SubArachnoid Hemorrhage - Which Grading Scale Best Predicts Outcome? (P5.336)

Abstract

OBJECTIVE:To determine which scale best predicts the outcome of SAH among H&H, GCS and Fischer grade, independent of demographic risk factors BACKGROUND:Multiple grading scales have been developed to grade severity of subarachonid haemorrhage(SAH). The most widely used scales, Hunt and Hess(H&H), Fischer and Glasgow Coma Scale(GCS scale were developed based on severity of symptoms,which has resulted in unclear correlation with outcome. DESIGN/METHODS:A retrospective chart review of 141 consecutive patients who presented to 2 academic institutions between July 2009 to April 2012 and met the criteria for spontaneous SAH was done. Demographics (age, gender and race) and comorbidities (diabetes, hypertension and hyperlipidemia) were recorded along with GCS, H&H and Fischer. Primary outcome was discharge modified Rankin Scale (mRS), further dichotomized into good (mRS≤2) and poor (mRS>2). Results were analyzed using SYSTAT software. Canonical correlation coefficients were calculated between each scale and mRS, controlling for demographics and risk factors. Efficacy of individual scales in predicting outcome(poor v. good) was assessed using binary logistic regression model, into which comorbidities as well as baseline demographics were incorporated.RESULTS:All three scales correlated with discharge mRS, even after controlling for the effect of age and other risk factors. The GCS (R2 = 0.369, p< 0.001) had the highest correlation coefficient, followed by H&H(R2=0.276, p< 0.001) and Fischer(R2=0.054, p=0.003), which had the least correlation. In logistic regression model,GCS was again best in predicting the dichotomous outcome measure, with ROC AIC of 0.878 (OR=1.440, [1.257-1.649], p < 0.001) followed by the H&H with ROC area of 0.850 (H&H OR= 0.384 [0.258-0.574], p< 0.001). Fischer scale did not reach statistical significance in predicting outcome (Fischer scale OR: 0.617 [0.374-1.018], (p=0.059)CONCLUSIONS:Our study shows that admission GCS score is best predictor of outcome in SAH patients when compared with other widely used scales (H&H and Fisher). Furthermore familiarity of GCS amongst all health care personnel makes it a better scale to follow compared to other scales which are user specific.

Disclosure: Dr. Suchdev has nothing to disclose. Dr. Doshi has nothing to disclose. Dr. Ibrahim has nothing to disclose. Dr. Seraji-Bozorgzad has nothing to disclose.