Blood Pressure Monitoring in the Elderly -...

865

Age is the main risk factor for hypertension and death.1 By 2050, the proportion of octogenarians will globally

increase from 1.5% to 4.3%.2 At all ages, high blood pressure is the major driver of cardiovascular complications.1 Worldwide, hypertension causes 9.4 million deaths, underlying over half of the estimated 17 million cardiovascular deaths and accounts for >20% of health loss in seniors aged ≥70 years.3

In randomized trials of antihypertensive treatment in the elderly,4–11 the systolic target pressure was ≥150 mm Hg, whereas diastolic targets were either not specified,5,6,8,10 or set at or > 80 mm Hg.9,11 After publication of the Hypertension in the Very Elderly Trial (HYVET), which demonstrated ben-efit of lowering office systolic pressure to 150 mm Hg in fit octogenarians,11 updated guidelines12–15 suggested that in very

Abstract—To generate outcome-driven thresholds for home blood pressure (BP) in the elderly, we analyzed 375 octogenarians (60.3% women; 83.0 years [mean]) enrolled in the International Database on home BP in relation to cardiovascular outcome. Over 5.5 years (median), 155 participants died, 76 from cardiovascular causes, whereas 104, 55, 36, and 51 experienced a cardiovascular, cardiac, coronary, or cerebrovascular event, respectively. In 202 untreated participants, home diastolic in the lowest fifth of the distribution (≤65.1 mm Hg) compared with the multivariable-adjusted average risk was associated with increased risk of cardiovascular mortality and morbidity (hazard ratios [HRs], ≥1.96; P≤0.022), whereas the HR for cardiovascular mortality in the top fifth (≥82.0 mm Hg) was 0.37 (P=0.034). Among 173 participants treated for hypertension, the HR for total mortality in the lowest fifth of systolic home BP (<126.9 mm Hg) was 2.09 (P=0.020). In further analyses of home BP as continuous variable (per 1-SD increment), higher diastolic BP predicted lower cardiovascular mortality and morbidity and cardiac and coronary risk (HR≤0.65; P≤0.039) in untreated participants. In those treated, cardiovascular morbidity was curvilinearly associated with systolic home BP with nadir at 148.6 mm Hg and with a 1.45 HR (P=0.046) for a 1-SD decrease below this threshold. In conclusion, in untreated octogenarians, systolic home BP ≥152.4 and diastolic BP ≤65.1 mm Hg entails increased cardiovascular risk, whereas diastolic home BP ≥82 mm Hg minimizes risk. In those treated, systolic home BP <126.9 mm Hg was associated with increased total mortality with lowest risk at 148.6 mm Hg. (Hypertension. 2015;66:865-873. DOI: 10.1161/HYPERTENSIONAHA.115.05800.) • Online Data Supplement

Key Words: aged ■ aged 80 and over ■ cardiovascular diseases ■ home blood pressure monitoring ■ hypertension

Received May 7, 2015; first decision May 18, 2015; revision accepted July 3, 2015.From the Division of Hypertension, Department of Internal Medicine, Hospital Italiano de Buenos Aires, Capital Federal, Argentina (L.S.A., Jessica

Barochiner, J.A., P.E.C., G.D.W.); Studies Coordinating Centre, Research Unit Hypertension and Cardiovascular Epidemiology, KU Leuven Department of Cardiovascular Sciences, University of Leuven, Leuven, Belgium (L.T., L.J., A.N.O., J.A.S.); Centro de Nefrología and Departamento de Fisiopatología, Hospital de Clínicas, Universidad de la República, Montevideo, Uruguay (José Boggia); Department of Internal Medicine, College of Health Sciences, University of Abuja, Abuja, Nigeria (A.N.O.); Department of Hygiene and Public Health, Teikyo University School of Medicine, Tokyo, Japan (K.A., K.N., T.O.); Department of Public Health, Tohoku University Graduate School of Medicine, Sendai, Japan (I.T.); Hypertension Center, Third Department of Medicine, University of Athens, Sotiria Hospital, Athens, Greece (G.S.S); Tohoku University Graduate School of Pharmaceutical Sciences, Sendai, Japan (M.K., Y.I.); and R & D VitaK Group, Maastricht University, Maastricht, The Netherlands (J.A.S.).

This article was sent to David A. Calhoun, Guest Editor, for review by expert referees, editorial decision, and final disposition.The online-only Data Supplement is available with this article at http://hyper.ahajournals.org/lookup/suppl/doi10.1161/HYPERTENSIONAHA.

115.05800/-/DC1.Correspondence to Jan A. Staessen, Studies Coordinating Centre, Research Unit Hypertension and Cardiovascular Epidemiology, KU Leuven

Department of Cardiovascular Sciences, University of Leuven, Campus Sint Rafaël, Kapucijnenvoer 35, PO Box 7001, BE-3000 Leuven, Belgium. E-mail [email protected]

Defining Thresholds for Home Blood Pressure Monitoring in Octogenarians

Lucas S. Aparicio, Lutgarde Thijs, José Boggia, Lotte Jacobs, Jessica Barochiner, Augustine N. Odili, José Alfie, Kei Asayama, Paula E. Cuffaro, Kyoko Nomura,

Takayoshi Ohkubo, Ichiro Tsuji, George S. Stergiou, Masahiro Kikuya, Yutaka Imai, Gabriel D. Waisman, Jan A. Staessen;

on behalf of the International Database on Home Blood Pressure in Relation to Cardiovascular Outcome (IDHOCO) Investigators

© 2015 American Heart Association, Inc.

Hypertension is available at http://hyper.ahajournals.org DOI: 10.1161/HYPERTENSIONAHA.115.05800

Blood Pressure Monitoring in the Elderly

by guest on July 9, 2018http://hyper.ahajournals.org/

Dow

nloaded from


Dow

nloaded from


Dow

nloaded from


Dow

nloaded from


Dow

nloaded from


Dow

nloaded from


Dow

nloaded from


Dow

nloaded from


Dow

nloaded from


Dow

nloaded from


Dow

nloaded from


Dow

nloaded from


Dow

nloaded from


Dow

nloaded from


Dow

nloaded from


Dow

nloaded from


Dow

nloaded from


Dow

nloaded from

http://hyper.ahajournals.org/lookup/suppl/doi:10.1161/HYPERTENSIONAHA.115.05800/-/DC1

http://hyper.ahajournals.org/lookup/suppl/doi10.1161/HYPERTENSIONAHA.115.05800/-/DC1

http://hyper.ahajournals.org/lookup/suppl/doi10.1161/HYPERTENSIONAHA.115.05800/-/DC1

mailto:[email protected]

http://hyper.ahajournals.org/


















866 Hypertension October 2015

elderly hypertensive patients systolic pressure should be low-ered to 140 mm Hg12,14 or 150 mm Hg 13–15 without stated goal diastolic pressure. Compared with office measurement, home blood pressure monitoring is a better prognosticator because of the greater number of readings, the minimization of the white coat effect, and the unmasking of masked hypertension.14,16,17 According to Japanese guidelines,17 home blood pressure measurement is useful for preventing an excessive or insuf-ficient antihypertensive effect of drugs. To our knowledge, the Predictive Values of Blood Pressure and Arterial Stiffness in Institutionalized Very Aged Population Study (PARTAGE)18,19 is the only survey based on the self-measured blood pres-sure in octogenarians. High blood pressure did not entail an increased risk of all-cause mortality or major cardiovascular events. In view of the uncertainty about blood pressure targets to be achieved in the very elderly, we assessed the predictive value of the self-measured home blood pressure in octogenar-ians enrolled in the International Database on Home Blood Pressure in Relation to Cardiovascular Outcome (IDHOCO).20 In line with the PARTAGE findings,19 we stratified our analy-sis according to treatment status.

Methods

Study PopulationExpanded Methods available in the online only Data Supplement provide detailed information on the construction of the IDHOCO database20 and the characteristics of enrolled cohorts (Table S1 in the online-only Data Supplement). For the present analysis, we selected 375 participants aged ≥80 years, of whom 202 were untreated at baseline. Hospitalized, institutionalized, incapacitated, or demented patients were excluded from recruitment. The analysis included 144 (untreated/treated, 106/38) residents of Ohasama, Japan21; 137 (75/62) inhabitants of the Tsurugaya district in Sendai, Japan22; 36 (21/15) inhabitants of Didima, Greece23; and 58 treated outpatients of a general hospital in Buenos Aires, Argentina.24 All studies received ethical approval. Participants provided written informed consent.

Collection of the Baseline DataOffice blood pressure was the average of 2 readings obtained with a standard mercury sphygmomanometer,23 or with an automated aus-cultatory21 or oscillometric device,22,24 using the appropriate cuff size. All home blood pressure measurements were performed in the sitting position with a validated oscillometric device,21–24 using the appropri-ate cuff size. The average of all available home measurements was used in the analysis. Throughout this article, unless otherwise speci-fied, blood pressure refers to the self-measured home blood pressure.

We obtained information on each participant’s medical history, use of medications, and smoking and drinking habits by means of ques-tionnaires21–23 or checking hospital files.24 The Expanded Methods in the online only Data Supplement provide information on the collec-tion of anthropometric and biochemical data.

Ascertainment of EventsWe ascertained vital status and the incidence of fatal and nonfatal diseases from the appropriate sources in each country (Table S1), as described in previous publications.20 Fatal and nonfatal stroke did not include transient ischemic attacks. Cardiac events comprised fatal and nonfatal myocardial infarction, death caused by ischemic heart disease or other cardiac causes, sudden death, fatal and nonfatal heart failure, coronary revascularization, and pacemaker implantation. The composite cardiovascular end point included cardiovascular mortality and the aforementioned cardiovascular end points. In outcome analy-ses of composite cardiovascular end points, we only considered the first event in each category.

Statistical AnalysisFor database management and statistical analysis, we used SAS software, version 9.3 (SAS Institute Inc, Cary, NC). For comparison of means and proportions, we applied the large sample z-test or ANOVA and the χ2 sta-tistic, respectively. We compared intercepts and slopes of regression lines between untreated and treated participants by ANCOVA.25 Statistical sig-nificance was a α-level of <0.05 on 2-sided tests.

