Endoscopic ultrasound-guided gastro-enteric anastomosis: A systematic review and meta-analysis

Endoscopic ultrasound-guided gastro-enteric anastomosis (EUS-GEA) using lumen-apposing metal stents (LAMS) is emerging as a minimally invasive alternative to surgery across several indications. Literature on this subject is heterogeneous, with variable reporting of techniques and outcomes. Our aim was to perform a meta-analysis of published data on EUS-GEA, providing a pooled estimate of technical and clinical outcomes.

The protocol was registered in PROSPERO (Reg. no. CRD42018111110). PubMed, Embase, Scopus, and Web of Science databases were searched until February 2019 for studies describing patients undergoing EUS guided enteric anastomosis. PRISMA methodology was used. Pooled technical and clinical success rates as well as pooled adverse events rates were calculated. Study quality, publication bias, and heterogeneity were explored.

Twelve studies including 290 patients were included, published between 2016 and 2019. All studies but one were retrospective. Main procedure indication was gastric outlet obstruction (62.4%), followed by ERCP access (27.9%) in patients with gastric bypass surgery. Direct puncture technique was the most frequently adopted (68.2%). Pooled technical success rate (12 studies, 290 patients) was 93.5% [95% confidence interval (CI) 89.7-6.0%; I²:0%], while clinical success rate (11 studies, 260 patients) was 90.1% [95%CI 85.5-93.4%; I²:0%]. Pooled total adverse events rate (11 cohorts, 261 patients) was 11.7% [95%CI 8.2-16.6%; I²:0%], mainly mild/moderate: 10.6% [95%CI 7 - 15.6%]. No publication bias or significant heterogeneity was found.

EUS-GEA has a high rate of technical and clinical success when performed in expert centers. The procedure appears to be relatively safe, and might represent a non-inferior minimally invasive alternative to surgery. The paucity of long-term clinical outcomes suggests prudency and need for further research, especially regarding non-malignant indications.