Estimated Glomerular Filtration Rate and Risk of Survival in Acute Stroke.

OBJECTIVE To assess the risk of survival in acute stroke using the MDRD equation derived estimated glomerular filtration rate. DESIGN A prospective observational cross-sectional study. SETTING Medical wards of a tertiary care hospital. SUBJECTS Eighty three acute stroke patients had GFR calculated within 48 hours of admission after basic data were captured. OUTCOME MEASURES Stroke outcome was defined as either discharged or still-in-care (survived) or all cause in-hospital death. GFR was estimated by the MDRD equation, stroke severity was assessed by the Canadian Neurological Scale (CNS). Data were compared between the GFR groups of < 60 ml/min and ≥ 60 ml/min. Relative risks (RR) and odds ratios (OR) for stroke outcomes (survival and death) were estimated between the GFR groups and the homogeneity of the odds ratios among the different layers of stroke severity (CNS < 6.5 and ≥ 6.5) was determined by Breslow-Day and Tarone's test. Matanel Hazensel and Cochran's tests were used to determine conditional independence and the common odds ratio with stroke severity as a layering variable. RESULTS No significant differences were found between the age and sex distribution of the two GFR groups. Serum urea and creatinine and CNS were significantly different between the GFR groups (p < 0.001, < 0.001, < 0.001). RR of survival and death for the GFR groups-less than 60 ml/min and above or equal to 60 ml/min were (0.425 and 1.204) and (2.360 and 0.830). The OR of survival for GFR below 60 ml/min compared to GFR above or equal to 60 ml/min was 0.353. There was homogeneity across the two layers of stroke severity (CNS score less than 6.5 and above or equal to 6.5), p = 0.612 and 0.612. CONCLUSION Independent of stroke severity, GFR is a surrogate in the assessment of the risk of survival in acute stroke.