This study demonstrated a five percent increase in DLI construction between 2007–2009 and 2016–2018. Surprisingly, a high number of DLI (71.6%) was registered from 2007 to 2009, although RECTODES-trial was not reported until 2007 . The increased usage of DLI did not reduce AL incidence nor reoperations due to AL. More than one-third of the DLI-patients in both periods had tumors located ≥ 11cm from the anal verge.
AL is deemed one of the most feared surgical complications after sphincter-preserving surgery. Significant efforts to preclude its occurrence are conducted by minimizing modifiable risk factors and implementing protective measures, including DS. There are inconsistent results on how DS affect AL rates despite well-conducted randomized controlled trials, prospective multicentre studies, and meta-analyses [20, 21]. The present national study indicates that too many AR patients receive a DLI without a beneficial effect on AL, suggesting that the selection process is too blunt. Similarly, a comparison of the Dutch TME-trial in 1996–1999 to the Dutch Surgical Colorectal Audit in 2010 demonstrated significantly increased defunctioning rates from 57–70%, albeit AL remained stable (12% vs. 11%).
Additionally, a Swedish regional study identified increase of DS construction from 15% (2002–2006) to 91% (2007–2011), the latter group probably influenced by RECTODES trial results, while AL lingered around 10% . This shift reflects the early adoption of the routine use of DLI in Sweden. There is no causal effect of DLI on AL, but our study could not detect a reduced rate of reoperations either. However, we may speculate that there might be a shift towards a lesser need for laparotomy and emergency procedures, although SCRCR data cannot prove this.
The interpretation of the results from RCTs advocating the protective role of DS must consider the circumstances that entail DS construction. In the case of the RECTODES-trial, several detrimental factors were considered. More than two-thirds of AR patients were not accepted for randomization. The most critical exclusion criterion, in our opinion, was anastomosis level > 7 cm above the anal verge or resection with a PME procedure. However, in Sweden, a high proportion (25%) of AR patients have high-located tumors (≥ 11cm from the anal verge), and about 34% subjected to PME were diverted with DLI in Sweden. This high proportion of PME is consistent with our findings which detected a diverting rate of 35% (2007–2009) and 37% (2016–2018) [22, 23]. Furthermore, the most frequent exclusion factor was intraoperative technical difficulties or intraoperative adverse events, which would create a selection bias and, consequently, decrease the external validity.
The short- and long-term stoma-related morbidities for a DS are not negligible. The few weeks after hospital discharge carry a high risk of readmission (17%) due to dehydration by high-output stoma . Moreover, the risk for chronic kidney injury (CKD) accompanying DLI is also time-related, as the incidence of severe CKD injury is higher during the first six months . Stoma reversal is another eventful step that conveys a high rate of 18–40% complications which might require reoperation in 3–8% [26, 27]. Two Swedish population-based cohort studies have investigated the permanent stoma rate, and up to 26% of AR patients would have a permanent stoma. Although AL is one of the most prominent risk factors for the permanent stoma, constructing a defunctioning stoma has no more than negligible effect on maintaining a permanent one [23, 28].
This observational study is strengthened by a large sample representing the national population of this patient group and thus enhances its generalizability and external validity. It was unbiased in selecting all consecutive patients from two periods. The study is limited by some unavailable variables, such as type of reoperation for AL, type of DS, and no follow-up of AL after 30-day, thus only reflecting the early AL rate.
Although our findings indicate that routine DLI construction is inefficient in diminishing the risk of AL in routine AR use, the role of selective DLI in mitigating AL consequences in low anastomoses should be emphasized. Other preventive measures include ghost ileostomies, an AL-check list, and scheduled postoperative AL surveillance, including sigmoidoscopy, labs, and rectography [29–31].
Thus, an urge for a decision algorithm regarding selective criteria for DLI is called for to spare DLI usage in this complex patient group with multiple risk factors for AL. Noteworthy, there is a significant shift from open to laparoscopic approach in TME-procedure. Therefore, new studies should explore the protective role of DLI specifically in the current surgical practice of laparoscopic and trans-anal TME procedures.