, Alli C, Avanzini F, et al. Arm position as a source of error in blood pressure measurement. Clinical cardiology. 1987 Oct;10(10):591-3.

20. Box-Steffensmeier JM, De Boef S. Repeated events survival models: the conditional frailty model. 20060927 DCOM- 20070125(0277-6715 (Print)).

21. Cui J, Forbes A Fau - Kirby A, Kirby A Fau - Marschner I, et al. Parametric conditional frailty models for recurrent cardiovascular events in the lipid study. 20081125 DCOM- 20090303(1740-7745 (Print)).

22. van Hateren KJ, Kleefstra N, Blanker MH, et al. Orthostatic hypotension, diabetes, and falling in older patients: a cross-sectional study. The British journal of general practice : the journal of the Royal College of General Practitioners. 2012 Oct;62(603):e696-702.

23. Weiss A, Grossman E Fau - Beloosesky Y, Beloosesky Y Fau - Grinblat J, Grinblat J. Orthostatic hypotension in acute geriatric ward: is it a consistent finding? 20021106 DCOM- 20021204(0003-9926 (Print)).


25. von Elm E, Altman DG, Egger M, et al. The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) Statement: Guidelines for reporting observational studies. International journal of surgery. 2014 Jul 18.

26. Allan LM, Ballard CG, Rowan EN, Kenny RA. Incidence and prediction of falls in dementia: a prospective study in older people. PLoS One. 2009;4(5):e5521.

27. Graafmans WC, Ooms ME, Hofstee HMA, et al. Falls in the elderly: A prospective study of risk factors and risk profiles. American Journal of Epidemiology. 1996;143(11):1129-36.

28. Luukinen H, Koski K, Kivela SL, Laippala P. Social status, life changes, housing conditions, health, functional abilities and life-style as risk factors for recurrent falls among the home-dwelling elderly. Public health. 1996 Mar;110(2):115-8.

29. Maurer MS, Burcham J, Cheng H. Diabetes mellitus is associated with an increased risk of falls in elderly residents of a long-term care facility. The journals of gerontology Series A, Biological sciences and medical sciences. 2005 Sep;60(9):1157-62.

30. Bunn F, Dickinson A Fau - Simpson C, Simpson C Fau - Narayanan V, et al. Preventing falls among older people with mental health problems: a systematic review. 20140306 DCOM- 20140306(1472-6955 (Electronic)).

31. Alli C, Avanzini F, Bettelli G, et al. Prevalence and variability of orthostatic hypotension in the elderly. Results of the ‘Italian study on blood pressure in the elderly (SPAA)’. The ‘Gruppo di Studio Sulla Pressione Arteriosa nell’Anziano’. European heart journal. 1992 Feb;13(2):178-82.

R1

R2

R3

R4

R5

R6

R7

R8

R9

R10

R11

R12

R13

R14

R15

R16

R17

R18

R19

R20

R21

R22

R23

R24

R25

R26

R27

R28

R29

R30

R31

R32

R33

R34

R35

R36

R37

R38

R39

Chapter 4

64

SUPPLEMENTAL DATA

Table A.1. Cox regression analyses; hazard ratios of OH for first fall incident and recurrent falls, within patients were postural change could be performed from lying to standing (n=186).

First fall

HR (95% CI)Recurrent FallsHR (95% CI)

OH (model 1) 1.18 (0.70-2.01) 1.49 (0.83-2.69)OH (model 2) 1.11 (0.64-1.91) 1.40 (0.76-2.58)OH (model 3) 1.02 (0.58-1.80) 1.27 (0.67-2.39)

Model 1 was unadjusted, model 2 adjusted for age and gender, model 3 adjusted for age, gender, body mass index (BMI), history of hypertension, previous fall (within last 12 months), the score on the Barthel Index questionnaire, Number of medication, and type of department.

Table A.2. Cox regression analyses; hazard ratios for first fall incident and recurrent falling.

First fall incidentHR (95%CI)

Recurrent FallingHR (95%CI)

OH (model 1) 1.04 (0.64-1.71) 1.31 (0.74-2.31)OH (model 2) 0.99 (0.60-1.64) 1.25 (0.69-2.24)OH (model 3) 1.01 (0.60-1.69) 1.21 (0.65-2.24)OH (model 4) 1.04 (0.62-1.76) 1.23 (0.65-2.33)OH (model 5) 0.93 (0.55-1.59) 1.18 (0.63-2.21)OH (model 6) 0.97 (0.56-1.66) 1.19 (0.62-2.27)

Model 1 is the unadjusted model. In model 2 is adjusted for age and gender. In model 3 we adjusted for age, gender, BMI, hypertension, previous fall, barthel index, number of medication, and type of department. In model 4 is adjusted for age, gender, BMI, hypertension, previous fall, barthel index, number of medication, type of department, and timing of the day (morning, afternoon, night). In model 5 is adjusted for age, gender, BMI, hypertension, previous fall, barthel index, number of medication, type of department, and consumption of a meal (yes/no). In model 6 is adjusted for age, gender, BMI, hypertension, previous fall, barthel index, number of medication, type of department, timing of the day, and consumption of a meal.

CHAPTER 5Is orthostatic hypotension related to falling?

A meta-analysis of individual patient data of prospective

observational studies

Submitted as: Hartog LC, Schrijnders D, Landman GWD, Groenier KH, Kleefstra N, Bilo HJG,

Van Hateren KJJ. Is orthostatic hypotension related to falling? A meta-analysis

of individual patient data of prospective observational studies.

R1

R2

R3

R4

R5

R6

R7

R8

R9

R10

R11

R12

R13

R14

R15

R16

R17

R18

R19

R20

R21

R22

R23

R24

R25

R26

R27

R28

R29

R30

R31

R32

R33

R34

R35

R36

R37

R38

R39

Chapter 5

66

ABSTRACT

ac ground: Orthostatic hypotension (OH) is one out of many risk factors believed to contribute to an increased fall risk in elderly subjects but it is unclear whether an independent association between OH and falling exists.

urpose: To perform an individual patient data (IPD) meta-analysis of prospective observational studies investigating the relationship between OH and falling.

Materials and Methods: MEDLINE, EMBASE, the Cochrane Library, and the abstracts of annual meetings of selected hypertension societies were searched. Both one-stage (analysing all IPD from all studies simultaneously) and two-stage (analysing IPD per study, and then pooling the results) methods were used, and both logistic and cox regression analyses were performed. The study protocol was published on PROSPERO (2015:CRD42015019178).

esults: From the 34 selected abstracts, 6 studies were included. IPD were provided in 1022 patients from 3 cohorts and were included in the IPD meta-analysis. The hazard ratio (HR) in the one-stage cox proportional hazard model was 1.52 (95% Confidence Interval (CI) 1.23-1.88).No significant relationship between OH and falling was found in the one-stage logistic regression analysis (Odds Ratio (OR) 1.21 (95% CI 0.87-1.68) and the two-stage logistic and cox regression analyses.

onclusions: This IPD meta-analysis of prospective observational studies showed a clear and significant relationship between OH and time to first fall incident. Although the ORs of falling was not significantly different for patients with and without OH, the width of the 95% CI does not exclude a relevant clinical association between OH and falling.

R1

R2

R3

R4

R5

R6

R7

R8

R9

R10

R11

R12

R13

R14

R15

R16

R17

R18

R19

R20

R21

R22

R23

R24

R25

R26

R27

R28

R29

R30

R31

R32

R33

R34

R35

R36

R37

R38

R39

s orthostatic h potension related to alling n eta anal sis o prospective observational studies

67

5

INTRODUCTION

The prevalence of orthostatic hypotension (OH), defined as a decrease in systolic blood pressure by at least 20 mmHg or a decrease in diastolic blood pressure by at least 10 mmHg within 3 minutes after changing to standing position, increases with advancing age [1]. OH is associated with cardiovascular disease and all-cause mortality, especially in elderly subjects [2-5]. Furthermore, OH is presumed to be associated with an increased fall risk. Especially in frail elderly, OH and the subsequent increased risk of falling are considered to potentially lead to severe morbidity [6-11]. Equivalent to the aetiology of OH, fall risk is a complex and multifactorial phenomenon and OH is one out of many risk factors believed to contribute to an increased fall risk in elderly subjects [7, 8, 12].Previous studies on the relationship between OH and fall risk were mostly performed in nursing homes and reported a positive association [9, 13, 14]. These results cannot be directly extrapolated to home dwelling elderly subjects. Two recent systematic reviews described the relationship between OH and falling in old age, although the absolute attributive risk could not established due to few included patients [7, 8]. It remains unclear whether there is an independent association between OH and falling. Therefore, we aimed to investigate whether OH contributes to falling in a meta-analysis of individual patient data (IPD). When individual patient data could not be retrieved, a secondary analysis was pre-planned that aimed to perform a meta-analysis of published study results.


rotocolThe prespecified objectives, eligibility criteria, quality assessment and main analyses were published on PROSPERO (2015:CRD42015019178). PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) recommendations were followed throughout the design, implementation, analysis and reporting of this study [15].A study was considered eligible if it was a prospective study that used the 1996 consensus guideline definition of OH [1], included adults subjects, and described the relationship between OH and fall incidents.

ata ources and earchesAn electronic search of MEDLINE (Pubmed), EMBASE, the Cochrane Library, and the abstracts of the 2012, 2013, and 2014 annual meetings of the International society of hypertension and American Society of Hypertension, was performed on 14 April 2015 and updated on 12 February 2016. The search was restricted to the English-language literature.

R1

R2

R3

R4

R5

R6

R7

R8

R9

R10

R11

R12

R13

R14

R15

R16

R17

R18

R19

R20

R21

R22

R23

R24

R25

R26

R27

R28

R29

R30

R31

R32

R33

R34

R35

R36

R37

R38

R39

Chapter 5

68

Additional studies were retrieved by hand searching references of selected articles. Clinical trials registries (www.clinicaltrials.gov, www.clinicaltrialsregister.eu and www.trialregister.nl) were searched for unpublished data. The following search terms were used: ‘Orthostatic hypotension’ OR ‘postural hypotension’ OR ‘orthostatic’ OR ‘Medical Subject Heading (MeSH) terms orthostatic hypotension’ AND ‘falls’ OR ‘falling’ OR ‘recurrent falls’ OR ‘accidental falls’ OR ‘fall-risk’ OR ‘MeSH terms accidental falls’. The complete search query is available in Appendix 1.

tud electionPublications retrieved from MEDLINE, EMBASE, and the Cochrane library were imported in Endnote reference management software. Duplicates were removed and, two reviewers (LH, DS) independently screened abstracts. For abstract selection and full-text selection inclusion criteria were used. The same two reviewers (LH, DS) extracted data and assessed the quality of each study. Differences in opinion between reviewers were resolved by consensus with a third reviewer (KH). Two reviewers (LH, DS) independently searched the trial registers.

ata ollection and ata te sFrom each study, data were extracted concerning; authors, year of publication, national clinical trial (NCT) number (if applicable), studied population, sample size, participants’ baseline characteristics (age, gender, blood pressure, OH, body mass index (BMI), medication, hypertension, Parkinson’s disease); and fall incidents (yes/no).

b ectiveThe primary objective was to investigate whether OH contributes to falling in a meta-analysis of IPD. When individual patient data could not be retrieved, a secondary analysis was pre-planned that aimed to perform a meta-analysis of published study results.

Missing ata and Multiple eportsAuthors of the selected papers were contacted and asked whether they were willing to share (anonymous) individual patient data. Initially, the first author was contacted by email or telephone (repeatedly in case of no response). The other authors were contacted when the first author did not respond. In case individual patient data were not provided, published data were used.

is o ias ssess entThe quality of each study was assessed using the Newcastle-Ottowa Scale (NOS) [16]. On a 9-point scale, the NOS evaluates the quality of observational studies on three broad

R1

R2

R3

R4

R5

R6

R7

R8

R9

R10

R11

R12

R13

R14

R15

R16

R17

R18

R19

R20

R21

R22

R23

R24

R25

R26

R27

R28

R29

R30

R31

R32

R33

R34

R35

R36

R37

R38

R39


69

5

categories: selection of the study groups (maximum of 4 stars); comparability of the groups (maximum of 2 stars); and ascertainment of the outcome of interest (maximum of 3 stars). Two reviewers (LH, DS) independently assessed quality; differences in opinion were resolved by consensus with a third reviewer (KH).

tatistical nal sisIPD analyses were performed using one-stage and two-stage methods [17-19]. In a one-stage method, analyses are performed with all IPD from all studies simultaneously, just as they belong to a single trial. In a two-stage method analyses are first performed for all individual studies separately and then pooled into a meta-analysis for estimating the overall effect [20]. Both one-stage and two-stage methods were used.

Univariate binary logistic regression analyses and subsequently multivariate binary logistic regression analyses were performed to assess the association of OH and falling (yes/no). A cox proportional hazard modelling was used to investigate the relation between OH and the first fall incident. Three different predefined models were used. In model 1, unadjusted analyses were performed. In model 2, only age and gender were taken into account as possible confounders. Model 3 was adjusted for age, gender, BMI, use of antihypertensive medication, systolic blood pressure (SBP), diabetes mellitus (DM), and the total number of drugs. Logistic regression and cox proportional hazard analyses were performed in both one-stage and two-stage methods. Odds ratios (OR) and hazard ratios (HR) between patients with OH compared to patients without OH and 95% confidence intervals (CI) were calculated. Because of the observational design of all studies heterogeneity was expected, therefore initially a random effects model was used. In case of no significant heterogeneity, a fixed effect model was also applied. Heterogeneity was considered relevant if the p value was <0.10.

Logistic regression and cox proportional hazard analyses were performed using SPSS, version 22. Pooling of results and analyses regarding heterogeneity were performed with RevMan 5.3.

ensitivit and subgroup anal sesIn case of significant heterogeneity, sensitivity analyses were planned to explore the possible source of heterogeneity. In the one-stage method, predefined subgroup analyses were planned and registered at PROSPERO regarding the following variables: age, patient group, community dwelling or nursing home patients, study quality, and specific patients groups; e.g. hypertension, DM, and Parkinson. The predefined subgroup analyses were only performed in case of an adequate number of patients to justify the subgroup analyses. In

R1

R2

R3

R4

R5

R6

R7

R8

R9

R10

R11

R12

R13

R14

R15

R16

R17

R18

R19

R20

R21

R22

R23

R24

R25

R26

R27

R28

R29

R30

R31

R32

R33

R34

R35

R36

R37

R38

R39

Chapter 5

70

the two-stage method, subgroup and meta-regression analyses were considered with 10 or more included studies [21].

thical approval and linical rial registrationThis study was performed in accordance with the Declaration of Helsinki. According to Dutch guidelines this study did not fall under the scope of the Medical Research Involving Human Subjects Act, and therefore this study did not need a formal approval of an accredited medical ethics committee. All data were analysed anonymously. The eligibility criteria, outcomes, and analyses were pre-specified and published on PROSPERO (2015: CRD42015019178).

RESULTS

earch resultsIn total, 34 abstracts [10, 11, 13, 22-52] were selected for full-text evaluation, 6 of these were included in the meta-analysis (Figure 1) [10, 11, 13, 23, 39, 41]. The authors of these studies were contacted for sharing IPD. IPD were provided from 3 studies [10, 11, 23]. The authors of one study did not reply to several attempts to make email (or telephone) contact [39], one was not able to participate in the study without giving a specific reason [41], and one could not provide individual patient data because data were no longer available [13]. Also, results as published could not be used from these studies because only HRs or ORs of subgroup analyses were published instead of results of the total study group regarding OH and falling. In addition, different endpoints were described and therefore these published results could not be included into the current meta-analysis.

tud characteristicsThe characteristics of the 6 selected prospective cohort studies are shown in table 1. The baseline characteristics of the 3 included studies are shown in table 2. The follow-up period in all included studies was approximately 1 year. The participants of the three studies included in the IPD meta-analysis were all community-dwelling elderly [10, 11, 23]. The sample size of the cohorts ranged from 70 to 736 subjects. The median (interquartile range) age was 77 years (73-81). The prevalence of OH ranged from 11% to 82 % and the prevalence of one or more fall incidents ranged from 51% to 62%. OH was measured in all 3 studies from supine to standing position. Fall incidents were retrieved by fall calendars [10], monthly questionnaires [11], or fall diaries [23].

R1

R2

R3

R4

R5

R6

R7

R8

R9

R10

R11

R12

R13

R14

R15

R16

R17

R18

R19

R20

R21

R22

R23

R24

R25

R26

R27

R28

R29

R30

R31

R32

R33

R34

R35

R36

R37

R38

R39


71

5

Figure 1. Flowchart.

is o ias ssess entThe three studies all scored 6 out of 9 stars, meaning that the overall quality of the included studies was moderate [10, 11, 23] (table 1). Because all studies had an equal NOS score, no subgroup analysis was performed regarding study quality. See appendix table 1 for the detailed NOS score of the 3 included studies in the IPD meta-analysis.

R1

R2

R3

R4

R5

R6

R7

R8

R9

R10

R11

R12

R13

R14

R15

R16

R17

R18

R19

R20

R21

R22

R23

R24

R25

R26

R27

R28

R29

R30

R31

R32

R33

R34

R35

R36

R37

R38

R39

Chapter 5

72

Tabl

e 1.

Cha

ract

eris

tics

of th

e se

lect

ed p

rosp

ectiv

e co

hort

stu

dies

.

Stud

yLo

cati

onTo

tal

Follo

w-u

p (m

onth

s)M

ean

Age

(SD

)Fe

mal

e N

(%)

Stud

y qu

alit

y(N

OS)

Alla

n et

al.

[23]

UK

179

12

76

(7)

73 (4

1%)

6 s

ta

rs

Gan

gava

ti et

al.

[10]

USA

722

12

78

(5)

462

(64%

)6

sta

rs

Hei

tter

achi

et a

l. [1

1]A

ustra

lia

701

2

77 (6

)56

(80%

)6

sta

rs

Mau

rer

et a

l. [4

1]U

SA1

11

9 88

(7)

91 (8

2%)

8 st

ars

Luuk

inen

et a

l. [3

9]Fi

nlan

d1

01

61

2

76 (5

)62

0 (6

1%)

7 st

ars

Ooi

et a

l. [1

3]U

SA84

414

No

t r

ep

orte

d67

7 (8

0%)

7 st

ars

NO

S: N

ewca

stle

-Ott

awa

Scal

e: on

a 9

-poi

nt s

cale

, the

NO

S ev

alua

tes

the

qual

ity o

f obs

erva

tiona

l stu

dies

on

thre

e br

oad

cate

gori

es: s

elec

tion

of t

he

stud

y gr

oups

(max

imum

of 4

sta

rs);

com

para

bilit

y of

the

gro

ups

(max

imum

of 2

sta

rs);

and

asce

rtai

nmen

t of

the

out

com

e of

inte

rest

(max

imum

of

3 st

ars)

.

Tabl

e 2.

Bas

elin

e ta

ble

of in

divi

dual

pati

ent d

ata.

Tota

l

(n=1

022)

Alla

n et

al,

[23]

20

09

(n=2

16)

Gan

gava

ti e

t al,

[10]

20

11

(n=7

36)

Hei

tter

achi

et a

l, [1

1]

2002

(n

=70)

OH

28

2 (2

8)17

8 (8

2)83

(11)

21 (3

0)Fa

lling

621

(61)

129

(60)

456

(62)

36 (5

1)A

ge (y

ears

)77

(73-

81)

76 (7

1-81

)77

(74-

82)

76 (7

2-80

)Fe

mal

e G

ende

r60

2 (5

8)85

(38)

461

(63)

56 (8

0)Bo

dy M

ass

Inde

x (k

g/m

2)

27 (2

4-29

)25

(23-

28)

27 (2

4-30

)26

(23-

28)

DM

, DM

vs

cont

rol

177

(17)

18 (8

)14

9 (2

0)10

(14)

Anti

hype

rten

sive

med

icati

on62

4 (6

1)75

(33)

519

(71)

30 (4

3)N

umbe

r of

med

icati

on

10 (4

-13)

4 (2

-6)

10 (7

-14)

3 (2

-5)

Syst

olic

blo

od p

ress

ure

(mm

Hg)

129

(119

-143

)13

5 (1

21-1

49)

131

(119

-140

)12

5 (1

08-1

47)

Dat

a ar

e m

eans

(± S

D),

med

ians

(int

erqu

artil

e ra

nge)

or

n (%

). O

H: o

rtho

stati

c hy

pote

nsio

n, D

M: D

iabe

tes

mel

litus

.

R1

R2

R3

R4

R5

R6

R7

R8

R9

R10

R11

R12

R13

R14

R15

R16

R17

R18

R19

R20

R21

R22

R23

R24

R25

R26

R27

R28

R29

R30

R31

R32

R33

R34

R35

R36

R37

R38

R39


73

5

ffect o on alling one stage ethod table No significant relationship between OH and falling was found in the one-stage logistic regression analysis.

Table 3. Adjusted odds ratios and Hazard ratios for the effect of orthostatic hypotension on the risk of falling with individual patient data (one-stage method). The odds ratios can be interpreted as a measure of the association of OH to falling (the dependent variables). Hazard ratios refer to time to first fall incident.

Odds ratio (95% CI)(3 studies)

Hazard ratio (95% CI)(2 studies)

OH (model 1) 0.92 (0.69-1.21) (n=1022) 1.30 (1.08-1.57) (n=952)OH (model 2) 0.94 (0.70-1.24) (n=1022) 1.30 (1.08-1.57) (n=952)OH (model 3) 1.21 (0.87-1.68) (n=954) 1.52 (1.23-1.88) (n=884)

Model 1:unadjusted. Model 2:adjusted for gender and age. Model 3:adjusted for gender, age, BMI, DM, antihypertensive medication, number of medication, and systolic blood pressure. OH = orthostatic hypotension. DM = Diabetes Mellitus.

From the three studies, two studies [10, 23] collected data on the time to first fall incident. These two studies could be used in the Cox analyses regarding the relationship between OH and first fall incident. A significant relationship was observed in the Cox regression analysis, HR 1.52 (95%CI 1.23-1.88). The chance of a first fall incident was 52% higher for patients with OH compared to those without. Figure 2 shows the cumulative proportion of first fall incident for patients with and without OH.

9

e .

1. 1. 1. 1. 1. 1. 1. 1. 1.

Figure

Figure 2. Cumulative proportion of first fall incident (survival curve).

R1

R2

R3

R4

R5

R6

R7

R8

R9

R10

R11

R12

R13

R14

R15

R16

R17

R18

R19

R20

R21

R22

R23

R24

R25

R26

R27

R28

R29

R30

R31

R32

R33

R34

R35

R36

R37

R38

R39

Chapter 5

74

Post-hoc logistic regression analyses were performed with IPD of the same two studies as used in the Cox regression analysis; results did not relevantly change (model 1: OR 0.87 (95%CI 0.65-1.16), model 2: OR 0.89 (95%CI 0.66-1.20), and model 3: OR 1.17 (95%CI 0.83-1.65)).

ffect o on alling t o stage ethodFor the two-stage method, ORs and HRs were analysed for each individual study separately and then pooled in the meta-analysis. No significant relationships were seen between OH and falling, in both the logistic and Cox regression analyses (figures 3 and 4).As no heterogeneity (I2 = 0%) was observed, also a fixed effect model was applied. When pooling the odds ratios, no significant relationship was seen between OH and falling, with both the random and fixed effect models. The unadjusted and adjusted ORs of the fixed effect models were 0.95 (95%CI 0.66-1.36; I2 = 0%; p = 0.39) and 0.98 (95%CI 0.66-1.44; I2 = 0%; p = 0.40), respectively (Fig 3). A Cox regression analysis was only performed for 2 studies [10, 23] because Heitterachi et al. did not report the time to first fall incident in their study [11]. No significant relationship between OH and time to first fall incident was seen. The pooled unadjusted and fully adjusted HRs were 0.95 (95%CI 0.74-1.22; I2 = 0%; p = 0.70) and 1.02 (95%CI 0.77-1.34; I2 = 0%; p = 0.98), respectively (Fig 4). Sensitivity analyses in both the logistic and Cox regression analysis were not performed because no heterogeneity was present.

ubgroup anal sesIn the one-stage method, only subgroup analyses regarding hypertension and diabetes status were performed. The three studies were comparable with respect to age, patient group, community dwelling or nursing home patients, and study quality. Subgroup analyses would not provide additional information and was thus not performed. A subgroup analysis on the covariate Parkinson’s disease could not be performed due to a low number of patients with Parkinson’s disease.No significant relation was seen between OH and falling within the logistic regression analysis stratified to hypertension and diabetes (Appendix table 2). When stratified according to diabetes status within the cox regression analysis, a significant relation was seen within non-diabetic patients in all three models; unadjusted HR 1.29 (95% CI 1.05-1.59), age- and gender adjusted HR 1.29 (95% CI 1.05-1.58), and fully adjusted HR 1.53 (95% CI 1.21-1.93). The stratified analysis regarding to hypertension did not relevantly change the results (Appendix table 2). Interaction between diabetes and OH, and hypertension and OH, was analysed and no interaction was seen, p=0.53 and p=0.86, respectively. In the two-stage method no subgroup or meta-regression analyses were performed because of the low number of studies included.

R1

R2

R3

R4

R5

R6

R7

R8

R9

R10

R11

R12

R13

R14

R15

R16

R17

R18

R19

R20

R21

R22

R23

R24

R25

R26

R27

R28

R29

R30

R31

R32

R33

R34

R35

R36

R37

R38

R39


75

5

Fi

gure

Fi

gure

.

Figu

re 3

. For

rest

Plo

t of

the

adj

uste

d od

ds r

atios

for

the

effec

t of

ort

host

atic

hypo

tens

ion

on t

he r

isk

of fa

lling

with

indi

vidu

al p

atien

t da

ta fo

r ea

ch

stud

y (t

wo-

stag

e m

etho

d). A

djus

ted

for

age,

gen

der,

BMI,

diab

etes

mel

litus

, anti

hype

rten

sive

med

icati

on, n

umbe

r of

med

icati

on, a

nd m

ean

syst

olic

b

loo

d p

re

ssu

re

.

Fi

gure

Fi

gure

.

Figu

re 4

. For

rest

Plo

t of t

he a

djus

ted

haza

rd ra

tios

for t

he e

ffect

of o

rtho

stati

c hy

pote

nsio

n on

the

risk

of f

allin

g w

ith in

divi

dual

pati

ent d

ata

for e

ach

stud

y (t

wo-

stag

e m

etho

d). A

djus

ted

for

age,

gen

der,

BMI,

diab

etes

mel

litus

, anti

hype

rten

sive

med

icati

on, n

umbe

r of

med

icati

on, a

nd m

ean

syst

olic

b

loo

d p

re

ssu

re

.

R1

R2

R3

R4

R5

R6

R7

R8

R9

R10

R11

R12

R13

R14

R15

R16

R17

R18

R19

R20

R21

R22

R23

R24

R25

R26

R27

R28

R29

R30

R31

R32

R33

R34

R35

R36

R37

R38

R39

Chapter 5

76

DISCUSSION

The one-stage meta-analysis showed a clear and significant relationship between OH and time to first fall incident. Although the odds ratio of falling was not significantly different for patients with and without OH, the width of the 95% confidence interval does not exclude a relevant clinical association. The results of the two-stage method, in which the results of the separate studies are pooled together, showed no significant relationship between OH and falling.

In the predefined protocol published on PROSPERO no preference was expressed about using logistic regression or cox proportional hazard analysis. Since both analyses can be useful to investigate the relationship between a causal factor and an outcome, both were used. While in the cox proportional hazard analysis the time to the first fall incident is the outcome measure from which the influence of OH on a fall incident is investigated, the ORs resulting from a logistic regression analysis only present cross-sectional information about this relationship after a (arbitrarily) fixed period of time. In the present meta-analysis, patients with OH had a 52% higher hazard on the first fall incident compared to patients without OH at any time during the follow-up period. Since patients can experience recurrent fall incidents over a longer period of time, a time to event analysis such as the cox regression analysis has a distinctive advantage over a binary logistic regression. In different studies OH seems to be related to recurrent falling in elderly patient [13, 32]. All things considered, the cox proportional hazard analysis provided more clinical relevant information in the present meta-analysis [53, 54].

In an IPD meta-analysis both one-stage and two-stage methods can be used. Although both types of analyses use similar IPD, the choice of a one- or two-stage method could result in different conclusions [19]. In the present meta-analysis this phenomenon was seen, HRs were different in one- and two-stage method, leading to different results. Debray et al. described differences between one- and two-stage methods and preferred the one-stage method, particularly in a meta-analysis with only a few studies [19]. Therefore, the conclusion in the present study is mainly based on the one-stage analyses.

This study is the first to aggregate results from observational studies into one IPD meta-analysis. In the prospective studies inconsistent results regarding the relationship between OH and fall incidents were reported [10, 11, 13, 23, 39, 41]. Furthermore, all studies had important limitations. Firstly, only in 1 out of the 6 selected studies the results for the total study population were reported [41]. The other selected studies mainly presented the results of subgroups

R1

R2

R3

R4

R5

R6

R7

R8

R9

R10

R11

R12

R13

R14

R15

R16

R17

R18

R19

R20

R21

R22

R23

R24

R25

R26

R27

R28

R29

R30

R31

R32

R33

R34

R35

R36

R37

R38

R39


77

5

or described different endpoints regarding the relationship between OH and fall incidents [10, 13, 23, 39, 41]. Maurer and colleagues described no relation of OH and time to first fall incident in nursing home residents [41]. Only HRs of systolic or diastolic blood pressure decrease at different time-points were presented. In the study of Gangavati et al. a significant relationship was only seen within a subgroup of participants with systolic OH at 1 minute [10] and Allan et al. described only symptomatic OH as a significant predictor of falls in patients with dementia [23]. Besides, Luukinen et al. described the relationship of OH and falling in the subgroup recurrent fallers [39]; no significant association between OH and recurrent falling was seen. However, Ooi et al. reported that OH was an independent risk factor for recurrent falls [13]. Secondly, 2 out of 6 studies did not adjust for important confounders [11, 41]. Heiterrachi et al., described an increased risk of falls in older people with a decrease in systolic blood pressure > 20 mmHg [11], but no confounders were taken into account. Maurer et al. did also not adjust for any confounders; a post-hoc analysis was performed in which was adjusted for available medications only [41]. Thirdly, 2 out of the 6 selected studies did not use the International consensus definition of OH when analysing results [10, 41]. Maurer et al. and Gangavati et al. only described HRs at 1 minute or 3 minutes after standing separately, while the International consensus definition includes a decrease of blood pressure within 3 minutes [1].

In addition, many previous studies investigated the relationship between OH and falling only with retrospective fall data [24, 55, 56], leading to several forms of bias, such as confounding or recall bias [55]. Considering the results of all above-mentioned studies together, none of the studies showed a clear relationship between OH and falling. Several reviews described a theoretical relationship [7, 12], but could not perform a meta-analysis considering the low number of studies [8]. By using IPD, we were able to investigate the relationship between OH and falling in a meta-analysis and adjust for the most important confounders.

trengths and i itationsDespite the small number of studies, the current study is the only study, to the best of our knowledge that used the consensus definition of OH, which included a representative group of elderly, and adjusted for important confounders. No heterogeneity was observed and follow-up of all included studies was sufficient.We acknowledge several limitations of our analysis. Firstly, only a small number of studies, and a limited number of patients were included in the meta-analysis. Unfortunately, one study did not report the time to first fall incident [11] and therefore the Cox regression analysis was only performed on two out of three studies [10, 23]. In addition, both results of the one-stage and two-stage methods showed wide confidence intervals that indicate additional information is needed for more precise estimates [21].

R1

R2

R3

R4

R5

R6

R7

R8

R9

R10

R11

R12

R13

R14

R15

R16

R17

R18

R19

R20

R21

R22

R23

R24

R25

R26

R27

R28

R29

R30

R31

R32

R33

R34

R35

R36

R37

R38

R39

Chapter 5

78

Another important limitation is the lack of subgroup and meta-regression analyses. Because of the limited number of studies included into the meta-analysis no subgroup or meta-regression analyses could be performed. A subgroup analysis according to Parkinson’s disease or dementia would have been useful since both could have influenced the relationship with falling [23]. Subgroup analyses of dementia were not performed because it was not prespecified and only one [23] of the three studies included patients with dementia or Parkinson’s disease. For the other two studies no data regarding dementia or Parkinson’s disease were available [10, 11]. In addition, subgroup analysis regarding study quality was preferred. However, all three included studies scored equal, allowing no subgroup analysis. Furthermore, OH was assessed within 3 min of postural change, thus not taking into account ‘delayed OH’. Also, we did not perform an analysis regarding recurrent falling. Only falling (yes/no) or time to first fall incident were analysed. Finally, both one and two-stage IPD was performed, leading to different results. It is known that one or two-stage IPD occasionally could lead to different conclusions [19].

CONCLUSIONS

This IPD meta-analysis of prospective observational studies showed a significant relationship between OH and time to first fall incident. However, since the small number of prospective studies included in present meta-analysis, more of these studies are needed for a more precise estimate of the relationship between OH and falling.

R1

R2

R3

R4

R5

R6

R7

R8

R9

R10

R11

R12

R13

R14

R15

R16

R17

R18

R19

R20

R21

R22

R23

R24

R25

R26

R27

R28

R29

R30

R31

R32

R33

R34

R35

R36

R37

R38

R39


79

5

REFERENCES

1. Lahrmann H, Cortelli P, Hilz M, Mathias CJ, Struhal W, Tassinari M. EFNS guidelines on the diagnosis and management of orthostatic hypotension. European journal of neurology: the official journal of the European Federation of Neurological Societies. 2006; 13 (9):930-6.

2. Verwoert GC, Mattace-Raso FU, Hofman A, Heeringa J, Stricker BH, Breteler MM, et al. Orthostatic hypotension and risk of cardiovascular disease in elderly people: the Rotterdam study. Journal of the American Geriatrics Society. 2008; 56 (10):1816-20.

3. Xin W, Lin Z, Mi S. Orthostatic hypotension and mortality risk: a meta-analysis of cohort studies. Heart. 2014; 100 (5):406-13.

4. Masaki KH, Schatz IJ, Burchfiel CM, Sharp DS, Chiu D, Foley D, et al. Orthostatic hypotension predicts mortality in elderly men: the Honolulu Heart Program. Circulation. 1998; 98 (21):2290-5.

5. Rose KM, Tyroler HA, Nardo CJ, Arnett DK, Light KC, Rosamond W, et al. Orthostatic hypotension and the incidence of coronary heart disease: the Atherosclerosis Risk in Communities study. American journal of hypertension. 2000; 13 (6 Pt 1):571-8.

6. Scheffer AC, van Hensbroek PB, van Dijk N, Luitse JS, Goslings JC, Luigies RH, et al. Risk factors associated with visiting or not visiting the accident & emergency department after a fall. BMC health services research. 2013; 13:286.

7. Shaw BH, Claydon VE. The relationship between orthostatic hypotension and falling in older adults. Clinical autonomic research : official journal of the Clinical Autonomic Research Society. 2014; 24 (1):3-13.

8. Angelousi A, Girerd N, Benetos A, Frimat L, Gautier S, Weryha G, et al. Association between orthostatic hypotension and cardiovascular risk, cerebrovascular risk, cognitive decline and falls as well as overall mortality: a systematic review and meta-analysis. Journal of hypertension. 2014; 32 (8):1562-71; discussion 71.

9. Graafmans WC, Ooms ME, Hofstee HM, Bezemer PD, Bouter LM, Lips P. Falls in the elderly: a prospective study of risk factors and risk profiles. American journal of epidemiology. 1996; 143 (11):1129-36.

10. Gangavati A, Hajjar I, Quach L, Jones RN, Kiely DK, Gagnon P, et al. Hypertension, orthostatic hypotension, and the risk of falls in a community-dwelling elderly population: the maintenance of balance, independent living, intellect, and zest in the elderly of Boston study. Journal of the American Geriatrics Society. 2011; 59 (3):383-9.

11. Heitterachi E, Lord SR, Meyerkort P, McCloskey I, Fitzpatrick R. Blood pressure changes on upright tilting predict falls in older people. Age and ageing. 2002; 31 (3):181-6.

12. Ambrose AF, Paul G, Hausdorff JM. Risk factors for falls among older adults: A review of the literature. Maturitas. 2013; 75 (1):51-61.

13. Ooi WL, Hossain M, Lipsitz LA. The association between orthostatic hypotension and recurrent falls in nursing home residents. The American journal of medicine. 2000; 108 (2):106-11.

14. Rubenstein LZ, Josephson KR, Robbins AS. Falls in the nursing home. Annals of internal medicine. 1994; 121 (6):442-51.

15. Moher D, Liberati A, Tetzlaff J, Altman DG, Group P. Preferred reporting items for systematic reviews and meta-analyses: the PRISMA statement. PLoS medicine. 2009; 6 (7):e1000097.

16. Wells GA SB, O’Connell D, Peterson J, Welch V, Losos M, Tugwell P,. The Newcastle-Ottowa Scale (NOS) for assessing the quality of nonrandomised studies in meta-analyses. 2010 [cited; Available from: http://www.ohri.ca/programs/clinical_epidemiology/oxford.asp

17. Bekkering GE, Cools F, Aertgeerts B, de Backer TL, Kellen E, van der Stichele RH, et al. [Meta-analyses based on individual patient data (IPD)]. Nederlands tijdschrift voor geneeskunde. 2009; 153 (9):402-6.

R1

R2

R3

R4

R5

R6

R7

R8

R9

R10

R11

R12

R13

R14

R15

R16

R17

R18

R19

R20

R21

R22

R23

R24

R25

R26

R27

R28

R29

R30

R31

R32

R33

R34

R35

R36

R37

R38

R39

Chapter 5

80

18. Simmonds MC, Higgins JP, Stewart LA, Tierney JF, Clarke MJ, Thompson SG. Meta-analysis of individual patient data from randomized trials: a review of methods used in practice. Clinical trials (London, England). 2005; 2 (3):209-17.

19. Debray TP, Moons Kg Fau - Abo-Zaid GMA, Abo-Zaid Gm Fau - Koffijberg H, Koffijberg H Fau - Riley RD, Riley RD. Individual participant data meta-analysis for a binary outcome: one-stage or two-stage? ; (1932-6203 (Electronic)).

20. Bowden J, Tierney JF, Simmonds M, Copas AJ, Higgins JP. Individual patient data meta-analysis of time-to-event outcomes: one-stage versus two-stage approaches for estimating the hazard ratio under a random effects model. Research synthesis methods. 2011; 2 (3):150-62.

21. Higgins JPT, Green S, Cochrane Collaboration. Cochrane handbook for systematic reviews of interventions: version 5.1.0. The Cochrane Collaboration. http://www.cochrane-handbook.org. Updated March 2011. Cochrane book series. Chichester, England ; Hoboken, NJ: Wiley-Blackwell; 2011. p. xxi, 649 p.

22. Alemdarolu E, Ucan H, Topcuolu AM, Sivas F. In-hospital predictors of falls in community-dwelling individuals after stroke in the first 6 months after a baseline evaluation: A prospective cohort study. Archives of Physical Medicine and Rehabilitation. 2012; 93 (12):2244-50.

23. Allan LM, Ballard CG, Rowan EN, Kenny RA. Incidence and prediction of falls in dementia: a prospective study in older people. PLoS One. 2009; 4 (5):e5521.

24. Blumenthal MD, Davie JW. Dizziness and falling in elderly psychiatric outpatients. The American journal of psychiatry. 1980; 137 (2):203-6.

25. Campbell AJ, Borrie MJ, Spears GF. Risk factors for falls in a community-based prospective study of people 70 years and older. Journal of gerontology. 1989; 44 (4):M112-7.

26. Chen YC, Chien SF, Chen LK. Risk factors associated with falls among Chinese hospital inpatients in Taiwan. Archives of gerontology and geriatrics. 2009; 48 (2):132-6.

27. Cleinman A, Griswold ME, Simonsick E, Meeks WM, Gregg K, Ferrucci L, et al. Falls and orthostatic hypotension: Re-examining limits. Journal of the American Geriatrics Society. 2012; 60:S11.

28. Davies AJ, Kenny RA. Falls presenting to the accident and emergency department: types of presentation and risk factor profile. Age and ageing. 1996; 25 (5):362-6.

29. Diaz Gravalos GJ, Gil Vazquez C, Andrade Pereira V, Alonso Payo R, Alvarez Araujo S, Reinoso Hermida S. [Risk factors for falls amongst older people living in nursing home. A cohort study]. Revista espanola de geriatria y gerontologia. 2009; 44 (6):301-4.

30. Gangavati A. Erratum: Hypertension, orthostatic hypotension, and the risk of falls in a community-dwelling elderly population: The maintenance of balance, independent living, intellect, and zest in the elderly of Boston study (Journal of the American Geriatrics Society (2011) 59 (383-389)). Journal of the American Geriatrics Society. 2011; 59 (5):960.

31. Gangavati A, Hajjar I, Quach L, Jones RN, Kiely DK, Gagnon P, et al. Hypertension, orthostatic hypotension and the risk of falls in a community-dwelling elderly population: The MOBILIZE Boston study. Clinical Autonomic Research. 2010; 20 (5):293-4.

32. Graafmans WC, Ooms ME, Hofstee HMA, Bezemer PD, Bouter LM, Lips P. Falls in the elderly: A prospective study of risk factors and risk profiles. American Journal of Epidemiology. 1996; 143 (11):1129-36.

33. Kamel HK, Iqbal MA, Malekgoudarzi B. Postprandial hypotension and relation to falls in institutionalized elderly persons. Annals of internal medicine. 2001; 135 (4):302.

34. Kario K, Tobin JN, Wolfson LI, Whipple R, Derby CA, Singh D, et al. Lower standing systolic blood pressure as a predictor of falls in the elderly: A community-based prospective study. Journal of the American College of Cardiology. 2001; 38 (1):246-52.

35. Kikuchi R, Kozaki K, Nakamura T, Toba K. [Muscle and bone health as a risk factor of fall among the elderly. An approach to identify high-risk fallers by risk assessment]. Clinical calcium. 2008; 18 (6):784-7.

R1

R2

R3

R4

R5

R6

R7

R8

R9

R10

R11

R12

R13

R14

R15

R16

R17

R18

R19

R20

R21

R22

R23

R24

R25

R26

R27

R28

R29

R30

R31

R32

R33

R34

R35

R36

R37

R38

R39


81

5

36. Kruse WH. Falls in community-dwelling older persons. Journal of the American Geriatrics Society. 1996; 44 (8):1010.

37. Kwok T, Liddle J, Hastie IR. Postural hypotension and falls. Postgraduate medical journal. 1995; 71 (835):278-80.

38. Liu BA, Topper AK, Reeves RA, Gryfe C, Maki BE. Falls among older people: relationship to medication use and orthostatic hypotension. Journal of the American Geriatrics Society. 1995; 43 (10):1141-5.

