Systematic Review With Meta-analysis

Endoscopic Dilation Is Highly Effective and Safe in Children and Adults With Eosinophilic Oesophagitis

F. J. Moawad; J. Molina-Infante; A. J. Lucendo; S. E. Cantrell; L. Tmanova; K. M. Douglas


Aliment Pharmacol Ther. 2017;46(2):96-105. 

In This Article


This systematic review and meta-analysis was performed to assess the clinical efficacy and safety of dilation in EoE. Our previous meta-analysis was limited to a relatively small number of studies and strictly to an adult population.[16] Since then, several other centers have published their data on complications rates with dilation, including the first paediatric series. Patients who have undergone dilation have more than doubled and a threefold increase in endoscopic procedures has been observed since our previous meta-analysis. We demonstrated that dilation was highly effective in improving dysphagia. The frequency of major complications, which consisted of perforation, haemorrhage, hospitalisation, or death, was rare and occurred in well below 1% of patients.

Our aim was to assess the clinical efficacy of dilation in EoE and dilation was found to improve symptoms in 95% of patients. One of the earliest studies by Straumann et al. followed the natural history of 30 patients with EoE, 11 of whom were treated with dilation.[54] Clinical improvement was observed in 91% patients. Several more recent studies with a larger population of patients have also demonstrated efficacy. In the study by Runge et al. in which 164 EoE patients were dilated, 85% achieved clinical response.[24] Compared to our previous meta-analysis, the effectiveness rate has increased by 15% and importantly, heterogeneity in clinical improvement has dropped from 86% to 10%, hinting at a highly consistent effectiveness among recent studies published over the past 3 years.

Stricture formation is a consequence of long-term untreated or under treated EoE.[12] Studies have demonstrated that length of delay in diagnosis correlates with the presence of fibrostenotic features.[12,13] Endoscopic dilation does not impact the underlying inflammation and should be combined with an anti-inflammatory therapy. Since topical steroids and diet have shown their ability to reverse fibrotic remodelling,[55–57] it is plausible to speculate whether anti-inflammatory therapies may also reverse endoscopic features like strictures and narrow calibre oesophagus. This has been recently shown in small series or case reports and should be further explored.[58,59]

Initial case reports and small studies cautioned endoscopists about dilation in that was a newly encountered cause of food impaction and strictures.[19,20,60] Reports of perforation surfaced, as did the dramatic appearance of mucosal shearing following dilation or passage of an endoscope.[18,61] A large number of institutional studies is now available in children and adults. These include prospective cohorts and one randomised controlled trial demonstrating safety of dilation with a complication rate comparable to dilation for other types of oesophageal strictures.[62]

In our review, we only found seven cases of perforation out of a total of 1820 dilations, which were reported in three studies. One study was a retrospective design at a single institution which resulted in three perforations from 293 dilations, a second was an audit on endoscopic complications from dilations from a tertiary care center in which three perforations occurred, and the final perforation was reported in a case report of a 17-year-old woman. We excluded cases that had perforation reported by history as we wanted to ensure endoscopic dilation was the actual cause rather than spontaneous perforation from a delayed food disimpaction.[17] In our review, there was only one case of haemorrhage reported following dilation and there was not a single death reported.

The most frequent complication reported in studies exploring the safety of dilation in EoE was mucosal laceration. However, it is noteworthy that mucosal lacerations or even deeper rents are not actually complications, but rather the intended outcome of dilation and patients may not experience clinical improvement unless a tear develops. Given the variation in studies describing this feature, we were not able to calculate a summary effect. In addition, mucosal laceration is likely underreported as second-look endoscopy is not always performed following Savary or Maloney dilation. With mucosal tears of the oesophageal mucosa, post-procedural chest pain may develop. This is also most likely an underreported finding that patients may not seek medical attention for pain and endoscopists may not follow-up with patients within the days after dilation. Among the studies included, 9% of patients experienced chest pain, however less than 1% required hospitalisation for pain management. The only study to demonstrate a discrepancy between chest pain found in patient records versus in a post-procedural survey was by Schoepfer et al.[21] While medical records documented chest pain in only 7% of cases, in actuality, it occurred in 74% of patients, albeit mild. In most cases, this pain is self-limited, and can be managed with topical analgesics.

In our review, two cases were dilated with EndoFLIP, which has shown that reduced oesophageal distensibility can increase risk of food impaction, regardless of eosinophilic inflammation,[63] EndoFLIP may accurately identify patients with recurrent food impaction or dysphagia and therefore may identify candidates for oesophageal dilation. Whether the addition of this tool will enhance our understanding of dysphagia in EoE and refine our indications for endoscopic dilation remains to be elucidated.

Last, one of our aims was to explore whether factors can predict improvement in symptoms. Some studies reported a target oesophageal diameter following dilation, however, the data were not sufficient to obtain a summary estimate. We also wanted to compare the safety and efficacy of the three types of dilators, but given the limited number of studies with very low number of events, we could not perform meta-regression.

Our meta-analysis has several strengths. We performed a comprehensive search strategy of all major databases, as well as, abstracts from major meetings, without any restriction to language. Professional librarians with experience in meta-analysis assisted us in our search strategy. All studies were screened by two teams in duplicate and data abstraction was performed in duplicate as well to minimise bias. Our review included children and adults therefore making our results more generalisable to all EoE patients.

The major limitation of our review was the quality of evidence included. There was only one small randomised controlled trial and two prospective cohorts. The majority of studies were retrospective and case reports, therefore, we were not able to grade the quality of studies included. This is particularly a limitation when attempting to assess clinical effectiveness of a procedure. The duration of follow-up after dilation was not reported in all studies and was limited to a relatively short period of time. We aimed to explore difference in types of dilators in complications and response, however, due to the limited number of studies with a low number of events, we could not compare differences.

In summary, oesophageal dilation is highly effective in improving clinical symptoms, at least in the short term, and is a safe procedure with a very low major complication rate.