【Abstract】ObjectiveTo compare the reliability of acute physiology and chronic health evaluation Ⅱ (APACHE Ⅱ) and APACHE Ⅲ to estimate mortality of critical patients in abdominal surgery. MethodsTwo hundred and sixtyone critical patients in abdominal surgery were included in this study. The clinical data of the first day in ICU were collected and evaluated with both APACHE Ⅱand APACHE Ⅲ prognostic systems and statistical analysis were performed. Probability of survival (Ps) was compared with actual mortality. ResultsThe scores of APACHE Ⅱ and APACHE Ⅲ of death group were significantly higher than those of survival group respectively (P＜0.01). The actual mortality of patients whose Ps was no more than 0.5 was higher than that whose Ps was over 0.5 (P＜0.01). With two prognostic systems, the scores and mortality were the highest in pancreatitis patients and the lowest in patients with gastrointestinal malignant tumor. ConclusionAPACHE Ⅱ and APACHE Ⅲ prognostic systems can be effectively applied to the estimation of mortality of critical patients in abdominal surgery. For certain diagnostic categories, APACHE Ⅲ is better than APACHE Ⅱprognostic system.