Colonic cancer: The current role of stent insertion

Abstract

Introduction

Colorectal cancer is the third most common cancer in men and second most common in women; it accounts for 8% of all cancer deaths globally.1 Eight to thirteen percent of all colonic cancers present with large bowel obstruction.2,3 Of these obstructing cancers 75% are distal to the splenic flexure; the sigmoid colon being the most common location.4 The traditional management of malignant colonic obstruction is surgical resection (with primary anastomosis or stoma formation) which has both high morbidity (40%–50%) and mortality (15%–20%).5,6 In patients unfit for surgical resection, a defunctioning stoma only was typically performed. Endoscopically placed self expandable metallic stents (SEMS) have been increasingly used to relieve malignant enteral obstruction (oesophageal, biliary). SEMS have been used in colonic obstruction since the early 1990s, as both a bridge to elective surgery and as palliation. Early cohort studies suggested that SEMS use was associated with reduced mortality, morbidity and colostomy rates when compared with emergency surgery.7–10

Randomized controlled trials (RCTs), and meta-analyses have demonstrated convincing benefits for the use of SEMS in the palliative setting, but not where the SEMS has been used as a bridge to surgery in potentially curative cases.11 Importantly, in the latter scenario, concern has been raised following analysis of RCT follow-up data regarding the oncological impact of SEMS use.12–14 It is only recently that there has been some clarity as to the most appropriate clinical indications for SEMS in malignant colonic obstruction.15

Proximal obstructing cancers (proximal to splenic flexure) are generally treated with resectional surgery.4,16 Although there are no RCTs to support this practice, surgery for right-sided colonic obstruction has lower morbidity and mortality than that for left sided obstruction, and SEMS placement is more challenging.15 Rectal cancers have been excluded from all but one of the published RCTs on stenting. Stenting for low rectal cancer is associated with stent migration, tenesmus, pain and incontinence. Thus, much of the evidence applies to left sided colonic (distal to splenic flexure) cancers only.

The European Society of Gastrointestinal Endoscopy (ESGE) has published a clinical guideline on the use of SEMS in malignant colonic obstruction.15 This guideline is based on evidence and consensus opinion. General considerations based on clinical consensus rather than robust evidence are also included.

General Considerations and Technique

Contrast enhanced computed tomography (CT) scan is advised to diagnose malignant colonic obstruction (sensitivity 96%, specificity 93%).15 CT is also useful for accurately defining the level (94%) and aetiology (81%) of the obstruction, as well as local and distant disease staging in the majority of patients.15–17 CT scanning will also reliably demonstrate if there is perforation at the site of tumour or caecum when stenting would be inappropriate. The diameter of the caecum also provides an indication of the urgency of decompression; caecal perforation is much more likely if there is a closed loop obstruction and when the diameter of the caecum is 14 cm or greater.

ESGE advise bowel preparation with an enema in order to clean the colon and rectum distal to the stricture prior to colonic SEMS insertion.15 Oral bowel preparation is relatively contraindicated with symptomatic bowel obstruction.

Colonic SEMS can be inserted using either through-the-scope (TTS) or the over-the-guidewire (OTW) technique. The TTS technique utilizes combined endoscopy (usually a double-channel gastroscope) and fluoroscopy to locate and cannulate the stricture. OTW technique uses fluoroscopic guidance and can be performed with additional endoscopic visualization or purely radiologically (uses a stiff guide-wire). Comparable success rates of combined endoscopic and fluoroscopic guidance and purely fluoroscopy have been demonstrated by retrospective studies.18–20 However there is a trend towards better success rates with the combined technique hence ESGE recommend colonic SEMS placement with combined endoscopy and fluoroscopy.15

ESGE discourage dilatation of malignant colonic stricture, either before or after SEMS placement.15 It is suggested that dilatation may adversely affect patient outcomes following SEMS.21–23 Systematic review and meta-analysis of retrospective data suggests an increased risk of perforation colonic perforation with stricture dilatation.24,25

Enteral SEMS may be covered or uncovered. In the setting of malignant colonic obstruction; the two designs are equally safe and effective. Two recent meta-analyses26,27 found similar technical success rate, clinical success and complication rates. Covered SEMS suffer higher slippage rates (21.3% vs 5.5%), but significantly lower tumour ingrowth (0.9% vs 11.4%).26,27

