Endoscopy 2016; 48(10): 939-948
DOI: 10.1055/s-0042-114210
Guideline
© Georg Thieme Verlag KG Stuttgart · New York

Esophageal stenting for benign and malignant disease: European Society of Gastrointestinal Endoscopy (ESGE) Clinical Guideline

› Author Affiliations
Further Information

Corresponding author

M.C.W. Spaander, MD PhD
Department of Gastroenterology and Hepatology
Room Hs-308
Erasmus Medical Centre Rotterdam
PO Box 2040
3000 CA Rotterdam
3015 CE Rotterdam
The Netherlands
Fax: +31-10-7035172

Publication History

Publication Date:
14 September 2016 (online)

This Guideline is an official statement of the European Society of Gastrointestinal Endoscopy (ESGE), endorsed by the European Society for Radiotherapy and Oncology (ESTRO), the European Society of Digestive Endoscopy (ESDO), and the European Society for Clinical Nutrition and Metabolism (ESPEN). The Grading of Recommendations Assessment, Development, and Evaluation (GRADE) system was adopted to define the strength of recommendations and the quality of evidence.

Main recommendations for malignant disease

1 ESGE recommends placement of partially or fully covered self-expandable metal stents (SEMSs) for palliative treatment of malignant dysphagia over laser therapy, photodynamic therapy, and esophageal bypass (strong recommendation, high quality evidence).

2 For patients with longer life expectancy, ESGE recommends brachytherapy as a valid alternative or in addition to stenting in esophageal cancer patients with malignant dysphagia. Brachytherapy may provide a survival advantage and possibly a better quality of life compared to SEMS placement alone. (Strong recommendation, high quality evidence.)

3 ESGE recommends esophageal SEMS placement as the preferred treatment for sealing malignant tracheoesophageal or bronchoesophageal fistula (strong recommendation, low quality evidence).

4 ESGE does not recommend the use of concurrent external radiotherapy and esophageal stent treatment. SEMS placement is also not recommended as a bridge to surgery or prior to preoperative chemoradiotherapy. It is associated with a high incidence of adverse events and alternative satisfactory options such as placement of a feeding tube are available. (Strong recommendation, low quality evidence.)

Main recommendations for benign disease

1 ESGE recommends against the use of self-expandable stents (SEMSs) as first-line therapy for the management of benign esophageal strictures because of the potential for adverse events, the availability of alternative therapies, and costs (strong recommendation, low quality evidence).

2 ESGE suggests consideration of temporary placement of SEMSs as therapy for refractory benign esophageal strictures (weak recommendation, moderate evidence).

Stents should usually be removed at a maximum of 3 months (strong recommendation, weak quality evidence).

3 ESGE suggests that fully covered SEMSs be preferred over partially covered SEMSs for the treatment of refractory benign esophageal strictures, because of their lack of embedment and ease of removability (weak recommendation, low quality evidence).

4 For the removal of partially covered esophageal SEMSs that are embedded, ESGE recommends the stent-in-stent technique (strong recommendation, low quality evidence).

5 ESGE recommends that temporary stent placement can be considered for treating esophageal leaks, fistulas, and perforations. The optimal stenting duration remains unclear and should be individualized. (Strong recommendation, low quality evidence.)

6 ESGE recommends placement of a SEMS for the treatment of esophageal variceal bleeding refractory to medical, endoscopic, and/or radiological therapy, or as initial therapy for patients with massive esophageal variceal bleeding (strong recommendation, moderate quality evidence).

#

Abbreviations

BMI: body mass index
CI: confidence interval
CSEMS: covered self-expandable metal stent
ESDO: European Society of Digestive Oncology
ESGE: European Society of Gastrointestinal Endoscopy
ESPEN: European Society for Clinical Nutrition and Metabolism
ESTRO: European Society for Radiotherapy and Oncology
FCSEMS: fully covered self-expandable metal stent
GEJ: gastroesophageal junction
HR: hazard ratio
PCSEMS: partially covered self-expandable metal stent
PEG: percutaneous endoscopic gastrostomy
PDT: photodynamic therapy
RBES: refractory benign esophgeal stricture
RR: risk ratio or relative risk
RCT: randomized controlled trial
SEMS: self-expandable metal stent
SEPS: self-expandable plastic stent
TIPS: transjugular intrahepatic portosystemic shunt
QoL: quality of life

Introduction

Esophageal cancer is the eighth most common cancer worldwide with an estimated 456 000 new cases and 400 000 deaths in 2012 [1]. More than 50 % of patients with esophageal carcinoma have metastatic disease at the time of diagnosis. Dysphagia is the most common symptom of incurable obstructive esophageal cancer and can be treated by esophageal stent placement. In recent years different designs of esophageal stents have emerged for improving dysphagia and quality of life in patients with malignant esophageal tumor, malignant fistula, or extrinsic compression [2] [3]. Esophageal stent placement in patients with incurable esophageal cancer is aimed at maintaining oral intake and improving quality of life, but it carries a risk of adverse events such as hemorrhage, pain, and fistula [4]. The current variety of commercially available stents for malignant disease comprises uncovered self-expandable metal stents (SEMSs); fully covered self-expandable metal stents (FCSEMSs), in which the entire length of the stent is covered; partially covered self-expandable metal stents (PCSEMSs), in which the proximal and distal ends of the stent are devoid of a covering; and fully covered self-expandable plastic stents (SEPSs). All currently available SEMSs are made of nitinol, a nickel and titanium alloy. In Europe, the types of stents that are predominately used in the treatment of malignant dysphagia are PCSEMSs and FCSEMSs.

Esophageal stents are also commonly used for the treatment of benign esophageal diseases, albeit most stents are not officially approved for this indication. Common and investigative indications include treatment of refractory benign esophageal stricture (RBES), sealing of perforations, leaks, and treatment of acute esophageal variceal bleeding. FCSEMS and PCSEMS as well as SEPS are used for this indication, but only the latter has formal approval for RBES. All stents used for benign esophageal conditions should be removed, except for the self-expandable biodegradable stents that have recently become available in Europe.

This clinical Guideline aims to describe the role of esophageal stents in patients with malignant or benign esophageal disease and makes recommendations on circumstances that warrant their use.

#

Methods

The European Society of Gastrointestinal Endoscopy (ESGE) commissioned this Guideline and appointed a Guideline leader (M.J.B.) who invited the listed authors. The key questions were prepared by the coordinating team (M.C.W.S., J.-M.D., C.H., M.J.B.) and then approved by the other members (see Appendix e1, available online). The coordinating team established task force subgroups, each with a leader (P.D.S. for malignant disease and T.H.B. for benign disease), and divided the key topics among these task forces.

Each task force performed a systematic literature search to prepare evidence-based and well-balanced statements on their assigned key questions. All selected articles were graded for the level of evidence and strength of recommendation according to the Grading of Recommendations Assessment, Development and Evaluation (GRADE) system [5]. The numbers of articles retrieved and selected by each task force are indicated in the evidence tables (see [Tables e1 – e6] in Appendix e2, available online).

Each task force proposed statements for their assigned key questions, which were discussed during a meeting in Amsterdam (April 2015). In August 2015, a draft prepared by the coordinating team was sent to all group members. It was also sent for review and endorsement to the European Society for Radiotherapy and Oncology (ESTRO), the European Society of Digestive Oncology (ESDO), and the European Society for Clinical Nutrition and Metabolism (ESPEN). The manuscript was reviewed by two members of the ESGE Governing Board and sent for further comments to the National Societies and ESGE Individual Members. After agreement on a final version, the manuscript was submitted to Endoscopy for publication, All authors agreed on the final revised manuscript.

This Guideline was issued in 2016 and will be considered for review and update in 2021 or sooner if new and relevant evidence becomes available. Any updates to the Guideline in the interim will be noted on the ESGE website: http://www.esge.com/esge-guidelines.html.

#

Recommendations and statements

ESOPHAGEAL STENTS IN MALIGNANT DISEASE

ESGE recommends placement of partially or fully covered self-expanding metal stents (SEMSs) for palliation of malignant dysphagia over laser therapy, photodynamic therapy, and esophageal bypass (strong recommendation, high quality evidence).

