Does one score fit all? Measuring risk in ulcerative colitis

Background The American College of Surgeons Surgical Risk Calculator was developed to improve risk stratification and surgical quality but has not been studied at the institutional level for specific disease states, like ulcerative colitis (UC). Methods UC patients undergoing colorectal resection had predicted risk calculator data compared with actual outcomes for length of stay (LOS), complications, reoperation, and death. Main outcome measures were the difference in actual vs predicted outcomes. Results Seventy patients were evaluated. The actual and predicted mean LOS was identical, but not representative of the actual LOS picture, which had 10 LOS outliers (14.3%). The actual incidence of any complication (P < .001) and major complications (P < .001) was higher than predicted. The most common complications actually encountered-intrabdominal abscess (14.3%), postoperative ileus (7.2%), and anastomotic leak (5.7%), were not even calculated by the tool. Conclusions For UC, the calculator poorly evaluates relevant risks, complications, and is greatly impacted by outliers. These limitations caution use for surgical quality reporting and determining specific patient outcomes, at least in UC.