Validation of a Mayo post-operative mortality risk prediction model in Korean cirrhotic patients

Abstract

Background Patients with cirrhosis have an increased risk of mortality after surgery. In 2007, a new model was suggested to calculate mortality risk at specific time points after surgery at the Mayo clinic. Aims We investigated the mortality risks in Korean cirrhotic patients who underwent various surgeries and applied the Mayo clinic model to our study populations. Methods We conducted a retrospective review of the charts of 160 patients with cirrhosis who underwent surgical procedures under general anaesthesia between January 1996 and December 2006 at two hospitals. Results The overall 30-, 90-day and 1-year mortality rates were 7.5, 9.4 and 10.6% respectively. In multivariate analysis, the Child-Turcotte-Pugh (CTP) score, model for end-stage liver disease (MELD) and the American Society of Anesthesiologists (ASA) physical status classification and age were significantly associated with mortality. The area under the receiver operating characteristic (AUROC) from the calculated value using Mayo model as a predictor of 30-, 90-day and 1-year mortality was 0.832, 0.803 and 0.822 respectively, of which, 1-year mortality was significantly different from AUROC of mortality prediction based on our patient's data (P=0.025). In addition, the mean of predicted 1-year mortality rate (22.6 +/- 12.0%) using Mayo model was significantly higher than that from observed (8.9 +/- 1.4%, P < 0.01). Conclusions The CTP score or MELD score or ASA physical class and age were found to be significant predictors of post-operative mortality in cirrhotic patients. The risk prediction model developed at the Mayo clinic showed good performance in Korean cirrhotic patients. However, we found that the model tended to overestimate mortality, especially 1 year after surgery.