Comparison of efficacy and safety of endoscopic and radiological interventions for gastric varices: A systematic review and network meta-analysis.

Aim of the study: There is a paradigm shift in the management of gastric varices with the availability of endoscopic ultrasound and radiologic interventions. The optimal choice of intervention remains a dilemma for most treating physicians. Material and methods: We searched MEDLINE, the Cochrane Central Register of Controlled Trials, and ScienceDirect for studies comparing endoscopic glue injection, endoscopic thrombin injection (THB), variceal band ligation, EUS-guided coiling, EUS-guided glue injection, EUS-guided coiling with glue (EUS-C+G), balloon occluded retrograde transvenous obliteration (BRTO), and transjugular intrahepatic portosystemic shunt (TIPS) for gastric varices in adults. The data on four outcomes - obliteration of varices, rebleeding, adverse effects, and mortality - were pooled using a random-effects model. Treatment estimates were calculated as odds ratios (ORs) along with their 95% confidence interval (CI). The relative ranking of interventions for various outcomes was calculated as their surface under the cumulative ranking curve (SUCRA). Results: We identified 34 studies (10 randomized controlled trials, 24 non-randomized trials) with 2783 patients. Based on SUCRA plots, BRTO (SUCRA 95.1) had the highest rate of variceal obliteration followed by EUS-C+G (SUCRA 80.9). The risk of rebleeding was lowest with BRTO (SUCRA 85.1) followed by EUS-C+G (SUCRA 78.8). Moderate-severe adverse effects were least likely with THB (SUCRA 92.5) and highest with TIPS (SUCRA 3.7). In terms of mortality, EUS-C+G (73.5) had the lowest probability of overall mortality followed by TIPS (69.1). Conclusions: In this network meta-analysis, we found BRTO and EUS-guided therapies to be superior to endoscopic glue injection. However, the level of evidence remains low.