Expandable Stents for Benign Esophageal Disease

Partially covered self-expandable esophageal stents have been associated with unacceptable complications when used for benign esophageal disorders. With the introduction of removable or potentially removable fully covered stents and biodegradable stents, interest in using expandable stents for benign indications has been revived. Although expandable stents can offer a minimally invasive alternative to surgery, they can be associated with serious complications; hence, this approach should be considered in carefully selected patients, preferably on a protocol basis.

The efficacy of self-expanding metal stents (SEMSs) in the palliation of malignant esophageal diseases has been extensively studied and published. With a greater than 95% technical success rate in placement and almost immediate relief in symptoms, SEMSs have become the most widely accepted modality used to palliate malignant dysphagia and/or malignant esophageal fistulae. Compared with the previous semirigid plastic tubes, which were associated with higher complication rates, SEMSs have a smaller diameter delivery catheter; thus, preinsertion dilatation of the esophageal stricture is not required in most patients. Despite this smaller diameter, once released, these stents can expand to large diameters; the expansion is gradual rather than abrupt and the stent can conform to the shape of the stricture. Where required, it is easy to deploy another stent into a previously placed stent. To prevent tumor in-growth, majority of the SEMS are partially covered with a plastic membrane. Since plastic covering makes SEMS susceptible to migration, the uncovered segments at the upper and lower ends of the stent allow for tissue in-growth and embedment ( Fig. 1 ). Although this makes the stent nonremovable, removability is a not an issue when stents are used in patients with limited life expectancy. Experience in using SEMSs in those with limited life expectancy has also not given opportunity to evaluate their efficacy when used on a long-term basis. A study from the MD Anderson Cancer Center in Houston, Texas, showed that if SEMSs are kept in place for several weeks, even for malignant indications, patients can start experiencing significant and life-threatening complication (37%); a suggestion has been made that palliation of malignant dysphagia may be better accomplished by a combination of SEMS (immediate symptom relief) with brachytherapy and removal of SEMS after 4 to 6 weeks to avoid complications.

With uncertainties regarding removability and complications associated with long-term use, SEMSs have not received widespread acceptance for benign esophageal disorders. Initial experience has been discouraging primarily because of granulation tissue in-growth into the uncovered part of SEMSs (see Fig. 1 ) as well as tissue embedment, making them difficult, if not impossible, to remove. Newer SEMSs that are fully covered, hence do not embed, are now available. These stents seem removable but are still not approved for benign indications. Recently, a fully covered self-expanding plastic esophageal stent (SEPS) (Polyflex, Boston Scientific, Natick, MA, USA) made of woven plastic strands (potentially inducing less tissue reaction compared with metal) was developed. This stent can be removed and is Food and Drug Administration (FDA) approved for treatment of benign refractory esophageal strictures. Another attractive concept has been the introduction of biodegradable stents, where the issue of removability does not exist because these stents eventually undergo metabolic degradation and absorption. With the availability of SEPSs, fully covered SEMSs (FC-SEMSs), and biodegradable stents, interest in using expandable stents for benign esophageal disorders has been revived. This review focuses on experience in using self-expanding stents for benign esophageal diseases with special emphasis on refractory benign esophageal strictures (RBESs) and benign esophageal perforations, leaks, and fistulae.

Refractory benign esophageal strictures

Before embarking on endotherapy for esophageal stricture, it is important to establish that the stricture is benign by multiple biopsies and by imaging studies where needed. Similarly, it is important to address the primary cause of the stricture. For example, a peptic stricture is labeled as refractory if a patient is not receiving or is not compliant with strict antireflux measures. In addition to cause, the stricture anatomic characteristics may have influence on the outcomes of endotherapy. Benign esophageal strictures can either be classified as simple (short [<2 cm], straight, and wide enough to allow a standard 9.5-mm diameter endoscope to pass) or complex (long [>2 m], tortuous, multiple sites, and too narrow to allow passage of a standard endoscope). Besides esophageal rings and webs, simple strictures are generally peptic in origin whereas complex strictures can develop after corrosive injuries, radiation therapy, surgery, and esophageal ablative treatments, such as photodynamic therapy and mucosal resections.