ESGE recommends against the placement of nonexpandable and expandable plastic stents for the palliation of malignant esophageal strictures (strong recommendation, high quality evidence).

Efficacy

Photodynamic therapy (PDT), laser therapy, and esophageal bypass have not been shown to be superior to SEMS placement for the palliation of malignant dysphagia in several randomized controlled trials (RCTs) [6] [7] [8] [9] [10] [11]. From 1993 up to 2005 several RCTs have compared SEMS versus rigid plastic stents [12] [13] [14] [15] [16] [17] [18]. One of the largest published RCTs including 217 patients [17] showed a better improvement in dysphagia score at 1 and 6 weeks with SEMS compared to rigid plastic stents and fewer late adverse events. Systematic reviews and meta-analyses showed that SEMS insertion was superior to rigid plastic stents in terms of improvement and recurrence of dysphagia, as well as occurrence of adverse events including perforation and migration [19] [20].

Multiple types of self-expandable stents are available. They differ in terms of design, luminal diameter, radial force, flexibility, and degree of shortening after deployment. In Europe, partially or fully covered SEMS are used for the treatment of malignant dysphagia because recurrent dysphagia due to tumor ingrowth has been a major drawback of uncovered SEMSs [21]. In most cases a 100 % technical success rate of stent placement has been reported with an improvement in dysphagia score of at least 2 points (from 3 [liquids only] to 1 [almost all solids]) within 1 – 2 days [20]. Most new stent designs have been evaluated in single-arm prospective or retrospective series. SEPS are similar to SEMS with regard to relief of dysphagia in the short term, but adverse events occurred more often with SEPS, especially migration, making SEMS preferable over SEPS for malignant dysphagia [22].

Safety

The most prevalent adverse events following stent placement are shown in [Table e7] (Appendix e3, available online) and details are also presented in Appendix e3. Analysis of pooled data from RCTs and prospective and retrospective studies showed that major adverse events occur in 18 %, 21 %, and 10 % of patients with PCSEMS, FCSEMS, and SEPS, respectively, while recurrent dysphagia develops in 41 %, 29 %, and 37 % of these patients, respectively [22] [23] [24] [25] [26] [27] [28] [29] [30] [31] [32] [33] [34] [35] [36] [37] [38] [39]. Stent insertion-related mortality is 0 % – 2 % [23] [40].

For patients with longer life expectancy, ESGE recommends brachytherapy as a valid alternative or in addition to stenting in esophageal cancer patients with malignant dysphagia. Brachytherapy may provide a survival advantage and possibly a better quality of life compared to SEMS placement alone. (Strong recommendation, high quality evidence.)

Two RCTs have compared SEMS versus brachytherapy. One RCT compared a PCSEMS (Ultraflex) with single-dose intraluminal brachytherapy in 202 patients with incurable esophageal cancer [4]. Compared to SEMS placement, brachytherapy improved dysphagia less rapidly, but after 1 month from treatment, dysphagia score improvement no longer differed significantly between stent placement and brachytherapy. With respect to survival, patients treated by brachytherapy had more days with almost no dysphagia during follow-up than those treated by stent placement. In addition, major complications (i. e., perforation, hemorrhage) occurred more frequently after stent placement than after brachytherapy. There was no difference in recurrent dysphagia and median survival. Quality-of-life (QoL) scores significantly favored brachytherapy, whereas total costs were similar across the two groups. In the other RCT (n = 65), insertion of SEMS offered a more immediate relief of dysphagia compared to brachytherapy, but quality of life was better with brachytherapy for patients with longer survival [41]. The main limitations of brachytherapy include limited availability, technical difficulty, and need for dedicated logistics and expertise. Therefore this treatment can only be considered in dedicated centers.

Esophageal stent placement for malignant tracheoesophageal or bronchoesophageal fistula

Esophageal SEMS placement is recommended as the preferred treatment for sealing malignant tracheoesophageal or bronchoesophageal fistula (strong recommendation, low quality evidence).

Application of double stenting (esophagus and airways) can be considered when fistula occlusion is not achieved by esophageal or airway prosthesis alone (strong recommendation, low quality evidence).

Malignant tracheoesophageal or bronchoesophageal fistula develops in 5 % to 15 % of patients with esophageal cancer and in less than 1 % of patients with lung carcinoma [42] [43]. Because of advances in palliative treatment, the incidence has increased over the last 30 years to above 10 % among all nonresected esophageal cancers [44].

Tracheoesophageal or bronchoesophageal fistulas are usually late developments of advanced cancer of the esophagus, lung, or mediastinum, caused by tumoral invasion or as an adverse event of cancer therapies, in particular chemoradiotherapy [45] [46] [47]. Importantly, the condition of such patients has often already significantly deteriorated when they develop a fistula and the remaining life expectancy is short (weeks to months). Rapid relief of disabling symptoms due to the fistula, preferably by minimally invasive treatment, is thus of pivotal importance in order to improve quality of life.

Esophageal stenting is the most widely used approach [48]. Multiple studies using SEMSs for sealing off esophageal – airway fistulas have reported improvement in symptoms and sealing of the fistula in 75 % – 100 % of patients [2] [42] [43] [49] [50] [51] [52] [53] [54] [55]. Application of double stenting (esophagus and airways) can be considered when fistula occlusion is not achieved by esophageal or airway prosthesis alone [51] [56] [57] [58]. In the largest prospective series, Shin et al. successfully placed SEMSs in 61 patients with malignant esophageal – airway fistulas, sealing off the fistula in 49 patients (80 %), while 10 patients (16 %) required concomitant airway stents [42]. Re-opening of the fistula occurred in 17 patients (35 %); of these, 8 were successfully re-treated with SEMSs, in 2 patients the fistula was closed spontaneously, and 7 patients did not undergo further treatment [42].

Procedure-related complications are reported in 0 % – 27 % of patients with a mortality rate of 0 % – 12 % [42] [43] [49] [50] [52] [54].

In another study that compared quality of life following placement of a SEMS versus gastrostomy or jejunostomy or best supportive care, quality of life improved more with SEMS placement, particularly for symptoms of dyspnea, dysphagia, other eating problems, dry mouth, cough, and hypersalivation [59]. In three large retrospective studies, esophageal stent placement was associated with a significant improvement in survival compared with no sealing of the fistula, a feeding gastrostomy or jejunostomy, or best supportive care [42] [43] [50].

#

Stent placement for malignant dysphagia as a bridge to surgery

ESGE does not recommend SEMS placement as a bridge to surgery or prior to preoperative chemoradiotherapy. It is associated with a high incidence of adverse events, and other satisfactory options such as placement of a feeding tube are preferable. (Strong recommendation, low quality evidence.)

It is now accepted that neoadjuvant chemotherapy or chemoradiotherapy should be administered to all patients with a resectable esophageal cancer, except for cancers staged 0 – IIA [60] [61] [62]. In a systematic review and meta-analysis of 9 studies (n = 180 patients) on esophageal stenting preceding or concomitant with neoadjuvant chemotherapy for esophageal cancer, the procedural success rate was 95 % (95 % confidence interval [95 %CI] 90 % – 98 %) [63]. There was a significant decrease in dysphagia score and a nonsignificant increase in patient weight (0.6 kg) and serum albumin. However, major adverse events were extremely frequent, including stent migration (incidence 32 %, 95 %CI 26 % – 40 %) and chest discomfort (incidence 51.4 %, 95 %CI 21 % – 81 %). SEPS were used in 5 of the 9 studies (41 % of patients). The negative impact on oncologic outcome of SEMS placement as bridge to surgery was also confirmed in a large European cohort of 2944 patients [64]. This study showed an in-hospital postoperative mortality and morbidity rate for the SEMS versus control groups of 13.2 % versus 8.6 % and 63.2 % versus 59.2 %, respectively. In addition, significant differences in R0 resection (71.0 % vs. 85.5 %), median time to recurrence (6.5 vs. 9.0 months), and 3-year overall survival (25 % vs. 44 %) were found, to the disadvantage of the SEMS group. The results remained significant after excluding SEMS-related esophageal perforations and after adjusting for confounding factors. Similar unfavorable results have been reported with biodegradable stents as a bridge to surgery [65].

