Risk of heart failure in ambulatory resistant hypertension: a meta-analysis of observational studies

The impact of ambulatory resistant hypertension (ARH) on the occurrence of heart failure (HF) is not yet completely known. We performed for the first time a meta-analysis, by using published data or available data from published databases, on the risk of HF in ARH. Patients with ARH (24-h BP >= 130/80 mmHg during treatment with >= 3 drugs) were compared with those with controlled hypertension (CH, clinic BP < 140/90 mmHg and 24-h BP < 130/80 mmHg regardless of the number of drugs used), white coat uncontrolled resistant hypertension (WCURH, clinic BP >= 140/90 mmHg and 24-h BP < 130/80 mmHg in treated patients) and ambulatory nonresistant hypertension (ANRH, 24-h BP >= 130/80 mmHg during therapy with <= 2 drugs). We identified six studies/databases including 21,365 patients who experienced 692 HF events. When ARH was compared with CH, WCURH, or ANRH, the overall adjusted hazard ratio for HF was 2.32 (95% confidence interval (CI) 1.45-3.72), 1.72 (95% CI 1.36-2.17), and 2.11 (95% CI 1.40-3.17), respectively, (all P < 0.001). For some comparisons a moderate heterogeneity was found. Though we did not find variables that could explain the heterogeneity, sensitivity analyses demonstrated that none of the studies had a significant influential effect on the overall estimate. When we evaluated the potential presence of publication bias and small-study effect and adjusted for missing studies identified by Duval and Tweedie's method the estimates were slightly lower but remained significant. This meta-analysis shows that treated hypertensive patients with ARH are at approximately twice the risk of developing HF than other ambulatory BP phenotypes.