39. Luukinen H, Koski K, Kivela SL, Laippala P. Social status, life changes, housing conditions, health, functional abilities and life-style as risk factors for recurrent falls among the home-dwelling elderly. Public health. 1996; 110 (2):115-8.

40. Maurer MS, Burcham J, Cheng H. Diabetes mellitus is associated with an increased risk of falls in elderly residents of a long-term care facility. The journals of gerontology Series A, Biological sciences and medical sciences. 2005; 60 (9):1157-62.

41. Maurer MS, Cohen S, Cheng H. The degree and timing of orthostatic blood pressure changes in relation to falls in nursing home residents. Journal of the American Medical Directors Association. 2004; 5 (4):233-8.

42. Oliveira Guerra R, Villaverde Gutierrez C, Coelho Bernardo G, Tirado B, Ruiz CA, Ruiz Villaverde R, et al. Intrinsical risk factors of fall on elderly institutionalized patients. Geriatrika. 2003; 19 (5):17-20.

43. Passant U, Warkentin S, Gustafson L. Orthostatic hypotension and low blood pressure in organic dementia: a study of prevalence and related clinical characteristics. International journal of geriatric psychiatry. 1997; 12 (3):395-403.

44. Radtke A, Lempert T, von Brevern M, Feldmann M, Lezius F, Neuhauser H. Prevalence and complications of orthostatic dizziness in the general population. Clinical autonomic research : official journal of the Clinical Autonomic Research Society. 2011; 21 (3):161-8.

45. Rudzinska M, Bukowczan S, Stozek J, Zajdel K, Mirek E, Chwala W, et al. Causes and consequences of falls in Parkinson disease patients in a prospective study. Neurologia i neurochirurgia polska. 2013; 47 (5):423-30.

46. Rutan GH, Hermanson B, Bild DE, Kittner SJ, LaBaw F, Tell GS. Orthostatic hypotension in older adults. The Cardiovascular Health Study. CHS Collaborative Research Group. Hypertension. 1992; 19 (6 Pt 1):508-19.

47. Ruthazer R, Lipsitz LA. Antidepressants and falls among elderly people in long-term care. American Journal of Public Health. 1993; 83 (5):746-9.

48. Stone CA, Lawlor PG, Savva GM, Bennett K, Kenny RA. Prospective study of falls and risk factors for falls in adults with advanced cancer. Journal of Clinical Oncology. 2012; 30 (17):2128-33.

49. Swanenburg J, de Bruin ED, Uebelhart D, Mulder T. Falls prediction in elderly people: A 1-year prospective study. Gait and Posture. 2010; 31 (3):317-21.

50. Valvanne J, Sorva A, Erkinjuntti T, Tilvis R. The occurrence of postural hypotension in age cohorts of 75, 80 and 85 years. A population study. Archives of Gerontology and Geriatrics. 1991; 12 (SUPPL. 2):421-4.

51. Van Helden S, Wyers CE, Dagnelie PC, Van Dongen MC, Willems G, Brink PRG, et al. Risk of falling in patients with a recent fracture. BMC Musculoskeletal Disorders. 2007; 8.

52. Yiu D, Gieroba Z, Hall K, Mackintosh S, Russell A, Mangoni AA. An early drop in systolic blood pressure predicts a positive tilt table test in elderly patients with unexplained falls or syncope. Age and ageing. 2006; 35 (2):183-5.

53. Green Ms Fau - Symons MJ, Symons MJ. A comparison of the logistic risk function and the proportional hazards model in prospective epidemiologic studies. (0021-9681 (Print)).

54. Hosmer DW LS. Applied survival analysis: Regression modeling of time to event data.: Wiley-Interscience; 2008.

R1

R2

R3

R4

R5

R6

R7

R8

R9

R10

R11

R12

R13

R14

R15

R16

R17

R18

R19

R20

R21

R22

R23

R24

R25

R26

R27

R28

R29

R30

R31

R32

R33

R34

R35

R36

R37

R38

R39

Chapter 5

82

55. Romero-Ortuno R, Cogan L, Foran T, Kenny RA, Fan CW. Continuous noninvasive orthostatic blood pressure measurements and their relationship with orthostatic intolerance, falls, and frailty in older people. Journal of the American Geriatrics Society. 2011; 59 (4):655-65.

56. van der Velde N, van den Meiracker AH, Stricker BH, van der Cammen TJ. Measuring orthostatic hypotension with the Finometer device: is a blood pressure drop of one heartbeat clinically relevant? Blood pressure monitoring. 2007; 12 (3):167-71.

R1

R2

R3

R4

R5

R6

R7

R8

R9

R10

R11

R12

R13

R14

R15

R16

R17

R18

R19

R20

R21

R22

R23

R24

R25

R26

R27

R28

R29

R30

R31

R32

R33

R34

R35

R36

R37

R38

R39


83

5

SUPPLEMENTAL DATA

Appendix 1: Search Strategy((((“hypotension, orthostatic”[MeSH Terms] OR (“hypotension”[All Fields] AND “orthostatic”[All Fields]) OR “orthostatic hypotension”[All Fields] OR (“orthostatic”[All Fields] AND “hypotension”[All Fields])) OR (“hypotension, orthostatic”[MeSH Terms] OR (“hypotension”[All Fields] AND “orthostatic”[All Fields]) OR “orthostatic hypotension”[All Fields] OR (“postural”[All Fields] AND “hypotension”[All Fields]) OR “postural hypotension”[All Fields])) OR orthostatic[All Fields]) OR “hypotension, orthostatic”[MeSH Terms]) AND ((((((“accidental falls”[MeSH Terms] OR (“accidental”[All Fields] AND “falls”[All Fields]) OR “accidental falls”[All Fields] OR “falls”[All Fields]) OR falling[All Fields]) OR (recurrent[All Fields] AND (“accidental falls”[MeSH Terms] OR (“accidental”[All Fields] AND “falls”[All Fields]) OR “accidental falls”[All Fields] OR “falls”[All Fields]))) OR (“accidental falls”[MeSH Terms] OR (“accidental”[All Fields] AND “falls”[All Fields]) OR “accidental falls”[All Fields])) OR fall-risk[All Fields]) OR “accidental falls”[MeSH Terms])

Appendix Table 1. Newcastle-Ottawa scale for study quality for all three included studies.

Study Selection Comparability Outcome/exposure1 2 3 4 1 1 2 3

Allan et al.[23] Ë Ë ËË Ë ËGangavati et al. [10] Ë Ë Ë ËË ËHeitterachi et al. [11] Ë Ë Ë ËË Ë

NOS: Newcastle-Ottawa Scale: on a 9-point scale, the NOS evaluates the quality of observational studies on three broad categories: selection of the study groups (maximum of 4 stars); comparability of the groups (maximum of 2 stars); and ascertainment of the outcome of interest (maximum of 3 stars).

Appendix Table 2. Subgroup analyses. Adjusted odds ratios and Hazard ratios for the effect of orthostatic hypotension on the risk of falling with individual patient data in a ‘one-stage’ model. The odds ratios can be interpreted as a measure of the association of OH to falling (the dependent variables).

Subgroup Odds ratio (95% CI) Hazard ratio (95% CI)OH (model 1) Diabetic patients 0.87 (0.44-1.73) 1.15 (0.72-1.86)OH (model 2) Diabetic patients 0.85 (0.42-1.69) 1.17 (0.72-1.90) OH (model 3) Diabetic patients 0.97 (0.44-2.15) 1.38 (0.80-2.38)OH (model 1) Non-diabetic patients 0.89 (0.65-1.21) 1.29 (1.05-1.59)OH (model 2) Non-diabetic patients 0.90 (0.66-1.24) 1.29 (1.05-1.58)OH (model 3) Non-diabetic patients 1.26 (0.88-1.81) 1.53 (1.21-1.93)OH (model 1) Hypertension 1.02 (0.57-1.85) 1.06 (0.68-1.64)OH (model 2) Hypertension 0.94 (0.51-1.71) 0.99 (0.64-1.55)OH (model 3) Hypertension 1.01 (0.53-1.93) 1.05 (0.65-1.71)OH (model 1) No Hypertension 0.98 (0.51-1.86) 0.92 (0.60-1.41)OH (model 2) No Hypertension 0.99 (0.52-1.91) 0.93 (0.60-1.43)OH (model 3) No Hypertension 1.07 (0.54-2.16) 0.99 (0.63-1.58)

OH = orthostatic hypotension. DM = Diabetes Mellitus.

CHAPTER 6The association between orthostatic hypotension, falling

and successful rehabilitation in a nursing home population

Published as: Hartog LC, Cimzar-Sweelssen M, Knipscheer A, Groenier KH, Kleefstra N, Bilo

HJG , Van Hateren KJJ. The association between Orthostatic hypotension,

Falling and Successful Rehabilitation in a nursing home population. Archives of

Gerontology and Geriatrics 2015; 61 (2):190-6.

R1

R2

R3

R4

R5

R6

R7

R8

R9

R10

R11

R12

R13

R14

R15

R16

R17

R18

R19

R20

R21

R22

R23

R24

R25

R26

R27

R28

R29

R30

R31

R32

R33

R34

R35

R36

R37

R38

R39

Chapter 6

86

ABSTRACT

urpose o the stud : Our objectives were to identify the prevalence of orthostatic hypotension (OH) in frail, elderly nursing home residents, and assess its possible association with falling and chances of successful rehabilitation.

Materials and Methods: A prospective observational cohort study. A total of 290 patients participated in this study, of which 128 were admitted to the rehabilitation department. OH was defined as a drop in systolic blood pressure of > 20 mmHg and diastolic blood pressure of > 10 mmHg after postural change within 3 minutes. The analyses regarding falling and successful rehabilitation were only performed in the rehabilitation group. Multivariate binary logistic regression analyses were used to describe risk factors related with falling. Cox proportional hazard modeling was used to investigate the relation between OH and the time to successful rehabilitation.

esults: The prevalence of OH in the studied nursing home population was 36.6% (95%CI (confidence interval): 31.1%-42.1%). The prevalence varied from 28.6% (95%CI:16.8%-40.4%) in somatic patients, 36.7% (95%CI: 28.4%-45.1) in rehabilitation patients, to 40.6% (95%CI: 31.3%-50.0%) in psychogeriatric patients. The association between orthostatic hypotension and previous falling was not significant; Odds ratio 0.66 (95%CI: 0.30-1.48). The Hazard ratio of the relationship between OH and successful rehabilitation was 2.88 (95% CI:1.77-4.69).

onclusions: OH is highly prevalent in nursing home residents. Surprisingly, patients with OH were found to have a higher chance of successful rehabilitation compared to patients without OH. If confirmed in other studies, these results may change our view of the implications of OH.

R1

R2

R3

R4

R5

R6

R7

R8

R9

R10

R11

R12

R13

R14

R15

R16

R17

R18

R19

R20

R21

R22

R23

R24

R25

R26

R27

R28

R29

R30

R31

R32

R33

R34

R35

R36

R37

R38

R39

rthostatic h potension alling and rehabilitation

87

6

INTRODUCTION

Orthostatic hypotension (OH) is frequently seen in the elderly population, and its prevalence increases with advancing age [1]. The etiology of OH is multifactorial; decrease of baroreceptor sensitivity, pure autonomic failure, and use of different medications like antihypertensive medications or antipsychotics are all considered being possible causes [2, 3]. Presence of OH increases the risk of cardiovascular disease and all-cause mortality in elderly people [4-7]. Furthermore, morbidity associated with OH is not limited to cardiovascular disease; some studies strongly suggest OH to be associated with falls, which in turn can lead to serious morbidity [8-10]. Most studies, which reported a positive association between falls and OH, were performed in nursing homes [11-13]. As the prevalence of OH in nursing home populations is high, and associations with fall incidents were observed in these populations, OH may be an important prognostic factor for chances of rehabilitation in nursing home patients. Although a few studies reported prevalence of OH in frail, elderly nursing home residents [11, 14, 15], the prevalence in these studies varied widely from 18-50% and the actual prevalence in different patient categories, such as psychogeriatric, somatic and rehabilitation patients, in nursing homes is unknown. In the current observational study we aimed to investigate the prevalence of OH and its association with previous fall incidents in different nursing home patient groups. Furthermore, we aimed to explore the influence of OH on the chances of successful rehabilitation. We hypothesized that the presence of OH would negatively influence the time to successful rehabilitation.


tud populationFor this prospective observational cohort study, patients were recruited from a nursing home facility in the northeastern region of the Netherlands (TriviumMeulenbeltZorg, Hengelo). Recruitment and all study procedures for patients on the somatic and psychogeriatric departments took place between September 2010 and December 2010. For patients on the rehabilitation department recruitment and all study procedures took place between September 2010 and December 2011, mostly within the first weeks of admission. Patients at the psychogeriatric department were diagnosed with severe to very severe dementia and behavioral disorders. Patients at the somatic department received prolonged or permanent care whereas patients at the rehabilitation department were rehabilitated to return to home as soon as possible. Exclusion criteria were a life expectancy of less than 4 weeks or admission to the hospice department.

R1

R2

R3

R4

R5

R6

R7

R8

R9

R10

R11

R12

R13

R14

R15

R16

R17

R18

R19

R20

R21

R22

R23

R24

R25

R26

R27

R28

R29

R30

R31

R32

R33

R34

R35

R36

R37

R38

R39

Chapter 6

88

ata collectionBaseline data involved demographic characteristics, a full medical history including a history of cardiovascular disease, diabetes mellitus, hypertension, falls in the previous year (yes of no), and medication use. All patients were subjected to questionnaires and blood pressure measurements. The activities of daily living were measured with the Barthel-Index [16]. Body mass index was calculated by measuring body weight and height. Blood pressure was measured following a standardized protocol, using an automated sphygmomanometer (Omron M6) [17]. Trained medical staff performed all tests. Blood pressure was measured two times in supine position after 5 minutes of rest, and two times each at 1 and 3 minutes after postural change. The forearm of the patient was supported at heart level during the measurements in upright position [18]. The postural change was from supine to standing position, with the exception of patients who were unable to stand. For these patients the postural change was from lying to sitting position. OH was defined as a drop in systolic blood pressure (SBP) of > 20 mmHg or diastolic blood pressure (DBP) of > 10 mmHg after postural change compared to the mean value of the baseline measurements in supine position [1]. Characteristic symptoms of OH like light-headedness, syncope, or dizziness after postural change were questioned and the combination of OH and orthostatic complaints was described as symptomatic OH. Successful rehabilitation was defined as discharge to patients’ own homes or to an adapted home for the elderly, where they functioned self-reliant.

tatistical anal sesContinuous variables are presented as mean and standard deviation for normally distributed variables, or as median and interquartile range for non-normally distributed variables. Univariate binary logistic regression analyses were performed to assess the association of OH, orthostatic complaints or symptomatic OH and the baseline characteristics. Subsequently, multivariate binary logistic regression analyses were used to describe risk factors associated with falling. In the multivariate binary logistic regression analyses we adjusted for age, gender, body mass index (BMI), a history of diabetes mellitus, the score on the Barthel questionnaire, previous macrovascular complications, and the use of antihypertensive medication. Cox proportional hazard modeling was used to investigate the relation between OH, orthostatic complaints, or symptomatic OH and successful rehabilitation. Since not all blood pressure measurements were performed in the first week of rehabilitation; time to successful rehabilitation started on the day of OH blood pressure measurement. In the multivariate Cox regression models we adjusted for the same variables as in the multivariate binary logistic analyses, including mean systolic blood pressure. The Schoenfeld residual plots were inspected for each predictor variable to check the assumption of proportional hazards. The analyses regarding falling and successful rehabilitation were only performed in the rehabilitation group. P values less than 0.05 were considered statistically significant.

R1

R2

R3

R4

R5

R6

R7

R8

R9

R10

R11

R12

R13

R14

R15

R16

R17

R18

R19

R20

R21

R22

R23

R24

R25

R26

R27

R28

R29

R30

R31

R32

R33

R34

R35

R36

R37

R38

R39


89

6

All statistical analyses were performed using SPSS software (version 22). The ‘Strengthening the Reporting of Observational studies in Epidemiology’ (STROBE) statement was used to describe this observational cohort study [19].

thical approval and linical rial registrationThis study was performed in accordance with the Declaration of Helsinki. According to Dutch guidelines this study did not fall under the scope of the Medical Research Involving Human Subjects Act, and therefore this study did not need a formal approval of an accredited medical ethics committee. Written informed consent was obtained for all patients by the participating medical doctor or nurse. All data were analyzed anonymously. The study was registered on ClinicalTrials.gov (NCT01362751).

RESULTS

A total of 290 patients was included in this cohort, 106 patients at the psychogeriatric department, 56 patients at the somatic department, and 128 patients at the rehabilitation department. The baseline characteristics are presented in table 1. Mean age of the total study population was 80.9 (+ 9.9) years. OH was present in 106 out of 290 patients, resulting in a prevalence of 36.6% (95%CI (confidence interval): 31.1%-42.1%). The prevalence ranged from 28.6% (95%CI: 16.8%-40.4%) on the somatic department, to 36.7% (95% CI: 28.4% to 45.1) and 40.6% (95% CI: 31.3% to 50.0%) on the rehabilitation and the psychogeriatric departments, respectively. The intended postural change could be performed from lying to standing in 35.7% of the somatic patients, 86.7% of the rehabilitation patients, and 66.0% of the psychogeriatric patients. The remaining patients performed the postural change from lying to sitting. The results of the univariate regression analyses are presented in appendix table A1 and A2. Both age and mean supine systolic blood pressure were related to OH. Mean systolic blood pressure was higher in the OH group. A total of 59 (20.3%) patients had fallen at least once in the previous year. A correlation between falls in the last year and symptomatic OH was seen, whereas the presence of OH and orthostatic complaints were not correlated.

R1

R2

R3

R4

R5

R6

R7

R8

R9

R10

R11

R12

R13

R14

R15

R16

R17

R18

R19

R20

R21

R22

R23

R24

R25

R26

R27

R28

R29

R30

R31

R32

R33

R34

R35

R36

R37

R38

R39

Chapter 6

90

Tabl

e 1.

Bas

elin

e ch

arac

teri

stics

tota

l pop

ulati

on a

nd d

iffer

ent g

roup

s of

pati

ents

.

Tota

lPs

ycho

geri

atri

c de

part

men

tSo

mati

c de

part

men

tRe

habi

litati

on d

epar

tmen

tCh

arac

teri

stic

N=2

90N

=106

N=5

6N

=128

Dem

ogra

phic

sA

ge

80

.8 (9

.9)

82.9

(8.9

)77

.1 (1

4.1)

80.7

(7.9

)Fe

mal

e G

ende

r20

6 (7

1.0)

78 (7

3.6)

36 (6

4.3)

92 (7

1.9)

Mea

n bo

dy m

ass

inde

x, k

g/m

2

26.3

(5.4

)25

.9 (5

.2)

28.2

(5.8

)25

.9 (5

.3)

Hyp

erte

nsio

n 22

6 (7

7.9)

79 (7

4.5)

42 (7

5)10

5 (8

2.0)

His

tory

of C

VD13

1 (4

5.2)

48 (4

5.3)

32 (5

7.1)

51 (3

9.8)

Dia

bete

s m

ellit

us10

4 (3

5.0)

30 (2

8.3)

19 (3

3.9)

55 (4

3.0)

Dem

entia

121

(41.

7)10

0 (9

4.3)

12 (2

1.4)

9 (7

.0)

Curr

rent

sm

oker

32

(11.

0)8

(7.5

)12

(21.

4)12

(9.4

)

Mea

sure

men

tsCo

nsum

ption

mea

l or

drin

k 19

4 (6

6.9)

77 (7

2.6)

30 (5

3.6)

87 (6

8.0)

Mea

n SB

P ly

ing

147

(22)

148.

5 (2

1.0)

147

(25)

146.

5 (2

2.2)

Mea

n D

BP ly

ing

74 (1

1)76

.6 (1

1.4)

75 (1

3)72

.4 (9

.8)

Mea

n pu

lse

freq

uenc

y 74

(13)

72.4

(14.

1)74

(13)

75.4

(12.

3)Fa

lls la

st y

ear

59 (2

0.3)

8 (7

.5)

4 (7

.1)

47 (3

6.7)

Ort

host

atic

hypo

tens

ion

106

(36.

6)43

(40.

6)16

(28.

6)47

(36.

7)O

rtho

stati

c co

mpl

aint

s71

(24.

7)26

(24.

5)11

(19.

6)34

(26.

6)Sc

ore

Bart

hel q

uesti

onna

ire

10 (6

-15)

9 (4

-14)

6 (2

-12)

13 (9

-16)

Med

icati

onM

ean

num

ber

of a

gent

s 8.

7 (3

.5)

8.5

(3.0

)9.

5 (4

.2)

8.6

(3.6

)A

ntihy

pert

ensi

ve m

edic

ation

- D

iure

tics

- Be

ta b

lock

ers

- Ca

lciu

m c

hann

el b

lock

ers

- AC

E in

hibi

tors

194

(66.

9)12

8 (4

4.1)

85 (2

9.3)

42 (1

4.5)

100

(34.

5)

68 (6

4.2)

43 (4

0.6)

23 (2

1.7)

13 (1

2.3)

42 (3

9.6)

35 (6

2.5)

23 (4

1.1)

17 (3

0.4)

8 (1

4.3)

15 (2

6.8)

91 (7

1.1)

62 (4

8.4)

45 (3

5.2)

21 (1

6.4)

43 (3

3.6)

Benz

odia

zepi

nes

150

(51.

7)52

(49.

1)34

(60.

7)64

(50)

Anti

psyc

hotic

s59

(20.

3)34

(32.

1)11

(19.

6)14

(10.

9)A

ntide

pres

sant

s69

(23.

8)32

(30.

2)18

(32.

1)19

(14.

8)O

pioi

ds32

(11.

0)10

(9.4

)9

(16.

1)13

(10.

2)

Dat

a ar

e m

eans

(±

SD),

med

ians

(in

terq

uarti

le r

ange

) or

n (

%).

CVD

= C

ardi

o va

scul

ar d

isea

se. S

BP =

sys

tolic

blo

od p

ress

ure.

DBP

= d

iast

olic

blo

od

pre

ssu

re

.

R1

R2

R3

R4

R5

R6

R7

R8

R9

R10

R11

R12

R13

R14

R15

R16

R17

R18

R19

R20

R21

R22

R23

R24

R25

R26

R27

R28

R29

R30

R31

R32

R33

R34

R35

R36

R37

R38

R39


91

6

Tabl

e 1.

Bas

elin

e ch

arac

teri

stics

tota

l pop

ulati

on a

nd d

iffer

ent g

roup

s of

pati

ents

.

Tota

lPs

ycho

geri

atri

c de

part

men

tSo

mati

c de

part

men

tRe

habi

litati

on d

epar

tmen

tCh

arac

teri

stic

N=2

90N

=106

N=5

6N

=128

Dem

ogra

phic

sA

ge

80

.8 (9

.9)

82.9

(8.9

)77

.1 (1

4.1)

80.7

(7.9

)Fe

mal

e G

ende

r20

6 (7

1.0)

78 (7

3.6)

36 (6

4.3)

92 (7

1.9)

Mea

n bo

dy m

ass

inde

x, k

g/m

2

26.3

(5.4

)25

.9 (5

.2)

28.2

(5.8

)25

.9 (5

.3)

Hyp

erte

nsio

n 22

6 (7

7.9)

79 (7

4.5)

42 (7

5)10

5 (8

2.0)

His

tory

of C

VD13

1 (4

5.2)

48 (4

5.3)

32 (5

7.1)

51 (3

9.8)

Dia

bete

s m

ellit

us10

4 (3

5.0)

30 (2

8.3)

19 (3

3.9)

55 (4

3.0)

Dem

entia

121

(41.

7)10

0 (9

4.3)

12 (2

1.4)

9 (7

.0)

Curr

rent

sm

oker

32

(11.

0)8

(7.5

)12

(21.

4)12

(9.4

)

Mea

sure

men

tsCo

nsum

ption

mea

l or

drin

k 19

4 (6

6.9)

77 (7

2.6)

30 (5

3.6)

87 (6

8.0)

Mea

n SB

P ly

ing

147

(22)

148.

5 (2

1.0)

147

(25)

146.

5 (2

2.2)

Mea

n D

BP ly

ing

74 (1

1)76

.6 (1

1.4)

75 (1

3)72

.4 (9

.8)

Mea

n pu

lse

freq

uenc

y 74

(13)

72.4

(14.

1)74

(13)

75.4

(12.

3)Fa

lls la

st y

ear

59 (2

0.3)

8 (7

.5)

4 (7

.1)

47 (3

6.7)

Ort

host

atic

hypo

tens

ion

106

(36.

6)43

(40.

6)16

(28.

6)47

(36.

7)O

rtho

stati

c co

mpl

aint

s71

(24.

7)26

(24.

5)11

(19.

6)34

(26.

6)Sc

ore

Bart

hel q

uesti

onna

ire

10 (6

-15)

9 (4

-14)

6 (2

-12)

13 (9

-16)

Med

icati

onM

ean

num

ber

of a

gent

s 8.

7 (3

.5)

8.5

(3.0

)9.

5 (4

.2)

8.6

(3.6

)A

ntihy

pert

ensi

ve m

edic

ation

- D

iure

tics

- Be

ta b

lock

ers

- Ca

lciu

m c

hann

el b

lock

ers

- AC

E in

hibi

tors

194

(66.

9)12

8 (4

4.1)

85 (2

9.3)

42 (1

4.5)

100

(34.

5)

68 (6

4.2)

43 (4

0.6)

23 (2

1.7)

13 (1

2.3)

42 (3

9.6)

35 (6

2.5)

23 (4

1.1)

17 (3

0.4)

8 (1

4.3)

15 (2

6.8)

91 (7

1.1)

62 (4

8.4)

45 (3

5.2)

21 (1

6.4)

43 (3

3.6)

Benz

odia

zepi

nes

150

(51.

7)52

(49.

1)34

(60.

7)64

(50)

Anti

psyc

hotic

s59

(20.

3)34

(32.

1)11

(19.

6)14

(10.

9)A

ntide

pres

sant

s69

(23.

8)32

(30.

2)18

(32.

1)19

(14.

8)O

pioi

ds32

(11.

0)10

(9.4

)9

(16.

1)13

(10.

2)

Dat

a ar

e m

eans

(±

SD),

med

ians

(in

terq

uarti

le r

ange

) or

n (

%).

CVD

= C

ardi

o va

scul

ar d

isea

se. S

BP =

sys

tolic

blo

od p

ress

ure.

DBP

= d

iast

olic

blo

od

pre

ssu

re

.

revious all incidentsBecause 80% of all patients who had fallen at least once in the previous year were admitted at the rehabilitation ward, multivariate analyses on the risk of previous falls were only performed in the rehabilitation group (table 2). In the multivariate analyses no significant relation was found between OH or orthostatic complaints and reported previous falling. The relationship between symptomatic OH and previous falls was also not significant; adjusted Odds ratio 2.1 (95% CI: 0.7-6.1). Adjustment for the postural position (standing or sitting) did not influence the results (data not shown).

uccess ul rehabilitationA total of 128 patients were admitted to the rehabilitation department. During a median follow-up period of 2.6 months (IQR 1.2-4.6), 106 patients were successfully rehabilitated. 94 patients were discharged to their own homes, and 12 patients were discharged to an adapted home for the elderly. Two patients died during admission. Median time between admission to the rehabilitation department and blood pressure measurements was 1.1 month (IQR 0.5-2.5). Table 3 presents the results of the Cox regression analyses, and figure 1 shows the cumulative proportion of successful rehabilitation curve. The chance of successful rehabilitation was 188% (95%CI: 77%-469%) higher for patients with OH compared to those without, when adjusted for orthostatic complaints and selected confounders. Also without adjustment for orthostatic complaints the chance of successful rehabilitation was higher for patients with OH. This relation was also significant when it was adjusted only for age and gender (HR 1.71 [95%CI 1.14-2.59]). Higher scores on the Barthel questionnaire were also related to successful rehabilitation in all multivariate models. A history of diabetes mellitus was inversely related to successful rehabilitation. The plots of the Schoenfeld residues showed that the assumptions of proportional hazards were met. Some post-hoc analyses were performed. As the postural position (standing or sitting), reason for admission, and the discharge destination (own home or adapted home for the elderly) may have influenced the results, we performed additional cox regression analyses in which we additionally adjusted for these variables. Results did not relevantly change (data not shown).As blood pressure may be a marker of frailty in old age [20], we performed post-hoc analyses in which we tested for interaction with blood pressure by analyzing different systolic blood pressure subgroups (99-130 mmHg, 130-160 mmHg, and 160-220 mmHg). The HR’s for each subgroup were 2.95 (95%CI: 0.55-15.78), 3.04 (95% CI: 1.54-5.99, and 1.95 (95%CI: 0.48-8.012), respectively. OH was only significantly related to successful rehabilitation in the group with blood pressure values between 130-160 mmHg (p=0.001). Figure 1 shows the cumulative proportion of successful rehabilitation curves for the total cohort and each of the subgroups.

R1

R2

R3

R4

R5

R6

R7

R8

R9

R10

R11

R12

R13

R14

R15

R16

R17

R18

R19

R20

R21

R22

R23

R24

R25

R26

R27

R28

R29

R30

R31

R32

R33

R34

R35

R36

R37

R38

R39

Chapter 6

92

Tabl

e 2.

Adj

uste

d od

ds ra

tios

for

the

effec

t of c

linic

al v

aria

bles

on

the

risk

of p

revi

ous

fall

inci

dent

s in

the

reha

bilit

ation

pati

ent g

roup

(n=1

28).

The

odds

ratio

s ca

n be

inte

rpre

ted

as a

mea

sure

of t

he a

ssoc

iatio

n of

the

vari

ous

vari

able

s to

pre

viou

s fa

ll in

cide

nts

(the

dep

ende

nt v

aria

bles

).

Prev

ious

fall

inci

dent

s in

last

yea

r as

the

depe

nden

t var

iabl

eM

odel

1M

odel

2M

odel

3M

odel

4O

dds

ratio

(95%

CI)

Odd

s ra

tio (9

5% C

I)O

dds

ratio

(95%

CI)

OH

0.

66 (0

.26-

1.48

)-

0.63

(0.2

8-1.

43)

-O

rtho

stati

c co

mpl

aint

s-

1.22

(0.5

1-2.

92)

1.35

(0.5

5-3.

29)

-Sy

mpt

omati

c O

H

--

-2.

10 (0

.72-

6.14

)A

ge

1.02

(0.9

6-1.

07)

1.02

(0.9

7-1.

07)

1.02

(0.9

7-1.

07)

1.02

(0.9

7-1.

07)

Gen

der,

fem

ale

vs m

ale

0.55

(0.2

2-1.

38)

0.53

(0.2

1-1.

32)

0.55

(0.2

2-1.

39)

0.53

(0.2

1-1.

33)

Body

Mas

s In

dex

0.99

(0.9

1-1.

07)

0.99

(0.9

1-1.

07)

0.99

(0.9

1-1.

07)

0.99

(0.9

1-1.

07)

DM

, DM

vs

cont

rol

0.78

(0.3

5-1.

74)

0.79

(0.3

6-1.

76)

0.79

(0.3

5-1.

76)

0.79

(0.3

5-1.

76)

Scor

e Ba

rthe

l que

stion

naire

1.03

(0.9

5-1.

13)

1.04

(0.9

5-1.

13)

1.04

(0.9

5-1.

13)

1.04

(0.9

5-1.

13)

His

tory

of C

VD d

isea

se

0.43

(0.1

9-0.

98)

0.45

(0.2

0-1.

03)

0.46

(0.2

0-1.

05)

0.45

(0.2

0-1.

03)

Anti

hype

rten

sive

med

icati

on1.

02 (0

.41-

2.52

)0.

92 (0

.37-

2.28

)0.

97 (0

.39-

2.42

)0.

91 (0

.37-

2.22

)

Mod

el 1

was

adj

uste

d fo

r OH

, mod

el 2

for o

rtho

stati

c co

mpl

aint

s, m

odel

3 fo

r OH

and

ort

host

atic

com

plai

nts,

and

mod

el 4

for s

ympt

omati

c or

thos

tatic

hy

pote

nsio

n. A

ll m

odel

s w

ere

addi

tiona

lly a

djus

ted

for

the

follo

win

g po

ssib

le p

redi

ctor

s of

fall

inci

dent

s: a

ge, g

ende

r, BM

I, di

abet

es m

ellit

us, s

core

Ba

rthe

l que

stion

naire

, pre

viou

s m

acro

vas

cula

r com

plic

ation

s, a

nd u

se o

f anti

hype

rten

sive

med

icati

ons.

The

ORs

can

be

inte

rpre

ted

as a

mea

sure

of

the

asso

ciati

on o

f the

var

ious

var

iabl

es to

pre

viou

s fa

ll in

cide

nts.

OH

= o

rtho

stati

c hy

pote

nsio

n. C

VD =

Car

dio

vasc

ular

dis

ease

.

R1

R2

R3

R4

R5

R6

R7

R8

R9

R10

R11

R12

R13

R14

R15

R16

R17

R18

R19

R20

R21

R22

R23

R24

R25

R26

R27

R28

R29

R30

R31

R32

R33

R34

R35

R36

R37

R38

R39


93

6

Tabl

e 3.

Cox

regr

essi

on a

naly

ses;

haz

ard

ratio

s fo

r su

cces

sful

reha

bilit

ation

in th

e re

habi

litati

on p

atien

t gro

up (n

=128

).

Mod

el 1

Mod

el 2

Mod

el 3

Mod

el 4

HR

(95%

CI)

HR

(95%

CI)

HR

(95%

CI)

HR

(95%

CI)

OH

2.

65 (1

.69-

4.17

)-

2.88

(1.7

7-4.

69)

-

Ort

host

atic

com

plai

nts

-1.

10 (0

.69-

1.74

)0.

79 (0

.48-

1.28

)-

Sym

ptom

atic

OH

--

-1.

42 (0

.81-

2.51

)A

ge

1.01

(0.9

8-1.

04)

1.01

(0.9

8-1.

04)

1.01

(0.9

8-1.

04)

1.01

(0.9

8-1.

04)

Gen

der,

fem

ale

vs m

ale

1.37

(0.8

5-2.

19)

1.38

(0.8

6-2.

22)

1.37

(0.8

5-2.

18)

1.40

(0.8

7-2.

26)

BMI

1.05

(1.0

0-1.

09)

1.04

(1.0

0-1.

09)

1.05

(1.0

0-1.

10)

1.04

(0.9

9-1.

10)

Scor

e Ba

rthe

l que

stion

naire

1.16

(1.1

0-1.

23)

1.14

(1.0

7-1.

20)

1.16

(1.1

0-1.

23)

1.14

(1.0

8-1.

20)

Anti

hype

rten

sive

med

icati

on0.

95 (0

.57-

1.57

)1.

01 (0

.61-

1.69

)0.

96 (0

.57-

1.59

)1.

02 (0

.62-

1.70

)H

isto

ry o

f CVD

dis

ease

1.

39 (0

.89-

2.17

)1.

33 (0

.84-

2.10

)1.

39 (0

.89-

2.19

)1.

35 (0

.85-

2.12

)T2

DM

, DM

vs

cont

rol

0.60

(0.3

8-0.

94)

0.59

(0.3

8-0.

93)

0.61

(0.3

9-0.

95)

0.59

(0.3

7-0.

92)

Mea

n SB

P ly

ing

1.01

(0.9

9-1.

02)

1.01

(0.9

9-1.

02)

1.01

(0.9

9-1.

02)

1.01

(0.9

9-1.

02)

Haz

ard

ratio

s fo

r su

cces

sful

reh

abili

ation

. All

mod

els

wer

e ad

ditio

nally

adj

uste

d fo

r th

e fo

llow

ing

poss

ible

pre

dict

ors

of fa

ll in

cide

nts:

age

, gen

der,

BMI,

diab

etes

mel

litus

, sco

re B

arth

el q

uesti

onna

ire, p

revi

ous

mac

ro v

ascu

lar

com

plic

ation

s, m

ean

SBP

lyin

g, a

nd t

he n

umbe

r of

anti

hype

rten

sive

m

edic

ation

s us

ed. O

H =

ort

host

atic

hypo

tens

ion.

CVD

= C

ardi

o va

scul

ar d

isea

se. S

BP =

sys

tolic

blo

od p

ress

ure.

R1

R2

R3

R4

R5

R6

R7

R8

R9

R10

R11

R12

R13

R14

R15

R16

R17

R18

R19

R20

R21

R22

R23

R24

R25

R26

R27

R28

R29

R30

R31

R32

R33

R34

R35

R36

R37

R38

R39

Chapter 6

94

99

Figure 1. Cumulative proportion of successful rehabilitation (Survival curves) for the total cohort (A), subgroups with systolic blood pressure 99-130mmHg (B), 130-160 mmHg (C), and 160-220 mmHg (D).

Figure 1. Cumulative proportion of successful rehabilitation (Survival curves) for the total cohort (A),

subgroups with systolic blood pressure 99-130mmHg (B), 130-160 mmHg (C), and 160-220 mmHg (D).

DISCUSSION

In the present study, the overall prevalence of OH in nursing home residents was 36.6%, varying from 28.6% at the somatic department, to 36.7% at the rehabilitation department, and 40.6% at the psychogeriatric department. No significant correlation was found between reported previous falling and OH or orthostatic complaints. Finally, the chance of successful rehabilitation was higher for patients with OH compared to patients without.

revalenceThe prevalence of OH in this study proved to be higher than in community-dwelling elderly of the same age [6, 21, 22]. In contrast, the prevalence was lower compared to a prospective

R1

R2

R3

R4

R5

R6

R7

R8

R9

R10

R11

R12

R13

R14

R15

R16

R17

R18

R19

R20

R21

R22

R23

R24

R25

R26

R27

R28

R29

R30

R31

R32

R33

R34

R35

R36

R37

R38

R39


95

6

study by Ooi et al. amongst nursing home residents. In this study, OH in frail, elderly nursing home residents occurred in more than 50% [14]. The most important difference with the current study could be the timing of the measurements. In the study by Ooi et al. the prevalence of OH was highest before breakfast, while in the present study two third of the patients had already eaten when measured. The prevalence of OH was lowest in the group of somatic patients. Because only 35.7% of these patients could perform the postural change from lying to standing, the actual number of patients with OH could be underestimated in this group of patients.

revious all incidentsNo significant correlation was found between reported previous falling and OH or orthostatic complaints. In a previous review and meta-analysis, no significant conclusions could be drawn concerning the relationship between OH and falling [10, 23]. In a prospective study of over 800 elderly patients, OH was not associated with subsequent falls. However, patients with OH and a history of previous falling had an increased risk of recurrent falls [12]. Romero-Ortuno et al. did find a correlation between initial OH, defined as blood pressure decrease of more than 40 mmHg in systolic or more than 20 mmHg in diastolic blood pressure within 15 seconds after standing with symptoms, and falls [24]. Furthermore, a strong correlation was found between initial OH and orthostatic complaints, but not between OH and complaints. Possibly, the current definition of OH is incorrect with regard to the relationship between OH and falling. Orthostatic complains may be more relevant than OH on itself [25].

is o success ul rehabilitationA rather counterintuitive association between OH and the hazard of successful rehabilitation was observed in the present study. As far as we are aware, there is only one study that has described functional outcome of patients undergoing rehabilitation stratified according to OH. Functional outcome was not different between stroke patients with and without OH [26]. Cox regression analyses were not performed in this study. We hypothesized about possible explanations for the positive relationship between OH and successful rehabilitation. Firstly, we tested for interaction with blood pressure, as lower values are a marker of frailty in old age. Results of these additional analyses showed that OH was only significantly related to a higher chance of successful rehabilitation in patients with SBP values between 130 and 160 mmHg (possibly because of the higher number of events in this group). Still, this is no explanation for the remarkable association. As OH is related to reduced standing balance [27], the homes and home environment of patients with OH may be more adapted to their disabilities, or patients with OH may be used to a less active life, resulting in a quicker achievement of their rehabilitation targets. However, the Barthel index was positively correlated with successful rehabilitation. This argues against

R1

R2

R3

R4

R5

R6

R7

R8

R9

R10

R11

R12

R13

R14

R15

R16

R17

R18

R19

R20

R21

R22

R23

R24

R25

R26

R27

R28

R29

R30

R31

R32

R33

R34

R35

R36

R37

R38

R39

Chapter 6

96

the above-mentioned arguments as patients with higher scores on the Barthel index are more independent. Finally, the result could be a matter of coincidence as the study group is rather small.

i itationsThe current study has some limitations. The main limitation is its observational design; therefore, establishing a causal relation is not possible. Besides, two important types of bias occurred in this study; selection and recall bias. Patients in a nursing home on a rehabilitation department are mainly rehabilitating after a fracture or fall, which creates bias with respect to falling and OH. Recall problems were caused because all previous fall incidents were retrospectively assessed. The information regarding previous falls was based on questioning patients or participants. It is very likely that the actual number of patients with previous fall incidents was higher.Another limitation is the definition of successful rehabilitation. This was defined as discharge to home or a home for the elderly, where they functioned self-reliant. As a consequence, patients with a worse outcome after rehabilitation but with a highly adapted home environment (e.g. stairlift, homecare, meal service) may be quickly sent home. Finally, a limitation of this study is the variation of point of time of the blood pressure measurements. Not all measurements were performed in the first weeks of rehabilitation. The actual number of patients with OH could be underestimated by this variation. In the cox regression analyses, time to successful rehabilitation started on the day of OH blood pressure measurement. We also performed the same analyses stratified according to timing of blood pressure measurements: hazard ratios were 3.66 (95%CI 1.82-7.36) for measurements < 30 days and 2.60 (95%CI 1.34-5.02) for measurements > 30 days after admission. Because we are investigating OH in relation to successful rehabilitation, the time at risk started at the date of OH measurement.

CONCLUSIONS

More than one third of the nursing home residents were diagnosed with OH. There was no significant correlation between reported previous falls and OH or orthostatic complaints. Remarkably, patients with OH were found to have a higher chance of successful rehabilitation compared to patients without OH. Confirmation of this rather remarkable and counterintuitive finding in other studies is necessary, preferably performed in prospective studies.

R1

R2

R3

R4

R5

R6

R7

R8

R9

R10

R11

R12

R13

R14

R15

R16

R17

R18

R19

R20

R21

R22

R23

R24

R25

R26

R27

R28

R29

R30

R31

R32

R33

R34

R35

R36

R37

R38

R39


97

6

REFERENCES













13. Rubenstein LZ, Josephson KR, Robbins AS. Falls in the nursing home. Annals of internal medicine. 1994 Sep 15;121(6):442-51.







