Perioperative outcome of colorectal cancer and validation of CR-POSSUM in a Caribbean country

ABSTRACT

OBJECTIVE: To evaluate the risk-adjusted perioperative outcome of colorectal cancer surgery, applying the Colorectal Physiological and Operative Severity Score for the enumeration of Mortality and morbidity (CR-POSSUM). METHODS: A retrospective chart review of patients who underwent colorectal cancer surgery from 2004 to 2007 was done. Data including demographics and physiological data for CR-POSSUM were recorded. Predicted mortality was calculated; validation of CR-POSSUM was done using Hosmer-Lemeshow goodness-of-fit and Receiver Operating Characteristic (ROC) Curve analyses. RESULTS: 232 patients were studied. The overall mean CR-POSSUM score was 18.3+/-3.8 (SD). Predicted mortality was 7.7%, observed mortality was 6.9% and the standardized mortality ratio was 0.9. 34.4% of patients presented with Duke's Stage C or D and had a higher risk of mortality (Odds Ratio (OR) 3.1, 95% Confidence Intervals (CI) 1.1, 9.1). Emergency surgery was associated with a higher risk of mortality (OR 4.7, 95% CI 1.5, 14.1). CR-POSSUM calibrated well (Hosmer-Lemeshow Chi-square value 4.3; df 8; p=0.82) and fairly discriminated outcome as shown by the area under the ROC Curve 0.69, (Standard Error 0.07). CONCLUSIONS: Perioperative outcome of colorectal surgery in Trinidad and Tobago is comparable to the developed countries as evaluated by the CR-POSSUM. Patients presenting for emergency surgery and those with advanced stages of cancer had higher perioperative mortality.