Abstract WP464: Stress-Induced Hyperglycemia After Intracerebral Hemorrhage Predicts In-Hospital Mortality and Temporary CSF Diversion

Abstract

Introduction: Stress-induced hyperglycemia (SIH) in critical illnesses, including intracerebral hemorrhage (ICH) is often associated with increased morbidity and mortality. Objective parameters to determine SIH and associated clinical outcome are not systematically evaluated due to lack of standard definition for SIH. Hypothesis: We investigated if recently standardized definition for SIH, glycemic gap (GG) correlates with disease severity at presentation and in-hospital morbidity and mortality after ICH. Methods: Five-year retrospective chart review of patients with spontaneous ICH. GG defined as the difference between admission glucose (AG) and an A1C derived average glucose (GG= AG-28.7*HbA1c+46.7). Receiver operating curve (ROC) analysis was used to determine a GG threshold to predict a adverse outcome defined as death or discharge to hospice. We used multivariate analyses to determine role of GG in patients with different ICH scores. Results: A GG threshold of 30 mg/dL determined worse clinical outcome in a cohort of 334 patients with ICH. Patients with higher ICH score of 3-5 were more likely to present with SIH as compared to those with ICH score of 0-2, 51/95 (53.7%) vs. 56/239 (23.4%), p<0.001. SIH also showed a dose-dependent correlation to ICH score. Patient with ICH score 0-2 had lower incidence of external ventricular drain (EVD) placement as compared to those with scores 3-5, 31/233 (13.3%) vs. 35/93 (37.6%), p<0.001. In multivariate model ICH score at presentation, history of hypertension, diabetes mellitus, in-hospital EVD placement and mortality were independently associated with SIH. Patients with ICH score 0-2 had 5.7 times more odds in having an EVD placed (p<0.0001). Conclusion: Using objective definition for SIH, i.e. GG our study shows definitive association of SIH with in-hospital morbidity and mortality. Such association alludes to adverse outcome related to neurohumoral response after ICH.