Background: There is no dedicated scoring system for predicting the risk of surgical-site infection (SSI) after resection of the colon or rectum. Generic scores, such as the National Nosocomial Infections Surveillance index, are not used by colorectal surgeons.
Methods: Multivariable analysis of risk factors for SSI was performed in patients who underwent resection of the colon or rectum, and were followed during the first month after operation. A logistic regression model was used to identify determinant variables and construct a predictive score.
Results: There were 534 patients of whom 114 (21·3 per cent) developed SSI. In multivariable analysis, four parameters correlated with an increased risk of SSI: obesity (odds ratio (OR) 2·93, 95 per cent confidence interval 1·71 to 5·03), contamination class 3-4 (OR 3·33, 2·08 to 5·32), American Society of Anesthesiologists grade III-IV (OR 1·82, 1·14 to 2·90) and open surgery (OR 2·22, 1·01 to 4·88). Each of these contributed 1 point to the risk score. The observed risk of SSI was 5 per cent for a score of 0, 12·0 per cent for a score of 1 point, 18·7 per cent for 2 points, 44 per cent for 3 points and 68 per cent for 4 points. The area under the receiver operating characteristic curve for the score was 0·729.
Conclusion: A simple clinical score based on four preoperative variables was clinically useful in predicting the risk of SSI in patients undergoing colorectal surgery.
Copyright © 2012 British Journal of Surgery Society Ltd. Published by John Wiley & Sons, Ltd.