Abstract Background and aims Spontaneous intracerebral haemorrhage (ICH) is associated with high morbidity and mortality. We investigated associations between cerebral and systemic haemodynamic parameters, including intracranial pressure (ICP), and mortality risk post-ICH in a UK intensive care unit (ITU). Methods This retrospective cohort included 120 patients ≥18 years old admitted to ITU between June 2013 and April 2019 with spontaneous ICH, who underwent continuous monitoring. Mean, standard deviation and variability independent of mean (VIM) were calculated for ICP and other haemodynamic variables in over 7,400 hours of data. Kruskal-Wallis tests assessed differences by ICH volume tertile and survival during monitoring (23 - 94.5 hours). Multivariate Cox regression evaluated predictors of survival. Results Eighty patients survived beyond 30 days; 40 died within 30 days; 27 died during the monitoring period. ICP VIM (p0.001) and mean arterial pressure VIM (p=0.017) were associated with ICH volume, with greater variability observed in smaller haemorrhages. Mean diastolic blood pressure (p=0.048) and mean heart rate (p=0.037) showed U-shaped associations with ICH volume (higher in the upper and lower tertiles). Cox regression identified elevated ICP (p=0.001) and female sex (p=0.019) as predictors of reduced survival. Higher central venous pressure (CVP) (p=0.019), systolic blood pressure (SBP) (p=0.001) and larger ICH (p=0.009) predicted improved survival. Conclusions In this cohort, a complex interplay of factors is observed, highlighted by analysis of ICH volume, whereby ICP variability was found to be significantly greater in smaller haemorrhages, which were associated with reduced survival. Higher CVP, SBP, and larger ICH volume were predictors of improved survival. Conflict of interest Jennifer Nicholls: nothing to disclose. Corry Gellatly: nothing to disclose. Meeriam Kadicheeni: nothing to disclose. Ronney Panerai: nothing to disclose. Thompson Robinson: nothing to disclose. Hiren Patel: nothing to disclose. Adrian Parry-Jones: nothing to disclose. Jatinder Minhas: nothing to disclose.
Nicholls et al. (2026) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: