This retrospective study showed no association between the time of closure and development of major complications. However, EC is likely associated with higher occurrence of minor complications. This study is uniquely valuable because of the heterogeneous characteristics of the patients, including a wide age range, the Eastern Cooperative Oncology Group Performance Status Scale (ECOG-PS), and postoperative complications after initial surgery [12, 13]. Furthermore, the study demonstrated two intriguing results.
First, early stoma closure within 90 days after the initial operation was not associated with fatal complications such as leakage of the colorectal/coloanal anastomosis, or abdominal abscess formation. The total complication rate of Clavien-Dindo Classification III or more was 20.7% and 16.7% in the EC and LC groups (Table 3), respectively, suggesting no significant difference between the groups. Thus, we can assume that early stoma closure is a feasible management option when preoperative radiological examinations suggest no anastomotic leakage. However, one concern is that the frequency of severe postoperative complications (Clavien-Dindo Classification ≥ III) in the study is greater than that reported in previous studies [12, 13]. Danielsen reported a complication rate of 3.6% and 7% in the EC and LC groups, respectively . Inclusion of old cases and a longer observation period after stoma closure may partially explain the high incidence of postoperative complications in our study. However, further improvement in the high rate of postoperative complications in our institute is needed.
The next clinical question to be answered is, what is the best timing for stoma closure within 90 days? A clinical trial demonstrated that stoma closure at 30 days after the initial operation increased the rate of severe postoperative complications when compared with that of late stoma closure . Although the authors could not explain the reasons for their results, adhesion at 30 days after the initial operation could be worse than after a longer period, and difficulties during the operation might result in higher occurrence of severe postoperative complications. In the present study, only three patients underwent stoma reversal 20–40 days after the initial surgery. However, we should note that one of these patients developed a pelvic abscess after stoma closure. Further studies should focus on analyzing the safety of stoma closure approximately 1 month after the initial operation. We believe that elective stoma closures within 1 month should be avoided until clear evidence of their safety has been established.
The second intriguing point is the possibility of an increased incidence in minor postoperative complications, Clavien-Dindo Classification I/II, after early stoma closure. For instance, superficial incisional SSI was observed in 17.2% (5/29) and 5.6% (1/18) of patients after early and delayed closure, respectively (Table 3). One explanation for this could be that early stoma closure in our patients was performed when stoma-related complications could not be conservatively resolved rather than when the postoperative course was uneventful. Therefore, patients who underwent early stoma closure could have had a worse general condition, or local inflammation around the stoma, leading to the high incidence of minor postoperative complications. Our institution adopts a purse-string closure technique, which is associated with significantly fewer SSIs and better cosmetic outcomes with stoma reversal than those associated with conventional primary closure , thus potentially reducing the postoperative wound infection rate even after early stoma closure. Furthermore, applying anti-adhesive materials  around the stoma could reduce the technical difficulties experienced and further improve the results of early stoma closure.
The limitations of this study are the small number of patients and the unadjusted comparison between the two groups. Because the occurrence of postoperative complications is multi-factorial, a simple comparison may exaggerate or mask the differences. Thus, we focused on the statistical comparison of severe complications, resulting in the obscurity of other complications. Further, we could not assess the safety of “very” early stoma closure, within 2 weeks of the initial surgery. Although clinical trials have suggested the safety of very early stoma closure in selected patients, further studies with the inclusion of older patients are necessary before recommending early stoma closure for all patients. Furthermore, a small number of patients who underwent preoperative chemotherapy (or chemo-radiation) were included in this study. Hence, our results cannot be extrapolated in these clinical settings because of the potential risks associated with early stoma closure due to the poor clinical condition of these patients .
We are currently conducting and awaiting the results of a clinical trial investigating whether very early stoma closure within 2 weeks of initial surgery is safe. (UMIN ID: 000036382, registered on 03/04/2019).