PROGNOSTIC IMPORTANCE OF LOW VISIT-TO-VISIT SYSTOLIC BLOOD PRESSURE VARIABILITY IN HIGH-RISK HYPERTENSION

Abstract

Objective: Besides high blood pressure variability (BPV), low BPV was associated with adverse cardiovascular (CV) prognosis in selected high-risk patients. We explored this issue among high-risk, nondiabetic, nonstroke hypertensive adults from the Systolic Blood Pressure Intervention Trial (SPRINT). Design and method: Long-term SBPV (coefficient of variation, CoV %) was calculated on quarterly visits from randomization until an outcome, excluding titration period and patients with missing visits. The primary outcome was any fatal/nonfatal CV event (myocardial infarction [MI]; acute coronary syndrome not resulting in MI; stroke, acute decompensated heart failure [HF], death from CV diseases [CVD]); secondary outcomes were all-cause mortality and the composite of CV events and mortality. Multivariate (treatment arm, pulse pressure, age > = 75 years, gender, race, CVD, chronic kidney disease [CKD], smoke, n.BP medications) Cox proportional hazard models with penalized smoothing splines were used to explore the potential nonlinear relationship between the continuum of CoV and the HR (95%CI) for outcomes. Sensitivity analysis based on 1:1 propensity score matching (PSM) was performed. Results: The association of SBPV with CV events was J-shaped (lowest risk for CoV 5–10%), while that with all-cause mortality and with the composite outcome was linear. After multivariate adjustment, only low SBPV (CoV < 5%) was associated with CV events (HR 1.85, 95%CI 1.24–2.75), especially MI (HR 2.76, p < 0.001). High SBPV (CoV>10%) was associated with the risk of the composite outcome (HR 1.35, p = 0.037). PSM, matching 566 non-smoker, non-CVD/CKD participants having SBPV 5–10% with an equal number of patients having different SBPV, confirmed the inverse association of low SBPV with CV events (HR 3.33, p = 0.035). Conclusions: Nonlinear modeling indicates that low visit-to-visit SBPV has prognostic relevance in high-risk hypertensives from SPRINT. Unmeasured factors, like inflammatory and metabolic changes, might contribute to the observed results. We treated SBPV as a continuous variable calculated on 16 visits and included 60% of the SPRINT cohort, which might explain the limited analogy with earlier findings. Dedicated trials are needed to test these results and related implications.