Profound weight loss and malnutrition as a consequence of severe dysphagia and cancer cachexia are cardinal symptoms in esophageal cancer [66] [67]. To detect nutritional disturbances at an early stage, the European Society for Clinical Nutrition and Metabolism (ESPEN) recommends regular evaluation of nutritional intake, weight change, and body mass index (BMI), at the time of cancer diagnosis and repeated according to the stability of the clinical situation [68]. In patients with digestive cancer, body composition may be quite easily assessed from computed tomography scans [69]. ESPEN recommends nutritional support prior to major surgery in patients with severe nutritional risk (e. g., those with weight loss > 10 – 15 % within 6 months) as a grade A recommendation [70]. If oral feed intake is inadequate despite counseling and oral nutritional supplements, supplemental enteral nutrition or, if the latter is not sufficient or possible, parenteral nutrition is recommended [68] [71]. In patients with severe dysphagia, this can be achieved through nasogastric tube placement, percutaneous feeding tube placement, or parenteral nutrition. Percutaneous endoscopic gastrostomy (PEG) or endoscopic jejunostomy is recommended by ESPEN in place of nasogastric tube placement if enteral feeding is scheduled to last more than 2 – 3 weeks [72] [73]. Furthermore, a Cochrane review of RCTs showed that intervention failure (e. g., feeding interruption, blocking or leakage of the tube, no adherence to treatment) was more frequent with nasogastric tube placement compared with PEG feeding (risk ratio [RR] 0.24, 95 %CI 0.08 – 0.76) [74]. However, in esophageal cancer patients who are scheduled to undergo a gastric tube reconstruction a PEG placement may be contraindicated, in which case a feeding tube is the preferred treatment. The proportion of patients who refuse placement of a feeding tube in the setting of head and neck cancer patients treated with (chemo)radiation has been found to be very low (4 % in an RCT) [75].

#

Esophageal stents and concomitant palliative treatment with radiotherapy

ESGE does not recommend the concurrent use of radiotherapy if an esophageal stent is present (strong recommendation, low quality evidence).

ESGE suggests that SEMS placement with concurrent single-dose brachytherapy is safe and effective for relief of dysphagia (weak recommendation, low quality evidence).

In contrast to the rapid improvement in dysphagia by stent placement, palliative radiotherapy improves dysphagia after 4 to 6 weeks [76]. Temporary and permanent placement of retrievable metallic stents with concurrent radiotherapy has been suggested as an effective method for increasing survival, immediately improving dysphagia and dietary intake in the period before the effects of radiotherapy become apparent [77] [78] [79] [80]. However, a higher risk of life-threatening adverse events has been reported, suggesting that palliative stenting should be delayed until radiotherapy has failed [81] [82] [83].

Potential scattering from the metal material in SEMSs may complicate radiation dosimetry. In a simulated clinical protocol measuring the effects of esophageal stents of various materials and designs on radiation effects on tissue adjacent to the stent in the radiation field, a dose enhancement was seen with SEPSs and stainless steel stents, and not with nitinol stents [84]. In another study, dose perturbation by SEMSs was related to the density of the mesh, with a higher density having greater effect, while SEPS and biodegradable stents had minimal-to-no dose effects outside of the radiopaque markers [85].

In contrast to external radiotherapy, the combination of SEMS and single-dose brachytherapy has been reported to be feasible and safe as a palliative treatment in patients with advanced esophageal cancer [77] [86]. In an RCT that included 53 patients, Guo et al. compared conventional SEMS treatment with SEMS loaded with iodine-125 seeds for brachytherapy; these authors reported a significantly longer dysphagia-free period and longer survival in the irradiation stent group [25] [87].

Data on the use of biodegradable stents in patients receiving brachytherapy are limited, but a high rate of major stent-related complications has been described and a normal diet could not be tolerated because of retrosternal pain and vomiting in more than one third of patients [78].

#

Esophageal stent placement after palliative chemotherapy and radiotherapy

Data are contradictory with respect to the risk of major adverse events in patients receiving a stent for recurrent malignancy following radiotherapy alone or combined with chemotherapy (RTCT). Some studies show an increased risk while other studies, including a meta-analysis, do not report any relationship between SEMS placement after RTCT and the incidence of life-threatening adverse events or survival; only minor adverse events such as chest pain are associated, suggesting stenting is safe in these patients [18] [22] [32] [88] [89] [90] [91] [92] [93] [94]. In detail, the reported rate of life-threatening adverse events ranged from 16 % to 77 % in patients treated with stents after RTCT compared to 0 % to 45 % in patients without previous treatment [18] [22] [32] [88] [89] [90] [91]. Reported stent-related mortality ranged from 0 % to 54 % in patients with prior RTCT compared to 0 % to 6 % in patients without prior RTCT.

It has been suggested that the increased risk, if any, of developing life-threatening adverse events, in patients with prior RTCT may be related to the radiation-induced damage on the esophageal wall, potentiated by chemotherapy. However, it is difficult to discern whether such stent-related adverse events are due to stents and radiation effects, the advanced nature of the disease process, or both. Radiotherapy can cause esophagitis, ulcerations, submucosal fibrosis, and vasculitis, with ischemic damage of the esophageal wall causing esophageal perforations and esophageal – respiratory fistulas via local hypoxemia. Although SEMS placement is effective for short-term palliation of malignant dysphagia, stent pressure on a damaged esophageal wall increases the risk of necrosis [89] [95] [96] [97] [98] [99]. The effect of radiation on the esophageal wall is dose-dependent, with serious damage especially when doses greater than 6 Gy have been administered [97] [99]. The risks of sudden fatal hemorrhage and formation of a respiratory fistula are relatively high in patients with invasive (T4) cancer [47] [96].

#
#

Refractory benign strictures

ESGE recommends against the use of SEMSs as first-line therapy for the management of benign esophageal strictures because of the potential for adverse events, the availability of alternative therapies, and costs (strong recommendation, low quality evidence).

Most studies have used expandable stents for treatment of refractory or recurrent esophageal strictures as defined by Kochman: generally when more than 3 to 5 dilations (either mechanical or pneumatic) have been performed without clinical and endoscopic response or when it was impossible to achieve a 14-mm lumen over 3 dilation sessions [100]. No studies have compared the clinical efficacy of different initial strategies (i. e., dilation vs. stent placement). Therefore, algorithms are mainly based on the experience of tertiary referral centers [101]. Most experts agree that stent placement should be considered when other treatment options (dilation with or without intralesional triamcinolone acetate injections and/or incisional therapy) have failed, though a clear definition of clinical failure has not been uniformly adopted.

ESGE suggests consideration of temporary placement of self-expandable stents for refractory benign esophageal strictures (weak recommendation moderate quality evidence).

ESGE does not recommend a specific type of expandable stent (covered metal, plastic, biodegradable) because none has been shown to be superior to any other for this indication (strong recommendation, moderate quality evidence).

A recent systematic review and meta-analysis (10 prospective and 8 retrospective studies; 444 patients) evaluated the clinical outcome of stent placement for refractory benign esophageal stricture (RBES) [102]. FCSEMS were used in 9 studies (227 patients), 8 trials used SEPS (140 patients) and 4 studies used biodegradable stents (77 patients). Overall, the pooled clinical success rate was 40.5 % (95 %CI 31.5 % – 49.5 %). Patients treated with SEPS and SEMS did not have significantly different success rates compared with patients treated with biodegradable stents. The overall migration rate was 28.6 % (95 %CI 21.9 % – 37.1 %). Stent removal was successful in 99 % of cases. Finally, the overall adverse event rate was 20.6 % (95 %CI, 15.3 % – 28.1 %) with no significant difference between the three types of stents. Only one patient died; this was due to massive bleeding.

