Applicability of the Colon Leakage Score as an anastomotic leakage predictor in colorectal cancer surgery

Abstract

Introduction: Anastomotic filtration increases morbidity and mortality in colorrectal surgery. Identification of risk factors and creation of a predictive model would help the decision of creating a defunctionalizing ostoma, that currently is taken by the surgeon. Dekker created de Colon Leakege Score (CLS) with objective criteria.

Objective: Establish CLS in patients that underwent left colon and rectum surgery with cancer diagnosis in Clinica INDISA, define the local cutting value, it’s specificity and sensibility. Patients and Methods: Corresponds to a diagnostic test’s study, that intervention is CLS application, comparing with the presence of anastomotic filtration (gold standard), defined by clinical and radiologic criteria. For the analysis, ROC curves, Youden’s index and logistic regression.

Results: From 180 patients, anastomotic filtration was present in 12 (6.6%). Average CLS score in patients with anastomotic filtration was 11.5 and in those without anastomotic filtration was 6.9 (p = 0.0001). Area under the curve for anastomotic filtration prediction using CLS was 0.829 (CI 95% 0.69-0.96) with a cutting value of 11, 67% of sensibility and 89% of specificity. Logistic regression analysis, OR for anastomotic filtration prediction using CLS was 1.48 (CI 95% 1.22-1.79 p < 0.001).

Conclusion: CLS is a tool that permits predicting anastomotic filtration risk in patients that underwent left colon and rectum surgery. With a CLS value equal or more than 11, we should create a protective ostoma, generating a clinical practice local change.