experience with 24-hour ambulatory blood pressure monitoring in hypertension

Experience with 249hour ambulatory blood pressure monitoring in hypertension

Traditional sphygmomanometric blood pressure measurements may lead to errors in the diagnosis of arterial hypertension due to a number of factors, among which are the alerting reaction and pressor response induced in patients by the doctor’s visit. This phenomenon, as quantified in our laboratory by continuous intraarterial recordings, is responsible for an average rise In systolic and diastolic blood pressure of 27115 mm Hg, a rise that does not seem to be reduced by simply desensitizing the patient by means of more frequent physician visits. Twenty-four hour ambulatory blood pressure monitoring may theoretically improve the diagnostic approach to hypertensive patients by overcoming some limitations of isolated cuff measurements. In recording intraarterial blood pressure in 108 ambulant hypertensive subjects, we have found that 24hour blood pressure values are able to discriminate among patfents with different degrees of target organ damage better than isolated sphygmomanometric readings. Moreover, these studies have indicated that 24-hour blood pressure variability may be as important as blood pressure mean values in the assessment of cardiovascular complications. In clinical practice, however, intraarterial blood pressure monitoring is not feasible, and only noninvasive recorders can be used. Use of these devices does not induce any alertlng reaction in the patients and does not Interfere with day-night blood pressure changes. Although it is characterized by intermittent readings, this approach is not Incompatible with a precise estimate of 24-hour blood pressure mean values. However, the discontinuous measurements provided by the presently available noninvasive devices do not allow the precise assessment of blood pressure variability and these measurements are characterized by a considerable degree of inaccuracy when used in ambulant subjects. These limitations, combined with the lack of prospective demonstrations of the diagnostic superiority of ambulatory blood pressure monltoring over office blood pressure readings, still suggest caution in recommending a generalized use of this nevertheless promising approach to the clinical management of hypertension. (AM HEART J 1988; 118: 1134.)

Giuseppe Mancia, MD, and Gianfranco Parati, MD. Milan, Italy

This article deals with data we obtained on the following three issues: (1) the errors associated with cuff blood pressure measurement by a physician because of the alarm reaction of the patient, (2) the evidence that ambulatory blood pressure monitoring may be useful in the diagnosis of hypertension and the evaluation of antihypertensive treatment, and (3) the advantages and limitations of ambulatory blood pressure monitoring performed by noninvasive means compared with that performed intraarterially.

PRESSOR REACTION TO SPHYGMOMANOMETRIC BLOOD PRESSURE MEASUREMENT BY PHYSICIANS

In 88 essential hypertensive patients blood pressure was recorded intraarterially by the Oxford

From the Center for Clinical Physiology and Hypertension, University of Milan.

Reprint requests: Guiseppe Par&i, MD, Ospedale Maggiore di Milano, University di M&no, Via Francesco Sforza 35, 20122 Milan, Italy.

technique’ for 24 hours.2-4 During the recording the patients were visited by a physician who was asked to measure their blood pressure. In nearly all instances the physician’s visit induced a rise in the patient’s intraarterial blood pressure and heart rata. These rises began as soon as the physician entered the patient’s room, reached a peak within 4 minutes, and then declined after the end of the visit. On average the peak blood pressure rise was large (26.7 +- 2.3/14.9 + 1.6 mm Hg, means +- SE), but an even more striking feature was represented by the pronounced interindividual variability of this response. There was no correlation between the pressor response elicited by the physician’s visit and patients’ blood pressure, age or response to laboratory tests devised to assess reactivity to stress.2*“7 Prlrfhprmnrn _ -, ..___._ int:a-arterial blood pressure n,hmm

recording was prolonged to 48 hours to allow the same physician to visit the patient several times, no attenuation of the pressor reaction was observed. (Fig. l).” Thus cufI blood pressure evaluation by the

1134

votums 116

Numbsr 4 Twenty-four hour monitoring 1135

Peak BP and HR responses to repeated visits (n=16)

mmHg so-

b SBP m m Hg A DBP so-

.

60- . 60- *

. . .a*

40- . .

:. i ; 40- . .

.* -7 - dr -e-

. . . . . 20- “::”

.‘).. l . . l **-* a : .

. 0:. 20. $ - ..: *

z * . l .y .==’ f.

