Introduction

Aging is a dominant risk factor for cardiovascular disease (CVD) and hypertension.[1,2] The increased arterial stiffness, usually accompanied by the aging of vascu latures and endothelial dysfunctions, plays a central role in CVD among older people.[3] Aortic pulse wave velocity (PWV) elevation has been recognized to correlate significantly with blood pressure progression[4,5] and cardiovascular events.[6,7] Carotid-femoral pulse wave velocity (cfPWV) is the gold standard for the evaluation of arterial stiffness, but it is not applicable to large clinical studies due to technical and equipment limitations. Brachial ankle pulse wave velocity (baPWV) has the advantages of being economical, non-invasive, and reproducible, which has been reported to be significantly correlated with cfPWV.[8,9] Higher baPWV has a reverse relationship with blood pressure control in response to antihypertensive treatments.[10]

The optimal systolic blood pressure (SBP) treatment target in hypertensive patients remains inconsistent according to current clinical guidelines. China has the most population in the world, including about 18.9% of the population older than 60 years, and the problem of aging is inescapable. Recently, the Strategy of Blood Pressure Intervention in the Elderly Hypertensive Patients (STEP) trial showed that intensive treatment (target, 110 mmHg ≤SBP <130 mmHg) resulted in a lower incidence of cardiovascular events than standard treatment (target, 130 mmHg ≤SBP <150 mmHg) among older patients with hypertension.[11] A post-hoc analysis of the STEP trial showed that arterial stiffening preceded blood pressure elevation and led to difficulty in reaching target SBP under intensive treatment.[12] A secondary analysis of the Systolic Blood Pressure Intervention Trial (SPRINT) showed that intensive treatment (target, SBP <120 mmHg) was associated with improvements in estimated aortic stiffness after 12 months of follow-up and improved risk prediction for incident CVD events and mortality.[13] An extended SPRINT-HEART study combined cardiovascular magnetic resonance imaging of aortic stiffness and suggested that improvement in aortic stiffness might be one of the mechanisms contributing to the CVD benefits of intensive blood pressure treatment.[14] However, the relationship between blood pressure and PWV is an interesting but complex topic in its own right, which needs more investigation in older people.

Therefore, in the present study, we used the longitudinal data of baPWV and ankle-brachial index (ABI) from the STEP trial to evaluate whether intensive treatment could delay or reverse the progression of arterial stiffness and to investigate the mechanistic link between blood pressure control, arterial stiffness, and the reduction of cardiovascular events in older hypertensive patients aged 60–80 years.

Methods Study design and participant population

This study was a post-hoc analysis of the STEP trial, and the details of the study design and rationale for STEP trial have been published previously.[11,15] Briefly, the STEP trial was a prospective, multicenter, randomized controlled trial (No. NCT03015311) aiming to compare two SBP treatment targets (an intensive target of 110 mmHg ≤SBP <130 mmHg vs. a standard target of 130 mmHg ≤SBP <150 mmHg). A total of 8511 patients were enrolled at 42 clinical centers throughout China from January 10 to December 31, 2017, and randomly assigned to the intensive treatment group (n = 4243) or the standard treatment group (n = 4268). The geographical distribution and affiliations of the STEP group members were presented in Supplementary Table 1, https://links.lww.com/CM9/B747. All eligible patients were required to be aged 60–80 years and have hypertension with 140 mmHg ≤SBP ≤190 mmHg during three screening visits or be taking antihypertensive medication. Patients with a history of ischemic or hemorrhagic stroke were excluded. The inclusion and exclusion criteria were shown in Supplementary Methods, https://links.lww.com/CM9/B747. After randomization, all participants were scheduled for follow-up once monthly for the first three months and every three months thereafter until the end of the study (December 31, 2020). This study was approved by the Ethics Committee of the Fuwai Hospital and collaborating centers (No. 2016-838). All participants provided written informed consent.