R1

R2

R3

R4

R5

R6

R7

R8

R9

R10

R11

R12

R13

R14

R15

R16

R17

R18

R19

R20

R21

R22

R23

R24

R25

R26

R27

R28

R29

R30

R31

R32

R33

R34

R35

R36

R37

R38

R39

Chapter 6

98

20. van Hateren KJ, Landman GW, Kleefstra N, et al. Lower blood pressure associated with higher mortality in elderly diabetic patients (ZODIAC-12). Age and ageing. 2010 Sep;39(5):603-9.






26. Kong KH, Chuo AM. Incidence and outcome of orthostatic hypotension in stroke patients undergoing rehabilitation. Archives of physical medicine and rehabilitation. 2003 Apr;84(4):559-62.


SUPP

LEM

ENTA

L D

ATA

Tabl

e A

1. U

niva

riat

e lo

gisti

c re

gres

sion

ana

lyse

s w

ith O

H, o

rtho

stati

c co

mpl

aint

s an

d sy

mpt

omati

c O

H a

s de

pend

ent v

aria

bles

. To

tal

OR

(95%

CI)

OR

(95%

CI)

OR

(95%

CI)

Char

acte

risti

cN

=290

rt

host

atic

copl

aint

spt

oati

c D

emog

raph

ics

Ag

e

80.8

(9.9

)1.

04 (1

.01-

1.07

)1.

00 (0

.97-

1.03

)1.

01 (0

.97-

1.05

)G

ende

r (%

fem

ale)

206/

290

(71.

0)1.

18 (0

.70-

1.99

)0.

79 (0

.43-

1.46

)1.

00 (0

.44-

2.28

)M

ean

body

mas

s in

dex,

kg/

m2

26.3

(5.4

)0.

99 (0

.94-

1.03

)1.

00 (0

.95-

1.05

)1.

02 (0

.96-

1.09

)H

yper

tens

ion

226/

290

(77.

9)1.

63 (0

.89-

2.99

)1.

75 (0

.86-

3.58

)1.

53 (0

.56-

4.17

)H

isto

ry o

f CVD

dis

ease

13

1/29

0 (4

5.2)

1.21

(0.7

5-1.

95)

0.64

(0.3

7-1.

11)

0.74

(0.3

5-1.

59)

Dia

bete

s m

ellit

us10

4/29

0 (3

5.0)

1.29

(0.7

9-2.

12)

1.42

(0.8

2-2.

46)

1.53

(0.7

2-3.

27)

Dem

entia

121/

290

(41.

7)1.

18 (0

.73-

1.92

)0.

82 (0

.47-

1.42

)0.

54 (0

.24-

1.21

)Cu

rrre

nt s

mok

er

32/2

90 (1

1.0)

0.54

(0.2

4-1.

26)

0.68

(0.2

7-1.

73)

0.53

(0.1

2-2.

32)

Mea

sure

men

tsCo

nsum

ption

mea

l or

drin

k 19

4/29

0 (6

6.9)

0.63

(0.3

8-1.

05)

0.97

(0.5

5-1.

70)

0.65

(0.3

1-1.

40)

Mea

n SB

P ly

ing

147

(22)

1.01

(1.0

0-1.

02)

0.99

(0.9

8-1.

01)

1.01

(0.9

9-1.

02)

Mea

n D

BP ly

ing

74 (1

1)1.

02 (0

.99-

1.04

)1.

01 (0

.98-

1.03

)1.

02 (0

.99-

1.05

)M

ean

puls

e fr

eque

ncy

74 (1

3)0.

98 (0

.97-

1.00

)1.

01 (0

.99-

1.03

)1.

00 (0

.98-

1.03

)Fa

lls la

st y

ear

59/2

90 (2

0.3)

0.95

(0.5

2-1.

72)

1.62

(0.8

6-3.

03)

2.42

(1.0

9-5.

38)

Ort

host

atic

hypo

tens

ion

106/

290

(36.

6)N

A1.

48 (0

.86-

2.55

)N

A

Ort

host

atic

com

plai

nts

71/2

88 (2

4.7)

1.48

(0.8

6-2.

55)

NA

NA

Scor

e Ba

rthe

l que

stion

naire

10

(6-1

5)0.

98 (0

.94-

1.02

)1.

01 (0

.96-

1.06

)1.

02 (0

.96-

1.09

)

Med

icati

onM

ean

num

ber

of a

gent

s 8.

7 (3

.5)

1.03

(0.9

6-1.

10)

0.97

(0.9

0-1.

05)

0.99

(0.8

9-1.

11)

Anti

hype

rten

sive

med

icati

on-

Diu

retic

s -

Beta

blo

cker

s -

Calc

ium

cha

nnel

blo

cker

s -

ACE

inhi

bito

rs

194/

290

(66.

9)12

8/29

0 (4

4.1)

85/2

90 (2

9.3)

42/2

90 (1

4.5)

100/

290

(34.

5)

1.32

(0.7

9-2.

21)

1.37

(0.8

5-2.

21)

0.99

(0.5

9-1.

68)

1.36

(0.7

0-2.

65)

1.34

(0.8

1-2.

20)

1.25

(0.7

0-2.

23)

1.77

(1.0

3-3.

03)

0.77

(0.4

2-1.

41)

0.95

(0.4

4-2.

05)

1.24

(0.7

1-2.

17)

0.76

(0.3

5-1.

64)

1.40

(0.6

6-2.

95)

0.99

(0.4

3-2.

24)

0.86

(0.2

9-2.

60)

0.76

(0.3

3-1.

71)

Benz

odia

zepi

nes

150/

290

(51.

7)0.

75 (0

.46-

1.21

)0.

96 (0

.56-

1.63

)0.

86 (0

.41-

1.82

)A

ntips

ycho

tics

59/2

90 (2

0.3)

1.14

(0.6

3-2.

05)

0.58

(0.2

8-1.

22)

0.73

(0.2

7-1.

99)

Anti

depr

essa

nts

69/2

90 (2

3.8)

0.83

(0.4

7-1.

47)

0.83

(0.4

4-1.

60)

0.93

(0.3

8-2.

25)

Opi

oids

32

/290

(11.

0)1.

87 (0

.89-

3.91

)0.

84 (0

.35-

2.05

)1.

65 (0

.59-

4.66

)

Base

line

char

acte

risti

cs a

nd r

esul

ts o

f un

ivar

iate

log

istic

reg

ress

ion

anal

yses

with

ort

host

atic

hypo

tens

ion

(OH

), or

thos

tatic

com

plai

nts,

and

sy

mpt

omati

c O

H a

s de

pend

ent v

aria

bles

. Dat

a ar

e m

eans

(SD

), m

edia

ns (i

nter

quar

tile

rang

e) o

r n

(%).

R1

R2

R3

R4

R5

R6

R7

R8

R9

R10

R11

R12

R13

R14

R15

R16

R17

R18

R19

R20

R21

R22

R23

R24

R25

R26

R27

R28

R29

R30

R31

R32

R33

R34

R35

R36

R37

R38

R39


99

6

SUPP

LEM

ENTA

L D

ATA

Tabl

e A

1. U

niva

riat

e lo

gisti

c re

gres

sion

ana

lyse

s w

ith O

H, o

rtho

stati

c co

mpl

aint

s an

d sy

mpt

omati

c O

H a

s de

pend

ent v

aria

bles

. To

tal

OR

(95%

CI)

OR

(95%

CI)

OR

(95%

CI)

Char

acte

risti

cN

=290

rt

host

atic

copl

aint

spt

oati

c D

emog

raph

ics

Ag

e

80.8

(9.9

)1.

04 (1

.01-

1.07

)1.

00 (0

.97-

1.03

)1.

01 (0

.97-

1.05

)G

ende

r (%

fem

ale)

206/

290

(71.

0)1.

18 (0

.70-

1.99

)0.

79 (0

.43-

1.46

)1.

00 (0

.44-

2.28

)M

ean

body

mas

s in

dex,

kg/

m2

26.3

(5.4

)0.

99 (0

.94-

1.03

)1.

00 (0

.95-

1.05

)1.

02 (0

.96-

1.09

)H

yper

tens

ion

226/

290

(77.

9)1.

63 (0

.89-

2.99

)1.

75 (0

.86-

3.58

)1.

53 (0

.56-

4.17

)H

isto

ry o

f CVD

dis

ease

13

1/29

0 (4

5.2)

1.21

(0.7

5-1.

95)

0.64

(0.3

7-1.

11)

0.74

(0.3

5-1.

59)

Dia

bete

s m

ellit

us10

4/29

0 (3

5.0)

1.29

(0.7

9-2.

12)

1.42

(0.8

2-2.

46)

1.53

(0.7

2-3.

27)

Dem

entia

121/

290

(41.

7)1.

18 (0

.73-

1.92

)0.

82 (0

.47-

1.42

)0.

54 (0

.24-

1.21

)Cu

rrre

nt s

mok

er

32/2

90 (1

1.0)

0.54

(0.2

4-1.

26)

0.68

(0.2

7-1.

73)

0.53

(0.1

2-2.

32)

Mea

sure

men

tsCo

nsum

ption

mea

l or

drin

k 19

4/29

0 (6

6.9)

0.63

(0.3

8-1.

05)

0.97

(0.5

5-1.

70)

0.65

(0.3

1-1.

40)

Mea

n SB

P ly

ing

147

(22)

1.01

(1.0

0-1.

02)

0.99

(0.9

8-1.

01)

1.01

(0.9

9-1.

02)

Mea

n D

BP ly

ing

74 (1

1)1.

02 (0

.99-

1.04

)1.

01 (0

.98-

1.03

)1.

02 (0

.99-

1.05

)M

ean

puls

e fr

eque

ncy

74 (1

3)0.

98 (0

.97-

1.00

)1.

01 (0

.99-

1.03

)1.

00 (0

.98-

1.03

)Fa

lls la

st y

ear

59/2

90 (2

0.3)

0.95

(0.5

2-1.

72)

1.62

(0.8

6-3.

03)

2.42

(1.0

9-5.

38)

Ort

host

atic

hypo

tens

ion

106/

290

(36.

6)N

A1.

48 (0

.86-

2.55

)N

A

Ort

host

atic

com

plai

nts

71/2

88 (2

4.7)

1.48

(0.8

6-2.

55)

NA

NA

Scor

e Ba

rthe

l que

stion

naire

10

(6-1

5)0.

98 (0

.94-

1.02

)1.

01 (0

.96-

1.06

)1.

02 (0

.96-

1.09

)

Med

icati

onM

ean

num

ber

of a

gent

s 8.

7 (3

.5)

1.03

(0.9

6-1.

10)

0.97

(0.9

0-1.

05)

0.99

(0.8

9-1.

11)

Anti

hype

rten

sive

med

icati

on-

Diu

retic

s -

Beta

blo

cker

s -

Calc

ium

cha

nnel

blo

cker

s -

ACE

inhi

bito

rs

194/

290

(66.

9)12

8/29

0 (4

4.1)

85/2

90 (2

9.3)

42/2

90 (1

4.5)

100/

290

(34.

5)

1.32

(0.7

9-2.

21)

1.37

(0.8

5-2.

21)

0.99

(0.5

9-1.

68)

1.36

(0.7

0-2.

65)

1.34

(0.8

1-2.

20)

1.25

(0.7

0-2.

23)

1.77

(1.0

3-3.

03)

0.77

(0.4

2-1.

41)

0.95

(0.4

4-2.

05)

1.24

(0.7

1-2.

17)

0.76

(0.3

5-1.

64)

1.40

(0.6

6-2.

95)

0.99

(0.4

3-2.

24)

0.86

(0.2

9-2.

60)

0.76

(0.3

3-1.

71)

Benz

odia

zepi

nes

150/

290

(51.

7)0.

75 (0

.46-

1.21

)0.

96 (0

.56-

1.63

)0.

86 (0

.41-

1.82

)A

ntips

ycho

tics

59/2

90 (2

0.3)

1.14

(0.6

3-2.

05)

0.58

(0.2

8-1.

22)

0.73

(0.2

7-1.

99)

Anti

depr

essa

nts

69/2

90 (2

3.8)

0.83

(0.4

7-1.

47)

0.83

(0.4

4-1.

60)

0.93

(0.3

8-2.

25)

Opi

oids

32

/290

(11.

0)1.

87 (0

.89-

3.91

)0.

84 (0

.35-

2.05

)1.

65 (0

.59-

4.66

)

Base

line

char

acte

risti

cs a

nd r

esul

ts o

f un

ivar

iate

log

istic

reg

ress

ion

anal

yses

with

ort

host

atic

hypo

tens

ion

(OH

), or

thos

tatic

com

plai

nts,

and

sy

mpt

omati

c O

H a

s de

pend

ent v

aria

bles

. Dat

a ar

e m

eans

(SD

), m

edia

ns (i

nter

quar

tile

rang

e) o

r n

(%).

R1

R2

R3

R4

R5

R6

R7

R8

R9

R10

R11

R12

R13

R14

R15

R16

R17

R18

R19

R20

R21

R22

R23

R24

R25

R26

R27

R28

R29

R30

R31

R32

R33

R34

R35

R36

R37

R38

R39

Chapter 6

100

Tabl

e A

2. R

ehab

ilita

tion

depa

rtm

ent:

Uni

vari

ate

logi

stic

regr

essi

on a

naly

ses w

ith O

H, o

rtho

stati

c co

mpl

aint

s, a

nd fa

lls la

st y

ear a

s dep

ende

nt v

aria

bles

.Re

habi

litati

on d

epar

tmen

tO

R (9

5%CI

)O

R (9

5%CI

)O

R (9

5%CI

)Ch

arac

teri

stic

N=1

28rt

host

atic

copl

aint

sal

ls la

st

ear

Dem

ogra

phic

sA

ge

80

.7 (7

.9)

1.01

(0.9

6-1.

05)

0.96

(0.9

1-1.

00)

1.02

(0.9

7-1.

07)

Fem

ale

Gen

der

92 (7

1.9)

1.57

(0.7

2-3.

46)

0.89

(0.3

7-2.

16)

0.47

(0.2

0-1.

12)

Mea

n bo

dy m

ass

inde

x, k

g/m

2

25.9

(5.3

)1.

01 (0

.95-

1.08

)1.

03 (0

.96-

1.11

)0.

99 (0

.92-

1.06

)H

yper

tens

ion

105

(82.

0)2.

40 (0

.83-

6.96

)4.

60 (1

.02-

20.8

1)0.

57 (0

.23-

1.42

)H

isto

ry o

f CVD

51 (3

9.8)

1.20

(0.5

8-2.

48)

0.45

(0.1

9-1.

06)

0.43

(0.2

0-0.

94)

Dia

bete

s m

ellit

us55

(43.

0)1.

12 (0

.54-

2.31

)0.

91 (0

.41-

2.01

)0.

64 (0

.31-

1.34

)D

emen

tia9

(7.0

)2.

29 (0

.58-

9.00

)0.

33 (0

.04-

2.71

)0.

85 (0

.20-

3.58

)Cu

rrre

nt s

mok

er

12 (9

.4)

0.55

(0.1

4-2.

12)

0.53

(0.1

1-2.

53)

0.85

(0.2

4-2.

99)

Mea

sure

men

tsM

ean

SBP

lyin

g 14

6.5

(22.

2)1.

00 (0

.99-

1.02

)0.

99 (0

.97-

1.01

)1.

01 (0

.99-

1.02

)M

ean

DBP

lyin

g 72

.4 (9

.8)

1.02

(0.9

9-1.

06)

1.01

(0.9

7-1.

05)

0.99

(0.9

6-1.

03)

Mea

n pu

lse

freq

uenc

y 75

.4 (1

2.3)

0.98

(0.9

6-1.

02)

0.99

(0.9

6-1.

03)

0.99

(0.9

7-1.

03)

Falls

last

yea

r47

(36.

7)0.

62 (0

.29-

1.33

)1.

29 (0

.58-

2.89

)N

A

Ort

host

atic

hypo

tens

ion

47 (3

6.7)

NA

2.13

(0.9

6-4.

75)

0.62

(0.2

9-1.

33)

Ort

host

atic

com

plai

nts

34 (2

6.6)

2.13

(0.9

6-4.

75)

NA

1.29

(0.5

8-2.

89)

Scor

e Ba

rthe

l que

stion

naire

13

(9-1

6)0.

98 (0

.91-

1.06

)0.

98 (0

.91-

1.07

)1.

06 (0

.97-

1.14

)

Med

icati

onM

ean

num

ber

of a

gent

s 8.

6 (3

.6)

1.08

(0.9

8-1.

20)

1.03

(0.9

2-1.

15)

0.98

(0.8

9-1.

09)

Anti

hype

rten

sive

med

icati

on-

Diu

retic

s-

Beta

blo

cker

s -

Calc

ium

cha

nnel

blo

cker

s-

ACE

inhi

bito

rs

91 (7

1.1)

62 (4

8.4)

45 (3

5.2)

21 (1

6.4)

43 (3

3.6)

1.85

(0.8

0-4.

28)

2.34

(1.1

2-4.

89)

0.93

(0.4

4-1.

97)

1.37

(0.5

3-3.

52)

1.39

(0.6

5-2.

95)

1.45

(0.5

9-3.

59)

1.77

(0.8

0-3.

92)

1.20

(0.5

3-2.

70)

1.48

(0.5

4-4.

05)

0.93

(0.4

0-2.

14)

0.94

(0.4

2-2.

06)

0.90

(0.4

4-1.

85)

1.07

(0.5

1-2.

27)

0.48

(0.1

7-1.

42)

0.76

(0.3

5-1.

65)

Benz

odia

zepi

nes

64 (5

0)0.

82 (0

.40-

1.68

)1.

62 (0

.73-

3.59

)1.

22 (0

.60-

2.51

)A

ntips

ycho

tics

14 (1

0.9)

0.95

(0.3

0-3.

03)

0.73

(0.1

9-2.

79)

0.43

(0.1

2-1.

64)

Anti

depr

essa

nts

19 (1

4.8)

1.01

(0.3

7-2.

76)

0.99

(0.3

3-2.

98)

1.01

(0.3

7-2.

76)

Opi

oids

13 (1

0.2)

3.12

(0.9

6-10

.17)

1.85

(0.5

6-6.

12)

0.74

(0.2

2-2.

56)

Base

line

char

acte

risti

cs a

nd re

sults

of u

niva

riat

e lo

gisti

c re

gres

sion

ana

lyse

s w

ith o

rtho

stati

c hy

pote

nsio

n (O

H),

orth

osta

tic c

ompl

aint

s, a

nd F

alls

last

ye

ar a

s de

pend

ent v

aria

bles

. Dat

a ar

e m

eans

(SD

), m

edia

ns (i

nter

quar

tile

rang

e) o

r n (%

). CV

D =

Car

dio

vasc

ular

dis

ease

. SBP

= sy

stol

ic b

lood

pre

ssur

e.

DBP

= d

iast

olic

blo

od p

ress

ure.

CHAPTER 7The association between orthostatic hypotension

and handgrip strength with successful rehabilitation

in elderly hip fracture patients

Accepted for publication as:

Hartog LC, Winters AM, Roijen H, Kamper AM, Inia H, Kleefstra N, Bilo HJG, Van

Hateren KJJ. The association between Orthostatic Hypotension and Handgrip

Strength with Successful Rehabilitation in elderly hip fracture patients.

Archives of Physical medicine and Rehabilitation.

R1

R2

R3

R4

R5

R6

R7

R8

R9

R10

R11

R12

R13

R14

R15

R16

R17

R18

R19

R20

R21

R22

R23

R24

R25

R26

R27

R28

R29

R30

R31

R32

R33

R34

R35

R36

R37

R38

R39

Chapter 7

102

ABSTRACT

b ective: Our objectives were to investigate the relationship between orthostatic hypotension (OH) and muscle strength versus time to successful rehabilitation within elderly hip fracture patients.

esign: A prospective observational cohort study. Handgrip strength was measured at the day of admission and OH as soon as possible after surgery. Cox proportional hazard modelling was used to investigate the relation between OH, or handgrip strength (kg) and time to successful rehabilitation, expressed as hazard ratios (HR). OH was defined as a drop in systolic blood pressure of > 20 mmHg or diastolic blood pressure of > 10 mmHg after postural change (dichotomous). Handgrip strength was measured with a hand dynamometer (continuous). The study was registered on trialregister.nl (NTR4940).

etting: General hospital

articipants: Patients of > 70 years with a hip fracture were recruited at the day of hospital admission. A total of 116 patients was included.

Main outco e easures: Primary outcome was time to successful rehabilitation, which was defined as discharge to patients’ own homes.

esults: During a median follow-up period of 36 days (IQR 9-57), 103 (89%) patients were successfully rehabilitated. No statistically significant relationships were found between OH and time to successful rehabilitation; HR 1.05 (95% confidence interval (CI) 0.67-1.66)). Also handgrip strength and successful rehabilitation were not statistically significantly related; HR 1.03 (95%CI 0.99-1.06).

onclusions: OH measured during the first days of hospitalization is not related to time to successful rehabilitation in operated hip fracture patients. Although no significant relationship was seen in the present study, the width of the confidence intervals does not exclude a relevant relationship between handgrip strength and time to successful rehabilitation

R1

R2

R3

R4

R5

R6

R7

R8

R9

R10

R11

R12

R13

R14

R15

R16

R17

R18

R19

R20

R21

R22

R23

R24

R25

R26

R27

R28

R29

R30

R31

R32

R33

R34

R35

R36

R37

R38

R39

rthostatic h potension handgrip strength and rehabilitation in elderl

103

7

INTRODUCTION

Hip fractures are a common cause of hospitalization and rehabilitation in elderly patients [1, 2]. The main purpose of rehabilitation in these patients is to regain their prefracture health status as much as possible [3, 4]. Dependence on medical care, decline in functional outcome, or admission to a nursing home may be the consequence when rehabilitation fails. The outcome of rehabilitation reflects the condition of the elderly patient and is a summation of many factors, including both physical and mental parameters [3-11]. The definition of successful rehabilitation or recovery varied widely; from regaining prior functional and/or mobility status, to functional independence leading to discharge to patients own home [4, 7, 9, 10, 12]. Examples of the numerous factors that negatively influence the response to rehabilitation are high age, presence of cognitive impairment or coexisting diseases, and high fear of falling (FOF) [3-5, 9, 13]. Also, orthostatic hypotension (OH) and muscle strength are amongst the factors that have been found to influence rehabilitation in elderly patients [7] [8, 10]. As the prevalence of OH and impaired muscle strength is high in elderly patients and are considered as important risk factors for falling and frailty, these variables are likely to negatively influence successful rehabilitation [14-20]. A previous study observed the counterintuitive finding that patients with OH were found to have a higher risk of successful rehabilitation compared to patients without OH [7]. Another study found no difference in functional outcome between stroke patients with and without OH [21]. Muscle strength is considered to be a strong positive predictor for functional outcome after rehabilitation in elderly hip fracture patients [8, 10]. OH and muscle strength separately influence outcome, but it is likely that these factors are also interrelated. Several causes of OH, such as the use of different medications, hypovolemic disorders, and bed rest, are potentially related to muscle strength [22-25]. As muscle strength and OH are both related to successful rehabilitation, and possibly also interrelated, these factors should be combined (and adjusted for) in analysing the association with rehabilitation. As far as we are aware, no previous studies investigated these combined associations. Therefore, we performed a study in which we aimed to investigate the relation between OH and muscle strength with time to successful rehabilitation within elderly hip fracture patients. We hypothesized that the presence of OH or low muscle strength would negatively influence the time to successful rehabilitation. Furthermore, we hypothesized that the relationship between OH and time to successful rehabilitation would be influenced by muscle strength.

R1

R2

R3

R4

R5

R6

R7

R8

R9

R10

R11

R12

R13

R14

R15

R16

R17

R18

R19

R20

R21

R22

R23

R24

R25

R26

R27

R28

R29

R30

R31

R32

R33

R34

R35

R36

R37

R38

R39

Chapter 7

104

METHODS

tud populationThis prospective observational cohort study was performed in a general hospital (the Isala hospital, Zwolle, the Netherlands). All patients of 70 years of age or older, admitted to the hospital with a hip fracture and treated by surgery, were recruited. Recruitment and all study procedures took place between November 2014 and December 2015. Exclusion criteria were a life expectancy of less than 3 months, unable to mobilize before hospitalization, and being institutionalized in a nursing home facility before hospitalization. By performing a prospective study in a general hospital we tried to minimize the chance of selection bias.

ata collectionBaseline data involved demographic characteristics, a full medical history including a history of cardiovascular disease (CVD), diabetes mellitus (DM), hypertension, FOF, and medication use. Patients were considered to have cardiovascular disease when they had a history of angina pectoris, myocardial infarction, percutaneous transluminal coronary angioplasty, coronary artery bypass grafting, stroke or transient ischaemic attack. Blood pressure was measured following a standardized protocol, using an automated sphygmomanometer (A&D UA-767 Plus) [26]. If the automated sphygmomanometer displayed an error message, blood pressure was manually measured with a Heine Gamma XXL-T sphygmomanometer [27]. Blood pressure was measured two times in supine position after 5 minutes of rest, and two times each at 1 and 3 minutes after postural change. The forearm of the patient was supported at heart level during the measurements in upright position [28]. The postural change was from supine to standing position, or from supine to sitting position for patients who were unable to stand. Blood pressure was measured as soon as possible after surgery. OH was defined as a drop in systolic blood pressure (SBP) of > 20 mmHg or diastolic blood pressure (DBP) of > 10 mmHg after postural change compared to the mean value of the baseline measurements in supine position [29]. Characteristic symptoms of OH like light-headedness, syncope, or dizziness after postural change were questioned and the combination of OH and orthostatic complaints was described as symptomatic OH. Handgrip strength was measured with a Jamar hand dynamometer [30] in kilogram (kg) within 2 days of hospital admission, preferably at the day of admission. When a patient was operated on the day of admission, the handgrip strength was measured postoperatively but always within two days after admission.Testing was performed with the participant in a comfortable sitting position in the hospital bed. The forearms were resting with the elbow flexed at 90°, the forearm in neutral

R1

R2

R3

R4

R5

R6

R7

R8

R9

R10

R11

R12

R13

R14

R15

R16

R17

R18

R19

R20

R21

R22

R23

R24

R25

R26

R27

R28

R29

R30

R31

R32

R33

R34

R35

R36

R37

R38

R39


105

7

position, and thumbs facing up. Both dominant and non-dominant hand was tested, both 3 times. The best of 6 attempts of maximal voluntary contraction was used for statistical analysis [30]. OH was expressed as dichotomous (OH vs no OH) and handgrip strength as a continuous variable.To measure FOF, a numeric scale (1-10) was used, with 1 representing no FOF and with 10 representing an extreme FOF [31]. The FOF was measured at the day of admission. Activities of daily living were measured with the Barthel-20 index at the day of admission [32] to evaluate prefracture status. Body mass index was calculated by measuring body weight and height. All tests were part of usual clinical care. Four trained medical staff members performed all tests to reduce the change on inter-observer disagreement. It was intended to measure all variables per patient by the same medical staff member. Primary outcome was time to successful rehabilitation, which was defined as discharge to patients’ own homes, where they functioned self-reliant and lived by themselves. Time to successful rehabilitation started on the day of OH blood pressure measurements, which were performed as soon as possible after surgery. Patients were considered as self-reliant if a patient regained his or her prefracture health status. As a consequence, patients with an already highly adapted home environment (e.g. stairlift, homecare, meal service) may be sent home earlier than others. In the trial register, successful rehabilitation was predefined as having the same functional status compared to the prefracture status, evaluated by using the mobility component of the Barthel index. Because all patients reached the prefracture mobility score on the Barthel index in a few days after surgery (despite the fact they were not discharged home, but had to be admitted to a rehabilitation facility), we evaluated this definition and decided to change it into the current clinically more relevant definition.

tatistical anal sesContinuous variables are presented as mean and standard deviation for normally distributed variables, or as median and interquartile range for non-normally distributed variables. Cox proportional hazard modelling was used to investigate the relation between OH, orthostatic complaints, symptomatic OH, or handgrip strength and time to successful rehabilitation. Two separate cox proportional hazard analyses were performed; one regarding the relationship between OH and successful rehabilitation and one between muscle strength and successful rehabilitation. We used three different models. In model 1, unadjusted analyses were performed. In model 2, only age and gender were taken into account as possible confounders. In model 3, regarding the relationship between OH and rehabilitation, we additionally adjusted for the following variables: body mass index (BMI), a history of diabetes mellitus, the score on the Barthel index, previous macrovascular complications, mean systolic blood pressure, the use of antihypertensive medication, and

R1

R2

R3

R4

R5

R6

R7

R8

R9

R10

R11

R12

R13

R14

R15

R16

R17

R18

R19

R20

R21

R22

R23

R24

R25

R26

R27

R28

R29

R30

R31

R32

R33

R34

R35

R36

R37

R38

R39

Chapter 7

106

baseline handgrip strength. For the analyses regarding the relationship between handgrip strength and rehabilitation, we adjusted for age, gender, BMI, the score of the Barthel index, previous macrovascular complications, and OH. These confounders were chosen based on clinical grounds, since all confounders were likely to be related to successful rehabilitation and OH or handgrip strength. By adjusting for potential confounding factors the risk of confounding bias was reduced.The confounding effect of FOF on the relationship between OH, handgrip strength and successful rehabilitation was explored by adding FOF to model 3 in both analyses. FOF was added separately because of missing values (n=8). There were missing values of FOF (n=8), BMI (n=6), and Barthel index (n=2). The hazard ratios (HRs) regarding systolic blood pressure refer to a pressure increase in steps of 10 mmHg.The Schoenfeld residual plots were inspected for each predictor variable to check the assumption of proportional hazards.P-values less than 0.05 were considered statistically significant. Collinearity diagnostics were tested for each confounder; co-variables are considered to be highly correlated with a variance inflation factor (VIF) of 10 or more [33, 34]. When necessary, interaction was tested between different variables. Interaction was considered to be significant, with a p value less than 0.05.All statistical analyses were performed using SPSS software (version 22). The ‘Strengthening the Reporting of Observational studies in Epidemiology’ (STROBE) statement was used to describe this observational cohort study [35]. The study was registered on trialregister.nl (NTR4940).

thical approval and linical rial registrationThis study was performed in accordance with the Declaration of Helsinki. According to Dutch guidelines this study did not fall under the scope of the Medical Research Involving Human Subjects Act, and therefore this study did not need a formal approval of an accredited medical ethics committee. Written informed consent was obtained for all patients by the participating medical doctor or nurse. All data were analysed anonymously. The study was registered on Trialregister.nl (NTR4940).

RESULTS

A total of 116 patients was included in this cohort. The baseline characteristics are presented in table 1. Median age of the total study population was 82 (IQR (interquartile range) 76-86) years. Various surgical techniques were used to treat the hip fractures; 37% intramedullary nail, 50% hemi- or total hip arthroplasty, 13% (sliding) hip screws. 39 patients (34%) were

R1

R2

R3

R4

R5

R6

R7

R8

R9

R10

R11

R12

R13

R14

R15

R16

R17

R18

R19

R20

R21

R22

R23

R24

R25

R26

R27

R28

R29

R30

R31

R32

R33

R34

R35

R36

R37

R38

R39


107

7

discharged to their own homes and 77 patients (66%) to a nursing home facility for further rehabilitation. During a median follow-up period of 36 days (IQR 9-57), 103 (89%) patients were successfully rehabilitated. Three patients died during rehabilitation. Ten patients could not return home and stayed at a long-term nursing home facility. Patients who did not successfully rehabilitate were found to have a higher prevalence of macrovascular disease and hypertension compared to patients who were successfully rehabilitated.

Table 1. Baseline characteristics total population.

TotalBaseline

Characteristic N=116DemographicsAge (years) 82 (76-86)Female gender 86 (74)Mean body mass index (kg/m2) 25 (23-28)Hypertension 77 (66)History of CVD 27 (23)Diabetes mellitus 23 (20)Current smoker 17 (15)

MeasurementsConsumption meal or drinka

113 (97)No. Days between operation and BPM 2 (1-3)Mean SBP lying (mmHg) 130 (22)Mean DBP lying (mmHg) 65 (11)Mean pulse frequency (beats/min) 81 (18)Orthostatic hypotension 39 (34)Orthostatic complaints 22 (19)Symptomatic hypotension 16 (14)Percentage postoperative handgrip strength measurement 21 (18)Handgrip Strength (kg) 20 (15-26)Score Barthel index 19 (17-20)Fear of Fallingb 1 (1-4)

Medication during admissionMean number of agents 6 (3-9)Antihypertensive medication- Diuretics- Beta blockers - Calcium channel blockers- ACE inhibitors

70 (60)40 (35)31 (27)17 (15)44 (38)

Benzodiazepines 22 (19)Antipsychotics 3 (3)Antidepressants 12 (10)

Data are means (± SD), medians (interquartile range) or n (%). BPM = blood pressure measurement. MSM = muscle strength measurement. SBP = systolic blood pressure. DBP = diastolic blood pressure. a Meal < 2 hours or drink < 1 hour prior to the measurements. b Missing values in 8 patients.

R1

R2

R3

R4

R5

R6

R7

R8

R9

R10

R11

R12

R13

R14

R15

R16

R17

R18

R19

R20

R21

R22

R23

R24

R25

R26

R27

R28

R29

R30

R31

R32

R33

R34

R35

R36

R37

R38

R39

Chapter 7

108

and uccess ul rehabilitationOH was present in 39 out of 116 patients, resulting in a prevalence of 34% (95% confidence interval (CI) 25-43%). The postural change was performed mostly from lying to sitting (n=114 (98%)) due to decreased mobility; only 2% of the tested population could perform postural change from lying to standing position. Blood pressure measurement took place with a median of 2 (IQR 1-3) days after surgery.Table 2 presents the results of the Cox regression analyses regarding the relationship between OH and successful rehabilitation. In the present study no statistically significant relationships were seen between OH (HR 1.05 (95%CI 0.67-1.66)) and time to rehabilitation. The confounders systolic blood pressure (HR 1.01 (95%CI 1.00-1.03)), diabetes mellitus (HR 0.47 (95%CI 0.26-0.85)), and handgrip strength (HR 1.05 (95%CI 1.01-1.08)) were statistically significantly related to time to successful rehabilitation. Adding FOF to the multivariate model did not change the association between OH and time to rehabilitation. The hazard ratio of FOF was 0.87 (95%CI 0.79-0.97). Orthostatic complaints (HR 1.06 (95%CI 0.62-1.83)) and symptomatic OH (HR 1.15 (95%CI 0.62-2.13)) were also not related to time to successful rehabilitation in the multivariate analyses.

Table 2. Hazard ratios of OH for successful rehabilitation (n=116).

Model 1 Model 2 Model 3HR (95% CI) HR (95% CI) HR (95% CI)

OH 1.35 (0.90-2.05) 1.28 (0.85-1.94) 1.05 (0.67-1.66)Age - 0.96 (0.93-0.99) (p=0.02) 0.99 (0.96-1.03)Gender, female vs male - 0.99 (0.63-1.54) 0.61 (0.31-1.18)BMI - - 0.99 (0.95-1.05)Score Barthel Index - - 1.05 (0.94-1.18)Antihypertensive medication - - 0.90 (0.58-1.40)History of CVD disease - - 0.64 (0.37-1.11)DM, DM vs control - - 0.47 (0.26-0.85) (p=0.01)Mean SBP lying a

- - 1.01 (1.00-1.03) (p=0.01)Handgrip Strength - - 1.05 (1.01-1.08) (p<0.01)

Hazard ratios for successful rehabilitation. Model 1 unadjusted. Model 2 adjusted for age and gender. Model 3 adjusted for age, gender, BMI, score Barthel index, the number of antihypertensive medications, previous macro vascular complications, diabetes mellitus, mean SBP lying, and Handgrip Strength. OH = orthostatic hypotension. BMI = Body mass index. CVD = Cardio vascular disease. DM = Diabetes Mellitus. SBP = systolic blood pressure. a The hazard ratio refers to a pressure increase of 10 mmHg.

As blood pressure may be a marker of frailty in old age [36], we performed analyses in which we tested for interaction between systolic blood pressure and OH. No statistically significant interaction was seen. The plots of the Schoenfeld residuals showed that the assumptions of proportional hazards were met. Collinearity was tested and no serious multicollinearity was seen, because the mean VIF value was 1.37 (range 1.01-2.22).

R1

R2

R3

R4

R5

R6

R7

R8

R9

R10

R11

R12

R13

R14

R15

R16

R17

R18

R19

R20

R21

R22

R23

R24

R25

R26

R27

R28

R29

R30

R31

R32

R33

R34

R35

R36

R37

R38

R39


109

7

andgrip strength and uccess ul rehabilitationMedian handgrip strength of the dominant arm was 20 kilograms (IQR 15-26). All handgrip strength measurements were performed within 2 days of admission. For the majority (82%) of the patients the handgrip strength measurements were performed preoperatively.Table 3 present the results of the Cox regression analyses regarding the relationship between muscle strength and time to successful rehabilitation. None of the models showed a significant relationship between handgrip strength and time to successful rehabilitation. The confounder CVD was related to time to successful rehabilitation (HR 0.57 (95%CI 0.33-0.99). Adding FOF to the multivariate model did not change the association of handgrip strength with rehabilitation (HR 1.03 (95%CI 0.99-1.06)). As a confounder, FOF was significantly related to time to successful rehabilitation (HR 0.87 (95%CI 0.78-0.97)).

Table 3. Hazard ratios of handgrip strength for successful rehabilitation (n=116).

Model 1 Model 2 Model 3HR (95% CI) HR (95% CI) HR (95% CI)

Handgrip strength 1.02 (1.00-1.04)* 1.02 (0.99-1.05) 1.03 (0.99-1.06) Age - 0.97 (0.94-1.00) 0.99 (0.95-1.02)Gender, female vs male - 0.72 (0.39-1.33) 0.83 (0.44-1.57)BMI - - 0.97 (0.93-1.02)Score Barthel Index - - 1.09 (0.98-1.21)History of CVD disease - - 0.57 (0.33-0.99)OH - - 1.09 (0.71-1.68)

Hazard ratios for successful rehabilitation. Model 1 unadjusted. Model 2 adjusted for age and gender. Model 3 adjusted for age, gender, BMI, score Barthel index, previous macro vascular complications, and OH. CVD = Cardio vascular disease. OH = orthostatic hypotension. * p=0,054.

DISCUSSION

OH, measured in the immediate postoperative phase, was not related to time to successful rehabilitation in hospitalized elderly with a hip fracture. Although increased muscle strength was not significantly related to time to successful rehabilitation in the present study, the width of the confidence interval does not exclude a relevant relationship between handgrip strength and time to successful rehabilitation. Besides, muscle strength as a confounder, in the model with OH as the variable of interest, was significantly related to time to successful rehabilitation.

R1

R2

R3

R4

R5

R6

R7

R8

R9

R10

R11

R12

R13

R14

R15

R16

R17

R18

R19

R20

R21

R22

R23

R24

R25

R26

R27

R28

R29

R30

R31

R32

R33

R34

R35

R36

R37

R38

R39

Chapter 7

110

and success ul rehabilitationIn contrast to the current study, a previous study performed by the same authors showed that patients with OH were found to have a higher hazard of successful rehabilitation compared to patients without OH [7]. Although our previous study reported a positive relation between OH and time to successful rehabilitation, we hypothesized prior to the present study that the presence of OH would negatively influence the time to successful rehabilitation. The prevalence of OH and successful rehabilitation was similar in both studies. When comparing both study populations, patients of the present study seemed to have less comorbidity, used less medication, and baseline blood pressure was lower, which reflects the setting of the previous study (nursing home).The high prevalence of OH in the present study could be partially caused by hip fracture or hospital admission-related factors like bed rest, surgery, effects of anaesthesia, inadequate water intake, and blood loss. In these circumstances, OH may very well be a temporarily phenomenon and therefore not a predictor for an outcome such as time to successful rehabilitation [23]. In the study of Weiss et al., the impact of OH in hospitalized patients on mortality was described, and they advised to divide patients in 2 groups; patients with episodic OH, as is seen during hospitalization, and established OH (repeated measurements) [22]. Measuring OH in the first week of rehabilitation within a nursing home might possibly be a more accurate predictor for successful rehabilitation. Analogous to the association with mortality, as assessed in the study by Weiss et al., one may hypothesize that episodic and sustained OH have different associations with rehabilitation. Episodic OH may have no consequences for chances of rehabilitation, whereas sustained OH may be much more relevant.The confounders DM, SBP, handgrip strength and FOF were significantly related to time to successful rehabilitation. The hazard of successful rehabilitation in patients with DM was lower than patients without DM, as was also seen in the previous (mentioned) study [7]. The hazard of successful rehabilitation increased by 15% (95%CI 3-28%) for every 10 mmHg increase in SBP. In a previous study, higher blood pressure in frail patients was related to lower all-cause mortality while the opposite relationship was seen in non-frail patients [37]. Therefore, it was not unexpected that higher SBP is associated with a higher hazard of successful rehabilitation. Poor muscle strength and FOF are frequently seen in elderly patients, and these factors are also related with the level of frailty [14-16]. Successful rehabilitation increased by 5% (95% CI 1-8) for every 1 kg increase in handgrip strength measurement. The relationship between handgrip strength and rehabilitation will be discussed in 4.2.Time to successful rehabilitation decreased by 13% (95% CI 3-22%) for every 1-point increase on the VAS-FOF scale. These results support previous studies regarding the impact of FOF on functional outcome [11, 38], which describes an association between FOF with negative

R1

R2

R3

R4

R5

R6

R7

R8

R9

R10

R11

R12

R13

R14

R15

R16

R17

R18

R19

R20

R21

R22

R23

R24

R25

R26

R27

R28

R29

R30

R31

R32

R33

R34

R35

R36

R37

R38

R39


111

7

outcomes as falling and functional impairment (e.g. IADL). In the study by Oude Voshaar et al. FOF seems to be an important predictor for functional recovery after hip fracture surgery [9]. Previous studies described that fear after falling may restrict physical activity, which causes immobility and further loss of functional independence and risk of falling [39, 40]. FOF can be divided into three components; physiological, behavioural, and cognitive [11]. Prevention and treatment of FOF by intervening all of those three components is an important clinical treatment goal.

andgrip strength and success ul rehabilitationAlthough increased muscle strength was not significantly related to time to successful rehabilitation in the present study, a relationship cannot be excluded based on the width of the confidence interval. In the model with OH as the variable of interest, a statistically significant association was observed. Previous studies also observed positive relationships between handgrip strength and rehabilitation [8] [41]. Di Monaco et al. described a significant relationship between handgrip strength and functional outcome in hip fracture patients [8]. Another study showed a relationship between handgrip strength during hospital admission and walking independently [41]. An important difference between the study of Di Monaco and the present study is the timing of the handgrip strength measurement; at the rehabilitation division after discharge from the hospital versus preoperatively in the present study. Measuring handgrip strength preoperatively reflects the baseline condition of a patient and is a predictor for complications or length of stay [42, 43]. Therefore, handgrip strength measurement can be used to identify those patients who are frailer and need a different approach during hospital admission [42].

tud trengths and i itationsAs the present study took place in a general hospital, and only a few exclusion criteria were used, our study population is a representative group of elderly patients with a hip fracture. The timing of inclusion and the homogenous study population of the present study were major strengths compared to our previous study [7]. Recruitment and testing took place within 2 days after admission to the hospital, preferably at the day of admission. The current study has also some limitations. The main limitation was that 18% of the handgrip strength measurements were not measured preoperatively. However, we performed the same analyses in the group of patients with preoperatively measured handgrip strength, and the results did not relevantly change (data not shown). Although OH should be measured from lying to standing, this was not possible in 98% of patients. It is very likely that the actual number of patients with OH was higher. Furthermore, OH was only measured once during the follow-up period, which probably biased the results. By repeated OH measurements not only episodic OH but also established OH would have been diagnosed. Future studies are needed to evaluate the clinical implications of sustained OH on rehabilitation.

