Acute left-sided malignant colonic obstruction: Is there a role for endoscopic stenting?

Abstract

The therapy of left-sided malignant colonic obstruction continues to be one of the largest problems in clinical practice. Numerous studies on colonic stenting for neoplastic colonic obstruction have been reported in the last decades. Thereby the role of self-expandable metal stents (SEMS) in the treatment of malignant colonic obstruction has become better defined. However, numerous prospective and retrospective investigations have highlighted serious concerns about a possible worse outcome after endoscopic colorectal stenting as a bridge to surgery, particularly in case of perforation. This review analyzes the most recent evidence in order to highlight pros and cons of SEMS placement in left-sided malignant colonic obstruction.

Key Words: Colorectal neoplasm; Intestinal obstruction; Endoscopy; Self expandable metallic stents; Colorectal surgery; Chemotherapy

Core Tip: Self-expandable metal stents (SEMS) should be considered as a primary option in palliative treatment of malignant left-sided colonic obstruction. In patients with conceivably curable left-sided colon cancer, SEMS placement as a bridge to surgery should be carefully discussed, specifically focusing on lower risk and lower permanent stoma rates, but potentially higher recurrence rates when compared to surgery. In this scenario the endoscopic expertise has a significant impact on the complication rate.

INTRODUCTION

Colorectal cancer (CRC) is the third most frequently diagnosed malignancy in the world and the second cause of cancer-related mortality[1]. CRC is still among the most common reason for large bowel obstruction in adults and about 20% of patients with CRC are admitted with emergency[2-4]. Obstructive CRC most frequently develops in the sigmoid colon, with 75% of tumors located distal to the splenic flexure[5]. Emergency surgery (ES) is the standard approach for obstructive right-sided colon cancer, along with primary resection and ileocolic anastomosis[6]. However, it is debatable whether emergency or radical surgery following stenting as a bridge to surgery (BTS) should be considered for obstructive left-sided colorectal cancer[7]. Self- expandable metal stents (SEMS) for BTS (Figure 1) have shown excellent short-term results, but related complications such as perforations may be disastrous and long-term outcomes are still a matter of debate[8-11].

Figure 1 Left-sided colorectal cancer obstruction treated with self-expandable metal stents. A: Obstructing cancer of the sigmoid colon; B: Endoscopic view after self-expandable metal stents (SEMS) deployment; C: Radiological view of the deployed SEMS.

STENT AS A BRIDGE-TO-SURGERY

Clinical aspects

Over the last decades, many papers have been published on colonic stenting for neoplastic obstruction, including randomized controlled trials (RCT), post-hoc analysis and systematic reviews. Moreover, in 2020 the European Society of Gastrointestinal Endoscopy (ESGE) released updated guidelines on this topic[7]. Even though the role of SEMSs in the management of malignant colonic obstruction has been better defined, several issues still remain. Although screening programs are widespread in developed countries, large bowel obstruction is one of the most common causes of ES in patients with CRC[7,12]. For example, in the United Kingdom, the rate of colorectal cancer presenting as an emergency remains at 20%[13]. Colonic SEMS placement is mainly suggested for patients who have obstructive symptoms and CT-results compatible with obstructing CRC. Acute colorectal obstruction (ACRO) is a medical emergency related to CRC that occurs more frequently in patients with advanced disease, in whom ES is responsible of significant morbidity and mortality than elective surgery, particularly in aged patients[14,15]. These patients usually present to the emergency department with nausea, vomiting, constipation and/or abdominal distention, often combined with poor intake of food from the previous days[16].

In ACRO, the main therapeutic aim is to decrease colonic distension and to prevent complications (i.e. necrosis, perforation), generally associated with pneumoperitoneum and systemic inflammatory response syndrome. Therefore, colonic stenting is an interesting option to obtain this goal in ACRO, as a BTS and for palliative purposes in patients with advanced and/or unfit for surgery CRC[7,15].

Effective stent placement makes it feasible to perform non-surgical intestinal decompression and prepare the colon for a forthcoming elective oncologic resection. Furthermore, in CRC obstruction, the proximal colon is frequently dilated with vascular insufficiency, with an increased risk of colostomy/ileostomy in case of ES. As shown in many studies, in this situation SEMSs may decompress the dilated proximal colon, thus obviating the requirement of ES with colostomy/ileostomy[17].

