Prediction of postoperative complications and optimization of surgical technique

BACKGROUND: Potential benefits and risks of new operation techniques can show up differently in the individual patients. The aim of this pilot study was to verify on the basis of reliable prediction of postoperative complications the possibility to create model for optimization of the operation technique in the cohort of patients operated for rectal carcinoma. MATERIAL AND METHODS: The study involved patients operated in the course of five years for rectal carcinoma by means of laparoscopic or open technique. The mathematical models of prediction of postoperative complications of individual operative techniques were derived from the Physiological and Operative Severity Score for enUmeration of Mortality and Morbidity scoring system (POSSUM). The reliability of prediction for the given technique was tested and the patients of the cohort were analyzed with regard to the "suitability" of the operation technique used. RESULTS: The study involved 91 patients operated using open technique and 67 patients operated laparoscopically with the occurrence of 45% and 39% of the postoperative complications respectively. The statistically relevant variables tested for the occurrence of postoperative complications were the "cardiac symptoms" and "severity of the surgery" for open operations, and "leukocytes" and "severity of the surgery" for the laparoscopic operations. The prediction models based on these variables showed statistically high reliability. The complications in the entire cohort would in case of ideal selection of surgical technique drop by 36%. CONCLUSION: Reliable prediction of the postoperative complications can be potentially effective tool to optimize surgical treatment for an individual patient.