R1

R2

R3

R4

R5

R6

R7

R8

R9

R10

R11

R12

R13

R14

R15

R16

R17

R18

R19

R20

R21

R22

R23

R24

R25

R26

R27

R28

R29

R30

R31

R32

R33

R34

R35

R36

R37

R38

R39

Chapter 7

112

Another limitation is the definition of successful rehabilitation; in our study, this was defined as discharge to patients’ own homes. Patients with a worse outcome after rehabilitation but with a highly adapted home environment may be sent home earlier than others.

CONCLUSIONS

In conclusion, this study showed that orthostatic hypotension measured during the first days of hospitalization was not related to time to successful rehabilitation. Although no significant relationship was seen in the present study, the width of the confidence interval does not exclude a relevant relationship between handgrip strength and time to successful rehabilitation

R1

R2

R3

R4

R5

R6

R7

R8

R9

R10

R11

R12

R13

R14

R15

R16

R17

R18

R19

R20

R21

R22

R23

R24

R25

R26

R27

R28

R29

R30

R31

R32

R33

R34

R35

R36

R37

R38

R39


113

7

REFERENCES

1. Peeters CM, Visser E, Van de Ree CL, et al. Quality of life after hip fracture in the elderly: A systematic literature review. Injury. 2016 Jul; 47(7): 1369-82.

2. Polinder S, Haagsma J, Panneman M, et al. The economic burden of injury: Health care and productivity costs of injuries in the Netherlands. Accid Anal Prev. 2016 Aug;93:92-100. doi: 10.1016/j.aap.2016.04.003. Epub 2016 May 10.

3. Lieberman D, Friger M, Lieberman D. Inpatient rehabilitation outcome after hip fracture surgery in elderly patients: a prospective cohort study of 946 patients. Archives of physical medicine and rehabilitation. 2006 Feb;87(2):167-71.


5. Clague JE, Craddock E, Andrew G, Horan MA, Pendleton N. Predictors of outcome following hip fracture. Admission time predicts length of stay and in-hospital mortality. Injury. 2002 Jan;33(1):1-6.

6. Muir SW, Yohannes AM. The impact of cognitive impairment on rehabilitation outcomes in elderly patients admitted with a femoral neck fracture: a systematic review. Journal of geriatric physical therapy (2001). 2009;32(1):24-32.

7. Hartog LC, Cizmar-Sweelssen M, Knipscheer A, et al. The association between orthostatic hypotension, falling and successful rehabilitation in a nursing home population. Arch Gerontol Geriatr. 2015 Sep-Oct;61(2):190-6. doi: 10.1016/j.archger.2015.05.005. Epub 2015 May 18.

8. Di Monaco M, Castiglioni C Fau - De Toma E, De Toma E Fau - Gardin L, et al. Handgrip strength is an independent predictor of functional outcome in hip-fracture women: a prospective study with 6-month follow-up. Medicine (Baltimore). 2015 Feb;94(6):e542. doi: 10.1097

9. Oude Voshaar RC, Banerjee S Fau - Horan M, Horan M Fau - Baldwin R, et al. Fear of falling more important than pain and depression for functional recovery after surgery for hip fracture in older people. Psychol Med. 2006 Nov;36(11):1635-45. Epub 2006 Jul 25.

10. Di Monaco M, Castiglioni C, De Toma E, et al. Handgrip strength but not appendicular lean mass is an independent predictor of functional outcome in hip-fracture women: a short-term prospective study. Arch Phys Med Rehabil. 2014 Sep;95(9):1719-24. doi: 10.1016/j.apmr.2014.04.003. Epub 2014 Apr 21.

11. Visschedijk JH, Caljouw MA, Bakkers E, van Balen R, Achterberg WP. Longitudinal follow-up study on fear of falling during and after rehabilitation in skilled nursing facilities. BMC Geriatr. 2015 Dec 4;15:161. doi: 10.1186/s12877-015-0158-1.

12. Vaughan L, Leng X, La Monte MJ, et al. Functional Independence in Late-Life: Maintaining Physical Functioning in Older Adulthood Predicts Daily Life Function after Age 80. J Gerontol A Biol Sci Med Sci. 2016 Mar;71 Suppl 1:S79-86. doi: 10.1093/gerona/glv061.

13. Marottoli RA, Berkman LF, Leo-Summers L, Cooney LM, Jr. Predictors of mortality and institutionalization after hip fracture: the New Haven EPESE cohort. Established Populations for Epidemiologic Studies of the Elderly. American journal of public health. 1994 Nov;84(11):1807-12.

14. Fried LP, Tangen Cm Fau - Walston J, Walston J Fau - Newman AB, et al. Frailty in older adults: evidence for a phenotype. J Gerontol A Biol Sci Med Sci. 2001 Mar;56(3):M146-56.

15. O’Connell MD, Savva GM, Fan CW, Kenny RA. Orthostatic hypotension, orthostatic intolerance and frailty: The Irish Longitudinal Study on Aging-TILDA. Arch Gerontol Geriatr. 2015 May-Jun;60(3):507-13. doi: 10.1016/j.archger.2015.01.008. Epub 2015 Jan 17.

16. Rockwood K, Abeysundera Mj Fau - Mitnitski A, Mitnitski A. How should we grade frailty in nursing home patients? J Am Med Dir Assoc. 2007 Nov;8(9):595-603. Epub 2007 Oct 22.

R1

R2

R3

R4

R5

R6

R7

R8

R9

R10

R11

R12

R13

R14

R15

R16

R17

R18

R19

R20

R21

R22

R23

R24

R25

R26

R27

R28

R29

R30

R31

R32

R33

R34

R35

R36

R37

R38

R39

Chapter 7

114




20. Moreland JD, Richardson Ja Fau - Goldsmith CH, Goldsmith Ch Fau - Clase CM, Clase CM. Muscle weakness and falls in older adults: a systematic review and meta-analysis. J Am Geriatr Soc. 2004 Jul;52(7):1121-9.

21. Kong KH, Chuo AM. Incidence and outcome of orthostatic hypotension in stroke patients undergoing rehabilitation. Archives of physical medicine and rehabilitation. 2003 Apr;84(4):559-62.



24. Spruit MA, Sillen MJ, Groenen MT, Wouters EF, Franssen FM. New normative values for handgrip strength: results from the UK Biobank. Journal of the American Medical Directors Association. 2013 Oct;14(10):775 e5-11.

25. Dodds RM, Syddall HE, Cooper R, et al. Grip strength across the life course: normative data from twelve British studies. PloS one. 2014;9(12):e113637.

26. Rogoza AN, Pavlova TS, Sergeeva MV. Validation of A&D UA-767 device for the self-measurement of blood pressure. Blood pressure monitoring. 2000 Aug;5(4):227-31.




30. Roberts HC, Denison HJ, Martin HJ, et al. A review of the measurement of grip strength in clinical and epidemiological studies: towards a standardised approach. Age and ageing. 2011 Jul;40(4):423-9.

31. Scheffer Ac Fau - Schuurmans MJ, Schuurmans Mj Fau - vanDijk N, vanDijk N Fau - van der Hooft T, van der Hooft T Fau - de Rooij SE, de Rooij SE. Reliability and validity of the visual analogue scale for fear of falling in older persons. J Am Geriatr Soc. 2010 Nov;58(11):2228-30. doi: 10.1111/j.1532-5415.2010.03105.x.


33. G. D. The little handbook of statistical practice.: Amazon digital services; 2002.

34. Tu YK, Kellett M Fau - Clerehugh V, Clerehugh V Fau - Gilthorpe MS, Gilthorpe MS. Problems of correlations between explanatory variables in multiple regression analyses in the dental literature. Br Dent J. 2005 Oct 8;199(7):457-61.


R1

R2

R3

R4

R5

R6

R7

R8

R9

R10

R11

R12

R13

R14

R15

R16

R17

R18

R19

R20

R21

R22

R23

R24

R25

R26

R27

R28

R29

R30

R31

R32

R33

R34

R35

R36

R37

R38

R39


115

7


37. van Hateren KJ, Hendriks SH, Groenier KH, et al. Frailty and the relationship between blood pressure and mortality in elderly patients with type 2 diabetes (Zwolle Outpatient Diabetes project Integrating Available Care-34). Journal of hypertension. 2015 Jun;33(6):1162-6.

38. Lach HW, Parsons JL. Impact of fear of falling in long term care: an integrative review. J Am Med Dir Assoc. 2013 Aug;14(8):573-7. doi: 10.1016/j.jamda.2013.02.019. Epub 2013 Apr 16.

39. Myers AM, Powell LE, Maki BE, et al. Psychological indicators of balance confidence: relationship to actual and perceived abilities. The journals of gerontology Series A, Biological sciences and medical sciences. 1996 Jan;51(1):M37-43.

40. Tinetti ME, Powell L. Fear of falling and low self-efficacy: a case of dependence in elderly persons. Journal of gerontology. 1993 Sep;48 Spec No:35-8.

41. Savino E, Martini E Fau - Lauretani F, Lauretani F Fau - Pioli G, et al. Handgrip strength predicts persistent walking recovery after hip fracture surgery. Am J Med. 2013 Dec;126(12):1068-75.e1. doi: 10.1016/j.amjmed.2013.04.017. Epub 2013 Sep 18.

42. Shyam Kumar AJ, Beresford-Cleary N Fau - Kumar P, Kumar P Fau - Barai A, et al. Preoperative grip strength measurement and duration of hospital stay in patients undergoing total hip and knee arthroplasty. Eur J Orthop Surg Traumatol. 2013 Jul;23(5):553-6. doi: 10.1007/s00590-012-1029-5. Epub 2012 Aug 29.

43. Webb AR, Newman La Fau - Taylor M, Taylor M Fau - Keogh JB, Keogh JB. Hand grip dynamometry as a predictor of postoperative complications reappraisal using age standardized grip strengths. JPEN J Parenter Enteral Nutr. 1989 Jan-Feb;13(1):30-3.

CHAPTER 8Orthostatic changes in blood pressure and mortality

in a nursing home population

Published as: Hartog LC, Hendriks SH, Cimzar-SweelssenM , Knipscheer A, Groenier KH,

Kleefstra N, Bilo HJG, Van Hateren KJJ. Orthostatic changes in blood pressure

and mortality in a nursing home population. Journal of hypertension 2016; 34

(6):1068-74

R1

R2

R3

R4

R5

R6

R7

R8

R9

R10

R11

R12

R13

R14

R15

R16

R17

R18

R19

R20

R21

R22

R23

R24

R25

R26

R27

R28

R29

R30

R31

R32

R33

R34

R35

R36

R37

R38

R39

Chapter 8

118

ABSTRACT

b ective: Hypertension, orthostatic hypotension (OH) and orthostatic hypertension (OHT), are highly prevalent in old age. The associations in the very elderly and frail patients between blood pressure, and especially orthostatic changes in blood pressure and mortality, are unclear. We aimed to investigate the relationships between orthostatic changes in blood pressure, blood pressure and mortality in nursing home residents.

esign and Methods: A prospective observational cohort study. Cox proportional hazard modeling was used to investigate the relation between OH, OHT, the various blood pressure variables, and mortality with adjustment for confounders. In case of significant associations in the models, risk prediction capabilities were assessed with Harrell’s C statistics and the proportion of explained variance (R2).

esults: 290 patients with a mean age of 80.8 (SD 9.9) years participated in this study. The overall mortality risk increased by 17% (95%CI: 2-34%) for every 10-mmHg increase in DBP. Adding DBP did not change Harrell’s C values and increased R2 with < 0.03. Only in patients at the psychogeriatric department, OH was associated with an increased all-cause mortality risk (HR 1.71 [95%CI 1.08-2.71]). The hazard ratio of OHT in this patient group was 0.61 [95%CI 0.32-1.19].

onclusion: DBP was related to all-cause mortality in a nursing home population. OH was related to all-cause mortality in the most frail group of nursing home patients. The predictive capabilities of both DBP and OH are rather small with respect to mortality. A beneficial effect of OHT could not be excluded based on the width of the confidence interval.

R1

R2

R3

R4

R5

R6

R7

R8

R9

R10

R11

R12

R13

R14

R15

R16

R17

R18

R19

R20

R21

R22

R23

R24

R25

R26

R27

R28

R29

R30

R31

R32

R33

R34

R35

R36

R37

R38

R39

rthostatic changes in blood pressure and ortalit in a nursing ho e population

119

8

INTRODUCTION

Hypertension and orthostatic changes in blood pressure, i.e. orthostatic hypotension (OH) and hypertension (OHT), are highly prevalent in old age [1-3]. Some studies describe that the presence of OH increases the risk of cardiovascular disease, cardiovascular mortality or all-cause mortality in elderly people [2, 4-11]. However, the evidence supporting an independent relationship between OH and mortality in the elderly population is sparse. Furthermore, many studies were not able to adjust for important confounders in the relationship between OH and mortality [8]. Although there is no official definition of OHT, many studies use an increase in systolic blood pressure (SBP) of 20 mmHg after postural change as a cut-off point. OHT can be considered as a form of prehypertension and is associated with hypertension associated target-organ damage [12, 13]. Furthermore, OHT is related to all-cause and cardiovascular mortality in a group of community-dwelling elderly patients [2]. The relationship between blood pressure level itself and mortality in old age has been investigated in many observational studies, and mostly inverse associations were reported [14, 15]. The only randomized controlled trial in octogenarians, involving only very healthy patients, showed that lowering blood pressure is beneficial [16]. To our knowledge, only the observational PARTAGE study and the study of Askari et al. had specifically investigated this relationship in nursing home populations. An inverse relationship was observed between SBP and mortality in the PARTAGE study [17, 18]. No association between blood pressure levels and cardiovascular outcome was seen in the study of Askari et al. [19]. The association between blood pressure levels, and especially orthostatic changes in blood pressure and mortality within the very elderly and frail patients remains unclear. Therefore, we aimed to investigate these associations in nursing home residents.

DESIGN AND METHODS

tud populationFor this prospective observational cohort study, patients were recruited from three departments of a nursing home facility in the northeastern region of the Netherlands (TriviumMeulenbeltZorg, Hengelo) [20]. Recruitment and all study procedures took place between September 2010 and December 2011. Patients at the psychogeriatric department were diagnosed with severe to very severe dementia and/or behavioral disorders. Patients at the somatic department received prolonged or permanent care whereas patients at the rehabilitation department were rehabilitated to return to their home situation as soon as possible. Exclusion criteria were a projected life expectancy of less than 4 weeks or admission to the hospice department.

R1

R2

R3

R4

R5

R6

R7

R8

R9

R10

R11

R12

R13

R14

R15

R16

R17

R18

R19

R20

R21

R22

R23

R24

R25

R26

R27

R28

R29

R30

R31

R32

R33

R34

R35

R36

R37

R38

R39

Chapter 8

120

ata collectionBaseline data included demographic characteristics, a full medical history including a history of cardiovascular disease, diabetes mellitus, hypertension, falls in the previous year (yes or no), and medication use. Patients were considered to have cardiovascular disease when they had a history of angina pectoris, myocardial infarction, percutaneous transluminal coronary angioplasty, coronary artery bypass grafting, stroke or transient ischaemic attack. The activities of daily living were measured with the Barthel-Index [21]. Blood pressure was measured following a standardized protocol, using an automated sphygmomanometer (Omron M6) [22]. If the automated sphygmomanometer displayed an error message, blood pressure was manually measured with a sphygmomanometer Heine Gamma XXL-T [23]. Renal function was not routinely assessed when patients were admitted to the nursing home. Creatinine clearance was calculated by the Cockcroft-Gault equation [24]. Trained medical staff performed all tests. Blood pressure was measured twice in supine position after 5 minutes of rest, and twice each at 1 and 3 minutes after postural change. The forearm of the patient was supported at heart level during the measurements in upright position [25]. The postural change was from supine to standing position, with the exception of patients who were unable to stand. For these patients the postural change was from lying to sitting position. OH was defined as a drop in SBP of > 20 mmHg or diastolic blood pressure (DBP) of > 10 mmHg after postural change compared to the mean value of the baseline measurements in supine position [3]. The presence or absence of characteristic symptoms of OH like light-headedness, syncope, or dizziness after postural change was scored, and the combination of OH and orthostatic complaints was described as symptomatic OH. For OHT, the most commonly used definition was used: an increase of > 20 mmHg in SBP [2, 13]. Pulse pressure (PP) was defined as the difference between mean SBP and DBP.

linical endpointsClinical endpoint was all-cause mortality. In 2015, the life status and cause of death were retrieved from records maintained by the nursing home and the general practitioners.

tatistical anal sesContinuous variables are presented as mean and standard deviation for normally distributed variables, or as median and interquartile range for non-normally distributed variables. Normality was evaluated using Q-Q plots and histograms. Cox proportional hazard modeling was used to investigate the relation between OH, OHT, the various blood pressure variables, and mortality with adjustment for selected confounders. We used four different models. In model 1, unadjusted analyses were performed. In model 2, only age and gender were taken into account as possible confounders. In model 3, we adjusted for age, gender, and Barthel index. In model 4, we additionally adjusted for the

R1

R2

R3

R4

R5

R6

R7

R8

R9

R10

R11

R12

R13

R14

R15

R16

R17

R18

R19

R20

R21

R22

R23

R24

R25

R26

R27

R28

R29

R30

R31

R32

R33

R34

R35

R36

R37

R38

R39


121

8

following variables: smoking, body mass index, hypertension (defined as a medical history of hypertension or use of antihypertensive medication), diabetes mellitus, history of cardiovascular disease and the total number of drugs. Cox regression analyses were repeated with the orthostatic changes in blood pressure categorized into OH, OHT or normal change (normal change used as the reference group). In case of significant associations in the Cox regression models, risk prediction capabilities were assessed with Harrell’s C statistics and the proportion of explained variance (R2) [26, 27]. The Schoenfeld residual plots were inspected for each predictor variable to check the assumption of proportional hazards. P values less than 0.05 were considered statistically significant. All statistical analyses were performed using SPSS software (version 22) and STATA (version 13).

thical approval and linical rial registrationThis study was performed in accordance with the Declaration of Helsinki. According to Dutch guidelines this study did not fall under the scope of the Medical Research Involving Human Subjects Act, and therefore this study did not need a formal approval of an accredited medical ethics committee. Written informed consent was obtained for all patients by the participating medical doctor or nurse. All data were analyzed anonymously. The ‘Strengthening the Reporting of Observational studies in Epidemiology’ (STROBE) statement was used to describe this observational cohort study [28]. The study was registered on ClinicalTrials.gov (NCT01362751).

RESULTS

Baseline characteristics of the study population are presented in table 1 and appendix table 1. A total of 290 patients were included: 106 patients at the psychogeriatric department, 56 patients at the somatic department, and 128 patients at the rehabilitation department. Mean age of the total study population was 80.8 (SD 9.9) years. OH was present in 106 out of 290 patients, resulting in a prevalence of 37% (95% confidence interval (CI): 31%-42%). OHT was present in 13% (95%CI: 10%-17%) of the patients. During a median follow-up period of 2.7 years, 175 (60%) patients died, of whom 28 patients (16%) from cardiovascular causes. Cause of death was unknown for 15 patients (9%). Due to the low cardiovascular disease (CVD) death rate, analyses were only performed for all-cause mortality. Deceased patients were older, had more often hypertension and dementia, and had lower scores on the Barthel index.

R1

R2

R3

R4

R5

R6

R7

R8

R9

R10

R11

R12

R13

R14

R15

R16

R17

R18

R19

R20

R21

R22

R23

R24

R25

R26

R27

R28

R29

R30

R31

R32

R33

R34

R35

R36

R37

R38

R39

Chapter 8

122


Total Deceased patients Survived patients P-valueCharacteristic N=290 N= 175 N= 115DemographicsAge 80.8 (9.9) 83.2 (8.4) 77.1 (10.9) <0.01Female Gender 206 (71.0) 127 (72.6) 79 (68.7) 0.48Mean body mass index, kg/m2

26.3 (5.4) 26.0 (5.2) 26.9 (5.7) 0.17Hypertension 226 (77.9) 144 (82.3) 82 (71.3) 0.03History of CVD 131 (45.2) 87 (49.7) 44 (38.3) 0.06

Diabetes mellitus 104 (35.0) 66 (37.7) 38 (33) 0.42Dementia 121 (41.7) 94 (53.7) 27 (23.5) <0.01Current smoker 32 (11.0) 16 (9.1) 16 (13.9) 0.21

Orthostatic hypotension 106/290 (36.6) 67 (38.3) 39 (33.9) 0.45Orthostatic complaints 71/288 (24.7) 37 (21.1) 34 (29.6) 0.11

Orthostatic hypertension 39/290 (13.4) 23 (13.1) 16 (13.9) 0.85

MeasurementsMean SBP supine 147 (22) 149 (23) 144 (21) 0.08Mean DBP supine 74 (11) 75 (12) 73 (10) 0.07Mean pulse frequency 74 (13) 74 (14) 74 (12) 0.68Mean pulse pressure 70 (58-87) 71 (58-87) 68 (58-85) 0.44Mean SBP standing 1 min 142 (24) 144 (26) 140 (22) 0.04Mean DBP standing 1 min 75 (14) 74 (13) 77 (14) 0.72Mean pulse frequency standing 1 min 81 (15) 80 (16) 82 (14) 0.23Mean SBP standing 3 min 144 (24) 146 (25) 141 (21) 0.02Mean DBP standing 3 min 76 (13) 75 (13) 76 (13) 0.98Mean pulse frequency standing 3 min 80 (15) 78 (16) 82 (14) 0.20

Creatinine clearance, ml/min* 57 (41-75) 54 (40-68) 65 (50-84) <0.01Score Barthel questionnaire 10 (6-15) 8 (4-13) 14 (8-17) <0.01

MedicationMean number of agents 8.7 (3.5) 9.0 (3.4) 8.4 (3.4) 0.13Antihypertensive medication 194 (66.9) 121 (69.1) 73 (63.5) 0.32- ACE-inhibitor 100 (34.5) 63 (36.0) 37 (32.2) 0.50- Beta-blocker 85 (29.3) 48 (27.4) 37 (32.2) 0.39- Diuretics 128 (44.1) 83 (48.0) 45 (38.5) 0.11

- Calcium channel blockers 42 (14.5) 27 (15.6) 15 (12.8) 0.51Benzodiazepines 150 (51.7) 85 (49.1) 65 (55.6) 0.28Antipsychotics 59 (20.3) 39 (22.5) 20 (17.1) 0.26

Antidepressants 69 (23.8) 44 (25.4) 25 (21.4) 0.43Opiods 32 (11.0) 14 (8.1) 18 (15.4) 0.05

Data are means (± SD), medians (interquartile range) or n (%). CVD = Cardiovascular disease. SBP = systolic blood pressure. DBP = diastolic blood pressure. * 70 missing values for Creatinine clearance.

R1

R2

R3

R4

R5

R6

R7

R8

R9

R10

R11

R12

R13

R14

R15

R16

R17

R18

R19

R20

R21

R22

R23

R24

R25

R26

R27

R28

R29

R30

R31

R32

R33

R34

R35

R36

R37

R38

R39


123

8

lood ressureTable 2 presents the hazard ratios of the various blood pressure variables with regard to mortality. In the fully adjusted model a significant relationship between DBP and all-cause mortality was observed. The overall mortality risk increased by 17% (95%CI: 2%-34%) for every 10-mmHg increase in DBP (Figure 1). No other significant associations between mortality and the various blood pressure indices were observed. A significant relationship between DBP and all-cause mortality was not seen in the separate patient groups. The results of the analyses regarding the risk prediction capabilities are presented in Appendix table 2. The Harrell’s C value for the crude model of DBP for all-cause mortality was 0.54 (95%CI 0.50-0.59). There were no differences between the Harrell’s C values within all models with and without DBP (appendix table 2). The R2 was slightly higher in the models with DBP compared to the models without DBP; 0.21 (95%CI: 0.15-0.33) and 0.19 (95%CI: 0.13-0.31), respectively. We performed additional cox regression analyses in which we additionally adjusted for renal function, peripheral artery disease, and diabetes mellitus. Results did not relevantly change (data not shown), except for the stratified analyses according to diabetes mellitus. The hazard ratio (HR) for DBP and mortality was 1.04 (95%CI 0.80-1.36) in diabetic patients and 1.21 (95%CI 1.02-1.42) in non-diabetic patients.

Figure 1.

Figure .

Figure 1. Survival curve; diastolic blood pressure and all-cause mortality.

R1

R2

R3

R4

R5

R6

R7

R8

R9

R10

R11

R12

R13

R14

R15

R16

R17

R18

R19

R20

R21

R22

R23

R24

R25

R26

R27

R28

R29

R30

R31

R32

R33

R34

R35

R36

R37

R38

R39

Chapter 8

124

and The prevalence of OH ranged from 28.6% (95%CI: 16.8%-40.4%) at the somatic department, to 36.7% (95%CI: 28.4%-45.1%) and 40.6% (95%CI: 31.3%- 50.0%) at the rehabilitation and the psychogeriatric departments, respectively. The prevalence of OHT was 8.9% (95%CI: 1.4%-16.4%) at the somatic department, 13.3% (95%CI: 7.4%-19.2%) at the rehabilitation department, and 16.0% (95%CI: 9.0%-22.9%) at the psychogeriatric department.OH and OHT were not associated with all-cause mortality after adjustment for the selected confounders (table 2). For patients at the psychogeriatric department, OH was associated with an increased all-cause mortality risk (HR 1.71 [95%CI: 1.08-2.71]) (figure 2). The hazard ratio of OHT was 0.61 [95%CI 0.32-1.19] in this subgroup. Figure 2 illustrates the relationship for OH and OHT in psychogeriatric patients. No associations between OH and mortality were seen in the other patient groups. The Harrell’s C values were higher in all the models with OH (Appendix table 2). The largest increase in Harrell’s C value were observed for adding OH to the age- and gender adjusted model. The Harrell’s C values for the fully adjusted models with and without OH were 0.68 (95%CI: 0.62-0.75) and 0.66 (95%CI: 0.60-0.73), respectively. The R2 was higher in the models with OH compared to the models without OH; 0.17 (95%CI: 0.10-0.42) and 0.15 (95%CI: 0.07-0.34), respectively.

Figure 1.

Figure .

Figure 2. Survival curve; OH and OHT and all-cause mortality in psychogeriatric patients.

Using patients with a normal orthostatic change in blood pressure as the reference group, OH was only associated with an increased all-cause mortality risk in the unadjusted model in the psychogeriatric patient group (HR 1.61 [95%CI 1.02-2.53]) (Appendix table 3). No other significant associations between OH, OHT and mortality were observed, compared to the reference group.

R1

R2

R3

R4

R5

R6

R7

R8

R9

R10

R11

R12

R13

R14

R15

R16

R17

R18

R19

R20

R21

R22

R23

R24

R25

R26

R27

R28

R29

R30

R31

R32

R33

R34

R35

R36

R37

R38

R39


125

8

Tabl

e 2.

Cox

regr

essi

on a

naly

ses;

haz

ard

ratio

s fo

r al

l-cau

se m

orta

lity.

Tota

l Gro

upSo

mati

c w

ard

N=5

6 (†

30)

Psyc

hoge

riat

ric

war

dN

=106

(†85

)Re

habi

litati

on w

ard

N=1

28 (†

60)

HR

(95%

CI)

HR

(95%

CI)

HR

(95%

CI)

HR

(95%

CI)

SBP

(mod

el 1

)1.

06 (1

.00-

1.14

)1.

13 (0

.98-

1.31

)1.

06 (0

.95-

1.17

)1.

02 (0

.90-

1.15

)SB

P (m

odel

2)

1.04

(0.9

7-1.

12)

1.10

(0.9

4-1.

31)

1.05

(0.9

5-1.

17)

0.99

(0.8

9-1.

12)

SBP

(mod

el 3

)1.

04 (0

.97-

1.12

)1.

10 (0

.94-

1.29

)1.

06 (0

.95-

1.18

)0.

99 (0

.89-

1.12

)SB

P (m

odel

4)

1.03

(0.9

6-1.

12)

1.12

(0.9

3-1.

33)

1.06

(0.9

5-1.

20)

0.98

(0.8

7-1.

10)

DBP

(mod

el 1

)1.

16 (1

.01-

1.33

)0.

98 (0

.71-

1.34

)1.

04 (0

.87-

1.27

)1.

23 (0

.94-

1.61

)D

BP (m

odel

2)

1.24

(1.0

8-1.

42)

1.04

(0.7

7-1.

42)

1.10

(0.9

0-1.

34)

1.33

(1.0

3-1.

74)

DBP

(mod

el 3

)1.

17 (1

.02-

1.34

)1.

04 (0

.77-

1.41

)1.

03 (0

.84-

1.26

)1.

29 (1

.00-

1.68

)D

BP (m

odel

4)

1.17

(1.0

2-1.

34)

1.17

(0.8

2-1.

66)

1.00

(0.8

2-1.

22)

1.21

(0.9

3-1.

57)

Puls

e Pr

essu

re (m

odel

1)

1.04

(0.9

6-1.

13)

1.21

(1.0

1-1.

44)

1.05

(0.9

4-1.

18)

0.97

(0.8

4-1.

12)

Puls

e Pr

essu

re (m

odel

2)

0.98

(0.9

0-1.

06)

1.12

(0.9

3-1.

34)

1.03

(0.9

1-1.

17)

0.91

(0.7

9-1.

06)

Puls

e Pr

essu

re (m

odel

3)

1.00

(0.9

2-1.

08)

1.13

(0.9

3-1.

36)

1.08

(0.9

5-1.

23)

0.92

(0.8

1-1.

06)

Puls

e Pr

essu

re (m

odel

4)

0.99

(0.9

0-1.

07)

1.12

(0.9

0-1.

38)

1.10

(0.9

5-1.

27)

0.92

(0.8

0-1.

06)

OH

(mod

el 1

)1.

21 (0

.89-

1.64

)1.

03 (0

.47-

2.25

)1.

81 (1

.18-

2.79

)0.

88 (0

.51-

1.52

)O

H (m

odel

2)

1.05

(0.7

7-1.

44)

0.69

(0.3

1-1.

55)

1.80

(1.1

6-2.

80)

0.78

(0.4

5-1.

35)

OH

(mod

el 3

)1.

05 (0

.77-

1.44

)0.

77 (0

.34-

1.75

)1.

56 (0

.99-

2.42

)0.

76 (0

.44-

1.32

)O

H (m

odel

4)

1.01

(0.7

4-1.

38)

0.81

(0.3

3-1.

97)

1.71

(1.0

8-2.

71)

0.70

(0.4

0-1.

23)

OH

T (m

odel

1)

0.81

(0.5

2-1.

27)

1.21

(0.3

7-4.

01)

0.60

(0.3

2-1.

10)

0.80

(0.3

6-1.

76)

OH

T (m

odel

2)

0.80

(0.5

1-1.

25)

1.59

(0.4

7-5.

41)

0.55

(0.3

0-1.

02)

0.84

(0.3

8-1.

87)

OH

T (m

odel

3)

0.87

(0.5

5-1.

39)

1.89

(0.4

1-8.

68)

0.57

(0.3

1-1.

05)

0.90

(0.4

0-2.

02)

OH

T (m

odel

4)

0.86

(0.5

4-1.

37)

1.41

(0.2

8-7.

07)

0.61

(0.3

2-1.

19)

0.88

(0.3

8-2.

00)

The

haza

rd r

atios

of

puls

e pr

essu

re, S

BP a

nd D

BP r

efer

to

a pr

essu

re in

crea

se o

f 10

mm

Hg.

Mod

el 1

: Una

djus

ted,

Mod

el 2

: Adj

uste

d fo

r ag

e an

d ge

nder

, mod

el 3

: Adj

uste

d fo

r age

, gen

der,

Bart

hel i

ndex

. mod

el 4

: Adj

uste

d fo

r age

, gen

der,

smok

ing,

DM

, his

tory

of C

VD, h

yper

tens

ion,

BM

I, N

umbe

r of

med

icati

on, B

arth

el in

dex.

† O

bser

ved

num

ber

of d

eath

s.

R1

R2

R3

R4

R5

R6

R7

R8

R9

R10

R11

R12

R13

R14

R15

R16

R17

R18

R19

R20

R21

R22

R23

R24

R25

R26

R27

R28

R29

R30

R31

R32

R33

R34

R35

R36

R37

R38

R39

Chapter 8

126

DISCUSSION

In this prospective observational study, higher DBP was related to increased all-cause mortality in elderly nursing home patients. In specifically the group of psychogeriatric patients, OH was found to be related to all-cause mortality.

lood pressurePrevious studies showed that a low DBP was associated with an increased all-cause mortality risk, especially in the oldest and frail individuals [29, 30]. Concerning SBP, a combination of a SBP less than 130 mmHg and use of antihypertensive drugs was associated with a higher mortality risk in the PARTAGE study [17]. No association between blood pressure levels and cardiovascular outcome was seen in the study of Askari et al. That study described that the role of multiple medical diseases in this frail nursing home population could be more important predictors of cardiovascular outcome than blood pressure itself [19]. We only observed a significant relationship between DBP and all-cause mortality in the total study population and no significant relationship between DBP and all-cause mortality was seen in the separate patients groups. Remarkably, in an observational study amongst patients aged 65 years and older from the general population, both a positive and inverse relation between DBP and mortality has been described. In patients without cognitive and physical dysfunction, higher DBP was related to increased mortality, whereas for frail elderly patients this relationship was the opposite [29]. Furthermore, the influence of frailty on the relationship between blood pressure levels and mortality was also shown in other studies [31-33]. In the study by Odden et al. stratified analyses according to frailty were performed; in non-frail patients a positive relationship was described and in frail patients blood pressure was not related to mortality [32]. The level of frailty may also be the explanation for the positive relationship found in our study, as the overall results were comparable to the results in the rehabilitation group. These patients were indeed the most vital patients as their scores on the Barthel index were the highest, and the prevalence of dementia (7%) was the lowest. Strikingly, the relationship between DBP and mortality was only seen in the non-diabetic patient group. We hypothesized that the level of frailty of the diabetic patient group could be an explanation. However, the majority of the non-diabetic patients (60%) were patients admitted at the psychogeriatric or somatic department, which are frail patients. In our opinion, the observed results are a matter of coincidence.Although our study showed that DBP is an independent risk factor for mortality, its practical implications still remain to be determined. Based on the lack of increase in Harrell’s C values when adding DBP to the adjusted models, one may conclude that the additional value of

R1

R2

R3

R4

R5

R6

R7

R8

R9

R10

R11

R12

R13

R14

R15

R16

R17

R18

R19

R20

R21

R22

R23

R24

R25

R26

R27

R28

R29

R30

R31

R32

R33

R34

R35

R36

R37

R38

R39


127

8

DBP in mortality prediction apparently is very limited. It is important to realize that this study group is a frail group of patients with much comorbidity, probably more determinative to mortality than DBP itself. Even in the fully adjusted models, the C value was lower than 0.70, indicating the poor predictive capability of the overall model. To our knowledge, only the observational PARTAGE study and the study of Askari et al. had specifically investigated the relationship between blood pressure, pulse pressure and mortality in a nursing home population [17-19]. Results of present study were comparable to these previous studies.

and In our study, only OH was related to increased all-cause mortality, specifically in patients at the psychogeriatric department. The direction of the hazard ratio for OHT was quite the opposite, although this association was not significant. Dementia, and its (cardiovascular) causes, may explain these observations. Arterial stiffness and lower blood pressure will cause hypoperfusion of the brain, and are associated with accelerated cognitive decline [34, 35]. Also, autonomic dysfunction within neurodegenerative diseases like dementia could give OH [34]. Apparently, within a cognitive impaired population, a decrease in blood pressure upon standing is related to increased mortality, whereas an increase may be protective. Because the prevalence of dementia on the psychogeriatric department (94%) is much higher than is seen at the somatic (21%) and rehabilitation (7%) department, this could explain the different impact of OH on all-cause mortality. In a meta-analysis of observational studies, OH was related to an increased risk of all-cause mortality but not to cardiovascular mortality [8]. It could be that OH is a marker of underlying CVD or frailty, rather than an independent risk factor for CVD [2, 6, 7]. Amongst community-dwelling patients in Italy, OH was related to higher non-CVD mortality and OHT to higher all-cause and CVD-related mortality [2]. Other studies also found that OHT is associated with hypertensive target-organ damage, cardiovascular events and all-cause and CVD-related mortality [12, 13]. Besides analyzing OH as risk factor for total mortality, we performed Harrell’s C and R2

statistics to study if OH improved the risk model for mortality in the psychogeriatric patient group. Because both Harrell’s C and R2 showed only small improvements when adding OH to the adjusted models, one may conclude that the additional values of OH in risk prediction seems limited. As there is no universally accepted definition for OHT, we used the most commonly used definition: an increase of > 20 mmHg in SBP [2, 13]. The definition is arbitrary because other studies used varying definitions [13]. Most definitions have one common feature: they are solely based on change in SBP, whereas OH is defined as a change in SBP or DBP. Therefore, we performed a post-hoc analysis in which OHT was defined as an increase in SBP of > 20

R1

R2

R3

R4

R5

R6

R7

R8

R9

R10

R11

R12

R13

R14

R15

R16

R17

R18

R19

R20

R21

R22

R23

R24

R25

R26

R27

R28

R29

R30

R31

R32

R33

R34

R35

R36

R37

R38

R39

Chapter 8

128

mmHg or DBP of > 10 mmHg after postural change compared to the mean value of the baseline measurements in supine position (comparable to the OH definition). The hazard ratio of OHT, in the fully adjusted model, regarding total mortality in the total study group was 0.58 (95%CI 0.41-0.82). Strikingly, with this definition of OHT, 21 patients met both the criteria of OH and of OHT. Other study groups may therefore have chosen to define OHT solely on SBP.

trengths and li itationsThe main strength of this study is the variety in the nursing home population. To the best of our knowledge, there are no previous studies that have investigated different orthostatic changes and its relationship with mortality in different groups (psychogeriatric, rehabilitation, and somatic patients) of nursing home residents. The main limitations are the small study sample (decreasing the precision of the observed findings) and the observational design; therefore, establishing a causal relation is not possible. Furthermore, blood pressure was only measured once during the follow-up period, which also influenced accurate classification of patients as having OH or OHT or not.

In summary, higher DBP was related to higher all-cause mortality rates in a nursing home population, but this association is probably only applicable to the most vital nursing home residents (the rehabilitation group). OH was related to all-cause mortality in the psychogeriatric group of nursing home patients, whereas a beneficial effect of OHT could not be excluded. The predictive capabilities of both DBP and OH are rather small with respect to mortality.

R1

R2

R3

R4

R5

R6

R7

R8

R9

R10

R11

R12

R13

R14

R15

R16

R17

R18

R19

R20

R21

R22

R23

R24

R25

R26

R27

R28

R29

R30

R31

R32

R33

R34

R35

R36

R37

R38

R39


129

8

REFERENCES

1. Franklin SS, Gustin Wt, Wong ND, et al. Hemodynamic patterns of age-related changes in blood pressure. The Framingham Heart Study. Circulation. 1997 Jul 1;96(1):308-15.











12. Kario K. Orthostatic hypertension: a measure of blood pressure variation for predicting cardiovascular risk. Circulation journal : official journal of the Japanese Circulation Society. 2009 Jun;73(6):1002-7.

13. Kario K. Orthostatic hypertension-a new haemodynamic cardiovascular risk factor. Nature reviews Nephrology. 2013 Dec;9(12):726-38.

14. Rastas S, Pirttila T, Viramo P, et al. Association between blood pressure and survival over 9 years in a general population aged 85 and older. Journal of the American Geriatrics Society. 2006 Jun;54(6):912-8.

15. van Bemmel T, Gussekloo J, Westendorp RG, Blauw GJ. In a population-based prospective study, no association between high blood pressure and mortality after age 85 years. Journal of hypertension. 2006 Feb;24(2):287-92.




R1

R2

R3

R4

R5

R6

R7

R8

R9

R10

R11

R12

R13

R14

R15

R16

R17

R18

R19

R20

R21

R22

R23

R24

R25

R26

R27

R28

R29

R30

R31

R32

R33

R34

R35

R36

R37

R38

R39

Chapter 8

130






24. Cockcroft DW, Gault MH. Prediction of creatinine clearance from serum creatinine. Nephron. 1976;16(1):31-41.


26. Schemper M, Henderson R. Predictive accuracy and explained variation in Cox regression. Biometrics. 2000 Mar;56(1):249-55.

27. Choodari-Oskooei B, Royston P Fau - Parmar MKB, Parmar MK. A simulation study of predictive ability measures in a survival model I: explained variation measures. 20120918 DCOM- 20130305(1097-0258 (Electronic)).


29. Post Hospers G, Smulders YM, Maier AB, Deeg DJ, Muller M. Relation between blood pressure and mortality risk in an older population: role of chronological and biological age. Journal of internal medicine. 2015 Apr;277(4):488-97.



32. Odden MC, Peralta CA, Haan MN, Covinsky KE. Rethinking the association of high blood pressure with mortality in elderly adults: the impact of frailty. Archives of internal medicine. 2012 Aug 13;172(15):1162-8.

33. Rockwood MR, Howlett SE, Rockwood K. Orthostatic hypotension (OH) and mortality in relation to age, blood pressure and frailty. Archives of gerontology and geriatrics. 2012 May-Jun;54(3):e255-60.

34. Frewen J Fau - Savva GM, Savva Gm Fau - Boyle G, Boyle G Fau - Finucane C, Finucane C Fau - Kenny RA, Kenny RA. Cognitive performance in orthostatic hypotension: findings from a nationally representative sample. 20140829 DCOM- 20140912(1532-5415 (Electronic)).

