Pathophysiology and Incidence

The normal process of wound healing after creation of the esophageal anastomosis involves the creation of scar tissue. During the tissue remodeling phase of wound healing, fibroblasts promote wound contraction. Tissue contraction of open wounds is beneficial to close the injury; however, wound contraction in the setting of a circular end-to-end anastomosis creates narrowing. Therefore, it is quite natural to see a degree of narrowing at the site of the esophageal anastomosis after EA repair.

The reported incidence of anastomotic stricture after EA repair has varied in case series from as low as 9% to as high as 80%. There are several factors implicated in the pathogenesis of anastomotic stricture. These include creation of the esophageal anastomosis under excessive tension, ischemia at the ends of the esophageal pouches, creation of the anastomosis with 2 suture layers, use of silk suture material, anastomotic leak, esophageal gap length greater than 4 cm (long gap EA), and postoperative gastroesophageal reflux.

Esophageal Stricture Symptoms and Definitions

When a swallowed food bolus becomes too large to pass through the narrowed portion of the esophagus, symptoms of dysphagia will occur. Typical symptoms of an esophageal stricture include feeding difficulties, coughing and choking during feeds, food impaction, and regurgitation of undigested material. In younger children, apnea may be a presenting symptom as well as feeding refusal. If a patient with EA develops any of these symptoms he or she should undergo a contrast fluoroscopy study and/or endoscopy to evaluate for a possible stricture. An esophageal stricture therefore is defined as an intrinsic luminal narrowing that leads to the patient becoming clinically symptomatic.

In older children and adolescents, an esophageal luminal diameter of 13 mm or smaller typically results in dysphagia to solids. Dysphagia to liquids occurs when the esophageal diameter is even narrower. Typically, liquid dysphagia will occur with esophageal diameters of 5 mm or smaller; however, we have seen children drink normally with stricture diameters much smaller than this. Dysphagia to purées will occur generally when the esophageal diameter is 8 mm or smaller to 10 mm depending on the viscosity of the purée.

In adults, strictures are classified as simple or complex. A simple stricture is defined as one that is, short in length, focal, straight, and allows passage of an adult-diameter endoscope (≈9 mm). A complex stricture is classified as having one of the following characteristics: long in length (>2 cm), angulated, irregular, and or has a severely narrow diameter. In general, simple strictures are more easily dilated and require fewer treatments to remediate the lumen size. Complex strictures are more likely to become refractory or recalcitrant.

In adults, a stricture is defined as refractory or recalcitrant when there is an inability to remediate the esophageal lumen to a diameter of 14 mm during 5 dilation sessions at 2-week intervals. A recurrent stricture is defined as the inability to maintain a lumen patency for 4 weeks once the target diameter of 14 mm has been achieved. An alternative definition of a refractory stricture is a stricture that requires ongoing dilation sessions (more than 7–10).

There are no agreed on definitions for benign refractory strictures in pediatrics. It is our practice to follow the adult definitions; however, we modify the goal diameter based on the patient’s age. For infants 6 months of age or younger, our goal lumen diameter to achieve is 10 mm. For infants older than 6 months and children to approximately the age of 7, we use a diameter of 12 mm. For older children approximately 7 years of age or older, we use the adult definition with a diameter of 14 mm.

In an attempt to better standardize and quantify strictures in children, Said and colleagues developed an anastomotic stricture index (SI) defined as follows: SI = ( D − d ) D × 100 , where D is the diameter of esophagus below the stricture and d is the diameter of the stricture. In this study, dilations were considered successful if the SI decreased to 10% or less. Parolini and colleagues used the SI to classify strictures into mild (>0.1) and severe categories (≥0.3). It is important to note that Parolini and colleagues did not specify whether they measured the upper or lower esophagus in their calculation of the SI. Not surprisingly, both mild and severe groups were at increased risk of requiring esophageal dilations. A definition for a recalcitrant and recurrent pediatric stricture based on the SI has been proposed. A recalcitrant stricture would be a persistent SI greater than 10% after 5 dilation sessions, and a recurrent stricture would be an SI greater than 50% 4 weeks after SI less than 10% had been achieved.