We applied the deviation from mean coding to compute hazard ra-tios (HRs) in fifths of the distribution of the home blood pressure. This approach expresses the risk in each fifth relative to the overall risk in the whole study population and allows computing 95% con-fidence intervals (CIs) for the HRs in all fifths without defining an arbitrary reference group. Next, we used Cox proportional hazards models to compute HRs expressing the risk associated with blood pressure analyzed as continuous variable, while adjusting for cohort, sex, age, body mass index, smoking, drinking alcohol, serum cho-lesterol, and a history of cardiovascular disease and diabetes melli-tus. Fully adjusted models included both systolic and diastolic blood pressures. We checked the proportional hazards assumption and the functional forms of the covariables by the Kolmogorov-type supre-mum test. We tested heterogeneity in the HRs across subgroups by introducing the appropriate interaction term in the Cox model. We applied the generalized R2 statistic26 to assess the risks additionally explained by home blood pressure over and beyond other risk factors.

Results

Characteristics of ParticipantsTable 1 shows baseline characteristics of the 202 untreated and 173 treated octogenarians included in IDHOCO. The baseline characteristics of each cohort appear in Tables S3 and S4 for untreated and treated participants, respectively. The median number (5th to 95th percentile interval) of home blood pres-sure measurements amounted to 48 (8–56) in Ohasama, 10 (2–34) in Tsurugaya, 12 (12–12) in Didima, and 24 (22–28) in Buenos Aires. Among treated participants belonging to the Ohasama, Tsurugaya, and Buenos Aires cohorts, 80 (50.6%) were on monotherapy and 78 (49.4%) were on combination therapy taking a median of 1.0 drug per day. In treated partici-pants of these 3 cohorts, 37 (23.4%) used diuretics, 40 (25.3%) β-blockers, 33 (20.9%) angiotensin-converting enzyme inhib-itors, 38 (24.1%) angiotensin-II type-1 receptor blockers, 113 (71.5%) calcium-channel blockers, and 22 (13.9%) other vari-ous classes of antihypertensive agents. Information on drugs use was unavailable from the Didima cohort.

Analysis of Untreated Participants

Incidence of End PointsAmong 202 untreated octogenarians, median follow-up was 5.5 years (5th to 95th percentile interval, 0.9–11.8 years). Over 1164 person-years of follow-up, 103 participants died (88.5 per 1000 person-years) and 58 experienced a fatal or nonfatal cardiovascular complication (51.5 per 1000 person-years). The cause of death was cardiovascular in 46 partici-pants, noncardiovascular in 55, end-stage renal disease in 1, and unknown in 1. Considering cause-specific first cardio-vascular events, the incidence of stroke and cardiac events amounted to 28 (24.8 per 1000 person-years) and 30 (25.8 per 1000 person-years), respectively.

Risk Associated With Home Blood PressureWith adjustments applied for cohort, sex, age, body mass index, smoking and drinking, serum cholesterol, and history


Dow

nloaded from


Aparicio et al Home Blood Pressure Monitoring in Octogenarians 867

of cardiovascular disease and diabetes mellitus, the risk across fifths of the distribution of blood pressure versus the average risk in all untreated octogenarians (Figure 1) reached statistical sig-nificance in the top fifth (≥152.4 mm Hg) of systolic pressure for cardiovascular mortality and for all fatal plus nonfatal cardiovas-cular events combined. The HRs were 2.19 (95% CI, 1.04–4.64; P=0.040) and 2.09 (95% CI, 1.11–3.91; P=0.022), respectively.

For diastolic pressure, the HRs were significant in the lower fifth (≤65.1 mm Hg) for cardiovascular mortality (2.66; 95% CI, 1.39–5.10; P=0.003) and for fatal plus nonfatal cardiovascular events (1.96; 95% CI, 1.10–3.47; P=0.022), and in the upper fifth (≥82.0 mm Hg) for cardiovascular mortality (0.37; 95% CI, 0.15–0.93; P=0.034). The association between outcome and blood pressure in untreated octogenarians did not deviate from log-linearity (P≥0.051). Table 2 shows the standardized HRs associated with blood pressure in the untreated population. The incidence of all coronary events increased with lower baseline diastolic pressure. After further adjustment for systolic pressure, diastolic pressure predicted cardiovascular mortality (HR, 0.54; 95% CI, 0.33–0.87; P=0.011), all fatal plus nonfatal cardiovas-cular events (0.65; 95% CI, 0.43–0.98; P=0.039), all cardiac events (0.48; 95% CI, 0.26–0.87; P=0.016), and all coronary events (0.37; 95% CI, 0.16–0.82; P=0.015). The increase in R2 by adding diastolic pressure to the model already including the other risk factors ranged from 1.54% to 2.36% for cardiovascu-lar mortality and cardiac events, respectively (Table 3). Systolic pressure added only from 0.01% to 0.21% on top of the basic model, not reaching significance.

Figure 2 shows the 5-year risk of cardiovascular events associated with systolic or diastolic pressure analyzed as continuous variables and across percentiles (10th, 50th, and 90th) of the alternative component of home blood pressure in untreated octogenarians. The association was significant for diastolic blood pressure (P=0.039). The findings were similar for cardiovascular mortality (Figure S1).

Analysis of Treated Participants

Incidence of End PointsAmong 173 untreated octogenarians, median follow-up was 5.3 years (5th to 95th percentile interval, 1.2–11.8 years). Over 900 person-years of follow-up, 52 participants died (57.8 per

Table 1. Baseline Characteristics of Participants

CharacteristicUntreated

ParticipantsTreated

Participants

n 202 173

Number with characteristic

Japanese 181 (89.6) 100 (57.8)†

Women 118 (58.4) 108 (62.4)

Current smoking 18 (9.0) 13 (8.9)

Drinking alcohol 27 (23.1) 24 (25.8)

Diabetes mellitus 11 (5.5) 21 (12.1)*

Cardiovascular disease 38 (18.8) 68 (39.3)†

Mean of characteristic (±SD)

Age, y 83.1±4.5 82.6±3.9*

Body mass index, kg/m2 23.0±3.0 25.2±4.2†

Office blood pressure, mm Hg

Systolic 144.0±21.4 145.8±20.9

Diastolic 78.2±12.2 77.5±11.9

Home blood pressure, mm Hg

Systolic 138.2±18.7 143.0±18.1*

Diastolic 74.1±10.4 73.8±9.5

Home heart rate, bpm 69.0±8.9 68.4±10.0

Total cholesterol, mmol/L 4.9±1.0 5.1±0.8

Significance of the difference between untreated and treated participants: *P≤0.05 and †P≤0.0001.

12 (30) 10 (25) 7 (17) 8 (20) 9 (22) 13 (33) 12 (30) 10 (24) 10 (25) 13 (31)

N°. of events (%) 22 (55) 23 (58) 22 (54) 15 (38) 21 (51)

15 (38) 11 (28) 9 (22) 5 (13) 6 (15) 16 (40) 13 (33) 12 (29) 7 (18) 10 (24)

26 (65) 22 (55) 21 (51) 14 (35) 20 (49)

110 120 130 140 150 160 170

0.5

1.0

1.5

2.0

2.5

3.0

2.19 (1.04-4.64) P=0.040

2.09 (1.11-3.91) P=0.022

Haz

ard

Ratio

CV mortalityCV events

Total mortality

Systolic Pressure (mm Hg)

0

Diastolic Pressure (mm Hg)

2.66 (1.39-5.10) P=0.003

1.96 (1.10-3.47) P=0.022

0.37 (0.15-0.93)P=0.034

60 65 70 75 80 85 90 95

A B

Figure 1. Multivariable-adjusted hazard ratios by fifths of the distributions of the systolic (A) and diastolic (B) home blood pressure in 202 untreated octogenarians. The hazard ratios express the risk in fifths compared with average risk in all untreated participants and were adjusted for cohort, sex, age, body mass index, smoking and drinking, serum cholesterol, and history of cardiovascular disease and diabetes mellitus. Point estimates, 95% confidence interval and P value are provided for hazard ratios that significantly deviated from the reference. CV indicates cardiovascular.


Dow

nloaded from



1000 person-years) and 46 experienced a fatal or nonfatal car-diovascular complication (53.9 per 1000 person-years). The cause of death was cardiovascular in 30 participants, noncar-diovascular in 20, end-stage renal disease in 1, and unknown

in 1. Considering cause-specific first cardiovascular events, the incidence of stroke and cardiac events amounted to 23 (26.4 per 1000 person-years) and 25 (28.3 per 1000 person-years), respectively.

Table 3. Generalized R 2-Statistics for Home Blood Pressure as Predictor by Treatment Status

Group Model

Mortality Fatal Plus Nonfatal End Points

Total Cardiovascular Cardiovascular Cardiac

LR P R 2 LR P R 2 LR P R 2 LR P R 2

Untreated (n=202)

Basic 57.5 … 24.8 42.0 … 18.8 31.8 … 14.6 20.4 … 9.60

+SBP 0.03 0.86 0.01 0.03 0.87 0.01 0.15 0.70 0.06 0.48 0.49 0.21

+DBP 0.57 0.45 0.21 3.85 0.049 1.54 1.49 0.22 0.63 5.35 0.02 2.36

+SBP and DBP

1.86 0.39 0.69 6.47 0.04 2.56 4.32 0.12 1.80 6.51 0.04 2.87

Treated (n=173)

Basic 24.0 … 13.0 29.3 … 15.6 19.9 … 10.9 25.3 … 13.6

+SBP 5.90 0.02 2.92 2.19 0.14 1.06 6.82 0.03 3.45 8.01 0.005 3.91

+DBP 0.50 0.48 0.25 0.60 0.44 0.29 0.03 0.86 0.02 1.62 0.20 0.81

+SBP and DBP

6.04 0.049 2.99 2.20 0.33 1.07 6.84 0.08 3.45 8.01 0.02 3.91

DBP indicates diastolic home blood pressure; LR, likelihood ratio; R2, generalized R2 statistic expressed in percent; and SBP, systolic home blood pressure. The basic model included cohort, sex, age, body mass index, smoking and drinking, serum cholesterol, and history of cardiovascular disease and diabetes mellitus. P values are for the improvement of the fit versus the basic model. Values are likelihood ratios and associated generalized R2-statistics for adding systolic home blood pressure and diastolic home blood pressure to the basic model. For the basic model LR and R2 are for the comparison between the basic model and the null model.

