Abstract WP191: DM is Not Related With Poor Outcome in Acute Ischemic Stroke. the GLIAS-II Study

Abstract

Backgrounds: Diabetes Mellitus (DM) has been widely associated with poor outcome after an ischemic stroke (IS). However the underlying mechanisms are not well elucidated. We aim to study whether the outcome in stroke patients with DM is independent of comorbidity, the development of poststroke hyperglycaemia or due to higher glycaemic variability. Methods: Secondary analysis of the GLIAS-II study (multicenter, prospective, observational academic cohort study). We compared outcomes (in-hospital complications; mortality; death or dependence (mRS >2); and stroke recurrence at 3 months) in two study groups: patients with prior diagnosis of DM (DM group) vs. patients without DM (non DM group). We performed multivariate analysis in order to adjust the effect of DM on stroke outcomes by comorbidity (Charlson index), poststroke hyperglycaemia (>155 mg/dL) during the fist 48 h and glycaemic variability (standard deviation and coefficient of variation). Results: We included 213 patients, 64 (30%) with DM. The baseline characteristics (age, sex, stroke severity and the proportion of patients treated with intravenous thrombolysis) were similar between groups. There were no differences in in-hospital complications: pneumonia (4.7% vs. 3.4%; P=.699); urine infection (4.7% vs.1.3%; P =.161); cerebral edema (6.2% vs. 4%; P=.492); or symptomatic hemorrhagic transformation of the stroke (1.6 vs. 0.7%; P =.512) between DM and non DM groups. There was no difference between groups in mortality (12.7 vs. 5.7%; P=.096); death or dependency at 3 months (31.7% vs. 26.4%; P =.500), neither in stroke recurrence (3.1 vs. 4%; P =1.000) at 3 months. After adjusting for other prognostic factors, DM was not associated with any of the studied outcomes, whilst glycaemic variability was independently associated with mortality (p=0.032) and hyperglycemia within 48 hours with death or dependence (p=0.049). Conclusion: DM itself was not associated with poor outcome in acute stroke patients admitted to SU, being glycaemic variability and poststroke hyperglycemia risk factors for mortality and death or dependence respectively.