Nambirajan and colleagues described an anastomotic index (AI) calculated as a ratio of the diameter of the upper pouch to the diameter of the anastomosis on contrast study. In this study, the AI did not predict the development of a recalcitrant stricture.

Recently, a new SI, the Esophageal Anastomotic SI (EASI), was described by Sun and colleagues. In this study, the EASI was calculated at the time of postoperative contrast esophagram (5–10 days post anastomosis), and defined as follows: lateral d D + Anteroposterior d D 2 . In this equation, d is the diameter of the stricture and D is the diameter of esophagus either below the stricture or above the stricture. The investigators concluded that EASI was more accurate using the diameter of the esophagus below the stricture as opposed to the calculations using the diameters above the stricture. This is likely due the upper pouch generally being more dilated secondary to chronic obstruction in utero, as well as increased pressure in the upper esophagus secondary to increased resistance across the stricture area. In their study, an EASI of 0.30 or less highly correlated with requiring a course of dilatations.

Although all of the stricture indices show merit in standardizing stricture evaluation in pediatrics, further validation is required before any one can be accepted into clinical practice. In the meantime, it remains unclear whether these indices are clinically more useful than just estimating the stricture diameter. Indeed, it is possible that these indices may only prove useful for academic purposes. Furthermore, although estimating the size of the stricture is important, the patient’s clinical symptoms should always be taken into account before starting any stricture therapy.

Dilation

The cornerstone of esophageal stricture treatment is dilation. The goal of esophageal dilation is to increase the luminal diameter of the esophagus while also improving dysphagia symptoms. This is achieved through circumferential stretching and splitting of the scar tissue within the stricture. Dilators can be characterized as mechanical (bougie) dilators or balloon-based dilators.

Mechanical (Bougie) Dilators

There are several different types of bougie-based dilators. Today, the most common types of bougie dilators used are guidewire based. These dilators are tapered cylindrical solid tubes made of polyvinyl chloride with a central channel to accommodate a guidewire. These dilating tubes have varying lengths of tapering at the tip and also have radiopaque markers to allow for fluoroscopic guidance (eg, Savary-Gilliard dilators (Cook Medical, Bloomington, IN), American Dilators (CONMED, Utica, NY), and Safeguide (Medovations, Milwaukee, WI)).

There are also non-guidewire mechanical dilators that are tungsten weighted to allow for gravity assistance when the patient is in a seated position. The 2 commonly used non-guidewire bougie dilators are Hurst (Medovations, Milwaukee, WI) and Maloney dilators (Medovations, Milwaukee, WI). These 2 dilators differ by their tips. The Hurst dilators have a rounded blunt tip, whereas the Maloney dilators have a tapered tip. Both dilators were designed for self-dilation at home.

Another type of mechanical dilator is the Tucker dilator (Medovations, Milwaukee, WI). These are small silicone bougies that are tapered at each end. There are loops on each end with a string attached to each end to allow for the dilator to be pulled antegrade or retrograde across strictures. These dilators are used when the patient has a gastrostomy tube. Tucker dilators can remain inside the patient for periodic serial dilations.

The basic technique of mechanical dilations involves the passage of a bougie dilator across the stricture. This results in both longitudinal shearing force as well as radial force on the stricture area. The goal of mechanical dilation is to pass serial bougie dilators of incremental size across the stricture site. Although fluoroscopy can be helpful to confirm your position as you pass the bougie dilator, it is not mandatory. It is generally recommended to use fluoroscopy in complex strictures.

Mechanical dilation is a tactile technique. As the bougie is advanced across the stricture site, it should be possible for the proceduralist to be able to feel a degree of resistance. The object is to feel and then overcome the resistance across the strictured area. Once moderate resistance is encountered with the bougie dilator, it is generally recommend passing no more than 3 consecutive dilators in increments of 1 mm in a single session for a total of 3 mm. Although there are no published studies, this consensus, known as the “rule of 3,” is a well-established approach for mechanical dilations that is believed to improve safety and efficacy.

