Delayed anastomotic leakage following laparoscopic intersphincteric resection for lower rectal cancer: report of four cases and literature review

The introduction of intersphincteric resection (ISR) is one of the recent advances in the surgical treatment of lower rectal cancer. ISR is a surgical technique to preserve sphincter function that was first described by Schiessel et al. [1]. Several studies have demonstrated the acceptable outcomes of ISR in terms of morbidity, oncologic safety, and postoperative anal functions, and ISR has been proposed as an alternative to abdominoperineal resection (APR) for selected patients with lower rectal cancer [2‐5].

Anastomotic leakage (AL) is one of the most dreadful postoperative complications of colorectal cancer because it can result in increased morbidity and mortality as well as poorer long-term prognosis. Reported incidence rates of colorectal AL vary between 3 and 20% [6‐8]. Although most studies of AL have limited their investigation time to a period of postoperative 30 days, some studies have shown that AL can occur more than 30 days postoperatively [9‐14]. To date, there is no consensus as to an exact definition of delayed AL, and there is little information on delayed AL. Delayed AL has been defined as AL diagnosed after hospital discharge [11] or as AL diagnosed more than 30 days postoperatively [12‐14]. Shin et al. proposed the following criteria for delayed AL: (1) AL was detected more than 3 weeks postoperatively, (2) a normal diet and defecatory function was resumed within 1 week of surgery, (3) AL developed without the occurrence of any signs or symptoms of peritonitis within the postoperative 3 weeks, and (4) no local recurrence developed during the follow-up period [10].

Here, we reported four cases of patients with delayed AL following laparoscopic ISR and conducted a review of the literature about delayed AL following colorectal surgery. To the best of our knowledge, this is the first report of delayed AL following ISR.

The gold standard of surgical technique for rectal cancer is total mesorectal excision (TME), which results in improved survival and reduced local recurrence. In recent years, ISR for lower rectal cancer has been performed in selected patients as an alternative to APR. ISR involves the transanal division of the distal rectum, removal of part or all of the internal anal sphincter, and restoration of bowel continuity by performing handsewn coloanal anastomosis. By performing TME simultaneously, ISR is thought to afford adequate oncological resection margins while preserving sphincter function [2‐5]. In addition, laparoscopic ISR is touted as a minimally invasive technique [5].

AL is one of the most serious complications following colorectal surgery. While AL is commonly believed to occur within 30 days postoperatively, recent studies have shown that AL can occur beyond the 30-day postoperative period. Here, we reported four cases of patients with delayed AL following laparoscopic ISR. In all four cases, AL occurred more than 1 month after surgery: postoperative months 4, 9, 14, and 57. The literature search yielded only a few English-language publications on delayed AL following colorectal surgery (Table 1). Importantly, there was no previous report on delayed AL following ISR. The incidence rate of delayed AL was reported to be relatively high (i.e., 0.3–4.3%), approximately one third of all AL cases [9‐14]. A retrospective review of our prospective database from July 2005 to June 2015 suggested that a total of 41 rectal cancer patients underwent laparoscopic ISR at our institution and that the incidence rate of delayed AL was 9.8% (4/41), whereas that of early AL (within less than 30 days postoperatively) was 0% (0/41). Preoperative chemoradiotherapy or chemotherapy was not performed in these 4 patients with delayed AL, indicating delayed AL was not associated with preoperative chemoradiotherapy or chemotherapy in this series (Table 2). In addition, no correlation was found in terms of sex, UICC-TNM stage, and lateral lymph node dissection (Table 2). In that same period, 179 patients with rectal cancer underwent laparoscopic low anterior resection (LAR) with double stapling technique anastomosis at our institution. Regarding laparoscopic LAR, early AL occurred in 23 patients (23/179: 12.8%), while delayed AL did not occur at all. These findings suggest that delayed AL cannot actually be considered a rare complication; therefore, surgeons should provide long-term follow-up and remain alert to the possible development of delayed AL.

There is a lack of understanding as to whether or not delayed AL is different from early AL. Reported risk factors for early AL following rectal surgery are low level of anastomosis, male gender, and the presence of intraoperative adverse events [6‐8], which may correlate to the degree of surgical difficulty. In terms of the timeframe in which delayed AL develops, delayed AL does not seem to be attributable to technical factors, but rather to other predisposing factors. According to the findings of previous reports, there were no obvious differences in patient characteristics and surgical factors between early AL and delayed AL [10‐14]. The main difference may lie in the extent of leakage, i.e., more severe leakages give rise to symptoms earlier, whereas less severe leakages take longer to develop. The causes or predisposing factors associated with delayed AL have not yet been elucidated. Tan et al. reported that patients with delayed AL were much younger and more prone to present with fistulas compared to those with early AL, while no significant difference was found between the two groups in terms of other factors including gender, body mass index, smoking, hypertension, preoperative albumin, and duration of surgery [13]. Floodeen et al. reported that leakage from the anterior side of the circular stapler line was more common in patients with delayed AL than in those with early AL and that there was a larger proportion of an anastomotic-vaginal fistula in delayed AL [11]. Shin et al. reported that the independent risk factors for delayed AL were female gender, low-level anastomosis, and preoperative chemoradiation therapy and that the rate of anastomotic-vaginal fistula was relatively high (42%) in delayed AL [10]. Recently, Lim et al. reported that delayed AL following LAR was associated with preoperative radiotherapy, fistula formation, and the less frequent need for reoperation [14]. In the procedure of ISR, the perineal approach (i.e., intersphincteric dissection and coloanal anastomosis) is commonly performed by direct vision. However, the field of vision in the perineal approach is poor, especially at the anterior side, rendering it one of the most difficult parts of the ISR procedure. In our cases, all four patients with delayed AL presented with fistulas at the anterior side of the coloanal anastomosis: two anastomotic-urethral fistulas, one anastomotic-vaginal fistula, and one anastomotic-perineal fistula. These occurrences may be related to the technical difficulty caused by poor visualization of surgical field at the anterior side. To resolve this problem, a promising alternative approach can be transanal TME. The transanal approach in ISR can provide better surgical field especially at the anterior side, which may reduce the incidence of postoperative complications, such as delayed AL.

We report four cases of delayed AL following laparoscopic ISR with a review of the literature. Delayed AL is not a rare postoperative complication, and therefore, surgeons should provide long-term follow-up and remain alert to the possible development of delayed AL.