Table 2. Standardized Hazard Ratios Relating Outcome to Home Blood Pressure

Hazard Ratio (95% CI)

End Point

Untreated Participants (n=202) Treated Participants (n=173)

Model No (E)Systolic Home BP

(18.7 mm Hg)Diastolic Home BP

(10.4 mm Hg) Model No (E)Systolic Home BP

(18.1 mm Hg)Diastolic Home BP

(9.5 mm Hg)

Mortality

Total A 103 1.02 (0.83–1.26) 0.92 (0.74–1.15) A 52 0.62 (0.42–0.92)* 0.89 (0.63–1.24)

FA 1.20 (0.88–1.64) 0.80 (0.58–1.11) FA 0.60 (0.39–0.93)* 1.08 (0.74–1.57)

Cardiovascular A 46 0.97 (0.70–1.35) 0.70 (0.48–1.01) A 30 0.66 (0.37–1.16) 0.83 (0.51–1.34)

FA 1.44 (0.93–2.23) 0.54 (0.33–0.87)* FA 0.67 (0.36–1.26) 0.96 (0.57–1.64)

Fatal and nonfatal events

Cardiovascular A 58 1.06 (0.80–1.41) 0.82 (0.60–1.13) A 46 … 0.97 (0.68–1.38)

FA 1.40 (0.95–2.04) 0.65 (0.43–0.98)* FA … 1.03 (0.69–1.53)

Cardiac A 30 0.99 (0.66–1.50) 0.65 (0.40–1.05) A 25 0.51 (0.29–0.89)* 0.75 (0.43–1.32)

FA 1.56 (0.91–2.67) 0.48 (0.26–0.87)* FA 0.50 (0.27–0.93)* 1.02 (0.55–1.90)

Coronary A 18 0.81 (0.45–1.48) 0.47 (0.24–0.93)* A 18 0.41 (0.21–0.80)† 0.65 (0.33–1.27)

FA 1.52 (0.71–3.25) 0.37 (0.16–0.82)* FA 0.40 (0.19–0.85)* 1.02 (0.49–2.12)

Stroke A 28 1.06 (0.71–1.60) 0.97 (0.64–1.48) A 23 1.36 (0.79–2.34) 1.04 (0.67–1.61)

FA 1.17 (0.67–2.03) 0.87 (0.49–1.54) FA 1.47 (0.79–2.72) 0.89 (0.54–1.47)

Hazard ratios express the risk associated with a 1-SD increase in blood pressure (given between parentheses in column headings). All models were adjusted for cohort, sex, age, body mass index, smoking and drinking, serum cholesterol, and history of cardiovascular disease and diabetes mellitus. Adjusted models include either systolic pressure or diastolic pressure, while fully adjusted models include both systolic and diastolic blood pressure in addition to other covariables. In treated participants, the relation of cardiovascular events with systolic pressure was curvilinear and is reported separately in the Results section of this article and Figure 3. A indicates adjusted; BP, blood pressure; CI, confidence interval; FA, fully adjusted; and No (E), number of events.

Significance of the hazard ratios: *P<0.05 and †P<0.01.


Dow

nloaded from



Risk Associated With Home Blood PressureThe analyses were adjusted as in the untreated participants. The risk of all-cause mortality versus the average risk in all treated participants was higher in the lower fifth of the distri-bution of systolic pressure (<126.9 mm Hg; HR, 2.09; 95% CI, 1.10–3.98; P=0.020; Figure S2). Table 2 shows the stan-dardized HRs associated with blood pressure in the treated population. The risk of all-cause mortality, cardiac, and coronary events significantly increased with lower baseline systolic home blood pressure. After further adjustment for diastolic blood pressure, lower systolic pressure predicted

all-cause mortality (HR, 0.60; 95% CI, 0.39–0.93; P=0.022), cardiac events (0.50; 95% CI, 0.27–0.93; P=0.027), and coro-nary events (0.40; 95% CI, 0.19–0.85; P=0.017). According to the Kolmogorov-type supremum test, the log-linear model was appropriate for all end points. However, in relation to all fatal and nonfatal cardiovascular events, the fit of the Cox model for systolic pressure significantly improved (P<0.008) by adding a quadratic term to a model already including the linear term. As shown in Figure 3, in treated participants, the relationship between all cardiovascular events and systolic blood pressure was U shaped with the nadir at 148.6 mm Hg.

110 120 130 140 150 160

PS=0.088PD=0.039

Systolic Pressure (mm Hg)

5-ye

ar ri

sk (%

)

87 mm Hg 73 mm Hg 62 mm Hg

Diastolic Pressure (mm Hg)


60 70 80 90

10

20

30

0

40 BA

Figure 2. Five-year risk of a cardiovascular event associated with systolic (A) or diastolic (B) home blood pressure analyzed as continuous variables and across percentiles (10th, 50th, and 90th) of the alternative component of blood pressure in 202 untreated octogenarians. The risk was standardized to the distribution in the whole untreated population (ratio or mean) of cohort, sex, age, body mass index, smoking and drinking, serum cholesterol, and history of cardiovascular disease and diabetes mellitus. PS and PD indicate the significance of the association with systolic and diastolic home blood pressure, respectively.

5-ye

ar ri

sk (%

)

Diastolic Pressure (mm Hg)Systolic Pressure (mm Hg)

PS=0.008PD=0.90


60 70 80 90


10

20

30

25

15

120 130 140 150 160 170

BA

Figure 3. Five-year risk of a cardiovascular event associated with systolic (A) or diastolic (B) home blood pressure analyzed as continuous variables and across percentiles (10th, 50th, and 90th) of the alternative component of blood pressure in 173 treated octogenarians. The risk was standardized to the distribution (ratio or mean) in the whole treated population of cohort, sex, age, body mass index, smoking and drinking, serum cholesterol, and history of cardiovascular disease and diabetes mellitus. The relation with systolic blood pressure was U-shaped with nadir at 148.6 mm Hg).


Dow

nloaded from



Decreasing the nadir by 1 SD (18.1 mm Hg), increased the risk for a cardiovascular event by 45% (1% to 108%; P=0.046) in the adjusted model and by 45% (−3% to 116%; P=0.071) in the fully adjusted model. The corresponding estimates for a 1-SD increment in systolic blood pressure above the nadir were 44% (−12% to 103%; P=0.092) and 44% (−15% to 104%; P=0.112), respectively.

The increase in the generalized R2 by adding systolic blood pressure to the model already including the other risk factors ranged from 2.92% to 3.91% for all-cause mortality and cardiovascular and cardiac events (Table 3). Diastolic blood pressure added only from 0.02% to 0.81% on top of the basic model, not reaching significance.

Sensitivity AnalysesWe checked the consistency of our results for all-cause and cardiovascular mortality (Table S5) and for the combination of fatal and nonfatal cardiovascular events (Table S6) accord-ing to sex and ethnicity. For untreated (0.083≤P≤0.68) and treated (0.096≤P≤0.96) participants, the HRs were not statis-tically different across strata. Tables S7 and S8 show the HRs obtained after excluding one cohort at a time for untreated and treated participants, respectively. Statistical significance was mainly driven by the Ohasama cohort with the largest number of events. However, all HRs were within the 95% CI for all available participants. Thus, no cohort had an excessive influ-ence on the estimate based on all participants combined.

Finally, in untreated and treated participants, ANCOVA showed that the regression lines relating office with home blood pressure were coincident (P values for differences in slopes and intercepts ≥0.10). In all participants (Figure 4), a home blood pressure of 127 mm Hg systolic and 65 mm Hg diastolic corresponded with office blood pressures of 136.3 mm Hg (95% CI, 134.0–138.6 mm Hg) and 72.0 mm Hg (95% CI, 70.6–73.5 mm Hg), respectively.

DiscussionCurrent guidelines for diagnosis and management of hyper-tension provide either no special recommendations for the very elderly27,28 or recommendations exclusively based on office blood pressure.12–15 Our current study provides pre-liminary thresholds for the self-measured blood pressure in risk stratification among the very elderly. In untreated partici-pants, a home blood pressure >150 mm Hg (rounded) or <65 mm Hg (rounded) conferred an increased risk of a cardiovas-cular death or a composite cardiovascular end point, whereas a diastolic home pressure >80 mm Hg was associated with low cardiovascular mortality. These findings are in line with the vast epidemiological knowledge database showing that in the elderly a wide pulse pressure, an index of arterial stiffness, is associated with a dire prognosis.29 In treated octogenarians, there was a curvilinear association between cardiovascular risk and systolic home blood pressure, independent from dia-stolic pressure, with nadir at ≈150 mm Hg. The risk of death or a cardiovascular event increased below a systolic home pres-sure of ≈130 mm Hg.

Only one prior study in the very elderly, PARTAGE, addressed the association of cardiovascular risk with the self-measured blood pressure in the very elderly.18,19 The study included 1127 frail nursing home residents (78.1% women; mean age, 87.6 years). Among 227 PARTAGE participants with low home systolic pressure (<130 mm Hg) on combi-nation antihypertensive treatment (≥2 drugs) compared with the 900 others, home blood pressure averaged 119/65 mm Hg versus 142/75 mm Hg with mortality rates of 32.2% versus 19.7%, respectively. The HRs for all-cause mortality and car-diovascular events associated with low blood pressure were 1.36 (95% CI, 1.06–1.75) and 1.28 (95% CI, 0.99–1.65).19 In another analysis of the same cohort,18 patients with a dia-stolic home blood pressure ranging from 49.3 to 68.5 mm Hg

40

60

80

100

120

Offi

ce D

iast

olic

Pre

ssur

e (m

m H

g)

Home Diastolic Pressure (mm Hg)

50 60 70 80 90 100 110 120

140

Home Systolic Pressure (mm Hg)

70

90

110

130

150

170

190

110 130 150 170 190

Offi

ce S

ysto

lic P

ress

ure

(mm

Hg)

210

90

230

100 120 140 160 180 200

BA

Figure 4. Office systolic (A) and diastolic (B) pressure plotted as function of the corresponding home blood pressure. Home levels of 127 mm Hg systolic and 65 mm Hg diastolic corresponded with office values of 136.3 mm Hg (95% confidence interval, 134.0–138.6 mm Hg) and 72.0 mm Hg (70.6–73.5 mm Hg), respectively.


