WHEN IS DIVERSION INDICATED AFTER RIGHT SIDED COLON RESECTION?
Nicholas P. McKenna*, Katherine Bews, Kellie L. Mathis, Elizabeth B. Habermann
General Surgery, Mayo Clinic , Rochester, MN
Background: Ileocolic anastomoses have a lower risk of anastomotic leak (AL) than other more distal anastomoses resulting in lower utilization of diversion. However, some patients perceived to be at high risk of AL are still diverted, and exposed to the negative sequelae of an ileostomy including an increased risk of dehydration and an additional operation for reversal. It is challenging in practice to identify patients at high enough risk of AL to justify diversion secondary to the low overall rate and the many proposed risk factors. Therefore, we used a multicenter database to develop and validate an anastomotic leak risk calculator that can assist surgical decision-making on when diversion should be used.
Methods: The American College of Surgeons National Surgical Quality Improvement Program Colectomy Targeted PUF was queried from 2012-2017 for patients undergoing elective right-sided colectomy for cancer, benign neoplasm, or Crohn’s disease (CD). Multivariable logistic regression identified independent predictors of AL in patients who were not diverted. Point values were assigned to these variables based on their maximum likelihood estimates to develop a risk score. Boot-strap analysis (200 replications) was performed for internal validation. Visual correspondence between predicted anastomotic leak rate and actual anastomotic leak rate and 95% confidence interval (CI) was assessed within each risk score decile. Lastly, risk scores were calculated for patients who were diverted at the time of elective right-sided colectomy.
Results: A total of 24,056 patients (53% female) with a median age of 66 years (range, 18-90) underwent right-sided resection without diversion. 2.6% of patients experienced an anastomotic leak, and independent risk factors for AL included a diagnosis of CD, male sex, immunosuppressant use, bowel preparation other than combination mechanical and antibiotic, an open approach, and a wound class of III or IV (TABLE). A risk calculator incorporating these variables showed excellent calibration and fair discrimination (raw c-index 0.65, boot-strap validated c-index 0.64) for prediction of anastomotic leak. There was strong visual correspondence with the predicted risk within the 95% CI of actual AL rate for nine of the ten deciles. The 95% CI for risk decile 10 overlapped the CIs for risk deciles 8 and 9 (FIGURE). An additional 530 patients underwent elective right-sided colectomy and were diverted, but 78% of these patients would have been in risk decile 8 or less indicating a predicted AL rate of less than 6%.
Conclusion: Applying the validated scores revealed that the majority of diverted patients are likely not at high enough risk to justify diversion and that even patients in the highest risk decile may not warrant diversion. Therefore, diversion after right-sided resections should be reserved for extreme circumstances.
Point values assigned to variables in final risk score.
Comparison between mean model-predicted and observed (95% confidence interval) anastomotic leak rate across the risk score deciles in non-diverted patients.
Back to 2020 Abstracts