Intensive blood pressure lowering after spontaneous intracerebral haemorrhage for secondary stroke prevention (RECAP-ICH): a systematic review and individual participant data meta-analysis

BACKGROUND Blood pressure-lowering treatment prevents stroke after spontaneous intracerebral haemorrhage. However, there is uncertainty over the consistency of the effects across clinically relevant subgroups as well as the time course over which benefits accrue. We aimed to pool individual-level data across diverse settings and treatment strategies to quantify the size, heterogeneity, and timing of the effects of blood pressure-lowering treatment for the secondary prevention of major cardiovascular endpoints in patients with a history of intracerebral haemorrhage. METHODS We did a systematic review and meta-analysis of identified randomised controlled trials evaluating long-term blood pressure-lowering for secondary prevention in adults with a history of spontaneous intracerebral haemorrhage. We searched Embase and Ovid MEDLINE, from database inception to Jan 25, 2026, with two authors (YG and MJV) independently screening titles, abstracts, and full-text articles for eligibility. Trials were eligible if they enrolled adults (aged 18 years or older) with previous spontaneous intracerebral haemorrhage, randomly assigned participants to fixed-dose antihypertensive therapy or an intensive, titrated target-based blood pressure-lowering strategy versus placebo or standard blood pressure management, and reported clinical outcomes. Stroke trials had available individual participant data for subgroups of at least 100 participants with acute intracerebral haemorrhage. Methodological quality was assessed using the Cochrane Risk of Bias 2 tool. The primary outcome was first recurrent stroke of any type. We pooled individual participant data with a one-stage intention-to-treat meta-analysis using Cox proportional hazards models, incorporating a random effect for trial and adjustment for prespecified covariates. Prespecified subgroup analyses included demographic variables and baseline blood pressure, with formal tests for treatment-by-subgroup interaction. We estimated the time to benefit for a clinically meaningful absolute risk reduction, defined as 1% risk difference. The protocol is registered with PROSPERO (CRD420251274994). FINDINGS Of 236 records screened, 37 trials were retrieved, 24 were assessed for eligibility after duplicates were removed, and four eligible trials provided individual participant data for analysis. 2944 participants were included in the meta-analysis. The mean age was 59·6 years (SD 10·6), 972 (33·0%) were female, 1972 (67·0%) were male, 2218 (75·3%) were Asian, and the median follow-up was 42·0 months (IQR 24·0-66·0). The mean systolic blood pressure difference between blood pressure-lowering treatment and control over follow-up was 11·2 mm Hg (95% CI 10·7-11·7). Intensive blood pressure-lowering reduced the hazard of first recurrent stroke of any type compared with control (96 [6·5%] of 1487 participants versus 152 [10·4%] of 1457 participants; adjusted hazard ratio [HR] 0·62, 95% CI 0·48-0·80; p=0·0002). This finding was driven by fewer recurrent intracerebral haemorrhage events (33 [2·2%] of 1487 participants on blood pressure lowering-treatment versus 81 [5·6%] of 1457 control participants; adjusted HR 0·39, 95% CI 0·26-0·59; p<0·0001). Serious adverse events occurred in 429 (28·9%) of 1487 patients in the intensive group and 481 (33·0%) of 1457 patients in the control group. Treatment effects were consistent across prespecified subgroups (eg, age, sex, region, background blood pressure-lowering treatment, baseline blood pressure, and time since index event). The time to achieve a 1% absolute benefit was estimated to be 6·1 months (95% CI 3·5-14·8). INTERPRETATION Intensive blood pressure-lowering treatment after spontaneous intracerebral haemorrhage reduces recurrent stroke, mainly by preventing recurrent haemorrhage, without evidence of excess serious adverse events. Effects were consistent across patient characteristics, including time since the index event and baseline level of blood pressure. The early accrual and durability of benefit after treatment initiation strengthens the case for sustained long-term blood pressure control as the cornerstone of secondary prevention after intracerebral haemorrhage.