Use of Primary Anastomosis With Diverting Ileostomy in Patients With Acute Diverticulitis Requiring Urgent Operative Intervention

Abstract

Background: Previous studies suggest that urgent colectomy and primary anastomosis with diversion is safe for perforated diverticulitis. Current guidelines support this approach.

Objective: The purpose of this study was to describe the use of urgent or emergent primary anastomosis with diversion in diverticulitis before the 2014 American Society of Colon and Rectal Surgeons guidelines and compare national outcomes of primary anastomosis with diversion to the Hartmann procedure.

Design: This was a national retrospective cohort study.

Settings: The study was conducted with a national all-payer US sample from 1998 to 2011.

Patients: Patients included those admitted and treated with urgent or emergent colectomy for diverticulitis. Exclusion criteria were age <18 years, concurrent diagnosis of colorectal cancer or IBD, no fecal diversion performed, and operations >24 hours after admission.

Main outcome measures: In-hospital mortality was measured.

Results: A total of 124,198 patients underwent emergent or urgent colectomy for acute diverticulitis; 67,721 underwent concurrent fecal diversion, including 65,084 (96.1%) who underwent end colostomy and 2637 (3.9%) who underwent anastomosis with ileostomy. The rate of primary anastomosis with diverting ileostomy increased from 30 to 60 diverting ileostomy cases per 1000 operative diverticulitis cases in 1998 versus 2011 (incidence rate ratio = 2.04 (95% CI, 1.70-2.50). However, overall use remained low, with >90% of patients undergoing end colostomy. Complication rates were higher (32.1% vs 23.3%; p < 0.001) and in-hospital mortality rates were higher (16.0% vs 6.4%; p < 0.001) for primary anastomosis with diversion patients compared with end colostomy. These findings were consistent on multivariable logistic regression. Other factors that contributed to in-hospital mortality included increasing age, increasing comorbid disease burden, and socioeconomic status.

Limitations: Billing data can be inaccurate or biased because of nonmedically trained professional data entry. Selection bias could have affected the results of this retrospective study.

Conclusions: The use of primary anastomosis with proximal diversion for urgent colectomy in diverticulitis increased over our study period; however, overall use remained low. Poor national outcomes after primary anastomosis with proximal diversion might affect compliance with new guidelines. See Video Abstract at http://links.lww.com/DCR/A600.