35. Conway KS, Forbang N, Beben T, et al. Relationship Between 24-Hour Ambulatory Blood Pressure and Cognitive Function in Community-Living Older Adults: The UCSD Ambulatory Blood Pressure Study. American journal of hypertension. 2015 Apr 19.

R1

R2

R3

R4

R5

R6

R7

R8

R9

R10

R11

R12

R13

R14

R15

R16

R17

R18

R19

R20

R21

R22

R23

R24

R25

R26

R27

R28

R29

R30

R31

R32

R33

R34

R35

R36

R37

R38

R39


131

8

SUPP

LEM

ENTA

L D

ATA

App

endi

x ta

ble

1. T

ypes

of m

edic

ation

tota

l pop

ulati

on a

nd d

iffer

ent g

roup

s of

pati

ents

.

Tota

lPs

ycho

geri

atri

c de

part

men

tSo

mati

c de

part

men

tRe

habi

litati

on d

epar

tmen

tCh

arac

teri

stic

N=2

90N

=106

N=5

6N

=128

Med

icati

onM

ean

num

ber

of a

gent

s 8.

7 (3

.5)

8.5

(3.0

)9.

5 (4

.2)

8.6

(3.6

)A

ntihy

pert

ensi

ve m

edic

ation

- D

iure

tics

- Be

ta b

lock

ers

- Ca

lciu

m c

hann

el b

lock

ers

- AC

E in

hibi

tors

194

(66.

9)12

8 (4

4.1)

85 (2

9.3)

42 (1

4.5)

100

(34.

5)

68 (6

4.2)

43 (4

0.6)

23 (2

1.7)

13 (1

2.3)

42 (3

9.6)

35 (6

2.5)

23 (4

1.1)

17 (3

0.4)

8 (1

4.3)

15 (2

6.8)

91 (7

1.1)

62 (4

8.4)

45 (3

5.2)

21 (1

6.4)

43 (3

3.6)

Benz

odia

zepi

nes

150

(51.

7)52

(49.

1)34

(60.

7)64

(50.

X)A

ntips

ycho

tics

59 (2

0.3)

34 (3

2.1)

11 (1

9.6)

14 (1

0.9)

Anti

depr

essa

nts

69 (2

3.8)

32 (3

0.2)

18 (3

2.1)

19 (1

4.8)

Opi

oids

32 (1

1.0)

10 (9

.4)

9 (1

6.1)

13 (1

0.2)

Dat

a ar

e m

eans

(± S

D),

med

ians

(int

erqu

artil

e ra

nge)

or

n (%

).

R1

R2

R3

R4

R5

R6

R7

R8

R9

R10

R11

R12

R13

R14

R15

R16

R17

R18

R19

R20

R21

R22

R23

R24

R25

R26

R27

R28

R29

R30

R31

R32

R33

R34

R35

R36

R37

R38

R39

Chapter 8

132

App

endi

x ta

ble

2. C

ox re

gres

sion

ana

lyse

s; h

azar

d ra

tios

for

all-c

ause

mor

talit

y. T

he c

ompa

riso

n of

pre

dicti

ve c

apab

ility

for

mor

talit

y de

term

ined

by

the

Har

rel’s

C s

tatis

tic, a

nd th

e R2

for

addi

ng D

BP o

r O

H to

mod

els

2,3

and

4, re

spec

tivel

y.

Har

rell’

s C

R2

Har

rell’

s C

R2

HR

(95%

CI)

(95%

CI)

(95%

CI)

(95%

CI)*

(95%

CI)*

Tota

l stu

dy s

ampl

e N

=290

Dia

stol

ic B

P (m

odel

1)

1.16

(1.0

1-1.

33)

0.54

(0.5

0-0.

59)

0.01

(0.0

0-0.

05)

NA

NA

Dia

stol

ic B

P (m

odel

2)

1.24

(1.0

8-1.

42)

0.62

(0.5

8-0.

66)

0.09

(0.0

4-0.

18)

0.61

(0.5

6-0.

65)

0.06

(0.0

2-0.

14)

Dia

stol

ic B

P (m

odel

3)

1.17

(1.0

2-1.

34)

0.67

(0.6

3-0.

71)

0.20

(0.1

2-0.

30)

0.67

(0.6

3-0.

71)

0.18

(0.1

1-0.

28)

Dia

stol

ic B

P (m

odel

4)

1.17

(1.0

2-1.

34)

0.68

(0.6

5-0.

72)

0.21

(0.1

5-0.

33)

0.68

(0.6

4-0.

72)

0.19

(0.1

3-0.

31)

Psyc

hoge

riat

ric

pati

ents

N=1

06O

H (m

odel

1)

1.81

(1.1

8-2.

79)

0.58

(0.5

3-0.

64)

0.07

(0.0

0-0.

21)

NA

NA

OH

(mod

el 2

)1.

80 (1

.16-

2.80

)0.

62 (0

.55-

0.68

)0.

06 (0

.00-

0.19

)0.

57 (0

.50-

0.63

)0.

01 (0

.00-

0.17

)O

H (m

odel

3)

1.56

(0.9

9-2.

42)

0.65

(0.5

9-0.

71)

0.15

(0.0

6-0.

32)

0.64

(0.5

7-0.

70)

0.15

(0.0

6-0.

29)

OH

(mod

el 4

)1.

71 (1

.08-

2.71

)0.

68 (0

.62-

0.75

)0.

17 (0

.10-

0.42

)0.

66 (0

.60-

0.73

)0.

15 (0

.07-

0.34

)

The

haza

rd ra

tios

of D

BP re

fer t

o a

pres

sure

incr

ease

of 1

0 m

mH

g. M

odel

1: U

nadj

uste

d, M

odel

2: A

djus

ted

for a

ge a

nd g

ende

r, m

odel

3: A

djus

ted

for

age,

gen

der,

Bart

hel i

ndex

. mod

el 4

: Adj

uste

d fo

r age

, gen

der,

smok

ing,

DM

, his

tory

of C

VD, h

yper

tens

ion,

BM

I, N

umbe

r of m

edic

ation

, Bar

thel

inde

x.

* H

arre

ll’s

C va

lues

and

R2 fo

r th

e m

odel

s w

ithou

t DBP

or

OH

.

R1

R2

R3

R4

R5

R6

R7

R8

R9

R10

R11

R12

R13

R14

R15

R16

R17

R18

R19

R20

R21

R22

R23

R24

R25

R26

R27

R28

R29

R30

R31

R32

R33

R34

R35

R36

R37

R38

R39


133

8

App

endi

x ta

ble

3. C

ox re

gres

sion

ana

lyse

s; h

azar

d ra

tios

for

all-c

ause

mor

talit

y.

Tota

l Gro

upSo

mati

c w

ard

N=5

6 (†

30)

Psyc

hoge

riat

ric

war

dN

=106

(†85

)Re

habi

litati

on w

ard

N=1

28 (†

60)

HR

(95%

CI)

HR

(95%

CI)

HR

(95%

CI)

HR

(95%

CI)

Nor

mal

cha

nge

(mod

el 1

)1

[ref

eren

ce]

1 [r

efer

ence

]1

[ref

eren

ce]

1 [r

efer

ence

]O

H

1.15

(0.8

4-1.

59)

1.06

(0.4

8-2.

36)

1.61

(1.0

2-2.

53)

0.83

(0.4

7-1.

45)

OH

T 0.

80 (0

.49-

1.32

)1.

24 (0

.36-

4.20

)0.

62 (0

.31-

1.25

)0.

72 (0

.30-

1.72

)N

orm

al c

hang

e (m

odel

2)

1 [r

efer

ence

]1

[ref

eren

ce]

1 [r

efer

ence

]1

[ref

eren

ce]

OH

0.

99 (0

.72-

1.38

)0.

72 (0

.32-

1.63

)1.

55 (0

.97-

2.47

)0.

73 (0

.42-

1.30

)O

HT

0.75

(0.4

6-1.

24)

1.43

(0.4

1-4.

97)

0.57

(0.2

8-1.

15)

0.72

(0.3

0-1.

74)

Nor

mal

cha

nge

(mod

el 3

)1

[ref

eren

ce]

1 [r

efer

ence

]1

[ref

eren

ce]

1 [r

efer

ence

]O

H

1.02

(0.7

4-1.

40)

0.80

(0.3

5-1.

86)

1.33

(0.8

4-2.

13)

0.73

(0.4

1-1.

28)

OH

T 0.

83 (0

.50-

1.38

)1.

74 (0

.37-

8.23

)0.

56 (0

.28-

1.13

)0.

78 (0

.32-

1.90

)N

orm

al c

hang

e (m

odel

4)

1 [r

efer

ence

]1

[ref

eren

ce]

1 [r

efer

ence

]1

[ref

eren

ce]

OH

0.

97 (0

.70-

1.34

)0.

82 (0

.34-

2.03

)1.

53 (0

.93-

2.49

)0.

67 (0

.37-

1.19

)O

HT

0.80

(0.4

8-1.

35)

1.34

(0.2

6-6.

82)

0.64

(0.3

0-1.

37)

0.74

(0.3

0-1.

81)

The

haza

rd r

atios

of p

ulse

pre

ssur

e, S

BP a

nd D

BP r

efer

to

a pr

essu

re in

crea

se o

f 10

mm

Hg.

Cox

reg

ress

ion

anal

yses

wer

e re

peat

ed w

ith o

rtho

stati

c bl

ood

pres

sure

cha

nges

cat

egor

ized

into

OH

, OH

T or

nor

mal

cha

nge

(nor

mal

cha

nge

used

as

the

refe

renc

e gr

oup)

. Mod

el 1

: Una

djus

ted,

Mod

el 2

: A

djus

ted

for

age

and

gend

er, m

odel

3:

Adj

uste

d fo

r ag

e, g

ende

r, Ba

rthe

l ind

ex, m

odel

4:

Adj

uste

d fo

r ag

e, g

ende

r, sm

okin

g, D

M, h

isto

ry o

f CV

D,

hype

rten

sion

, BM

I, N

umbe

r of

med

icati

on, B

arth

el in

dex.

† O

bser

ved

num

ber

of d

eath

s.

CHAPTER 9The clinical relevance of orthostatic

hypotension in elderly patients

Accepted for publication as:

Hartog LC, Kleefstra N, Luigies RH, De Rooij SE, Bilo HJG, Van Hateren KJJ. The

clinical relevance of orthostatic hypotension in elderly patients. Geriatrics &

Gerontology International.

R1

R2

R3

R4

R5

R6

R7

R8

R9

R10

R11

R12

R13

R14

R15

R16

R17

R18

R19

R20

R21

R22

R23

R24

R25

R26

R27

R28

R29

R30

R31

R32

R33

R34

R35

R36

R37

R38

R39

Chapter 9

136

ABSTRACT

i : Orthostatic hypotension (OH) is highly prevalent in old age. The impact of OH on orthostatic complaints and falling is questionable. We wondered if the consensus definition of OH plays an essential role in the accuracy and direction of the prediction of these endpoints. We aimed to explore the relation between different definitions of OH, including relative decrease of blood pressure, and orthostatic complaints and falling.

Methods: A cross-sectional study. 1415 participants aged > 65 years visiting a mobile fall-prevention team. Data on blood pressure, orthostatic complaints and previous fall incidents were collected. The CAREFALL Triage Instrument was administered. Multivariate binary logistic regression analyses were performed to assess the association of different definitions of OH and orthostatic complaints or falling. Ten different definitions of OH based on different relative declines of systolic blood pressure (SBP) and diastolic blood pressure (DBP) were defined.

esults: The 2011 consensus definition of OH was not related to orthostatic complaints (Odds Ratio (OR) 1.07 (95% Confidence Interval (CI) 0.68-1.69)) or previous fall incidents (OR 1.08 (95%CI 0.83-1.41)). A >25% SBP decrease was significantly related to orthostatic complaints (OR 2.81 (95%CI 1.31-6.05)) and a >25% DBP decrease was related to previous fall incidents (OR 2.56 (95%CI 1.08-6.09)).

onclusions: With the exception of a decrease of > 25% SBP or DBP, the clinical relevance of incidental OH blood pressure measurements seems very limited with respect to orthostatic complaints or fall incidents in elderly patients. Using relative decreases may be more appropriate in clinical practice.

R1

R2

R3

R4

R5

R6

R7

R8

R9

R10

R11

R12

R13

R14

R15

R16

R17

R18

R19

R20

R21

R22

R23

R24

R25

R26

R27

R28

R29

R30

R31

R32

R33

R34

R35

R36

R37

R38

R39

he clinical relevance o orthostatic h potension in elderl patients

137

9

INTRODUCTION

Orthostatic changes in blood pressure are highly prevalent in old age [1]. The prevalence of orthostatic hypotension (OH) varies from 8-67%, depending on the population studied and the definition used [2-7].In 1996, OH was defined as a drop in systolic blood pressure (SBP) of >20 mmHg or diastolic blood pressure (DBP) of >10 mmHg after postural change [1, 8]. This definition was based on clinical judgment, as epidemiologic data were unavailable at that time. In response to the study of Fedorowski et al. in 2011 [9], an updated consensus statement was published [10], and a higher cut-off point (DSBP>30 mmHg) was advised for patients with supine hypertension (SBP>160 mmHg). The evidence supporting an independent relationship between this widely accepted definition of OH and falling or mortality is still sparse. Many studies were not able to adjust for important confounders and thereby meta-analyses have not yet been performed [11, 12]. We wondered if the standard 1996 and 2011 definitions of OH are the best definitions in relation to orthostatic complaints or falling. Since current definitions only discriminate for hypo- and hypertension broadly, it might be more accurate to define relative decline to describe OH [9, 10]. A previous study has shown that different definitions of OH defined as absolute declines were not independent predictors of falling [13]. No study has investigated the relationship between a relative decline of blood pressure, to define OH, and falling or orthostatic complaints. Therefore, we aimed to explore the relation between different definitions of OH and orthostatic complaints and falling.

METHODS

For this cross-sectional study, community-dwelling people aged 65 years or older were recruited between 2007 and 2009. A healthcare insurance company in the Netherlands randomly invited their clients aged >60 years by mail to participate in the fall risk screening and prevention program. Mail was based on random selection of zip codes across the country. A total of approximately 2250 invitations were mailed. Interested clients could make a phone call for an appointment and received a visit from a fall prevention team with a mobile diagnostic centre at their home addresses. Only patients aged 65 years or older were included in present study.Baseline data included demographic characteristics, a medical history including a history of cardiovascular disease, diabetes mellitus, hypertension, and medication use. All participants were asked to complete the CAREFALL Triage Instrument (CTI) to obtain fall history and risk factors for recurrent falling [14]. This instrument contains 45 questions to identify risk factors for recurrent falling in elderly patients [14]. Although the CTI has been developed to obtain fall history, not all included participants had fallen.

R1

R2

R3

R4

R5

R6

R7

R8

R9

R10

R11

R12

R13

R14

R15

R16

R17

R18

R19

R20

R21

R22

R23

R24

R25

R26

R27

R28

R29

R30

R31

R32

R33

R34

R35

R36

R37

R38

R39

Chapter 9

138

Blood pressure was measured after 5 minutes of rest (lying), and 3 minutes after postural change. The consensus definition of 1996 and 2011 was used to define OH [1, 8, 10]. Orthostatic complaints were defined as symptoms of light-headedness, syncope, or dizziness questioned by the CTI questionnaire (see Appendix I in the supplementary data). The questions regarding orthostatic complaints were related to a fall incident rather than a postural change. Multivariate binary logistic regression analyses were performed to assess the association of different definitions of OH and orthostatic complaints or falling. Next to the consensus definitions 1996 and 2011, we defined 10 different definitions of OH based on different relative declines of SBP and DBP after postural change. Participants were categorized as having OH based on the different definitions as a categorical variable. Additionally, the absolute decline of SBP and DBP as a continuous variable was investigated. We adjusted for the following confounders: age, gender, diabetes mellitus, history of stroke, myocardial infarction, difficulty walking, and use of antihypertensive drugs.

thical approvalThis study was performed in accordance with the Declaration of Helsinki. This study was part of a larger, on-going study and was approved by the medical ethics committee of the Academic Medical Center, Amsterdam, the Netherlands. The treating staff obtained written informed consent from all patients and data were analyzed anonymously.

undingFunding was provided by the Dutch healthcare insurance company OHRA, who also provided the mobile diagnostic centre.

RESULTS

In total 1415 participants aged 65 years and older were included in this study (table 1). 504 participants were excluded from the analyses regarding the association of different definitions of OH and orthostatic complaints because of missing values of orthostatic complaints resulting in 911 (64.4%) participants. Besides, 8 participants were excluded from the analyses regarding previous fall incidents because of missing values resulting in 1407 participants (see Appendix II in the supplementary data).

R1

R2

R3

R4

R5

R6

R7

R8

R9

R10

R11

R12

R13

R14

R15

R16

R17

R18

R19

R20

R21

R22

R23

R24

R25

R26

R27

R28

R29

R30

R31

R32

R33

R34

R35

R36

R37

R38

R39


139

9

Tabl

e 1.

Bas

elin

e Ch

arac

teri

stics

Tota

l Stu

dy g

roup

N=1

415

Stud

y gr

oup

Ana

lysi

s O

H a

ndO

rtho

stati

c co

mpl

aint

s

N=9

11

Stud

y gr

oup

Ana

lysi

s O

H a

nd

Prev

ious

fall

inci

dent

s

N=1

407

Dem

ogra

phic

sA

ge (y

ears

) (SD

)78

(7)

79 (6

)78

(7)

Gen

der

(% F

emal

e)88

8 (6

3)57

3 (6

3)88

3 (6

3)H

yper

tens

ion

(%)

577

(41)

371

(41)

574

(41)

Dia

bete

s M

ellit

us (%

)16

5 (1

2)83

(9)

165

(12)

Stro

ke/c

ereb

ral i

nfar

ction

(%)

106

(8)

68 (8

)10

5 (8

)M

yoca

rdia

l inf

arcti

on (%

)12

7 (9

)60

(7)

127

(9)

Mea

sure

men

tsSB

P ly

ing

mm

Hg

(SD

)15

9 (2

2)15

9 (2

3)15

9 (2

2)D

BP ly

ing

mm

Hg

(SD

)86

(12)

86 (1

2)86

(12)

SBP

3 m

in a

fter

pos

tura

l cha

nge

mm

Hg

(SD

)14

9 (2

5)14

8 (2

5)14

9 (2

5)D

BP 3

min

aft

er p

ostu

ral c

hang

e m

mH

g (S

D)

83 (1

3)83

(13)

83 (1

3)Fa

lls la

st y

ear

(%)

359

(25)

284

(31)

359

(26)

OH

con

sens

us d

efini

tion

1996

[1] (

%)

531

(38)

361

(40)

528

(38)

Ort

host

atic

com

plai

nts

(%)

102/

911

(11)

a10

2 (1

1)10

1/91

1 (1

1)a

Diffi

culty

wal

king

(%)

528

(37)

344

(38)

527

(38)

Med

icati

onN

umbe

r of

age

nts

(IQR)

0 (0

-2)

1 (0

-2)

0 (0

-2)

Anti

hype

rten

sive

med

icati

on (%

)41

9 (3

0)34

2 (3

8)41

7 (3

0)

Dat

a ar

e m

eans

(st

anda

rd d

evia

tion)

or

med

ian

(Inte

rqua

rtile

ran

ge)

or N

(%

). SB

P= s

ysto

lic b

lood

pre

ssur

e; D

BP=

dias

tolic

blo

od p

ress

ure;

CTI

= CA

REFA

LL T

riag

e In

stru

men

t. a

504

mis

sing

val

ues.

R1

R2

R3

R4

R5

R6

R7

R8

R9

R10

R11

R12

R13

R14

R15

R16

R17

R18

R19

R20

R21

R22

R23

R24

R25

R26

R27

R28

R29

R30

R31

R32

R33

R34

R35

R36

R37

R38

R39

Chapter 9

140

Orthostatic complaints were experienced in 102 participants (11%), and 359 participants (25%) had undergone a previous fall incident. The 1996 and 2011 definitions of OH were both not related to previous fall incidents or orthostatic complaints. Only a >25% decrease in SBP was significantly related to orthostatic complaints (table 2). A >25% decrease in SBP was seen in 55 of the 1415 participants (4%). The median decrease in the group of patients with >25% SBP decrease was 48 mmHg (interquartile range (IQR) 42-53). No significant relation was seen between a >25% decrease in SBP and previous fall incidents. An association with previous fall incidents was only observed for a >25% decrease in DBP. A >25% decrease in DBP was seen in 41 of the 1415 participants (3%). The median decrease in this group was 25 mmHg (interquartile range (IQR) 22-30). No significant relation was seen between a >25% decrease in DBP and orthostatic complaints.

Table 2. Adjusted odds ratios for the effect of OH definitions on the risk of orthostatic complaints or previous fall incidents.

Orthostatic complaints as dependent variable

N=911Odds ratio (95% CI)

Previous fall incidents as dependent variable

N=1407Odds ratio (95% CI)

OH consensus definition 1996 [1] 0.84 (0.54-1.30) 1.04 (0.81-1.33)OH consensus definition 2011 [10] 1.07 (0.68-1.69) 1.08 (0.83-1.41)SBP decrease 10-15% 0.66 (0.34-1.27) 1.08 (0.77-1.51)SBP decrease 15-20% 0.77 (0.34-1.76) 0.92 (0.59-1.42)SBP decrease 20-25% 1.09 (0.44-2.70) 1.03 (0.59-1.80)SBP decrease > 25% 2.81 (1.31-6.05) 1.05 (0.56-1.95)DBP decrease 10-15% 1.05 (0.56-1.96) 1.10 (0.60-2.00)DBP decrease 15-20% 0.55 (0.21-1.44) 0.64 (0.25-1.62)DBP decrease 20-25% 1.16 (0.39-3.45) 1.32 (0.46-3.83)DBP decrease > 25% 2.29 (0.92-5.72) 2.56 (1.08-6.09)Δ SBP 1.00 (0.99-1.02) 1.00 (0.99-1.01)Δ DBP 0.99 (0.97-1.01) 1.00 (0.99-1.02)

All models were adjusted for age, gender, diabetes mellitus, history of stroke/ cerebral infarction, myocardial infarction, difficulty walking, and use of antihypertensive medications. The ORs can be interpreted as a measure of association of the various variables to orthostatic complains or previous fall incidents. OH= orthostatic hypotension; SBP= systolic blood pressure; DBP= diastolic blood pressure; CTI= CAREFALL triage instrument.

DISCUSSION

The 1996 and 2011 consensus definition of OH are not related to orthostatic complaints or previous fall incidents in community dwelling participants aged 65 years and older. Out of all other studied definitions, only a >25% decrease in SBP was related to complaints, whereas only a >25% decrease in DBP was related to previous fall incidents. The group of participants

R1

R2

R3

R4

R5

R6

R7

R8

R9

R10

R11

R12

R13

R14

R15

R16

R17

R18

R19

R20

R21

R22

R23

R24

R25

R26

R27

R28

R29

R30

R31

R32

R33

R34

R35

R36

R37

R38

R39


141

9

who had a >25% decrease in SBP after postural change, had a significant higher prevalence of hypertension, history of myocardial infarction, and used higher number of medications than participants without a >25% decrease in SBP (data not shown). In a previous study of Maurer et al. the relation between the degree of orthostatic change and falling was investigated; no relation was found between degree of orthostatic change and falling [13]. Although Fedorowski has shown that patients with severe hypertension (SBP >160 mmHg) have larger orthostatic hemodynamic reserve and these patients had higher decrease of SBP after postural change no studies has been published which describe the influence of a relative orthostatic decrease of BP [9]. Since a SBP decrease of >40 mmHg or DBP decrease of >20 mmHg within the first 15s of postural change, frequently triggers syncope it seems reasonable that the amount of BP decrease after postural change influences outcome like orthostatic complaints or falling [15]. Several limitations need to be addressed. First, blood pressure measurement in the supine position was only performed once at 3 minutes. This may have led to an underestimation of the actual prevalence of OH, because OH is defined as a decrease of SBP or DBP within 3 minutes. Therefore, all drops of blood pressure before 3 minutes were not measured in present study. Secondly, two types of bias could have occurred; recall and selection bias. Previous fall incidents were based on participants’ report, which could have resulted in recall bias. Selection bias may have occurred as participants subscribed actively to this study when they were interested. Another limitation is the definition of orthostatic complaints. This was defined as symptoms of light-headedness, syncope, or dizziness questioned by the CTI questionnaire. These questions were not specifically related to postural change but to falling, which could have led to underestimation of the prevalence of orthostatic complaints. Finally, establishing a causal relation is not possible due to the cross-sectional design.

In conclusion, the consensus definition of OH was not a predictor for previous fall incidents and orthostatic complaints. A decrease of SBP or DBP of >25% were significantly related to orthostatic complaints or previous falling, respectively. Using relative decreases may be more appropriate in clinical practice. Although OH blood pressure measurements are widely used for clinical and research goals, its clinical relevance seems very limited. More studies investigating relative decreases should be performed.

R1

R2

R3

R4

R5

R6

R7

R8

R9

R10

R11

R12

R13

R14

R15

R16

R17

R18

R19

R20

R21

R22

R23

R24

R25

R26

R27

R28

R29

R30

R31

R32

R33

R34

R35

R36

R37

R38

R39

Chapter 9

142

REFERENCES






6. Feldstein C, Weder AB. Orthostatic hypotension: a common, serious and underrecognized problem in hospitalized patients. Journal of the American Society of Hypertension: JASH. 2012 Jan-Feb;6(1):27-39.

7. Low PA. Prevalence of orthostatic hypotension. Clinical autonomic research: official journal of the Clinical Autonomic Research Society. 2008 Mar;18 Suppl 1:8-13.

8. Kaufmann H. Consensus statement on the definition of orthostatic hypotension, pure autonomic failure and multiple system atrophy. Clinical autonomic research: official journal of the Clinical Autonomic Research Society. 1996 Apr;6(2):125-6.

9. Fedorowski A, Burri P, Melander O. Orthostatic hypotension in genetically related hypertensive and normotensive individuals. Journal of hypertension. 2009 May;27(5):976-82.





14. Boele van Hensbroek P, van Dijk N, van Breda GF, et al. The CAREFALL Triage instrument identifying risk factors for recurrent falls in elderly patients. The American journal of emergency medicine. 2009 Jan;27(1):23-36.

15. Wieling W, Krediet Ct Fau - van Dijk N, van Dijk N Fau - Linzer M, Linzer M Fau - Tschakovsky ME, Tschakovsky ME. Initial orthostatic hypotension: review of a forgotten condition. 20070103 DCOM- 20070222(1470-8736 (Electronic)).

R1

R2

R3

R4

R5

R6

R7

R8

R9

R10

R11

R12

R13

R14

R15

R16

R17

R18

R19

R20

R21

R22

R23

R24

R25

R26

R27

R28

R29

R30

R31

R32

R33

R34

R35

R36

R37

R38

R39


143

9

SUPPLEMENTAL DATA

Appendix I. CTI questions regarding orthostatic complaints

Do you feel the fall coming up? If yes, what are your complaints before you fall?- Dizziness: yes/no/I do not know- light-headedness: yes/no/I do not know

During, or just after the fall, do you experience any of the following complaints?- Loss of consciousness: yes/no/I do not know

CTI= CAREFALL Triage Instrument.

Total study population

N=1415

Analyses regarding OH and previous fall

incidentes

Analyses regarding OH and orthostatic

complaints

Included N=1407

Included N=911

Missing values N=8

Missing values N=504

Appendix II. Flow diagram

CHAPTER 10Health-related quality of life, rehabilitation

and mortality in a nursing home population

Published as: Hartog LC, Landman GWD, Cimzar-Sweelssen M, Knipscheer A, Groenier

KH, Kleefstra N, Bilo HJG, Van Hateren KJJ. Health-related quality of life,

rehabilitation and mortality in a nursing home population. Neth J Med 2016

Jul; 74 (6):247-56

R1

R2

R3

R4

R5

R6

R7

R8

R9

R10

R11

R12

R13

R14

R15

R16

R17

R18

R19

R20

R21

R22

R23

R24

R25

R26

R27

R28

R29

R30

R31

R32

R33

R34

R35

R36

R37

R38

R39

Chapter 10

146

ABSTRACT

ac ground: Health-related quality of life (HRQOL) in nursing home residents is generally low. The purpose of this study was to investigate the associations between HRQOL and two clinical relevant outcome measures, all-cause mortality and successful rehabilitation, in a nursing home population.

Methods: In an observational prospective cohort study in nursing home population, HRQOL was assessed with the RAND-36. A total of 184 patients were included, 159 (86%) completed the RAND-36 and were included in the study. A cox pr{Schemper, 2000 #8941}oportional hazard model was used to investigate the independent association between HRQOL, rehabilitation and mortality with adjustment for confounders. Risk prediction capabilities were assessed with Harrell’s C statistics and the proportion of explained variance (R2).

esults: The median age (interquartile range) was 79 (75-85) years. The health dimensions vitality (HR 0.88 (95%CI 0.77-0.99)) and mental health (HR 0.86 (95%CI 0.75-0.98)) were inversely associated and role functioning-physical (HR 1.08 (95%CI 1.02-1.15)) was positively associated with mortality. The Harrell’s C value and the R2 were ≤0.02 and ≤0.03 higher in the adjusted models with the dimensions role functioning-physical, mental health or vitality compared to the models without these dimensions. None of the health dimensions or summary scales was related to successful rehabilitation.

onclusion: HRQOL was significantly associated with mortality for three dimensions, but partly in opposite directions. Additional value of HRQOL in mortality prediction is very limited. There were no independent associations between HRQOL and successful rehabilitation. Although HRQOL is an important outcome, this study did not provide evidence for an association between HRQOL and successful rehabilitation.

R1

R2

R3

R4

R5

R6

R7

R8

R9

R10

R11

R12

R13

R14

R15

R16

R17

R18

R19

R20

R21

R22

R23

R24

R25

R26

R27

R28

R29

R30

R31

R32

R33

R34

R35

R36

R37

R38

R39

ealth related ualit o li e rehabilitation and ortalit in a nursing ho e population

147

10

INTRODUCTION

Comorbidities, depression, cognitive impairment and other geriatric problems, are highly prevalent in old age and can greatly impact health-related quality of life (HRQOL) [1-6]. As a consequence, HRQOL is generally low in nursing home residents [2-4, 7-9]. Nursing homes in the Netherlands provide care and long-stay services for elderly with chronic mental or physical diseases, and the majority also provide rehabilitation services. Measuring HRQOL in nursing home patients could lead to an increased understanding on factors that negatively impact HRQOL, ultimately aiming to improve HRQOL of this patient group, characterized by a overall very low HRQOL.The evaluation of HRQOL in individual patients can be used to measure disease-related distress and overall perception of health. Next to the evaluation of HRQOL as a separate outcome measure, HRQOL also has prognostic value in non-nursing home settings. Furthermore, a lower HRQOL has been associated with increased mortality risk in non-nursing home settings, also in elderly patients. Besides, HRQOL is also used to evaluate therapeutic interventions. Therefore, HRQOL could have a variety of implications in decision-making processes regarding patients and medical interventions.Increased understanding about HRQOL in nursing home patients could improve HRQOL and outcome within these patients. As mortality risk is already very high in old age, other clinical outcomes, next to HRQOL itself, like successful rehabilitation, may be more relevant. Several studies have reported a relationship between HRQOL and rehabilitation in for example stroke patients [10, 11]. Low HRQOL corresponds to substantial limitations in physical, emotional and social well-being due to a medical condition or its treatment [12]. These aspects of HRQOL can negatively influence successful rehabilitation. The associations between HRQOL and rehabilitation may be bidirectional: on the one hand successful rehabilitation itself can improve HRQOL [13-15] and on the other hand a higher HRQOL may improve motivation and increase the chance of successful rehabilitation. To our knowledge, there are no studies that have investigated the relationship between HRQOL and rehabilitation in a nursing home. Only one study investigated the relationship between HRQOL and mortality in a nursing home setting [16]. Furthermore, no studies reported whether measuring HRQOL had a discriminatory value, using for example the Harrell’s C, to assess the predictive capability of HRQOL in a nursing home population. The continuous growth of the elderly population in combination with the severely impaired HRQOL in our oldest old, underline the importance to gain more understanding of the implications of HRQOL in nursing home populations. The purpose of this study was to investigate the associations between HRQOL and two clinical relevant outcome measures, all-cause mortality and successful rehabilitation, in patients admitted to a nursing home.

R1

R2

R3

R4

R5

R6

R7

R8

R9

R10

R11

R12

R13

R14

R15

R16

R17

R18

R19

R20

R21

R22

R23

R24

R25

R26

R27

R28

R29

R30

R31

R32

R33

R34

R35

R36

R37

R38

R39

Chapter 10

148


The study is reported according to the STROBE (Strengthening the reporting of observational studies in epidemiology) recommendations [17].

tud sa pleThe design and details of this prospective observational cohort study have been described previously [18]. Only patients from rehabilitation and somatic departments were included. Patients admitted to a somatic department receive prolonged or permanent care whereas patients at the rehabilitation department are admitted with the intention to rehabilitate and return home. Recruitment and all study procedures, for example administering the HRQOL questionnaire, for patients on the somatic department took place between September 2010 and December 2010. Timing of inclusion was different for each somatic patient. For patients on the rehabilitation department recruitment and all study procedures took place between September 2010 and December 2011, mostly within the first weeks of admission.

Patients admitted to the psychogeriatric department were excluded, because these patients are generally not able to complete HRQOL questionnaires [19]. Other exclusion criteria were a life expectancy less than 4 weeks and an impending transfer to a hospice department.

ata collectionAn elderly care physician collected all baseline data directly after inclusion. Baseline data included demographic characteristics, full medical histories (including cardiovascular disease (CVD), diabetes mellitus and hypertension), and medication use. Trained physicians or nurses administered the questionnaires at baseline. HRQOL was measured using the RAND-36 questionnaire [12, 20]. The RAND-36 is a generic instrument to measure aspects of health that are relevant to functional status and well-being [20, 21]. The RAND-36 consists of nine aspects of health status: physical functioning, role limitations due to physical problems (role functioning-physical), bodily pain, general health, vitality, social functioning, role limitations due to emotional problems (role functioning-emotional), mental health, and health change. Each dimension has a 100-point scale, where higher scores indicate better HRQOL. Of the nine scales two component summary scores can be calculated: a physical component summary (PCS) and a mental component summary (MCS). In 2015, vital status and cause of death were retrieved from records maintained by the nursing home and general practitioners.

R1

R2

R3

R4

R5

R6

R7

R8

R9

R10

R11

R12

R13

R14

R15

R16

R17

R18

R19

R20

R21

R22

R23

R24

R25

R26

R27

R28

R29

R30

R31

R32

R33

R34

R35

R36

R37

R38

R39


149

10

linical outco e easuresAll-cause mortality and successful rehabilitation were the primary endpoints. Successful rehabilitation was defined as either discharge to patients’ own home or an adapted home for the elderly, where they functioned self-reliant. Successful rehabilitation was investigated in the subgroup of patients admitted to the rehabilitation department.

tatistical anal sesContinuous variables are presented as mean with the standard deviation for normally distributed variables, or as median with an interquartile range (IQR) for non-normally distributed variables. Cox proportional hazard modelling was used to investigate the relationship between HRQOL and 1. all-cause mortality and 2. successful rehabilitation, with and without adjustment for the following confounders: age, gender, smoking, body mass index, a history of cardiovascular, or pulmonary disease, hypertension, psychiatric disease, length of stay before study inclusion, and the number of drugs. Collinearity was tested between each of the RAND-36 dimensions and the confounders. All hazard ratios refer to a 10-point higher score on the RAND-36 dimensions. In case of significant associations in the Cox regression models, risk prediction capabilities were assessed with Harrell’s C statistics and the proportion of explained variance (R2)[22, 23]. The Schoenfeld residual plots were inspected for each predictor variable to check the assumption of proportional hazards; all assumptions were met unless stated otherwise. A two-sided p<0.05 was considered significant. All statistical analyses were performed using SPSS software version 22 (SAS Institute, Cary, NC, USA).

thical approval and linical rial registrationBefore starting the study, the central committee of research involving human subjects (CCMO) in the Netherlands was contacted. The CCMO agreed that with the current design, no formal approval of an accredited medical ethics committee was needed. According to Dutch guidelines this research does not fall under the scope of the Medical Research Involving Human Subjects Act. This study was performed in accordance with the Declaration of Helsinki and the treating physicians and nurses obtained written informed consent from all patients and data were analysed anonymously.The study was registered on ClinicalTrials.gov (NCT01362751).

R1

R2

R3

R4

R5

R6

R7

R8

R9

R10

R11

R12

R13

R14

R15

R16

R17

R18

R19

R20

R21

R22

R23

R24

R25

R26

R27

R28

R29

R30

R31

R32

R33

R34

R35

R36

R37

R38

R39

Chapter 10

150

RESULTS

A total of 184 patients were included in this cohort. HRQOL data were completed for 159 (86%) patients. From the 159 included patients, 123 patients were admitted to the rehabilitation department and 36 to the somatic department. Baseline characteristics of the total study population are presented in table 1. The median age (IQR) was 79.2 (75.2-85.9) years. Median PCS and MCS (IQR) scores were 44 (34-57) and 59 (48-74), respectively. Patients on the rehabilitation department were older, had a lower BMI, hypertension was seen more frequent, and psychiatric disease was seen less often compared to patients on the somatic department (Appendix table 1). No difference in mortality hazard was observed between the rehabilitation department and the somatic department (hazard ratio (HR) 0.90 (95% confidence interval (CI) 0.66-1.21). Patients with missing data had a higher history of dementia and CVD compared to patients without missing data (Appendix table 1). Missing data was not significantly associated with mortality or successful rehabilitation, hazard ratio (HR) 0.89 (95% confidence interval (CI) 0.47-1.71) and HR 1.41 (95%CI 0.47-4.23), respectively.

and ortalitDuring a median follow-up period of 3.4 years, 75 (47%) patients had died. Three out of 9 health dimensions were independently associated with all-cause mortality (after adjusting for confounders). The health dimensions vitality and mental health were inversely associated with mortality; HR 0.88 (95% CI 0.77-0.99) and HR 0.86 (95%CI 0.75-0.98), respectively. The dimension role functioning-physical was positively associated with mortality, HR 1.08 (95%CI 1.02-1.15). PCS and MCS component scores were not significantly associated with all-cause mortality. In model 2, age was the only covariate besides above-mentioned three health dimensions, which was significantly related to mortality (HR 1.08 (95%CI 1.04-1.12)). The results of the Cox regression analyses are presented in table 2. Post-hoc analyses were performed according to type of department. Because of the small number of somatic long-term patients (n=36), we first adjusted for age and gender in both groups (Appendix table 2). The analyses with the fully adjusted model were only performed for the rehabilitation group (Appendix table 3). In additional analyses, stratified according to type of department, no significant associations were seen between HRQOL and mortality within the group of somatic patients. Within the rehabilitation group significant relations with mortality were observed for the dimensions mental health and general health perception; HR 0.80 (95%CI 0.69-0.94) and HR 0.85 (95%CI 0.74-0.99), respectively. After adjustment for all selected confounders, a significant relationship in the rehabilitation group was only seen between the dimension role functioning-physical and mortality (HR 1.08 (95%CI 1.00-1.18)).

R1

R2

R3

R4

R5

R6

R7

R8

R9

R10

R11

R12

R13

R14

R15

R16

R17

R18

R19

R20

R21

R22

R23

R24

R25

R26

R27

R28

R29

R30

R31

R32

R33

R34

R35

R36

R37

R38

R39


151

10

Tabl

e 1.

Bas

elin

e ch

arac

teri

stics

. Tota

l HRQ

OL

Dec

ease

d pa

tien

tsSu

rviv

ed p

atien

tsp-

valu

eSu

cces

sful

re

habi

litati

on

No

succ

essf

ul

reha

bilit

ation

p-

valu

e

Char

acte

risti

cN

= 15

9N

=75

N=8

4N

=102

N=2

1D

emog

raph

ics

Ag

e a

79

(75-

85)

85 (7

9-88

)78

(72-

84)

<0.0

0581

(76-

86)

85 (7

9-89

)0.

04Fe

mal

e G

ende

r b

111

(70%

)49

(65%

)62

(74%

)0.

2574

(73%

)15

(71%

)0.

92Bo

dy m

ass

inde

x, k

g/m

2 a

27 (2

3-29

)25

(22-

29)

26 (2

3-29

)0.

7325

(23-

28)

24 (2

2-27

)0.

38H

yper

tens

ion

b12

4 (7

8%)

66 (8

8%)

58 (6

9%)

<0.0

0581

(79%

)20

(95%

)0.

09H

isto

ry o

f CVD

b67

(42%

)38

(51%

)29

(35%

)0.

0439

(38%

)8

(38%

)0.

99D

iabe

tes

mel

litus

b68

(43%

)36

(48%

)32

(38%

)0

.21

45 (4

4%)

9 (4

3%)

0.92

Dem

entia

b11

(7%

)7

(9%

)4

(5%

)0

.26

4 (4

%)

4 (1

9%)

0.01

Psyc

hiat

ric

dise

ase

b54

(34%

)27

(36%

)27

(32%

)0

.61

20 (2

0%)

12 (5

7%)

<0.0

05Pu

lmon

ary

dise

ase

b40

(25%

)20

(27%

)20

(24%

)0.

6829

(28%

)4

(19%

)0.

38Cu

rren

t sm

oker

b21

(13%

)8

(11%

)13

(16%

)0.

3712

(12%

)0

(0%

)0.

09N

umbe

r of

med

icin

e a

9 (6

-11)

9 (7

-11)

8 (5

-10)

0.0

69

(6-1

0)8

(6-1

1)0.

70

Mea

sure

men

tsPh

ysic

al fu

nctio

ning

a10

(5-3

0)10

(0-3

0)13

(5-3

0)0.

3615

(5-3

5)10

(0-2

8)0

.10

Soci

al fu

nctio

ning

a50

(50-

63)

63 (5

0-75

)50

(50-

63)

0.19

50 (5

0-63

)50

(50-

75)

0.76

Role

func

tioni

ng-p

hysi

cal a

0 (0

-50)

0 (0

-75)

0 (0

-44)

0.1

20

(0-2

5)0

(0-3

8)0.

94Ro

le-f

uncti

onin

g-em

otion

al a

83 (0

-100

)67

(0-1

00)

83 (0

-100

)0.