Factors predicting successful stent treatment

A systematic review demonstrated that the clinical success of stenting in RBES was significantly lower in patients with cervical strictures and for strictures longer than 2 cm [100]. The latter finding was confirmed by a prospective study showing stricture length as the only factor associated with success, with longer strictures being at higher risk of recurrence (hazard ratio [HR] 1.37, 95 %CI 1.08 – 1.75) [103]. The previously mentioned review and meta-analysis by Fuccio et al. [102] showed that the etiology of the stricture might influence outcome, with esophageal strictures that had developed after surgical resection or radiation therapy being potentially more responsive to stent treatment. However, no firm conclusion can be drawn because many etiologies of stricture were under-represented and, in many studies, the results were not stratified according to the stricture etiology.

ESGE does not recommend permanent stent placement for refractory benign esophageal stricture; stents should usually be removed at a maximum of 3 months (strong recommendation, weak quality evidence).

No studies have compared different strategies in terms of stenting duration. It is generally accepted that FCSEMSs or SEPSs should remain in place for at least 6 – 8 weeks and no more than 12 weeks, to maximize success and to minimize the risk of hyperplastic tissue reaction and stent embedment. Indeed, a large multicenter study that specifically addressed the safety of endoscopic removal of self-expandable stents inserted to treat RBES found no association between indwelling time and the risk of major adverse events [104].

ESGE suggests that FCSEMSs be preferred over PCSEMSs for the treatment of refractory benign esophageal stricture, because of their lack of embedment and ease of removability (weak recommendation, low quality evidence).

The use of partially covered or uncovered SEMS in benign strictures should be avoided because the hyperplastic tissue reaction of the esophageal mucosa to the bare metal mesh often results in recurrent dysphagia. Furthermore, complete embedding of the uncovered metal wires in the esophageal wall may preclude safe stent removal [105] [106].

ESGE recommends the stent-in-stent technique to remove PCSEMSs that are embedded in the esophageal wall (strong recommendation, low quality evidence).

In the case of embedded PCSEMSs, temporary placement of a second, fully covered, stent in the first stent (“stent-in-stent” technique) has been shown to facilitate safe removal of the embedded stent, by induction of pressure necrosis of the overgrowing and ingrowing mucosa [103] [107] [108] [109] [110]. Stents used for the stent-in-stent technique should have a fully covered design and a diameter at least equal to that of the partially covered embedded stent in order to provide sufficient pressure at the site of embedment. In addition, the fully covered stent needs to overlap completely tissue ingrowth inside the lumen of the partially covered stent. The second stent is left in place for 10 – 14 days, before it is retrieved and removal of the embedded PCSEMS is attempted. The success rate of the stent-in-stent technique is above 90 %; in the case of failure, a second FCSEMS should be placed and left in place for 10 – 14 days before a second attempt to remove the stent is performed [103].

Esophageal stent placement in combination with other dilation approaches

ESGE suggests that a combined approach of stent placement with additional techniques (e. g., corticosteroid injection, chemotherapeutic topical application) should not be used in an attempt to improve the long-term benefit of temporary stenting (weak recommendation, very low quality evidence).

Endoscopic incisional therapy has been proposed as either an alternative or additional treatment to endoscopic dilation. Initially proposed for the treatment of recurrent Schatzki rings, it has also been used for the treatment of anastomotic strictures. To our knowledge there have been no studies that have reported using a combined or sequential approach with incisional therapy followed by stent placement.

In order to prevent stricture recurrence, corticosteroid injection into the stricture followed by dilation was proposed more than 10 years ago [111]. Small retrospective studies reporting corticosteroid injection before stent placement do not allow conclusions to be drawn on the additional clinical value for prolonging efficacy following temporary stent placement [112].

Topical application of mitomycin-C has been proposed for refractory corrosive esophageal strictures. Mitomycin-C is a chemotherapeutic agent that inhibits the proliferation of fibroblasts and collagen synthesis and has been proposed to prevent stricture relapse. There are few available studies, mainly case reports and small series, to support its use, and no studies of this treatment combined with stent placement [113] [114].

Treatment options after stent failure for refractory benign esophageal stricture

If refractory benign esophageal stricture has not satisfactorily improved after 2 separate treatments with temporary stenting, ESGE suggests alternative treatment strategies such as self-dilation or surgical treatment (weak recommendation, low quality evidence).
In poor surgical candidates, ESGE recommends self-dilation with rigid dilators (strong recommendation, low quality evidence).

Stent placement for treatment of RBES may be repeated although the majority of studies have demonstrated that additional stent placement does not produce significant incremental benefit [106] [115]. If sustained stricture resolution is not obtained after temporary stenting on two occasions, the suggested treatment options are self-dilation and surgery. Surgery is advised when possible according to anatomical extent as well as patient condition and willingness to undergo such a complex surgical procedure. The best candidates for self-dilation are those who are self-motivated, compliant, and poor surgical candidates [116] [117]. Based on two retrospective studies, esophageal self-dilation was successful in treating 90 % of patients, with significant improvement in global dysphagia scores and overall quality of life [116] [117].

#

Benign esophageal leaks, fistulas, and perforations

ESGE recommends that temporary stent placement can be considered for treatment of leaks, fistulas, and perforations. No specific type of stent can be recommended and the duration of stenting should be individualized. (Strong recommendation, low quality of evidence).

SEMSs have been used for management of perforations and leaks [118] [119]. Closure of an iatrogenic perforation can also be performed by other endoscopic methods [120]. [Table e5] (Appendix e2, available online) shows the results of the published studies on the efficacy and safety of SEMS placement for benign rupture and leakage. In two systematic reviews, the clinical success after placement of temporary stents (FCSEMSs, PCSEMSs, and SEPSs) for benign rupture and anastomotic leaks of the esophagus was similar with different stent types (FSEMS 85 %, PSEMS 86 %, SEPS 84 % [121] [122]. The mean duration of stenting was 7 weeks. Stent migration occurred in 25 %, and it occurred more often with SEPS (26 %) and FCSEMS (26 %).

Data on the use of biodegradable stents are limited. In a small study, 4 of 5 patients with an esophageal leak or anastomotic perforation achieved long-term leak sealing after placement of a covered biodegradable stent [123].

The optimal duration of stenting remains unknown. In most studies stent removal was performed 6 – 8 weeks (range 4 – 10 weeks) after insertion. Stent-associated esophagorespiratory fistula is a serious adverse event that may occur as a consequence of SEMS placement for benign disease. In one retrospective study of 397 patients, 20 patients developed esophagorespiratory fistulas after a median of 5 months following stent placement [124]. Most fistulas occurred at the proximal edge of the stent and in the setting of prior external radiation therapy; thus the cause may have been ischemic pressure necrosis.

#

Acute variceal bleeding

ESGE recommends considering placement of a SEMS for the treatment of esophageal variceal bleeding refractory to medical, endoscopic, and/or radiological therapy, or as initial therapy for patients with massive bleeding (strong recommendation, moderate quality evidence).

[Table e6] (Appendix e2, available online) shows the results of the published studies to date on the applicability, efficacy and safety of covered SEMS for acute esophageal variceal bleeding [125] [126] [127] [128] [129] [130] [131] Most published studies are observational studies [125] [126] [127] [128] [129] [130] [131]. Results from these studies are in agreement with a recently published systematic review and meta-analysis showing that treatment with SEMSs is successful in controlling severe or refractory acute variceal bleeding, without the occurrence of severe adverse events and with a 1-month survival of more than 60 %; these findings confirmed that this therapy can be used as a bridge to transjugular intrahepatic portosystemic shunt (TIPS) or liver transplantation in a significant proportion of patients [132].

An RCT compared patient outcome after SEMS placement (SX-Ella Danis stent; n = 13) versus balloon tamponade (Sengstaken-Blakemore tube; n = 15) in patients with esophageal variceal bleeding refractory to medical and endoscopic treatment [133]. Successful therapy was significantly more frequent in the stent than in the balloon tamponade group (66 % vs. 20 %) with a significantly higher rate for control of bleeding (85 % vs. 47 %), lower transfusion requirements (3 ± 3.4 vs. 6 ± 4.8 packed red blood cell units), and a lower incidence of serious adverse events (15 % vs. 47 %), mainly due to differences in aspiration pneumonia (0 vs. 5) and esophageal tear (1 patient in the balloon tamponade group). No significant difference in 6-week survival was observed (54 % vs. 40 %).