At baseline and during the 3rd follow-up year (from December 17, 2019 to December 31, 2020) of the STEP trial, enrolled patients underwent repeated measurements of baPWV and ABI to evaluate arterial stiffness. In the present analysis, patients who did not complete arterial stiffness measurements at baseline (n = 799 in the intensive group and n = 847 in the standard group) or at the follow-up year (n = 715 in the intensive group and n = 739 in the standard group), and patients who were lost to follow-up (n = 34 in the intensive group and n = 36 in the standard group) or discontinued interventions (n = 2) were excluded. Finally, a total of 5339 (62.7%) patients were included for analyzing the effects of intensive blood pressure control on progression of arterial stiffness and cardiovascular risk. The flowchart is shown in Figure 1.

Figure 1:

Flowchart showing the patient selection process in this study. ABI: Ankle-brachial index; baPWV: Brachial-ankle pulse wave velocity; SBP: Systolic blood pressure; STEP: Strategy of Blood Pressure Intervention in the Elderly Hypertensive Patients.

Assessment for arterial stiffness

The baPWV and ABI were examined using an automated waveform analyzer (Omron BP-203RPEIII, Omron Healthcare, Kyoto, Japan) to assess arterial stiffness.[16,17] Measurements were performed from 7:00 to 9:00 in the morning on the examination day by well-trained clinicians. After being seated for at least 5 minutes, participants were asked to lie down on the examination couch in a supine position and remain quiet during measurement. Cuffs were wrapped on both arms and ankles. BaPWV is automatically calculated as the length of an arterial segment between brachium and ankle (which is estimated from body height) divided by the transit time of pulse wave from the brachial to ankle arteries. Examinations were performed on both the left and right sides, and the maximum of the right and left baPWV values was used in the analysis. Values of baPWV ≤1800 cm/s were considered normal, while values of baPWV >1800 cm/s indicated the presence of arterial stiffness.[18] Based on the SBP and diastolic blood pressure (DBP) levels measured by the arterial stiffness detection device, the mean arterial pressure (MAP) was calculated as DBP + 1/3 (SBP–DBP).

The ABI was typically calculated as the ratio of the higher of the dorsal pedis and posterior tibial arterial systolic pressure to the higher of the left and right brachial systolic pressures. The lower of right and left ABI values recorded were considered as the final ABI value. Values of ABI >0.90 and ≤1.40 in both legs were considered normal, while values of ABI ≤0.90 or >1.40 indicate the presence of peripheral arterial disease.[19] Details of baPWV and ABI measurements are described in Supplementary Methods, https://links.lww.com/CM9/B747.

Assessment of covariates

At baseline, all participants completed a standardized questionnaire, including age, sex, weight, height, educational level, smoking status, alcohol intake, medical history, and current medication treatment. In this study, represented antihypertensive drugs were provided to patients free of charge, including angiotensin receptor blockers (ARBs), calcium channel blockers (CCBs), and thiazide-type diuretics, all of which have robust evidence to reduce blood pressure and prevent cardiovascular events. For all participants, olmesartan medoxomil (Nanjing Chia Tai Tianqing Pharmaceutical Co., Ltd, Nanjing, China) was the preferred ARB at a daily dose of 20 mg (once daily), and amlodipine besylate (China Resources Saike Pharmaceutical Co., Ltd, Beijing) was the preferred CCB at a daily dose of 5–10 mg (once daily), and hydrochlorothiazide was not used as the initial therapy. Other antihypertensive drugs such as beta-adrenergic blockers might also be used when an investigator deemed it necessary and appropriate. A detailed antihypertensive treatment algorithm to reach the SBP targets has been previously published.[11]

Office blood pressure measurements were performed by a trained trial staff member (physician or nurse). Patients were required to rest for at least 5 minutes in a seated position, and then the blood pressure was measured by a staff three times at 1-minute intervals. This process was standardized, and the same validated office blood pressure monitor[20] was used at all participating centers at baseline and follow-up visits.

