Meeting Highlights


Meta-analysis of Hypertension Trials Confirms Benefits of Blood Pressure Lowering

Presented by Costas Thomopoulos, Italy

Costas Thomopoulos, MD, San Luca Hospital, IRCCS Istituto Auxologico, Milan, Italy, presented the results of a meta-analysis of blood pressure (BP)-lowering trials that have been published over the past 47 years. Dr. Thomopoulos and colleagues conducted this meta-analysis of hypertension randomized trials to determine the effect of differential BP lowering on hard clinical outcomes.

BP-lowering randomized controlled trials can utilize an intentional or nonintentional design. Intentional design trials can be further subdivided into the following classifications: active treatment versus placebo or no treatment; more intensive versus less intensive active treatment; or a more intensive lowering strategy versus a less intensive lowering strategy, including predefined systolic and diastolic BP. For the present meta-analysis, both intentional and nonintentional design trials were included. However, the following types of trials were excluded: nonrandomized trials; those with a mean follow-up <6 months; trials with <5 events reported; trials with different additional interventions among the 2 study arms; heart failure trials; studies of acute myocardial infarction (MI) and acute stroke; studies where the baseline hypertension prevalence was <40%; and comparison trials between active treatments of similar intensity.

MEDLINE was the primary source used to identify appropriate trials. Secondary sources included the Cochrane Database, the Scopus Database, and reference lists of previous meta-analyses.

Clinical outcomes included MI (fatal and nonfatal), stroke (fatal and nonfatal), heart failure, cardiovascular death, all-cause death, a composite of MI and stroke, and a composite of MI, stroke, and heart failure. The key findings of active treatment versus placebo or no treatment from intentional trials are presented in Table 1.

The findings from meta-analysis conducted after pooling all types of intentional and nonintentional trials (Table 2) were consistent with the results from intentional trials of active treatment versus placebo or no treatment.

A metaregression analysis indicated that the difference in achieved BP was negatively correlated with all outcomes (all p≤0.005) except MI (p=0.09 for systolic BP change and p=0.26 for diastolic BP change).

Dr. Thomopoulos concluded that lowering BP per se, regardless of whether an intentional or nonintentional design was employed, led to a reduction in all hard clinical endpoints. In addition, achieved systolic and diastolic BP changes were correlated with a reduction in all events except MI.