Balloon Dilation

Balloons deliver equal radial force simultaneously across the entire length of the stricture. They are designed to pass through the endoscope with or without a guidewire. Through-the-scope dilation allows the endoscopist to directly visualize the stricture during and immediately after the dilation. A potential drawback for through-the-scope balloon dilation in EA stricture therapy is that it requires the use of an adult gastroscope, which has a minimum working channel of 2.8 mm. This size gastroscope can be difficult to use in younger infants weighing less than 10 kg.

In younger patients, the balloon can be passed over a guidewire under fluoroscopic guidance. This is performed by passing a 0.035-mm guidewire across the stricture through the endoscope working channel. A wire exchange under fluoroscopy is performed, leaving the wire in place as the scope is removed. The balloon is then passed over the wire.

Dilating balloons expand by the injection of liquid (eg, water, radiopaque contrast) under pressure using a handheld inflation device. A manometer on the device will measure the fluid pressure in the balloon to allow for accurate radial expansion force.

Balloon dilators are designed to inflate to one set diameter or allow for sequential inflation to multiple sizes (typically 3 sizes per balloon). Multisized balloons will inflate to different sizes based on the amount of pressure infused into the balloon.

The basic approach to balloon dilation is to first estimate the size of the stricture. Once the size is estimated, the “rule of 3” can be applied to balloon dilators by choosing a balloon that will increase in size by increments of 1 mm in a single session for a total of 3 mm above the originally estimated stricture size. The balloon is advanced across the stricture either with endoscopic and/or fluoroscopic guidance. Ideally the balloon should be centered so that the middle of the balloon is centered across the stricture. Balloons are available with or without a wire.

The typical approach in our program is to have a wire across the stricture and typically into the stomach. The goal of the wire is to make certain that the tip of the balloon remains within the lumen of the esophagus. Having a wire across a complex stricture is recommended. If a complex stricture is encountered that is very narrow and/or torturous and a wire is not used, it is possible for the tip of the balloon to dissect through the esophageal wall.

Once the balloon is properly positioned, it can be inflated to the desired size. The optimal inflation time has not been established. Balloon inflation times of 30 to 60 seconds are generally accepted. One recent study examined the effect of duration of balloon dilation by randomizing stricture patients into 2 groups. The first group had balloon inflation times lasting only 10 seconds and the second group had balloon inflation times of 2 minutes. In this study, there was no significant difference in effectiveness of dilations in either group. Therefore, it appears that the act of inflation that tears the scar tissue is more important than the duration the balloon is inflated.

The use of fluoroscopy during balloon dilation is helpful. In the setting of a complex stricture, fluoroscopy is useful in advancing the wire and balloon safely across the stricture. In addition, inflating the balloon with contrast will allow the endoscopist to see if the stricture is being effectively dilated. It is useful to see the appearance of the stricture forming a waist around the balloon and the subsequent obliteration of a waist as the balloon is further inflated ( Fig. 1 ). It is a practice of this author to use fluoroscopy frequently with our EA stricture dilations. In addition, there is the added benefit of using fluoroscopy to conduct a postdilation contrast study to evaluate for a postdilation esophageal leak or perforation.

Fluoroscopic image of esophageal stricture waist around the balloon.

Outcomes and Comparative Data

A recent systematic review analyzed 5 studies that looked at outcomes of balloon dilation in children with EA. This study looked at total of 139 children with a total of 401 balloon dilation sessions. The reported success rate ranged from 70% to 100%, with approximately 3 dilation sessions per child. The reported perforation rate for the combined studies was 1.8% following balloon dilation. The reported success rate for bougie dilations in EA has been reported from 87% to 90% with mean number of 3.2 dilations per patient in both studies. There was one reported esophageal perforation in the study with bougie dilation that was fatal to the patient. In a large study of bougie dilations in children (n = 107), the reported perforation rate was 0.9%. In this study, the stricture population was mixed, with most patients having caustic strictures.

There have been several studies comparing the effectiveness of bougie dilation with balloon dilation. Lang and colleagues retrospectively looked at patients with EA who underwent bougie dilation compared with balloon dilation. The balloon dilation group required fewer procedures compared with the bougie group: median 2.0 versus 8.5, respectively, P = .002. Jayakrishnan and colleagues looked at technical success defined as being able to perform the dilation in 37 children (24 patients with EA). There were fewer technical failures in the balloon group compared with the bougie group (0/125 vs 4/88, P <.02). In addition, there were fewer perforations in the balloon group compared with the bougie group 1.6% versus 5.7% ( P = .1).

