6 Results
6.2 Ankle and exercise blood pressures as predictors of coronary
A total of 436 coronary events occurred during the follow-up of 18 years. There were 145 deaths due to CHD, 122 non-fatal MIs, 77 percutaneous coronary angioplasties and 147 persons with coronary artery bypass grafting. Differences in baseline characteristics between persons with and without a CHD event during the follow-up are presented in Table 3. Persons with a CHD event were older, especially women (62 years). In persons without a CHD event, almost all risk factors were more favourable compared with the CHD event group.
Table 4 compares cardiovascular risk factors in different blood pressure groups. The reference group was younger and all risk factors were more favourable compared with the other groups. In the reference group, 4% of individuals had positive exercise test without a CHD event in the follow-up.
Figure 5 shows the Kaplan-Meier curves of the five blood pressure groups for all CHD events. The curves clearly show the best prognosis of the group with normal ankle and exercise brachial blood pressures and the worst prognosis of the group with poor physical performance (group 5). The curves of the other groups (groups 2–4) are very close to each other between the two extremes.
The most significant predictors of CHD event were age (HR 1.06, 95% CI 1.04–
1.07, P<0.0001), gender (HR, women compared with men, 0.27, 95% CI 0.21–0.36, P<0.0001), SBP (HR 1.01, 95% CI 1.01–1.02, P<0.0001), physical performance (METs, HR 0.79, 95% CI 0.75–0.84, P<0.0001), positive family history (HR 1.45, 95% CI 1.19–
1.76, P=0.0002) and pack-years of smoking (HR 1.01, 95% CI 1.01–1.02, P=0.0003) when self-reported elevated cholesterol and self-reported elevated blood glucose were adjusted for.
The age- and sex-adjusted and the multivariate adjusted HRs of a CHD event in the different blood pressure groups are shown in Table 5. Compared with the reference group, the most elevated risk was seen in group with poor physical performance (group 5). The persons with elevated ABP without exaggerated exercise brachial blood pressure and normal or slightly elevated resting brachial blood pressure (144 ± 16/90 ± 10mmHg) had 2.10-fold risk of a CHD event compared with the reference group.
Further adjustment for traditional risk factors, such as smoking, BMI, parental history of early CVD, selfreportedelevated cholesterol and blood glucose attenuated the HR slightly, but it nevertheless remained clearly significant (1.68-fold). The persons with elevated ABP and exaggerated brachial exercise blood pressure at a moderate exercise level had a 2.09-fold risk in the basic model and 1.80-fold risk in the larger model compared with the reference group. In this group, the resting brachial BP was also elevated (153 ± 17/94 ± 10mmHg). In persons with a discrepancy between the ankle and exercise brachial blood pressures, that is, normal ABP and elevated exercise brachial blood pressure (group 4), the risk of a future coronary event was even higher (2.63-fold
in the age and sex adjusted model and 2.29-fold in the multivariate adjusted model). The resting brachial blood pressure was normal or slightly elevated (135 ± 15/86 ± 9mmHg).
In general, the HRs were higher for fatal than for non-fatal CHD events. The persons with poor physical performance (group 5) had the highest risk for fatal outcome (15.9-fold in the basic model and 7.02-(15.9-fold in the larger model). In groups 2–4, the risk of CHD death was about 3.03- to 4.62-fold in the basic model and about 3.03- to 4.16-fold in the larger model. In addition, the risk of a non-fatal CHD event was significant for all groups in both models.
Also the HRs remains significant, when the brachial systolic blood pressure was added in the multivariate model (Group 2, HR 1.4, 95% CI 1.05–1.87, P=0.02; Group 3, HR 1.37 95% CI 1.00–1.90, P=0.05; Group 4 HR 2.1, CI 95% 1.47–3.0, P<0.0001;
Group 5, HR 3.2, 95% CI 2.2–4.7, P<0.0001).
For comparison, we also calculated the HRs for the usual brachial systolic blood pressure, for exercise brachial blood pressure and for ABP alone. The resting brachial systolic blood pressure was dichotomized at 160mmHg (≤160 vs >160mmHg). In the basic model, that is, adjusted for age and sex only, the HR for a CHD event was 1.83 (1.39–2.41, P<0.0001). In the larger model, the HR was 1.40 (1.05–1.85, P=0.02). The same comparison was repeated using the fatal CHD event as the outcome. In the basic model, the HR was 2.80 (1.87– 4.17, P<0.0001) and in the larger model the HR was 2.02 (1.33–3.06, P=0.0009). In persons with elevated exercise blood pressure (>215mmHg) at the moderate exercise level, the HRs for fatal or nonfatal CHD event were also significant elevated (1.3–fold).
