systolic vs diastolic blood pressure control in the ... · systolic vs diastolic blood pressure...

Systolic vs Diastolic Blood Pressure Control inthe Hypertensive Patients of the PAMELA PopulationGiuseppe Mancia, MD; Michele Bombelli, MD; Arturo Lanzarotti, MD; Guido Grassi, MD;GianCarlo Cesana, MD; Alberto Zanchetti, MD; Roberto Sega, MD

Background: Previous studies have shown that in thetreated fraction of the hypertensive population, blood pres-sure (BP) control is less common for systolic BP (SBP)than for diastolic BP (DBP) as measured in the physi-cian’s office. Whether this phenomenon is artifactuallyattributable to a temporary increase in BP owing to a“white-coat” effect or represents a true rarity of SBP con-trol in daily life is unknown.

Methods: Data were obtained from the PAMELA (Pres-sioni Arteriose Monitorate E Loro Associazioni) studypopulation, which involved individuals ranging in agefrom 25 to 74 years who were representative of the resi-dents of Monza (a city near Milan, Italy) and who werestratified according to sex. Office (an average of 3 sphyg-momanometric measurements), home (an average ofmorning and evening self-measurements using a semi-automatic device), and 24-hour ambulatory (average ofmeasurements performed every 20 minutes during theday and at night) BP values were obtained in all studysubjects. In the treated hypertensive patients, BP was re-garded as controlled if office values were less than 140

(SBP) or 90 (DBP) mm Hg. Home and 24-hour averageSBP and DBP were regarded as controlled if the valueswere lower than 132/83 and 125/79 mm Hg, respec-tively.

Results: In the study participants (n=2051), the numberof patients with hypertension who were receiving antihy-pertensive treatment was 398, or approximately 42% of allindividuals with hypertension. In-office SBP control by treat-ment was less frequent than DBP control (29.9% vs 41.5%,P�.05). This was also the case when home and 24-hourSBP and DBP control was considered (38.3% vs 54.6% and50.8 vs 64.9%, respectively, P�.05 for both).

Conclusions: In the PAMELA population, SBP controlby treatment was much less frequent than DBP controlby treatment. This was the case not only for office BP val-ues but also for home and 24-hour BP values, demon-strating that inadequate SBP control is not limited to ar-tificial BP-measuring methods but occurs in daily life.

Arch Intern Med. 2002;162:582-586

S EVERAL STUDIES performed ina variety of countries haveshown that blood pressure(BP) values are satisfactorilycontrolled by treatment in

only a small fraction of the hypertensivepopulation.1-9 Some of these studies havealso shown that control by treatment iseven less common for systolic BP (SBP)than for diastolic BP (DBP); ie, that in anumber of patients, DBP values are lowerthan 90 mm Hg, while SBP values remainabove 140 mm Hg.10-12 This finding has im-portant adverse implications for publichealth, because in the general populationSBP has been shown to be an equal or evenmore important cardiovascular risk fac-tor than DBP.13,14 Furthermore, in elderlyindividuals with hypertension, a reduc-tion in SBP by treatment has often beenshown to be a more important determi-nant of a patient’s protection from cardio-vascular disease than a reduction in DBP.15

We previously showed that a notice-able number of patients with hyperten-sion in an urban population living in atown (Monza) northeast of Milan, Italy,were untreated and that among those whowere treated only a small percentage hadoffice BP values that were lower than140/90 mm Hg, as has been documentedin other countries.16 Herein we report theresults obtained by analyzing the status ofSBP control vs DBP control by treatmentin the hypertensive fraction of the samepopulation. This analysis provided data onSBP and DBP control based not only onoffice values but also on home and ambu-latory BP values, which enabled us to de-termine the degree of SBP and DBP con-trol in conditions of daily life.

RESULTS

In the PAMELA population, the percent-age of subjects with an increase in office

ORIGINAL INVESTIGATION

From Clinica Medica(Drs Mancia, Bombelli,Lanzarotti, Grassi, and Sega)and Clinica del Lavoro(Dr Cesana), Università diMilano-Bicocca, OspedaleSan Gerardo, Monza, Italy;and Istituto Auxologico Italiano(Drs Mancia, Grassi, andZanchetti) and CentroFisiologia Clinica eIpertensione, Università Milano(Drs Mancia, Grassi, andZanchetti), Milano, Italy.