Despite the efficacy of stent placement in controlling acute variceal bleeding, a mortality rate of 25 % has been described in these patients, reflecting the seriousness of the underlying condition of the patient in the case of refractory acute variceal bleeding [129]. In published studies SEMSs have remained in place for up to 2 weeks [125] [131] [134] [135]. When a dedicated SEMS is used retrieval is done using a specifically designed system (PEX-Ella or extractor for SX-Ella Stent Danis).

This Guideline, produced by ESGE and endorsed by the European Society for Radiotherapy and Oncology (ESTRO), the European Society of Digestive Oncology (ESDO), and the European Society for Clinical Nutrition and Metabolism (ESPEN), represents a consensus of best practice based on the available evidence at the time of preparation. The Guideline may not apply in all situations and should be interpreted in the light of specific clinical situations and resource availability. Further controlled clinical studies may be needed to clarify aspects of the statements, and revision may be necessary as new data appear. Clinical consideration may justify a course of action at variance to the recommendations. The Guidelines is intended to be an educational device to provide information that may assist endoscopists in providing care to patients. It is not a set of rules and should not be construed as establishing a legal standard of care or as encouraging, advocating, requiring, or discouraging any particular treatment.

#
#
#

Appendix e1 Esophageal stenting for benign and malignant disease

 Stent placement for benign esophageal disease: task forces and key questions Task force 1: Can stenting be used for benign esophageal strictures? 1 What are the results of stent placement for (refractory) esophageal strictures? 2 What are the results for stent placement compared to other treatment modalities? 3 What are the indications for stent placement compared to other modalities? 4 Which type of stent should be used? 5 What is the optimal indwelling time for stent placement? Task force 2: Can stenting be used for benign esophageal strictures? 1 What are the results of stent placement combined with other treatment modalities for refractory strictures? 2 What are the predictive factors for a successful outcome for stent placement in esophageal strictures? 3 What are the treatment options if stent placement fails in refractory esophageal strictures? Task force 3: Can stents be used to treat benign esophageal leaks and perforations? 1 What are the results of stent placement for esophageal leaks and perforations? 2 Which type of stent should be used? 3 What are the adverse events of stent placement for leaks and perforations? 4 What is the optimal indwelling time of stent placement for leaks and perforations? Task force 4: Can stents be used for variceal bleeding? 1 What are the results of stent placement for variceal bleeding? 2 What are the indications for stent placement for variceal bleeding? 3 What are the adverse events of stent placement for variceal bleeding? 4 Which type of stent should be used? 5 What is the optimal indwelling time of stent placement for variceal bleeding?

#

Table e1(a)

Comparing plastic tubes versus self-expanding metal stents (SEMS).

First author,
year [ref.]

Study design

Participants

Outcome measure

Results

Level of evidence

Dai, 2014 [20]

Cochrane review

7 RCTs

433 patients
– 192 plastic tube vs.
– 241SEMS

Primary:
Dysphagia improvement

Secondary:
Overall survival
Technical success
QoL

Mean difference of dysphagia ≥ 4 weeks post-intervention: – 0.36 (95 %CI – 0.63 to – 0.09; P  = 0.009) in favor of SEMS

Adverse events were significantly more frequent in plastic stent group.
No significant differences in survival, technical success rate, procedure-related mortality, and QoL

High

Conio, 2007 [22]

RCT

101 patients
– 46 Polyflex vs.
– 54 Ultraflex

Dysphagia improvement
Technical success

No significant difference, except for significantly more complications in plastic stent group

Moderate

Table e1(b)

Comparing covered versus uncovered SEMS.

First author,
year [ref.]

Study design

Participants

Outcome measure

Results

Level of evidence

Vakil,
2001 [21]

RCT

62 patients

Primary:
Need for re-intervention for recurrent dysphagia

Secondary:
Dysphagia improvement
Overall survival
Technical success
QoL

Re-intervention rate significantly higher in uncovered stent group

No statistical difference in technical success, dysphagia improvement, overall survival, and adverse events.

Moderate

Saranovic D,
2005 [39]

Retrospective

152 patients

Primary:
Dysphagia improvement

Secondary:
Re-intervention

Dysphagia improved significantly in favor of the covered stent

Stent migration was significantly lower in the uncovered stent group
Recurrent dysphagia was significantly higher in the uncovered stent group

Low

Table e1(c)

Comparing SEMS with other modalities.

Modality
First author,year [ref.]

Study design

Participants

Outcome measure

Results

Level of evidence

SEMS and plastic stent vs. laser

Alderson, 1990 [6]
Carter, 1992 [7]
Fuchs, 1991 [8]
Dallal, 2001 [10]

RCTs

2 RCTs:
SEMS vs. laser (125 patients)

3 RCTs
Plastic stent vs. laser (120 patients)

Primary:
Dysphagia improvement

Secondary:
Overall survival
Technical success
QoL

SEMS/plastic vs. laser: no statistical difference in dysphagia, overall survival, procedure-related mortality, and adverse events

Tumor regrowth, fistula, and perforation occurred in laser therapy, while migration and hemorrhage occurred in SEMS placement

Technical success rate significantly higher for SEMS, with fewer interventions needed

QoL: SEMS vs. laser uncertain and no difference for plastic.

High

SEMS vs. brachytherapy

Homs, 2004 [4]
Bergquist, 2005 [41]

RCTs

274 patients

Primary:
Dysphagia improvement

Secondary:
Overall survival
Technical success
QoL

Dysphagia-free survival significantly higher in brachytherapy group

No statistical difference in overall survival
Significant more adverse events in stent group (late hemorrhage)
QoL: significant difference in favor of brachytherapy.

High

RCT, randomized controlled trial; QoL, quality of life.

Table e2

Early and late adverse events: randomized controlled trials comparing two or more esophageal stents.

First author,
year [ref.]

Total patients, n

Patients, by stent type, n

Technical success, %

CT/RT Pre
Patients, n

CT/RT Post
Patients, n

Pre-dilation
Type
Mean extent, mm

Overall survival, median, days or Follow-up, mean (range), days

Level of evidence

Zhu,
2014 [25]

148

Moderate

73

100 %

28 RTCT

None

None

Severe pain: 23 %
Fistula: 8 %
Pneumonia: 15 %
Hemorrhage: 7 %
Recurrent dysphagia: 28 %

177 days
(153 – 201)

Covered stent

75

100 %

31 RTCT

Severe pain: 20 %
Fistula: 7 %
Pneumonia: 19 %
Hemorrhage: 7 %
Recurrent dysphagia 20 %

147 days
(124 – 170)

Van Heel,
2012 [24]

80

Moderate

Ultraflex (Boston)

40

100 %

30 CT/RT

1 RT

Balloon dilation

NR

Stent dysfunction: 40 %
Migration: 7.5 %
Ingrowth/overgrowth: 20 %
Bolus impaction: 17.5 %
Leakage: 2.5 %
Pneumonia: 7.5 %
Hemorrhage: 17.5 %
Mild pain: 12.5 %

77 days
(1 – 90)

Evolution (Cook)

40

100 %

25 CT/RT

Balloon dilation

NR

Stent dysfunction: 8 %
Migration: 2.5 %
Ingrowth/overgrowth: 2.5 %
Severe pain: 2.5 %
Mild pain: 12.5 %
Reflux: 2.5 %
Foreign-body sensation: 2.5 %

75 days
(1 – 90)

Blomberg,
2010 [26]

65

Moderate

ARS Esophageal- Z Dua stent
(Cook)

28

100 %

NR

NR

Balloon

Up to 15 mm

NR

Migration: 7 %
Obstruction: 18 %
Hemorrhage: 7 %
Severe pain: 4 %
Other: 14 %

63 days
(4 – 393)

CS Esophageal-Z stent
(Cook)
Ultraflex
(Boston)
Wallstent
(Boston)

37

100 %

NR

NR

Balloon

Up to 15 mm

NR

Perforation: 3 %
Migration: 11 %
Obstruction: 11 %
Hemorrhage: 3 %
Severe pain: 14 %
Other: 11 %

