Comparative efficacy and safety of endoscopic treatments for gastric varices: a systematic review and network meta-analysis

Background The best endoscopic treatment for gastric varices is currently unclear. We performed a network meta-analysis combining direct and indirect comparisons among the different techniques currently in use. Methods We identified 10 randomized controlled trials (RCTs; 760 patients) comparing endoscopic cyanoacrylate injection, endoscopic thrombin injection, endoscopic ultrasound (EUS)-guided cyanoacrylate injection, EUS-guided coil injection, EUS-guided coil+cyanoacrylate injection, and large-volume band ligation. The primary outcome was rebleeding rate, and secondary outcomes were complete obliteration of gastric varices and adverse event rate. The results were expressed in terms of risk ratio (RR) with 95%CIs. Results EUS-guided cyanoacrylate and EUS-guided coil+cyanoacrylate injection showed significantly lower rates of rebleeding than direct endoscopic cyanoacrylate injection (RR 0.39, 95%CI 0.27-0.58 and RR 0.15, 95%CI 0.03-0.90, respectively). None of the treatments were significantly superior in terms of complete obliteration of gastric varices but EUS-guided cyanoacrylate injection, large-volume band ligation, and thrombin injection had lower adverse event rates compared with endoscopic cyanoacrylate injection (RR 0.59, 95%CI 0.48-0.72; RR 0.38, 95%CI 0.19-0.75; and RR 0.30, 95%CI 0.15-0.61, respectively). Conclusions EUS-guided cyanoacrylate injection with or without coils was associated with lower rebleeding rates than conventional endoscopic cyanoacrylate injection. However, the quality of evidence was very low and further RCTs are needed before firm conclusions can be drawn regarding the relative efficacy and safety of these treatments.