302 Outcomes in Aneurysmal Subarachnoid Hemorrhage Patients with Diabetes on Sulfonylureas: A Propensity Match Analysis

Abstract

INTRODUCTION: Aneurysmal subarachnoid hemorrhage (aSAH) is associated with an increase blood-brain-barrier permeability, disruption of tight junctions, and increased cerebral edema. Sulfonylureas has been associated with less disturbance in tight junctions and reduced edema following aSAH in animal models. METHODS: Patients from the Post-Barrow Ruptured Aneurysm Trail (PBRAT) databse from August 1st, 2007, to July 31st, 2019, were retrospectively reviewed. Initial analysis grouped patients into two cohorts: diabetes and no-diabetes. A second analysis in diabetic patients separated patients based on prescription of sulfonylureas on hospital admission. A 4:1 nearest neighbor propensity match (match for age, gender, Hunt and Hess (HH), Fisher Grade, aneurysm location, aneurysm size, aneurysm type, Charlson Comorbidity Score (CCI), and treatment) was performed. Primary outcome was neurological status at last follow-up, with a poor outcome defined as a modified Rankin Score >2 and a good outcome as mRS </= 2. Significance was established as p < 0.05. RESULTS: A total of 1014 aSAH patients were treated during the study period. Of these patients, 129 (13%) were found to have a history of diabetes on admission. Compared to non-diabetic patients, the diabetic cohort had an increase age (59, sD 14. vs 55, sD 14.1; p < 0.006) and a higher CC I(2.5 vs 1.5; p < 0.001). Patients with diabetes (N = 77, 60%) were found to have a significantly higher odd of a mRS >2 on last follow-up compared to non-diabetics (N = 43, 49%) (p = 0.02). In the diabetic cohort, 16 patients (12%) were found to be on a sulfonylurea medication. Following propensity score matching, diabetic patients on a sulfonylurea (31%, N = 5) were found to have a significantly decrease percentage of a mRS >2 on last follow-up compared to non-sulfonylurea patients (63%, N = 40/64) (p = 0.02). On univariate analysis, besides the administration of a sulfonylurea, a HH < 4 was found to be associated with a decrease percentage of a mRS >2 (49%, N = 28 of 57) (p = 0.04). However, on multivariable logistic regression analysis, only the administration of a sulfonylurea was found to be a predictor of a good neurological outcome (OR 3.4, 95% CI: 1.1-11.3, p = 0.04). CONCLUSION: Diabetes appears to be associated with poor neurological outcomes in aSAH patients. However, the poor neurological outcome in diabetic patients appears to be mitigated by the administration of sulfonylureas.