Comparison of the long‐term outcomes of the self‐expandable metallic stent and transanal decompression tube for obstructive colorectal cancer

Abstract Aim Endoscopic decompression using the self‐expandable metallic colonic stent (SEMS) or transanal decompression tube (TDT) can convert emergency surgery into elective one‐stage surgery for obstructive colorectal cancer (OCRC). The aim of the present study was to clarify the effect of SEMS and TDT on long‐term oncological outcomes. Methods We retrospectively analyzed 76 consecutive pathological stage II and III OCRC patients who were inserted with SEMS or TDT as a bridge to curative surgery between 2009 and 2018. Results There were 53 SEMS cases and 23 TDT cases. The tumor was located in the left colon in 58 cases and in the right colon in 18 cases. The interval between the decompression and the surgery was 16.5 days in the SEMS group and 13.0 days in the TDT group (P = 0.09). Technical and clinical success rates were 100% and 100% for SEMS, and 95% and 91% for TDT, respectively. Stoma was created in four patients in the SEMS group, and in five in the TDT group (P = 0.08). Three‐year overall survival rates of the SEMS and TDT groups were 82% and 86% (P = 0.94), and disease‐free survival rates were 68% and 62% (P = 0.79), respectively. The recurrence pattern was not significantly different. Conclusion This study found no statistically significant differences between the effects of SEMS and TDT for OCRC as a bridge to surgery on long‐term outcomes.


| INTRODUC TI ON
Intestinal obstruction is one of the common presenting symptoms of colorectal cancer. Its incidence is reported as high as 30%, 1,2 and obstructive colorectal cancer (OCRC) accounted for 85% of colonic emergency. 3,4 Emergency surgery is usually indicated and this is associated with increased morbidity, mortality, and stoma rate compared to elective surgery. Stoma creation is permanent in up to 40% of patients, and significantly diminishes patient's quality of life (QOL). 5 Further, emergency surgery might result in oncologically suboptimal resection. 6 With regard to right-sided OCRC, resection with primary anastomosis is considered the treatment of choice. 7 However, in some series, reported anastomotic leak rate was 2.5%~16.4%, 8,9 and mortality was higher compared to left-sided OCRC, 9,10 suggesting choosing safer therapeutic options might be feasible in some cases. Management options for left-sided OCRC, which accounts for 70% of OCRC, 6,11 are more diverse. The surgical strategies range from three-stage surgery (proximal colostomy, tumor resection, and stoma closure) to one-stage procedure.
To avoid anastomotic leakage, Hartmann's procedure, subtotal colectomy with ileocolic anastomosis, and segmental resection followed by primary anastomosis with diverting stoma are occasionally selected based on surgeon's preference and patient's condition. 8 Endoscopic decompression can convert emergency surgery into elective one-stage surgery. Self-expandable metallic colonic stent (SEMS) and transanal decompression tube (TDT) were both shown to be effective as a bridge to elective surgery, and associated with reduced morbidity and stoma rate compared to emergency surgery. [12][13][14] Long-term outcomes comparing SEMS and TDT have not been reported. The aim of the present study was to clarify the effect of SEMS and TDT on long-term oncological outcomes.

| Patients
We retrospectively analyzed 76 consecutive pathological stage II and III OCRC patients who were inserted with SEMS or TDT as "a bridge to surgery" at Sendai City Medical Center between 2009 and 2018. All patients subsequently underwent curative surgical resection. Patients with benign disease, distant metastasis, positive surgical margin, and invasion from a non-colonic malignancy were excluded from the study. There were 40 men and 36 women.

| Statistical analysis
Continuous variables are presented as mean ± SEM and were tested using the Mann-Whitney U-test. Associations between decompression modalities and clinicopathological parameters were evaluated in a cross-

| RE SULTS
Clinicopathological findings of the 76 patients are summarized in  As shown in Table 3 Three-year OS rates in the SEMS and TDT groups were 82% and 86% (P = 0.94), and DFS rates were 68% and 62% (P = 0.79), respectively ( Figure 1). When the cases were divided into T3 and T4, differences in OS and DFS were still non-significant ( Figure 2).
When the cases were stratified by lymph node status, OS and DFS the recurrence pattern was not significantly different (P = 0.17; Table 4).

