Research in context
Evidence before this study
We did a systematic literature search in PubMed for articles published from inception to Jan 17, 2019, with the keywords “diverticulitis”, “peritonitis”, “Hartmann*”, “primary”, and “anastomosis”, without language restrictions. We specifically included randomised controlled trials that compared Hartmann's procedure with sigmoidectomy with primary anastomosis for perforated diverticulitis with purulent or faecal peritonitis (Hinchey III or Hinchey IV); three of 127 articles identified by our search met this criterion. Quality assessment of these studies is given in the appendix (p 6). Overall, these trials randomly allocated 116 patients to primary anastomosis and 138 patients to Hartmann's procedure, of whom 204 (80%) had Hinchey III diverticulitis. All three studies were prematurely terminated, either because of slow patient accrual (two studies) or for safety reasons (one study). No significant differences in mortality or overall morbidity were reported after the index procedure or reversal procedure. Two studies found a significant difference in stoma reversal rates in favour of primary anastomosis.
Added value of this study
To our knowledge, the LADIES trial is the largest study to date on primary anastomosis in Hinchey III and Hinchey IV diverticulitis and has several methodological differences compared with previous randomised trials. First, to our knowledge, this is the first trial to report on stoma-free survival as a primary endpoint and to incorporate patient-reported outcomes. Second, the decision to construct a defunctioning ileostomy was left to the discretion of the surgeon, whereas in previous studies, by design, a defunctioning ileostomy had to be constructed in all patients undergoing sigmoidectomy with primary anastomosis. However, in one previous trial, a third of patients underwent primary anastomosis without construction of an ileostomy, thereby deviating from the study protocol. Furthermore, patients in the present study were randomly assigned after diagnostic laparoscopy, allowing for a more accurate distinction between Hinchey III and Hinchey IV diverticulitis and, consequently, this is the first study to report on outcomes in Hinchey III and IV disease separately. Finally, although not all non-included patients could be registered during the trial period, baseline demographics and preoperative disease severity data for 235 eligible non-included patients were available to compare with included patients. This comparison improves the external generalisability of our study.
Implications of all the available evidence
The LADIES trial provides strong support in favour of sigmoidectomy with primary anastomosis as the most appropriate surgical treatment for diverticulitis with purulent or faecal peritonitis in patients who are haemodynamically stable and immunocompetent. This finding is important because, in combination with existing evidence, it has the potential to fundamentally change current practice and reduce both patient and socioeconomic burden.