APACHE II, SAPS 3, and SOFA scores demonstrated better accuracy in predicting in-hospital mortality (AUROC 0.850, 0.840, and 0.791, respectively) compared to biological markers such as CRP and lactate in surgical critically ill patients.
Cohort (n=847)
No
Do clinical prognostic indices (APACHE II, SOFA, SAPS 3) perform better than biological markers (CRP, albumin, lactate) in predicting mortality among surgical critically ill patients?
Clinical prognostic scores like APACHE II and SAPS 3 are superior to individual biological markers such as CRP, albumin, and lactate for predicting in-hospital mortality in postoperative critically ill patients.
Effect estimate: AUROC 0.850 (95% CI 0.824-0.873)
p-value: p=<0.0001
Several prognostic indices have been employed to predict the outcome of surgical critically ill patients. Among them, acute physiology and chronic health evaluation (APACHE) II, sequential organ failure assessment (SOFA) and simplified acute physiology score (SAPS 3) are widely used. It seems that biological markers such as C-reactive protein (CRP), albumin, and blood lactate levels correlate with the degree of inflammation during the immediate postoperative phase and could be used as independent predictors. The objective of this study is to compare the different predictive values of prognostic indices and biological markers in the outcome of 847 surgical patients admitted to the intensive care unit (ICU) in the postoperative phase.The patients were divided into survivors (n = 765, 57.4% males, age 61, interquartile range 51-71) and nonsurvivors (n = 82, 57.3% males, age 70, interquartile range 58-79). APACHE II, APACHE II death probability (DP), SOFA, SAPS 3, SAPS 3 DP, CRP, albumin, and lactate were recorded on ICU admission (first 24 hours). The area under the ROC curve (AUROC) and 95% confidence interval (95% CI) were used to measure the index accuracy to predict mortality.The AUROC and 95% CI for APACHE II, APACHE II DP, SOFA, SAPS 3, SAPS 3 DP, CRP/albumin ratio, CRP, albumin, and lactate were 0.850 (0.824-0.873), 0.855 (0.829-0.878), 0.791 (0.762-0.818), 0.840 (0.813-0.864), 0.840 (0.813-0.864), 0.731 (0.700-0.761), 0.708 (0.676-0.739), 0.697 (0.665-0.728), and 0.601 (0.567-0.634), respectively. The ICU and overall in-hospital mortality were 6.6 and 9.7%, respectively. The APACHE II, APACHE II DP, SAPS 3, SAPS 3 DP, and SOFA scores showed a better performance than CRP/albumin ratio, CRP, albumin, or lactate to predict in-hospital mortality of surgical critically ill patients.Even though all indices were able to discriminate septic from nonseptic patients, only APACHE II, APACHE II DP, SOFA and to a lesser extent SAPS 3, SAPS 3 DP, and blood lactate levels could predict in the first 24-hour ICU admission surgical patients who have survived sepsis.
Basile-Filho et al. (Sat,) conducted a cohort in Surgical critically ill patients in the immediate postoperative phase (n=847). Prognostic indices (APACHE II, SOFA, SAPS 3) and biological markers was evaluated on Accuracy to predict mortality (AUROC) (AUROC 0.850, 95% CI 0.824-0.873, p=<0.0001). APACHE II, SAPS 3, and SOFA scores demonstrated better accuracy in predicting in-hospital mortality (AUROC 0.850, 0.840, and 0.791, respectively) compared to biological markers such as CRP and lactate in surgical critically ill patients.