Location and Timing of Recurrent, Nontraumatic Intracerebral Hemorrhage.

Abstract

Importance: : The spatial and temporal distribution of intracerebral hemorrhage (ICH) recurrence are largely unknown.

Objective: To assess timing and location of recurrent ICH events in relation to the index ICH event (adjacent ICH [adjICH] vs remote ICH [remICH]).

Design, setting, and participants: This cohort study was a pooled analysis of individual cohort studies from 2002 to 2021 among hospital-based European cohorts. Patients with 2 or more clinically distinguishable (≥1 recurrent) small vessel disease-related ICH events were included. Data analysis was performed from December 2023 to December 2024.

Exposures: ICH location and underlying small vessel disease type.

Main outcomes and measures: The primary outcome was adjICH, defined by anatomical ICH location and side, and the secondary outcome was time to recurrence. Multivariable regression analyses were conducted adjusting for ICH location, cerebral amyloid angiopathy according to Boston 2.0 or simplified Edinburgh criteria, convexity subarachnoid hemorrhage extension, hypertension, and antihypertensive treatment, including an interaction term for hypertension and antihypertensive treatment.

Results: Among 733 patients (median [IQR] age, 72.4 [65.2 to 79.0] years; 346 female [47.2%]), there were 1616 ICH events, including 733 index and 883 recurrent ICH events (range, 1 to 6 recurrences) over a median (IQR) follow-up of 2.53 (0.66 to 4.92) years. There were 340 patients (46.4%) with adjICH and 393 patients (53.6%) with remICH. Among recurrent ICH events, there were 476 adjICH events and 407 remICH events. In multivariable regression analyses, lobar index ICH (adjusted odds ratio [aOR], 2.08; 95% CI, 1.32 to 3.27) and cerebral amyloid angiopathy at index ICH (aOR, 2.21; 95% CI, 1.57 to 3.11) were associated with higher odds of adjICH, while cerebellar index ICH was associated with lower odds of adjICH (aOR, 0.25; 95% CI, 0.07 to 0.89). The median (IQR) time to recurrence was 1.25 (0.36 to 3.38) years for adjICH and 2.21 (0.66 to 4.85) years for remICH. Previous lobar or convexity subarachnoid hemorrhage (coefficient, -0.75; 95% CI, -1.25 to -0.25; P = .003 ), adjICH (coefficient, -0.60; 95% CI, -1.02 to -0.18; P = .005), and the number of previous ICH events (coefficient per 1-event increase, -0.62; 95% CI, -0.93 to -0.32; P < .001) were independently associated with a shorter time to recurrence.

Conclusions and relevance: This study found that early recurrence and cerebral amyloid angiopathy were associated with adjICH. These findings suggest that regional, tissue-based factors may facilitate recurrence and that identifying and targeting local vasculopathic changes may represent potential novel treatment targets.