Risk Stratification by Self-Measured Home Blood Pressure across Categories of Conventional Blood Pressure: A Participant-Level Meta-Analysis

Background: The Global Burden of Diseases Study 2010 reported that hypertension is worldwide the leading risk factor for cardiovascular disease, causing 9.4 million deaths annually. We examined to what extent self-measurement of home blood pressure (HBP) refines risk stratification across increasing categories of conventional blood pressure (CBP). Methods and Findings: This meta-analysis included 5,008 individuals randomly recruited from five populations (56.6% women; mean age, 57.1 y). All were not treated with antihypertensive drugs. In multivariable analyses, hazard ratios (HRs) associated with 10-mm Hg increases in systolic HBP were computed across CBP categories, using the following systolic/diastolic CBP thresholds (in mm Hg): optimal, <120/<80; normal, 120-129/80-84; high-normal, 130-139/85-89; mild hypertension, 140-159/90-99; and severe hypertension, >= 160/>= 100. Over 8.3 y, 522 participants died, and 414, 225, and 194 had cardiovascular, cardiac, and cerebrovascular events, respectively. In participants with optimal or normal CBP, HRs for a composite cardiovascular end point associated with a 10-mm Hg higher systolic HBP were 1.28 (1.01-1.62) and 1.22 (1.00-1.49), respectively. At high-normal CBP and in mild hypertension, the HRs were 1.24 (1.03-1.49) and 1.20 (1.06-1.37), respectively, for all cardiovascular events and 1.33 (1.07-1.65) and 1.30 (1.09-1.56), respectively, for stroke. In severe hypertension, the HRs were not significant (p >= 0.20). Among people with optimal, normal, and high-normal CBP, 67 (5.0%), 187 (18.4%), and 315 (30.3%), respectively, had masked hypertension (HBP >= 130 mm Hg systolic or >= 85 mm Hg diastolic). Compared to true optimal CBP, masked hypertension was associated with a 2.3-fold (1.5-3.5) higher cardiovascular risk. A limitation was few data from low-and middle-income countries. Conclusions: HBP substantially refines risk stratification at CBP levels assumed to carry no or only mildly increased risk, in particular in the presence of masked hypertension. Randomized trials could help determine the best use of CBP vs. HBP in guiding BP management. Our study identified a novel indication for HBP, which, in view of its low cost and the increased availability of electronic communication, might be globally applicable, even in remote areas or in low-resource settings.