Dow

nloaded from



had a 2-year shorter survival than those with higher levels (P=0.021).

Other cohort studies in the very elderly addressed the risk associated with office blood pressure.30,31 In Tampere, Finland, 561 octogenarians (82% women; mean age, 88.4 years) were stratified according to systolic (from <120 to >200 mm Hg) or diastolic (from <70 to >110 mm Hg) blood pressure and fol-lowed up for 5 years.30 The strong point of this study was that the population was representative of both fit and frail individu-als because they were living either at their home (n=272) or were institutionalized or had been hospitalized (n=289). The key finding was that mortality was highest in the low blood pressure group (<120/70 mm Hg) and lowest in subjects with office blood pressure levels in excess of 160/90 mm Hg. A retrospective cohort study enrolled 4071 hypertensive octoge-narians at 10 US Department of Veterans Affairs sites (3.4% women; mean age, 82.6 years).31 Follow-up spanned 5 years. The association between survival and treated blood pressure was U shaped. Lowering blood pressure <140 mm Hg systolic or 90 mm Hg diastolic was associated with shorter survival. The HRs for a 10-mm Hg increase in blood pressure were 0.84 (95% CI, 0.78–0.89) systolic and 0.91 (95% CI, 0.87–0.96) diastolic, suggesting that aggressive treatment might be harm-ful at very old age.31

The HYVET trial,11 enrolled 3845 physically fit patients (60.5% women; mean age, 83.6 years). The blood pressure at enrolment averaged 173/91 mm Hg. Eligible patients were aged ≥80 years and had a sustained systolic pressure of ≥160 mm Hg. They were randomized to receive either indapamide or matching placebo with the possible addition of perindo-pril or matching placebo to achieve a target blood pressure of 150/80 mm Hg.11 At median follow-up (1.8 years), blood pres-sure was 15.0/6.1 mm Hg lower in the active-treatment group than in the placebo. In the intention-to-treat analysis,11 reduc-tions in the rates were 30% for fatal or nonfatal stroke (95% CI, −1 to 51), 39% for fatal stroke (95% CI, 1–62), 21% for total mortality (95% CI, 4–35), 23% for cardiovascular mor-tality (95% CI, −1 to 40), and 64% for heart failure (95% CI, 42–78). HYVET11 did not provide information on whether a further reduction of office blood pressure would be beneficial and was not representative of the oldest old in whom physical or cognitive disabilities are common.

Current recommendations for the target office blood pres-sure in very elderly hypertensive patients are ambiguous. The American College of Cardiology Foundation/American Heart Association expert consensus document12 suggests an office systolic pressure of 140 to 145 mm Hg if tolerated. It states that levels <130/65 mm Hg should be avoided. The European guidelines14 recommend that those with an office systolic of ≥160 mm Hg should be considered for treatment if consid-ered fit. The target systolic pressure should be <140 mm Hg, or between 140 and 150 mm Hg, according to the level of fit-ness. Both the Canadian13 and International15 guidelines rec-ommend systolic office pressure levels <150 mm Hg without mentioning a diastolic threshold. All guidelines, including one especially focusing on the home blood pressure,32 suggest a self-measured threshold of 135 mm Hg systolic or 85 mm Hg diastolic irrespective of age.

Strong points of our current report are the involvement of old individuals living in 4 different countries, a follow-up of >5 years, and the high albeit variable number of home measurements with medians ranging across cohorts from 10 to 48. On the contrary, our study has some potential limita-tions. First, the sample size was relatively small, but very old people only represent 1.5% of the population and are difficult to recruit.2 Second, one should be cautious in extrapolating our current findings to other ethnic groups, especially blacks, to populations in which the prevalence of obesity is rampant, or even to the general Japanese or European population. Indeed, our analyses comprised mainly slender Japanese and whites with body mass index averaging 23.0 and 27.1 kg/m2, respectively. Finally, methods of ascertaining events and mea-surements were not identical among cohorts. However, we adjusted for cohort in the Cox proportional hazards models, and our findings remained consistent when we excluded one cohort at a time.

PerspectivesNotwithstanding potential limitations, our current study might have implications for clinical practice. We provide for the first time a reference frame for home blood pressure that may prevent overtreatment by avoiding decreasing home blood pressures <125 to 130 mm Hg. In untreated patients, physi-cians should be aware that home diastolic pressure levels <65 mm Hg are associated with a worse cardiovascular prognosis, whereas values above ≈80 mm Hg predict a better outcome. These results do not contradict current recommendations for the management of office blood pressure in the very elderly. However, they differ from the recommendations for the man-agement of home blood pressure that suggest the same thresh-old irrespective of age.32 A rounded home blood pressure of 125 mm Hg systolic or 65 mm Hg diastolic corresponded with office blood pressures of ≈135 systolic and 70 mm Hg dia-stolic. Our study, therefore, provides indirect information on the thresholds for office blood pressure that might be optimal in the very elderly.

AcknowledgmentsAnnick De Soete and Renilde Wolfs (Leuven) and Silvia Yeha and Ayelen Margarella (Buenos Aires) provided expert clerical assistance.

Sources of FundingThe European Union (HEALTH-F7-2011–278249 EU-MASCARA, HEALTH-F7-305507 HOMAGE, and the European Research Council Advanced Research Grant 294713 EPLORE) and the Fonds voor Wetenschappelijk Onderzoek Vlaanderen, Ministry of the Flemish Community, Brussels, Belgium (G.0881.13 and G.088013 N) supported the Studies Coordinating Centre (Leuven, Belgium). The Ohasama study was supported by Grants for Scientific Research (23249036, 23390171, 24390084, 24591060, 24790654, 25253059, 25461083, 25461205, 25860156, 26282200, and 26860093) from the Ministry of Education, Culture, Sports, Science, and Technology, Japan; a Grant-in-Aid from the Japan Society for the Promotion of Science (JSPS) fellows (25.7756, 25.9328, and 26.857); the Japan Arteriosclerosis Prevention Fund; Intramural Research Fund (22-4-5) for Cardiovascular Diseases of National Cerebral and Cardiovascular Center; and a Health Labor Sciences Research Grant (H26-Junkankitou [Seisaku]-Ippan-001) from the Ministry of Health, Labour, and Welfare; and A Scheme to Revitalize Agriculture and


Dow

nloaded from



Fisheries in Disaster Area through Deploying Highly Advanced Technology (NouEi 2–02) from the Ministry of Agriculture, Forestry and Fisheries, Japan. The Tsurugaya study was supported by a Health Sciences Research Grant for Health Service (H21-Choju-Ippan-001) from the Ministry of Health, Labor and Welfare, Japan, and the Japan Arteriosclerosis Prevention Fund. The Department of Internal Medicine of the Hospital Italiano de Buenos Aires (Capital Federal, Argentina) supported the study of home blood pressure measurement in Buenos Aires.

DisclosuresNone.

References 1. Lewington S, Clarke R, Qizilbash N, Peto R, Collins R; Prospective

Studies Collaboration. Age-specific relevance of usual blood pressure to vascular mortality: a meta-analysis of individual data for one million adults in 61 prospective studies. Lancet. 2002;360:1903–1913.

2. World Population Prospects. The 2012 revision. Key findings and advance tables. Working Paper No. ESA/P/WP.227. 2013. http://esa.un.org/wpp/Documentation/pdf/WPP2012_%20KEY%20FINDINGS.pdf. Accessed July 3, 2015.

3. Lim SS, Vos T, Flaxman AD, et al. A comparative risk assessment of bur-den of disease and injury attributable to 67 risk factors and risk factor clus-ters in 21 regions, 1990-2010: a systematic analysis for the Global Burden of Disease Study 2010. Lancet. 2012;380:2224–2260. doi: 10.1016/S0140-6736(12)61766-8.

4. Coope J, Warrender TS. Randomised trial of treatment of hypertension in elderly patients in primary care. BMJ. (Clin Res Ed). 1986;293:1145–1151.

5. Amery A, Birkenhäger W, Brixko P, Bulpitt C, Clement D, Deruyttere M, De Schaepdryver A, Dollery C, Fagard R, Forette F. Mortality and mor-bidity results from the European Working Party on High Blood Pressure in the Elderly trial. Lancet. 1985;1:1349–1354.

6. Dahlöf B, Lindholm LH, Hansson L, Scherstén B, Ekbom T, Wester PO. Morbidity and mortality in the Swedish Trial in Old Patients with Hypertension (STOP-Hypertension). Lancet. 1991;338:1281–1285.

7. SHEP Cooperative Research Group. Prevention of stroke by antihyperten-sive drug treatment in older persons with isolated systolic hypertension. Final results of the Systolic Hypertension in the Elderly Program (SHEP). JAMA. 1991;265:3255–3264.

8. Staessen JA, Fagard R, Thijs L, et al. Randomised double-blind compari-son of placebo and active treatment for older patients with isolated sys-tolic hypertension. The Systolic Hypertension in Europe (Syst-Eur) Trial Investigators. Lancet. 1997;350:757–764.

9. Hansson L, Lindholm LH, Ekbom T, Dahlöf B, Lanke J, Scherstén B, Wester PO, Hedner T, de Faire U. Randomised trial of old and new antihy-pertensive drugs in elderly patients: cardiovascular mortality and morbid-ity the Swedish Trial in Old Patients with Hypertension-2 study. Lancet. 1999;354:1751–1756.

10. Liu L, Wang JG, Gong L, Liu G, Staessen JA. Comparison of active treatment and placebo in older Chinese patients with isolated systolic hypertension. Systolic Hypertension in China (Syst-China) Collaborative Group. J Hypertens. 1998;16(12 Pt 1):1823–1829.

11. Beckett NS, Peters R, Fletcher AE, et al; HYVET Study Group. Treatment of hypertension in patients 80 years of age or older. N Engl J Med. 2008;358:1887–1898. doi: 10.1056/NEJMoa0801369.

12. Aronow WS, Fleg JL, Pepine CJ, et al; ACCF Task Force. ACCF/AHA 2011 expert consensus document on hypertension in the elderly: a report of the American College of Cardiology Foundation Task Force on Clinical Expert Consensus Documents. Circulation. 2011;123:2434–2506. doi: 10.1161/CIR.0b013e31821daaf6.

