Effects of blood pressure lowering on outcome incidence in hypertension. 1. Overview, meta-analyses, and meta-regression analyses of randomized trials

Abstract

Background: Antihypertensive treatment is based on randomized controlled trials (RCTs) started since 1966. Meta-analyses comprehensive of all RCTs but limited to RCTs investigating blood pressure (BP) lowering in hypertensive patients are lacking.

Objectives: Two clinical questions were investigated: the extent of different outcome reductions by BP lowering in hypertensive patients, and the proportionality of outcome reductions to SBP, DBP, and pulse pressure (PP) reductions.

Methods: PubMed between 1966 and December 2013 (any language), Cochrane Collaboration Library and previous overviews were used as data sources for identifying and selecting all RCTs comparing the antihypertensive drugs with placebo or less intense BP lowering (intentional BP-lowering RCTs); comparing BP-lowering drugs with placebo without BP-lowering intention, but with BP difference (nonintentional BP-lowering RCTs); and enrolling at least 40% hypertensive patients. RCTs on acute myocardial infarction, heart failure, acute stroke, and dialysis were excluded. RCT quality was assessed by scoring. Risk ratios and 95% confidence interval (CI), standardized to 10/5 mmHg SBP/DBP reduction, of seven fatal and nonfatal outcomes were calculated (random-effects model). The relationships of different outcome reductions to SBP, DBP, and PP reductions were investigated by meta-regressions.

Results: A total of 68 RCTs (245,885 individuals) were eligible, of which 47 (153,825 individuals) were 'intentional' RCTs. All outcomes were reduced (P < 0.001) by BP lowering, stroke [-36% (-29, -42)], and heart failure [-43% (-28, -54)] to a greater extent, with smaller reductions for coronary events [coronary heart disease (CHD): -16% (-10, -21)], cardiovascular [-18% (-11, -24)], and all-cause mortality [-11% (-5, -16)]. Absolute risk reductions were 17 (14, 20) strokes, 28 (19, 35) cardiovascular events, and 8 (4, 12) deaths prevented every 1000 patients treated for 5 years. Logarithmic risk ratios were related to SBP, DBP, and PP reductions (P = 0.001-0.003) for stroke and composite cardiovascular events, but not for CHD.

Conclusion: Meta-analyses of all BP-lowering RCTs involving hypertensive patients provide precise estimates of benefits (larger for stroke and heart failure, but also significant for CHD and mortality). Absolute risk reductions are substantial. Relationships of logarithmic risk ratios with BP reductions imply risk reduction increases progressively to a smaller extent the larger the BP reduction.