To evaluate the severity of obstruction, in Japan a modified point score system called ColoRectal Obstruction Scoring System (CROSS) (Table 1) is widely used. CROSS 0 patients need ES or SEMS placement. CROSS 1 or 2 patients are candidates for elective surgery. In CROSS 3 and 4 patients SEMS placement is not required because they can receive food. A post hoc analysis of two prospective, observational, single-arm multicenter clinical trials demonstrated the short-term high efficacy and safety of SEMS placement as a BTS for patients with obstructive CRC classified as CROSS 0, 1, and 2[18].

Table 1 ColoRectal Obstruction Scoring System adapted from Ohki et al[18].

Level of oral intake	Score
Requiring continuous decompression	0
No oral intake	1
Liquid or enteral nutrient intake	2
Soft solids, low-residue, and full diet with symptoms of stricure1	3
Soft solids, low-residue, and full diet without symptoms of stricure1	4

¹Symptoms of stricture include abdominal pain/cramps, abdominal distension, nausea, vomiting, constipation, and diarrhea, which are related to gastrointestinal transit.

Clinical success and adverse events

In a large cohort prospective study, the clinical success rate of SEMS placement was 95.5% and the technical success rate 97.9%. Major adverse events included perforation (2.1%), stent migration (1.0%), and stent occlusion (0.8%)[19]. The primary cause of perforation was the procedure itself (0.8%) followed by comorbidities (impending perforation, obstructive colitis) not manifest prior to SEMS insertion (0.6%). In a retrospective study, the technical success rate for stent placement for left-sided malignant colonic obstruction (LS-MCO) and rectal obstruction did not differ, but the clinical success rate was lower in patients with rectal obstruction (85.4% vs 92.1%; P = 0.02). In addition, the latter group of patients had a higher complication rate (37.4% vs 25.1%; P = 0.01), due to an increased risk of extra-intestinal cancer[20]. Furthermore, it is well established from the literature that expertise, method, lesion characteristics, and the location of the obstruction or architecture of the colon, such as tortuosity, have a significant impact on the technical and clinical failure rates for colonic stenting[7,21]. Since there have been growing concerns about protracted and technically challenging stent placement in complex patients, the Colonic Stent Safe Procedure Research Group, in collaboration with the Japan Gastroenterological Endoscopy Society, has developed mini-guidelines to ensure the procedural safety and efficacy for colonic stent placement. A post-hoc analysis[22] of a large multicenter clinical trial identified the risk factors for difficult colonic stenting cases such as a CROSS score of 0 before SEMS placement, evidence of peritoneal carcinomatosis, tumor site in the right colon, stricture length ≥5 cm and placement of multiple SEMSs[22]. In light of this evidence, Kuwai et al[22] concluded that before attempting SEMS placement for obstructive CRC clinicians must anticipate technical challenges.

The choice of the stent

Various SEMS have been developed, but they can be classified as covered and uncovered. A recent meta-analysis examined the effectiveness of uncovered vs covered stents in treating colonic obstruction either as a curative BTS or palliative option. Uncovered SEMSs presented less complications (e.g. tumor overgrowth and displacement), longer SEMS patency (mean duration 18 mo), while the risk of tumor ingrowth was higher, as expected. Rates of technical success, clinical success, perforation, stool impaction and stent obstruction were similar in both groups[21].

It is difficult to make recommendations regarding the SEMS length or diameter, as few studies have shown conflicting results. When selecting a stent after fluoroscopic measurement of colonic stricture length, it is widely accepted in clinical practice to follow a simple rule: to prepare for stent foreshortening, the distal edge of the SEMS should be placed proximal to the obstruction. Furthermore, the SEMS length should include 1-2 cm on each side beyond the stricture, considering the extent of shortening once deployed[7,17,21,23].

Is bridge-to-surgery stenting a safe alternative to emergency surgery?

Emergency surgery is burdened by high anastomotic leakage rates, up to 33%[12]. Furthermore a recent study suggests that emergency presentation remains an independent poor prognostic indicator after curative colorectal resection[24]. The optimal management of left-sided malignant large bowel obstruction is less clear than the right-sided cancer where the surgical approach is highly recommended[25].

Several surgical options exist for left-sided bowel obstruction including primary resection (with or without anastomosis), subtotal colectomy (with or without anastomosis) or unfunctioning ileostomy/colostomy with interval resection[24,25].

For the first time in 1994 Tejero et al[26] described the technique of SEMS placement in 2 patients with ACRO as a BTS. Nearly twenty years after this initial description, the debate is still open regarding the role of SEMSs as a BTS for symptomatic LS-MCO because interpretation of the literature on this subject is still challenging.