When analysing the ABP alone, we considered four mutually exclusive categories: (1) ABP <175mmHg was taken as the ‘normal’ or reference category; (2) ABP between 175–
215mmHg was taken as the ‘moderately elevated ABP’ category; (3) ABP >215mmHg was taken as the ‘high ABP’ category; and (4) persons with abnormal ABI (<0.97) were considered as their own category. In persons with moderately elevated ABP, the HRs for fatal and non-fatal CHD event did not reach statistical significance in any of the models (data not shown). In persons with high ABP, the HR for a fatal and non-fatal CHD event was significantly elevated in both models, 2.11 (1.05–4.27, P=0.04) and 1.49 (1.12–1.98, P=0.006). Also, in persons with abnormal ABI, the HR for a fatal and non-fatal CHD event was significant in both models, 3.34 (1.34–8.31, P=0.01) and 1.91 (1.24–2.96, P=0.004).
Table 3. Baseline Characteristics of Participants With and Without Incident Coronary Heart Disease Event During the Follow-up.
No CHD Event CHD Event
n=3423 n=436 ρ
Age (years) 50 (10) 56 (9) 0.0001*
Men/women, n (%) 2217 (65) / 1206 (35) 344 (79) / 92 (21) 0.0001†
Body mass index (kg/m2) 26.0 (4) 26.9 (4) 0.0001 *
Syst. blood pressure (mmHg) 132 (18) 142 (19) 0.0001*
Diast. blood pressure (mmHg) 85 (11) 88 (10) 0.0001*
Pulse /min 74 (13) 73 (12) NS
SBP80 (women) 181 (25) 198 (21) 0.0001**
SBP150 (men) 200 (26) 213(27) 0.0001**
SBPMax 203 (27) 207 (31) 0.005*
Ankle blood pressure (mmHg) 165 (26) 172 (32) 0.0001*
ABI 1.21 (0.22) 1.18 (0.2) 0.01*
Current smokers, n (%) 543 (15%) 103 (28%) 0.0001†
Self-reported abnormal total
cholesterol, n (%) 1256 (36%) 198 (45%) 0.0001†
Self-reported abnormal blood
glucose, n (%) 230 (7%) 60 (14%) 0.0001†
Pos. family history, n (%) 1256 (37%) 183 (42%) 0.009 †
METs 7.9 (3) 6.5 (2) 0.0001*
Blood pressure groups Group I (reference) Group II
Group III Group IV Group V
2085 (61%) 662 (19%) 425 (12%) 181 (5%) 70 (3%)
128 (29%) 116 (27%) 85 (19%) 45 (10%)
62 (15%) 0.0001†
Data are mean (S.D.) or n and proportions (%), * t-Test,** Mann-Whitney U Test, † χ2 test SBP80=systolic blood pressure at the exercise level of 80 Watts
SBP150=systolic blood pressure at the exercise level of 150 Watts SBPMax=maximum systolic blood pressure during exercise ABI=ankle brachial index
METs=physical working capacity in metabolic equivalents
Please see the Methods-section for the explanation for the blood pressure groups.
Table 4. Baseline characteristics of participants by the ankle and exercise blood pressure group Group 1
n=2213 Group 2
n=778 Group 3
n=510 Group 4
n=226 Group 5
132 P
Age 47(10) 55(9) 54(7) 51(9) 59(10) 0.0001a
Male/female 1379/800 428/344 417/80 206/20 101/34 0.0001b
BMI (kg/m2) 25.4(4) 26.8(4) 27.5(4) 26.8(4) 25.6(4) 0.0001a
SBP (mmHg) 124(14) 144(17) 152(17) 135(16) 137(20) 0.0001a
DBP(mmHg) 81(9) 90(10) 94(10) 85(9) 84(11) 0.0001a
ABP(mmHg) 150(17) 194(14) 198(18) 157(14) 141(37) 0.0001a
Antihypertens.
medication (n, %) 214(10) 241(31) 133(27) 27(12) 53(39) 0.0001b
Elevated
cholesterol (n, %) 727(33) 338(43) 233(46) 106(47) 50(38) 0.0001b
DM(n,%) 91(4) 92(12) 65(13) 18(8) 24(18) 0.001b
Pack-years of
smoking 9(11) 9(12) 13(14) 14(13) 20(16) 0.0001a
Please see the Materials and methods section for the explanation for blood pressure groups Data are mean (s.d.) or n and proportions (%)
a Analysis of variance b χ2-Test
Ref.Group, n=2179 2147 2094 1632 142
Group 2, n=772 720 680 467 38
Group 3, n=497 468 440 274 13
Group 4, n=226 216 200 145 14
Group 5, n=135 98 78 52 7
The numbers indicate people remaining in the follow-up at different points of time.
Figure 5. Kaplan-Meier Curves for CHD Event
Table 5. Hazard Ratios (HR, 95% Confidence Interval (CI)) of nonfatal, fatal and all CHD events according to specified blood pressure groups.
Model 1 Model 2 Model 1: adjusted for age and sex adjusted
Model 2: adjusted for age, sex, BMI, smoking, early parental cardiovascular disease and physical working capacity(METs), self-reported elevated cholesterol and abnormal blood glucose