gms | German Medical Science

Objective: Prediction of outcome in patients suffering from subarachnoid hemorrhage (SAH) is still uncertain and most prediction models ignore extra cerebral organ functions and pre-existing diseases. Aim of our study was to analyze the predictive value of the simplified acute physiology score II (SAPS II) a highly validated intensive care (ICU) scoring system to predict hospital mortality and of the SAPS-DRG (Diagnosis Related Groups) in a German SAH cohort.

Method: In 258 consecutive patients with aneurysmal SAH SAPS II and SAPS-DRG were retrospectively calculated for the first 24 hours after admission to the ICU. The SAPS II value spans from 0 to 154 and contains the age, heart rate, systolic blood pressure, body temperature, PaO₂ for ventilated patients, urinary output, serum urea level, white platelet count, serum potassium, sodium, bicarbonate and bilirubin level, Glasgow coma score (GCS), chronic diseases and the type of admission. SAPS-DRG contains the same parameters without GCS (value of GCS 0-26). The predictive values of SAPS II and SAPS-DRG for hospital mortality or poor outcome (Glasgow outcome scale (GOS) 1 and 2) were related to Hunt & Hess grade (H&H), gender and the highest intracranial pressure (ICP), if measured, within 24 hours after admission to ICU. SAPS II values of our cohort were divided into tertiles for further statistical analysis, 0-30 (T1), 31-45 (T2) and > 45 (T3) points, by logistic regression models and Receiver Operating Characteristics (ROC).

Results: Mean age of the 258 (180 female / 78 male) patients was 54 ± 14 years. SAPS II had a higher predictive value than SAPS-DRG with a ROC value of 0.825 vs. 0.661 for SAPS-DRG. Patients with a SAPS II value of > 45 (T3) had the highest risk for hospital mortality (Odds ratio (OR) 19.710 [5.723-67.881] p < 0.001). Beyond that age higher H&H (p < 0.001) and higher ICP in the first 24 hours (p = 0.002) were related with hospital mortality. Almost same results were calculated for poor outcome (GOS 1+2) with the highest predictive value in T3 with an OR of 24.5 [8.177-73.410] (p < 0.001). Multivariate analysis demonstrated SAPS II (p = 0.016) and Hunt & Hess V (p = 0.015) as independent predictors of hospital mortality. The combination of both predictors improved the ROC to 0.848.

Conclusions: The SAPS II is an independent and useful predictor of hospital mortality and poor outcome in SAH patients and has the potential to improve the clinical management of SAH patients.