Assessment of endpoints

The primary outcome was a composite of stroke (ischemic or hemorrhagic), acute coronary syndrome (acute myocardial infarction and hospitalization for unstable angina), acute decompensated heart failure, coronary revascularization, atrial fibrillation, or death from cardiovascular causes. Definitions and evaluation criteria of the endpoints were outlined in Supplementary Methods, https://links.lww.com/CM9/B747. The follow-up time for each patient was determined from the date of cardiovascular events, death, loss to follow-up, or the last follow-up, whichever came first.

Statistical analysis

Clinical characteristics of participants were compared between two treatment groups using Chi-squared tests for categorical variables (expressed as n [%]), and t-tests or Mann–Whitney non-parametric tests for quantitative variables (expressed as mean ± standard deviation or median [interquartile range]). Values of baPWV and ABI were presented as the mean (95% confidence interval [CI]). To evaluate the effect of intensive treatment on arterial stiffness, changes in baPWV and ABI during the follow-up period were compared between the intensive and standard treatment groups by using a multivariate linear regression model, which adjusted for baseline clinical features including age, sex, blood lipid levels, blood glucose, current smoking and drinking status, history of CVD and diabetes, and baseline values of baPWV or ABI when appropriate. In addition, considering that change in PWV has a close association with change in blood pressure,[21,22] and that the MAP calculated as DBP + 1/3 (SBP–DBP) is more accurate for reflection of the peripheral compliance than either SBP or DBP[23,24] and strongly predicts the CVD events,[25] we accordingly also adjusted baseline MAP and change in MAP during follow-up as covariates in the analytical models to analyze arterial stiffness in the present study. The improvement of arterial stiffness was considered when baPWV or ABI values were abnormal at baseline but normal during follow-up. The proportions of patients with arterial stiffness improvement in the two treatment groups were compared using the Chi-squared test.

The incidence of the primary composite CVD outcomes was estimated using Kaplan–Meier curves. The hazard ratio (HR) with 95% CI was calculated using the Cox proportional hazard regression model after adjusting for the covariates mentioned above to evaluate the effect of intensive treatment on the primary outcomes. Subgroup analyses were also conducted, including age (<70 years vs. ≥70 years), sex (male vs. female), SBP at baseline (<140 mmHg, 140–150 mmHg, and >150 mmHg), history of diabetes (yes vs. no), the 10-year risk of CVD based on the Framingham risk score (<15% vs. ≥15%), and baseline baPWV (<1800 cm/s vs. ≥1800 cm/s). For each subgroup analysis, P value for interaction between the treatment effect and the subgroups mentioned above was reported.

All analyses were performed using R software, version 3.6.3 (R Foundation for Statistical Computing, Vienna, Austria). A two-sided P value <0.05 was considered statistically significant.

Results Clinical characteristics of studied patients

Of 8511 patients in the original STEP trial, 5339 (62.7%) patients who had repeated measurements of arterial stiffness at baseline and during follow-up were included in this study. The included patients were younger (66.0 years vs. 66.7 years), and had a lower proportion of patients ≥70 years (22.2% vs. 27.5%) and current alcohol intake (25.8% vs. 27.4%) compared with the excluded patients (P <0.05) [Supplementary Table 2, https://links.lww.com/CM9/B747]. Among the included patients, 2693 and 2646 patients, respectively, were in the intensive and standard treatment groups. Overall, the mean age of patients was 66.0 ± 4.7 years, 45.9% (n = 2453) were male, 19.7% (n = 1052) had prior diabetes mellitus, 6.0% (n = 320) had prior coronary heart disease, and 65.7% (n = 3510) were at high risk (10-year Framingham risk score ≥15%). There were no statistically significant differences in baseline characteristics between the two treatment groups [Table 1]. All patients were provided antihypertensive drugs free of charge, including olmesartan, amlodipine, and hydrochlorothiazide. In general, patients receiving the intensive treatment had higher proportions of two drugs combination therapy of olmesartan and amlodipine after enrollment (40.2% [n = 1082] in the intensive group and 31.9% [n = 844] in the standard group). A total of 24 (0.4%, 24/5339) patients used other kinds of ARB and CCB.

