Objective The objective of this study was to assess the heterogeneity in treatment effect of intensive blood pressure (BP)‐lowering across hematoma volume after acute intracerebral hemorrhage (ICH). Methods We undertook a pooled analysis of individual patient data from the pivotal trials of early intensive BP‐lowering in ICH (the Intensive Blood Pressure Reduction in Acute Cerebral Hemorrhage Trial 4 INTERACT4 and Antihypertensive Treatment of Acute Cerebral Hemorrhage 2 ATACH‐2 studies). The primary outcome was functional recovery, defined by the distribution of scores on modified Rankin scale (mRS). Secondary outcomes were hematoma expansion (HE) over 24 hours, defined by absolute (12.5 ml) and relative HE (66%). Generalized linear mixed models with trial as a random effect were conducted. We further assessed effect modification by hematoma volume and plotted the treatment effect curve. Results Among 6,125 individuals with available hematoma volume, intensive BP‐lowering improved functional recovery (odds ratio OR for unfavorable shift in mRS score = 0.90, 95% confidence interval CI = 0.82 to 0.99, p = 0.027). In 3,897 participants with available HE, intensive BP‐lowering reduced the risk of absolute (OR = 0.88, 95% CI = 0.78 to 0.99, p = 0.043) and relative (OR = 0.88, 95% CI = 0.78 to 0.99, p = 0.034) HE. We found effect modification of treatment on functional outcome and absolute HE by hematoma volume (p for interaction = 0.043 and 0.025, respectively). U‐shaped curves were observed, with benefits seen in cases with hematoma volume of 7.5 to 27.5 and 7.0 to 32.5 ml, respectively, both peaking at 20 ml. Interpretation Early intensive BP‐lowering improves functional outcome and reduces HE in ICH. Heterogeneity by hematoma volume indicates the importance of patient selection in future trials and clinical practice. ANN NEUROL 2026
Building similarity graph...
Analyzing shared references across papers
Loading...
Xinwen Ren
Qiang Li
Menglu Ouyang
Annals of Neurology
University of Minnesota
Universidade de São Paulo
University of Leicester
Building similarity graph...
Analyzing shared references across papers
Loading...
Ren et al. (Wed,) studied this question.
www.synapsesocial.com/papers/69cf5e2e5a333a821460c571 — DOI: https://doi.org/10.1002/ana.78216