13. Canadian Hypertension Education Program. The 2014 Canadian Hypertension Education Program recommendations for blood pressure measurement, diagnosis, assessment of risk, prevention, and treatment of hypertension. Can J Cardiol. 2014;30:485–501.

14. Mancia G, Fagard R, Narkiewicz K, Redon J, Zanchetti A, Böhm M, Christiaens T, Cifkova R, De Backer G, Dominiczak A, Galderisi M, Grobbee DE, Jaarsma T, Kirchhof P, Kjeldsen SE, et al: The Task Force for the management of arterial hypertension of the European Society of Hypertension (ESH) and of the European Society of Cardiology (ESC). 2013 ESH/ESC Guidelines for the Management of Arterial Hypertension: Eur Heart J. 2013;34:2159–2219.

15. Weber MA, Schiffrin EL, White WB, et al. Clinical practice guidelines for the management of hypertension in the community: a statement by

the American Society of Hypertension and the International Society of Hypertension. J Clin Hypertens (Greenwich). 2014;16:14–26. doi: 10.1111/jch.12237.

16. Staessen JA, Thijs L, Ohkubo T, Kikuya M, Richart T, Boggia J, Adiyaman A, Dechering DG, Kuznetsova T, Thien T, de Leeuw P, Imai Y, O’Brien E, Parati G. Thirty years of research on diagnostic and therapeutic thresh-olds for the self-measured blood pressure at home. Blood Press Monit. 2008;13:352–365. doi: 10.1097/MBP.0b013e3283108f93.

17. Imai Y, Kario K, Shimada K, Kawano Y, Hasebe N, Matsuura H, Tsuchihashi T, Ohkubo T, Kuwajima I, Miyakawa M; Japanese Society of Hypertension Committee for Guidelines for Self-monitoring of Blood Pressure at Home. The Japanese Society of Hypertension Guidelines for Self-monitoring of Blood Pressure at Home (Second Edition). Hypertens Res. 2012;35:777–795. doi: 10.1038/hr.2012.56.

18. 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;60:1503–1511. doi: 10.1016/j.jacc.2012.04.055.

19. Benetos A, Labat C, Rossignol P, Fay R, Rolland Y, Valbusa F, Salvi P, Zamboni M, Manckoundia P, Hanon O, Gautier S. Treatment With Multiple Blood Pressure Medications, Achieved Blood Pressure, and Mortality in Older Nursing Home Residents: The PARTAGE Study. JAMA Intern Med. 2015;175:989–995. doi: 10.1001/jamainternmed.2014.8012.

20. Niiranen TJ, Asayama K, Thijs L, Johansson JK, Ohkubo T, Kikuya M, Boggia J, Hozawa A, Sandoya E, Stergiou GS, Tsuji I, Jula AM, Imai Y, Staessen JA; International Database of Home blood pressure in rela-tion to Cardiovascular Outcome Investigators. Outcome-driven thresholds for home blood pressure measurement: international database of home blood pressure in relation to cardiovascular outcome. Hypertension. 2013;61:27–34. doi: 10.1161/HYPERTENSIONAHA.111.00100.

21. Ohkubo T, Imai Y, Tsuji I, Nagai K, Kato J, Kikuchi N, Nishiyama A, Aihara A, Sekino M, Kikuya M, Ito S, Satoh H, Hisamichi S. Home blood pressure measurement has a stronger predictive power for mortality than does screening blood pressure measurement: a population-based observa-tion in Ohasama, Japan. J Hypertens. 1998;16:971–975.

22. Niu K, Hozawa A, Awata S, Guo H, Kuriyama S, Seki T, Ohmori-Matsuda K, Nakaya N, Ebihara S, Wang Y, Tsuji I, Nagatomi R. Home blood pres-sure is associated with depressive symptoms in an elderly population aged 70 years and over: a population-based, cross-sectional analysis. Hypertens Res. 2008;31:409–416. doi: 10.1291/hypres.31.409.

23. Stergiou GS, Baibas NM, Kalogeropoulos PG. Cardiovascular risk pre-diction based on home blood pressure measurement: the Didima study. J Hypertens. 2007;25:1590–1596.

24. Aparicio LS, Barochiner J, Cuffaro PE, Alfie J, Rada MA, Morales MS, Galarza CR, Marin MJ, Waisman GD. Determinants of the morning-evening home blood pressure difference in treated hypertensives: the HIBA-Home Study. Int J Hypertens. 2014;2014:569259. http://dx.doi.org/10.1155/2014/569259. Accessed July 3, 2015.

25. Kleinbaum DG, Kupper LL, Muller KE. Applied Regression Analysis and Other Multivariate Methods. Boston, MA, PWS-Kent Publishing Company, 1988:41–79.

26. The SAS Institute. Usage note 24441: How can I compute an R-square statistic for models that are fit by PROC PHREG? http://support.sas.com/kb/24–441.html. Accessed July 2, 2015.

27. Chobanian AV, Bakris GL, Black HR, Cushman WC, Green LA, Izzo JL Jr, Jones DW, Materson BJ, Oparil S, Wright JT Jr, Roccella EJ; National Heart, Lung, and Blood Institute Joint National Committee on Prevention, Detection, Evaluation, and Treatment of High Blood Pressure; National High Blood Pressure Education Program Coordinating Committee. The Seventh Report of the Joint National Committee on Prevention, Detection, Evaluation, and Treatment of High Blood Pressure: the JNC 7 report. JAMA. 2003;289:2560–2572. doi: 10.1001/jama.289.19.2560.

28. 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). JAMA. 2014;311:507–520. doi: 10.1001/jama.2013.284427.

29. Staessen J, Amery A, Fagard R. Isolated systolic hypertension in the elderly. J Hypertens. 1990;8:393–405.

30. Mattila K, Haavisto M, Rajala S, Heikinheimo R. Blood pressure and five year survival in the very old. BMJ. (Clin Res Ed). 1988;296:887–889.

31. Oates DJ, Berlowitz DR, Glickman ME, Silliman RA, Borzecki AM. Blood pressure and survival in the oldest old. J Am Geriatr Soc. 2007;55:383–388. doi: 10.1111/j.1532-5415.2007.01069.x.


Dow

nloaded from

http://esa.un.org/wpp/Documentation/pdf/WPP2012_%20KEY%20FINDINGS.pdf

http://esa.un.org/wpp/Documentation/pdf/WPP2012_%20KEY%20FINDINGS.pdf

http://dx.doi.org/10.1155/2014/569259

http://dx.doi.org/10.1155/2014/569259

http://support.sas.com/kb/24 � 441.html

http://support.sas.com/kb/24 � 441.html



32. Parati G, Stergiou GS, Asmar R, et al; ESH Working Group on Blood Pressure Monitoring. European Society of Hypertension guide-lines for blood pressure monitoring at home: a summary report of the

Second International Consensus Conference on Home Blood Pressure Monitoring. J Hypertens. 2008;26:1505–1526. doi: 10.1097/HJH.0b013e328308da66.

What Is New?• In octogenarians, the optimal self-measured blood pressure (BP) has not

yet been established. Guidelines only provide recommendations for office BP based on clinical trials not representative for the general population and with short follow-up. We, therefore, analyzed 375 octogenarians en-rolled in the International Database on HOme blood pressure in relation to Cardiovascular Outcome, who were followed up for 5.5 years (median).

What Is Relevant?• In untreated octogenarians, a home diastolic BP of <65 mm Hg predicted

increased cardiovascular events.• In untreated octogenarians, a home systolic pressure exceeding 150

mm Hg worsened cardiovascular risk.

• In treated octogenarians, a home systolic BP <125 mm Hg was associ-ated with increased total mortality.

• In treated octogenarians, home systolic pressure showed a U-shaped pattern for cardiovascular events with nadir ≈150 mm Hg.

Summary

In untreated octogenarians, low home diastolic pressure should be considered before antihypertensive therapy is initiated. In treated octogenarians, overtreatment should be avoided by not lowering the home systolic pressure to <125 mm Hg.

Novelty and Significance


Dow

nloaded from


Outcome (IDHOCO) Investigatorson behalf of the International Database on Home Blood Pressure in Relation to Cardiovascular

StaessenTsuji, George S. Stergiou, Masahiro Kikuya, Yutaka Imai, Gabriel D. Waisman and Jan A.

Odili, José Alfie, Kei Asayama, Paula E. Cuffaro, Kyoko Nomura, Takayoshi Ohkubo, Ichiro Lucas S. Aparicio, Lutgarde Thijs, José Boggia, Lotte Jacobs, Jessica Barochiner, Augustine N.


Print ISSN: 0194-911X. Online ISSN: 1524-4563 Copyright © 2015 American Heart Association, Inc. All rights reserved.

is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX 75231Hypertension doi: 10.1161/HYPERTENSIONAHA.115.05800

2015;66:865-873; originally published online July 27, 2015;Hypertension.

http://hyper.ahajournals.org/content/66/4/865World Wide Web at:

The online version of this article, along with updated information and services, is located on the

http://hyper.ahajournals.org/content/suppl/2015/07/27/HYPERTENSIONAHA.115.05800.DC1Data Supplement (unedited) at:

http://hyper.ahajournals.org//subscriptions/

is online at: Hypertension Information about subscribing to Subscriptions:

http://www.lww.com/reprints Information about reprints can be found online at: Reprints:

document. Permissions and Rights Question and Answer this process is available in the

click Request Permissions in the middle column of the Web page under Services. Further information aboutOffice. Once the online version of the published article for which permission is being requested is located,

can be obtained via RightsLink, a service of the Copyright Clearance Center, not the EditorialHypertensionin Requests for permissions to reproduce figures, tables, or portions of articles originally publishedPermissions:


Dow

nloaded from

http://hyper.ahajournals.org/content/66/4/865

http://hyper.ahajournals.org/content/suppl/2015/07/27/HYPERTENSIONAHA.115.05800.DC1

http://www.ahajournals.org/site/rights/

http://www.lww.com/reprints

http://hyper.ahajournals.org//subscriptions/


HYPERTENSION. ONLINE SUPPLEMENT


Lucas S. Aparicio, Lutgarde Thijs, José Boggia, Lotte Jacobs, Jessica Barochiner, Au-gustine N. Odili, José Alfie, Kei Asayama, Paula E. Cuffaro, Kyoko Nomura, Takayoshi

Ohkubo, Ichiro Tsuji, George S. Stergiou, Masahiro Kikuya , Yutaka Imai, Gabriel D. Waisman, Jan A. Staessen, on behalf of the International Database on HOme blood

pressure in relation to Cardiovascular Outcome (IDHOCO) Investigators

Correspondence: Jan A. Staessen, MD, PhD, Studies Coordinating Centre, Research Unit Hypertension and Cardiovascular Epidemiology, KU Leuven Department of Cardiovascular Sciences, University of Leuven, Campus Sint Ra-faël, Kapucijnenvoer 35, Box 7001, BE-3000 Leuven, Belgium

Telephone: +32-16-34-7104 (office) +32-15-41-1747 (home) +32-47-632-4928 (mobile) Facsimile: +32-16-34-7106 (office) +32-15-41-4542 (home) email: [email protected] [email protected]



-Page 2 of 16-

Data Supplement

This Data Supplement formed part of the original submission and has been peer reviewed. We posted it as supplied by the authors. Supplement to Defining Thresholds for Home Blood Pres-sure Monitoring in Octogenarians.