Table 1 - Baseline characteristics of patients in this study. Characteristics Total (N = 5339) Intensive treatment (N = 2693) Standard treatment (N = 2646) t/ χ 2/H values P values Age (years) 66.0 ± 4.7 66.0 ± 4.7 66.0 ± 4.6 –0.153‡ 0.88 Distribution of age 0.100§ 0.77 60–69 years 4155 (77.8) 2091 (77.6) 2064 (78.0) 70–80 years 1184 (22.2) 602 (22.4) 582 (22.0) Male 2453 (45.9) 1228 (45.6) 1225 (46.3) 0.261§ 0.62 Body mass index (kg/m2) 25.6 ± 3.2 25.6 ± 3.1 25.7 ± 3.2 –1.595‡ 0.11 Baseline blood pressure (mmHg) SBP 146.1 ± 17.0 146.4 ± 17.0 145.9 ± 16.9 1.109‡ 0.27 DBP 82.6 ± 10.7 82.5 ± 10.7 82.6 ± 10.5 –0.063‡ 0.95 Distribution of SBP 4.485§ 0.11 <140 mmHg 1980 (37.1) 962 (35.7) 1018 (38.5) 140–150 mmHg 1330 (24.9) 691 (25.7) 639 (24.1) >150 mmHg 2029 (38.0) 1040 (38.6) 989 (37.4) Fasting serum glucose (mmol/L) 6.2 ± 1.7 6.2 ± 1.7 6.2 ± 1.7 –0.045‡ 0.96 Lipids profile (mmol/L) Total cholesterol 4.9 ± 1.2 4.9 ± 1.2 4.9 ± 1.1 0.615‡ 0.54 Triglycerides 1.4 (1.0–1.9) 1.3 (1.0–2.0) 1.4 (1.0–1.9) –0.329‖ 0.74 HDL-C 1.3 ± 0.3 1.3 ± 0.3 1.3 ± 0.3 0.007‡ 0.99 LDL-C 2.7 ± 0.9 2.7 ± 0.9 2.7 ± 0.9 –0.441‡ 0.66 Missing data 3 (0) 1 (0) 2 Educational level 1.480§ 0.23 Middle school or below 3009 (56.4) 1496 (55.6) 1513 (57.2) High school or above 2327 (43.6) 1196 (44.4) 1131 (42.7) Missing data 3 (0) 1 (0) 2 (0.1) Smoking status 0.379§ 0.83 Never 3848 (72.1) 1951 (72.4) 1897 (71.7) Former 626 (11.7) 312 (11.6) 314 (11.9) Current 853 (16.0) 424 (15.8) 429 (16.2) Missing data 12 (0.2) 6 (0.2) 6 (0.2) Alcohol consumption 1.487§ 0.48 Never 3694 (69.2) 1882 (69.9) 1812 (68.5) Former 257 (4.8) 123 (4.6) 134 (5.1) Current 1376 (25.8) 682 (25.3) 694 (26.2) Missing data 12 (0.2) 6 (0.2) 6 (0.2) Medical history Diabetes mellitus 1052 (19.7) 535 (19.9) 517 (19.5) 0.090§ 0.78 Hyperlipidemia 2002 (37.5) 1029 (38.2) 973 (36.8) 1.177§ 0.28 CVDs 320 (6.0) 165 (6.1) 155 (5.9) 0.172§ 0.69 Ten-year risk of CVD ≥15%* 3510 (65.7) 1770 (65.7) 1740 (65.8) 0.001§ 0.99 Antihypertensive drug use after enrollment Amlodipine alone 1677 (31.5) 766 (28.4) 911 (34.4) 22.192§ <0.001 Olmesartan alone 993 (18.6) 456 (16.9) 537 (20.3) 9.964§ 0.002 HCZ alone 7 (0.1) 4 (0.1) 3 (0.1) 0.126§ 0.72 Amlodipine and olmesartan 1926 (36.1) 1082 (40.2) 844 (31.9) 39.688§ <0.001 Amlodipine and HCZ 33 (0.6) 19 (0.7) 14 (0.5) 0.676§ 0.41 Olmesartan and HCZ 74 (1.4) 39 (1.4) 35 (1.3) 0.154§ 0.70 Amlodipine, olmesartan and HCZ 185 (3.5) 105 (3.9) 80 (3.0) 3.059§ 0.08 Other ARB or CCB 24 (0.4) 8 (0.3) 16 (0.6) 2.822§ 0.10 Other drugs† 253 (4.7) 124 (4.6) 129 (4.9) 0.461§ 0.67

