See “Commentary on ‘Esophageal Endoscopic Dilations”’ by Dall’Oglio and De Angelis on page 716.
Esophagus strictures are rare but serious in children and require intensive management. The main challenge is to allow proper nutrition. Peptic stricture is the principal etiology in our context followed by caustic stricture and stricture subsequent to surgical treatment of esophageal atresia. Congenital stricture is rare. Esophageal dilation is one of the major therapeutic endoscopic procedures performed in our unit. The technique used is based on Savary-Gilliard bougies. Efficiency is important, and the risks are minimal.
During a period of 7 years (2004–2010), we conducted a study of esophageal stricture cases collected within the pediatric service of the University Hospital Hassan II of Fez. Patient information was gathered from the gastrointestinal endoscopy register and patient records.
A total of 60 patients were included in the present study. They had esophageal strictures confirmed by radiology and endoscopy. Our patients were divided into 3 groups: group A consequent to peptic exposure, group B for caustic agents, and group C for anastomotic stricture after surgical treatment of esophageal atresia.
We used video endoscope and Savary-Gilliard bougies for endoscopic dilations. No balloon dilation was performed. Dilators with a diameter ranging from 5 to 14 mm were selected based on age and degree of stricture. The most used dilators were 8 to 12 mm. Several rules were strictly respected: dilation was exclusively performed by 2 well-trained operators; it was performed carefully on children under general sedation using propofol, midazolam, and halogen gases (halothane or isoflurane) with appropriate intensive care–anesthesia equipment by an anesthesiologist; the patients fasted for at least 6 hours; the patients of group A and group B were first treated with proton pump inhibitors (PPI) such as omeprazole at 2 mg · kg−1 · day−1 at least 1 month before the dilation. The same drug was maintained for the duration of management; the maintenance dose was reduced to 1 mg · kg−1 · day−1; in group B, a corticosteroid such as methylprednisolone (1 g · 1.73 m2 · d−1) was administered intravenously during the first 2 weeks after ingestion; the diameter of the dilator should be appropriate without exceeding 3 increasing caliber bougies. The dilation session took an average of 15 minutes.
Initially, the pediatric endoscope was introduced to locate the stricture and, if possible, determine its extent, the state of the mucosa upstream and downstream, and determine whether hiatal hernia was associated. The guidewire is placed under endoscopic control through the stricture and then is pushed into the stomach, and the endoscope then withdrawn gradually. In the second step, the dilators were slipped one after the other on the wire, achieving longitudinal forces on the stricture. At the end, an endoscopic control was necessary to determine the diameter of the dilated stricture and to check the integrity of the esophageal and gastric mucosa. Checking the antireflux efficacy was systematic. After each session, patients remained at least 5 hours in the recovery room. Monitoring of breathing, heart rate, blood pressure, temperature, and subcutaneous emphysema or respiratory distress was always performed. The chest radiography was performed only when there was doubt about a complication.
Of the 60 esophageal strictures, 52 cases belonged to group A (86.6%). During the present study period and >2031 endoscopies performed, 824 endoscopic examinations were for gastroesophageal reflux (GER). Peptic complications were found in 640 cases with a prevalence of 77.7% and of 31.5% for the entire endoscopy performed. Peptic stricture represented 8.1% of these complications with an incidence of 0.6 cases per month.
In group B, 4 cases of caustic strictures were included. Two occurred after white spirit ingestion and 2 after lye ingestion. Four patients representing group C developed stricture at the site of the esophageal anastomosis subsequent to surgery during the neonatal period to correct esophageal atresia. Our patients were between 10 months and 17 years old for group A, with an average age of 5.7 years; between 13 months and 4 years old for group B, with an average age of 2.6 years; and between 12 months and 5 years old, with an average age of 2.2 years for group C. A male predominance was found in all of the groups, with a sex ratio of 1.3. The delay between the onset of clinical signs and the first endoscopic examination ranged from 3 months to 12 years for group A, from 1 day to 1 month for group B, and from 12 months to 5 years for group C. Dysphagia was the main clinical symptom often associated with chronic vomiting in all of the groups and gastrointestinal bleeding in group A. Half of the patients had failure to thrive at diagnosis. Respiratory signs were observed exclusively in the peptic group, mainly repetitive bronchopulmonary disease or chronic cough (Table 1). In group A, 30% of patients had already received a radiological diagnosis. In all of the groups, esophageal opacification revealed regular stricture in 91% of cases, with an average size of 3.7 cm (Figs. 1 and 2). Endoscopically, stricture was impassable in the first expansion in 73% of patients of group A, 25% of group B, and 100% of group C (Table 2). Of these impassable strictures, the expansion allowed passage through in 86.6% of the cases. In total, 247 dilation sessions were carried out, including 186 in group A, 42 in group B, and 19 in group C. In group A, the number of sessions ranged from 1 to 14, with an average of 3.6 sessions per patient. In contrast, in the caustic group, this average reached 10.5 dilations per patient, with a total session number ranging from 7 to 15 (Table 3). The interval between sessions was at least 3 weeks and averaged 4 weeks.