Od o- .

mmHg A MhP b/min b HR

~i/e i s + jf i f p

1st 2nd 3’* 4t”visit 1-t 2”“ 3rd 4thvisit

Fig. 1. Peak rises in intra-arterial systolic blood pressure @BP), diastolic blood pressure @BP), mean arterial pressure (MAP), and heart rate (HR) observed in 16 essential hypertensive subjects during four visits by the same physician aimed at evaluating subjects’ blood pressure. The visits were performed over 36 to 48 hours, during which time blood pressure was measured intraarterially by the Oxford method. Data are expressed as changes from a control value taken 4 minutes before the physician’s visit. (Reproduced with permission from Mancia G, Parati G, Pomidossi G, Grassi G, Casadei R, Zanchetti A. Alerting reaction and rise in blood pressure during measurement by physician and nurse. Hypertension 1987;9:209-15.)

n,30 mmlig 5 SBP max 5 DBP max b/min A HR inax

+50,

40 40 40 5

4 ; 30 30 30 L ; 0

o 20 20 20

I 2

: 10 10 10

F 9 6 0 0 0

- 10 I , -ioJ r I -10 J , 0octc.r Nurse Doctor ““MB OoCtOr llurse * p < 0.005

Fig. 2. Maximum rise in SBP, DBP, and HR occurring in 30 essential hypertensive subjects during a physician’s and a nurse’s visit aimed at evaluating subjects’ blood pressure. The visits were performed over 24 hours, during which time blood pressure was measured intraarterially by the Oxford method. In 15 subjects the nurse’s visit preceded the physician’s visit whereas in the remaining subjects the opposite occurred. Data are expressed as changes from a control value taken 4 minutes before each visit. Symbols are the same as those in Fig. 1. (Reproduced w&h permission from Mancia G, Parati G, Pomidossi G, Grassi A, Casadei R, Zanchetti A. Alerting reaction and rise in blood pressure during measurement by physician and nurse. Hypertension 1987;9:209-15.)

1136 Mancia and Parati October 1oBB

Amerlcen Heerl Journal

t-t=106

mmH9 P<O 01

160.

70-

%

0 loo- : E . :

g 76- CD L -

160-

0 p co.06 p .x 0.01

Group: lst 2nd ci’d 4”’ 1”’

0 24 h MAP c Group Average

m 24 h MAP z Group Average

1st 2nd 3rd 4’” 5’”

Fig. 3. Rate, and severity of target organ damage of hypertension observed in 108 essential hypertensive patients. The patients were subdivided into five groups according to the increasing value of their mean arterial pressure (MAP) as measured by the. cuff method. Within each group, the patients were further subdivided into two classes, according to whether 24-hour MAP was below or above the 24-hour average MAP of the group. Target organ damage was assessed by history and clinical and laboratory examinations and expressed as (1) the percentage of subjects exhibiting the damage (rate) and (2) the score obtained by assigning a predetermined value to the clinical data. Note that within each group the two classes had a similar cuff blood pressure but that the rate and the severity of target organ damage was less in the classes in which 24-hour MAP was lower. The p refers to the difference between all couples of classes. (Reproduced with permission from Parati G, Pomidossi G, Albini F, Malaspina D, Mancia G. Relationship of 24-hour blood pressure mean and variability to severity of target organ damage in hypertension. J Hypertens 1987;5:93-8.)

physician may largely overestimate patients’ blood pressure. This overestimation is extremely variable among patients and not easily predictable. It is also not easily extinguishable by increasing the patient’s familiarity with the procedure. All this represents a serious limitation in the diagnosis of hypertension.

However, the error inherent to cuff blood pressure evaluations can be reduced in two ways. In 30 patients the peak rise in blood pressure observed during blood pressure measurements by a physician was reduced by about 50% when blood pressure was assessed by a nurse (Fig. 2).* In addition, the blood

pressure rise observed at the tenth or fifteenth minute of the physician’s visit was about 30% of the initial blood pressure response. This emphasizes the importance of taking blood pressure measurements under conditions in which the alerting reaction of the patient is minimized. However, even under these conditions, an overestimation of blood pressure cannot be completely avoided. .I._... .-em.. -. aha.- -a--CI..-- CII..I-CI ~~DULM~UFIX b~uuir rtwaaute MURI t uFiiNG iN THE CLINICAL EVALUATION OF HYPERTENSION

Although justified on theoretical grounds, the use of ambulatory blood pressure monitoring for the

Volume 116

Number 4

mmlig

170,

Twenty-four hour monitoring 1137

n ,108

mmlig pco.02

5 12

.- 1

/O pco.05 8 100

;

g 3.0

E

B ,” 1

c B 75 !i

s $2.0 1

i; F 50

; D

2 .’

z ; 1.0.