(REPRINTED) ARCH INTERN MED/ VOL 162, MAR 11, 2002 WWW.ARCHINTERNMED.COM582

©2002 American Medical Association. All rights reserved.

Downloaded From: on 07/08/2018

BP values who were unaware of their status or un-treated was 26.6% of the overall number of subjects inwhom office measurements were available. The corre-sponding figures for home and 24-hour BP values were22.1% and 22.1%, respectively.

The number of subjects with hypertension who weretreated was 398. As shown in Figure 1, (1) office SBPand DBP values were controlled in only 21.1% of the casesand (2) SBP control was less frequent than DBP control.Although the percentages were somewhat greater (Fig-ure 1), similar findings were observed when SBP and DBPcontrol was based on home and 24-hour BP values. Thisobservation is further illustrated in Figure2 which showsindividual office, home, and 24-hour BP values in thetreated subjects with hypertension. Figure 3 andFigure 4 show the data separately for sex and age. Foroffice, home, and 24-hour BP values, systolic control wasinvariably less frequent than diastolic control in both

women and men (Figure 3). Office, home, and 24-hourSBP control became progressively less frequent as age in-creased (Figure 4), whereas DBP control did not showany age-related change.

COMMENT

The PAMELA study has shown that in subjects who wereidentified as hypertensive by screening a large sample ofan urban population living in the northeast district of Mi-lan, BP control by treatment is rare.16 It has also shownthat this is the case not only for office values but also forhome and ambulatory values; therefore, this phenom-enon is attributable not to a “white-coat” effect causinga temporary increase in BP22 but to a true rarity of BP con-trol in daily life. The present analysis of the PAMELA dataadds 2 pieces of evidence to these results: (1) the rareBP control of the treated hypertensive subjects of the

SUBJECTS AND METHODS

The data reported herein originate from the PAMELA (Pres-sioni Arteriose Monitorate E Loro Associazioni) study, whichwas performed to gather information on the normal val-ues of ambulatory and home BP in a representative sampleof the population of Monza, a town in the northeast dis-trict of Milan, Italy. The methodological aspects of thePAMELA study have been reported elsewhere.16-19 Briefly,3200 individuals aged 25 to 74 years were selected fromthe Monza residents according to the age (decades), sex,and socioeconomic criteria of the World Health Organiza-tion’s MONICA (Monitoring Trends and Determinants inCardiovascular Disease) Project, which was performed inthe same geographical area.20 The individuals selected wereasked by letter to report to the local hospital on a workdaymorning (Monday-Friday). The participants were inter-viewed and visited so that demographic data could be re-corded and information on medical history, lifestyle, car-diovascular risk factors, and diseases could be obtained. Thesame information was obtained from nonparticipants whowere interviewed by telephone and found to be similar tothe participants in age, sex, socioeconomic status, and preva-lence of cardiovascular risk and diseases. A trained physi-cian measured the participants’ sitting BP 3 times using amercury sphygmomanometer. The sitting BP was furthermeasured (1) over 24 hours with an oscillometric devicemodel 90207; SpaceLabs Inc, Redmond, Wash) that wasprogrammed to start the monitoring period after comple-tion of the office BP measurements and to automaticallyread BP every 20 minutes21 and (2) in the morning and inthe evening at home with a semiautomatic oscillometric de-vice (model HP 5331; Philips, Tokyo, Japan) whose cuffhad to be positioned on the arm contralateral to that usedfor ambulatory BP monitoring. The participants had to re-cord the values shown on the digital display of the devicein a diary and give it to the investigator. A similar numberof subjects were studied during the 4 different seasons andduring the 5 different workdays.19

For each participant, the 3 clinic BP values and the 2home BP values were averaged, with the SBP and DBP val-ues being averaged separately. Ambulatory SBP and DBP

values were edited from artifacts21 and averaged over the 24hours. There were 2051 participants in the study (64.1% ofthe overall sample). Home BP values were available for 1867subjects. Ambulatory BP data were available for 2027 sub-jects and were found to be adequate (at least 70% of the 72readings over the 24 hours were valid) in 96% of the sub-jects. In the group as a whole, the valid SBP and DBP read-ings were 94.7%±2.4% (mean±SD) and 93.5%±2.2%,respectively, of the 72 readings during the 24-hour period.