In adults, there have been 2 randomized controlled trials comparing bougie dilations with balloon dilations in benign esophageal strictures. In these 2 studies, there was no significant difference in efficacy and safety. Therefore, it is generally recommended that the method of choice between balloon dilation and bougie depends on operator experience and comfort with the equipment. The only time that bougie dilation is completely contraindicated is in the setting of patients with epidermolysis bullosa. In this setting, the longitudinal sheering force may result in further damage to other segments of the esophagus.

Intralesional Steroid Injection

The proposed mechanism of intralesional steroid injection in the treatment of esophageal strictures is to locally inhibit the inflammatory response, which in turn results in reduced collagen formation. The efficacy of intralesional triamcinolone injection in peptic strictures has been shown by Ramage and colleagues in a randomized double-blind placebo-controlled trial. In this study, 2 (13%, 95% confidence interval [CI] 4%–38%) of 15 patients in the steroid group and 9 (60%, 95% CI 36%–80%) of 15 in the sham group required repeat dilation ( P = .021). Those patients who received intralesional steroids were administered 4 quadrant injections of 0.5 mL of triamcinolone acetate (40 mg per mL) for total of 80 mg.

There have been multiple studies that have shown the benefit of intralesional steroids in reducing recurrent stricture formation. However, most reports are small uncontrolled studies evaluating strictures of diverse etiology. Hirdes and colleagues, in a multicenter double-blind placebo-controlled trial involving 60 patients with benign esophagogastric anastomotic strictures, reported no statistically significant decrease in frequency of repeat dilations with a median number of 2 dilations (range, 1–7) performed in the corticosteroid group versus 3 dilations (range, 1–9) in the control ( P = .36). In addition, there was no improvement in dysphagia-free symptoms in each group; 45% of the patients remained dysphagia-free for 6 months in the triamcinolone group, compared with 36% of controls (relative risk, 1.26; 95% CI 0.68–2.36; P = .46).

Potential complications of intralesional steroid therapy include adrenal suppression. Therefore, some investigators suggest surveillance for adrenal suppression ; however, this is currently not standard of care practice. In addition, there have been reports of increased Candida esophagitis. Last, there has been one report of intralesional steroids contributing to the spontaneous rupture of a right aortic arch, presumably secondary to the steroids weakening the arterial wall.

It is our group’s practice to start intralesional steroid injections once a stricture is at its third or fourth dilation, if there has been no significant improvement in intraluminal diameter. We use triamcinolone acetate at a concentration of 10 mg/mL at a dose of 1 to 2/mg/kg per dose with a maximum dose of 80 mg. The injections are typically 4 quadrant; however, if the scar tissue is uneven, a preponderance of steroid will be injected at the site where the scar tissue is at its thickest.

Injection is typically recommended directly into the scar tissue; however, we also like to inject subcutaneously, just above the stricture, to allow for diffusion of the medication into the scar tissue area. The rationale behind this is that by directly injecting steroid into the scar tissue, you are more likely to extravasate steroid outside of the esophagus. This may potentially lead to increased risk of complications, as well as decreased effectiveness.

Our group typically will perform a combination of direct injection into the scar tissue and subcutaneous injection above the stricture. Typically patients will undergo anywhere from 1 to 3 steroid injection sessions in combination with dilation before considering additional adjunct therapy.

Mitomycin C

Mitomycin C is an antineoplastic agent that disrupts base paring of DNA molecules and inhibits fibroblast proliferation and reduces fibroblastic collagen synthesis by inhibiting DNA-dependent RNA synthesis. It also induces apoptosis at higher doses by suppressing cellular proliferation during the late G1 and S phases. It has been proposed as an adjunct treatment to manage esophageal strictures. Mitomycin C has been mainly placed topically in the literature; in addition, there are reports of injection of mitomycin C.