7333

(0-1

00)

100

(0-1

00)

0.33

Men

tal h

ealth

a68

(56-

80)

64 (5

2-80

)72

(64-

84)

0.02

72 (6

3-84

)56

(44-

72)

<0.0

05Bo

dily

pai

n a

67 (2

2-80

)67

(33-

88)

62 (2

1-80

)0.

6550

(20-

78)

78 (4

7-90

)0.

02Vi

talit

y a

65 (4

5-75

)60

(45-

75)

65 (5

0-80

)0.

0865

(55-

78)

50 (3

5-68

)<0

.005

Gen

eral

hea

lth p

erce

ption

a55

(40-

75)

50 (4

0-70

)60

(45-

75)

0.0

658

(45-

75)

55 (4

0-70

)0.

53H

ealth

cha

nge

a50

(25-

50)

50 (2

5-50

)50

(25-

50)

0.14

50 (2

5-50

)25

(0-5

0)0.

03PC

S Sc

ore

a44

(34-

57)

44 (3

4-57

)44

(34-

56)

0.90

42 (3

3-54

)42

(34-

54)

0.97

MCS

Sco

re a

69 (4

8-74

)58

(43-

73)

61 (5

1-75

)0.

3258

(49-

74)

59 (4

0-74

)0.

38

Dat

a ar

e m

edia

ns (i

nter

quar

tile

rang

e) o

r N

(%).

a M

ann-

Whi

tney

U te

st w

as u

sed

to c

ompa

re g

roup

s. b

Chi S

quar

e te

st w

as u

sed

to c

ompa

re g

roup

s.

Abb

revi

ation

s CV

D =

Car

dio

vasc

ular

dis

ease

. PCS

= p

hysi

cal c

ompo

nent

sum

mar

y. M

CS =

Men

tal c

ompo

nent

sum

mar

y.

R1

R2

R3

R4

R5

R6

R7

R8

R9

R10

R11

R12

R13

R14

R15

R16

R17

R18

R19

R20

R21

R22

R23

R24

R25

R26

R27

R28

R29

R30

R31

R32

R33

R34

R35

R36

R37

R38

R39

Chapter 10

152

Tabl

e 2.

Rel

ation

ship

bet

wee

n H

RQO

L, a

ll-ca

use

mor

talit

y, a

nd s

ucce

ssfu

l reh

abili

tatio

n.

All-

caus

e m

orta

lity

(N=1

59)

Succ

essf

ul re

habi

litati

on (N

=123

)M

odel

1H

R (9

5%CI

)M

odel

2H

R (9

5%CI

)M

odel

1H

R (9

5%CI

)M

odel

2H

R (9

5%CI

)Ph

ysic

al fu

nctio

ning

0.97

(0.8

8-1.

07)

0.99

(0.8

9-1.

10)

1.06

(0.9

7-1.

16)

1.05

(0.9

6-1.

16)

Soci

al fu

nctio

ning

1.04

(0.9

1-1.

17)

1.03

(0.9

1-1.

17)

1.02

(0.9

3-1.

13)

1.00

(0.9

0-1.

12)

Role

func

tioni

ng-p

hysi

cal

1.03

(0.9

8-1.

09)

1.08

(1.0

2-1.

15)

0.99

(0.9

3-1.

05)

0.98

(0.9

2-1.

05)

Role

func

tioni

ng-e

moti

onal

1.01

(0.9

6-1.

06)

1.01

(0.9

6-1.

06)

1.00

(0.9

6-1.

04)

0.99

(0.9

4-1.

03)

Men

tal h

ealth

0.83

(0.7

5-0.

94)

0.86

(0.7

5-0.

98)

1.08

(0.9

7-1.

21)

1.06

(0.9

4-1.

20)

Bodi

ly p

ain

1.01

(0.9

4-1.

08)

1.01

(0.9

4-1.

08)

0.94

(0.8

9-1.

00)

0.96

(0.9

0-1.

03)

Vita

lity

0.87

(0.7

8-0.

97)

0.88

(0.7

7-0.

99)

1.08

(0.9

7-1.

21)

1.03

(0.9

2-1.

16)

Gen

eral

hea

lth p

erce

ption

0.90

(0.8

0-1.

00)

0.91

(0.8

1-1.

03)

1.00

(0.9

0-1.

10)

1.00

(0.8

9-1.

12)

Hea

lth c

hang

e0.

93 (0

.85-

1.02

)0.

95 (0

.87-

1.05

)1.

08 (1

.00-

1.17

)1.

07 (0

.99-

1.17

)PC

S0.

97 (0

.84-

1.13

)1.

02 (0

.88-

1.20

)1.

00 (0

.87-

1.15

)0.

99 (0

.84-

1.15

)M

CS0.

90 (0

.78-

1.03

)0.

91 (0

.88-

1.06

)1.

05 (0

.92-

1.20

)1.

00 (0

.87-

1.15

)

Mod

el 1

is t

he u

nadj

uste

d m

odel

. In

Mod

el 2

we

adju

sted

for

age,

gen

der,

smok

ing,

DM

, his

tory

of

CVD

, hyp

erte

nsio

n, B

MI,

hist

ory

of p

ulm

onar

y di

seas

e, h

isto

ry o

f psy

chia

tric

dis

ease

, len

gth

of s

tay,

and

the

num

ber o

f med

icati

ons.

The

haz

ard

ratio

s re

fer t

o a

10-p

oint

hig

her s

core

on

the

RAN

D-

36 d

imen

sion

s. A

bbre

viati

ons

HR:

haz

ard

ratio

; CI:

confi

denc

e in

terv

al. P

CS =

phy

sica

l com

pone

nt s

umm

ary.

MCS

= M

enta

l com

pone

nt s

umm

ary.

Bol

d va

lues

cor

resp

ond

to a

p-v

alue

of 0

.05

or le

ss.

Tabl

e 3.

Cox

regr

essi

on a

naly

ses

and

pred

ictiv

e ca

pabi

lity

for a

ll-ca

use

mor

talit

y. T

he H

arre

l’s C

sta

tistic

and

the

R2 fo

r eva

luati

ng p

redi

ctive

cap

abili

ty

the

addi

tion

of H

RQO

L di

men

sion

role

func

tioni

ng-p

hysi

cal,

men

tal h

ealth

or

vita

lity

to m

odel

2.

All-

caus

e m

orta

lity

(N=1

59)

Haz

ard

Ratio

Har

rell’

s C

R2

Har

rell’

s C

*R

2 *

(95%

CI)

(95%

CI)

(95%

CI)

(95%

CI)*

(95%

CI)*

Role

func

tioni

ng-p

hysi

cal (

mod

el 1

)1.

03 (0

.98-

1.09

)0.

55 (0

.49-

0.61

)0.

01 (0

.01-

0.11

)N

AN

A

Role

func

tioni

ng-p

hysi

cal (

mod

el 2

)1.

08 (1

.02-

1.15

)0.

69 (0

.62-

0.75

)0.

15 (0

.09-

0.43

)0.

67 (0

.61-

0.73

)0.

14 (0

.07-

0.41

)M

enta

l hea

lth (m

odel

1)

0.83

(0.7

5-0.

94)

0.61

(0.5

5-0.

68)

0.06

(0.0

0-0.

19)

NA

NA

Men

tal h

ealth

(mod

el 2

)0.

86 (0

.75-

0.98

)0.

69 (0

.63-

0.75

)0.

17 (0

.10-

0.42

)0.

67 (0

.61-

0.73

)0.

14 (0

.07-

0.41

)Vi

talit

y (m

odel

1)

0.87

(0.7

8-0.

97)

0.59

(0.5

2-0.

66)

0.05

(0.0

1-0.

17)

NA

NA

Vita

lity

(mod

el 2

)0.

88 (0

.77-

0.99

)0.

68 (0

.62-

0.74

)0.

17 (0

.10-

0.42

)0.

67 (0

.61-

0.73

)0.

14 (0

.07-

0.41

)

Mod

el 1

is t

he u

nadj

uste

d m

odel

. In

Mod

el 2

we

adju

sted

for

age,

gen

der,

smok

ing,

DM

, his

tory

of

CVD

, hyp

erte

nsio

n, B

MI,

hist

ory

of p

ulm

onar

y di

seas

e, h

isto

ry o

f psy

chia

tric

dis

ease

, and

the

num

ber o

f med

icati

ons.

The

haz

ard

ratio

s re

fer t

o a

10-p

oint

hig

her s

core

on

the

RAN

D-3

6 di

men

sion

s.

Abb

revi

ation

s H

R: h

azar

d ra

tio; C

I: co

nfide

nce

inte

rval

. PCS

= p

hysi

cal c

ompo

nent

sum

mar

y. M

CS =

Men

tal c

ompo

nent

sum

mar

y. *

Har

rell’

s C

valu

es

and

R2 fo

r th

e m

odel

s w

ithou

t the

HRQ

OL

dim

ensi

ons

role

func

tioni

ng-p

hysi

cal,

men

tal h

ealth

or

vita

lity.

R1

R2

R3

R4

R5

R6

R7

R8

R9

R10

R11

R12

R13

R14

R15

R16

R17

R18

R19

R20

R21

R22

R23

R24

R25

R26

R27

R28

R29

R30

R31

R32

R33

R34

R35

R36

R37

R38

R39


153

10

Tabl

e 2.

Rel

ation

ship

bet

wee

n H

RQO

L, a

ll-ca

use

mor

talit

y, a

nd s

ucce

ssfu

l reh

abili

tatio

n.

All-

caus

e m

orta

lity

(N=1

59)

Succ

essf

ul re

habi

litati

on (N

=123

)M

odel

1H

R (9

5%CI

)M

odel

2H

R (9

5%CI

)M

odel

1H

R (9

5%CI

)M

odel

2H

R (9

5%CI

)Ph

ysic

al fu

nctio

ning

0.97

(0.8

8-1.

07)

0.99

(0.8

9-1.

10)

1.06

(0.9

7-1.

16)

1.05

(0.9

6-1.

16)

Soci

al fu

nctio

ning

1.04

(0.9

1-1.

17)

1.03

(0.9

1-1.

17)

1.02

(0.9

3-1.

13)

1.00

(0.9

0-1.

12)

Role

func

tioni

ng-p

hysi

cal

1.03

(0.9

8-1.

09)

1.08

(1.0

2-1.

15)

0.99

(0.9

3-1.

05)

0.98

(0.9

2-1.

05)

Role

func

tioni

ng-e

moti

onal

1.01

(0.9

6-1.

06)

1.01

(0.9

6-1.

06)

1.00

(0.9

6-1.

04)

0.99

(0.9

4-1.

03)

Men

tal h

ealth

0.83

(0.7

5-0.

94)

0.86

(0.7

5-0.

98)

1.08

(0.9

7-1.

21)

1.06

(0.9

4-1.

20)

Bodi

ly p

ain

1.01

(0.9

4-1.

08)

1.01

(0.9

4-1.

08)

0.94

(0.8

9-1.

00)

0.96

(0.9

0-1.

03)

Vita

lity

0.87

(0.7

8-0.

97)

0.88

(0.7

7-0.

99)

1.08

(0.9

7-1.

21)

1.03

(0.9

2-1.

16)

Gen

eral

hea

lth p

erce

ption

0.90

(0.8

0-1.

00)

0.91

(0.8

1-1.

03)

1.00

(0.9

0-1.

10)

1.00

(0.8

9-1.

12)

Hea

lth c

hang

e0.

93 (0

.85-

1.02

)0.

95 (0

.87-

1.05

)1.

08 (1

.00-

1.17

)1.

07 (0

.99-

1.17

)PC

S0.

97 (0

.84-

1.13

)1.

02 (0

.88-

1.20

)1.

00 (0

.87-

1.15

)0.

99 (0

.84-

1.15

)M

CS0.

90 (0

.78-

1.03

)0.

91 (0

.88-

1.06

)1.

05 (0

.92-

1.20

)1.

00 (0

.87-

1.15

)

Mod

el 1

is t

he u

nadj

uste

d m

odel

. In

Mod

el 2

we

adju

sted

for

age,

gen

der,

smok

ing,

DM

, his

tory

of

CVD

, hyp

erte

nsio

n, B

MI,

hist

ory

of p

ulm

onar

y di

seas

e, h

isto

ry o

f psy

chia

tric

dis

ease

, len

gth

of s

tay,

and

the

num

ber o

f med

icati

ons.

The

haz

ard

ratio

s re

fer t

o a

10-p

oint

hig

her s

core

on

the

RAN

D-

36 d

imen

sion

s. A

bbre

viati

ons

HR:

haz

ard

ratio

; CI:

confi

denc

e in

terv

al. P

CS =

phy

sica

l com

pone

nt s

umm

ary.

MCS

= M

enta

l com

pone

nt s

umm

ary.

Bol

d va

lues

cor

resp

ond

to a

p-v

alue

of 0

.05

or le

ss.

Tabl

e 3.

Cox

regr

essi

on a

naly

ses

and

pred

ictiv

e ca

pabi

lity

for a

ll-ca

use

mor

talit

y. T

he H

arre

l’s C

sta

tistic

and

the

R2 fo

r eva

luati

ng p

redi

ctive

cap

abili

ty

the

addi

tion

of H

RQO

L di

men

sion

role

func

tioni

ng-p

hysi

cal,

men

tal h

ealth

or

vita

lity

to m

odel

2.

All-

caus

e m

orta

lity

(N=1

59)

Haz

ard

Ratio

Har

rell’

s C

R2

Har

rell’

s C

*R

2 *

(95%

CI)

(95%

CI)

(95%

CI)

(95%

CI)*

(95%

CI)*

Role

func

tioni

ng-p

hysi

cal (

mod

el 1

)1.

03 (0

.98-

1.09

)0.

55 (0

.49-

0.61

)0.

01 (0

.01-

0.11

)N

AN

A

Role

func

tioni

ng-p

hysi

cal (

mod

el 2

)1.

08 (1

.02-

1.15

)0.

69 (0

.62-

0.75

)0.

15 (0

.09-

0.43

)0.

67 (0

.61-

0.73

)0.

14 (0

.07-

0.41

)M

enta

l hea

lth (m

odel

1)

0.83

(0.7

5-0.

94)

0.61

(0.5

5-0.

68)

0.06

(0.0

0-0.

19)

NA

NA

Men

tal h

ealth

(mod

el 2

)0.

86 (0

.75-

0.98

)0.

69 (0

.63-

0.75

)0.

17 (0

.10-

0.42

)0.

67 (0

.61-

0.73

)0.

14 (0

.07-

0.41

)Vi

talit

y (m

odel

1)

0.87

(0.7

8-0.

97)

0.59

(0.5

2-0.

66)

0.05

(0.0

1-0.

17)

NA

NA

Vita

lity

(mod

el 2

)0.

88 (0

.77-

0.99

)0.

68 (0

.62-

0.74

)0.

17 (0

.10-

0.42

)0.

67 (0

.61-

0.73

)0.

14 (0

.07-

0.41

)

Mod

el 1

is t

he u

nadj

uste

d m

odel

. In

Mod

el 2

we

adju

sted

for

age,

gen

der,

smok

ing,

DM

, his

tory

of

CVD

, hyp

erte

nsio

n, B

MI,

hist

ory

of p

ulm

onar

y di

seas

e, h

isto

ry o

f psy

chia

tric

dis

ease

, and

the

num

ber o

f med

icati

ons.

The

haz

ard

ratio

s re

fer t

o a

10-p

oint

hig

her s

core

on

the

RAN

D-3

6 di

men

sion

s.

Abb

revi

ation

s H

R: h

azar

d ra

tio; C

I: co

nfide

nce

inte

rval

. PCS

= p

hysi

cal c

ompo

nent

sum

mar

y. M

CS =

Men

tal c

ompo

nent

sum

mar

y. *

Har

rell’

s C

valu

es

and

R2 fo

r th

e m

odel

s w

ithou

t the

HRQ

OL

dim

ensi

ons

role

func

tioni

ng-p

hysi

cal,

men

tal h

ealth

or

vita

lity.

The results of the analyses regarding the risk prediction capabilities are presented in table 3. The Harrell’s C values for the adjusted model of the dimensions role functioning-physical, mental health and vitality, were 0.69 (95%CI 0.62-0.75), 0.69 (95%CI 0.63-0.75) and 0.68 (95%CI 0.62-0.74), respectively. The Harrell’s C values and the R2 were ≤0.02 and ≤0.03 higher in the models with the HRQOL dimensions role functioning-physical, mental health or vitality compared to the models without these three dimensions.

and success ul rehabilitationData on HRQOL were missing for 5 (4%) out of 128 patients admitted to the rehabilitation department. During a median follow-up period of 36 days (IQR 7–88), 102 patients successfully rehabilitated; 90 patients were discharged to their own home and 12 were discharged to an adapted home for the elderly. Patients who successfully rehabilitated had higher scores on the subscales mental health, vitality and health change, while they scored lower on the subscale bodily pain. None of the health dimensions or summary scales was significantly associated to successful rehabilitation in the regression analyses (table 2).

DISCUSSION

HRQOL was significantly associated with mortality for three dimensions, but partly in opposite directions. We observed no independent association between HRQOL and successful rehabilitation.

and ortalitDuring a median follow-up period of 3.4 years, 75 (47%) patients had died. There is great variation in mortality rates in nursing home studies [16, 24, 25]. Because nursing homes could provide care for patients with chronic mental or physical diseases or provide rehabilitation services or combined, mortality rates strongly depend on the type of nursing home. Taken this all together it is difficult to make a reliable comparison of the mortality rates between the current and previous studies.Higher scores on the dimensions vitality and mental health were related to a lower mortality risk, whereas a higher score on the dimension role functioning-physical was related to a higher mortality risk. The Harrell’s C values and the R2 were ≤0.02 and ≤0.03 higher in the adjusted models with the dimensions role functioning-physical, mental health or vitality compared to the models without these dimensions.The score on the role functioning-physical dimension expresses the problems in daily life caused by a physical condition. The positive association with mortality implies that experiencing fewer problems in daily life was associated with an increased mortality risk,

R1

R2

R3

R4

R5

R6

R7

R8

R9

R10

R11

R12

R13

R14

R15

R16

R17

R18

R19

R20

R21

R22

R23

R24

R25

R26

R27

R28

R29

R30

R31

R32

R33

R34

R35

R36

R37

R38

R39

Chapter 10

154

which is counterintuitive and not previously reported. A possible explanation could be that these frail patients are accustomed to live with substantial limitations in functioning, and therefore scored low on this dimension. In addition, and probably even more relevant, this scale may be inappropriate for nursing home residents as the scale is composed from four individual questions which relate to work or daily activities. Finally, the results could be a matter of type 1 error.The mental health and vitality dimensions were inversely associated with mortality, which was as we had expected and confirmatory to results from prior studies in different populations [26-30]. The mental health dimension is related to depression and anxiety and the vitality scale is related to fatigue and apathy. In a previous study in a nursing home setting, only physical functioning was significantly related to mortality [16]. The study population was, despite the fact it was a nursing home population, very different compared to present study. Mainly long-term residents were included and a high percentage (26%) of these residents had cancer, which may have resulted in different complaints and limitations, resulting in different HRQOL scores, but also in a different mortality risk. Within the rehabilitation department, the dimension role functioning-physical was positively related to mortality. Within the somatic department no relation with HRQOL and mortality was found. This difference in results between the somatic and rehabilitation department could be explained by the fact patients rehabilitating in a nursing home are potentially physically more frail patients after a recent acute hospital admission. Patients at the somatic department are chronic patients and used to their physical status. Furthermore, patients admitted to the rehabilitation department were older compared to patients on the somatic ward. Because of the sample size, we cannot exclude that the stratified results concerning the somatic group were subject to a type 2 error.Another study in community-dwelling elderly described a significant relationship between all subscales and mortality [31]. Comparing to the present study, these community-dwelling elderly were not admitted patients but were selected by a demographic registration system, which probably explains the difference in HRQOL scores and may explain the other relationship with mortality. In the present study the physical and mental component summary scores were not significantly related to all-cause mortality. This can be explained by the fact the component summary scores are calculated from the nine health dimensions, while only three health dimensions were significantly related to mortality. Besides, this could be also due to the sample size or the duration of follow-up. A previous study in community-dwelling elderly patients with type 2 diabetes showed that MCS was associated with mortality, only after an extended and long-term follow-up period [29]. If HRQOL is indeed only related to mortality after a long follow-up period, using HRQOL for these prognostic capabilities will be irrelevant as a long follow-up is not feasible in an old and frail population.

R1

R2

R3

R4

R5

R6

R7

R8

R9

R10

R11

R12

R13

R14

R15

R16

R17

R18

R19

R20

R21

R22

R23

R24

R25

R26

R27

R28

R29

R30

R31

R32

R33

R34

R35

R36

R37

R38

R39


155

10

Although our study showed that the dimensions role functioning-physical, vitality and mental health were an independent risk factor for mortality. Based on minimal increase in Harrell’s C values when adding role functioning-physical, vitality or mental health to the adjusted models, one may conclude that the additional value of these dimensions in mortality prediction apparently is very limited. It is important to realize that this study group is a group of frail patients with much comorbidity. Even in the fully adjusted models, the C value was lower than 0.70, indicating the poor predictive capability of the overall model.

and success ul rehabilitationThe results of the present study showed no significant association between HRQOL and successful rehabilitation. We hypothesized that higher HRQOL scores would be associated with successful rehabilitation. Due to better physical, emotional and social well-being, rehabilitation targets would be achieved sooner. Assessing changes in HRQOL could be used to measure improvements in relation to the rehabilitation process. Although HRQOL was not significantly associated with successful rehabilitation, there was an inverse relationship between a history of psychiatric disease and successful rehabilitation (HR 0.41 [95%CI 0.24-0.69]). In previous studies the relationship between depressive symptoms and functional recovery has been described in post-stroke patients [32, 33]. Depression has a negative effect on recovery in functional status and treatment of depressive symptoms leads to enhanced rehabilitation. Several previous studies showed improvements in HRQOL after different types of rehabilitation in diverse patient groups (stroke, COPD, cardiac problems, cancer) [11, 14, 34, 35]. But the aims of these studies were different compared to the aim of the present study. In the present study we aimed to investigate the exact opposite, whether HRQOL could influence the rate of successful rehabilitation. To the best of our knowledge, there are no validation studies regarding the use of the RAND-36 in rehabilitation patients. Concerning the use in a nursing home population, a validation study has been performed [19]. It can be questioned if the SF-36 or RAND-36 is a valid instrument in a nursing home population. Possible, the SF-36 or RAND-36 could only be used for subgroups of rehabilitation patients, like within patients with a higher cognitive and physical functioning.Although we cannot exclude the possibility that rehabilitation itself may have positive consequences for HRQOL, the current study shows that baseline HRQOL is not related to an increased chance of successful rehabilitation.

trengths and li itationsThe main strengths of this study were the prospective design, the possibility to take into account the number of variables adjusted for in the multivariate model, and the representative group of nursing home patients.

R1

R2

R3

R4

R5

R6

R7

R8

R9

R10

R11

R12

R13

R14

R15

R16

R17

R18

R19

R20

R21

R22

R23

R24

R25

R26

R27

R28

R29

R30

R31

R32

R33

R34

R35

R36

R37

R38

R39

Chapter 10

156

Representativeness was based on the fact 86% of all admitted somatic or rehabilitation patients during the study period participated in the present study. In addition, admission to a Dutch nursing home requires approval of a central indication committee and finally, the nursing home facility in the present study was a general nursing home, with somatic, psychogeriatric and rehabilitation departments, comparable to other Dutch nursing homes.There were also limitations. Firstly, due to the observational design it was not possible to establish a causal relation between HRQOL and mortality. Secondly, the RAND-36 questionnaires were not completed in 14% of the sample and this could have led to an uncertainty in the effect estimate. On the other hand, missing data was not significantly associated with mortality or successful rehabilitation, HR 0.89 (95%CI 0.47-1.71) and HR 1.41 (95%CI 0.47-4.23), respectively. The inability to complete questionnaires is a frequently observed problem in geriatric studies and indicative for sever morbidity. However, previous studies that investigated the RAND-36 had a lower response rate compared to present study [29, 30, 36]. Thirdly, the adequacy of using the RAND-36 questionnaire within a nursing home population has been questioned in different studies [4, 19]. An important issue is that the RAND-36 entails several potential inappropriate questions for this population [4, 19]. Due to the high heterogeneity in the nursing home population in general, the use of the RAND-36 could be more suitable for subgroups of rehabilitation patients, like e.g. those with a higher cognitive and physical functioning [19].Fourthly, we did not investigate changes in HRQOL. A change in HRQOL could possibly have predicted mortality more accurate [37]. Fifthly, successful rehabilitation was defined as discharge to home or a home for the elderly, where they remained self-reliant. As a consequence, patients with a worse outcome after rehabilitation but with a highly adapted home environment (e.g. stairlift, homecare, meal service) might have been discharged sooner. Finally, our study sample is rather small and therefore our results may be a matter of coincidence. Confirmation of our results in other studies is necessary, preferably performed with HRQOL performed at several moments during rehabilitation.

CONCLUSIONS

HRQOL was significantly associated with mortality for three dimensions, but partly in opposite directions. Additional value of HRQOL in mortality prediction is very limited. There were no independent associations between HRQOL and successful rehabilitation. The evaluation of HRQOL is important as a goal on its own; however, this study did not provide evidence for an association between HRQOL and successful rehabilitation within a nursing home population.

R1

R2

R3

R4

R5

R6

R7

R8

R9

R10

R11

R12

R13

R14

R15

R16

R17

R18

R19

R20

R21

R22

R23

R24

R25

R26

R27

R28

R29

R30

R31

R32

R33

R34

R35

R36

R37

R38

R39


157

10

REFERENCES










10. Chen CM, Tsai CC, Chung CY, et al. Potential predictors for health-related quality of life in stroke patients undergoing inpatient rehabilitation. Health and quality of life outcomes. 2015;13(1):118.

11. Katona M, Schmidt R, Schupp W, Graessel E. Predictors of health-related quality of life in stroke patients after neurological inpatient rehabilitation: a prospective study. 20150530(1477-7525 (Electronic)).

12. van der Zee KI SR. Het meten van de algemene gezondheidstoestand met de Rand-36, een handleiding. Tweede herziene druk. . UMCG / Rijksuniversiteit Groningen, Research Institute SHARE.; 2012 [December 2014]; Available from: http://www.rug.nl/share.

13. Marques A, Lourenco O Fau - da Silva JAP, da Silva JA. The burden of osteoporotic hip fractures in Portugal: costs, health related quality of life and mortality. 20150519(1433-2965 (Electronic)).

14. Rugbjerg M, Iepsen UW, Jorgensen KJ, Lange P. Effectiveness of pulmonary rehabilitation in COPD with mild symptoms: a systematic review with meta-analyses. 20150506(1178-2005 (Electronic)).

15. Kurfirst V, Mokracek A Fau - Krupauerova M, Krupauerova M Fau - Canadyova J, et al. Health-related quality of life after cardiac surgery--the effects of age, preoperative conditions and postoperative complications. 20140326 DCOM- 20141208(1749-8090 (Electronic)).

16. Drageset J, Eide Ge Fau - Ranhoff AH, Ranhoff AH. Mortality in nursing home residents without cognitive impairment and its relation to self-reported health-related quality of life, sociodemographic factors, illness variables and cancer diagnosis: a 5-year follow-up study. 20130220 DCOM- 20140418(1573-2649 (Electronic)).

17. von Elm E, Altman DG, Egger M, et al. The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement: guidelines for reporting observational studies. Journal of clinical epidemiology. 2008 Apr;61(4):344-9.

R1

R2

R3

R4

R5

R6

R7

R8

R9

R10

R11

R12

R13

R14

R15

R16

R17

R18

R19

R20

R21

R22

R23

R24

R25

R26

R27

R28

R29

R30

R31

R32

R33

R34

R35

R36

R37

R38

R39

Chapter 10

158


19. Andresen EM, Gravitt GW, Aydelotte ME, Podgorski CA. Limitations of the SF-36 in a sample of nursing home residents. Age and ageing. 1999 Oct;28(6):562-6.

20. VanderZee KI, Sanderman R, Heyink JW, de Haes H. Psychometric qualities of the RAND 36-Item Health Survey 1.0: a multidimensional measure of general health status. International journal of behavioral medicine. 1996;3(2):104-22.

21. Ware JE KM, Keller SD. SF-36 Physical and Mental Health Summary Scales: A Users Manual. Boston: The Health Institute; 1994.

22. Schemper M, Henderson R. Predictive accuracy and explained variation in Cox regression. Biometrics. 2000 Mar;56(1):249-55.

23. Choodari-Oskooei B, Royston P, Parmar MK. A simulation study of predictive ability measures in a survival model I: explained variation measures. Statistics in medicine. 2012 Oct 15;31(23):2627-43.



26. Ul-Haq Z Fau - Mackay DF, Mackay Df Fau - Pell JP, Pell JP. Association between physical and mental health-related quality of life and adverse outcomes; a retrospective cohort study of 5,272 Scottish adults. 20141203(1471-2458 (Electronic)).

27. Xie G, Laskowitz DT, Turner EL, et al. Baseline health-related quality of life and 10-year all-cause mortality among 1739 Chinese adults. 20140710 DCOM- 20150330(1932-6203 (Electronic)).

28. Ediebah DE, Coens C, Zikos E, et al. Does change in health-related quality of life score predict survival? Analysis of EORTC 08975 lung cancer trial. 20140514 DCOM- 20140715(1532-1827 (Electronic)).

29. Landman GW, van Hateren KJ, Kleefstra N, et al. Health-related quality of life and mortality in a general and elderly population of patients with type 2 diabetes (ZODIAC-18). Diabetes care. 2010 Nov;33(11):2378-82.

30. Kleefstra N, Landman GW, Houweling ST, et al. Prediction of mortality in type 2 diabetes from health-related quality of life (ZODIAC-4). Diabetes care. 2008 May;31(5):932-3.

31. Tsai SY, Chi LY, Lee CH, Chou P. Health-related quality of life as a predictor of mortality among community-dwelling older persons. European journal of epidemiology. 2007;22(1):19-26.

32. Morris PL, Raphael B Fau - Robinson RG, Robinson RG. Clinical depression is associated with impaired recovery from stroke. 19921201 DCOM- 19921201(0025-729X (Print)).

33. Saxena SK, Ng TP, Koh G, Yong D, Fong NP. Is improvement in impaired cognition and depressive symptoms in post-stroke patients associated with recovery in activities of daily living? Acta neurologica Scandinavica. 2007 May;115(5):339-46.

34. Passchier E, Stuiver Mm Fau - van der Molen L, van der Molen L Fau - Kerkhof SIC, et al. Feasibility and impact of a dedicated multidisciplinary rehabilitation program on health-related quality of life in advanced head and neck cancer patients. 20150530(1434-4726 (Electronic)).

35. Ter Hoeve N, van Geffen ME, Post MW, et al. Participation in society in patients with coronary artery disease before and after cardiac rehabilitation. 20150605(1532-821X (Electronic)).

36. Chang HT, Liu LF, Chen CK, et al. Correlates of institutionalized senior veterans’ quality of life in Taiwan. Health and quality of life outcomes. 2010;8:70.

37. Fan VS, Au DH, McDonell MB, Fihn SD. Intraindividual change in SF-36 in ambulatory clinic primary care patients predicted mortality and hospitalizations. J Clin Epidemiol. 2004 Mar;57(3):277-83.

R1

R2

R3

R4

R5

R6

R7

R8

R9

R10

R11

R12

R13

R14

R15

R16

R17

R18

R19

R20

R21

R22

R23

R24

R25

R26

R27

R28

R29

R30

R31

R32

R33

R34

R35

R36

R37

R38

R39


159

10

SUPP

LEM

ENTA

L D

ATA

App

endi

x Ta

ble

1. B

asel

ine

char

acte

risti

cs fo

r so

mati

c an

d re

habi

litati

on d

epar

tmen

t and

pati

ents

with

mis

sing

and

with

out m

issi

ng d

ata.

Som

atic

pati

ents

Reha

bilit

ation

pati

ents

p-va

lue

Mis

sing

dat

aW

itho

ut m

issi

ng d

ata

p-va

lue

Char

acte

risti

cN

=25

N=

159

Dem

ogra

phic

sA

ge

a

80 (6

1-85

)82

(76-

86)

0.01

82 (7

8-91

)79

(75-

85)

0.1

6

Fem

ale

Gen

der

b22

(61%

)89

(72%

)0

.20

17 (6

8%)

111

(70%

)0.

86Bo

dy m

ass

inde

x, k

g/m

2 a

28 (2

3-32

)25

(23-

28)

0.02

26 (2

3-29

)27

(23-

29)

0.64

Hyp

erte

nsio

n b

23 (6

4%)

101

(82%

)0.

0223

(92%

)12

4 (7

8%)

0.1

0

His

tory

of C

VD b

20 (5

6%)

47 (3

8%)

0.0

616

(64%

)67

(42%

)0.

04D

iabe

tes

mel

litus

b14

(39%

)54

(44%

)0.

596

(24%

)68

(43%

)0.

08D

emen

tia b

3 (8

%)

8 (7

%)

0.70

10 (4

0%)

11 (7

%)

<0.0

01Ps

ychi

atri

c di

seas

e b

22 (6

1%)

32 (2

6%)

<0.0

0113

(52%

)54

(34%

)0.

08Pu

lmon

ary

dise

ase

b7

(19%

)33

(27%

)0.

378

(32%

)40

(25%

)0.

47Cu

rren

t sm

oker

b9

(25%

)12

(8%

)0.

023

(12%

)21

(13%

)0.

87N

umbe

r of

med

icin

e a

9 (7

-12)

9 (6

-10)

0.57

9 (8

-13)

9 (6

-11)

0.15

Mea

sure

men

tsPh

ysic

al fu

nctio

ning

a5

(0-2

9)15

(5-3

0)0.

02-

10 (5

-30)

-So

cial

func

tioni

ng a

50 (5

0-63

)50

(50-

63)

0.99

-50

(50-

63)

-Ro

le fu

nctio

ning

-phy

sica

l a63

(0-1

00)

0 (0

-25)

<0.0

01-

0 (0

-50)

-Ro

le-f

uncti

onin

g-em

otion

al a

100

(0-1

00)

67 (0

-100

)0

.12

-83

(0-1

00)

-M

enta

l hea

lth a

66 (4

8-84

)68

(60-

80)

0.29

-68

(56-

80)

-

Bodi

ly p

ain

a68

(45-

100)

57 (2

2-78

)0.

02-

67 (2

2-80

)-

Vita

lity

a55

(31-

79)

65 (5

0-75

)0.

19-

65 (4

5-75

)-

Gen

eral

hea

lth p

erce

ption

a53

(40-

79)

55 (4

5-75

)0.

72-

55 (4

0-75

)-

Hea

lth c

hang

e a

50 (2

5-50

)50

(25-

50)

0.2

0-

50 (2

5-50

)-

PCS

Scor

e a

53 (3

9-69

)42

(33-

54)

0.03

-44

(34-

57)

-M

CS S

core

a62

(46-

76)

59 (4

8-74

)0.

89-

69 (4

8-74

)-

Dat

a ar

e m

edia

ns (i

nter

quar

tile

rang

e) o

r N

(%).

a M

ann-

Whi

tney

U te

st w

as u

sed

to c

ompa

re g

roup

s. b

Chi S

quar

e te

st w

as u

sed

to c

ompa

re g

roup

s.

Abb

revi

ation

s CV

D =

Car

dio

vasc

ular

dis

ease

. PCS

= p

hysi

cal c

ompo

nent

sum

mar

y. M

CS =

Men

tal c

ompo

nent

sum

mar

y.

R1

R2

R3

R4

R5

R6

R7

R8

R9

R10

R11

R12

R13

R14

R15

R16

R17

R18

R19

R20

R21

R22

R23

R24

R25

R26

R27

R28

R29

R30

R31

R32

R33

R34

R35

R36

R37

R38

R39

Chapter 10

160

Appendix Table 2: Relationship between HRQOL and all-cause mortality stratified by type of department, only adjusted for age and gender.

Somatic department (N=36)HR (95%CI)

Rehabilitation department (N=123)HR (95%CI)

Physical functioning 0.90 (0.70-1.15) 1.01 (0.90-1.14)Social functioning 1.07 (0.79-1.47) 1.04 (0.90-1.20)Role functioning-physical 1.05 (0.93-1.18) 1.05 (0.98-1.14)Role functioning-emotional 1.01 (0.91-1.12) 1.00 (0.94-1.05)Mental health 0.93 (0.76-1.14) 0.80 (0.69-0.94)Bodily pain 0.94 (0.82-1.08) 1.03 (0.95-1.13)Vitality 0.85 (0.70-1.02) 0.90 (0.77-1.04)General health perception 0.93 (0.78-1.12) 0.85 (0.74-0.99)Health change 0.85 (0.70-1.02) 0.96 (0.86-1.06)PCS 0.91 (0.71-1.20) 1.01 (0.83-1.22)MCS 0.93 (0.74-1.17) 0.87 (0.71-1.04)

The hazard ratios refer to a 10-point higher score on the RAND-36 dimensions. HR: hazard ratio; CI: confidence interval. PCS = physical component summary. MCS = Mental component summary. Bold values correspond to a p-value of 0.05 or less.

Appendix table 3: Relationship between HRQOL and all-cause mortality stratified to rehabilitation department, adjusted for all selected confounders (the results for the total group are presented for comparison).

All-cause mortality (N=159)Total group (N=159)HR (95%CI)

Rehabilitation department (N=123)HR (95%CI)

Physical functioning 0.99 (0.89-1.10) 1.02 (0.91-1.15)Social functioning 1.03 (0.91-1.17) 1.02 (0.88-1.18)Role functioning-physical 1.08 (1.02-1.15) 1.08 (1.00-1.18)Role functioning-emotional 1.01 (0.96-1.06) 1.00 (0.94-1.06)Mental health 0.86 (0.75-0.98) 0.85 (0.72-1.01)Bodily pain 1.01 (0.94-1.08) 1.04 (0.95-1.13)Vitality 0.88 (0.77-0.99) 0.94 (0.80-1.10)General health perception 0.91 (0.81-1.03) 0.90 (0.78-1.05)Health change 0.95 (0.87-1.05) 1.00 (0.90-1.13)PCS 1.02 (0.88-1.20) 1.05 (0.87-1.28)MCS 0.91 (0.88-1.06) 0.91 (0.75-1.12)

Adjusted for age, gender, smoking, DM, history of CVD, hypertension, BMI, history of pulmonary disease, history of psychiatric disease, and the number of medications. The hazard ratios refer to a 10-point higher score on the RAND-36 dimensions. Abbreviations HR: hazard ratio; CI: confidence interval. PCS = physical component summary. MCS = Mental component summary. Bold values correspond to a p-value of 0.05 or less.

CHAPTER 11Discussion

R1

R2

R3

R4

R5

R6

R7

R8

R9

R10

R11

R12

R13

R14

R15

R16

R17

R18

R19

R20

R21

R22

R23

R24

R25

R26

R27

R28

R29

R30

R31

R32

R33

R34

R35

R36

R37

R38

R39

Chapter 11

162

R1

R2

R3

R4

R5

R6

R7

R8

R9

R10

R11

R12

R13

R14

R15

R16

R17

R18

R19

R20

R21

R22

R23

R24

R25

R26

R27

R28

R29

R30

R31

R32

R33

R34

R35

R36

R37

R38

R39

Discussion

163

11

The objectives of this thesis were: to study the methods of orthostatic hypotension (OH) measurement itself; to study the clinical implications of OH in elderly patients; and to study factors (including OH) that are related to mortality and successful rehabilitation in nursing home residents. In this chapter, the most important findings and the possible implications of findings will be presented and limitations and strengths in general and of the specific studies will be discussed. The discussion is structured in three parts: first the epidemiology, diagnostic strategies and definition of OH will be discussed; second the relationship between OH and different endpoints; and finally risk prediction capabilities of various supposed risk factors in a nursing home population will be discussed.

MAIN FINDINGS AND IMPLICATIONS FOR DAILY PRACTICE

pide iologOH is a clinical sign commonly seen in elderly subjects. Population aging is associated with an increasing prevalence of OH. In chapter 6, a prevalence of OH of 37% in a nursing home population was found. The prevalence ranged from 29% on the somatic department, to 37% and 41% on the rehabilitation and the psychogeriatric department, respectively [1]. The prevalence of OH is also high in patients with a history of hypertension [2], cardiovascular diseases [3, 4], diabetes [5], and during acute hospitalization [6, 7]. The latter is described in chapter 7 of this thesis; a prevalence of OH of 34% was reported in hospitalized elderly patients with a hip fracture. Despite the fact that OH frequently occurs in older subjects and seems to represent physical vulnerability and may lead to adverse outcomes [8, 9], OH is also described in comparatively healthy younger individuals [2, 10]. In chapter 2, a prevalence of OH of 66% was seen in adults >50 years old visiting an outpatient clinic of internal medicine, measured with a continuous blood pressure (BP) device. Even a higher prevalence of OH (94.1%) was seen in the study of Romero-Ortuno within community-dwelling participants aged 60 years and older, measured with a continuous BP device [11]. Patients with OH can experience orthostatic complaints like light-headedness, dizziness, or syncope. In our studies, the prevalence of orthostatic complaints ranged from 19% in elderly admitted to the hospital, to 22% in community-dwelling elderly, and to 25% in nursing home patients [1].

R1

R2

R3

R4

R5

R6

R7

R8

R9

R10

R11

R12

R13

R14

R15

R16

R17

R18

R19

R20

R21

R22

R23

R24

R25

R26

R27

R28

R29

R30

R31

R32

R33

R34

R35

R36

R37

R38

R39

Chapter 11

164

iagnostic strategies and de nition o rthostatic h potensionIn 1996, a consensus committee of the American Autonomic Society and the American Academy of Neurology defined OH as a drop in systolic blood pressure (SBP) of >20 mmHg and/or diastolic blood pressure (DBP) of >10 mmHg after postural change [12, 13]. This definition was based on clinical judgment, as epidemiologic data were unavailable at that time. In response to the study of Fedorowski et al. in 2011 [2], an updated consensus statement was published [14], and a higher cut-off point (DSBP >30 mmHg) was advised for patients with supine hypertension (SBP >160 mmHg). The impact of OH, as defined according to the consensus definition, on different clinical endpoints is questionable. The unclear association of OH with endpoints such as falling, rehabilitation and mortality, might partly be due to a poorly chosen definition of OH. In chapter 9 it was shown that OH, defined according to the 1996 and 2011 consensus definitions, is not related to orthostatic complaints or previous fall incidents in community-dwelling participants aged 65 years and older. Only a decrease of >25% in SBP was related to complaints, whereas a decrease of >25% in DBP was related to previous fall incidents. We performed a post-hoc analysis within a nursing home population, described in chapter 5, to assess whether a >25% decrease in DBP and SBP were indeed related to previous fall incidents and orthostatic complaints. Neither the standard 1996 definition nor a >25% decrease in DBP and SBP were found to be associated with previous fall accidents or orthostatic complaints. As the latter results were examined in a much smaller study population, it is quite evident that larger studies are necessary to replicate the present study. Besides, more research is necessary to either develop useful cut-off points for OH, or refute the notion that such measurements do have any value at all for clinical decision-making or prediction of incidents. Although the studies presented in this thesis do not provide sufficient evidence to favour one particular definition, it can be suggested that it is worthwhile to at least include a relative decrease instead of an absolute fixed value in the definition. This approach is supported by a previous study of Fedorowski et al. [2], although they chose fixed cut-off points because this seemed more feasible for clinical practice. A decrease of >25% in DBP and SBP seems a reasonable approach, although more prospective studies are needed to investigate this suggested relative decrease of BP as definition for OH.