Table of Contents

Expanded Methods p 3

References p 5

Table S1. IDHOCO Cohorts Included in Present Analysis p 7

Table S2. Protocols of Office and Home Blood Pressure Measurement p 8

Table S3. Baseline Characteristics of 202 Untreated Participants by Cohort p 9

Table S4. Baseline Characteristics of 173 Treated Participants by Cohort p10

Table S5. Hazard Ratios for Fatal Endpoints in Relation to Home Blood Pressure in Various Strata p11

Table S6. Hazard Ratios for Fatal plus Nonfatal Endpoints in Relation to Home Blood Pressure in Various Strata p12

Table S7. Hazard Ratios Excluding One Cohort at a Time in 202 Untreated Participants p13

Table S8. Hazard Ratios Excluding One Cohort at a Time in 173 Treated Participants p14

Fig. S1. Five-year risk of cardiovascular mortality associated with systolic or diastolic home blood pressure p15

Fig. S2. Multivariable-adjusted hazard ratios for total mortality by fifths of the distribution of systolic home blood pressure in 173 treated octogenarians. p16

-Page 3 of 16-

Expanded Methods

Study Population

At the time of writing this report, IDHOCO1,2 included 8912 participants recruited in the framework of six prospective population studies3-8 and one cohort study mainly including hypertensive patients9 (Table S1). All studies received ethical approval and have been de-scribed in peer-reviewed publications. Hospitalized, institutionalized or incapacitated pa-tients were excluded from recruitment. We discarded one cohort because data on cause-specific mortality was still being collected8 and two cohorts because they did not include oc-togenarians.4,7 Among the remaining 4704 participants, 375 were aged 80 years or over, had at least two home blood pressure readings, and known status for antihypertensive treatment. Therefore, the number of analyzed participants totaled 375, of whom at enrolment 202 were untreated and 173 were taking antihypertensive drugs. They included 144 (un-treated/treated, 106/38) residents of Ohasama, Japan;3 137 (75/62) inhabitants of the Tsuru-gaya district in Sendai, Japan;6 36 (21/15) inhabitants of Didima, Greece;5 and 58 treated outpatients of a general hospital in Buenos Aires, Argentina.9

Blood Pressure Measurement Office blood pressure was the average of two readings obtained with a standard mercury sphygmomanometer,5 or with an automated auscultatory3,5,9 or oscillometric device,6,9 using the appropriate cuff size, with the participants resting in the sitting3 or supine6 position. In each cohort, researchers implemented a quality assurance program for home blood pressure monitoring. Doctors,3,6 nurses3 or specially trained technicians9 first explained the rationale of blood pressure self-monitoring. Participants obtained the home blood pressure readings on the left arm after they had rested for 1 to 5 minutes in the sitting position. The automated or semi-automated devices used for office10-12 and home11,13,14 blood pressure measure-ment had all been validated. Further details on the devices and the procedures for office and home blood pressure measurements appear in Table S2. In all cohorts morning measure-ments were obtained within 1 hour of waking up, before breakfast, and if applicable, before the intake of medications. The average of all available measurements at home, at least two, was used in the analysis. Throughout this manuscript, unless specified otherwise, blood pressure refers to the self-measured home blood pressure.

Other Measurements We obtained information on each participant’s medical history, use of medications, and smoking and drinking habits by means of questionnaires3,5,6 or checking hospital files.9 Body mass index was body weight in kilograms divided by height in meters squared. Serum cholesterol and blood glucose were measured by automated enzymatic methods. As de-scribed previously,1 information on serum total cholesterol level was unavailable for the Didima population and was extrapolated from data provided by the Attica study investigators by sex and 10-year age strata.15 The Attica study population was a large population cohort examined at the same time (2001-2002) and in the same geographical area as the Didima cohort. Diabetes mellitus was the use of antidiabetic drugs, a self-reported diagnosis or a fasting or random blood glucose concentration of at least 7.0 mmol/L (126 mg/dL)16 or 11.1 mmol/L (200 mg/dL),16 respectively.

-Page 4 of 16-

Ascertainment of Events We ascertained vital status and the incidence of fatal and nonfatal diseases from the appro-priate sources in each country (Table S1), as described in previous publications.1,2 Fatal and nonfatal stroke did not include transient ischemic attacks. Cardiac events comprised fatal and nonfatal myocardial infarction, death due to ischemic heart disease or other cardiac causes, sudden death, fatal and nonfatal heart failure, surgical or percutaneous coronary revascularization, and pacemaker implantation. The composite cardiovascular endpoint in-cluded cardiovascular mortality and the aforementioned cardiovascular endpoints. In out-come analyses of composite cardiovascular endpoints, we only considered the first event in each category.

Statistical Analysis For database management and statistical analysis, we used SAS software, version 9.3 (SAS Institute Inc., Cary, NC, USA). For comparison of means and proportions, we applied the

large sample z-test or analysis of variance and the 2 statistic, respectively. We compared intercepts and slopes of regression lines between untreated and treated participants by anal-ysis of covariance.17 Statistical significance was a α-level of less than 0.05 on two-sided tests.

We applied the deviation from mean coding to compute hazard ratios in fifths of the distri-bution of the home blood pressure. This approach expresses the risk in each fifth relative to the overall risk in the whole study population and allows computing 95% confidence intervals for the hazard ratios in all fifths without defining an arbitrary reference group. Next, we used Cox proportional hazards models to compute hazard ratios expressing the risk associated with blood pressure analyzed as continuous variable, while adjusting for cohort , sex, age, body mass index, smoking, drinking alcohol, total serum cholesterol, and a history of cardio-vascular disease and diabetes mellitus. Fully adjusted models included both systolic and diastolic blood pressures. We checked the proportional hazards assumption and the func-tional forms of the covariables by the Kolmogorov-type supremum test, as implemented in the ASSESS statement of the PROC PHREG procedure of the SAS package. We tested heterogeneity in the hazard ratios across subgroups by introducing the appropriate interac-tion term in the Cox model. We applied the generalized R2 statistic18 to assess the risks additionally explained by home blood pressure over and beyond other risk factors.

-Page 5 of 16-

References

1. Niiranen TJ, Asayama K, Thijs L, Johansson JK, Ohkubo T, Kikuya M, Boggia J, Hozawa A, Sandoya E, Stergiou GS, Tsuji I, Jula AM, Imai Y, Staessen JA, for the Database of HOme blood pressure in relation to Cardiovascular Outcome Investigators. Outcome-driven thresholds for home blood pressure Measurement: International Database for HOme blood pressure in relation to Cardiovascular Outcome. Hypertension. 2013;61:27–34.

2. Asayama K, Thijs L, Brguljan Hitij J, Niiranen TJ, Hozawa A, Boggia J, Aparicio LS, Hara A, Johansson JK, Ohkubo T, Tzourio C, Stergiou GS, Sandoya E, Tsuji I, Jula AM, Imai Y, Staessen JA, on behalf of the International Database on HOme blood pressure in relation to Cardiovascular Outcome (IDHOCO) Investigators. Risk stratification by self-measured home blood pressure across categories of the conventional blood pressure: a participant-level meta-analysis. PLOS Medicine. 2014;11:e1001591.

3. Ohkubo T, Imai Y, Tsuji I, Nagai K, Kato J, Kikuchi N, Nishiyama A, Aihara A, Sekino M, Kikuya M, Ito S, Satoh H, Hisamichi S. Home blood pressure measurement has a stronger predictive power for mortality than does screening blood pressure measurement: a population-based observation in Ohasama, Japan. J Hypertens. 1998;16:971–975.

4. Schettini C, Bianchi M, Nieto F, Sandoya E, Senra H, Hypertension Working Group. Ambulatory blood pressure. Normality and comparison with other measurements. Hypertension. 1999;34:818–825.

5. Stergiou GS, Baibas NM, Kalogeropoulos PG. Cardiovascular risk prediction on home blood pressure measurement: the Didima Study. J Hypertens. 2007;25:1590–1596.

6. Niu K, Hozawa A, Awata S, Guo H, Kuriyama S, Seki T, Ohmori-Matsuda K, Nakaya N, Ebihara S, Wang Y, Tsuji I, Nagatomi R. Home blood pressure is associated with depressive symptoms in an elderly population aged 70 years and over: a population-based, cross-sectional analysis. Hypertens Res. 2008;31:409–416.

7. Niiranen TJ, Johansson JK, Reunanen A, Jula AM. Optimal schedule for home blood pressure measurement based on prognostic data: the Finn-Home study. Hypertension. 2011;57:1081–1086.

8. Cacciolati C, Tzourio C, Dufouil C, Alperovitch A, Hanon O. Feasibility of home blood pressure measurement in elderly individuals: cross-sectional analysis of a population-based sample. Am J Hypertens. 2012;25:1279–1285.

9. Aparicio LS, Barochiner J, Cuffaro PE, Alfie J, Rada MA, Morales MS, Galarza CR, Marin MJ, Waisman GD. Determinants of the morning-evening home blood pressure difference in treated hypertensives: the HIBA-Home Study. Int J Hypertens. 2014;2014:569259.