Data are shown as mean ± standard deviation, n (%), or median (Q1–Q3). ARB: Angiotensin receptor blocker; CCB: Calcium channel blocker; CVD: Cardiovascular disease; DBP: Diastolic blood pressure; HCZ: Hydrochlorothiazide; HDL-C: High-density lipoprotein cholesterol; LDL-C: Low-density lipoprotein cholesterol; SBP: Systolic blood pressure. *Ten-year CVD risk was estimated with Framingham risk scoring, and patients with a ≥15% risk score were considered to be at high risk. †Other drugs included traditional Chinese drugs or polypills, which are a combination of antihypertensive agents. ‡Student's t-test. §Chi-squared test for qualitative variables. ||H values.

Trajectory pattern of blood pressure during the follow-up period

The blood pressure trend in the intensive and standard treatment groups in this post-hoc analysis was consistent with that in the original STEP trial [Figure 2]. The two treatment strategies led to a rapid and sustained between-group difference in SBP in older patients with hypertension. During the median follow-up period of 3.36 years, the mean decreased in SBP from baseline were 20.6 mmHg and 10.4 mmHg in the intensive and standard treatment groups, respectively. Throughout follow-up, the mean SBP values were 127.7 mmHg and 136.2 mmHg, while the mean DBP values were 76.6 mmHg and 79.4 mmHg in the intensive and standard treatment groups, respectively [Supplementary Figure 1, https://links.lww.com/CM9/B747]. Patients in the intensive treatment group had more antihypertensive agents used, and the mean number of antihypertensive agents used per patient was 1.9 in the intensive treatment group and 1.5 in the standard treatment group at 42 months [Figure 2].

Figure 2:

Trajectory pattern of office systolic blood-pressure and antihypertensive medications during the follow-up period. The SBP target was 110 mmHg to <130 mmHg in the intensive treatment group and 130 mmHg to <150 mmHg in the standard treatment group. The mean number of medications is based on the number of blood-pressure medications administered at each visit per patient. ISBP: Systolic blood pressure.

Since patients with SBP <150 mmHg in the standard group usually did not require additional antihypertensive medication, and this difference in treatment strategy might be greater in patients without high SBP (<140 mmHg) and affect the clinical benefits from the SBP <140 mmHg subgroup, we compared the number of antihypertensive medications for patients with SBP <140 mmHg or within 140–150 mmHg in the two treatment groups, showing that the number of antihypertensive agents used in the intensive treatment group was significantly higher than that used in the standard treatment group during follow-up [Supplementary Table 3, https://links.lww.com/CM9/B747].

Effect of intensive treatment on the longitudinal changes of arterial stiffness

There were statistically significant differences in baPWV change but not in ABI change between the two treatment groups [Table 2]. The changes in baPWV (∆baPWV) from baseline to the follow-up visit were 61.5 cm/s (95% CI: 49.8–73.2 cm/s) in the intensive treatment group and 98.4 cm/s (95% CI: 86.7–110.1 cm/s) in the standard treatment group (P <0.001), after adjustment for vascular risk factors including age, sex, blood lipid levels, blood glucose, current smoking and drinking status, history of CVD and diabetes, baseline baPWV, MAP, and change in MAP during follow-up. The annual changes of baPWV were 23.1 cm·s–1·year–1 and 36.7 cm·s–1·year–1 in the intensive and standard treatment groups, respectively.

