대한대장항문학회

Purpose: The Physiological and Operative Severity Score for the enumeration of Morbidity and Mortality (POSSUM), the Portsmouth-POSSUM (P-POSSUM), and the colorectal-POSSUM (Cr-POSSUM) are relative scoring systems for the prediction of postoperative morbidity and mortality. This study is designed to evaluate the usefulness of each scoring system in elderly colorectal cancer patients undergoing major colorectal surgery.

Methods: From January 2000 to May 2008, the authors retrospectively analyzed the medical records of 251 elderly colorectal cancer patients who had undergone surgery. Collected data were analyzed using the Mann-Whitney U-test, a risk stratification analysis, and a receiver-operator characteristic (ROC) curve to evaluate the usefulness and the accuracy of each scoring system.

Results: All the predicted morbidity and mortality rates calculated by using the three POSSUM systems were higher than the observed morbidity and mortality rates. A risk stratification analysis showed a considerable correlation in risk prediction between the observed data and the calculated data. The ROC curves showed that all three POSSUM scoring systems had quite high accuracies as predictors of postoperative morbidity and mortality. POSSUM and P-POSSUM were more accurate than Cr-POSSUM.

Conclusion: All three scoring systems have a tendency for overestimation. The accuracies of POSSUM, P-POSSUM, and Cr-POSSUM as predictors are acceptance, and POSSUM and P-POSSUM are more accurate than Cr-POSSUM for prediting postoperative morbidity and mortality.