-Page 6 of 16-

10. Imai Y, Nagai K, Sakuma M, Sakuma H, Nakatsuka H, Satoh H, Minami N, Munakata M, Hashimoto J, Yamagishi T, Watanabe N, Yabe T, Nishiyama A, Abe K. Ambulatory blood pressure of adults in Ohasama, Japan. Hypertension. 1993;22:900–912.

11. O'Brien E, Mee F, Atkins N, Thomas M. Evaluation of three devices for self-measurement of blood pressure according to the revised British Hypertension Society Protocol: the Omron HEM-705 CP, Philips HP5332 and Nissei DS-175. Blood Press Monit. 1996;1:55–62.

12. Yamashina A, Tomiyama H, Takeda K, Tsuda H, Arai T, Hirose K, Koji Y, Hori S, Yamamoto Y. Validity, reproducibility, and clinical significance of noninvasive brachial-ankle pulse wave velocity measurement. Hypertens Res. 2002;25:359–364.

13. Imai Y, Abe K, Sasaki S, Minami N, Munakata M, Sakuma H, Hashimoto J, Sekino H, Imai K, Yoshinaga K. Clinical evaluation of semiautomatic and automatoc devices for home blood pressure measurement: comparison beween cuff-oscillometric and microphone methods. J Hypertens. 1989;7:983–990.

14. Bortolotto LA, Henry O, Hanon O, Sikias P, Mourad JJ, Girerd X. Validation of two devices for self-measurement of blood pressure by elderly patients according to the revised British Hypertension Society protocol: the Omron HEM-722C and HEM-735C. Blood Press Monit. 1999;4:21–25.

15. Pitsavos C, Panagiotakos DB, Chrysohoou C, Stefanadis C. Epidemiology of cardiovascular risk factors in Greece: aims, design and baseline characteristics of the ATTICA study. BMC Public Health. 2003;3:32.

16. Asmar RG, Girerd XJ, Brahimi M, Safavian A, Safar ME. Ambulatory blood pressure measurement, smoking and abnormalities of glucose and lipid metabolism in essential hypertension. J Hypertens. 1992;10:181–187.

17. Kleinbaum DG, Kupper LL, Muller KE. Applied Regression Analysis and Other Multivariate Methods. Boston, Massachusetts, USA, PWS-Kent Publishing Company, 1988, pp 41–79.

18. The SAS Institute. Usage note 24441: How can I compute an R-square statistic for models that are fit by PROC PHREG? http://support.sas.com/kb/24-441.html (accessed 2 July 2015).

http://support.sas.com/kb/24-441.html

-Page 7 of 16-

Table S1. IDHOCO Cohorts Included in Present Analysis

Characteristic Ohasama, Iwate

prefecture, Japan

Didima,

Argolida of

Peloponnesus,

Greece

Tsurugaya, Sendai,

Japan

Buenos Aires,

Argentina

Study design

Recruitment period 1988-1995 1997 2002 2008-2010

Age (years) ≥35 y ≥18 y ≥70 y ≥18 y

Sampling starting point address list local registry address list hospital registry

Participants

All ages 2777 665 836 426

≥80 years 144 36 140 60

≥80 years analyzed 144 36 137 58

Follow-up of octogenarians (years)* 6.53 (0.83–13.8) 6.29 (0.99–8.26) 5.51 (1.62–5.59) 3.52 (0.74–4.53)

Ascertainment of events death certificate stroke registry

medical records

death certificate interviews

medical records

National Health Institute database medical records

hospital files

Median (5th to 95th percentile interval).

-Page 8 of 16-

Table S2. Protocols of Office and Home Blood Pressure Measurement

Study Device

type

Device brand name

(reference for

validation study)

Days (Nº) Times per

day

Readings (Nº) Qualification

of observer

Position Minutes rest

before/between

readings

Office

Ohasama AA USF-700F (10) 1 1 2 N/T SI 2/0

Tsurugaya AO FORM ABI/PWV (12) 1 1 2 T SU 2/–

Didima MM Baumomanometer 2 1 2 D SI 5/1

Buenos Aires AO OMRON 705CP (11) 1 1 2 D/T SI 5/1

Home

Ohasama SAO OMRON HEM401c (13) 28 2 (M,E) 1 P SI 2/…

Tsurugaya AO OMRON HEM747 (14)a 30 2 (M,E) 2 P SI 5/…

Didima AO OMRON 705CP (11) 3 2 (M,E) 2 P (83%) R (17%)

SI 5/1

Buenos Aires AO OMRON 705CP (11) 4-5 2 (M, A, E) 2 P SI 5/1

Abbreviations: AA, automatic auscultatory; AO, automatic oscillometric; SAO, semi-automatic oscillometric; E, evening; A, afternoon, M, morning; MM, manually operated mercury manometer; D, doctor; N, nurse; T, technician; R, relative; P, participant; SI, sitting; SU, supine. An ellipsis indicates not applicable.

aSame device as the OMRON HEM-735C, except that it does not incorporate an integrated circuit memory.

-Page 9 of 16-

Table S3. Baseline Characteristics of 202 Untreated Participants by Cohort

Characteristic Ohasama,

Iwate Prefec-

ture, Japan

Didima,

Argolida of

Peloponnesus,

Greece

Tsurugaya,

Sendai, Japan

Number 106 21 75


Women 63 (59.4) 14 (6.9) 41 (57.7)

Current smoking 13 (12.3) 0 (0.0) 5 (6.7)

Drinking alcohol 8 (7.5) … 19 (25.3)

Diabetes mellitus 4 (3.8) 1 (4.8) 6 (5.7)

Cardiovascular disease 24 (22.6) 6 (28.6) 8 (10.7)*

Mean of characteristic (±SD)

Age (years) 83.5±4.9 84.9±3.7 82.4±3.8*

Body mass index (kg/m2) 22.3±2.2 25.9±4.7* 23.3±2.8*

Office blood pressure (mm Hg)

Systolic 142.3±20.8 138.5±25.0 147.5±20.5

Diastolic 74.8±11.3 72.8±12.4 83.5±11.3‡

Home blood pressure (mm Hg)

Systolic 135.5±17.1 133.7±21.0 143.2±19.3

Diastolic 72.7±9.43 72.5±10.6 76.6±11.3

Home heart rate (bpm) 69.2±8.7 76±6.6‡ 66.8±8.7§

Total cholesterol, mmol/L 4.9±0.9 5.4±0.3§ 5.1±0.9†

Significance of the difference with the adjacent left column: * P≤0.05; † P≤0.01; ‡ P≤0.001; and § P≤0.0001. An ellipsis indicates that the information was not available.

-Page 10 of 16-

Table S4. Baseline Characteristics of 173 Treated Participants by Cohort

Characteristic Ohasama, Iwate Prefec-ture, Japan

Didima, Argolida of Pel-

oponnesus, Greece

Tsurugaya, Sendai, Japan

Buenos Aires, Capital Federal,

Argentina

Number 38 15 62 58


Women 18 (47.4) 8 (53.3) 35 (56.5) 47 (81.0)†

Current smoking 4 (10.5) 0 (0) 6 (9.7) 3 (5.2)

Drinking alcohol 7 (18.4) … 16 (25.8) 1 (1.7)

Diabetes mellitus 3 (7.9) 0 (0) 13 (21.0) 5 (8.6)

Cardiovascular disease 7 (18.4) 8 (53.3)* 20 (32.3) 33 (56.9)†

On antihypertensive agent

Diuretics 2 (5.3) … 11 (17.7) 24 (41.4)*

-blockers 5 (13.2) … 4 (6.5) 31 (53.4)§

ACEIs 0 (0.0) … 11 (17.7)† 22 (37.9)*

ARBs 0 (0.0) … 10 (16.1)† 28 (48.3)§

Calcium-channel blockers 37 (97.4) … 42 (67.7)‡ 34 (58.6)

Other 1 (2.6) … 14 (22.6)† 7 (12.1)†

On combination therapy 7 (18.4) … 22 (35.5)§ 49 (84.5)§

Median number of drugsa 3 (2-3) … 3 (2-3) 3 (2-5)

Mean of characteristic (SD)

Age (years) 81.7±3.1 83.9±5.7 83.1±3.7 82.3±4.0

Body mass index (kg/m2) 22.6±2.3 27.3±4.3‡ 24.2±3.9† 27.4±4.1§

Office blood pressure (mm Hg)

Systolic 147.1±18.9 137.7±23.9 145.8±20.9 147.0±21.6

Diastolic 75.4±13.4 68.3±7.9* 81.4±11.1§ 77.0±11.0*

Home blood pressure (mm Hg)

Systolic 139.0±10.9 138.8±20.4 150.0±21.0 139.4±16.0†

Diastolic 76.9±9.4 69.9±7.5* 76.1±10.3* 70.3±7.8‡

Home heart rate (bpm) 67.7±9.4 74.7±10.1* 67.6±10.1* 68.1±10.0

Serum cholesterol, mmol/L 5.1±0.9 5.4±0.3 5.4±0.8 4.7±0.8§

Significance of the difference with the adjacent left column: * P≤0.05; † P≤0.01; ‡ P≤0.001; and § P≤0.0001. Information on antihypertensive drug use was unavailable in the Didima cohort.

aMedian number of drugs (5th–95th percentile interval) refers to participants on combination therapy.

