Background/aims: The aim of this study was to evaluate the predictive accuracy of P-POSSUM and O-POSSUM models on patients undergoing elective gastric resection.
Methodology: P-POSSUM and O-POSSUM predictor equations for mortality were applied retrospectively to 126 patients who had undergone elective gastrectomy for cancer. Observed mortality rates were compared with rates predicted by P-POSSUM and O-POSSUM using the Hosmer-Lemeshow goodness-of-fit test. Evaluation of the discriminative capability of both models was performed using receiver-operating characteristic (ROC) curve analysis.
Results: Overall fourteen deaths were observed. O-POSSUM predicted 15 deaths (chi2=14.61, p=0.13) and P-POSSUM predicted 20 deaths (chi2=25.41, p=0.002) using the Hosmer-Lemeshow test. ROC curves analysis revealed that O-POSSUM had better discriminatory power for mortality compared to P-POSSUM (area under curve=0.880, for O-POSSUM and area under curve=0.703 for P-POSSUM).
Conclusions: These data suggest that O-POSSUM predicts mortality more accurately than P-POSSUM in patients undergoing elective gastrectomy for cancer.