The fundamental hypotheses driving the growing interest in SEMS placement are that it can turn ES into elective surgery, reducing preoperative morbidity. Webster et al[25] analyzed 19 international guidelines for the treatment of LS-MCO from 2010 to 2018 and asked whether ES or stent placement as a bridge to surgery was the best procedure in terms of morbidity, mortality and long-term oncological outcomes. They concluded that there was a lack of high-quality evidence[25]. The more recent guidelines of the European Society of Gastrointestinal Endoscopy recommend to reserve colonic stenting in case of clinical symptoms and radiological signs of obstructing CRC, without evidence of perforation (strong recommendation, low quality evidence)[7].

In 2011, one of the first multicenter randomized trials comparing ES with colonic stenting as a BTS for left-sided CRC showed that colonic stenting had no decisive clinical advantages for global health status, mortality, morbidity and stoma rates. Moreover their results raised concerns about overt and silent perforations responsible for tumor spread[27].

A systematic review and meta-analysis of RCTs on colonic stenting as a BTS vs ES for acute symptomatic malignant left sided colonic obstruction[12] showed that patients treated with SEMS as a BTS had less short-term overall morbidity and reduced rates of both permanent and transient stoma. Albeit influenced by local expertise, level of obstruction and patient’s clinical status, stenting as a BTS for LS-MCO showed lower risk than ES in the short-term morbidity (60 d after surgery). However, recurrence rate data between the two groups showed a clear trend in favour of ES over stenting as a BTS (26% vs 40%), although this was not statistically significant.

In a subsequent multicenter randomized controlled trial (ESCO trial) comparing stenting as a BTS to ES for malignant colonic obstruction, Arezzo et al[28] reported a similar short term complications rate between the two groups but a higher stoma rate in the ES group (P = 0.031). Looking at the long term oncologic results of the ESCO trial, no difference was observed between the two groups in terms of overall survival, time to progression and disease free survival[29]. These results have also been confirmed in a more recent meta-analysis by Cirocchi et al[30].

While the majority of studies tried to understand if SEMS placement is more convenient than ES[12,31,32], there are few studies comparing the bridge to elective surgery approach such as decompressive stoma (DS) vs SEMS placement. Creation of a DS is a quite simple procedure with a near 100% success rate and can be performed in almost all patients while, as mentioned above, colonic stenting is an intervention requiring specific technical skills and expertise (in both colonoscopy and fluoroscopic techniques), including the ability to select correctly the patient based on stricture’s length and location, and carries risks of adverse events. A population-based cohort study[33] comparing the two bridge to elective surgery approaches showed that SEMS appears to be a safest procedure, with a shorter hospital admission, as well as in palliative care. In a recent meta-analysis of seven studies (1 prospective, 6 retrospective), involving 646 and 712 patients who underwent SEMS and DS approaches respectively, Zhang et al found a significantly lower complication rate in the SEMS group than in the DS group (8.68 vs 16.85%; P = 0.004), without differences in short-term mortality and permanent stoma rates. In line with the previously cited study[33], the authors concluded that SEMSs may be a better alternative to DS for obstructive CRC, but highlighted the lack of high-quality RCTs[34].

Finally, a newly published randomized trial with a longer follow-up (3 y) and larger population compared to prior studies, randomized patients with left-sided obstructive colon cancer to colonic stenting or surgical decompression. The authors showed that among patients undergoing potentially curative treatment, there were no significant differences in 30-d postoperative mortality or duration of hospital stay between stenting followed by delayed elective surgery and emergency surgery group. Moreover the use of a stoma resulted more frequent in patients treated with immediate surgery than in patients treated with SEMS (67.9% vs 47.5%; P = 0.003), without substantial differences in peri-operative morbidity, intensive care use, quality of life and 3-y recurrence or mortality[35].

Timing of surgery

The proper timing of surgery subsequent to SEMS placement as a BTS is not clear yet. Adequate radial stent expansion, ischemia reversibility of the colon proximal to the stricture and colon cleansing require sufficient time after SEMS deployment. In order to reduce the risk of stoma and postoperative complications, such as anastomotic leaks, abscesses, and wound’s problems, surgery should be postponed for at least 2 wk after SEMS placement. However, long delays in surgery could increase the complications rate related to SEMS. Therefore, surgery is suggested approximately 14 d after SEMS insertion[7,17].

