Association between carotid-femoral pulse wave velocity and cardiovascular disease in individuals with moderate blood pressure: a systematic review and individual participant meta-analysis

Abstract

Objectives: The predictive value of carotid-femoral pulse wave velocity (cfPWV) for cardiovascular (CV) events in individuals with blood pressure (BP) 120-159/80-99 mm Hg, where more accurate risk stratification has the greatest clinical effect, is unknown. This study aims to determine whether cfPWV improves the prediction of CV events beyond traditional risk factors in individuals with moderate BP.

Design: A systematic review and meta-analysis.

Data sources: PubMed and EMBASE were searched through April 2023.

Eligibility criteria: We included prospective, population-based cohort studies with ≥1 year follow-up that directly measured cfPWV as an index of arterial stiffness and reported incident CV disease (CVD), atherosclerotic CVD (ASCVD), coronary heart disease, stroke or all-cause mortality outcomes.

Data extraction and synthesis: Individual participant data from 11 cohorts (n=15 987) were harmonised and analysed using two-stage random-effects meta-analysis. Incremental predictive and clinical utility analyses compared 10-year risk models with and without cfPWV.

Results: There were 1279 first atherosclerotic CV events over a median follow-up of 9.9 years. A 1-SD increase in log_e(cfPWV) was associated with a 1.21-fold (95% CI 1.08 to 1.36) increase in risk of ASCVD. Adding cfPWV to traditional risk factors improved ASCVD prediction: change in discrimination (C-index): 0.0048 (95% CI 0.0002 to 0.0094), p=0.041. In hypothetical populations of 100 000 individuals with moderate BP, cfPWV-guided treatment could reduce event rates by 2.7% and 3.1% under European and US guidelines, respectively.

Conclusions: Adding cfPWV to traditional CV risk factors may improve the prediction and classification of first CV events in individuals with moderate BP. Additional screening with cfPWV could enhance risk stratification for antihypertensive treatment initiations.

Keywords: EPIDEMIOLOGY; Hypertension; Meta-Analysis.

Figure 1. Forest plot of the HRs for the effect of a 1-SD higher log_e(cfPWV) on the risk of ASCVD, the primary outcome. HRs adjusted for age, sex, SBP, HDLC, total cholesterol, smoking status, diabetes and antihypertensive medications. Pooled HR is estimated from a random effects meta-analysis, weighted by the number of events. N=15 987. ACCT, anglo-cardiff collaboration trial; ASCVD, atherosclerotic cardiovascular disease; BLSA, Baltimore Longitudinal Study of Ageing; CaPS, Caerphilly Prospective Study; cfPWV, carotid-femoral pulse wave velocity; DanMONICA, Danish-Multinational MONitoring of trends and determinants in Cardiovascular disease; EDIVA, The Estudo de DIstensibilidade VAscular project; Health ABC, The Health, Aging and Body Composition Study; HDLC, high-density lipoprotein cholesterol; NSHD, National Survey of Health and Development; SBP, systolic blood pressure.

Figure 2. Flowchart showing the screening and treatment pathway for a hypothetical population of 100 000 individuals aged 40–79 years based on the PCEs and the novel cfPWV risk model for ASCVD, the primary outcome. ACC, American College of Cardiology; AHA, American Heart Association; ASCVD, atherosclerotic cardiovascular disease; cfPWV, carotid-femoral pulse wave velocity; CV, cardiovascular; DBP, diastolic blood pressure; PCE, pooled cohort equations; SBP, systolic blood pressure.