Two cases of esophageal perforation were observed. The first perforation occurred during an initial dilation of a peptic stricture, which had no need for additional dilations because it was cured as shown by endoscopy. The second perforation occurred in 1 case in the caustic group, a 4-year-old child who ingested a small quantity of lye. The incidence of perforation was then calculated for group A and group B, respectively, to 2% and 25% per patient and to 0.5% and 2.4% per dilation session. These patients responded well to conservative treatment.
The effectiveness of endoscopic treatment was evaluated based on clinical and endoscopic signs: improvement of nutritional status, which meant the disappearance of dysphagia and vomiting; disappearance of esophagitis lesions and regression of stricture in group A and B; and its disappearance in group C. Healing was reported after an average management duration of 1 year and a mean hindsight of 2 years. In group A, we obtained a favorable response in 70% of cases. Indeed, 29 patients (55.8%) were completely cured and 8 of 16 who continued to benefit from additional sessions started to respond well. Among all of these patients, 17 had a hiatal hernia and underwent a Nissen intervention, and 9 who had a small hernia were not operated on. Ten patients were scheduled for surgery after nutritional improvement. The cure rate of surgical patients was 69.6%, which means that 30.4% required postoperative dilation sessions. The average of dilation sessions was 2.8 preoperatively and 4.6 postoperatively. No cases of Barrett esophagus or other type of strictures were found. In group B, 2 patients were cured, 1 after 7 sessions and the other after 15. One failure was noted after 11 dilation sessions and that patient underwent esophagoplasty. The fourth case required additional dilations that were increasingly spaced out after 2 years of evolution. In group C, 2 patients responded favorably to endoscopic treatment, 1 after 10 months and the other after 1 year of evolution. Two other patients are still being followed up with repeated sessions (Table 4). In parallel, both group A and group B patients received medical treatment with a PPI before and after dilation to reduce the degree of underlying esophagitis. This treatment was administered to only 1 patient in group C because, in general, patients in this group had no evolving esophagitis. In addition, in group B, all of the patients initially received systemic injection corticosteroid therapy (methylprednisolone 1 g · 1.73 m2 · d−1) with a cessation of feeding for a variable period depending on the lesions. Only 1 child in the same group required a gastrostomy tube feeding concomitant with early dilation. No patient was treated with mitomycin C.
Esophageal pH monitoring was performed for all of the patients in group B and has shown an associated GER reflux in all of the cases. In group A, monitoring was done after clinical and endoscopic healing. The decision to stop treatment was made after a normal pH monitoring.
Esophageal strictures were mainly caused by peptic strictures in our context. The management of the pathology of gastroesophageal reflux disease (GERD) is still not up to our ambitions. Many of our patients had never consulted a clinician before the onset of dysphagia. These patients are generally living in remote rural areas, far from hospitals (>60% of cases). For others, antireflux treatment was either prescribed later after the start of symptoms or taken irregularly. In <10% of cases, patients developed a peptic stricture despite regular monitoring and appropriate treatment. Endoscopic explorations were requested by their physician only after signs of complication. These patients had relatively large hiatal hernias.