The size of the stent is important. Stent diameter of less than 24 mm increases the risk of stent related complications (perforation, re-obstruction and migration), especially stent migration.21,22 Whilst stent length per se does not seem to be important, the ESGE guidance is that the stent should traverse the stricture such that it extends 20 mm either side (after deployment).15

As with all technical procedures, there is a learning curve for colonic SEMS placement. Retrospective series22,28 and non-comparative studies29,30 have shown that stent outcome is affected by operator experience. Technical success rate (and use of fewer stents) significantly improves after 20 procedures.29,30 Thus colonic SEMS insertion should be performed or supervised by an operator with experience of 20 or more cases.15

Synchronous colorectal cancers occur in 3% to 4% of patients,31,32 thus expedient (within three months) visualization of the remaining colon after potentially curative resection for malignant obstruction is recommended.15 Preoperative CT colonography and colonoscopy through the stent both appear to be safe and feasible.33–35

SEMS as a Bridge to Elective Surgery

The European Guidelines state that colonic SEMS placement is not recommended for routine use in the treatment of left-sided colonic malignant obstruction as a bridge to elective surgery.15 This guideline is based on the evidence from eight systematic reviews with meta-analysis comparing preoperative stenting with emergency resection for left-sided colonic malignant obstruction,11,36–42 with assimilated data from up to seven RCTs.43–49 Three of those RCTs were closed early due to adverse outcomes in the SEMS group (higher 30-day morbidity related to stent perforation),43,44 or high anastomotic leak rate in the emergency surgery group.48 The most recently published meta-analysis found that SEMS as a bridge to surgery (n = 195) compared to emergency surgery (n = 187) had a lower overall morbidity (33.1% vs 53.9%, P = 0.03), a higher successful primary anastomosis rate (67.2% vs 55.1%, P < 0.01), and a lower permanent stoma rate (9% vs 27.4%, P < 0.01); however, there was no statistical difference in post-operative mortality (10.7% vs 12.4%). This data suggests some advantages of SEMS use over emergency surgery, although with no difference in short-term mortality, the benefit of SEMS placement in this setting is less clear.

Importantly, there is some limited evidence that there may be worse long-term oncological outcomes after SEMS use. Early studies suggested an oncological compromise associated with tumour perforation or tumour cell dislodgement during SEMS insertion resulting in seeding of malignant cells.13 Manipulation of the tumour by the shearing forces of endoscopy and colonic insufflation is thought to disseminate tumour cells locally and into the circulation.43 Three RCTs have investigated long-term oncological outcome, all the sample sizes were small (15–26 patients in the SEMS groups), and found increased disease recurrence in the SEMS groups.12,46,48 Sloothaak et al12 reanalyzed the largest data set (from the Stent-In-2 trial in 2014). They excluded the benign cases and those with advanced disease (palliative). Median follow-up was 36 months. Loco-regional recurrence was reported in 9/32 (28%) in the surgery group and 13/26 (50%) in the SEMS group (8/20 unperforated SEMS [40%] and 5/6 perforated SEMS [83%]). Five-year recurrence was 25% in the surgery group and 42% in the SEMS group (P < 0.01). Cumulative incidence of overall disease recurrence was 26% in the surgery group, 34% in the unperforated SEMS group and 83% in the perforated SEMS group (P < 0.01).12 The concern therefore is that short-term gains would be negated by poorer long-term outcomes. If true, this would be an inappropriate trade off in patients fit enough to undergo surgery in whom the intent is curative. None of the studies have thus far, shown a reduced survival associated with increased disease recurrence.

Morbidity and mortality of emergency colonic resection remains high5,6 thus it is important to balance the long-term oncological risks of SEMS against the risks of surgery. In patients in whom the risk of surgery is higher than average (for example American Association of Anesthesiologists classification [ASA] III, or age > 70 years),50,51 SEMS as a bridge to elective surgery may still be appropriate management.15 It can convert an emergency into an elective operation and provide time for preoperative optimization of comorbidities and improvement in physiology. There is minimal data to suggest what the ideal time interval is between SEMS placement and elective surgery. The balance between allowing adequate recovery from the episode of bowel obstruction and improving nutritional status must be weighed against the increasing risk of stent complications with time. The ESGE guidelines suggest 5 to 10 days.15 An interval of 10 to 28 days is usual, but longer periods may be considered if the obstruction has coincided with serious medical complications, such as myocardial infarction.