70 days
(4 – 511)

Shenfine,
2009 [27]

207

High

Gianturco Z stent
(Cook)

104

98 %

NR

NR

Bougienage

Up to 15 mm

Migration: 10.5 %
Perforation: 0.9 %

Severe pain: 19 %
Overgrowth: 5 %
Bleeding: 19 %
Migration: 3 %
Aspiration: 7 %
Bolus impaction: 4 %

90 days

Rigid stent
(Cook)

57

98 %

NR

NR

Bougienage

Up to 15 mm

Migration 8.7 %
Perforation 1.7 %

Severe pain 14 %
Overgrowth 12 %
Bleeding 21 %
Migration 1 %
Aspiration 3,5 %
Stricture 2 %
Bolus impaction 16 %

119 days

Non-stent therapies

50

Kim,
2009 [28]

37

Moderate

Niti-S
Covered

(Taewoong)

19

100 %

9 CT/RT

2 CT/RT

None

Pneumonia: 5 %

Migration: 5 %
Hemorrhage: 11 %
Overgrowth: 26 %
Reflux: 11 %

62 days

Niti-S
Double-layered

(Taewoong)

18

94 %

7 CT/RT

3 CT/RT

None

Fistula: 6 %
Bolus impaction: 6 %

74 days

Sabharwal,
2008 [29]

48

Moderate

Ultraflex
(Boston)

26

26/26
(100 %)

6 RTCT
13 CT
1 RT

NR

Balloon

Up to 14 mm

Migration: 8 %
Reflux: 8 %
Severe pain: 35 %

Migration: 15 %
Hemorrhage: 8 %
Fistula: 4 %
Obstruction: 27 %

Mean follow-up
101 days

FerX Ella-valve
(Ella CS)

22

22/22
(100 %)

5 RTCT
6 CT
2 RT NR

NR

Balloon

Up to 14 mm

Reflux: 9 %
Severe pain: 9 %

Migration: 32 %
Hemorrhage: 5 %
Reflux: 4 %
Obstruction: 9 %

Mean follow-up
78 days

Verschuur,2008 [30]

125

Polyflex
(Boston)

41

83 %

7 CT
5 RTCT

2 CT

Savary

Up to 12 mm

Perforation: 5 %
Pneumonia: 2 %

Hemorrhage: 12 %
Mild pain: 2 %
Reflux: 5 %
Migration: 29 %
Tissue growth: 10 %
Bolus impaction: 5 %

102 days

High

Ultraflex
(Boston)

42

100 %

8 CT
3 RT
3 RTCT

7 CT

Savary

Up to 12 mm

Severe pain: 2 %
Fever: 2 %

Hemorrhage: 12 %
Fistula: 5 %
Mild pain: 5 %
Reflux: 2 %
Migration: 17 %
Tissue growth: 31 %
Bolus impaction: 24 %

132 days

Niti-S
(Taewoong)

42

95 %

7 CT
1 RT
3 RTCT

15 CT

Savary

Up to 12 mm

Hemorrhage: 2 %
Severe pain: 5 %

Hemorrhage: 2 %
Fistula: 2 %
Mild pain: 5 %
Reflux: 7 %
Migration: 12 %
Tissue growth: 24 %
Bolus impaction: 2 %

159 days

Conio,
2007 [22]

101

High

Polyflex
(Boston)

47

98 %

18 RT/CT

1 RT
4 CT

Savary

Up to 11 – 14 mm

Migration: 2 %
Perforation: 2 %
Hemorrhage: 4 %

Migration: 11 %
Overgrowth: 20 %
Hyperplastic reaction: 11 %
Fistula: 2 %

134 days
(100 – 168)

Ultraflex Partially covered
(Boston)

54

100 %

18
RT/CT

4 RT
5 CT

Savary

Up to 7 – 15 mm

Migration: 2 %
Perforation: 2 %

Migration: 2 %
Overgrowth: 19 %
Hyperplastic reaction: 7 %
Fistula: 4 %
Bolus impaction: 2 %

122 days
(84 – 160)

Power,
2007 [34]

49

Moderate

Hanarostent-valve
(M.I. Tech)

24

100 %

1 RT

NR

NR

Migration: 8 %

Severe pain: 8 %

NR

Ultraflex
(Boston)

25

100 %

1 CT/RT

NR

NR

Bolus impaction: 4 %
Severe pain: 4 %

NR

Wenger,
2006 [35]

41

68 days
(4 – 511)

Esophageal- Z
(Cook)
Ultraflex
(Boston)
Wallstent
(Boston)

22

100 %

NR

NR

Balloon
Up to 15 mm

NR

Migration: 14 %
Perforation: 9 %
Stent occlusion: 14 %
Bleeding: 5 %
Other: 14 %

Moderate

Esophageal- Z
Dua stent Covered

(Cook)

19

100 %

NR

NR

Balloon
Up to 15 mm

NR

Migration: 11 %
Stent occlusion: 5 %
Bleeding: 5 %
Other: 21 %

Homs,
2004 [36]

30

Moderate

FerX Ella-valve
(Ella CS)

15

87 %

5 CT

NR

Savary
Up to 9 mm

Severe pain: 7 %

Hemorrhage: 13 %
Mild pain: 7 %
Reflux: 20 %
Migration: 33 %
Inward folded valve:7 %

107 days
(11- – 03)

FerX Ella
(Ella CS)

13

93 %

3 CT

NR

Savary
Up to 9 mm

Severe pain: 7 %

Hemorrhage: 7 %
Aspiration pneumonia: 7 %
Reflux: 13 %
Migration: 13 %

87
(58 – 116)

Sabharwal,
2003 [29]

53

Moderate

Ultraflex
(Boston)

31

100 %

NR

NR

Balloon
Up to 15 mm

Migration: 6 %
Reflux: 6 %

Ingrowth/overgrowth: 6 %
Hemorrhage: 3 %

Mean follow-up
96.5 days

Flamingo
(Boston)

22

100 %

NR

NR

Migration: 5 %
Perforation: 5 %
Reflux: 5 %

Ingrowth/overgrowth: 9 %
Hemorrhage: 5 %

Mean follow-up
97.1 days

O’Donnell,
2002 [15]

48

Moderate

Plastic
(Wilson-Cook)

23

96 %

2 RT

1 RT

Balloon

Up to 14 – 19 mm

Reflux: 14 %
Pain: 63 %

Migration: 14 %
Ingrowth/overgrowth: 27 %
Bolus impaction: 23 %
Other: 4 %

62 days
(0 – 431)

Flamingo
Standard
Telestep
Partially covered

(Total 24 patients)

Ultraflex
(1 patient)
(Boston)

25

100 %

4 RT

5 CT

Balloon

Up to 12 mm

Migration: 4 %
Hemorrhage: 4 %
Perforation: 4 %
Reflux: 16 %
Pain: 44 %

Migration: 8 %
Ingrowth/overgrowth: 36 %
Fistula: 4 %
Other: 4 %

107 days
(4 – 462)

Laash,
2002 [38]

50

Moderate

Flamingo
(Boston)

25

100 %

NR

NR

Persistent pain: 4 %
Pneumonia-death: 4 %

Reflux: 96 %
Bolus impaction: 4 %
Perforation: 4 %
Migration: 12 %

102 days
(9 – 375)

Esophageal-Z
Dua stent

(Cook)

25

100 %

NR

NR

Balloon

15 mm

Reflux: 12 %
Bolus impaction: 4 %
Perforation: 4 %
Migration: 16 %

103 days
(8 – 272)

Siersema,
2001 [31]

100

High

Ultraflex
(Boston)

34

97 %

1 RT

5 CT

5 CT

Perforation: 6 %
Fever. 3 %

Migration: 18 %
Overgrowth: 3 %
Bolus impaction: 6 %
Bleeding: 15 %
Reflux: 3 %
Pain: 9 %

104 days

Flamingo
(Boston)

