Comparing the performance of SOFA, TPA combined with SOFA and APACHE-II for predicting ICU mortality in critically ill surgical patients: A secondary analysis

Summary Introduction Total psoas muscle area (TPA) can indicate the status of the entire human body's skeletal muscle mass. It has been reported that lower TPA can increase the risk of mortality in critically ill patients. The aim of our study was to evaluate the relationship between TPA and ICU mortality and to compare the performance of Sequential Organ Failure Assessment (SOFA), TPA combined with SOFA and Acute Physiology, Chronic Health Evaluation (APACHE-II) for predicting ICU mortality in critically ill surgical patients. Methods This study was a retrospective observational cohort study with a total of 96 critically ill surgical patients, ages 21–96 years old. Main outcome measures included difficult-to-wean (DTW), operation methods, ICU mortality, ICU stay, APACHE II, sepsis and SOFA. CT-scan assessed the TPA. It is acknowledged that the entire study was completed by Hao-Wei Kou et al. and the data were uploaded from plosone.com . The authors used this data only for secondary analysis. Results The results showed that TPA is a protective factor for ICU mortality (OR: 0.99 95% [0.99, 1.00], P = 0.0269). In addition, when we defined sarcopenia-based TPA, our study showed that sarcopenia increased the risk of ICU mortality (OR:3.73 (1.27, 10.98) P = 0.0167. Furthermore, discrimination of ICU mortality was significantly higher using SOFA (AUROC, 0.7810 [99% CI, 0.6658–0.8962]) than either TPA (AUROC, 0.7023 [99% CI, 0.5552–0.8494]) or APACHE II score (AUROC, 0.7447 [99% CI, 0.6289–0.8604]). Additionally, when we combined TPA with SOFA score, the ROC of TPA + SOFA (AUROC, 0.8647 [99% CI, 0.7881–0.9412]) was the highest when compared to the other three models. Conclusion The relationship between TPA and ICU mortality is negative in critically ill surgical patients. In addition, the combination of TPA and SOFA was the best tool among the three scoring systems in providing significant discriminative ability when predicting ICU mortality in critically ill surgical patients.