Hartmann's procedure versus primary anastomosis for Hinchey III or IV sigmoid diverticulitis: a propensity score analysis of a retrospective national cohort study.

INTRODUCTION

Primary anastomosis with a diverting stoma (PA) has been evaluated relative to Hartmann's procedure (HP) for Hinchey III/IV peritonitis in four randomized trials, all of which lacked power. We aimed to evaluate the safety of PA for this indication in real-life settings.

PATIENTS AND METHODS

Patients who underwent surgery for Hinchey III/IV diverticulitis at 43 French centers from 2010 to 2021 were included in this retrospective multicenter study. The primary endpoint was the rate of severe complications at 90 days (Clavien-Dindo III-V). Secondary endpoints included the overall complication rate and stoma-free survival at one year. Two groups of patients (PA vs. HP) were created by propensity score matching (PSM).

RESULTS

In total, 651 patients were included (112 with PA and 539 with HP). We matched 102 of the patients with PA with 102 patients who underwent HP. On day 90, there was no difference in the rate of severe complications (31% vs. 29%, p = 0.75) between these groups in subgroup analyses for Hinchey III (p = 0.99) or Hinchey IV (p = 0.35) diverticulitis. The percentages of patients with Hinchey III or IV diverticulitis did not differ significantly between the groups (SMD = 0.03). After inverse probability treatment weighting, mortality was significantly higher in the HP group than in the PA group (6% vs. 1%, p = 0.019). Stoma-free survival at one year was higher in the PA group (85% vs. 73%, p = 0.03). The reoperation rate at 90 days was higher in the PA group than in the HP group (unadjusted: 21% vs. 13%, p = 0.028; PSM: 21% vs. 7%, p = 0.007, respectively).

CONCLUSION

PA did not increase the overall rate of complications. Conversely, it decreased mortality and the rate of definitive stomas in a large cohort of patients with Hinchey III/IV sigmoid diverticulitis.