The Effect of Surgical Training and Operative Approach on Outcomes in Acute Diverticulitis: Should Guidelines Be Revised?

Abstract

Current guidelines accept partial colectomy and primary anastomosis with proximal diversion for select patients with perforated diverticulitis based on low-quality evidence. This study aimed to compare the effect of operative approach and surgeon training on outcomes following urgent/emergent colectomy for diverticulitis. This is a statewide retrospective cohort study. Data were obtained from the New York State all-payer sample from 2000 to 2014. All patients who underwent an urgent/emergent sigmoid colectomy for diverticulitis with creation of an end colostomy or primary anastomosis with proximal diversion were included. We excluded all patients age <18 years, with IBD, colorectal cancer, ischemic colitis, or elective operations. The main outcomes measured were postoperative in-hospital mortality and complications, RESULTS:: A total of 10,780 patients underwent urgent/emergent colectomy for diverticulitis: 10,600 (98.3%) received a Hartmann procedure and 180 (1.7%) received primary anastomosis with proximal diversion. Colorectal surgeons performed 6.0% of all operations. Utilization of primary anastomosis with proximal diversion was greater among colorectal surgeons but remained low overall (4.2% vs 1.5%; p < 0.001). Postoperative mortality was 2-fold greater when noncolorectal surgeons performed primary anastomosis vs Hartmann procedure (15% vs 7.4%; p < 0.001) and 1.4 times greater among noncolorectal surgeons than among colorectal surgeons (7.5% vs 5.3%; p = 0.04). On multivariable logistic regression (adjusting for patient demographics/characteristics, year, hospital academic status, and surgeon training) primary anastomosis with proximal diversion remained associated with increased mortality (OR, 2.7; 95% CI,1.7-4.4; p < 0.001), complications (OR, 1.8; 95% CI, 1.3-2.5; p < 0.001), and reoperation (OR, 3.4; 95% CI, 1.8-6.3; p < 0.001), whereas colorectal board certification was associated with decreased mortality (OR, 0.66; 95% CI, 0.46-0.95; p = 0.03). Selection bias secondary to retrospective nature and absence of disease severity were limitations of this study. Despite current recommendations for primary anastomosis with proximal diversion for perforated diverticulitis, this operation in New York State was associated with increased postoperative morbidity and mortality when performed by general surgeons. Given that the majority of urgent/emergent colectomies for diverticulitis are not performed by colorectal surgeons, guidelines for operative management of perforated diverticulitis should be reevaluated. See Video Abstract at http://links.lww.com/DCR/A772.