The subjects with hypertension were classified as un-aware or untreated if (1) the office BP value was greaterthan or equal to 140 mm Hg systolic and/or 90 mm Hg di-astolic, (2) the home BP value was greater than or equal to132 mm Hg systolic and/or 83 mm Hg diastolic, and (3)the 24-hour average BP value was greater than or equal to125 mm Hg systolic and/or 79 mm Hg diastolic with nohistory of antihypertensive drug treatment in the preced-ing 15 days. Selection of the home and ambulatory BP val-ues was based on previous calculation of the upper limitof normal of the home and 24-hour average BP values forthe whole PAMELA population.17 The subjects with hy-pertension were classified as treated, regardless of the BPvalues, if current antihypertensive drug treatment was re-ported. They were further classified as treated with SBP andDBP control if the report of current antihypertensive treat-ment was accompanied by an office SBP value of less than140 mm Hg and an office DBP of less than 90 mm Hg, re-spectively. The SBP or DBP control was also established byhome (systolic, �132 mm Hg; diastolic, �83 mm Hg) and24-hour average (systolic, �125 mm Hg; diastolic, �79 mmHg) BP values. Antihypertensive treatment was reportedto consist of the use of a diuretic (23%), a �-blocker (8%),a calcium antagonist (9%), an angiotensin-converting en-zyme inhibitor (18%), or a combination of 2 drugs or othertreatments (42%).

Data were also analyzed in a similar fashion for groupsof different age ranges (25-49, 50-64, and 65-74 years) andfor men and women. Mean±SD values from the variousgroups were compared by 2-way analysis of variance us-ing the t test for unpaired observations and the Bonferronicorrection for multiple comparisons to locate between-group differences. A P value of .05 was taken as the levelof statistical significance.




PAMELA population is more commonly attributable toan inadequate reduction of SBP than of DBP; (2) onceagain, the rare BP control involves not only office valuesbut home and ambulatory BP values as well. Therefore,we may conclude that in the hypertensive fraction of thePAMELA population (1) an effective decrease in SBP isunquestionably less common than an effective reduc-tion of DBP and (2) this also occurs when BP is mea-sured at home or in ambulatory conditions, thus indi-cating that SBP control is indeed less common than DBPcontrol in daily life.16

Our data are in line with the results of previous stud-ies that have also observed a less frequent SBP than DBPcontrol in treated patients with hypertension, althoughdata were based only on office BP values.10-12 Therefore, itis important to discuss 2 possible explanations for aphenomenon that mainly accounts for the poor rate of over-all BP control that has consistently been reported in vari-ous populations. One explanation is that current hyper-tensive drugs are less effective in lowering SBP than DBP,because the alteration involved in an SBP elevation, ie, anincrease in arterial stiffness,23 is less easily reversible thanthe alteration that is involved in a DBP elevation, ie, anincrease in arteriolar resistance to blood flow.23 The sec-ond explanation is that physicians titrate antihyperten-sive treatment of DBP and terminate further therapeuticefforts once DBP values are lower than 90 mm Hg, evenwhen SBP values are higher than 140 mm Hg. It should

SBP + DBP SBP DBP

P<.05

Office BP(n = 398)

21.1%29.9% 41.5%

Home BP(n = 339)

28.0% 38.3% 54.6%

P<.05

24-h BP(n = 390)