STENT AS PALLIATIVE TREATMENT

Three randomized controlled trials compared SEMS and decompressive stoma as palliative treatments for malignant bowel obstructions[36-38]. Palliative situations included patients unfit for surgery, as well as patients with inoperable primary lesions or metastatic disease. Given its effectiveness and the enhanced quality of life (QoL) that comes from avoiding a stoma, colonic stenting has been judged to be superior in both investigations. In a randomized prospective trial, Fiori et al[37,38] found that the mortality and morbidity rates following palliative stenting and colostomies were comparable. However, in the stenting group a shorter hospital stay, a faster return to oral intake, and a shorter operating time were recorded. On the other hand, a Dutch trial with a similar study design was prematurely stopped because of the unacceptable high mortality rate due to perforations in the stenting group. The authors hypothesized that the unpredictable high frequency of perforation in the nonsurgical arm could be associated with the type of stent used at that time[39].

Stent and chemotherapy

Data about the effects and safety of systemic chemotherapy alone or in association with biological agents (anti-VEGF or anti-EGFR) combined with palliative stenting in metastatic colorectal cancer (mCRC) patients are lacking.

In a metanalysis including 837 mCRC patients, patients treated with SEMS had similar overall survival compared to surgery-treated patients (7.64 mo vs 7.88 mo respectively), shorter time before starting chemotherapy (33.36 d vs 15.53 d, P < 0.00001) and lower 30-d mortality (4.2% vs 10.5%, P = 0.01)[40]. Tumor response to chemotherapy could increase the rate of complications related to stent placement, such as stent migration or late perforation, but, on the other hand, could reduce the risk of obstruction by maintaining its luminal patency, especially in a palliative setting. A multicenter retrospective study included 38 mCRC patients treated with only chemotherapy; major complications related to stenting were: Perforation (8%), stent migration (5%), and re-obstruction secondary to tumor ingrowth (13%)[41]. A retrospective trial including 72 mCRC patients compared long-term outcomes of palliative SEMS in patients treated with chemotherapy or with best supportive care. In the chemotherapy group, there was a higher rate of late migration (20% vs 2.4%, P = 0.018, for chemotherapy and best supportive care group respectively); patients refractory to chemotherapy reported a higher rate of late obstruction in comparison to patients who reached disease control during treatment (35.7% in disease progression, 0% in disease control, P = 0.014)[42]. A recent metanalysis evaluated the impact of systemic treatment (chemotherapy alone or in association with targeted therapy) on the risk of complications after SEMS deployment and on outcome in terms of survival rates. Chemotherapy was shown to not be related to a higher risk of SEMS-related complications nor a reduction in the survival rates[43].

The introduction of bevacizumab improved outcome of mCRC patients[44], although data about its effect on stent placement are still controversial. Moreover, some authors raised the hypothesis of an increased risk to develop SEMS-related complications (such as perforation) in patients on bevacizumab[45,46]. Conversely, other authors demonstrated that the addition of bevacizumab to chemotherapy was not related to a higher perforation rate in comparison to chemotherapy alone[47,48]. In an Italian retrospective, multicenter study including 91 mCRC patients treated with chemotherapy plus anti-VEGF or anti-EGFR agents, no correlation between chemotherapy with or without biological therapy, K-RAS status or risk of SEMS-related complications was shown[46].

These studies had several limitations: Retrospective nature, different outcomes and small sample size, patients with heterogeneous characteristics and different settings. At the state of the art more prospective and randomized trials to define the outcome and safety of the association of SEMS placement and systemic treatment are needed.

CONCLUSION

Colonic stenting is a well-recognized palliative approach for treating malignant left-sided colonic obstruction, with high rates of technical and clinical success. Especially in patients with poor general condition and limited life expectancy, it may allow for an early hospital discharge, an improved QoL and prolonged survival in comparison to surgery.

SEMS placement as a BTS has the advantage to convert an ES into an elective one, reducing preoperative morbidity, allowing for adequate oncological staging, good colonic preparation and faster initiation of chemotherapy. Although numerous prospective and retrospective investigations have highlighted serious concerns about tumor seeding after endoscopic colorectal stent placement, particularly in cases of perforation, recent high quality studies displayed encouraging results. Operator expertise remains a key element to ensure accurate stent placement and restoration of bowel function with a low rate of complications. For this reason, this approach should be considered a standard practice only in experienced high-volume referral centers and clinicians should carefully select the patients fit for an endoscopic decompressing approach before starting the procedure.

In conclusion, further evidence from prospective, ideally randomized trials on the probability of tumor recurrence following stenting is necessary to show the long-term safety of stenting as a BTS. Until then, the evident short-term advantages, combined with the high mortality rate in frail and elderly patients, should be weighed against the potential long-term threats of tumor recurrence.