Is the ICH score a valid predictor of mortality in intracerebral hemorrhage?

Abstract

The intracerebral hemorrhage (ICH) score utilizes a 0- to 6-point scoring system to predict 30-day mortality in ICH patients. The purpose of this analysis was to (a) validate the ICH score in an international, heterogeneous population of ICH patients; and (b) assess the usefulness of a 72-h ICH score. Analyses were based on data from 399 patients in the Novo Nordisk trial F7ICH-1371. The ICH score's ability to predict mortality was determined by calculating the sensitivity, specificity, and positive predictive value (PPV). Both the baseline and 72-h ICH score had high specificity but low sensitivity resulting in an overall PPV of 57%-76%. Specificity of the ICH score was higher in the baseline ICH score (95%) as compared to the 72-h score (89%). Sensitivity of the ICH score was higher in the 72-h ICH score (75%) as compared to the baseline score (36%). The baseline ICH score provides reasonable PPV while the 72-h score provides higher sensitivity. ICH scores obtained at baseline and/or 72 h are valid and may help practitioners to more accurately predict 30-day mortality in ICH patients.