Predication of post-operative outcome of colectomy in ulcerative colitis patients using Model of End-Stage Liver Disease Score.

PURPOSE

Model of End-Stage Liver Disease (MELD) score was developed to predict mortality in patients with liver disease. The aim of this study was to investigate the relationship between preoperative MELD score and 30-day surgical outcomes using the American College of Surgeons National Surgical Quality Improvement Program.

METHODS

Patients with ulcerative colitis (UC) (ICD: 556.X) who underwent colectomy were identified from NSQIP 2005 to 2013. The primary outcomes were bleeding complications, and overall morbidity and mortality.

RESULTS

A total of 7534 UC patients undergoing colectomy were identified. Patients with a higher MELD score had a longer hospital stay; more bleeding; and cardiac, respiratory, renal, thromboembolic, and septic complications as well as mortality. Patients were stratified into 4 groups by MELD score: < 7, 7-11, 12-15, and > 15 and a stratified multivariate analysis was done. Patients with a MELD score 12-15 (odds ratio [OR] 1.9, 95% confidence interval [CI] 1.1-1.3) and MELD > 15 (OR 2.6, 95%CI 1.5-4.7) were at significant risk for bleeding complication. Apart from the MELD score, the presence of ascites (OR 2.5, 95%CI 1.2-5.1) or varices (OR 1.0, 95%CI 1.01-1.03) was also significantly associated with post-operative bleeding complication. MELD 12-15 and MELD > 15 were also found to be risk factors for overall morbidity (OR 5.3, 95%CI 1.8-15.7; OR 10.3, 95%CI 3.6-29.7, respectively) and mortality (OR 3.3, 95%CI 1.3-8.4; OR 5.9, 95%CI 2.4-14.6, respectively).

CONCLUSION

UC patients with a higher MELD score were associated with a higher post-colectomy morbidity and mortality. MELD score > 11 was an independent indicator for post-operative bleeding, and overall complications and mortality.