Theoretically, the time of peptic stricture onset is variable but most frequently occurs during the first 8 months of GERD; however, the outcome may be slower or even years after the beginning of symptoms, as shown in our series. The clinical diagnosis of peptic stricture is still uncertain. Gastrointestinal endoscopy, in some early forms, is the only examination that can confirm the stricture (1,2). Caustic stenosis was rare in our series. This cannot be explained by the nature of recruited patients because our center, which receives >25% of all Moroccan cases, manages all types of esophageal stricture. We estimate that caustic ingestions are common and certainly not all of them are seen in our center. In another study we conducted on caustic ingestion, the majority of patients ingested bleach, which generally does not cause injury. Ingestion of white spirit comes second and is not always complicated with stricture. Esophageal stricture occurred in 28% of white spirit ingestion cases and in 100% of lye-ingestion cases. Lye is highly corrosive.
The occurrence of anastomotic stricture after esophageal atresia surgery is frequently reported in the literature but not in our series. The frequency depends on the study and the type of atresia and on the quality of the surgical repair (3,4). Other types of stricture (congenital stricture, postinfection stricture, or those occurring after sclerosis of esophageal varicose veins) were not included in our study.
All of the symptomatic strictures could be dilated endoscopically, with good tolerability and ease of implementation (1,5). Two techniques are presently used in children: mechanical dilation by Savary-Gilliard bougies and pneumatic balloon dilation. The Savary-Gilliard technique is the oldest and is the one we adopted in our unit. The choice of dilation technique depends on personal experience. Balloon dilation appears, according to some authors, to be more effectively used and safe in caustic and postatresia strictures, although mechanical dilation in these indications also seems efficient (3,4,6–9). Norberto showed that balloon dilators are most useful for 2 situations: circumstances under which it is desirable to minimize esophageal trauma and short strictures. Savary-Gilliard dilators are useful for strictures resistant to balloon dilation and for long strictures that require carefully controlled and graded dilation (10). Studies comparing the response of peptic and caustic strictures using balloon dilation conclude that peptic strictures respond well to a smaller number of sessions (1,11). This observation is also true for Savary-Gilliard bougies, as shown in our work (8); however, there have been no studies comparing the 2 techniques in peptic strictures, and to the present time, we have no experience with hydrostatic dilation.
Endoscopic dilation of peptic stricture can improve outcomes and may avoid unnecessary surgical repair; however, long-term PPI treatment is required to cure GERD. The risk of relapse is still possible once the PPI treatment is interrupted, and a regular and long-term follow-up of these patients is required. Cases with a large hiatal hernia or those not responding to endoscopic treatment require surgery. For these patients, endoscopic treatment is still indicated to ensure proper nutritional status. Half of our patients had, at diagnosis, failure to thrive, usually resulting from a delay in disease management. After Nissen intervention, endoscopic control should occur; significant cases require more dilation sessions. Pearson et al (5) noted that children and adolescents with esophageal reflux stricture frequently require antireflux surgery (94%), repeat antireflux surgery (17%), and multiple dilations for recurrent symptoms. In our study, 30.4% of surgical cases still needed an average of 4.6 sessions of expansion. We noticed that patients who had several sessions after surgery were those who received laparoscopy instead of laparotomy.
Perforations occurred in 3% to 8% of cases depending on the series. This rarely occurred in our study if we consider the total number of sessions performed. The finding of a slight parietal break or moderate bleeding during postdilation control demonstrated the effectiveness of the expansion.
Caustic strictures are difficult to manage. Dilations must start after the healing of the esophageal mucosa, which is variable. In practice, as in the literature, we had not started the expansions before 1 month after caustic ingestion. Endoscopic treatment can be effective in 75% to 90% cases depending on the series, but it should be repeated over a long period of time (6,7,12). In our group, this period ranged from 18 to 24 months. Its purpose is to preserve the esophagus and prevent the use of esophagoplasty, which should be reserved for cases of failure of dilation or too-large strictures. Long-term careful follow-up must be ensured because of the risk of esophageal carcinoma, which remains poorly evaluated.
The risk of perforation in caustic stricture is higher (up to 15%). Such a perforation is not a contraindication for further dilations, once healing is completed. Finally, we must consider and treat GERD that occurs during treatment probably caused by dilations and contraction of the fibrotic esophagus (2,12). In our work, esophageal pH recording performed in the caustic stricture after an average period of 1 year showed GER.
