Abstract DP262: Cocaine Use Does Not Impact Outcomes in Intracerebral Hemorrhage

Background: Cocaine use has been attributed to the occurrence of intracerebral hemorrhage (ICH) and worse clinical outcomes. However, data is limited to small, single-center cohorts without long-term follow-up. We assessed long-term outcomes of cocaine-associated ICH in a large, pooled analysis of multiple ICH studies that included active cocaine users. Methods: We harmonized individual and patient-level data from two randomized clinical trials (Antihypertensive Treatment of Acute Cerebral Hemorrhage 2 [ATACH 2] and Minimally Invasive Surgery Plus Alteplase for Intracerebral Hemorrhage Evacuation III [MISTIE III]) and one prospective observational study (Ethical/Racial Variations of Intracerebral Hemorrhage [ERICH]) with available modified Rankin scale (mRS) between 90-180 days. Cocaine-associated ICH was defined as self-reported active use within three months prior to ICH or physician determined cocaine-associated ICH. Logistic regression was used to assess the association between cocaine and poor functional outcomes (mRS>3) with adjustment for demographics, comorbidities, ICH characteristics (volume, location, initial GCS, blood pressure, time from onset), in-hospital events, and follow-up time. Secondary outcomes included hematoma expansion, discharge disposition, and mortality. Results: From a pooled cohort of 4,702 ICH patients, 128 patients had cocaine-associated ICH. Cocaine users were younger (53.7 vs 62.1 years), more frequently Black (60.2% vs 25%), smokers (37.3% vs 12.8%), and less likely to have comorbidities or use anticoagulants. Cocaine use was associated with non-lobar hematoma (83.6% vs 71.1%) and more frequently received ventricular drainage or ICP monitoring (26.6% vs 17.1%). A poor outcome occurred in 37.5% of cocaine-associated ICH and 44.9% of spontaneous ICH (Table 1). No association between cocaine and poor outcomes (mRS 4-6) was identified in the adjusted multivariable model (OR 0.86, CI 0.49-1.50) (Table 2). We found no association between cocaine use and hematoma expansion, discharge disposition, withdrawal of care, or mortality. Results were similar when accounting for trial population and intervention. Discussion: In the largest study of cocaine-associated ICH to date, we found no evidence to suggest cocaine use is a poor prognostic factor. Cocaine use alone should not influence prognostication and limitations of care after ICH.