Blood Pressure and Outcomes in Patients With Different Etiologies of Intracerebral Hemorrhage: A Multicenter Cohort Study

BACKGROUND: We aimed to investigate the association between blood pressure (BP) and outcomes in intracerebral hemorrhage (ICH) subtypes with different etiologies. METHODS AND RESULTS: A total of 5656 in‐hospital patients with spontaneous ICH were included between January 2012 and December 2016 in a prospective multicenter cohort study. Etiological subtypes of ICH were assigned using SMASH‐U (structural lesion, medication, amyloid angiopathy, systemic/other disease, hypertension, undetermined) classification. Elevated systolic BP was defined as ≥140 mm Hg. Hypertension was defined as elevated BP for >1 month before the onset of ICH. The primary outcomes were measured as 1‐month survival rate and 3‐month mortality. A total of 5380 patients with ICH were analyzed, of whom 4052 (75.3%) had elevated systolic BP on admission and 3015 (56.0%) had hypertension. In multinomial analysis of patients who passed away by 3 months, systolic BP on admission was significantly different in cerebral amyloid angiopathy (P<0.001), structural lesion (P<0.001), and undetermined subtypes (P=0.003), compared with the hypertensive angiopathy subtype. Elevated systolic BP was dose‐responsively associated with higher 3‐month mortality in hypertensive angiopathy (P (trend)=0.013) and undetermined (P (trend)=0.005) subtypes. In cerebral amyloid angiopathy, hypertension history had significant inverse association with 3‐month mortality (adjusted odds ratio, 0.37, 95% CI, 0.20–0.65; P<0.001). Similarly, adjusted Cox regression indicated decreased risk of 1‐month survival rate in the presence of hypertension in patients with cerebral amyloid angiopathy (adjusted hazard ratio, 0.47; 95% CI, 0.24–0.92; P=0.027). CONCLUSIONS: This study suggests that the association between BP and ICH outcomes might specifically depend on its subtypes, and cerebral amyloid angiopathy might be pathologically distinctive from the others. Future studies of individualized BP‐lowering strategy are needed to validate our findings.