BACKGROUND

Uncertainty remains about the effects of intensive blood pressure (BP) lowering in acute intracerebral haemorrhage, particularly the impact of treatment timing. This study aimed to assess the safety and effectiveness of early intensive BP-lowering treatment and its dependence on timing in patients with intracerebral haemorrhage.

METHODS

We undertook an individual patient-data pooled analysis of the four Intensive Blood Pressure Reduction in Acute Cerebral Haemorrhage Trials: INTERACT1 (n=404), INTERACT2 (n=2829), INTERACT3 (n=7036), and INTERACT4 (n=1043). INTERACT1-3 included adults with acute intracerebral haemorrhage who presented within 6 h of the onset of symptoms and had an elevated systolic BP (>150 mm Hg). INTERACT4 included patients with suspected acute stroke that caused a motor deficit and an elevated systolic BP (≥150 mm Hg) within 2 h after the onset of symptoms, among whom 1029 had a haemorrhagic form of stroke. Patients were randomly assigned to receive intensive (target systolic BP <140 mm Hg within 1 h) or guideline-recommended (target systolic BP <180 mm Hg within 1 h) BP-lowering treatment using locally available drugs. The primary outcome was functional recovery, defined by the distribution of scores on the modified Rankin scale (mRS). In a CT substudy, radiological outcomes were relative (≥33%) and absolute (≥6 mL) changes in haematoma volume from baseline to 24 h. The treatment effects were determined in logistic regression models adjusting for trial and baseline haematoma volume. Heterogeneity in the effects across groups by time from onset to randomisation (continuous) and baseline severity according to the intracerebral haemorrhage score were assessed by adding interaction terms to the models. These trials are registered at ClinicalTrials.gov (INTERACT1 NCT00226096; INTERACT2 NCT00716079; INTERACT3 NCT03209258; INTERACT4 NCT03790800). This pooled analysis is registered with PROSPERO (CRD420251001539).

FINDINGS

Among 11 312 patients (mean age 63 years [SD 12·7], 4066 [35·9%] female and 7246 [64·1%] male), the median time from the onset of symptoms to randomisation was 2·9 h (IQR 1·8-4·1). At 1 h, the mean systolic BP was 149·6 mm Hg (SD 21·8) in the intensive treatment group and 158·8 mm Hg (22·8) in the guideline group (difference 9·13 mm Hg, 95% CI 8·28-10·00; p<0·0001). Intensive BP-lowering treatment significantly decreased the chances of poor physical function (mRS scores of 3-6; odds ratio [OR] 0·85, 95% CI 0·78-0·91). Compared with guideline treatment, intensive BP-lowering treatment significantly reduced odds of neurological deterioration within 7 days (OR 0·76, 95% CI 0·66-0·88; p=0·0002), death (0·83, 0·75-0·94; p=0·002), and any serious adverse event (0·84, 0·76-0·92; p=0·0003). In the CT substudy involving 2921 patients, there was no apparent effect of intensive treatment on relative (0·85, 0·70-1·03; p=0·09) or absolute (0·84, 0·68-1·04; p=0·12) haematoma growth compared with guidelines treatment. In the same substudy, the treatment effects on functional recovery and relative haematoma growth decreased with increasing time from onset to randomisation, with a cutoff point in the effect crossing unity at 3 h (p=0·002 and p=0·01 for interaction, respectively).

INTERPRETATION

Intensive BP-lowering initiated within several hours of intracerebral haemorrhage onset was safe and improved functional recovery, without a clear effect on haematoma growth. The greatest benefits for both outcomes occurred when treatment was commenced within 3 h of symptom onset. These findings underscore the importance of early intervention and inform the design of future trials targeting patients at highest risk of haematoma expansion.

FUNDING

National Health and Medical Research Council of Australia (NHMRC); Department of Health and Social Care, the Foreign, Commonwealth & Development Office, Medical Research Council, and Wellcome Trust (all UK); the West China Hospital; Sichuan Credit Pharmaceutical; Takeda Pharmaceuticals China; the George Institute for Global Health; Shanghai East Hospital of Tongji University, National Natural Science Foundation of China, Sichuan Science and Technology Program, Project of Neurology Key Discipline of Sichuan (2018-53), Chengdu Science and Technology Bureau, the talent fund of Sichuan Provincial People's Hospital, and the talent fund of the First Affiliated Hospital of Chengdu Medical College.

TRANSLATION

For the Chinese translation of the abstract see Supplementary Materials section.