Prognostic Factors for Post-surgical Complications in Inflammatory Bowel Diseases: A Novel Predictive Score: ACG IBD Award

Abstract

Purpose: Despite advances in the medical management of patients with inflammatory bowel disease (IBD), surgical procedures are still frequently required. Predicting post-surgical complications and outcomes is important when deciding which patients may be best suited for surgical intervention. Several surgical risk scores have been proposed, but none of them has been specially designed for patients with IBD, which usually represent a unique group due to the heterogeneity of age, immune and nutritional status, phenotype of disease and environmental exposures. We aim to create a predictive model for risk of post-operative morbidity in patients undergoing IBD-related surgeries. Methods: Patients undergoing non-emergent intra-abdominal IBD-related surgery between January 1998 and March 2011 were included. Demographics, IBD phenotype (based on Montreal classification), nutritional status, comorbidities, laboratory parameters, histologic findings and medical treatment for IBD were considered for use in the predictive model. The primary outcome considered was development of a post-operative medical or surgical complication, defined as wound infection or dehiscence, intraabdominal abscess, anastomotic leak, urinary tract infection, pneumonia, deep venous thrombosis or death. Secondary outcomes were length of postoperative inpatient stay, post-operative need for total parenteral nutrition, and admission to an intensive care unit. Logistic regression models were used to identify individual variables and their coefficients associated with surgical morbidity. Interaction terms were considered. The final model was internally validated. Results: 91 patients were identified. 61% (56) had Crohn's disease (CD) and 39% (35) had ulcerative colitis (UC). Differences among the groups with and without morbidity are shown in Table 1. The risk score (RS) was calculated by adding the regression coefficients of the variables that were found to be predictive in the analysis (Table 2). The predictive risk of developing a post-operative complication in an IBD-related surgery was calculated using the RS in the formula presented in Figure 1. The sensitivity of the model was 86.4%, specificity was 97.1%, positive predictive value 90.5% and negative predictive value was 95.7%.Table 1: . No Caption available.Table: Risk score calculated by adding the following regression coefficientsFigure: RS: Risk score calculated with coefficients of Table 2.Conclusion: The predictive model described herein is a novel formula that may be useful in identifying which IBD patients are at risk for post-operative complications and poor outcomes. Future studies validating this score in different populations are warranted.