The international consensus definition recommends continuous beat-to-beat BP measurement to diagnose OH, and it is advised to compare BP measurements after 5 minutes of rest in supine position to BP measurement within 3 minutes in standing position [12]. A previous study suggested a lack of specificity for diagnosing OH by using the continuous beat-to-beat BP that could lead to incorrect diagnosis [11]. However, in daily practice automated sphygmomanometers are commonly used for this purpose, but due to the inability to

R1

R2

R3

R4

R5

R6

R7

R8

R9

R10

R11

R12

R13

R14

R15

R16

R17

R18

R19

R20

R21

R22

R23

R24

R25

R26

R27

R28

R29

R30

R31

R32

R33

R34

R35

R36

R37

R38

R39

Discussion

165

11

register BP at each heartbeat, this results in a quite significant risk of underreporting with regard to the lowest BP readings compared to continuous measurement of BP [15, 16]. Finally, as many elderly patients are not able to stand for several minutes, orthostatic BP measurements in standing position after 5 minutes supine rest cannot always be performed. In chapter 2 a difference was seen in BP response between standing and sitting postural change. Although no significant difference in prevalence of OH was observed, the positive and negative proportion of agreement of the prevalence of OH were at most moderate which indicates relevant differences in diagnosing OH on an individual level depending on the method used. In addition, in chapter 3 no difference in prevalence of OH was observed between OH measurements using the continuous or the interval BP measurement device but the positive and negative proportion of agreement were low. Therefore, the continuous BP measurement cannot be replaced by and compared to an interval BP measurement to diagnose OH. Taken all together, postural change to a standing position cannot be substituted by allowing a sitting position and a continuous BP measurement cannot be replaced by an interval BP measurement to diagnose OH in elderly patients, since these methods will result in the identification of partially different populations to have OH.

rthostatic h potension and clinical i plications in elderlIn previous literature, OH has been found to be associated with different clinical endpoints like falling, cardiovascular diseases, and mortality. Although such assumptions were made, those supposed associations have become questionable given the lack of evidence in the current literature. In the section below, the relationship between OH and the possible clinical implications in the elderly population will be discussed.

rthostatic h potension and allingIn chapters 4, 5, and 6 the previously poorly characterized relationship between OH and falling was described. In the prospective cohort study in chapter 4, no relationship between OH and falling or recurrent falling was found in a nursing home population. However, the individual patient data (IPD) meta-analysis presented in chapter 5 showed a significant relationship between OH and time to first fall incident; the hazard ratio (HR) in the one-stage cox proportional hazard model was 1.52 (95% Confidence Interval (CI) 1.23-1.88). No significant relationship between OH and falling was found in the one-stage logistic regression analysis (Odds Ratio (OR) 1.21 (95% CI 0.87-1.68) and the two-stage logistic and cox regression analyses. Since the number of eligible studies was limited, only a small number of prospective studies were included in this meta-analysis. Within the scope of this small number of studies, this meta-analysis is the only study that used the consensus definition of OH, included a representative group of elderly, and adjusted for important confounders within individual patient data.

R1

R2

R3

R4

R5

R6

R7

R8

R9

R10

R11

R12

R13

R14

R15

R16

R17

R18

R19

R20

R21

R22

R23

R24

R25

R26

R27

R28

R29

R30

R31

R32

R33

R34

R35

R36

R37

R38

R39

Chapter 11

166

In chapter 6, the relation between OH, orthostatic complaints, and previous falling was investigated in a nursing home population; no significant relationships were found.

Several reviews described a theoretical relationship [17, 18], but did not perform a meta-analysis with the available data due to the small number of studies [19]. Besides, important limitations were seen in previous studies that described the relationship between OH and falling. The lack of prospective fall data [11, 20, 21], the lack of adjustment for important confounders [22, 23], and the lack of using the international consensus definition of OH [23-25] are examples of limitations in previous studies. In the previous studies OH was mentioned as an independent risk factor for recurrent falls [26, 27], a decrease in SBP was reported to be related to falling [22, 25], and symptomatic OH or orthostatic complaints were described as a predictor for falling [5, 28]. In our prospective studies, orthostatic complaints and symptomatic OH were investigated; no significant relationships were seen with (recurrent) falling or previous fall incidents (chapters 4 and 6) [1].Because of the small number of studies included in the meta-analysis (chapter 5), we performed a post-hoc analysis in which we included the results of our prospective study into the meta-analysis (table 2). Patients with OH had a HR of 1.26 (95%CI 1.04-1.52) on a first fall incident in the study periods assessed. Although a significant relationship was seen in the post-hoc analysis, the HR decreased considerably compared to the HR in the original meta-analysis presented in chapter 5 (HR 1.52 (95%CI 1.23-1.88).By performing an IPD meta-analysis including prospective studies it can be concluded that OH could still be a predictor for a first fall incident. The lower boundary of the 95% confidence interval, after including the results of chapter 5, does not exclude the possibility of the association being irrelevant. Furthermore, the odds ratios of falling were not significantly different between patients with and without OH. Therefore, more prospective studies are needed for a precise estimate of the relationship between OH and falling.

Table 1. Adjusted odds ratios and Hazard ratios for the effect of orthostatic hypotension on the risk of falling with individual patient data (one-stage method). The odds ratios can be interpreted as a measure of the association of OH to falling (the dependent variables). Hazard ratios refer to time to first fall incident.

Odds ratio (95% CI)(4 studies)

Hazard ratio (95% CI)(3 studies)

OH (model 1) 0.84 (0.66-1.07) (n=1271) 1.09 (0.92-1.30) (n=1199)OH (model 2) 0.83 (0.65-1.06) (n=1271) 1.08 (0.91-1.28) (n=1199)OH (model 3) 1.05 (0.80-1.38) (n=1203) 1.26 (1.04-1.52) (n=1131)

Model 1: unadjusted. Model 2: adjusted for gender and age. Model 3: adjusted for gender, age, BMI, DM, antihypertensive medication, number of medication, and systolic blood pressure. OH = orthostatic hypotension. DM = Diabetes Mellitus.

R1

R2

R3

R4

R5

R6

R7

R8

R9

R10

R11

R12

R13

R14

R15

R16

R17

R18

R19

R20

R21

R22

R23

R24

R25

R26

R27

R28

R29

R30

R31

R32

R33

R34

R35

R36

R37

R38

R39

Discussion

167

11

rthostatic h potension and success ul rehabilitationNursing homes in the Netherlands provide care and long-stay facilities for elderly with chronic mental or physical diseases and the majority also provides rehabilitation services. Since the prevalence of OH in elderly staying in nursing homes is high and OH considered to be related to falling, we hypothesized that OH could be an important prognostic factor for chances of rehabilitation in a nursing home population. In chapters 6 and 7 the relationship between OH and successful rehabilitation was described in a nursing home population and in hospitalized elderly with a hip fracture, respectively. In nursing home residents, patients with OH were found to have a higher chance of successful rehabilitation compared to patients without OH (chapter 6) [1]. A possible explanation for this rather counterintuitive relationship could be that either patients with OH are more adapted to their disabilities (including having already adjusted their home environment to their disabilities) or may be used to a less active life, allowing a quicker discharge to their own home. The results could also be a matter of coincidence. Conflicting results, as presented in chapter 6, were seen in hospitalized patients with a hip fracture; OH measured during the first days of hospitalization was not related to rehabilitation (chapter 7). When comparing both study populations, patients included in the study with hospitalized hip fracture patients seemed to have less multi-morbidity, used less medication, and baseline blood pressure was lower. The high prevalence of OH in the hospitalized patients could be partially caused by the hip fracture itself and hospital admission-related factors like bed rest, surgery, inadequate water intake, and blood loss. In these circumstances, OH may very well be a temporary phenomenon and therefore not a predictor for an endpoint like successful rehabilitation [7]. In the study of Weiss et al., the impact of OH in hospitalized patients on mortality was described, and they advised to divide patients in 2 groups; patients with episodic OH, as is seen during hospitalization, and established OH (repeated measurements) [6]. Analogous to the association with mortality, one may hypothesize that episodic and sustained OH have different associations with rehabilitation. Episodic OH may have no consequences for chances of rehabilitation, whereas sustained OH may be much more relevant. Measuring OH in the first week of rehabilitation within a nursing home might possibly be a more accurate predictor for successful rehabilitation. Also, relationships between different clinically relevant factors and successful rehabilitation were described in chapter 7. The factors diabetes mellitus (DM), SBP, handgrip strength and fear of falling (FOF) were significantly related to successful rehabilitation. The hazard of successful rehabilitation in patients with DM was lower than patients without DM (HR 0.47 (95%CI 0.26-0.85)), which was also reported in the previous study (chapter 6). It is known that coexisting disease negatively influences the outcome of rehabilitation [29]. Several studies described a clear relationship between DM and poor rehabilitation outcome [30, 31].

R1

R2

R3

R4

R5

R6

R7

R8

R9

R10

R11

R12

R13

R14

R15

R16

R17

R18

R19

R20

R21

R22

R23

R24

R25

R26

R27

R28

R29

R30

R31

R32

R33

R34

R35

R36

R37

R38

R39

Chapter 11

168

The chances of successful rehabilitation increased by 15% (95%CI 3-28%) for every 10-mmHg increase in SBP. In a previous study, higher blood pressure in frail patients was related to lower all-cause mortality while the opposite relationship was seen in non-frail patients [32]. Therefore, it was not unexpected that higher SBP is associated with a higher chance of successful rehabilitation. Handgrip strength (HR 1.05 (95%CI 1.01-1.08)) was positively and FOF (HR 0.87 (95%CI 0.79-0.97)) was negatively associated with time to successful rehabilitation. Poor muscle strength and FOF are frequently seen in elderly patients, and these factors are also related with the level of frailty [33-35] and functional outcome [36-39]. Although no relation was found regarding handgrip strength and successful rehabilitation, in the analysis with handgrip strength as a confounder in the analysis of OH in relation to successful rehabilitation, handgrip strength turned out to be a significant confounder (chapter 7). Despite the fact that no evidence was provided for a relationship between handgrip strength and time to successful rehabilitation in chapter 7, previous

studies described that measuring handgrip strength preoperatively reflects the baseline condition of a patient and is a predictor of complications or length of hospital stay [40, 41]. Handgrip strength measurement can be used to identify those patients who are frailer and might need a different approach during hospital stay [40]. Normative values of handgrip strength are available due to the large study of Dodds et al [42]. In elderly aged >65 years a handgrip strength <26 kg in men and <16 kg in women is considered as weak [43]. In the study described in chapter 7, 31% of the patients met these cut-off points of weak handgrip strength.FOF is common, being present in up to 50% of elderly patients who are rehabilitating, and is related to outcome [37, 44]. Previous studies described that FOF may restrict physical activity, which causes immobility and further loss of functional independence and falling [45, 46].

rthostatic h potension and ortalit in elderlIn chapter 8 it was concluded that OH was related to increased all-cause mortality, but only in patients at the psychogeriatric department [47]. In previous studies it was shown that the presence of OH is associated with an increase in the risk of cardiovascular disease, cardiovascular mortality or all-cause mortality in elderly people [19, 48-55]. Although several explanations of the pathophysiological link between OH and mortality are described, it is questionable whether OH is the independent causal factor or the result of one or more other components within the pathophysiological process [3, 4]. Diseases like DM, atrial fibrillation, hypertension, and Parkinson disease are all related to both OH and mortality. Furthermore, baroreflex dysfunction and impaired hemodynamic homeostasis play an important role in mediating cardiovascular disease and mortality, but are also mentioned in the pathogenesis of OH [3].

R1

R2

R3

R4

R5

R6

R7

R8

R9

R10

R11

R12

R13

R14

R15

R16

R17

R18

R19

R20

R21

R22

R23

R24

R25

R26

R27

R28

R29

R30

R31

R32

R33

R34

R35

R36

R37

R38

R39

Discussion

169

11

In chapter 8 only in patients at the psychogeriatric department a relationship was seen between OH and all-cause mortality [47]. Apparently, a decrease in blood pressure upon standing is related to increased mortality within a cognitive impaired population. Although a recent retrospective study showed no associations between dementia and OH [56], dementia and its (cardiovascular) causes may explain the observations in chapter 8.

Arterial stiffness and lower blood pressure will cause hypoperfusion of the brain, and are associated with accelerated cognitive decline [57, 58]. Also, autonomic dysfunction caused by neurodegenerative diseases like dementia could lead to OH [57]. In spite of the fact that OH was related to mortality within this specific population, both Harrell’s C and R2 showed only small improvements when adding OH to the multivariate models, thus the additional value of OH in risk prediction is limited. Two recent meta-analyses of observational studies described a significant relationship between OH and mortality. In both analyses, OH was related to an increased risk of all-cause mortality [3, 53]. In the most recent meta-analysis, the presence of OH, compared to the absence of OH, was associated with all-cause mortality with a relative risk (RR) of 1.50 (95%CI 1.24-1.81) in the total study population followed by a RR of 1.26 (95%CI 0.99-1.62) in patients aged 65 years or older [3]. These meta-analyses are limited by the lack of sufficient appropriate studies. Both meta-analyses described that the overall effect of OH on mortality is possibly mediated by classic risk factors instead of an independent causal mechanism [3, 53]. Taken this all together, a relationship between OH and mortality was seen, but the additional value of OH with respect to mortality in elderly patients is limited.

is prediction capabilities o blood pressure and in nursing ho e patients

lood pressure in elderlInternational guidelines regarding blood pressure targets and treatment in elderly are inconsistent due to limited outcome data in this population [59]. In chapter 8, we described the relationship between blood pressure and mortality in a nursing home population. A significant relationship was only observed between DBP and all-cause mortality. The overall mortality risk increased by 17% (95%CI 2-34%) for every 10-mmHg increase in DBP [47]. Previous studies showed that a low DBP was associated with an increased all-cause mortality risk, especially in the oldest and frailest individuals [60, 61]. In the study by Odden et al. stratified analyses according to frailty were performed; in non-frail patients a positive relationship was described and in frail patients blood pressure was not related to mortality [62]. Van Hateren et al. described that in elderly patients with diabetes, higher blood pressure was related to increased cardiovascular mortality in non-frail patients, while an inverse relationship was seen in frail patients [32]. Taken this all together, a relationship between blood pressure and mortality in the elderly apparently exists, but this seems to be affected by the level of frailty.

R1

R2

R3

R4

R5

R6

R7

R8

R9

R10

R11

R12

R13

R14

R15

R16

R17

R18

R19

R20

R21

R22

R23

R24

R25

R26

R27

R28

R29

R30

R31

R32

R33

R34

R35

R36

R37

R38

R39

Chapter 11

170

Although the study described in chapter 8 showed that DBP is an independent risk factor for mortality, its practical implications still remain to be determined. Based on the lack of increase in Harrell’s C values when adding DBP to the adjusted models, one may conclude that the additional value of DBP in mortality prediction is very limited. It is important to realize that this study group is a frail group of patients with multi-morbidity, probably much more determinative for mortality than DBP itself. Only the authors of the observational PARTAGE study and the study of Askari et al. had specifically investigated this relationship in a nursing home population before [63, 64]. The authors of the PARTAGE study described that lower DBP was associated with higher mortality (RR 0.84 (95%CI 0.72-0.99)) [65]. No association between blood pressure and cardiovascular outcome was reported in the study of Askari et al. [64].The predictive value of BP in nursing home patients is not only limited, but the need for intensive treatment of hypertension in elderly remains questionable as well. The results from both the HYVET and the SPRINT studies suggest intensive treatment of hypertension in elderly to be beneficial [66, 67]. Both studies showed a reduction of cardiovascular events and deaths from any cause after antihypertensive therapy. However, the generalizability of both studies appeared to be limited because mostly healthy elderly patients were included. In the HYVET study, patients with heart failure, dementia and/or requiring nursing care were excluded [66]. In the SPRINT study, patients with a history of DM, stroke, symptomatic heart failure or left ventricular ejection fraction <35%, a clinical diagnosis of dementia, living in a nursing home, an expected survival of less than 3 years, and a SBP of less than 110 mmHg following 1 minute of standing were excluded [67]. By excluding large groups of elderly subjects, the patients in these studies do not adequately represent the general elderly population. This selection bias is illustrated by the low mortality rates in both treatment groups in the SPRINT trial. We compared the mortality rates in the SPRINT trial to the mortality rates of the age and gender-matched general populations in the Netherlands in 2012 [68] and the United States (US) in 2010 [69] (figure 1). Remarkably, both the standard and the intensive treatment groups in the SPRINT trial had a two-fold better life expectancy than the life expectancy of the age and gender-matched general populations in the Netherlands and the US. Therefore, the results of the SPRINT trial will only have implications for a very select group of vital elderly patients. At this moment, the Dutch guideline ‘Cardiovascular risk management’ advises antihypertensive treatment in case of SBP >140 mmHg or SBP >160 mmHg within patients above 70 or 80 years old, respectively [70]. Treatment goals described in the Dutch guideline are mainly based on studies like the HYVET trial and do not represent the frail elderly patient like the patients living in a nursing home. The US guideline recommends a SBP treatment target of 150 mmHg for adults >60 years of age [71].

R1

R2

R3

R4

R5

R6

R7

R8

R9

R10

R11

R12

R13

R14

R15

R16

R17

R18

R19

R20

R21

R22

R23

R24

R25

R26

R27

R28

R29

R30

R31

R32

R33

R34

R35

R36

R37

R38

R39

Discussion

171

11

Since the predictive value of BP in nursing home patients is limited and hypertension trials within this population are lacking, a higher SBP target value (also even higher than 160 mmHg) can be considered as acceptable in specific subjects and antihypertensive treatment might be minimized. Beside BP, one should take into account quality of life in this specific population compared to younger (and more vital) patients since side effects of antihypertensive treatment could have serious consequences. Large trials investigating antihypertensive treatment in elderly, without excluding large groups of elderly, are needed.

Health related uality of life in a nursing home population

ue

ri

F u e r

Figure 1. All-cause mortality rates of the SPRINT trial, and the age- and gender matched general populations in the Netherlands and United States of America.

ealth related ualit o li e in a nursing ho e populationHealth related quality of life (HRQOL) is considered to be an important outcome in the evaluation of interventions or treatment in different diseases [72]. Low HRQOL corresponds to substantial limitations in physical, emotional and social well-being and will to a varying degree be due to the presence of a specific medical condition or its treatment [73]. HRQOL in individual patients can be used to measure disease-related distress and overall perception of health. Next to the evaluation of HRQOL as a separate outcome measure, HRQOL also has a prognostic value in non-nursing home settings [74-76]; a lower HRQOL score has been associated with an increased mortality risk, also in elderly patients. In summary, deviations in

R1

R2

R3

R4

R5

R6

R7

R8

R9

R10

R11

R12

R13

R14

R15

R16

R17

R18

R19

R20

R21

R22

R23

R24

R25

R26

R27

R28

R29

R30

R31

R32

R33

R34

R35

R36

R37

R38

R39

Chapter 11

172

HRQOL could have a variety of implications in decision-making processes regarding patients and medical interventions, especially in elderly patients. In chapter 10, we discussed the relationship between HRQOL, rehabilitation, and mortality in nursing home patients. HRQOL as assessed by the RAND-36 was significantly associated with mortality in three dimensions, but partly in opposite directions. Higher scores on the dimensions vitality (HR 0.88 (95%CI 0.77-0.99)) and mental health (HR 0.86 (95%CI 0.75-0.98)) were related to a lower mortality risk, whereas a higher score on the dimension role functioning-physical (HR 1.08 (95%CI 1.02-1.15)) was related to a higher mortality risk. However, based on the minimal increase in Harrell’s C values when adding role functioning-physical, vitality or mental health to the adjusted models, the additional value of these dimensions in mortality prediction is limited. HRQOL represents a person’s perspective about his or her physical and emotional wellbeing and changes in health. Knowledge of a person’s perspective could have implications for decision-making by nursing home care givers [77]. Although the predictive value of HRQOL on mortality seems limited, it can still be advised to assess HRQOL for information about the degree of wellbeing experienced by a patient as such. For example, the AgeCoDe study showed that management of chronic diseases in elderly benefits from focusing on factors like HRQOL. Improvement in HRQOL was seen when the number of physical activities increased. Besides, HRQOL declined with increasing age, walking disability, and hearing impairment [78]. By using a HRQOL questionnaire regularly during nursing home admission, changes in HRQOL can be monitored and relatively simple approaches, like commencing physical activities, could improve HRQOL. Despite the minimal predictive value of HRQOL in nursing home residents, there are sufficient reasons, as mentioned above, to be informed about patients’ HRQOL in a nursing home population.

STRENGTHS AND LIMITATIONS

Several strengths and limitations need to be addressed. The main strengths of the studies presented in this thesis were the well-defined study populations, the prospective nature, and the number of clinically relevant variables adjusted for in multivariate analyses.The main limitations are the observational design and the small study samples. Because of the observational design, establishing a causal relation was not possible, only relationships were described. The small study sample decreases the precision and generalizability of the observed findings. In 1996, OH was defined as a drop in systolic blood pressure (SBP) of >20 mmHg or diastolic blood pressure (DBP) of >10 mmHg after postural change [12, 13]. This definition was based on clinical judgment, as epidemiologic data were unavailable at that time. In response to

R1

R2

R3

R4

R5

R6

R7

R8

R9

R10

R11

R12

R13

R14

R15

R16

R17

R18

R19

R20

R21

R22

R23

R24

R25

R26

R27

R28

R29

R30

R31

R32

R33

R34

R35

R36

R37

R38

R39

Discussion

173

11

the study of Fedorowski et al. in 2011 [2], an updated consensus statement was published [14], and a higher cut-off point (DSBP >30 mmHg) was advised for patients with supine hypertension (SBP >160 mmHg). In all studies described in this thesis the 1996 definition of OH was used. Although chapter 9 showed no relationship between OH defined by the update consensus statement and previous fall incidents or orthostatic complaints, it cannot be excluded that some results would be slightly different when using OH results as defined by the updated consensus statement.

In addition to these general strengths and limitations, specific strengths and limitations were present in the specific study designs. This thesis is based on different studies in three different populations; nursing home patients, hospitalized elderly, and community-dwelling elderly. The strengths and limitations of studies form each of these populations will be discussed separately below.

ursing ho e patientsA large part of this thesis was based on a representative Dutch nursing home population. This specific population is often excluded in studies in the elderly. The statement on representativeness is based on the fact that admission to a Dutch nursing home requires approval of a central indication committee, which implies that all patients are assessed using the same admission criteria. Besides, the nursing home facility in the present study was a general nursing home, with somatic, psychogeriatric and rehabilitation departments, comparable to most other Dutch nursing homes. No groups of patients were excluded in the present nursing home population. Several limitations need to be addressed. Firstly, OH measurements in elderly can be difficult due to mobility problems. Therefore, the postural change was performed from lying to standing in 36% of the somatic patients, 87% of the rehabilitation patients, and 66% of the psychogeriatric patients. The remaining patients performed the postural change from lying to sitting. Post-hoc analyses stratified according to postural position were performed, which did not relevantly change results. Based on chapter 2, where it was stated that postural change to standing position cannot be substituted by allowing a sitting position, it can not be excluded that differences in postural position has influenced the results. Secondly, recall problems regarding previous and prospective fall incidents need to be mentioned. Information regarding previous falls was based on questioning patients or participants and prospective fall incidents were registered by the medical staff. In patients with cognitive problems, recall bias could play a more important role than in a younger population. It is very likely that the actual number of patients with fall incidents was higher. Thirdly, although a validation study had been performed, the adequacy of using the RAND-36 questionnaire within a nursing home population has been questioned [79, 80]. An

R1

R2

R3

R4

R5

R6

R7

R8

R9

R10

R11

R12

R13

R14

R15

R16

R17

R18

R19

R20

R21

R22

R23

R24

R25

R26

R27

R28

R29

R30

R31

R32

R33

R34

R35

R36

R37

R38

R39

Chapter 11

174

important issue is that the RAND-36 entails several potential inappropriate questions for this population [79, 80]. Due to the high heterogeneity in the nursing home population in general, the use of the RAND-36 could be more suitable for subgroups of rehabilitation patients, like e.g. those with higher cognitive and physical functioning [80]. Finally, the impact of multi-morbidity in a frail group of elderly is considerable, influencing different endpoints like falling, successful rehabilitation, HRQOL, and mortality. Although we have adjusted for an important set of relevant variables in the multivariate model, multi-morbidity could still have relevantly influenced the results as presented in this thesis.

ospitali ed elderl patientsThe main strength of the study within hospitalized elderly patients was the setting in a general hospital, and only a few exclusion criteria were used, whereby this group can be seen as a representative group of elderly hip fracture patients. There were also important limitations. Firstly, although recruitment and testing took place preferably on the day of admission, 18% of the handgrip strength measurements were not measured preoperatively. However, stratified analyses according to timing of handgrip strength measurement did not alter the results. Secondly, as expected, the proportion of immobile patients was high in this study group. In 98% of the patients an OH measurement could not be performed from lying to standing. Like discussed before, the two postural changes identify different patients having OH.Finally, OH was only measured once during the follow-up period, which only detected episodic OH. By using repeated OH measurements sustained OH could also have been established. Future studies are needed to evaluate the clinical implications of sustained OH on rehabilitation.

o unit d elling elderlThe studies including community-dwelling elderly were performed in a general hospital on an outpatient clinic of internal medicine, and in elderly aged >60 years participating in a fall risk screening and prevention program.The strengths of the studies within the outpatient clinic were the randomization of the measurements (non-blinded randomization for both the sequence of the postural changes and the side of the BP measurements) and the fact that all measurements were performed and evaluated by the same individual to overcome inter-observer bias. The generalizability is limited to elderly patients visiting the internal medicine outpatient clinic. Limitations of these studies were the relatively small study sample. Due to the fact that the patients included in this study had to be able to stand for five minutes without assistance, the study group was slightly biased compared to the general population and the results are, of course, only useful in patients who are able to stand.

R1

R2

R3

R4

R5

R6

R7

R8

R9

R10

R11

R12

R13

R14

R15

R16

R17

R18

R19

R20

R21

R22

R23

R24

R25

R26

R27

R28

R29

R30

R31

R32

R33

R34

R35

R36

R37

R38

R39

Discussion

175

11

The strength of the study, including elderly subjects participating in the fall risk screening and prevention program, was that this study not only focused on elderly patients but also on older people without specific medical problems. Several limitations need to be addressed. Firstly, blood pressure measurement in the supine position was only performed once, namely at 3 minutes. This probably has led to an underestimation of the actual prevalence of OH, because OH is defined as a decrease of SBP or DBP within 3 minutes. Second, two types of bias could have occurred: recall and selection bias. Previous fall incidents were based on participants’ report, which could have resulted in recall bias. Selection bias may have occurred as patients participated in this study, were the selection of participants willing to participate and who actively made the effort to join this program. This could have led to registration of participants who had complaints more often or had gone through a previous fall incident. This might limit generalization to the general population of community-dwelling elderly.

CONCLUSION

The general aim of this thesis was to study the OH measurements itself, to study the clinical implications of OH in elderly patients, and to study factors (including OH) that are related to mortality and successful rehabilitation in nursing home residents.This thesis shows that the approach and interpretation of the finding of OH in elderly patients needs to be adjusted. OH measurements do not add much to clinical decision-making or predicting outcomes in the majority of elderly patients, in both outpatient and nursing home settings. OH defined by the current International consensus definition is not related to important clinical endpoints, at least not in the populations studied for this thesis. Moreover, it can be questioned whether an OH measurement is necessary in any circumstance in elderly patients, or that instead potential actions (e.g. de-prescribing antihypertensive treatment) should be based on the presence of orthostatic complaints. Furthermore, the mortality prediction capabilities of BP and HRQOL are very limited in a nursing home population. Since the predictive value of BP in nursing home patients is limited and hypertension trials within this population are lacking, the need for antihypertensive treatment should be individually assessed, taking into account factors like the degree of frailty, the estimated survival time, and the chances or presence of side effects of medication.Although the mortality prediction capability of HRQOL is limited in a nursing home population, measuring HRQOL in this patient category could lead to increased understanding of factors that negatively impact HRQOL. Since morbidity is already very high and life expectancy is short in nursing home patients, quality of life is probably the most important variable to evaluate.

R1

R2

R3

R4

R5

R6

R7

R8

R9

R10

R11

R12

R13

R14

R15

R16

R17

R18

R19

R20

R21

R22

R23

R24

R25

R26

R27

R28

R29

R30

R31

R32

R33

R34

R35

R36

R37

R38

R39

Chapter 11

176

Based on the results of this thesis it can be concluded that:1. Deviation from the current consensus on OH measurement will result in the

identification of a partially different population to be diagnosed with OH.a. Postural change during OH measurements in supine followed by a standing

position cannot be substituted by allowing a sitting position in the elderly population, since this will result in the identification of a partially different population (supposed) to have OH.

b. A continuous BP measurement cannot be replaced by and compared to an interval BP measurement to diagnose OH in elderly patients.

2. The clinical relevance of OH in elderly patients, using the consensus definition, is limited.

a. OH, defined according to the consensus, is not related to orthostatic complaints or previous fall incidents.

b. The relationship between OH and clinical endpoints like falling, rehabilitation, and mortality is minimal.

3. The mortality prediction capabilities of blood pressure and HRQOL are very limited in a nursing home population.

Finally, I have formulated the following recommendations for daily practice and future research.

1. he approach o in elderl patients needs to be changed: a. easure ent should not be part o usual care in clinical decision

a ing regarding OH measurements in elderly patients do not have additional value in clinical decision-making or predicting outcomes in the majority of elderly patients.

b n uiring a er orthostatic co plaints instead o easuring It can be questioned whether an OH measurement is necessary in any circumstance in elderly patients, or that instead potential actions (e.g. de-prescribing antihypertensive treatment) should be based on the presence of orthostatic complaints.

c n the e ceptional situation that an easure ent is per or ed it should be per or ed according to the current consensus de nition Despite that I recommend that OH measurements should not be part of usual care, most health care providers will not change the approach (in order to follow current guideline recommendations). In those cases, the clinician must take into account the facts that postural change to standing position cannot be replaced by postural change to sitting position and a continuous BP measurement cannot be replaced by and compared to an interval BP measurement in elderly patients.

R1

R2

R3

R4

R5

R6

R7

R8

R9

R10

R11

R12

R13

R14

R15

R16

R17

R18

R19

R20

R21

R22

R23

R24

R25

R26

R27

R28

R29

R30

R31

R32

R33

R34

R35

R36

R37

R38

R39

Discussion

177

11

2. urther research is needed to allo the identi cation o possible clinicall relevant cut off points or orthostatic h potension Since the current consensus definition is not related to important endpoints, its clinical relevance seems very limited, at least for the population studied in this thesis. More prospective studies are needed to investigate different definitions, including relative decreases, and its clinical relevance.

he role o orthostatic h potension on alling needs to be re uted or con r ed By performing an individual patient data meta-analysis with prospective studies it can be concluded that OH plays a role to experience a first fall incident. Since the results of this meta-analysis are imprecise, more prospective studies are needed.

4. ntih pertensive treat ent in elderl patients re uires an individual approach

Since the predictive value of BP in nursing home patients is limited and hypertension trials within this population are lacking, the need for antihypertensive treatment should be individually assessed, taking into account factors like the degree of frailty, the estimated survival time, and the chances or presence of side effects of medication. Furthermore, large trials investigating antihypertensive treatment in elderly, without excluding large groups of frail elderly and nursing home patients, are needed to develop tailor-made treatment.

5. tudies on in a nursing ho e are needed Knowledge of HRQOL in a nursing home could have implications for nursing home care givers. Especially, the influence of de-medicalizing, e.g. de-prescribing medicine, on HRQOL is of interest. Therefore prospective studies are needed to evaluate the use of HRQOL questionnaires as a helpful tool supporting clinical decision-making in nursing homes.

R1

R2

R3

R4

R5

R6

R7

R8

R9

R10

R11

R12

R13

R14

R15

R16

R17

R18

R19

R20

R21

R22

R23

R24

R25

R26

R27

R28

R29

R30

R31

R32

R33

R34

R35

R36

R37

R38

R39

Chapter 11

178

REFERENCES


2. Fedorowski A, Burri P, Melander O. Orthostatic hypotension in genetically related hypertensive and normotensive individuals. Journal of hypertension. 2009 May;27(5):976-82.

3. Ricci F, Fedorowski A, Radico F, et al. Cardiovascular morbidity and mortality related to orthostatic hypotension: a meta-analysis of prospective observational studies. European heart journal. 2015 Jul 1;36(25):1609-17.





8. O’Connell MD, Savva GM, Fan CW, Kenny RA. Orthostatic hypotension, orthostatic intolerance and frailty: The Irish Longitudinal Study on Aging-TILDA. Archives of gerontology and geriatrics. 2015 May-Jun;60(3):507-13.

9. Romero-Ortuno R, Cogan L, O’Shea D, Lawlor BA, Kenny RA. Orthostatic haemodynamics may be impaired in frailty. Age and ageing. 2011 Sep;40(5):576-83.

10. Low PA, Denq Jc Fau - Opfer-Gehrking TL, Opfer-Gehrking Tl Fau - Dyck PJ, et al. Effect of age and gender on sudomotor and cardiovagal function and blood pressure response to tilt in normal subjects. 19971223 DCOM- 19971223(0148-639X (Print)).


12. Lahrmann H, Cortelli P, Hilz M, et al. EFNS guidelines on the diagnosis and management of orthostatic hypotension. European journal of neurology: the official journal of the European Federation of Neurological Societies. 2006 Sep;13(9):930-6.

13. Kaufmann H. Consensus statement on the definition of orthostatic hypotension, pure autonomic failure and multiple system atrophy. Clinical autonomic research: official journal of the Clinical Autonomic Research Society. 1996 Apr;6(2):125-6.


15. Caine Se Fau - Alsop K, Alsop K Fau - Mac Mahon M, Mac Mahon M. Overlooking orthostatic hypotension with routine blood-pressure equipment. 19980827 DCOM- 19980827(0140-6736 (Print)).


17. Shaw BH, Claydon VE. The relationship between orthostatic hypotension and falling in older adults. Clinical autonomic research: official journal of the Clinical Autonomic Research Society. 2014 Feb;24(1):3-13.

R1

R2

R3

R4

R5

R6

R7

R8

R9

R10

R11

R12

R13

R14

R15

R16

R17

R18

R19

R20

R21

R22

R23

R24

R25

R26

R27

R28

R29

R30

R31

R32

R33

R34

R35

R36

R37

R38

R39

Discussion

179

11

18. Ambrose AF, Paul G, Hausdorff JM. Risk factors for falls among older adults: A review of the literature. Maturitas. 2013;75(1):51-61.







25. Gangavati A, Hajjar I, Quach L, et al. Hypertension, orthostatic hypotension, and the risk of falls in a community-dwelling elderly population: the maintenance of balance, independent living, intellect, and zest in the elderly of Boston study. Journal of the American Geriatrics Society. 2011 Mar;59(3):383-9.

26. Graafmans WC, Ooms ME, Hofstee HMA, et al. Falls in the elderly: A prospective study of risk factors and risk profiles. American Journal of Epidemiology. 1996;143(11):1129-36.


28. Allan LM, Ballard CG, Rowan EN, Kenny RA. Incidence and prediction of falls in dementia: a prospective study in older people. PLoS One. 2009;4(5):e5521.


30. Semel J, Gray Jm Fau - Ahn HJ, Ahn Hj Fau - Nasr H, Nasr H Fau - Chen JJ, Chen JJ. Predictors of outcome following hip fracture rehabilitation. 20100927 DCOM- 20101221(1934-1482 (Print)).

31. Lieberman D, Friger M Fau - Lieberman D, Lieberman D. Rehabilitation outcome following hip fracture surgery in elderly diabetics: a prospective cohort study of 224 patients. 20070316 DCOM- 20070412(0963-8288 (Print)).


33. Fried LP, Tangen Cm Fau - Walston J, Walston J Fau - Newman AB, et al. Frailty in older adults: evidence for a phenotype. 20010316 DCOM- 20010329(1079-5006 (Print)).

34. O’Connell MD, Savva GM, Fan CW, Kenny RA. Orthostatic hypotension, orthostatic intolerance and frailty: The Irish Longitudinal Study on Aging-TILDA. 20150413 DCOM- 20151117(1872-6976 (Electronic)).

35. Rockwood K, Abeysundera Mj Fau - Mitnitski A, Mitnitski A. How should we grade frailty in nursing home patients? 20071113 DCOM- 20071213(1538-9375 (Electronic)).

36. Lach HW, Parsons JL. Impact of fear of falling in long term care: an integrative review. 20130805 DCOM- 20131213(1538-9375 (Electronic)).

R1

R2

R3

R4

R5

R6

R7

R8

R9

R10

R11

R12

R13

R14

R15

R16

R17

R18

R19

R20

R21

R22

R23

R24

R25

R26

R27

R28

R29

R30

R31

R32

R33

R34

R35

R36

R37

R38

R39

Chapter 11

180

37. Visschedijk JH, Caljouw MA, Bakkers E, van Balen R, Achterberg WP. Longitudinal follow-up study on fear of falling during and after rehabilitation in skilled nursing facilities. 20151205(1471-2318 (Electronic)).

38. Oude Voshaar RC, Banerjee S Fau - Horan M, Horan M Fau - Baldwin R, et al. Fear of falling more important than pain and depression for functional recovery after surgery for hip fracture in older people. 20061016 DCOM- 20070206(0033-2917 (Print)).

39. Di Monaco M, Castiglioni C Fau - De Toma E, De Toma E Fau - Gardin L, et al. Handgrip strength is an independent predictor of functional outcome in hip-fracture women: a prospective study with 6-month follow-up. 20150214 DCOM- 20150423(1536-5964 (Electronic)).

40. Shyam Kumar AJ, Beresford-Cleary N Fau - Kumar P, Kumar P Fau - Barai A, et al. Preoperative grip strength measurement and duration of hospital stay in patients undergoing total hip and knee arthroplasty. 20130619 DCOM- 20140313(1633-8065 (Print)).

41. Webb AR, Newman La Fau - Taylor M, Taylor M Fau - Keogh JB, Keogh JB. Hand grip dynamometry as a predictor of postoperative complications reappraisal using age standardized grip strengths. 19890502 DCOM- 19890502(0148-6071 (Print)).

42. Dodds RM, Syddall HE, Cooper R, et al. Grip strength across the life course: normative data from twelve British studies. PloS one. 2014;9(12):e113637.

43. Alley DE, Shardell MD, Peters KW, et al. Grip strength cutpoints for the identification of clinically relevant weakness. The journals of gerontology Series A, Biological sciences and medical sciences. 2014 May;69(5):559-66.

44. Visschedijk J, Achterberg W Fau - Van Balen R, Van Balen R Fau - Hertogh C, Hertogh C. Fear of falling after hip fracture: a systematic review of measurement instruments, prevalence, interventions, and related factors. 20100924 DCOM- 20101012(1532-5415 (Electronic)).

45. Myers AM, Powell LE, Maki BE, et al. Psychological indicators of balance confidence: relationship to actual and perceived abilities. The journals of gerontology Series A, Biological sciences and medical sciences. 1996 Jan;51(1):M37-43.

46. Tinetti ME, Powell L. Fear of falling and low self-efficacy: a case of dependence in elderly persons. Journal of gerontology. 1993 Sep;48 Spec No:35-8.

47. Hartog LC, Hendriks SH, Cimzar-Sweelssen M, et al. Orthostatic changes in blood pressure and mortality in a nursing home population. Journal of hypertension. 2016 Mar 30.









R1

R2

R3

R4

R5

R6

R7

R8

R9

R10

R11

R12

R13

R14

R15

R16

R17

R18

R19

R20

R21

R22

R23

R24

R25

R26

R27

R28

R29

R30

R31

R32

R33

R34

R35

R36

R37

R38

R39

Discussion

181

11

56. Punchick B, Freud T, Press Y. The association between orthostatic hypotension and cognitive state among adults 65 years and older who underwent a comprehensive geriatric assessment. Medicine. 2016 Jul;95(29):e4264.

57. Frewen J Fau - Savva GM, Savva Gm Fau - Boyle G, Boyle G Fau - Finucane C, Finucane C Fau - Kenny RA, Kenny RA. Cognitive performance in orthostatic hypotension: findings from a nationally representative sample. 20140829 DCOM- 20140912(1532-5415 (Electronic)).

58. Conway KS, Forbang N, Beben T, et al. Relationship Between 24-Hour Ambulatory Blood Pressure and Cognitive Function in Community-Living Older Adults: The UCSD Ambulatory Blood Pressure Study. American journal of hypertension. 2015 Apr 19.

59. Butt DA, Harvey PJ. Benefits and risks of antihypertensive medications in the elderly. Journal of internal medicine. 2015 Dec;278(6):599-626.

60. Post Hospers G, Smulders YM, Maier AB, Deeg DJ, Muller M. Relation between blood pressure and mortality risk in an older population: role of chronological and biological age. Journal of internal medicine. 2015 Apr;277(4):488-97.


62. Odden MC, Peralta CA, Haan MN, Covinsky KE. Rethinking the association of high blood pressure with mortality in elderly adults: the impact of frailty. Archives of internal medicine. 2012 Aug 13;172(15):1162-8.





67. Williamson JD, Supiano MA, Applegate WB, et al. Intensive vs Standard Blood Pressure Control and Cardiovascular Disease Outcomes in Adults Aged >/=75 Years: A Randomized Clinical Trial. Jama. 2016 May 19.

68. Data published by Statistics Netherlands. In Dutch: “‘Centraal Bureau voor de Statistiek’”. . [updated 2016, March 31; cited 2016]; Available from: http://statline.cbs.nl/StatWeb/.

69. E. A. United States Life Tables, 2010. National Vital Statistics Reports. 2014 November 6;63(7).

70. update). TN-GCrms. Huisarts Wet. 2012;55:14-28.