Endoscopic dilatation using a bougie or balloon dilators is the most widely accepted method for treating benign esophageal strictures. Up to 40% of benign strictures may recur, requiring periodic dilatations. Simple strictures tend to respond better to dilatations and eventually the intervals between periodic dilatations become longer. Complex strictures, alternatively, are difficult to treat, carry a higher procedural complication rate, and tend to recur within weeks. These strictures are considered RBESs. Various investigators have defined RBESs in different ways and this may have had some bearing on the confounding results of endotherapy published in the literature. To standardize the characteristics of RBESs, Kochman and colleagues proposed the following definition: an anatomic fibrotic esophageal restriction, absence of inflammation or motility disorder, and inability to achieve a diameter of greater than or equal to 14 mm in 5 sessions of dilatations at 2-week intervals or inability to maintain a diameter of greater than or equal to 14 mm for 4 weeks once greater than or equal 14-mm diameter is achieved. This definition does not address the influence of the cause of the strictures on the outcomes of therapy because RBES from corrosive injury may behave differently from RBES secondary to radiation therapy.

Besides repeated high-risk dilatations, several endoscopic approaches have been tried for treating RBES. Some of these include intralesional steroid injection, electrocautery incision, argon-plasma coagulation, expandable stents, and a combination of these modalities. Surgeries with associated morbidity and mortality or gastrostomy tube feeding are alternatives for those not responding to endoscopic interventions.

The Concept of Using Expandable Stents for RBESs

If during endoscopic dilatation, a few seconds of stretching with a bougie or a balloon can relieve dysphagia for a few weeks, then, conceptually speaking, stretching the stricture continuously with a dilator in place for several weeks may give longer-lasting benefit by allowing tissue to remold around the dilator. Expandable esophageal stents are ideal in achieving this because they not only function as dilators but also maintain luminal patency while stretching the stricture continuously for weeks ( Fig. 2 ). Unlike for malignant strictures, one of the major issues for using stents for benign indications is the need to subsequently remove the stent. The types of self-expanding esophageal stents that have been used for RBESs are metal stents (partially covered and fully covered), plastic stents, and biodegradable stents. Unfortunately, most of the studies evaluating these stents for RBESs have been small, retrospective case series and there have been no major prospective, randomized studies comparing various alternative endoscopic techniques with stenting or comparing one type of stent with another type. Similarly, no major conclusions can be made on the influence of the cause of RBESs on the outcomes from stenting because most of the case series are small and included patients with RBESs from a variety of causes.

Partially Covered Self-Expanding Esophageal Metal Stent

Partially covered SEMSs (PC-SEMSs) have been primarily designed for malignant esophageal strictures. Although they can be uncovered, the majority of SEMSs used are those coated with plastic to prevent tumor in-growth. To reduce the risks of migration, short segments at the upper and lower ends of the stent are left uncovered for tissue in-growth and tissue embedment to allow for better anchoring (see Fig. 1 ). Tissue embedding of SEMSs was demonstrated by Bethge and colleagues and, although this feature reduces the risks of migration, it makes removal of PC-SEMSs difficult and a high-risk procedure. This may not be an issue when PC-SEMSs are placed permanently for palliation of malignant dysphagia. For RBESs, if stents are not removed, long-term stent-related complications, including ulceration, bleeding, fistula, and recurrent dysphagia due to migration, granulation tissue, and new stricture formation may occur. Stents may erode into mediastinal structures, including the aorta, with fatal outcomes. PC-SEMSs are not FDA approved for treatment of benign strictures. Nevertheless, anecdotally, PC-SEMSs have been placed in patients with RBESs and there are several small retrospective series in the literature where these stents were associated with poor results. In one series, 4 of 8 patients developed major complications and 1 patient died from bleeding secondary to the stent eroding into the aorta. Similar discouraging results have been reported in several other small series. In a review of 29 patients where SEMSs were placed for RBESs, Sandha and colleagues reported an overall complication rate of 80%, which included new stricture formation (40%), migration (31%), and trachea-esophageal fistula (6%).

Due to unacceptable complications associated with tissue in-growth and embedding of the stent, PC-SEMSs are not recommended for RBESs.