Recurrence after dilation is frequently observed in both peptic and caustic strictures. Several techniques have been proposed to prevent fibrosis and are under development in our unit: the placement of esophageal prosthesis or stent, infiltration of corticosteroids, or application of mitomycin C in the lesion. Extractable metal prosthesis or biodegradable implants may soon become a good alternative for eradicating recurrent stricture that requires several rounds of dilation (13–15).
For esophageal atresia stricture, esophageal dilation with Savary-Gilliard bougies is an effective technique, but this usually requires several sessions to remove the stricture and circular fibrosis. Surgery of recurrent stricture on anastomosis consists of a resection of the stricture and the creation of a new anastomosis. This new anastomosis reexposes patients to the risk of stricture. Therefore, it is not proposed in initial treatment; however, the surgical treatment of GERD may be necessary in cases of recurrent stricture associated with persistent GERD despite well-followed treatment (3,4).
Esophageal strictures are severe and require extensive treatment. Endoscopic dilation using Savary-Gilliard bougies is an effective technique that ensures normal nutrition. Peptic stricture and esophageal atresia show better response to endoscopic dilation, with a smaller number of sessions than caustic stricture. Esophageal dilation may be an alternative to surgery; otherwise, it remains a complementary treatment that is almost mandatory either before or after surgery. We insist on the prevention of peptic complications whose frequency highlights the critical need for the intensification of management of GER.
We are grateful to Dr Mohammed El-Azami El-Idrissi (Laboratory of Immunology, Faculty of Medicine and Pharmacy, Fez) for the English revision of the manuscript.
1. Heresbach D, Michaud L, Lachaud A. Dilatation des sténoses bénignes de l’œsophage chez l’adulte et l’enfant. Acta Endosc 2010; 40:282–286.
2. De Lagausie P. Main esophagial pathology in children. EMC-Pédiatrie 2004; 1:109–117.
3. Serhal L, Gottrand F, Sfeir R, et al. Anastomotic stricture after surgical repair of esophageal atresia: frequency, risk factors, and efficacy of esophageal bougie dilatations. J Pediatr Surg 2010; 45:1459–1462.
4. Said M, Mekki M, Golli M, et al. Balloon dilatation of anastomotic strictures secondary to surgical repair of oesophageal atresia. Br J Radiol 2003; 76:26–31.
5. Pearson EG, Downey EC, Barnhart DC, et al. Reflux esophageal stricture—a review of 30 years’ experience in children. J Pediatr Surg 2010; 45:2356–2360.
6. Youn BJ, Kim WS, Cheon JE, et al. Balloon dilatation for corrosive esophageal strictures in children: radiologic and clinical outcomes. Kor J Radiol 2010; 11:203–210.
7. Temiz A, Oguzkurt P, Ezer SS, et al. Long-term management of corrosive esophageal stricture with balloon dilation in children. Surg Endosc 2010; 24:2287–2292.
8. Novais P, Lemme E, Equi C, et al. Benign strictures of the esophagus: endoscopic approach with Savary-Gilliard bougies. Arq Gastroenterol 2008; 45:290–294.
9. Jayakrishnan VK, Wilkinson AG. Treatment of oesophageal strictures in children: a comparison of fluoroscopically guided balloon dilatation with surgical bouginage. Pediatr Radiol 2001; 31:98–101.
10. Rodriquez-Baez N, Andersen JM. Management of esophageal strictures in children. Curr Treat Options Gastroenterol 2003; 6:417–425.
11. Bittencourt PF, Carvalho SD, Ferreira AR, et al. Endoscopic dilatation of esophageal strictures in children and adolescents. J Pediatr (Rio J) 2006; 82:127–131.
12. Contini S, Garatti M, Swarray-Deen A, et al. Corrosive oesophageal strictures in children: outcomes after timely or delayed dilatation. Dig Liver Dis 2009; 41:263–268.
13. Foschia F, De Angelis P, Torroni F, et al. Custom dynamic stent for esophageal strictures in children. J Pediatr Surg 2011; 46:848–853.
14. Reinshagen K, Kähler G, Manegold BC, et al. Evaluation of stents in treating childhood benign esophageal strictures. Klin Padiatr 2009; 221:25–30.
15. Ancona E, Guido E, Cutrone C, et al. A new endoscopic technique for suspension of esophageal prosthesis for refractory caustic esophageal strictures. Dis Esophagus 2008; 21:262–265.