71. James PA, Oparil S, Carter BL, et al. 2014 evidence-based guideline for the management of high blood pressure in adults: report from the panel members appointed to the Eighth Joint National Committee (JNC 8). 20140205 DCOM- 20140320(1538-3598 (Electronic)).

72. Fan VS, Au DH, McDonell MB, Fihn SD. Intraindividual change in SF-36 in ambulatory clinic primary care patients predicted mortality and hospitalizations. J Clin Epidemiol. 2004 Mar;57(3):277-83.

73. van der Zee KI SR. Het meten van de algemene gezondheidstoestand met de Rand-36, een handleiding. Tweede herziene druk. . UMCG / Rijksuniversiteit Groningen, Research Institute SHARE.; 2012 [December 2014]; Available from: http://www.rug.nl/share.

74. Landman GW, van Hateren KJ, Kleefstra N, et al. Health-related quality of life and mortality in a general and elderly population of patients with type 2 diabetes (ZODIAC-18). Diabetes care. 2010 Nov;33(11):2378-82.

R1

R2

R3

R4

R5

R6

R7

R8

R9

R10

R11

R12

R13

R14

R15

R16

R17

R18

R19

R20

R21

R22

R23

R24

R25

R26

R27

R28

R29

R30

R31

R32

R33

R34

R35

R36

R37

R38

R39

Chapter 11

182

75. Dominick KL, Ahern FM, Gold CH, Heller DA. Relationship of health-related quality of life to health care utilization and mortality among older adults. Aging clinical and experimental research. 2002 Dec;14(6):499-508.

76. Tibblin G, Svardsudd K, Welin L, Erikson H, Larsson B. Quality of life as an outcome variable and a risk factor for total mortality and cardiovascular disease: a study of men born in 1913. Journal of hypertension Supplement : official journal of the International Society of Hypertension. 1993 Jun;11(4):S81-6.

77. Naylor MD, Hirschman KB, Hanlon AL, et al. Factors Associated With Changes in Perceived Quality of Life Among Elderly Recipients of Long-Term Services and Supports. Journal of the American Medical Directors Association. 2016 Jan;17(1):44-52.

78. Eisele M, Kaduszkiewicz H, Konig HH, et al. Determinants of health-related quality of life in older primary care patients: results of the longitudinal observational AgeCoDe Study. The British journal of general practice : the journal of the Royal College of General Practitioners. 2015 Nov;65(640):e716-23.


80. Andresen EM, Gravitt GW, Aydelotte ME, Podgorski CA. Limitations of the SF-36 in a sample of nursing home residents. Age and ageing. 1999 Oct;28(6):562-6.

CHAPTER 12Summary

R1

R2

R3

R4

R5

R6

R7

R8

R9

R10

R11

R12

R13

R14

R15

R16

R17

R18

R19

R20

R21

R22

R23

R24

R25

R26

R27

R28

R29

R30

R31

R32

R33

R34

R35

R36

R37

R38

R39

Chapter 12

184

R1

R2

R3

R4

R5

R6

R7

R8

R9

R10

R11

R12

R13

R14

R15

R16

R17

R18

R19

R20

R21

R22

R23

R24

R25

R26

R27

R28

R29

R30

R31

R32

R33

R34

R35

R36

R37

R38

R39

Summary

185

12

Orthostatic hypotension (OH) occurs frequently in the elderly population, and its prevalence increases with advancing age [1, 2]. International consensus defines OH as a decrease in systolic blood pressure (SBP) by at least 20 mmHg or a decrease in diastolic blood pressure (DBP) by at least 10 mmHg within 3 minutes after changing from supine to a standing position [1]. There is limited information with regard to differences between the various methods to accurately determine OH, about the consequences whenever OH is established, and about the therapy regarding OH in the elderly population. It is advised to perform a continuous blood pressure (BP) measurement, while automated sphygmomanometers with intermittent measurements are commonly used for this purpose in daily practice. It is generally assumed that OH is causally related to falling, cardiovascular complications and mortality, despite the sparse evidence. The lack of evidence regarding OH underlines the need for gaining more knowledge about the implications of OH. OH is partially caused by antihypertensive treatment. However, information regarding the relation between BP and mortality risk is lacking in a nursing home population, which adds to the discussion that the need for strict antihypertensive treatment in this population is questionable. Finally, the role of health related quality of life (HRQOL) in this specific population is important and perhaps even more important than numerous clinical parameters. However in what way HRQOL is related to cardiovascular morbidity and mortality is unknown.

The aims of this thesis were to study the OH measurements itself, to study the clinical implications of OH in elderly patients, and to study factors (including OH) that might be related to mortality and successful rehabilitation in nursing home residents. A large part of this thesis is based on the results of a prospective study performed in a nursing home. Data of this study were used to investigate the relationship between OH and different endpoints like falling, successful rehabilitation, and mortality. Also, health related quality of life (HRQOL) was investigated in this population. The other chapters of this thesis were based on performed cross-sectional cohort study amongst community-dwelling elderly visiting the outpatient clinic of internal medicine, and a prospective cohort study amongst elderly hospitalized with a hip fracture. Various methods to measure OH were studied and the relationship between OH, muscle strength, and successful rehabilitation were described in hospitalized elderly. The results of the studies as described in the different chapters of this thesis will be summarized below.

R1

R2

R3

R4

R5

R6

R7

R8

R9

R10

R11

R12

R13

R14

R15

R16

R17

R18

R19

R20

R21

R22

R23

R24

R25

R26

R27

R28

R29

R30

R31

R32

R33

R34

R35

R36

R37

R38

R39

Chapter 12

186

In chapters 2 and 3 various methods to measure OH are described. We compared the postural change to standing with the postural change to the sitting position and continuous versus interval blood pressure measurement. A difference was seen in blood pressure response between standing and sitting postural change. Although no significant difference in the prevalence of OH was observed, there was at most moderate agreement between the methods in diagnosing or refuting OH on an individual level. In addition, no difference in the prevalence of OH was observed between OH measurements using the continuous or the interval BP measurement device. However, agreement between methods was low.

In chapters 4 and 5 we described the relation between OH and falling. In chapter 4, no relationship between presence of OH and falling or recurrent falling was found within a nursing home population. The meta-analysis, as described in chapter 5, showed a significant relationship between orthostatic hypotension and time to first fall incident. By performing an individual patient data meta-analysis with prospective studies it can be concluded that OH is associated with the time to a first fall incident. However, the lower boundary and the width of the 95% confidence interval do not exclude the possibility there is no relevant association. Therefore, more prospective studies are needed for a precise estimate of the relationship between OH and falling.

In chapters 6 and 7 we focused on the relationship between OH and rehabilitation in elderly patients. In nursing home residents, patients with OH were found to have a higher chance of successful rehabilitation compared to patients without OH [3]. However, in hospitalized patients with a hip fracture, OH measured during the first days of hospitalization did not predict successful rehabilitation.

The relationship between blood pressure, OH, and mortality in a nursing home population was described in chapter 8. OH was related to increased all-cause mortality, but only in patients at the psychogeriatric department. Besides, a significant relationship was only observed between DBP and all-cause mortality. Although the study showed that DBP and OH were both an independent risk factor for mortality, its practical implications seem to be small. Adding DBP to other important risk factors for mortality, DBP hardly improved risk prediction.

In chapter 9 the discussion about the standard consensus definition of OH is described. We explored different definitions of OH in relation to the endpoints orthostatic complaints and falling. It was shown that the consensus definitions of OH are not related to orthostatic complaints or previous fall incidents in community-dwelling participants aged 65 years and older. Out of all studied definitions, only a marked >25% decrease in SBP was related to complaints, whereas only a >25% decrease in DBP was related to previous fall incidents.

R1

R2

R3

R4

R5

R6

R7

R8

R9

R10

R11

R12

R13

R14

R15

R16

R17

R18

R19

R20

R21

R22

R23

R24

R25

R26

R27

R28

R29

R30

R31

R32

R33

R34

R35

R36

R37

R38

R39

Summary

187

12

In chapter 10, the relationship between HRQOL, rehabilitation and mortality in a nursing home population was described. HRQOL, as assessed with the RAND-36 questionnaire, was significantly associated with mortality in three dimensions. However the additional values of these dimensions in mortality prediction were found to be very limited.

Based on the results of this thesis one may conclude that the approach of OH in elderly patients needs to be changed. OH measurements according to the current International consensus definition are not related to important clinical endpoints and therefore not useful in daily practice. Furthermore, the mortality prediction capabilities of BP and HRQOL are very limited in a nursing home population.

R1

R2

R3

R4

R5

R6

R7

R8

R9

R10

R11

R12

R13

R14

R15

R16

R17

R18

R19

R20

R21

R22

R23

R24

R25

R26

R27

R28

R29

R30

R31

R32

R33

R34

R35

R36

R37

R38

R39

Chapter 12

188

REFERENCES


2. Veronese N, De Rui M, Bolzetta F, et al. Orthostatic Changes in Blood Pressure and Mortality in the Elderly: The Pro.V.A Study. Am J Hypertens. 2015 Oct;28(10):1248-56.

3. Hartog LC, Cizmar-Sweelssen M, Knipscheer A, et al. The association between orthostatic hypotension, falling and successful rehabilitation in a nursing home population. Arch Gerontol Geriatr. 2015 Sep-Oct;61(2):190-6.

CHAPTER 13Nederlandse Samenvatting

(Summary in Dutch)

R1

R2

R3

R4

R5

R6

R7

R8

R9

R10

R11

R12

R13

R14

R15

R16

R17

R18

R19

R20

R21

R22

R23

R24

R25

R26

R27

R28

R29

R30

R31

R32

R33

R34

R35

R36

R37

R38

R39

Chapter 13

190

R1

R2

R3

R4

R5

R6

R7

R8

R9

R10

R11

R12

R13

R14

R15

R16

R17

R18

R19

R20

R21

R22

R23

R24

R25

R26

R27

R28

R29

R30

R31

R32

R33

R34

R35

R36

R37

R38

R39

ederlandse sa envatting u ar in utch

191

13

ORTHOSTATISCHE HYPOTENSIE

Orthostatische hypotensie (OH) wordt gedefinieerd als een bloeddrukdaling van minstens 20 mmHg systolisch of 10 mmHg diastolisch binnen 3 minuten na een positieverandering van liggend naar staand [1]. OH wordt veelvuldig gezien bij ouderen, en de prevalentie ervan neemt toe met het ouder worden [1]. De prevalentie bij verpleeghuispatiënten varieert van 18 tot 50% [2-4]. De etiologie van OH is multifactorieel en kan veroorzaakt worden door een afname van de baroreceptorgevoeligheid, falen van het autonome zenuwstelsel, hypertensie, het gebruik van verschillende soorten medicatie (bijvoorbeeld antihypertensiva), hypovolemie resulterend in onvoldoende ventriculaire vulling, en bedrust [5-9]. Met het ouder worden neemt de baroreceptorgevoeligheid af en neemt de prevalentie van multimorbiditeit en hoeveelheid medicatie toe. Dit alles kan een hogere prevalentie van OH tot gevolg hebben [8, 10]. OH kan vergezeld worden door orthostatische klachten zoals een licht gevoel in het hoofd, duizeligheid en flauwvallen [1, 10, 11].

DE ORTHOSTASE METING

De Internationale richtlijn geeft aan dat OH gemeten moet worden met een continu bloeddrukmeter, met een positieverandering van liggend naar staand [1]. Echter, het gebruik van deze richtlijn is in de klinische praktijk niet altijd haalbaar. Aangezien in de eerste lijn een continu bloeddrukmeter meestal niet aanwezig is, wordt in de dagelijkse praktijk de orthostatische bloeddruk bijna altijd met een automatische bloeddrukmeter (niet continu) gemeten. Daarnaast blijkt dat veel oudere patiënten geen staande bloeddrukmeting kunnen ondergaan in verband met mobiliteitsproblemen, wat er dan toe leidt dat de meting in staande houding wordt vervangen door een meting in zittende houding. Zowel het type bloeddrukmeter als het soort houdingsverandering kunnen van invloed zijn op de uitkomst van de meting.

KLINISCHE CONSEQUENTIES VAN ORTHOSTATISCHE HYPOTENSIE

In verschillende onderzoeken wordt gesuggereerd dat de aanwezigheid van OH geassocieerd is met een verhoogde kans op vallen, het vaker manifest worden van cardiovasculaire ziekten en zelfs met voortijdige sterfte bij ouderen [9, 12-23]. Deze relatie tussen OH en de verschillende klinische eindpunten is echter twijfelachtig. In veel van deze studies wordt namelijk de relatie tussen OH en deze eindpunten beschreven zonder te corrigeren voor

R1

R2

R3

R4

R5

R6

R7

R8

R9

R10

R11

R12

R13

R14

R15

R16

R17

R18

R19

R20

R21

R22

R23

R24

R25

R26

R27

R28

R29

R30

R31

R32

R33

R34

R35

R36

R37

R38

R39

Chapter 13

192

belangrijke risicofactoren, terwijl juist deze relatie beïnvloed kan worden door een aantal van deze risicofactoren. Naast de onduidelijke invloed van OH op vallen en sterfte, zou OH ook invloed kunnen hebben op de snelheid van revalideren of de ligduur in het ziekenhuis dan wel op de revalidatieafdeling.

HYPERTENSIE BIJ OUDEREN

Zowel hypertensie als het gebruik van antihypertensiva worden veelal genoemd als oorzaken van OH bij ouderen [1, 10, 24, 25]. Op dit moment wordt er wereldwijd gediscussieerd over het nut en de noodzaak van strikte behandeling van hypertensie bij ouderen; het is onduidelijk of intensieve behandeling van hypertensie zinvol is bij alle ouderen. De vraag is allereerst of een verhoogde bloeddruk wel een voorspellende factor is voor voortijdige sterfte bij kwetsbare ouderen.

KWALITEIT VAN LEVEN

Door de hoge prevalentie van multimorbiditeit bij ouderen is het belangrijk dat er, naast eventuele medische behandeling, aandacht is voor gezondheidsbeleving en kwaliteit van leven bij deze groep patiënten. Kwaliteit van leven gerelateerd aan ziekten en beperkingen wordt ook wel ‘health related quality of life’ (HRQOL) genoemd. Multimorbiditeit, depressie en verminderde cognitie zijn frequent aanwezig bij ouderen en hebben grote invloed op HRQOL [26-31]. Een consequentie is dan ook dat HRQOL vaak erg laag is bij verpleeghuispatiënten [27-29, 32-34]. Kennis over HRQOL bij een verpleeghuispatiënt kan helpen een medisch beleid in te stellen, aan te passen of juist te stoppen. Gezien het feit dat het sterfterisico binnen verpleeghuispatiënten al zo hoog is, kan HRQOL mogelijk wel de belangrijkste onderbouwing zijn voor het instellen dan wel afzien van een medische (be)handeling.

De doelen van dit proefschrift waren om onderzoek te doen naar de OH meting zelf, naar de klinische relevantie van OH bij de oudere patiënt en naar verschillende factoren (waaronder OH) welke van invloed zijn op revalidatie en sterfte bij verpleeghuispatiënten. Een belangrijk deel van het proefschrift is gebaseerd op de uitkomsten van een prospectieve studie die is uitgevoerd in een verpleeghuis. Gebruikmakend van de resultaten van deze studie werd de relatie tussen OH en vallen, succesvolle revalidatie en mortaliteit onderzocht. Daarnaast werd HRQOL in relatie tot mortaliteit in het verpleeghuis onderzocht. De overige hoofdstukken van dit proefschrift zijn gebaseerd op een prospectieve cross-sectionele

R1

R2

R3

R4

R5

R6

R7

R8

R9

R10

R11

R12

R13

R14

R15

R16

R17

R18

R19

R20

R21

R22

R23

R24

R25

R26

R27

R28

R29

R30

R31

R32

R33

R34

R35

R36

R37

R38

R39


193

13

studie uitgevoerd bij patiënten die de polikliniek interne geneeskunde bezochten, een prospectieve cohort studie uitgevoerd bij ouderen, opgenomen in het ziekenhuis met een heupfractuur, en tot slot een cross-sectionele studie uitgevoerd bij ouderen welke deelnamen aan een mobiel valpreventieprogramma. Met behulp van deze studies werden verschillende methoden om OH te meten onderzocht, werd de relatie tussen OH, spierkracht en revalidatie onderzocht, en werd de relatie tussen verschillende definities van OH en vallen of orthostatische klachten onderzocht.

RESULTATEN

In de hoofdstukken 2 en 3 worden verschillende methoden om OH te meten beschreven. De verschillen in hemodynamische veranderingen en de prevalentie van OH tussen twee verschillende positieveranderingen (staand versus zittend), en het gebruik van twee verschillende metingen (continu versus interval) zijn onderzocht. Hoewel er geen verschil in de prevalentie van OH wordt gezien tussen de staande en zittende metingen, blijkt dat de overlap van patiënten met (en zonder) OH bij het vergelijken van de resultaten van metingen bij beide positieveranderingen gering is. Ook bij het vergelijken van de continue bloeddrukmeting met de interval meter worden andere patiënten met OH gediagnostiseerd, ondanks dat er geen verschil in prevalentie wordt gezien.

In de hoofdstukken 4 en 5 wordt de relatie tussen OH en vallen beschreven. In een prospectieve studie uitgevoerd in het verpleeghuis is geen relatie waargenomen tussen OH en vallen. Echter, de meta-analyse in hoofdstuk 5 beschrijft wel een relatie tussen OH en tijd tot het eerste valincident. Patiënten met OH blijken daarbij sneller een eerste valincident door te maken dan patiënten zonder OH al is het onduidelijk of dat verband wel relevant is.

De relatie tussen OH en revalidatie bij oudere patiënten is beschreven in de hoofdstukken 6 en 7. Bij patiënten op een revalidatieafdeling in het verpleeghuis is het bestaan van OH gerelateerd aan een snellere revalidatie (hoofdstuk 6). Echter, bij patiënten met een heupfactuur blijkt dat het bestaan van OH, gemeten in de eerste dagen van de ziekenhuisopname, geen invloed heeft op de snelheid van revalidatie (hoofdstuk 7).

In hoofdstuk 8 is de relatie tussen bloeddruk en OH met sterfte in een verpleeghuispopulatie beschreven. Er blijkt een relatie tussen OH en sterfte te zijn, maar alleen bij patiënten op de afdeling psychogeriatrie: het bestaan van OH was hier gerelateerd aan een hogere sterftekans. Tevens is in de totale verpleeghuispopulatie een hogere diastolische bloeddruk gerelateerd aan een hogere sterfte. Echter, de additionele voorspellende waarde van zowel

R1

R2

R3

R4

R5

R6

R7

R8

R9

R10

R11

R12

R13

R14

R15

R16

R17

R18

R19

R20

R21

R22

R23

R24

R25

R26

R27

R28

R29

R30

R31

R32

R33

R34

R35

R36

R37

R38

R39

Chapter 13

194

OH als bloeddruk bovenop de overige risicofactoren is zeer gering. In hoofdstuk 11 wordt aan de hand van de studie in hoofdstuk 8 gediscussieerd over de relevantie van het behandelen van hypertensie bij ouderen in het verpleeghuis. Gezien het feit dat de gerandomiseerde studies naar hypertensie bij ouderen, zoals de HYVET en de SPRINT studies, alleen vitale ouderen includeren, zijn deze in ieder geval niet te vertalen naar ouderen in het verpleeghuis.

In hoofdstuk 9 wordt de huidige definitie van OH bediscussieerd. Er blijkt geen relatie te bestaan tussen OH gedefinieerd volgens de consensus definitie uit 1996 dan wel 2011 en vallen of orthostatische klachten. Alleen een afname van >25% in systolische bloeddruk blijkt gerelateerd te zijn aan orthostatische klachten, terwijl een afname van > 25% in diastolische bloeddruk gerelateerd blijkt te zijn aan valincidenten.

Tot slot wordt in hoofdstuk 10 de relatie tussen HRQOL, revalidatie en sterfte in het verpleeghuis beschreven. Ondanks dat HRQOL, gemeten met behulp van de RAND-36 vragenlijst, op verschillende subschalen gerelateerd blijkt te zijn aan sterfte, blijkt het geen additionele waarde te hebben in het voorspellen van sterfte in het verpleeghuis.

CONCLUSIES

Op basis van de studies in dit proefschrift kan geconcludeerd worden dat:1. wanneer afgeweken wordt van de Internationale richtlijn bij het meten van

orthostatische hypotensie bij oudere patiënten, andere patiënten gediagnosticeerd worden met orthostatische hypotensie dan wanneer de richtlijn gevolgd zou zijn.

2. de klinische relevantie van orthostatische hypotensie, gedefinieerd volgens de Internationale richtlijn, beperkt is bij oudere (verpleeghuis)patiënten.

3. bloeddruk en kwaliteit van leven een geringe tot geen voorspellende waarde hebben in het voorspellen van sterfte in het verpleeghuis.

R1

R2

R3

R4

R5

R6

R7

R8

R9

R10

R11

R12

R13

R14

R15

R16

R17

R18

R19

R20

R21

R22

R23

R24

R25

R26

R27

R28

R29

R30

R31

R32

R33

R34

R35

R36

R37

R38

R39


195

13

AANBEVELINGEN

De belangrijkste aanbevelingen van dit proefschrift zijn:1. De benadering van orthostatische hypotensie bij oudere patiënten moet aangepast

worden.a. Als bij mensen in een verpleeghuis op basis van bloeddrukmetingen

orthostatische hypotensie wordt vastgesteld, dan lijkt dat niet gerelateerd aan klinische consequenties. Het is daarom de vraag of het nuttig is om dit soort metingen te doen.

b. In plaats daarvan lijkt het verstandig om orthostase klachten uit te vragen en op basis van die klachten actie te ondernemen.

c. In het uitzonderlijke geval dat een orthostatische bloeddrukmeting wordt uitgevoerd, zal deze moeten plaatsvinden volgens de huidige Internationale richtlijn.

2. Er zijn meer studies nodig om de klinische relevantie van de huidige definitie van orthostatische hypotensie te onderzoeken.

3. De invloed van orthostatische hypotensie op vallen moet bevestigd worden door middel van meer prospectieve studies.

4. De hypertensie behandeling bij kwetsbare ouderen en verpleeghuispatiënten moet geïndividualiseerd worden, rekening houdend met klachten, levensverwachting en (kans op) bijwerkingen van de gebruikte medicatie.

5. Onderzoek naar het gebruik van HRQOL binnen het verpleeghuis, als één van de maten waarop beslissingen worden gefundeerd voor behandeling, is noodzakelijk.

R1

R2

R3

R4

R5

R6

R7

R8

R9

R10

R11

R12

R13

R14

R15

R16

R17

R18

R19

R20

R21

R22

R23

R24

R25

R26

R27

R28

R29

R30

R31

R32

R33

R34

R35

R36

R37

R38

R39

Chapter 13

196

REFERENTIES




















R1

R2

R3

R4

R5

R6

R7

R8

R9

R10

R11

R12

R13

R14

R15

R16

R17

R18

R19

R20

R21

R22

R23

R24

R25

R26

R27

R28

R29

R30

R31

R32

R33

R34

R35

R36

R37

R38

R39


197

13





24. Fedorowski A, Burri P Fau - Melander O, Melander O. Orthostatic hypotension in genetically related hypertensive and normotensive individuals. 20090428 DCOM- 20090723(1473-5598 (Electronic)).

25. Moonen JE, Foster-Dingley JC, de Ruijter W, et al. Effect of discontinuation of antihypertensive medication on orthostatic hypotension in older persons with mild cognitive impairment: the DANTE Study Leiden. Age and ageing. 2016 Jan 11.










Dankwoord

Curriculum Vitae

List of publications

Previous dissertations

R1

R2

R3

R4

R5

R6

R7

R8

R9

R10

R11

R12

R13

R14

R15

R16

R17

R18

R19

R20

R21

R22

R23

R24

R25

R26

R27

R28

R29

R30

R31

R32

R33

R34

R35

R36

R37

R38

R39

Dankwoord

200

R1

R2

R3

R4

R5

R6

R7

R8

R9

R10

R11

R12

R13

R14

R15

R16

R17

R18

R19

R20

R21

R22

R23

R24

R25

R26

R27

R28

R29

R30

R31

R32

R33

R34

R35

R36

R37

R38

R39

Dankwoord

201

DANKWOORD

“Look at situations from all angles, and you will become more open.” Dalai Lama

Mijn dank gaat uit naar alle personen die hebben bijgedragen aan dit proefschrift. Of wel in de woorden van de Dalai Lama te spreken; iedereen die mij heeft geholpen vanuit een andere invalshoek naar dit proefschrift te kijken. Door de input van al deze mensen is dit proefschrift geworden tot wat dit nu is. Op deze plaats wil ik graag een aantal mensen in het bijzonder bedanken voor hun bijdrage.

Allereerst gaat mijn dank uit naar alle patiënten uit verpleeghuis TriviumMeulenBelt zorg en de Isala die hun medewerking hebben verleend aan de studies die beschreven staan in dit proefschrift. Het onderzoek bij al deze patiënten had niet plaats kunnen vinden zonder alle medewerkers die belangeloos een bijdrage hebben geleverd bij metingen, werven van patiënten en ondersteunende taken. Bedankt daarvoor!

Hooggeleerde prof. Dr. Bilo. Beste Henk, als promotor heb je een hele belangrijke taak tijdens de totstandkoming van dit proefschrift. Jij hebt de gave om een heel betrokken begeleider te zijn, geïnteresseerd in de mens die voor je zit en toch heel kritisch. Ik heb er bewondering voor waar jij je energie vandaan haalt om een marathon te rennen, bergen te beklimmen, af te reizen naar verre oorden en poli te draaien terwijl je net een nachtvlucht hebt gehad. Een nieuw tijdperk breekt voor je aan, ik hoop dat je hier met net zoveel energie van gaat genieten. Ik vond het een voorrecht dat ik onder jouw leiding dit proefschrift heb mogen schrijven.

Weledelzeergeleerde dr. Kleefstra. Beste Nanno, wat heb ik veel geleerd van jou. Ik was altijd een beetje huiverig als ik een manuscript van jou terug zag komen, hoe rood zou het zijn? Maar 2 dingen wist ik altijd zeker, de kwaliteit van het manuscript werd er beter van en…ik moest er weer heel wat uren voor reserveren. Toch ben ik je heel erg dankbaar voor je kritische blik, het heeft dit proefschrift naar een hoger niveau getild. Daarnaast waardeer ik je als persoon, je bent altijd eerlijk en geïnteresseerd. De koffiegesprekken had ik niet willen missen.

Weledelzeergeleerde dr. Van Hateren. Beste Hans, wat had ik zonder jou gemoeten. Je bent zo’n gedreven dokter en wetenschapper, daar kunnen nog velen wat van leren. Iemand enthousiasmeren, moed inpraten, maar ook kritisch zijn had ik nodig als we ontevreden waren over een manuscript. Daarnaast de felicitaties bij elk geaccepteerd manuscript

R1

R2

R3

R4

R5

R6

R7

R8

R9

R10

R11

R12

R13

R14

R15

R16

R17

R18

R19

R20

R21

R22

R23

R24

R25

R26

R27

R28

R29

R30

R31

R32

R33

R34

R35

R36

R37

R38

R39

Dankwoord

202

alsof het de eerste was. Door jou ben ik de wetenschap echt leuk gaan vinden en lijkt de combinatie van dokter en wetenschapper minder ver weg dan ik ooit gedacht had. Dank je wel.

Weledelzeergeleerde dr. Groenier. Beste Klaas, wat ben jij belangrijk geweest bij alle analyses van dit proefschrift. Jij wist altijd zo duidelijk te omschrijven waar het eigenlijk om ging en kon in een paar zinnen antwoord geven op de statistische vragen die ik had. Bedankt.

Prof. dr. van der Horst, prof. dr. Slaets en prof. dr. Zuidema wil ik bedanken voor hun bereidwilligheid om zitting te nemen in de leescommissie en het goedkeuren van dit proefschrift.

Beste mede-auteurs van de hoofdstukken van dit proefschrift. Ik wil jullie allemaal bedanken voor jullie kritische blik, feedback en prettige samenwerking.

Beste Mateja en Astrid, dankzij jullie hebben we een mooie database kunnen opbouwen in een verpleeghuispopulatie. Naast de patiëntenzorg hebben jullie met tijd en aandacht patiënten onderzocht, statussen uitgeplozen en vragenlijsten afgenomen. Door jullie kennis over deze populatie is dit proefschrift naar een hoger niveau getild. Mede door jullie ben ik enthousiast geworden voor de ouderengeneeskunde. Bedankt voor al jullie hulp en inspiratie.

Beste Marian, Sanneke en Jolanda, bedankt voor de samenwerking bij de STOK studie. Een jaar lang hebben we met elkaar patiënten geïncludeerd. Iedereen op pad met een eigen onderzoeksvraag, samen hebben we de klus geklaard.

Beste collega onderzoekers. Steven, Dennis, Ilse, Gijs, Angelien en Leonie, wat heb ik prettig en gezellig met jullie samengewerkt. Steven en Dennis, van jullie heb ik verreweg het meeste geleerd. Een korte uitleg van statistiek, of een goede zin in het Engels, altijd hadden jullie geduld. Ilse, wat was het gezellig met jou. Gelukkig was jij er niet te vaak want dan was dit proefschrift zeker nog niet af geweest. Gijs, bedankt voor je altijd kritische blik op de manuscripten. Angelien en Leonie, wat was het gezellig in Stockholm.

Beste Anna. In het kader van je wetenschappelijke stage deed je onderzoek bij ons naar orthostase metingen. Dit heeft geleidt tot een enerverende samenwerking en 2 prachtige hoofdstukken in dit proefschrift. Bedankt.

R1

R2

R3

R4

R5

R6

R7

R8

R9

R10

R11

R12

R13

R14

R15

R16

R17

R18

R19

R20

R21

R22

R23

R24

R25

R26

R27

R28

R29

R30

R31

R32

R33

R34

R35

R36

R37

R38

R39

Dankwoord

203

Familie en vrienden. Vele telefoongesprekken onderweg van Zwolle naar huis. Sommigen weten precies waar dit proefschrift over gaat, anderen hebben geen idee. Wel hebben jullie allen gemeen dat jullie belangrijk voor me zijn. Gesprekken over werk, gezin, kinderen, vakanties, liefde, dood, het publiceren van artikelen tot maandelijks uit eten onder het mom van de stad leren kennen. Allen hebben jullie een andere maar toch zeer belangrijke rol gehad in de totstandkoming van dit proefschrift. Bedankt!

Lieve oma, inmiddels op hoge leeftijd maar nog altijd geïnteresseerd in alles wat uw kinderen, kleinkinderen en achterkleinkinderen mee maken. U bent mijn inspiratie geweest tijdens het schrijven van dit proefschrift. Daarom draag ik dit proefschrift aan u op.

Lieve pap en mam, dankzij jullie ben ik geworden zoals ik nu ben. Jullie staan altijd voor Patrick en mij klaar; met liefde, adviezen en praktische hulp. Zonder dat alles had ik hier niet gestaan. Bedankt daarvoor!

Lieve Ewoud, wat een tijd hebben we achter de rug. Trouwen, verhuizen naar het hoge noorden en beiden een nieuwe baan. Ondanks al deze drukte nemen we de tijd voor elkaar en onze twee allerliefste kinderen. Ook al ben je het niet altijd eens met mijn keuzes, je hebt me altijd gesteund in alles wat ik doe. Ik hoop dat we in de toekomst samen blijven genieten, groeien en liefhebben.

Pap en Ewoud, mijn paranimfen. Wie kan ik beter als paranimfen hebben dan de 2 belangrijkste mannen in mijn leven. Samen maken we er een mooie dag van!

Feline en Thom, mijn allerliefste kinderen. Wat geniet ik elke dag van jullie. Door jullie is geen dag hetzelfde, hier wil ik dan ook niets van missen. Mama houdt van jullie tot de maan en weer terug.

R1

R2

R3

R4

R5

R6

R7

R8

R9

R10

R11

R12

R13

R14

R15

R16

R17

R18

R19

R20

R21

R22

R23

R24

R25

R26

R27

R28

R29

R30

R31

R32

R33

R34

R35

R36

R37

R38

R39

Curriculum Vitae

204

R1

R2

R3

R4

R5

R6

R7

R8

R9

R10

R11

R12

R13

R14

R15

R16

R17

R18

R19

R20

R21

R22

R23

R24

R25

R26

R27

R28

R29

R30

R31

R32

R33

R34

R35

R36

R37

R38

R39

Curriculum Vitae

205

CURRICULUM VITAE

Laura Caroline Hartog werd geboren op 22 december 1980 in Rotterdam. Zij behaalde haar VWO diploma aan het Ashram College te Alphen aan den Rijn, waarna zij geneeskunde heeft gestudeerd aan de Vrije Universiteit van Amsterdam. In 2005 studeerde zij Cum Laude af voor haar artsexamen. Ze startte als ANIOS (arts niet in opleiding tot specialist) en later AIOS (arts in opleiding tot specialist) op de afdeling heelkunde van het Kennemer Gasthuis te Haarlem en VU Medisch Centrum. Halverwege 2010 besloot ze om de opleiding tot chirurg vroegtijdig te beëindigen om haar carrière anders vorm te geven. Ze deed ervaring op als arts in de jeugdgezondheidszorg, als arts-docent Masterfase Geneeskunde en als beleidsmedewerker bij een thuiszorgorganisatie. In 2014 kreeg zij de gelegenheid om onder begeleiding van prof. dr. HJG. Bilo, dr. KJJ van Hateren en dr. N. Kleefstra te starten met een promotieonderzoek naar orthostatische hypotensie bij ouderen, wat geleid heeft tot dit proefschrift. Naast haar promotieonderzoek is zij sinds 2016 tevens werkzaam als ANIOS in het verpleeghuis in Leeuwarden, wat zij zal vervolgen na de verdediging van haar proefschrift. Laura woont samen met haar echtgenoot Ewoud Jutte en hun 2 kinderen (Feline en Thom) in Leeuwarden.

R1

R2

R3

R4

R5

R6

R7

R8

R9

R10

R11

R12

R13

R14

R15

R16

R17

R18

R19

R20

R21

R22

R23

R24

R25

R26

R27

R28

R29

R30

R31

R32

R33

R34

R35

R36

R37

R38

R39

ist o publications

206

R1

R2

R3

R4

R5

R6

R7

R8

R9

R10

R11

R12

R13

R14

R15

R16

R17

R18

R19

R20

R21

R22

R23

R24

R25

R26

R27

R28

R29

R30

R31

R32

R33

R34

R35

R36

R37

R38

R39

ist o publications

207

LIST OF PUBLICATIONS

Hartog LC, Cimzar-Sweelssen M, Knipscheer A, Groenier KH, Kleefstra N, Bilo HJG , Van Hateren KJJ. The association between Orthostatic hypotension, Falling and Successful Rehabilitation in a nursing home population. Archives of Gerontology and Geriatrics 2015; 61 (2):190-6.

Hartog LC, Hendriks SH, Cimzar-SweelssenM , Knipscheer A, Groenier KH, Kleefstra N, Bilo HJG, Van Hateren KJJ. Orthostatic changes in blood pressure and mortality in a nursing home population. Journal of hypertension 2016; 34 (6):1068-74

Schrijnders D, Hartog LC, Kleefstra N, Groenier KH, Landman GW, Bilo HJG. Within –Sulfonylurea-Class evaluation of time to intensification with insulin (ZODIAC-43). Plos One 2016; 11(16): e0157668.

Hartog LC, Landman GWD, Cimzar-Sweelssen M, Knipscheer A, Groenier KH, Kleefstra N, Bilo HJG, Van Hateren KJJ. Health-related quality of life, rehabilitation and mortality in a nursing home population. Neth J Med 2016 Jul; 74 (6):247-56

Hartog LC, Cimzar-Sweelssen M, Knipscheer A, Groenier KH, Kleefstra N, Bilo HJG, Van Hateren KJJ. Orthostatic hypotension does not predict recurrent falling in a nursing home population. Archives of Gerontology and Geriatrics 2016 Sep 4; 68: 39-43.

Hendriks SH, Hartog LC, Groenier KH, Maas AHEM, Van Hateren KJJ, Kleefstra N, Bilo HJG. Patient activation in type 2 diabetes; does it differ between men and women? Accepted in Journal of Diabetes Research 2016[Epub ahead of print]

Hartog LC, Winters AM, Roijen H, Kamper AM, Inia H, Kleefstra N, Bilo HJG, Van Hateren KJJ. The association between Orthostatic Hypotension and Handgrip Strength with Successful Rehabilitation in elderly hip fracture patients. Accepted for publication in Archives of physical medicine and rehabilitation.

Hartog LC, Kleefstra N, Luigies RH, De Rooij SE, Bilo HJG, Van Hateren KJJ. The clinical relevance of orthostatic hypotension in elderly patients. Accepted for publication in Geriatrics & Gerontology International.

R1

R2

R3

R4

R5

R6

R7

R8

R9

R10

R11

R12

R13

R14

R15

R16

R17

R18

R19

R20

R21

R22

R23

R24

R25

R26

R27

R28

R29

R30

R31

R32

R33

R34

R35

R36

R37

R38

R39

revious dissertations

208

R1

R2

R3

R4

R5

R6

R7

R8

R9

R10

R11

R12

R13

R14

R15

R16

R17

R18

R19

R20

R21

R22

R23

R24

R25

R26

R27

R28

R29

R30

R31

R32

R33

R34

R35

R36

R37

R38

R39


209

PREVIOUS DISSERTATIONS

This thesis is published within the Diabetes Centre of the Isala in Zwolle. Previous dissertations at our Diabetes Centre:

Van Dijk P.R. (2015) Continuous intraperitoneal insulin infusion in the treatment of type I diabetes mellitus. Glycaemia and beyond. Promotores: prof. dr. H.J.G. Bilo, prof. dr. R.O.B. Gans. Copromotores: dr. N. Kleefstra, dr. S.J.J. Logtenberg.

Drion I. (2014) Renal function estimation: the implications for clnical practice. Promotores: prof. dr. H.JG. Bilo, prof. dr. J.F.M. Wetzels. Copromotor: dr. N. Kleefstra.

Joosten J.M.H. (2014) Defining risk factors associated with renal and cognitive dysfunction. Promotores: prof. dr. H.J.G. Bilo, prof. dr. J.P.J. Slaets. Copromotores: dr. R.T. Gansevoort, dr G.J. Izaks.

Hortensius J. (2013) Self-monitoring of blood glucose in insulin-treated patients with diabetes. Promotores: prof. dr. H.J.G. Bilo, prof. dr. R.O.B. Gans. Copromotores: dr. J.J. van der Bijl, dr. N. Kleefstra.

Alkhalaf A. (2013) Novel approaches in Diabetic Nephropathy. Promotores: prof. dr. G.J. Navis, prof. dr. H.J.G. Bilo. Copromotoes: dr. S.J.L. Bakker, dr. N. Kleefstra.

Van Hateren K.J.J. (2013) Diabetes Care in old age. Promotores: prof. dr. H.J.G. Bilo, prof. dr. K. Van der Meer. Copromotores: dr. N. Kleefstra, dr. S.T. Houweling.

Gerrits E.G. (2013) Cardiovascular risk and its determinants in high risk patients. Promotores: prof. dr. H.J.G. Bilo, prof. dr. R.O.B. Gans. Copromotores: dr. A.J. Smit, dr. H.L. Lutgers.

Landman G.W.D. (2012) Mortality predictors in patients with type 2 diabetes. Promotores: prof. dr. H.J.G. Bilo, prof. dr. R.O.B. Gans. Copromotores: dr. N. Kleefstra, dr. K.H. Groenier.

Lenters-Westra W.B. (2011) Hemoglobin A1c: Standardisation, analytical performance and interpretation. Promotores: prof. dr. H.J.G. Bilo, prof. dr. R.O.B. Gans. Copromotores: dr. R.J. Slingerland.

Kleefstra N. (2010) Self-care interventions in type 2 diabetes. Promotores: prof. dr. H.J.G. Bilo, prof. dr. R.O.B. Gans. Copromotores: dr. S.T. Houweling, dr. K.H. Groenier.

R1

R2

R3

R4

R5

R6

R7

R8

R9

R10

R11

R12

R13

R14

R15

R16

R17

R18

R19

R20

R21

R22

R23

R24

R25

R26

R27

R28

R29

R30

R31

R32

R33

R34

R35

R36

R37

R38

R39


210

Logtenberg S.J.J. (2010) Intensive insulin therapy and glucose management. Studies with the implantable pump and a glucose sensor. Promotores: prof. dr. H.J.G. Bilo, prof. dr. R.O.B. Gans.

Lutgers H.L. (2008) Skin autofluorescence in diabetes mellitus. Promotores: prof. dr. R.O.B. Gans, prof. dr. H.J.G. Bilo. Copromotores: dr. A.J. Smit, dr. Ir. R. Graaff, dr. T.P. Links.

Van der Horst I.C. (2005) Metabolic interventions in acute myocardial infarction. Promotoreres: prof. dr. F. Zijlstra, prof. dr. R.O.B. Gans. Copromotor: dr. H.J.G. Bilo.

Houweling S.T. (2005) Taakdelegatie in de eerste- en tweedelijns diabeteszorg; Resultaten van de DISCOURSE studies. Promotor: prof. dr. B. Meyboom-de Jong. Copromotor: dr. H.J.G. Bilo.

Ubink-Veltmaat L.J. (2004) Type 2 diabetes mellitus in a Dutch region. Epidemiology and shared care. Promotor: prof. dr. B. Meyboom-de Jong. Copromotor: dr. H.J.G. Bilo.

Hart H.E. (2004) Health related quality of life in patients with diabetes mellitus type I. Prmotoreres: prof. dr. M. Berg, prof. dr. B. Meyboom-de Jong. Copromotor: H.J.G. Bilo.

De Visser C.L. (2003) Health and health risk on Urk. A study about cardiovascular disease and type 2 diabetes. Promotor: prof. dr. B. Meyboom-de Jong. Copromotor: H.J.G. Bilo.

Assink J.H. (1998) Oxidative stress and health status in patients with insulindependent diabetes mellitus. Promotor: prof. dr. D. Grobbee. Copromotor: H.J.G Bilo.

Goddijn P.P.M. (1997) Improving metabolic control in NIDDM patients referred for insulin therapy. Promotor: prof. dr. B. Meyboom-de Jong. Copromotor: dr. H.J.G. Bilo.


L.C. Hartog

Orthostati

c hypotension in elderly patients

L.C

. Ha

rto

g

UITNODIGING

Voor het bijwonen van

de openbare verdediging van

het proefschrift


L.C. Hartog

woensdag 8 februari 2017

om 12:45 uur

Academiegebouw

Broerstraat 5

Groningen

university of groningen orthostatic hypotension in elderly ... · author(s) and/or copyright...

Documents