Fully Covered Self-Expanding Esophageal Metal Stent

Because granulation tissue in-growth into the uncovered segments of PC-SEMSs results in embedment of the stent, one way to overcome this problem is to completely cover the SEMS. Some of the FC-SEMSs available in the United States are listed in Table 1 . Most of these stents are dog-bone in shape and have a purse-string suture attached at the upper end for repositioning after deployment. This suture or the upper end of the stent can be grasped with a forceps and the stent can be withdrawn using a standard endoscope. Anecdotal and personal experience has shown that these stents can be removed (see Fig. 2 ); as of this review’s writing, these stents are not FDA approved for removability. Compared with plastic mesh, the wire mesh design of these stents can be constrained to a thinner diameter; hence, the delivery system of these stents are thinner than a self-expanding plastic stent (SEPS) (Polyflex—the only stent approved by FDA for use in benign strictures). Moreover, unlike the SEPS, these stents are user-friendly because they come preloaded and can be reconstrained during deployment for proper positioning.

Table 1

Some of the fully covered self-expanding esophageal stents available in the United States

Self-Expanding Plastic Esophageal Stent (Polyflex Stent) ^a
Diameter (mm)	Length (mm)	Delivery Catheter (mm)	Shortening	Recapture	Fully Covered	Company
16	90, 120, 150	12	Yes	No	Yes	Boston Scientific
18	90, 120, 150	13	Yes	No	Yes
21	90, 120, 150	14	Yes	No	Yes
Fully Covered Self-Expanding Metal Esophageal Stent ^b
18	103, 123, 153	6.2	Yes	Yes	Yes	Boston Scientific
23	105, 125, 155	6.3	Yes	Yes	Yes	Boston Scientific
Wallflex Stent
Alimaxx-ES Stent ^c
18	70, 100, 120	7.4	No	No	Yes	Merit Medical
22	70, 100, 120	7.4	No	No	Yes	Merit Medical
Evolution Stent
18	80, 100, 120	8	Yes	Yes	Yes	Cook Endoscopy
20	80, 100, 120	8	Yes	Yes	Yes	Cook Endoscopy
Niti-S Stent ^d
16	60, 80, 100, 120, 150	5.8	Yes	Yes	Yes	Taewoong Medical
18	60, 80, 100, 120, 150	6.5	Yes	Yes	Yes
20	60, 80, 100, 120, 150	6.5	Yes	Yes	Yes

a FDA cleared for use in benign esophageal strictures.

b Not yet FDA cleared for use in benign esophageal disorders.

c Recently, a 12-mm diameter Alimaxx-ES stent was also released by Merit Medicals (see Fig. 2 ).

d A 3.5-mm diameter delivery system, through-the scope, fully covered Niti-S esophageal stent is also available.

In a study on 8 pigs, Baron and colleagues showed that a PC-SEMS (Ultraflex, Boston Scientific) induced severe granulation tissue response embedding the uncovered segments of the stent, making its removal traumatic compared with an FC-SEMS (Alimaxx-ES, Merit Medical, South Jordan, UT, USA) that caused minimal tissue response and could be removed easily without significant trauma. There was, however, a higher migration rate with the FC-SEMSs. Song and colleagues published their study on a new fully covered nitinol SEMS that they placed in 5 patients with benign esophageal strictures. After 2 months, they were successful in removing the stent in all 5 patients with resolution of stricture. The stricture recurred, however, in 2 patients. The same group evaluated the safety and efficacy of this stent in an additional 25 patients with RBESs. The stent was successfully removed in 23 patients at 1 to 8 weeks after insertion. In 2 patients the stent had migrated (passed rectally in one and regurgitated in another). After removal, one patient developed a fistula that closed spontaneously. At mean 13 months’ follow up (range 2–25 months), 12 patients remained asymptomatic and in the remaining 13 patients, the strictures recurred requiring dilatations. In a small series of patients with RBESs, Lakhtakia and colleagues were successful in removing the Alimaxx-ES stent after 6 to 12 weeks in place. In another series, 36 Alimaxx-ES stents were placed in 31 patients, of which 7 were for RBESs. Stent removal was successful in all cases where attempted and the success rate in treating RBESs was 29%. In a recent retrospective review, Bakken and colleagues described their experience using an FC-SEMS (Alimaxx-ES) for benign esophageal diseases. There were 25 patients with strictures who underwent a total of 30 procedures. The stent were removed after an average of 67 days (range 0–279 days). Initial success rate in treating strictures was 56%. Stent migration was a major issue that was influenced by the cause of the stricture (radiation 25%, nonradiation 50%, and surgical anastomotic 60%) and the site of stricture (proximal 46%, middle 21%, and distal 38%). Other complications included nausea and emesis requiring hospital visits, neck and chest pain, gastroesophageal reflux, and dysphagia. Life-threatening major complications included stridor during 3 procedures in 2 patients requiring stent removal, respiratory distress in 1 patient (stent removed), and bradycardia/asystole in 1 patient (ICU admission). Another recent smaller retrospective study using Alimaxx-ES stent showed good success rates in treating RBES and fistula with migration rate of 39%.