Palliative SEMS

ESGE state that colonic SEMS placement is recommended in the palliation of malignant left-sided colonic obstruction.15

In the palliative setting, the clinical priorities for managing malignant colonic obstruction are clearly different to the potentially curative setting. The objectives of SEMS placement are shorter hospital stay, reduced morbidity and mortality and an improved quality of life. SEMS should be an ideal management option for obstructed patients who are unfit for surgery and those with either an unresectable primary tumour, and/or incurable metastatic disease and therefore reduced life-expectancy. An early RCT published in 2008 reported early and late stent related colonic perforation as up to 18% and 36% respectively raising concerns about SEMS use on this group.52

More recently two meta-analyses have assessed SEMS versus surgery in the setting of palliation of malignant left-sided colonic obstruction, in terms of short-term and long-term complications, hospital stay, 30-day mortality, and stoma rate.53,54 The larger study included data from 13 studies (n = 837; SEMS group n = 404, surgery group n = 433).54 Successful relief of obstruction was high in both groups, but lower in the SEMS group (99.8% vs 93.1%, P < 0.001). In SEMS treated patients, hospital stay was significantly shorter (18.8 days vs 9.6 days, P < 0.00001), and 30-day mortality was significantly lower (4.2% vs 10.5%, P = 0.01). Stoma formation rate is an important consideration in the palliative setting due to impact on quality of life. Permanent stoma rate was significantly lower in the SEMS group (13.0% vs 54.0%, P < 0.001). Stent related complications were common and included perforation (10.1%), migration (9.2%), and re-obstruction (18.3%). Surgery related complications included anastomotic leak (4.7%) and wound infection (5%). The overall complication rate was similar between these two groups (SEMS: 34.0% vs surgery: 38.1%, P = 0.60), but the stent-related complications occurred later than the surgery-related complications (early complications: 13.7% vs 33.7%, P = 0.03; late complications: 12.7% vs 32.3%, P < 0.0001). Overall survival time of SEMS- and surgery-treated patients was similar (7.6 months vs 7.8 months).54 The evidence is sufficiently strong for the benefit of stenting in this situation that this service should be available to patients at the time of need.

Colonic SEMS placement in the palliative setting is associated with some important adverse events which can be divided into early and late complications. Early complications include perforation (range 0%–12.8%), stent failure (range 0%–11.7%), stent migration (range 0%–4.9%), re-obstruction (range 0%–4.9%), pain (range 0%–7.4%), and bleeding (range 0%–3.7%).15 Late complications include perforation (range 0%–4.0%), re-obstruction (range 4.0%–22.9%), and stent migration (range 1.0%–12.5%).15 Perforation may be associated with guide-wire or catheter positioning, dilatation of the malignant stricture during stent placement, stent induced direct or local perforation, or proximal colonic perforation due to excessive insufflation or inadequate decompression.15,55 Early perforation necessitates emergency surgery and may be associated with high mortality (reported as 50% in prospective and retrospective studies.56,57 Stent migration and re-obstruction (caused by either tumour in-growth or inadequate stent expansion) can usually be managed endoscopically. Stent-in-stent placement has reported success rates of 75% to 86% but there is a lack of data on long-term outcomes.21

Chemotherapy is known to prolong survival in palliative colorectal cancer. Shorter hospitalization and better 30-day mortality rate in SEMS use allows earlier start of chemotherapy treatment. Early evidence regarding chemotherapy use in stented patients was contradictory21,58 but there was concern that chemotherapy use may increase perforation rate. Retrospective series suggested an increased perforation rate in those treated with bevacizumab (an angiogenesis inhibitor) (17% 50%).22,57,59 A recent meta-analysis investigated risk factors for increased perforation rate.25 Chemotherapy without bevacizumab was not associated with a significantly increased risk (7.0%), but use of bevacizumanb was (12.5%). Perforation rate in those who did not receive chemotherapy was 9.0%.25

Conclusions

The technique for colonic SEMS insertion in malignant colonic obstruction has been established, and technical and clinical success rates are high. There is now a well established role for SEMS use in patients with left sided malignant colonic obstruction treated with palliative intent with significant reductions in morbidity, mortality and stoma rates. Stenting related complications in this group are significant and are in the order of 30%. It is not known whether covered or bare metal stents are better in the palliative setting.

The evidence from RCTs has shown some benefit for SEMS as a bridge to surgery in patients with potentially curative disease, with a reduction in stoma rates, but there are concerns about the oncological safety. As such, current guidelines recommend that this treatment be used only when patients are considered to be a high surgical risk.

Article information

References