There have been numerous descriptions of methods for topically applying mitomycin C. These have included soaking pledgets or cotton swabs and placing them topically on the stricture area; dripping mitomycin via an injection needle onto the affected area; and using a spray catheter. A recent publication described a novel approach that uses a microporous polytetrafluoroethylene catheter balloon. The dose of mitomycin C used in these studies is also variable, ranging from 0.004 mg/mL to 1 mg/mL.

A systematic review of the literature identified 11 publications using mitomycin C in children. The results of the systematic review demonstrated complete relief of symptoms in 21 children (67.7%), with another 6 (19.4%) who had partial relief, and 4 with (12.9%) no benefit. There were no direct or indirect adverse effects of mitomycin C reported. The mean follow-up time for all of these studies was 22 months (range 6–60 months). It is important to note that most of the literature on mitomycin C involves children with caustic strictures, and therefore results may not be applicable to anastomotic strictures. Even in this systematic review, 61% of the patients had caustic strictures and 22% had anastomotic strictures secondary to EA.

A recent prospective study looking at 30 children with caustic strictures compared mitomycin C therapy to dilation therapy alone. Although this study was not blinded and does not appear to be randomized, the mitomycin C group did show a statistically significant improvement in dysphagia score ( P = .005), and an improvement in the median interval between dilations. The mitomycin group required dilations every 10 weeks after treatment compared with the control group, which required dilations every 4 weeks.

The largest published study to date of mitomycin C in patients with EA looked retrospectively at 21 patients. In this study, 11 patients received mitomycin C topically, and were compared with 10 historical EA controls who received a minimum of 3 dilations. The mitomycin C group received a mean total of 5.4 dilations per patient (range 3–11) and 8 of 11 patients achieved a resolution of their strictures. These results were not significantly different from the control group, which may contradict enthusiastic case reports of the benefit of mitomycin C in patients with EA. Of course, many of these studies lack strict definitions of recalcitrant strictures or criteria for success. Nevertheless, mitomycin C is still promising and likely should be considered a therapeutic modality in patients with recalcitrant strictures secondary to EA.

There is a hypothetical risk of secondary malignancy with mitomycin C, so this must be taken into account and should be discussed with the patient and parents before use. There have been reports of de novo gastric metaplasia around the areas of the anastomosis in 2 of the 6 cases that received topical mitomycin C. This author suggests that long-term follow-up with esophageal biopsies at the site of mitomycin C application should be performed. It is for the hypothetical risk of secondary malignancy with unproven efficacy that our group’s preference is first a trial of triamcinolone injections before attempting mitomycin C therapy.

Stents

The use of temporary externally removable stents has been reported to provide an alternative or adjunctive means of preventing stricture formation by providing a continuous means of dilating the esophagus for prolonged periods of time. The 2 types of externally removable temporary stents currently available are self-expanding plastic stent (SEPS) and fully covered self-expanding metal stents (FCSEMS). The current SEPS on the market is made of a woven polyester mesh with a silicon membrane coating to prevent ingrowth of tissue (Polyflex; Boston Scientific Corporation, Marlborough, MA). SEPS are packaged fully expanded and need to be manually loaded onto a delivery system before placement.

FCSEMS consist of a woven, knitted, or laser-cut metal mesh covered by a plastic or silicone membrane to prevent ingrowth. Most FCSEMS are made of Nitinol, which is an alloy of nickel and titanium that has the properties of shape memory as well as super elasticity. Therefore, Nitinol stents have the ability to come packaged greatly compressed on a loading device and once deployed it will self-expand to its predetermined shape. Both SEPS and FCSEMS are deployed via a delivery device catheter over a guidewire under fluoroscopic guidance. However, FCSEMS are easier to place due to increased flexibility and smaller size of the delivery device, as well as the small compressed nature of the nondeployed stent.

There are currently no esophageal stents on the market designed for pediatric patients. The smallest esophageal stent currently available in the United States is the FCSEMS ALIMAXX-ES Fully Covered Esophageal Stent (Merit Medical Systems, South Jordan, UT) with a diameter of 12 mm and a length of 7 cm. This is why most stents placed in the esophagus of pediatric patients are airway stents (ie, Polyflex [Boston Scientific Corporation], AERO Fully Covered Tracheobronchial Stents [Merit Medical Systems], and Taewoong Niti-S Tracheobronchial Stents [Taewoong Medical, Goyang, South Korea]).