a 25 s

;;; .z

c .’ 22

v) 0 O-

tiiroup: 3s’ 2nd 3rd 4”’ 5*h 1s’ 2nd 3rd 4th 5”’

0 Short t@rm MAP variability < Group Average

m Short term MAP variability z Group Average

Fig. 4. Rate and severity of target organ damage in the 108 patients of Fig. 3. The patients were divided into five groups according to the increasing value of their 24-hour MAP. Within each 24-hour MAP group, the patients were further subdivided into two classes according to whether their short-term blood pressure variability (see ref. 15), expressed as within half-hour standard deviation of MAP, was below or above the average short-term variability of the group. Note that within each group the two classes had a similar 24-hour MAP, but that the rate and the severity of target organ damage was less in the class in which blood pressure variability was lower. For other explanations, see Fig. 3. (Reproduced with permission from Parati G, Pomidossi G, Albini F, Malaspina D, Mancia G. Relationship of 24-hour blood pressure mean and variability to severity of target organ damage in hypertension. J Hypertens 1987;5:93-8.)

diagnosis of hypertension has not yet received unequivocal validation by prospective trials. Howev- er, a number of cross-sectional or retrospective studies have shown that the target organ damage of hypertension is more closely related to the blood pressure values obtained by this approach than to cuff blood pressure.g-13 This was found in 108 patients with untreated essential hypertension in whom blood pressure was measured by sphygmoma- nometry and recorded intraarterially for 24 hours in ambulatory conditions.14 The rate and severity of target organ damage, as assessed by a score based on clinical and laboratory examinations, showed a pro- gressive increase in five groups in which cuff blood pressure also progressively increased. However,

within each group patients with lower 24-hour blood pressure mean values had less target organ damage than patients with higher 24-hour blood pressure mean values (Fig. 3).

The same 108 patients were also subdivided into five groups with increasing 24-hour average blood pressure. Each group was further subdivided into two classes characterized by lower and higher standard deviations of all 24-hour blood pressure values, that is, with lower and higher blood pressure variabilities.15 Lower blood pressure variabilities were usually associated with less target organ damage than higher blood pressure variabilities, even in absence of any difference in mean blood pressure levels (Fig. 4). Thus ambulatory blood pressure

I 138 Mancia and Parati October 1969

American Heart Journal

AUTOMATIC CUFF INFLATION n=15

160-f

SEMI-AUTOMATIC CUFF INFLATION n, 20

j 1,111 t

Control -1 0 +lmin

Fig. 5. Intraarterial blood pressure and heart rate during the minute immediately preceding the beginning of cuff inflation (-1 to 0), the minute immediately after cuff inflation (0 to +l), and the l-minute cpntrol period taken 5 minutes before cuff inflation. Values were measured every 15 seconds. Data refer to mean + SEM results from patients in whom the inflations were applied every 15 minutes for 2 hours. In 15 patients the inflations were automatic, whereas in 20 patients they were semiautomatic, that is, the patients started the measurements by acting on the device. Arrows refer to the beginning of cuff inflation. S, systolic blood pressure; D, diastolic blood pressure; M, mean arterial pressure. (Reproduced with permission from Parati G, Pomidossi G, Casadei R, Marcia G. Lack of alerting reactions to intermitten cuff inflations during non-invasive blood pressure monitoring. Hypertension 1985;7:597-601).

reflects the untoward target organ consequences of hypertension better than isolated blood pressure values. This diagnostic superiority (which is enhanced by the ability of this approach to provide information on dynamic blood pressure events) may be accompanied by a prognostic superiority because in hypertension the extent of target organ consequences bears a clear relationship to the incidence of cardiovascular morbidity and mortality.16*17 How- ever, this needs confirmation by controlled prospective studies.

INVASIVE VS NONINVASIVE AMBULATORY BLOOD PRESSURE MONITORING

Although it provides exhaustive information on patients’ blood pressures, intraarterial 24-hour blood pressure recording can hardly be used in clinical practice, which is only suitable for noninva- hive ambuiatory biood pressure monitoring. In a series of studies designed to validate this approach, we have shown that periodic cuff inflations that

allow blood pressure to be measured for 24 hours on an automatic or semiautomatic basis do not cause an alerting reaction and a blood pressure rise in the patient (Fig 5).lg Furthermore, these inflations do not interfere with the marked physiologic hypoten- sion occurring at night (Fig 6).20,21 Thus noninvasive ambulatory blood pressure monitoring does not disturb the daytime and the nighttime blood pressure profile.

