Effect of antihypertensive drug treatment on long-term blood pressure reduction: An individual patient-level data meta-analysis of 352,744 Participants from 51 large-scale randomised clinical trials

Abstract

Objectives: Evidence from randomised trials on long-term blood pressure (BP) reduction from pharmacologic treatment is limited. To investigate the effects of antihypertensive drugs on long-term BP change and examine its variation over time and among people with different clinical characteristics Design: Individual participant-level data meta-analysis Setting and data source: The Blood Pressure Lowering Treatment Trialists Collaboration involving 51 large-scale long-term randomised clinical trials Participants: 352,744 people (42% women) with mean age of 65 years and mean baseline systolic/diastolic BP of 152/87 mmHg, of whom 18% were current smokers, 50% had cardiovascular disease, 29% had diabetes, and 72% were taking antihypertensive treatment at baseline Intervention: Pharmacological BP-lowering treatment Outcome: Difference in longitudinal changes in systolic and diastolic BP between randomised treatment arms over an average follow-up of four years Result: Drugs were effective in lowering BP, with the maximum effect becoming apparent after 12-month follow-up and with gradual attenuation towards later years. Based on measures taken [≥]12 months post-randomisation, more intense BP-lowering treatment reduced systolic/diastolic BP (95% confidence interval) by -11.2 (-11.4 to -11.0)/-5.6 (-5.8 to -5.5) mmHg than less intense treatment; active treatment by -5.1 (-5.3 to -5.0)/-2.3 (-2.4 to -2.2) mmHg lower than placebo, and active arm by -1.4 (-1.5 to -1.3)/-0.6 (-0.7 to -0.6) mmHg lower than the control arm for drug class comparison trials. BP reductions were consistently observed across a wide range of baseline BP values and ages, and by sex, history of cardiovascular disease and diabetes, and prior antihypertensive treatment use. Conclusion: Pharmacological agents were effective in lowering long-term BP among individuals with a wide range of characteristics, but the net between-group reductions were modest, which is partly attributable to the intended trial goals.