33

100 %

1 RT

4 CT

2 CT

Perforation: 6 %
Pressure necrosis: 3 %

Migration: 9 %
Overgrowth: 15 %
Bolus impaction: 9 %
Bleeding: 9 %
Reflux: 9 %
Pain: 18 %

113 days

Gianturco-Z
(Cook)

33

100 %

2 RT

4 CT

1 RT

3 CT

Perforation: 6 %
Bleeding: 6 %
Fever: 3 %
Severe pain: 3 %

Migration: 12 %
Overgrowth: 12 %
Bleeding: 18 %
Reflux: 6 %

110 days

Vakil,
2001 [21]

62

Moderate

Covered metal

32

97 %

3 RT

4 CT

4

Yes

NR

Improper positioning: 6 %
Bleeding: 26 %

Migration: 13 %
Ingrowth: 3 %
Reflux: 39 %

NR

Uncovered metal

30

100 %

4 RT

4 CT

4

Yes

NR

Perforation: 6 %
Bleeding: 6 %

Migration: 13 %
Overgrowth: 13 %
Bleeding: 19 %
Reflux: 6 %

NR

ARS, antireflux stent; CI, confidence interval; CS, conventional stent; CT/RT, chemotherapy or radiotherapy; RTCT, radiochemotherapy; NR, not reported; pre-dilation, dilation before stent placement;

Table e3

Stent placement for malignant esophageal fistula: included studies.

First author,
year [ref.]

Study design

Description

Participants

Outcome

Results

Level of evidence

Shin,
2004 [42]

Retrospective study

Stent for esophagorespiratory fistula

61 patients

Technical success

Fistula sealed: 49/61 (80 %)
Fistula re-opened during follow-up: 17/49 (35 %)

Mean survival significantly longer in patients with initial success than in patients with initial clinical failure (15.2 vs. 2 weeks, P ≤ 0.05)

Low

Balazs,
2008 [43]

Retrospective study

Stent for esophagorespiratory fistula

264 patients with esophageal and lung cancer and esophagorespiratory fistula

Incidence, causes, and characteristics of fistula formation and treatment options

Successful esophageal intubation, stent correction, or replacement: 188 patients (71 %)

Procedure-related mortality: 0.5 %

Mean survival in stent group higher than in gastrostomy, PEG, jejunostomy, and best supportive care: P ≤ 0.001

High

Hurtgen,
2014 [44]

Expert review on tracheo-esophageal and enterorespiratory fistula

Stent and surgery for esophagorespiratory fistula

–

Single or double stenting of the esophagus and trachea with SEMS is the established palliative treatment. Surgical interventions are justified only in very select cases and carry a high morbidity and mortality risk.

Low

Sarper,
2003 [49]

Prospective observational study

SEMS for malignant strictures and fistula (esophagotracheal and esophagorespiratory fistula).

41 patients of whom 14 with fistula

Fistula sealed: 12/14 (86 %)

Survival time of patients with fistulas: median 49 days (range 5 – 186 days)

Low

Chen,
2012 [50]

Retrospective observational study

SEMS, feeding gastrostomy or jejunostomy for tracheo-esophageal fistula

86 patients with tracheo-esophageal fistula
– 30 patients with stents vs. 35 patients with feeding stoma

Overall survival, serum albumin, mean body weight loss

Significant improvement in overall survival with SEMS placement: P = 0.03; OR 1.9.

No significant difference in serum albumin and mean body weight loss.

Low

Rodriguez,
2010 [51]

Review

Publications on malignant tracheo-esophageal and bronchoesophageal fistula

Causes of these fistulas are described.

Best palliative treatment is endoscopic placement of an esophageal stent, respiratory stent, or parallel stenting.

Dual stenting appeared to be better than single for palliation and safety.

Attention to tracheal compression/ erosion secondary to esophageal stenting.

Low-moderate

Dumonceau, 1999 [52]

Prospective study

Stainless steel and nitinol SEMS

17 patients, of whom 14 with malignant fistula
– 5 stainless steel vs.
– 12 nitinol

Efficacy and costs.

Fistula sealed:
– 1/5 (20 %) stainless steel covered stents vs.
– 12/12 (100 %) nitinol covered stents

Adverse events: 2/17: both were fatal

Costs $106 (stainless steel) vs.$57 nitinol

Low

van Heel, 2010 [2]

Prospective observational study

SEMS for palliation of malignant dysphagia and fistula recurrence after esophagectomy

81 patients:
– 66 malignant dysphagia
– 15 fistula

Safety and efficacy.

Fistula sealed: 14/15 (93 %)

Disease progression resulted in death after median 70 days

Low

Sharma, 2010 [55]

Guideline,
9 prospective case series

SEMS for malignant esophageal fistula

Efficacy, sealing, complication rate, overall survival

Sealing rate: 70 % – 100 %

Complication rate: 10 % – 30 %

Significantly longer survival in patients with successful fistula closure

Covered SEMS placement treatment of choice for malignant esophageal fistula

Moderate

Hu, 2009 [59]

Retrospective, comparative study

SEMS, gastrostomy for esophageal fistula

35 patients
– 17 SEMS
– 9 gastrostomy
– 9 controls

Survival, QoL

Survival in stent group longer: 93 days vs. 66 days (controls) and 62 days (gastrostomy group); P = 0.48

QoL/decrease in symptoms significantly better with SEMS placement, P ≤ 0.05

Low

OR, odds ratio; PEG, percutaneous endoscopic gastrostomy; SEMS, self-expandable metal stent; QoL, quality of life.

Table e4 a

First author,
year [ref.]

Study design

Participants, n

Tumor histology, n

Type of stent

Technical success, %

Stent-related complications, %

Siddiqui
2009 [92]

Retrospective study

12

SCC 2

Polyflex

RTCT

92 %

Migration: 36 %
Chest pain: 73 %

Bower
2009 [93]

Prospective study

25

SCC 5

Polyflex

RTCT

100 %

Migration: 25 %
Chest pain: 4 %

van der Berg
2014 [65]

Prospective study

10

SCC1

ELLA-SX

RTCT

100 %

Chest pain: 60 %
Stent obstruction: 10 %

RTCT, radiochemotherapy; SCC, squamous cell carcinoma.

Table e4 b

First author,
year [ref.]

Study design

Patients, n

Type of stent

Major complications, n/n (%)

Minor complications, %

Stent-related death, n/n (%)

Median survival

Level of evidence

Muto,
2001 [95]

Retrospective

Low

Prior RTCT

13

Z-stent
Wall stent
Ultraflex

10/13
(77 %)

7/13
(54 %)

69 days

Sumiyoshi,
2003 [96]

Retrospective

Low

Prior RTCT

22

Ultraflex
Wallflex

9/22
(41 %)

6/22
(27 %)

67 days

Homs,
2004 [88]

Prospective

Z-stent
Wallstent
Ultraflex

99 days

High

Prior RTCT

49

14/49 (29 %)

41 %

1/49 (2 %)

No prior RTCT

151

31/151 (21 %)

15 %

2/151 (1 %)

Iraha,
2006 [89]

Retrospective

SEMS

0 %

Low

Prior RTCT

12

3/12 (25 %)

No prior RTCT

8

0 %

Lecleire,
2006 [97]

Retrospective

Ultraflex
Choostent

Moderate

Prior RTCT

56

24/56 (43 %)

48 %

11/56 (20 %)

72 days

No prior RTCT

60

5/60 (8 %)

30 %

3/60 (5 %)

81 days

Conio
2007 [22]

RCT

Polyflex
Ultraflex

1/100 (1 %)

133 days

High

Prior RTCT

36

14/36 (39 %)

No prior RTCT

64

26/64 (41 % )

Park, 2012 [98]

Retrospective

208 patients including 134 radiotherapy patients. Of these:
– 23 with airway complication including 21 radiotherapy patients

Covered stents

21/134 (16 %) airway complications

Low

8

2 months

13

7 months

2

Didden,
2012 [47]

Retrospective

Prior RTCT

13

Ultraflex
Evolution Niti-S
Alimaxx-E
SX-Ella

8/13
(61 %)

8 %

1/13 (8%)

135 days

Low

Rueth
2012 [90]

Retrospective

Alimaxx
Ultraflex
Wallflex
Polyflex

2 (4.4 %)

Moderate

25

8/25 (47 %)

37 days

20

9/20
(45 %)

38 days

Qiu,
2012 [99]

Retrospective

MTN stent
CZES-II stent
Ultraflex

Moderate

Prior RTCT

57

32/57 (56 %)

16 (28 %)

77

No prior RTCT

35

4/35 11 %

2 (5.7 %)

246

RTCT, radiochemotherapy; RCT, randomized controlled trial

Table e5

Assessment of success and complication rate after placement of self-expandable metal stents (SEMSs) for benign rupture or leakage: included studies.

