Br J Hosp Med (Lond). 2026 Sep 21;87(9):58250. doi: 10.31083/BJHM58250.
ABSTRACT
AIMS/BACKGROUND: The 'obesity paradox' has been reported in intracerebral hemorrhage (ICH), but evidence on long-term mortality remains limited. We investigated the association between body mass index (BMI) and long-term all-cause mortality in a hospital-based ICH cohort.
METHODS: This retrospective cohort study included adult patients with primary ICH admitted within 48 h between 1 November 2016 and 28 February 2021. Patients with secondary ICH, neurosurgical procedures, isolated intraventricular hemorrhage, severe pre-existing disability, or loss to follow-up were excluded. Patients were categorised based on World Health Organization BMI criteria as follows: underweight (<18.5 kg/m2), normal weight (18.5-24.9 kg/m2), overweight (25.0-29.9 kg/m2), and obese (≥30.0 kg/m2). The primary outcome was all-cause mortality. Associations between BMI and mortality were assessed using Cox proportional hazards models, and potential nonlinearity was evaluated by modeling BMI with a spline term within the Cox proportional hazards framework.
RESULTS: Of the 947 screened patients, 839 were included (underweight n = 66; normal weight n = 465; overweight n = 261; obese n = 47). Median follow-up time varied across BMI groups (25.5-43.0 months). BMI demonstrated an inverse association with all-cause mortality in an exploratory multivariable model excluding age (hazard ratio (HR): 0.923, 95% confidence interval (CI): 0.864-0.985; p = 0.016). However, this association was attenuated and no longer statistically significant after adjustment for age (HR: 0.958, 95% CI: 0.898-1.022; p = 0.190), suggesting important age-related confounding. Restricted cubic spline analysis showed no evidence of nonlinearity (p = 0.475). Kaplan-Meier curves showed higher survival in overweight and obese groups than in underweight and normal-weight groups. BMI was also negatively associated with hematoma volume; this relationship remained after log transformation.
CONCLUSION: An apparent inverse association between BMI and long-term mortality was observed after primary ICH; however, this association was substantially influenced by age. These findings highlight the complex relationship between BMI, aging, and hemorrhagic stroke characteristics and should be interpreted cautiously given the observational design.
CLINICAL TRIAL REGISTRATION: ClinicalTrials.gov (NCT04803292).
PMID:42812084 | DOI:10.31083/BJHM58250