45.4% 50.8% 64.9%

P<.05

Controlled BPUncontrolled BP

Figure 1. The white part of the circles represents the percentage of treatedhypertensive subjects in the PAMELA (Pressioni Arteriose Monitorate E LoroAssociazioni) population with both systolic (S) and diastolic (D) bloodpressure (BP) control, SBP control only, and DBP control only. The shadedpart of the circle represents the percentage of treated hypertensive subjectswith uncontrolled SBP and DBP, SBP only, or DBP only. Control was definedbased on upper limits of office BP values (S, �140 mm Hg; D, �90 mmHg), home BP (S, �132 mm Hg; D, �83 mm Hg) or 24-hour average BPvalues (S, �125 mm Hg; D, �79 mm Hg). The upper limits of home and24-hour average BP values were derived from previous analysis of the datafrom the entire population. There were 398 treated hypertensive subjects,339 and 390 of whom also underwent monitoring of home and 24-houraverage BP values, respectively.

80

40

60

20

0

Office BP Home BP 24-h BP

Trea

ted

HTs

With

BP

Cont

rol,

%

Men(n = 185)

P <.05

Women(n = 213)

P <.05

Men(n = 170)

P <.05

Women(n = 169)

P <.05

Men(n = 180)

P <.05

Women(n = 210)

P <.05SBP ControlDBP Control

Figure 3. Percentage of treated hypertensive men and women (HTs) withsystolic (S) or diastolic (D) blood pressure (BP) control based on office,home, or 24-hour values.

230

170

200

140

110

80

5050 70 90 110 130

DBP, mm Hg

Office BP (n = 398)

SBP,

mm

Hg

A

230

170

200

132

110

80

5050 70 83 90 110 130

DBP, mm Hg

Home BP (n = 339)

SBP,

mm

Hg

B

210

150

180

125

90

6040 60 79 100 120

DBP, mm Hg

24-h BP (n = 390)

SBP,

mm

Hg

C

Figure 2. Individual systolic (S) and diastolic (D) blood pressure (BP) valuesin all treated hypertensive subjects of the PAMELA (Pressioni ArterioseMonitorate E Loro Associazioni) study. The horizontal and vertical linesindicate the upper normal limits for office, home, and 24-hour SBP andDBP values.




be emphasized that both these explanations may be valid,because large-scale studies of isolated systolic hyperten-sion have shown that drug treatment based on SBP reduc-tion rarely lowers SBP values below 140 mm Hg.24-26 Fur-thermore, in the large number of patients with systodiastolichypertension who were recruited for the Hypertension Op-timal Treatment (HOT) study,27 a reduction of DBP wellbelow 90 mm Hg (average value, 83 mm Hg) was achieved,even when the SBP remained above 140 mm Hg. This find-ing was also noted among the hypertensive patients in theInternational Nifedipine GITS (gastrointestinal therapeu-tic system) Study: Intervention as a Goal in HypertensionTreatment (INSIGHT) study, in which the average DBPvalue was reduced by treatment to 82 mm Hg, while theSBP value remained only slightly below 140 mm Hg.28 Thisfinding suggests that DBP control may be achieved atsmaller doses and/or with a lesser number of drugs thanSBP control, which may require more aggressive drugtreatment.

Several other points deserve to be mentioned. First,in the PAMELA study office, home and ambulatory SBPcontrol was less frequent than DBP control in both treatedmen and treated women. By and large, however, BP con-trol was more frequent in women than in men. This out-come confirms and extends the results of previous stud-ies that were based only on office BP values,10-12 whichhave suggested this phenomenon to be attributable notto a sex difference in the efficacy of antihypertensive drugsbut to a greater compliance by women to antihyperten-sive treatment. Second, in the PAMELA study patients,the percentage of treated subjects with SBP control wasprogressively less from the youngest to the oldest age stra-tum, regardless of whether office, home, or ambulatoryBP was considered. This result is unlikely to be ac-counted for by the fact that the compliance of elderly pa-tients to treatment is less than that of younger patients,because DBP control was similar at all ages. It is morelikely to depend on the fact that (1) physicians may adopta less aggressive therapeutic attitude when they face a BPincrease in the elderly (because of lack of full percep-tion of its risk and/or fear of a “J curve” phenomenon)and (2) in the elderly the BP increase is more frequentlysystolic, ie, the component of the BP profile that may bemore difficult to normalize, possibly because of the lim-ited reversibility of an increase in arterial stiffness. Third,