Airway stents range in diameter range from 8 mm to 20 mm and lengths ranging from 20 mm 80 mm. A drawback of airway stents is that they are more rigid than traditional esophageal stents, which in our experience leads to more ulceration to the esophagus and increase patient discomfort. An alternative to airway stents are biliary stents. There are several fully covered biliary stents on the market (ie, WallFlex [Boston Scientific Corporation], GORE VIABIL [Gore Medical, Flagstaff, AZ] and Taewoong Niti-S [Taewoong Medical]). Biliary stents are more flexible than airway stents, which we have found to cause less esophageal tissue injury, and to be associated with better patient comfort and tolerance. Biliary stents are limited only by their size, with available diameter sizes of 8 mm and 10 mm, and lengths of 40 mm, 60 mm, or 80 mm.

The use of removable stents to definitively treat benign esophageal strictures in adults has yielded mixed results, with a broad range of success rates from 12% to 80% for stricture resolution reflecting both retrospective and prospective study designs. Relevant to our discussion may be the study by Barthel and colleagues, which had the lowest success rate of 12% and looked primarily at anastomotic strictures.

In addition, a recent pooled analysis study of externally removable stents for benign esophageal strictures in adults (n = 232 patients) calculated a pooled success rate of externally removable stents in the treatment of refractory benign strictures to be 24.2%, with a pooled stent migration rate of 16.5% (77/468). Other complications associated with stent placement in this population included chest pain (0.8%), granulation tissue (2.7%), ulceration (0.8%), nausea and retching (0.8%), stent-related perforation (1%), and stent-related stricture formation (0.2%).

The pediatric data on stricture resolution after stent placement has been more promising, with success rates ranging from 50% to 86% ; however, most pediatric studies have been limited by small numbers, as well as mixed stricture populations. Our group has published the largest review to date of self-expanding stents for the treatment of anastomotic strictures with EA. In our study, 23 patients with EA underwent a total of 40 stenting sessions. Both SEPS (n = 14) and FCSEMS (n = 26) were used. Procedural success was defined as requiring no additional therapy after stent removal at 30 days or more and at 90 days or more. The rate of stricture resolution for 30 days or more after final stent removal was 39% (9/23) and a 90-day success rate of 26% (6/23). We also found the mean duration of stent placement was 9.7 days (with a range of 2–30 days). Complications of stent placement in our population included migration (21% of SEPS and 7% of FCSEMS), granulation tissue (37% of FCSEMS and 0% of SEPS), deep tissue ulcerations (22% of FCSEMS and 0% if SEPS), as well as pain and retching (26% of FCSEMS and 23% of SEPS).

The ideal duration of time a stent remains in place is unknown. In general, consensus is to leave the stent in place until inflammation has resolved, which is believed to reduce the risks of restricturing after stent removal. In the cases of severe adult strictures, the recommended duration varies from 8 to 16 weeks. In our study, the mean duration of time was 9.7 days, although our goal was to try to leave stents in place for 2 weeks.

Anecdotally, our group has noticed increased ulceration with longer stent duration. Therefore, our protocol is to keep patients in the hospital for the duration the stent is in place, using monitoring with periodic chest radiographs to assess for migration as well as for assessing the orientation of the stent to evaluate for erosion into the esophagus ( Fig. 2 A, B). We typically will endoscopically reevaluate a patient 1 week after the stent is placed to evaluate the tissue for any injury. At that time, we decide to leave the stent in place, replace it, or remove it completely.

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

Share this:

• Click to share on Twitter (Opens in new window)
• Click to share on Facebook (Opens in new window)

Related posts:

Training and Assessment in Pediatric Endoscopy Informed Consent for Pediatric Endoscopy Management of Upper Gastrointestinal Bleeding in Children Pediatric Considerations in Endoscopic Retrograde Cholangiopancreatography Role of Endoscopic Ultrasound in Pediatric Disease Advances in Pediatric Small Bowel Imaging

Stay updated, free articles. Join our Telegram channel

Join