Based on small case series and retrospective reviews, FC-SEMSs induce minimal tissue reaction and are potentially removable. Initial experience on using these stents for RBESs has been encouraging. Influence of the cause on the outcome and the duration for which the stent should be kept in place before removal has not yet been established. Migration has been a major issue and further design modifications are needed. FC-SEMSs are still not FDA cleared for benign indications.

Self-Expanding Esophageal Plastic Stents

The Polyflex stent is an SEPS that can be removed and is FDA approved for treating benign esophageal strictures. The stent is made of polyester mesh with an inner coating of a smooth silicone membrane along the entire length of the stent. Because of the full coating and presumably less tissue reaction to plastic compared with metal, this stent induces minimal granulation tissue formation facilitating easy removal. The upper and lower ends are smoothed by the silicone membrane to decrease tissue reaction above and below the stent. The proximal end is flared out to reduce the chances of migration. Since the stent is made of plastic, to enhance radiological visualization, the plastic is impregnated with barium. Also, there are radiopaque markers located at the upper and lower ends of the stents and an additional marker at the mid-point of the stent. The markers are colored blue for endoscopic positioning. Like SEMSs, SEPSs are available in various diameters and lengths (see Table 1 ). Unlike SEMSs, SEPSs are not preloaded. Using several simple steps, SEPSs are loaded into the delivery catheter and, although this feature may not be as user friendly as SEMSs, it has the advantage of removing, reloading, and reusing the same stent if deployed incorrectly. Unfortunately, the diameter of the delivery system is large (12 mm, 13 mm, and 14 mm for 16-mm, 18-mm, and 21-mm diameter stents, respectively), hence preinsertion dilatation is required in almost all patients. A newer SEPS is being developed to overcome some of these limitations. Unlike SEMSs, there is no suture at the upper end and the stent is removed by grabbing a spot on the upper rim of the stent and slowly withdrawing it.

Repici and colleagues treated 15 patients with esophageal strictures who had failed prior repeated dilatations. SEPSs could be placed in all patients and removed from all patients after 6 weeks. At a mean follow-up of 22.7 months, 12 patients (80%) remained symptom-free. There was one stent migration. In a slightly larger retrospective review, Evrard and colleagues found SEPSs curative in 17 of 21 patients with RBES at mean 21 (range 8–39) months’ follow-up. The stent could be removed in all patients and there was one major complication of tracheal compression requiring stent removal.

Subsequent experience from other centers using SEPSs for RBESs could not reproduce the results (discussed previously). On the contrary, studies showed poor cure rates with significant associated complication rates, which included migration, hemorrhage, chest pain, ulceration, and esophageal perforation. In a small series of 5 patients treated with SEPSs, all developed complications in the form of migration, chest pain, and one perforation. In another retrospective review, Holm reported a low success rate of only 6%. Complications included chest pain, nausea/emesis, and dysphagia. Stent migration was noted in up to 82% and was more common with proximal and distal stent placements compared with the midesophagus.