First author,
year [ref.]

Study design

Technical success, %

Stent type

Sealing rate, %

Migration, %

Tissue growth

Time stent in place

Re-intervention

Holm,
2008 [142]

Retrospective

98.8 %

SEPS

17 %

62.1 %

17.2 %

53 days

NR

van Heel,
2010 [118]

Prospective

100 %

FCSEMS
SEPS
PCSEMS

97 %

33 %

NR

6 – 84 weeks

NR

Cerna 2011 [123]

Prospective

100 %

80 %

60 %

NR

5 – 10 days

NR

D’Cunha,
2011 [143]

Retrospective

100 %

FCSEMS
SEPS

88 %

16.2 %

NR

33 days

45.2 %

van Boeckel,
2011 [107]

Retrospective

99 %

FCSEMS
SEPS
PCSEMS

65 %

17.3 %

3.8 %

39 days

NR

Dasari,
2014 [122]

Review (27 case series)

91 %

FCSEMS
SEPS
PCSEMS

81 %

20.8 %

NR

4 – 8 weeks

17 %

FCSEMS, fully covered self-expandable metal stent; NR, not reported; PCSEMS, partially covered self-expandable metal stent.

Table e6

Assessment of use of esophageal stents in the treatment of acute esophageal variceal bleeding: included studies.

First author,
year [ref.]

Design

n

Successful stent insertion

Efficacy

Stent migration

Mortality

Hubmann,
2006 [130]

Observational

20

100 %

100 %

5 %

25 %

10 % (in-hospital)

Zehetner,
2008 [131]

Observational

39[*]

100 %

97 %

3 %

18 %

26 % (30-day)

Wright,
2010 [125]

Observational

10

90 %

70 %

0

10 %

50 % (6-week)

Dechêne,
2012 [126]

Observational

8

100 %

88 %

12 % (SAE)

0

75 % (60-day)

Holster,
2013 [127]

Observational

5

100 %

80 %

40 % (no SAE)

20 %

40 % (in-hospital)

Fierz,
2013 [128]

Observational

9

78 %

89 %

0

22 %

78 % (6-week)

Zakaria,
2013 [129]

Observational

16

94 %

87.5 %

6 % (SAE)

37 %

25 % (in-hospital)

* Including the 20 previously reported cases

#

Table e7

Early and late adverse events associated with esophageal stent placement in malignant disease.

Mean (range), %

Proportion of studies reporting

Mean (range), %

Proportion of studies reporting

Severe pain

8.7 % (2 % – 35 %)

(7/13)

15 % (2 % – 23 %)

(6/17)

Hemorrhage/bleeding

7.6 % (2 % – 26 %)

(5/13)

11.3 % (2 % – 21 %)

(12/17)

Migration

6.6 % (2 % – 10 %)

(6/13)

11 %(1 % – 33 %)

(13/17)

Perforation

3.3 % (0.9 % – 6 %)

(7/13)

4.5 % (3 % – 9 %)

(3/17)

Ingrowth/overgrowth

–

–

14 % (2.5 % – 36 %)

(9/17)

Obstruction (bolus impaction)

–

–

9 % (2 % – 27 %)

(11/17)

Reflux

9.3 % (5 % – 16 %)

(3/13)

15 % (2 % – 96 %)

(9/17)

Bronchoaspiration/pneumonia

3.5 % (2 % – 5 %)

(3/13)

10.3 % (7 % – 19 %)

(5/17)

Fistula

–

–

5 % (2 % – 8 %)

(6/17)

Others
(e. g. fever, incorrect position, pressure necrosis, foreign-body sensation, stricture)

2.8 % (2 % – 6 %)

(3/13)

10 % (2 % – 21 %)

(5/17)

Most common adverse events of esophageal stenting

Severe pain

Selecting the appropriate size of stent may reduce this complication. Also use of stents that conform better to the esophageal anatomy, with good radial but low axial force, may diminish excessive pressure and thereby reduce pain. If the distance between the upper esophageal sphincter and the stricture is less than 2 cm, placement of large-diameter SEMSs is associated with more pain and discomfort including foreign-body sensation.

Bleeding

Early bleeding is caused by passage of the endoscope or stent catheter across the stricture and it is usually mild and self-limited. Delayed bleeding due to tumor progression or the formation of an aortoesophageal fistula may be fatal. Patients with aortic tumor involvement at the time of stent placement are particularly at risk and should be informed of this potentially fatal complication [47] [96]. Choice of an appropriate stent diameter minimizes the risk of bleeding due to ulcer formation caused by pressure necrosis at the levels of both ends of the device. Adequate prophylaxis with proton pump inhibitor is advisable to prevent bleeding due to severe esophagitis in patients with stent placement across the gastroesophageal junction.

Recurrent dysphagia (migration, ingrowth, overgrowth, and food obstruction)

Use of large-bore stents decreases the risk of stent migration and of food impaction; this can be recommended in mild strictures and in those involving the gastroesophageal junction (GEJ) [22] [31] [136]. Some stent designs may also reduce the migration risk (e. g. stents with increased resistance on the outside of the stent, stents with distinct shouldering of their upper end, and stents with a flip-flop collar) [28] [136]. Placement of a stent that is partially uncovered (both ends) can be suggested for patients in whom a covered stent has migrated. Techniques of stent fixation with clips have also been described, but without a clear advantage of this approach for the prevention of stent migration [137] [138] [139] [140] [141].

Placement of a FCSEMS or SEPS prevents ingrowth. Neoplastic overgrowth can be reduced by using a stent longer (at least 2 cm) than the stricture [136]. However, granulomatous overgrowth due to benign tissue hyperplasia may occur and has been described with all types of stent.

Fistula

Fistula may be caused by the progression of cancer, previous radiotherapy, and by erosion at the edge(s) of the stent, which are the widest parts of the device, into the esophageal wall. The use of an FCSEMS in the mid-esophagus for malignant strictures may prevent the development of late esophagorespiratory fistula [32] [88] [124]. In a large retrospective study no statistically significant relationship was detected between stent diameter or presence of a flange and the development of esophagorespiratory fistula [124]. Prior radiotherapy is the most important risk factor for the development of esophagorespiratory fistula after SEMS placement [97] [98] [124].

Perforation

Dilation before stent placement is associated with a significantly increased risk of major adverse events, in particular perforation [88]. Excessive manipulation of the guidewire, stricture dilation, and passage of the endoscope across the stricture should all be avoided to minimize the risk of perforation [22] [31].

#
#
#

Competing interests: T. H. Baron and M. J. Bruno have ongoing lecture/consultancy roles for Cook Medical and Boston Scientific. J-C. Garcia-Pagán has received grants from W. L Gore (2016 – 2020), Exalenz (2016 – 2017) and Novartis (2016 – 2017). P. D. Siersema received research support from Cook Medical, Ireland (2014 – 2016). M. C. W. Spaander supported a Boston Scientific esophageal stent trial (2012 – 2015). M. Conio, A. de Ceglie, J.-M. Dumonceau, A. Escorsell, L. Fuccio, C. Hassan, M. Nordsmark, A. Repici, B. Schumacher, T. Seufferlein, J. Skowronek, A. Van Gossum have no competing interests.

Corresponding author

M.C.W. Spaander, MD PhD
Department of Gastroenterology and Hepatology
Room Hs-308
Erasmus Medical Centre Rotterdam
PO Box 2040
3000 CA Rotterdam
3015 CE Rotterdam
The Netherlands
Fax: +31-10-7035172