in the treated hypertensive patients of the PAMELA popu-lation, SBP and DBP were more frequently found to becontrolled when home or ambulatory rather than officeBP normalization was considered, despite the very strin-gent criteria adopted to define the normal values for homeand ambulatory BP.17 This result may depend on the factthat in some patients office BP may appear not to be nor-malized by treatment because of the temporary increasein BP values that is caused by a white-coat effect.22 Thus,when the prevalence of uncontrolled hypertension in apopulation is being investigated, office BP measure-ments may be complemented by BP measurements ob-tained outside the office to avoid an overestimation as aresult of this phenomenon.

The very low frequency of SBP control in the hyper-tensive population has an obvious adverse significance forpublic health. Because SBP is now recognized as being anequal or even more important cardiovascular risk factorthan DBP,13,14 its more common lack of control means thatthe population in general will remain at a high risk for car-diovascular complications. Furthermore, a more com-mon reduction in DBP values than in SBP values inducesan increase in pulse pressure, ie, an increase in a variablethat has recently been shown to be a possible indepen-dent risk factor for cardiovascular disease.29

Accepted for publication July 17, 2001.Corresponding author and reprints: Giuseppe Man-

cia, MD, Clinica Medica, Universita di Milano-Bicocca, Os-pedale San Gerardo, Via Donizetti 106, 20052 Monza, Italy.

REFERENCES

1. Trenkwalder P, Ruland P, Stender M, et al. Prevalence, awareness, treatment andcontrol of hypertension in a population over the age of 65 years: results fromthe Starnberg Study on Epidemiology of Parkinsonism and Hypertension in theElderly (STEPHY). J Hypertens. 1994;12:709-716.

2. Menard J, Chatellier G. Limiting factors in the control of BP: why is there a gapbetween theory and practice? J Hum Hypertens.1995;9(suppl 2):S19-S23.

3. Marques-Vidal P, Tuomilehto J. Hypertension awareness, treatment and control inthecommunity: is the“ruleofhalves”still valid?JHumHypertens.1997;11:213-220.

4. Chockalingham A, Fodor JG. Treatment of raised blood pressure in the popula-tion: the Canadian experience. Am J Hypertens.1998;11:747-749.

5. Joint National Committee on Detection, Evaluation, and Treatment of High BloodPressure. The sixth report of the Joint National Committee on Prevention, De-tection, Evaluation, and Treatment of High Blood Pressure. Arch Intern Med. 1997;157:2413-2446.

80

40

60

20

0

SBP Control

Trea

ted

HTs

With

BP

Cont

rol,

%

Office BP Home BP 24-h BPDBP Control

Office BP Home BP 24-h BP

25-49 y50-64 y65-74 y

A B

Figure 4. Percentage of treated hypertensive patients (HTs) with systolic (S) or diastolic (D) blood pressure (BP) control grouped according to 3 differentage ranges.




6. Colhoun HM, Dong W, Poulter NR. Blood pressure screening, management andcontrol in England: results from the health survey for England 1994. J Hyper-tens. 1998;16:747-752.

7. Chamontin B, Poggi L, Lang T, et al. Prevalence, treatment and control of hy-pertension in the French population: data from a survey on high blood pressurein general practice, 1994. Am J Hypertens. 1998;11:759-762.

8. Coca A. Actual blood pressure control: are we doing things right? J HypertensSuppl. 1998;16:S45-S51.

9. Swales JD. Current clinical practice in hypertension: the EISBERG (Evaluationand Interventions for Systolic Blood pressure Elevation-Regional and Global)project. Am Heart J. 1999;138:231-237.

10. Burt VL, Cutler JA, Higgins M, Horan MJ, Labarthe D, Whelton P. Trends in preva-lence, awareness, treatment and control of hypertension in the adult US popu-lation: data from the health examination surveys, 1960 to 1991. Hypertension.1995;26:60-69.