| D ISCUSS I ON
Emergency surgery for OCRC is associated with increased morbidity and mortality compared to elective surgery. 5   Transanal decompression tube is also used as a bridge to surgery, and it was associated with improved primary anastomosis rate and reduced morbidity compared with emergency surgery. 12 Relative to SEMS, TDT is less costly, 24 and it was popular in Japan, especially until 2012, because SEMS was not covered by national health insurance.
TDT was comparable to SEMS in terms of postoperative morbidity and mortality, 24,25 but was associated with a higher permanent stoma rate. 26 Unlike SEMS, TDT does not mechanically expand the tumor, and possibly has a smaller risk of tumor spread, which might eventually affect long-term oncological outcomes. TDT might be more suitable therapeutic counterpart of SEMS than emergency surgery, since the therapeutic time course is similar. Only short-term outcomes were available comparing the effect of SEMS and TDT as a bridge to surgery 12,24,25,27 and, to the best of our knowledge, this is the first study showing long-term outcomes comparing these modalities.

Results of the present study showed that 3-year OS and DFS
were comparable between the SEMS group and the TDT group, and the recurrence pattern was not significantly different between the groups. Recent meta-analyses showed that SEMS did not adversely affect long-term results when compared with emergency surgery as a bridge to surgery, 4,28 and as palliative therapy. 29 It was also reported that incidence of local and distant recurrence was not significantly different. 4,28 Our results were in line with these previous studies, suggesting that SEMS did not adversely affect long-term oncological outcomes. TDT was equally effective in this regard, and could be another therapeutic option.
Studies on SEMS showed that perforation and subclinical silent perforation were associated with local recurrence and adverse long-term outcomes. 18,22 Morbidity rate of SEMS differed among studies, and it was suggested that SEMS should be inserted by an experienced endoscopist. 21 Reported perforation rate was 5.9%, 14 which has been decreasing, and reaching 0% in some studies 24,30 including ours. In the present study, we experienced one perforation case in the TDT group who developed peritoneal dissemination 33 months after the operation. Postoperative complication rate was comparable between the groups in this study. Matsuda et al 24 reported that SEMS was associated with reduced incidence of surgical site infection (SSI) compared to TDT. They attributed this to the high proportion of laparoscopic surgery in the SEMS group, which might explain the similar SSI rate observed in our study.
In the present study, endoscopic decompression was applied not only for left-sided OCRC but also for right-sided cases. As for right-sided OCRC, emergency colectomy with primary anastomosis is the standard of care in Western countries. 7 However, emergency colectomy for right-sided OCRC was associated with increased morbidity and mortality, 7,9,10,26 and it is a challenge for surgeons and anesthesiologists to manage the patient in a suboptimal condition in the emergency setting. In a retrospective study of 776 patients in France, postoperative morbidity and mortality rates for emergency surgery for right-sided OCRC were 51% and 10%, respectively, and age >70 years, American Society of Anesthesiology (ASA) score ≥3, and hemodynamic instability at admission were independent predictors for postoperative mortality. 9  further studies are warranted to evaluate the cost-effectiveness, and QOL.
This study was limited by the small sample size, and its retrospective, non-randomized design in a single institution. Although we limited patients to pathological stages II and III cases, heterogeneity existed in patients' backgrounds. Moreover, median follow-up time was relatively short, and there was a systematic difference in observation period between the SEMS and TDT groups (26.0 and 43.0 months as median values, respectively). The results therefore must be interpreted with caution.
In summary, the present study found no statistically significant differences between the effects of SEMS and TDT for OCRC as a bridge to surgery on long-term outcomes. Future research with a large sample size and a longer observation period is warranted to confirm the present findings. Considering QOL, the wider applicability including for the right colon, and the global popularity of SEMS make it a possibly better option, but the indications for this treatment need to be clarified. SEMS should be used in institutions with expertise in endoluminal stenting to minimize complications.

D I SCLOS U R E
Conflicts of Interest: Authors declare no conflicts of interest for this article.
Ethical Statement: The protocol for this research project was approved by the Ethics Committee of the institution (#2018-0027) and it conforms to the provisions of the Declaration of Helsinki.