Table 2 - Effect of intensive blood pressure treatment on change in arterial stiffness during the follow-up period. Variables Intensive treatment, mean (95% CI) Standard treatment, mean (95% CI) t/F value P value N of patients 2693 2646 baPWV (cm/s) Baseline baPWV 1781.2 (1768.4, 1793.9) 1818.3 (1805.7, 1830.9) –4.053 <0.001 Follow-up baPWV 1848.4 (1835.0, 1808.5) 1912.6 (1899.3, 1925.9) –6.682 <0.001 Unadjusted changes in baPWV 67.2 (54.8, 79.6) 94.3 (82.0, 106.7) 0.991 0.002 Adjusted changes in baPWV* 61.5 (49.8, 73.2) 98.4 (86.7, 110.1) 17.553 <0.001 ABI Baseline ABI 1.10 (1.09, 1.11) 1.10 (1.09, 1.11) 0.615 0.140 Follow-up ABI 1.09 (1.09, 1.10) 1.10 (1.09, 1.10) 0.277 0.190 Unadjusted changes in ABI –0.004 (–0.009, 0.001) –0.004 (–0.008, 0.001) 0.693 0.980 Adjusted changes in ABI* –0.004 (–0.009, 0.0003) –0.004 (–0.008, 0.001) 0.060 0.810

ABI: Ankle-brachial index; baPWV: Brachial ankle pulse wave velocity; CI: Confidence interval; CVD: Cardiovascular disease; MAP: Mean arterial pressure. *Linear regression model was used to compare changes in baPWV and ABI between two treatment groups after adjustment for baseline clinical features of age, sex, blood lipid levels, blood glucose, current smoking and drinking status, history of CVD and diabetes, baseline values of baPWV or ABI, MAP, and changes in MAP during the follow-up.

Subgroup analyses of ΔbaPWV were also conducted according to age, sex, baseline SBP, prior diabetes, and 10-year risk of CVD, showing a beneficial effect on arterial stiffness under intensive SBP control [Supplementary Table 4, https://links.lww.com/CM9/B747]. Particularly, intensive treatment significantly attenuated progression of baPWV in hypertensive patients at high risk (10-year Framingham risk score ≥15%) (P <0.001 and P for interaction = 0.04). For patients with hypertension aged 70–80 years, the effect size of ∆baPWV was 54.6 cm/s and 101.4 cm/s for the intensive and standard treatment groups, respectively (P = 0.03 and P for interaction = 0.21). The power would achieve 67% to detect a difference in ∆baPWV with a two-sided α level of 0.05 among 1184 patients in the subgroup of patients aged over 70 years. And for patients with baseline SBP under 140 mmHg, the effect size of ∆baPWV was 68.3 cm/s and 118.1 cm/s for the intensive and standard treatment groups, respectively (P <0.001 and P for interaction = 0.07). The power would achieve 95% to detect a difference in ∆baPWV with a two-sided α level of 0.05 among 1980 patients in the subgroup of patients with baseline SBP <140 mmHg. For patients with arterial stiffness (baPWV ≥1800 cm/s) at baseline, intensive treatment could significantly reverse the progression of arterial stiffness, showing ∆baPWV to be –48.5 (95% CI: –69.6 to 27.5) cm/s vs. –16.1 (95% CI: –36.1 to 3.9) cm/s in the intensive and standard treatment groups (P = 0.04), respectively. Patients who had baseline baPWV ≥1800 cm/s but regressed to be baPWV <1800 cm/s during follow-up accounted for 25.2% (287/1139) and 20.4% (254/1248), respectively, in the intensive and standard treatment groups (P = 0.005) [Supplementary Table 5,