Examination of prognostic factors in patients undergoing surgery for colorectal perforation: A case controlled study

OBJECTIVE: To determine if the POSSUM, SOFA, MPI, and SAS scores provide a better measure of severity for patients with prognostic factors undergoing surgery for colorectal perforation. SUBJECTS: Fifty-nine patients who underwent surgery between 1996 and 2012. METHODS: We retrospectively reviewed background factors, blood and physiological test results, and intraoperative findings of patients who survived and those who died. We also compared the POSSUM, SOFA, MPI, and SAS scores. Multivariate analysis was performed for factors that were significant by univariate analysis, and selected factors were used to produce a predictive prognostic model. RESULTS: Univariate analysis revealed significant differences in age, anticoagulant/steroid administration, serum creatinine level, PF ratio, base excess (BE), chest radiography, pulse rate, and severity of peritoneal soiling. Age, serum creatinine level, pulse rate, and severity of peritoneal soiling were selected for multivariate analysis; only pulse rate was significantly different. There were significant differences between the two groups in POSSUM PS, OSS, SOFA, and MPI scores, and a comparison in terms of the ROC curve showed that our model had the highest peak; the area under the curve was 94.8% compared with 70-80% for the other systems, suggesting that our model is better than those systems. CONCLUSIONS: POSSUM and SOFA are valid methods of evaluating risk from colorectal perforation, but our study revealed addition risk factors: (1) the PF ratio and BE, which are not included in POSSUM; (2) the pulse rate and severity of peritonitis, which are not included in SOFA; and (3) anticoagulant/steroid hormone administration.