Comparative study between colonic metallic stent and anal tube decompression for Japanese patients with left-sided malignant large bowel obstruction

Malignant large bowel obstruction is caused by a variety of advanced malignancies, and has traditionally been approached surgically with colonic resection and possible stoma creation or large intestine bypass. However, surgical management of malignant large bowel obstruction carries with high morbidity and mortality, high stoma creation rate, and prolonged hospital stay. Recently, colonic metallic stent (CMS) as a bridge to surgery and also for palliation for malignant large bowel obstruction, particularly left-sided malignant large bowel obstruction (LMLBO) caused by colorectal cancer, has been reported to be both a safe and feasible option [1]. However, CMS is not recommended routinely by the European Society of Gastrointestinal Endoscopy guidelines because of its lack of safety [2]. In Japan, placement of a trans-anal decompression tube (TDT) has traditionally been used for LMLBO as a bridge to surgery. Since CMS for malignant large bowel obstruction has been covered by insurance since 2012 in Japan, its feasibility has recently been evaluated. However, the clinical efficacy of CMS for LMLBO is not clear as there are few reports of comparative outcomes between CMS and TDT for LMLBO [3, 4].

The aim of this retrospective study is to evaluate the surgical outcomes and short- and mid-term results of CMS for Japanese patients with LMLBO compared with TDT.

We retrospectively evaluated surgical outcomes and short- and mid-term results of 26 patients with CMS for LMLBO as a bridge to surgery (CMS group, treated between 2013 and 2015) and compared with those of 33 patients with TDT (control group, treated between 2003 and 2011). In this study, LMLBO was defined as a large bowel obstruction caused by a colorectal cancer that was diagnosed by computed tomography in which urgent decompression was deemed necessary.

Regarding TDT insertion, the ArgyleTM Denis Colorectal tube (Medtronic Corp.) was used. We forcibly irrigated the large bowel using water twice daily for approximately 1 week until a planned surgery. Regarding CMS placement, we treated LMLBO with either the WallFlexTM Colonic Stent 22 mm (Boston Scientific Corp.) or the Niti-STM 22 mm (TaeWoong Corp.). Both TDT insertion and CMS placement were performed by colorectal surgeons. The insertion technique was via a combined endoscopic and fluoroscopic approach.

Postoperative complications were defined according to the Clavien-Dindo classification system. A hospital stay was defined as a duration to discharge after surgery in this study.

Malignant large bowel obstruction is caused by a variety of advanced malignant tumors, particularly pancreatic, gastric, colorectal, and peritoneal carcinomatosis with an ovarian primary. Malignant large bowel obstruction caused by colorectal cancer occurs in approximately 20% [5]: 9.0–17.8% in Japan. Traditionally, malignant large bowel obstruction caused by colorectal cancer has been approached surgically. For right-sided malignant large bowel obstruction, right hemicolectomy is performed, while for LMLBO, staged surgeries are usually necessary because the mortality for emergency surgery is much higher than that for elective surgery [6]. However, stoma that was created for initial decompression become permanent in patients with LMLBO because of the operative risk, advanced age, and patients’ unwillingness to proceed with further surgery. One-stage surgery with decompression including intraoperative lavage is thus an appealing option [7]. CMS and TDT strategies have been developed in order to achieve primary anastomosis without stoma at the initial surgery. Systematic reviews and meta-analyses have demonstrated the superiority of CMS for malignant large bowel obstruction as a bridge to surgery in terms of improved morbidity and shorter length of stay, among other benefits [8‐11]. Currently, CMS is considered a feasible option for malignant large bowel obstruction as a bridge to surgery. As CMS for malignant large bowel obstruction has been covered by insurance since 2012 in Japan, its feasibility has been evaluated in the literature. Recently, two retrospective studies of comparative outcomes between CMT and TDT for LMLBO were reported in Japan, but the benefits were not clear. Kawachi et al. [3] reported that treatment with CMS had benefits in terms of decreased stoma creation rate, as well as high rates of technical and clinical success of the stenting procedure itself. However, significant differences were not found in terms of mortality, morbidity, and shorter hospital stay compared with TDT. Additionally, in the report by Matsuda et al. [2], there were no differences between the two groups in terms of stoma creation rate, mortality, and morbidity. The CMS group did, however, demonstrate higher QOL including shorter postoperative hospital stay, higher rates of solid food intake, and temporary discharge prior to surgery. On the other hand, Li et al. in China compared the TDT group (n = 13) with the CMS group (n = 16) for acute LMLBO to evaluate the clinical effects. They concluded that both TDT and CMS can achieve preoperative colonic lavage for 1-stage operation for patients with acute LMLBO with no increase in complications [12]. However, these results were retrospective, single-center, and were carried out with a relatively small group of patients.

Optimizing technical success and minimizing perforation are critical if applying CMS to patients with malignant large bowel obstruction. A meta-analysis by Allievi et al. [10] demonstrated that technical success rate and perforation rate using CMS were 78.8% and 5.9%, respectively. In this study, perforation was found only in the TDT group (n = 3, 9.1%), while a technical success rate of 100% was observed in the CMS group. Also, all patients treated with CMS reported higher rates of solid food intake and temporary discharge prior to surgery compared with patients treated with TDT. However, the occurrence of postoperative complications more than grade 2 and permanent stoma creation rate were equivalent. It was advantageous that all patients treated with CMS were able to initiate solid food intake and were able to be discharged from the hospital for a short time when compared to patients treated with TDT. Additionally, surgeries performed after CMS had more complete pathologic staging in terms of more resected lymph nodes. However, this study was a retrospective study, comparing the CMS group with the TDT group by the historical study. Therefore, there were differences in the background between two groups.

In the CMS group, there were no peritoneal recurrences that could be associated with technical failure or perforation during CMS insertion, and only one patient had a local recurrence after surgery. The 5-year overall survival rate, including the analysis of 9 patients with stage IV disease, and disease-free survival rate of pathological stage II and III patients in the CMS group were similar to those in the TDT group. Recently, the multicenter, randomized controlled ESCO trial showed there was no difference in oncologic outcomes with a median follow-up of 36 months [9]. Also, the meta-analysis by Matsuda et al. showed no significant difference between the CMS group and the emergency surgery group in terms of overall survival, disease-free survival, and recurrence. On the other hand, Sabbagh et al. [13] reported negative outcomes including ulceration near the tumor, perineural invasion, and lymph node invasion associated with CMS placement and that overall survival was significantly lower in the CMS group. Broholm et al. [14] reported that delay of surgery after stent placement for resectable malignant colorectal obstruction was associated with a higher risk of recurrence. Takahashi et al. reported that CMS placement increased plasma levels of cfDNA and ctDNA by tumor manipulation despite no management with TDT [15]. The oncologic consequence of CMS placement for MLBO remains unclear.

This study indicates that treatment with CMS for LMLBO may have clinical benefits of its safety, higher rates of resumption of solid food intake, and temporary discharge prior to elective surgery compared to treatment with TDT. However, this study is limited by its small sample size, single-center retrospective design, and non-randomized nature. Furthermore, more than one type of stent was used. Future, randomized controlled trials are needed to clarify the superiority of treatment with CMS compared to TDT.