Finally, we have found that the 24-hour mean blood pressure value obtained by beat-to-beat anal- ysis of an intraarterial blood pressure tracing closely corresponds to the mean value obtained by sam- pling the same tracing every 5 to 30 minutczxz2 This means that although limited in number, the intermittent blood pressure measurements offered by noninvasive blood pressure monitoring are not incompatible with a precise estimation of the actual 24-hour mean blood pressure.

However, three features of noninvasive ambulatory blood pressure monitoring are less satisfactory:

Volumr 116

Number 4 Twenty-four hour monitoring 1139

MEAN VALUES (n=10)

mm Hg SBP mm HQ DBP

120- t 60- I I

mmHg 120-

MAP b/min

100- HR

80- 1 60 3h day 3h night 3h day 3h night

- Intra-arterial recording only

o-o Intra-arterial + non invasive automatic recording

Fig. 6. Average (+- SEM) SBP, DBP, MAP, and HR of 3 hours of wakefulness and 3 hours of night sleep in 10 subjects. All values were derived from 48-hour intra-arterial ambulatory recordings, performed for 24 hours without and for 24 hours with the addition of noninvasive ambulatory blood pressure monitoring (Spacelabs Monitor) performed on the contralateral arm. In five patients during the first 24 hours blood pressure was recorded only intraarterially, whereas the noninvasive blood pressure device was added during the second 24 hours. In the other five patients the reverse order was followed. (Reproduced with permission from Parati G, Pomidossi G, Malaspina D, Camesasca C, Mancia G. 24-Hour blood pressure measurements: methodological and clinical problems. Am J Nephrol 1986;6 (suppl 2):55-6.)

(1) The intermittent blood pressure measurements typical of this approach do not allow precise estimation of the actual 24-hour blood pressure variability.22 (2) Ambulatory blood pressure monitoring devices generate a number of artifactual blood pressure readings, the elimination of which requires lengthy visual editing.23 (3) The remaining blood pressure readings often show differences with ambulatory intra-arterial blood pressure so large as to indicate a substantial degree of inaccuracy.23-25 Finally, by analyzing the intraarterial blood pressure tracing of a large number of hypertensive subjects, we have shown that no subperiod of the 24 hours precisely corresponds to the average 24-hour blood pressure value.26 This means that there is no hope of obtaining information on 24-hour blood

pressure by shorter blood pressure monitorings. Therefore ambulatory blood pressure monitoring is likely to remain a time-consuming approach to the diagnosis of hypertension.

COffCLlJSlONS

Although cuff blood pressure measurements have been found to correlate with cardiovascular morbidity and mortality, their ability to reflect patients’ 24-hour blood pressures and closely predict hypertension-related complications is limited. This justi- fies the interest raised by 24-hour ambulatory blood pressure monitoring and worldwide research on this approach. At present ambulatory blood pressure monitoring has offered important contributions to hypertension research. It has also offered promising

I 140 Mancia and Parati October wee

American Heart Journal

data for the clinical practice. However, routine use of this approach in the clinical management of hypertension should still be discouraged because (1) new classifications of blood pressure values in ambulatory conditions are not available on a sufficiently large basis, (2) the prognostic superiority of ambulatory blood pressure monitoring over conventional blood pressure measurements has not yet been definitively proved and (3) noninvasive blood pressure monitoring (the only one eligible in the clinical practice) is still characterized by limited accuracy of individual blood pressure readings.

REFERENCES

1.

2.

3.

4.

5.

6.

7.

8.

9.

10.

11.

Bevan AT, Honour AJ, Stott FD. Direct arterial pressure recording in unrestricted man. Clin Sci 1969;36:328-44. Mancia G, Bertinieri G, Parati G, et al. Effects of blood pressure measurement by the doctor on patient’s blood pressure and heart rate. Lancet 1983;2:695-8. Mancia G, Zanchetti A. Blood pressure variability. In: Zan- chetti A, Tarazi R, eds. Handbook of hypertension, patho- physiology of hypertension: cardiovascular system, vol 7. Amsterdam: Elsevier, 1986;125-52. Mancia G. Ambulatory blood pressure monitoring in hypertension research and clinical practice. In: Hansson L, ed. Yearbook of hypertension. London: Gower, 1986:93-115. Mancia G, Parati G. Reactivity to physical and behavioural stress and blood pressure variability in hypertension. In: Julius S, Basset DR, eds. Handbook of hypertension, behavioural factors in hypertension, vol 9. Amsterdam: Elsevier, 1987:104-22. Parati G, Pomidossi G, Casadei R, et al. Limitations of laboratory stress testing in the assessment of subjects’ cardiovascular reactivity to stress. J Hypertens 1986;4 (suppl 6): s51-3. Parati G, Pomidossi G, Casadei R, et al. Comparison of the cardiovascular effects of different laboratory stressors and their relationship with blood pressure variability. J Hyper- tens 1988;6:481-8. Mancia G, Parati G, Pomidossi G, Grassi G, Casadei R, Zanchetti A. Alerting reaction and rise in blood pressure during measurement by physician and nurse. Hypertension 1987;9:209-15. Sokolow M, Werdegar S, Kain H, Hinman AT. Relationship between level of blood pressure measured casually and by portable recorders and severity of complications in essential hypertension. Circulation 1966;34:279-98. Floras JS, Jones JV, Hassan MD, Osikowska B, Sever PS, Sleight P. Cuff and ambulatory blood pressure in subjects with essential hypertension. Lancet lSBl;ii:107-9. Rowlands DB, Ireland MA, Glover DR, McLeay RAB, Stal- lard TJ, Littler WA. The relationship between ambulatory

