The accuracy of ICH-FOS for spontaneous intracerebral hemorrhage: An analysis of ATACH-2 and the Qatar stroke database

Abstract

Introduction

Spontaneous intracranial hemorrhage(sICH) comprise between 15–20% of stroke with high rates of mortality and disability. Multiple scoring systems have been developed for sICH prognostication.

Aims

We explored the Qatar stroke database and the ATACH-2 database to study the independent predictors of 90-day mortality and poor functional outcomes[modified rankin score(mRS) 4–6] in patients with sICH. We compared admission national institute of health stroke scale (NIHSS) and Glasgow coma scale (GCS) with original intracerebral hemorrage(o-ICH), modified(m)ICH, ICH-GS(grading scale), Lanseed(LS)-ICH, Max-ICH, and ICH-FOS(functional outcome score) scores to evaluate 90-day mortality and functional outcomes.

Methods

Data on baseline characteristics, premorbid condition, laboratory tests, imaging findings, and surgical treatment were included. Outcomes were mortality and poor functional outcome(mRS 4–6) at 90-day follow-up. Prognostic accuracy of each score was assessed using Receiver Operating Characteristic(ROC) curve analysis. ICH-FOS risk categories were created for mortality and poor functional outcomes.

Results

There were 1660 patients with sICH admitted to Hamad General Hospital available for analysis. Mean age 49(SD 12) years, median GCS 15(IQR 11–15), median NIHSS 12(IQR 5–19). At 90-day follow-up, 124(11.3 %) died and 396(36 %) had poor functional outcome. ATACH included 1000 non-anti-coagulant related sICH patients with supratentorial bleeds without intra-ventricular extension(IVH). Mean age 62(SD 12) years, median GCS 15(IQR 13–15), median NIHSS 11(IQR 6–16). At 90-day follow-up, 54(5.9 %) patients died and 342(37.3 %) had poor functional outcomes. Patients in ATACH-2 were older, had 4-fold increased odds of having a prior stroke, and were 4-times more likely to be smokers. There were more lobar bleeds in Qatar and a higher proportion of patients in Qatar died at 90-days(11.3 v 5.9, p < 0.001). No difference in the rate of poor functional outcomes was noted. ICH-FOS had the largest AUC in ATACH(0.83, 95 % CI 0.77–0.89) and Qatar(0.81, 95 % CI 0.77–0.84) databases for predicting 3-month mortality, and similarly for unfavourable functional outcomes(mRS 4–6), with AUC’s in ATACH(0.83, 95 % CI 0.79–0.85) and Qatar(0.85, 95 % CI 0.82–0.87), respectively.

Conclusion

ICH-FOS was the most accurate predictor of 90-day mortality and poor functional outcome in both Qatari and ATACH-2 patients with non-anticoagulant related supratentorial sICH without IVH. Our study extends the utility of the ICH-FOS from a Chinese population to both Middle-Eastern and Western populations.