Treatment of anastomotic leakage following Ivor Lewis esophagectomy—10 year experience from a Nordic center

TL;DR

This study examines the treatment outcomes of anastomotic leakage after a specific type of esophagectomy, finding that most cases can be managed endoscopically with good results.

Contribution

The paper presents a 10-year clinical experience from a Nordic center on managing anastomotic leakage with various endoscopic strategies.

Findings

Most anastomotic leakages healed with endoscopic treatment, preserving the anastomosis.

Airway fistulas occurred in 15% of patients, associated with leak severity.

EVT + stent treatment resulted in longer healing times compared to other methods.

Abstract

Anastomotic leakage (AL) is a dreaded complication following esophageal resection. No clear consensus exist for the optimal handling of this severe complication. The aim of this study was to describe the treatment outcome following AL. We conducted a retrospective cross-sectional study including all patients with AL operated with Ivor Lewis esophagectomy from 2010 to 2021 at Oslo University Hospital, Norway. 74/526 (14%) patients had AL. Patient outcomes were analyzed and categorized according to main AL treatment strategy; stent (54%), endoscopic vacuum therapy and stent (EVT + stent) (19%), nasogastric tube and antibiotics (conservative) (16%), EVT (8%) and by other endoscopic means (other) (3%). One patient had surgical debridement of the chest cavity. In 66 patients (89%), the perforation healed after median 27 (range: 4–174) days. Airway fistulation was observed in 11 patients (15%). Leak severity (ECCG) was associated with development of airway fistula (P = 0.03). The median hospital and intensive care unit stays were 30 (range: 12–285) and 9 (range: 0–60) days. The 90-days mortality among patients with AL was 5% and at follow up, 13% of all deaths were related to AL. AL closure rates were comparable across the groups, but longer in the EVT + stent group (55 days vs. 29.5 days, P = 0.04). Thirty-two percent developed a symptomatic anastomotic stricture within 12 months. Conclusion: The majority of AL can be treated endoscopically with preservation of the conduit and the anastomosis. We observed a high number of AL-associated airway fistulas.

Figures