-Page 11 of 16-

Table S5. Hazard Ratios for Fatal Endpoints in Relation to Home Blood Pressure in Various Strata

Stratum n/N Untreated participants Treated participants

Systolic home BP

(18.7 mm Hg)

Diastolic home BP

(10.4 mm Hg)

n/N Systolic home BP

(18.1 mm Hg)

Diastolic home BP

(9.5 mm Hg)

Total

All participants 103/202 1.20 (0.88–1.64) 0.80 (0.58–1.11) 52/173 0.65 (0.42–1.01) 1.05 (0.72–1.53)

Women 58/118 1.25 (0.78–1.98) 0.77 (0.49–1.24) 25/108 0.54 (0.26–1.15) 1.00 (0.53–1.89)

Men 45/84 1.18 (0.75–1.87) 0.68 (0.41–1.12) 27/65 0.74 (0.41–1.34) 0.82 (0.43–1.57)

Whites 10/21 0.17 (0.02–1.35) 2.68 (0.43–16.9) 15/73 2.57 (0.77–8.55) 0.13 (0.03–0.63)

Asians 93/100 1.25 (0.91–1.72) 0.80 (0.57–1.11) 37/100 0.62 (0.36–1.06) 1.25 (0.83–1.87)

Cardiovascular


Women 24/118 1.42 (0.72–2.82) 0.57 (0.28–1.16) 14/108 0.56 (0.17–1.79) 0.82 (0.29–2.35)

Men 22/84 1.60 (0.83–3.09) 0.52 (0.25–1.08) 16/65 0.81 (0.36–1.81) 0.50 (0.19–1.33)

Whites 7/73 0.25 (0.02–3.08) 1.47 (0.15–15.0) 8/73 … …

Asians 39/100 1.61 (1.02–2.54) 0.51 (0.31–0.84) 22/100 0.68 (0.33–1.41) 1.20 (0.68–2.13)

Abbreviations: CI, confidence interval; n/N, number of deaths/number of participants at risk; BP, blood pressure. Hazard ratios express the risk associated with a 1-SD increase in blood pressure (given between parentheses in column headings). All models include both systolic and diastolic pressure and were adjusted for cohort, sex, age, body mass index, smoking and drinking, serum cholesterol, history of cardiovascular disease and diabetes mellitus. Hazard ratios did not differ between strata (P≥0.075). An ellipsis indicates that the model did not converge.

-Page 12 of 16-

Table S6. Hazard Ratios for Fatal plus Nonfatal Endpoints in Relation to Home Blood Pressure in Various Strata

Stratum n/N Untreated participants Treated participants

Systolic

home BP

(18.7 mm Hg)

Diastolic

home BP

(10.4 mm Hg)

n/N Systolic

home BP

(18.1 mm Hg)

Diastolic

home BP

(9.5 mm Hg)

Cardiovascular


Women 30/118 1.60 (0.88–2.91) 0.47 (0.25–0.88) 22/108 1.22 (0.65–2.31) 1.18 (0.59–2.37)

Men 28/84 1.42 (0.80–2.53) 0.89 (0.51–1.55) 24/65 0.77 (0.43–1.36) 0.63 (0.30–1.29)

Whites 7/21 0.25 (0.02–3.08) 1.47 (0.15–15.00) 7/73 0.80 (0.40–1.62) 0.86 (0.37–1.99)

Asians 51/100 1.54 (1.03–2.29) 0.64 (0.42–0.97) 17/100 1.02 (0.58–1.80) 1.02 (0.65–1.61)

Cardiac


Women 17/118 2.05 (0.91–4.60) 0.40 (0.17–0.95) 12/108 0.64 (0.24–1.70) 1.72 (0.56–5.29)

Men 13/84 1.45 (0.59–3.57) 0.58 (0.23–1.46) 13/65 0.71 (0.29–1.72) 0.60 (0.20–1.83)

Whites 7/21 0.25 (0.02–3.08) 1.47 (0.15–15.00) 14/73 0.54 (0.23–1.30) 1.19 (0.44–3.20)

Asians 23/100 1.95 (1.05–3.65) 0.44 (0.22–0.87) 11/100 0.73 (0.30–1.77) 0.75 (0.30–1.85)

Coronary


Women 10/118 3.79 (0.94–15.2) 0.11 (0.01–1.55) 10/108 0.11 (0.01–1.55) 0.79 (0.14–4.45)

Men 8/84 1.35 (0.35–5.22) 0.77 (0.31–1.90) 8/65 0.77 (0.31–1.90) 0.67 (0.22–2.05)

Whites 4/73 … 0.17 (0.03–1.01) 4/73 0.17 (0.03–1.01) 1.68 (0.37–7.70)

Asians 14/100 2.39 (0.99–5.77) 0.77 (0.32–1.86) 14/100 0.77 (0.32–1.86) 0.61 (0.22–1.74)

Abbreviations: CI, confidence interval; n/N, number of events/number of participants at risk; BP, home blood pressure. Hazard ratios express the risk associated with a 1-SD increase in blood pressure (given between parentheses in column headings). All models include both systolic and diastolic pressure and were adjusted for cohort, sex, age, body mass index, smoking and drinking, serum cholesterol, history of cardiovascular disease and diabetes mellitus. Hazard ratios did not differ between strata (P≥0.075). An ellipsis indicates that the model did not converge.

-Page 13 of 16-

Table S7. Hazard Ratios Excluding One Cohort at a Time in 202 Untreated Participants

Endpoint

Excluded cohort

n/N Systolic home BP

(18.7 mm Hg)

Diastolic home BP

(10.4 mm Hg)

Hazard ratio (95% CI) P Hazard ratio (95% CI) P

Cardiovascular mortality

Ohasama 11/96 0.88 (0.26–3.00) 0.84 0.49 (0.14–1.78) 0.28

Sendai 42/127 1.33 (0.84–2.11) 0.22 0.63 (0.39–1.02) 0.059

Didima 39/181 1.59 (1.02–2.50) 0.043 0.51 (0.31–0.84) 0.008

Cardiovascular events

Ohasama 17/96 1.23 (0.54–2.83) 0.61 0.59 (0.25–1.37) 0.22

Sendai 48/127 1.34 (0.87–2.05) 0.18 0.70 (0.45–1.08) 0.11

Didima 51/181 1.53 (1.03–2.26) 0.034 0.64 (0.42–0.97) 0.035

Cardiac events

Ohasama 12/96 0.88 (0.30–2.55) 0.81 0.62 (0.20–1.86) 0.39

Sendai 25/127 1.57 (0.85–2.90) 0.15 0.51 (0.26–0.99) 0.049

Didima 23/181 1.93 (1.05–3.58) 0.036 0.44 (0.23–0.87) 0.018

Coronary events

Ohasama 7/96 1.17 (0.27–5.02) 0.83 0.33 (0.07–1.72) 0.19

Sendai 15/127 1.76 (0.70–4.45) 0.23 0.29 (0.10–0.80) 0.017

Didima 14/181 2.36 (0.99–5.62) 0.053 0.39 (0.16–0.95) 0.039

Abbreviations: CI, confidence interval; n/N, number of events/number of participants at risk; BP, home blood pressure. Hazard ratios express the risk asso-ciated with a 1-SD increase in blood pressure (given between parentheses in column headings). All models include both systolic and diastolic pressure and were adjusted for cohort, sex, age, body mass index, smoking and drinking, serum cholesterol, history of cardiovascular disease and diabetes mellitus.

-Page 14 of 16-

Table S8. Hazard Ratios Excluding One Cohort at a Time in 173 Treated Participants

Endpoint Excluded cohort

n/N Systolic home BP (18.1 mm Hg)

Diastolic home BP (9.5 mm Hg)

Hazard ratio (95% CI) P Hazard ratio (95% CI) P

Total mortality

Ohasama 25/135 0.87 (0.45–1.68) 0.68 0.77 (0.38–1.56) 0.47

Sendai 42/111 0.63 (0.40–0.99) 0.045 1.02 (0.67–1.54) 0.93

Didima 40/158 0.58 (0.35–0.96) 0.033 1.28 (0.85–1.92) 0.24

Buenos Aires 49/115 0.69 (0.43–1.12) 0.14 1.02 (0.69–1.52) 0.91

Cardiovascular mortality

Ohasama 12/135 0.75 (0.15–3.79) 0.73 0.35 (0.06–2.10) 0.25

Sendai 26/111 0.70 (0.37–1.34) 0.29 1.03 (0.60–1.76) 0.92

Didima 22/158 0.68 (0.33–1.41) 0.30 1.20 (0.68–2.14) 0.53

Buenos Aires 30/158 0.67 (0.35–1.29) 0.23 0.93 (0.53–1.63) 0.80

Cardiovascular events

Ohasama 28/135 0.87 (0.51–1.46) 0.59 1.02 (0.57–1.83) 0.96

Sendai 35/111 0.75 (0.48–1.17) 0.21 0.97 (0.61–1.54) 0.90

Didima 38/158 0.91 (0.58–1.43) 0.68 1.12 (0.74–1.69) 0.60

Buenos Aires 37/115 0.94 (0.56–1.58) 0.81 0.92 (0.58–1.45) 0.73

Cardiac events

Ohasama 19/135 0.47 (0.22–0.98) 0.046 1.13 (0.54–2.37) 0.74

Sendai 20/111 0.50 (0.26–0.97) 0.040 1.14 (0.57–2.29) 0.72

Didima 19/158 0.58 (0.30–1.11) 0.10 1.12 (0.57–2.21) 0.73

Buenos Aires 17/115 0.66 (0.28–1.52) 0.32 0.60 (0.25–1.47) 0.27

Abbreviations: CI, confidence interval; n/N, number of events/number of participants at risk; BP, home blood pressure. Hazard ratios express the risk asso-ciated with a 1-SD increase in home blood pressure (given between parentheses in column headings). All models include both systolic and diastolic pres-sure and were adjusted for cohort, sex, age, body mass index, smoking and drinking, serum cholesterol, history of cardiovascular disease and diabetes mellitus.

-Page 15 of 16-

Figure S1. Five-year risk of cardiovascular mortality associated with systolic (A) or diastolic (B) home blood pressure analyzed as continuous varia-bles and across percentiles (5th, 50th and 95th) of the alternative component of blood pressure in 202 untreated octogenarians. The risk was stand-ardized to average of the distribution in the whole untreated population (ratio or mean) of cohort, sex, age, body mass index, smoking and drinking, serum cholesterol, history of cardiovascular disease and diabetes mellitus. PS and PD indicate the significance of the association with systolic and

diastolic home blood pressure, respectively.

-Page 16 of 16-

Figure S2. Multivariable-adjusted hazard ratios for total mortality by fifths of the distribution of the systolic home blood pressure in 173 treated octogenarians. The hazard ratios express the risk in fifths compared with average risk in all treated participants and were adjusted for cohort, sex, age, body mass index, smoking and drinking, serum cholesterol, history of cardiovascular disease and diabetes mellitus. The point estimates, 95% confidence interval and P-value are provided for the hazard ratio that significantly deviated from the reference at systolic levels below 126.9 mm Hg.

Blood Pressure Monitoring in the Elderly -...

Documents

Transcript of Blood Pressure Monitoring in the Elderly -...