Management of iatrogenic esophageal perforations: a systematic review of non-surgical causes

Iatrogenic esophageal perforations (IEP) are uncommon but potentially fatal complications, most frequently related to endoscopic procedures. Previous systematic reviews have evaluated esophageal perforations as a heterogeneous entity, frequently combining spontaneous, traumatic, surgical, and iatrogenic causes. However, outcome estimates specific to non-surgical, iatrogenic perforations remain less clearly defined. This study aimed to systematically review and meta-analyze the causes, management strategies, and outcomes of patients managed for non-surgical IEPs.

A systematic search of MEDLINE, Embase, CENTRAL, Scopus, and ClinicalTrials.gov was performed from inception to September 2025 in accordance with PRISMA guidelines. Studies reporting clinical outcomes of adult patients with iatrogenic, non-surgical esophageal perforations were included. Postoperative surgical perforations and case reports were excluded. Primary outcomes were procedure type leading to perforation, overall mortality, and mortality according to treatment strategy. Random-effects meta-analyses were conducted for pooled estimates.

Twenty-five studies encompassing 596 patients were included. Interventional procedures accounted for 71.1% of perforations, while diagnostic procedures accounted for 28.9%. Diagnostic endoscopy was significantly less likely to cause perforation compared with interventional procedures (OR 0.20, 95% CI 0.08-0.45). The pooled overall mortality rate was 9.75% (95% CI 5.9%-13.6%). Among studies reporting treatment-specific outcomes, endoscopic-based management was employed in 48.1% of patients, most commonly involving esophageal stenting, and was associated with a pooled mortality of 11.37% (95% CI 2.84%-19.9%). Surgical management was used in 51.9% of cases, predominantly primary repair or esophagectomy, with a pooled mortality of 11.58% (95% CI 7.13%-16.03%).

Non-surgical IEP occurs predominantly after interventional procedures and remains associated with substantial mortality approaching 10%. Endoscopy-based management, particularly with stent placement, has become central to treatment and demonstrates outcomes comparable to surgery in selected patients, while surgery remains as a salvage option. Management should be individualized, and further high-quality studies are needed to refine treatment algorithms.