11. Cushman NC, Black HR, Probsfield J, et al. Blood pressure control in the Anti-hypertensive and Lipid Lowering Treatment to Prevent Heart Attack Trial (ALLHAT)[abstract]. Am J Hypertens. 1998;11(pt 2):17A.

12. Lapuerta P, l’Italien GJ. Awareness, treatment and control of systolic blood pres-sure in the United States [abstract]. Am J Hypertens. 1999;12(pt 2):92A.

13. Kannel WB, Gordon T, Schwarz MJ. Systolic versus diastolic blood pressure andrisk of coronary heart disease: the Framingham Study. Am J Cardiol. 1971;27:335-346.

14. Levy D. The role of systolic blood pressure in determining risk for cardiovascu-lar diseases. J Hypertens Suppl. 1999;17:S5-S18.

15. Amery A, Birkenhager W, Brixko P, et al. Mortality and morbidity results fromthe European Working Party on High Blood Pressure in the Elderly trial. Lan-cet.1985;1:1349-1358.

16. Mancia G, Sega R, Milesi C, Cesana G, Zanchetti A. Blood pressure control in thehypertensive population. Lancet. 1997;349:454-457.

17. Mancia G, Sega R, Bravi C, et al. Ambulatory blood pressure normality: resultsfrom the PAMELA Study. J Hypertens. 1995;13:1377-1390.

18. Sega R, Cesana G, Milesi C, Grassi G, Zanchetti A, Mancia G. Ambulatory andhome blood pressure normality in the elderly. Hypertension. 1997;30:1-6.

19. Sega R, Cesana G, Bombelli M, et al. Seasonal variations in home and ambula-

tory blood pressure in the PAMELA population. J Hypertens. 1998;16:1585-1592.

20. WHO Monica Project Principal Investigators. The World Health OrganizationMONICA Project: a major international collaboration. J Clin Epidemiol. 1988;41:105-114.

21. Groppelli A, Omboni S, Parati G, Mancia G. Evaluation of invasive blood pres-sure monitoring device Spacelabs 90202 and 90207 versus resting and ambu-latory 24-hour intra-arterial blood pressure. Hypertension. 1992;20:227-232.

22. Mancia G, Bertinieri G, Grassi G, et al. Effects of blood pressure measurementsby the doctor on patient’s blood pressure and heart rate. Lancet.1983;2:695-698.

23. Nichols WW, O’Rourke MF. McDonald’s Blood Flow in Arteries. 4th ed. London,England: Arnold; 1998.

24. SHEP Cooperative Research Group. Prevention of stroke by antihypertensive drugtreatment in older patients with isolated systolic hypertension: final results ofthe Systolic Hypertension in the Elderly Program (SHEP). JAMA. 1991;265:3255-3264.

25. Staessen JA, Fagard R, Thijs L, et al. Randomised double-blind comparison ofplacebo and active treatment for older patients with isolated systolic hyperten-sion: the Systolic Hypertension in Europe (Syst-Eur) Trial Investigators. Lancet.1997;350:757-764.

26. Liu L, Wang JG, Gong L, Liu G, Staessen J, for the Systolic Hypertension in China(Syst-China) Collaborative Group. Comparison of active treatment and placebofor older patients with isolated systolic hypertension. J Hypertens. 1998;16:1823-1829.

27. Hansson L, Zanchetti A, Carruthers SG, et al, for the HOT Study Group. Effectsof intensive blood-pressure lowering and low-dose aspirin in patients with hy-pertension: principal results of the Hypertension Optimal Treatment (HOT) ran-domised trial. Lancet. 1998;351:1755-1762.

28. Brown MJ, Palmer CR, Castaigne A, et al. Morbidity and mortality in patients ran-domised to double-blind treatment with a long-acting calcium-channel blockeror diuretic in the International Nifedipine GITS Study: Intervention as a Goal inHypertension Treatment (INSIGHT). Lancet. 2000;356:366-372.

29. Benetos A. Pulse pressure and cardiovascular risk. J Hypertens. 1999;17(suppl5):S21-S24.




systolic vs diastolic blood pressure control in the ... · systolic vs diastolic blood pressure...

Documents