12.

13.

14.

15.

16.

17.

18.

19.

20.

21.

22.

23.

24.

25.

26.

blood pressure and echocardiographically assessed left ventricular hypertrophy. Clin Sci 1981;61:1Ols-3s. Devereux R, Pickering TG, Harshfield GA, et al. Left ventricular hypertrophy in patients with hypertension: importance of blood pressure response to regularly recurring stress. Circulation 1983;68:470-6. Perloff D, Sokolow M, Cowan R. The prognostic value of ambulatory blood pressures. JAMA 1983;249:2792-8. Parati G, Pomidossi G, Albini F, Malaspina D, Mancia G. Relationship of 24-hour blood pressure mean and variability to severity of target organ damage in hypertension. J Hyper- ten 1987;5:93-8. Mancia G, Ferrari A, Gregorini L, et al. Blood pressure and heart rate variabilities in normotensive and hypertensive human beings. Circ Res 1983;53:96-104. Kannel WB, Abbott RD. A prognostic comparison of asymp- tomatic left ventricular hypertrophy and unrecognized myo- cardial infarction. AM HEART J 1986;111:391-7. Kannel WB, Sorlie P. Left ventricular hypertrophy in hypertension: prognostic and pathogenetic implications. In: Strov- er BE, ed. The heart in hypertension. Berlin: Springer- Verlag, 1981:223-48. Samuelsson D, Wilhelmsen L, Elmfeidt J, et al. Predictions of cardiovascular morbidity in treated hypertension: results from the primary preventive trial in Goteborg, Sweden. J Hypertens 1985;3:167-76. Parati G, Pomidossi G, Casadei R, Mancia G. Lack of alerting reactions to intermittent cuff inflations during non-invasive blood pressure monitoring. Hypertension 1985;7:597-601. Parati G, Pomidossi G, Casadei R, et al. Ambulatory blood pressure monitoring does not interfere with the haemody- namic effects of sleep. J Hypertens 1985;3(suppl 2):S107-9. Parati G, Pomidossi G, Malaspina D, Camesasca C, Mancia G. 24-Hour blood pressure measurements: methodological and clinical problems. Am J Nephrol 1986;6 (suppl 2):55- 60. Di Rienso M, Grassi G, Gregorini L, Pedotti A, Mancia G. Continuous vs intermittent blood pressure measurements in estimating 24 hours average blood pressure. Hypertension 1983;5:264-9. Casadei R, Parati G, Pomidossi G, Groppelli A, Trazzi S, Mancia G. Twenty-four hour blood pressure monitoring: evaluation of Spacelabs 5300 Monitor by comparison with intra-arterial blood pressure recording in ambulant subjects. J Hype&ens 1988 (in press). Gould BA, Hornung MB, Cashman PMM, Altman D, Raftery EB, An evaluation of the Avionics Pressurometer III 1978 at home and hospital. Clin Cardiol 1986;9:335-43. Palatini P, Pessina AC, Sperti G, et al. Comparison between and indirect and a direct method of ambulatory blood pressure monitoring. In: Stott FD, Raftery EB, Clement DL, eds. Proceedings of the 4th International Symposium on Ambulatory Monitoring. London: Academic Press, 1982: 449. Di Rienzo M, Parati G, Pomidossi G, Veniani M, Pedotti A, Mancia G. Blood pressure monitoring over short day and night times cannot predict 24-h average blood pressure. J Hypertens 198$3:343-S.

experience with 24-hour ambulatory blood pressure monitoring in hypertension

Documents