The exact reason for the excellent results from Europe and poor results from other centers is not clear. These variable results could be due to the studies enrolling patients with different definitions of RBES, different case mixes, variable periods for which the stents were kept in place and the retrospective nature of the design. To address some of these issues, a prospective (nonrandomized) study was conducted in which 40 patients were enrolled. All these patients fulfilled the definition of RBES. A SEPS (Polyflex) was deployed across the stricture and then removed after 4 to 6 weeks. Technical success rates of stent placement and removal were 93% and 94%, respectively. At a median of 53 weeks (range 11–153) follow-up, 30% (intention to treat) of patients were dysphagia-free. Another 30% of patients with recurrence of dysphagia after stent removal opted for insertion of a new Polyflex stent for a longer duration rather than returning to their baseline alternatives of repeated frequent dilations, gastrostomy tube feeding, or surgery. This approach was able to change the outcomes from baseline alternatives in 66% of patients, which also included those who did not want the stent to be removed. Unfortunately, complications in the form of migration (22%), severe chest pain (11%), bleeding (8%), and perforation (5.5%) were observed. One patient who refused stent removal died of severe bleeding probably related to an aortoesophageal fistula. A recent study compared Polyflex stents with repeated dilations. Both approaches were equally effective in relieving dysphagia but the Polyflex stent group required a lower number of dilations. In a pooled-data analysis of 10 studies using SEPSs for RBESs that involved a total of 130 patients, Repici and colleagues found that SEPSs were successfully inserted in 128 of 130 patients (98%). Clinical success using this approach was achieved in 68 patients (52%; 95% CI, 44%–61%) with a lower success rates in those with a cervical strictures (33% success rate in treating cervical strictures vs 54% for other sites; P <.05). Major complications occurred in 12 patients (9%; 95% CI, 4%–14%), resulting in one death (0.8%). Other complications included early (<4 weeks) migration in 19 patients (24%; 95% CI, 14%–32%), and 25 patients (21%) required postinsertion endoscopic reinterventions.

With the available data, the optimum duration for which SEPSs should remain in place is not certain. Although cleared by the FDA and approved for use in benign esophageal strictures for duration of up to 270 days, this approach can be associated with significant complication rates and should be considered in select patients and preferably in a protocol setting.

Biodegradable Esophageal Stents

Biodegradable stents are made of material that can be metabolized (polylactide/polydioxanone) by the body; hence, these stents do not need manual removal. The stents are uncovered, and tissue in-growth, if it occurs, can anchor the stent. Currently, Ella-CS, in Králové, Czech Republic, manufactures biodegradable stents of various lengths and diameters ( Table 2 ). Stents are packaged separately and require manual loading immediately before insertion. The stents are flexible and have radiopaque markers at the ends. There are limited data on their loss of radial force over time. According to company specifications, the stents maintain integrity and radial force for 6 to 8 weeks after placement and disintegration occurs by 11 to 12 weeks. Low pH (as can occur with gastroesophageal reflux) accelerates disintegration.

Table 2

Biodegradable esophageal stent (SX-ELLA biodegradable stent)

Diameter (mm)	Length (mm)	Delivery Catheter (mm)	Preloaded	Fully Covered	Company
16	60, 80, 100	Proximal 5.9 Distal 9.4	No (stent is manually loaded immediately before placement)	No	Ella-CS (Králové, Czech Republic)
18	60, 80, 100			No
23	60, 80, 100			No
25	60, 80, 100, 135			No

Saito and colleagues made a biodegradable stent by knitting poly-L-lactic acid monofilaments. They placed this stent in 13 patients (corrosive injury in 2, postsurgery in 4, and postendoscopic mucosal dissection in 7). Spontaneous migration occurred in 10 of 13 patients at 10 to 21 days postplacement. At 7 to 24 months’ follow-up, all patients were symptom-free. The same group was successful in treating 2 patients with strictures secondary to endoscopic mucosal dissection using biodegradable stents. A polydioxanone biodegradable stent (SX-ELLA BD stent, Ella-CS, Králové, Czech Republic) was placed in 4 patients with RBES. On follow-up endoscopic evaluation, stent disintegration was noted at 10 to 12 weeks after placement. During a short follow-up period of 4 to 17 weeks, all patients were dysphagia-free. In a recent prospective study from 2 European centers, 21 patients with RBES, as defined as Kochman and colleagues, were enrolled. A biodegradable stent (SX-ELLA) was placed and patients were clinically and endoscopically followed-up for 6 months and then only when symptoms recurred. Apart from 2 patients in whom the stent migrated, the stent was found fragmented in all patients endoscopically 3 months after deployment. At a median follow-up of 53 weeks (range 25–88), 45% patients were dysphagia-free. Severe pain (2 patients) and minor bleeding (1 patient) were some of the complications that occurred. Case reports of new esophageal fistula and tissue hyperplasia have been reported with biodegradable stents.

Only gold members can continue reading. Log In or Register to continue