The March 2020 issue of the scientific journal Digestive Liver Disease publishes a systematic review that assesses the clinical presentation, management, and outcomes of various interventions for esophageal perforation in eosinophilic esophagitis, in order to define the most appropriate management for patients who present this complication.

Dr. Laura Arias-González and colleagues, from the General Hospital of Tomelloso, in Spain, have evaluated 50 documents recovered in an exhaustive literature search, which describe 70 patients who suffered 76 esophageal perforations, being 13 of them children or adolescents.

A diagnosis of EoE had not been established at the moment 51 out of 76 patients suffered from perforation, therefore no treatment was instituted. In most of the remaining patients, the treatment had been found to be ineffective in controlling eosinophilic inflammation. Active eosinophilic inflammation was thus identified as a first risk factor for this complication.

Patients who consulted due to perforation mainly reported intense chest or epigastric pain after a food impaction episode. In some cases, fever or dyspnea also appeared. In 41 cases, the pain appeared after patient's efforts to induce vomiting to dislodge an impacted food. In 8 cases perforation appeared as a complication after endoscopic dilation with through-the scope balloons or Savary bougies. A CT scan was reported to be used to demonstrate esophageal perforation in 92% of episodes.

Regarding management, 67% were managed conservatively with no need for surgery, which mainly consisted of nil per os with parenteral diet, broad-spectrum antibiotics, proton pump inhibitors and analgesics. 10 patients underwent esophageal stenting as the primary measure to close the perforation, by using both full- and partially covered metallic stents. In two patients the primary mucosal tear was endoscopically closed with endoscopic clips or over-the-scope clips.

The remaining 33% of perforations underwent surgery, which mainly included repair of the perforation through thoracotomy or, less commonly, laparotomy or laparoscopy. An esophagectomy with gastroplasty was performed in 5 cases. Only 1 patient required an esophageal stoma and feeding jejunostomy, followed by reconstruction. The only finding that made surgical management more likely than a conservative one was pleural effusion. The date of publication of each report was not associated with the preferred management of perforation.

Most of the patients recovered uneventfully from perforation, and were discharged from hospital within a 1week period. Hospital discharge happened in the second week after perforation in only 7 patients and one additional patient was discharged later due to a more complex post-operative course which required continuous drainage from the site of rupture. Most of the esophageal stents were removed before the third week.

The authors conclude that, despite the potential severity of an esophageal perforation, two thirds of the EoE patients described in the literature were effectively treated conservatively, and none of them underwent surgery after failure of the primary treatment. Surgical repair of the perforation was, by contrast, more common in patients with complications of perforation. Importantly, no death was reported, and the majority of patients could be discharged with the first week after perforation. Preventing perforation in EoE requires improved identification and early diagnosis of patients with suggestive symptoms, treating them with effective medication or diets to suppress eosinophilic inflammation, and to perform careful endoscopic procedures in case of active inflammation.