Assessment and Comparison of Scoring Systems—SAPS III, APACHE II, and SOFA for Prediction of Mortality in Patients with Sepsis in a Mixed Medical and Surgical Indian ICU

Abstract

Background: Sepsis is a leading cause of intensive care unit (ICU) morbidity and mortality. Severity-of-illness scoring systems aid risk stratification, but most were derived from Western cohorts and require validation in Indian ICU settings.
Aim: To assess and compare the accuracy of the Simplified Acute Physiology Score (SAPS) III, Acute Physiology and Chronic Health Evaluation (APACHE) II, and Sequential Organ Failure Assessment (SOFA) scores for predicting mortality in patients with sepsis admitted to a mixed medical–surgical Indian ICU.
Methods: This prospective, single-centre, observational study enrolled 118 adults with sepsis admitted to a 40-bed mixed medical–surgical ICU over 18 months. SAPS III, APACHE II and SOFA scores were calculated within 24 hours of admission. Discrimination was assessed using the area under the receiver operating characteristic curve (AUROC), and sensitivity, specificity, positive and negative predictive values (PPV, NPV) and accuracy were computed at the optimal Youden’s index cut-off for each score.
Results: Overall in-hospital mortality was 39.8% (47/118). Mean SAPS III, APACHE II and SOFA scores were significantly higher among non-survivors than survivors (61.68 vs 49.23, 21.98 vs 16.39, and 7.91 vs 5.39 respectively; p<0.001 for SAPS III and SOFA, p=0.001 for APACHE II). The AUROC for SAPS III (0.73) was higher than that of APACHE II (0.66) and SOFA (0.66). At the optimal cut-off (≥60), SAPS III showed sensitivity 51.06%, specificity 85.92%, PPV 70.59%, NPV 72.62% and accuracy 72%. Presence of comorbidity was significantly associated with mortality (p=0.023), and lower Glasgow Coma Scale, higher serum creatinine and higher bilirubin were significantly associated with death.
Conclusion: SAPS III demonstrated better discriminatory ability than APACHE II and SOFA for predicting mortality in this mixed medical–surgical Indian ICU cohort of septic patients, though its sensitivity remained modest. A combination of scores, along with clinical judgement, is recommended for prognostication in resource-limited ICU settings.