Introduction: Endoscopic ultrasound-guided gallbladder (EUS-GBD) drainage has been described as an alternative palliative treatment for malignant biliary obstruction (MBO). We aim to assess the outcomes of EUS-GBD for MBO.
Methods: We conducted a comprehensive literature review of MEDLINE, EMBASE, Cochrane, and Scopus databases for studies published in the English language that addressed outcomes of EUS-GBD for MBO through December 2021. Technical success was defined as successful stent deployment. Clinical success was defined as the resolution of the indication to proceed with EUS-GBD which includes improvement of jaundice or significant improvement of bilirubin. Reintervention was defined as the need for reintervention after achieving clinical success. Immediate adverse events (AE) were defined as complications that occurred intra-procedural till the first 24 hours after the procedure. Delayed AE were defined as complications that occurred after the first 24 hours of doing the EUS-GBD drainage. Pooled estimates were calculated following the restricted maximum likelihood method using random effects model. We assessed heterogeneity using the I2 statistic.
Results: After excluding duplicates, 1597 articles were screened with 14 unique articles included in the meta-analysis (Fig 1A). A total of 143 patients were included (Table). All studies had reported the technical success. The pooled technical success was 91.66% [95% confidence interval (CI) 83.13 - 96.08%]. Thirteen articles (139 patients) reported the clinical success and immediate AE. The pooled clinical success rate was 82.32% (95% CI 74.90-87.89%). The pooled immediate AE rates were 8.45% (95% CI 4.38-15.68) (Fig 1B). 9 articles (124 patients) reported delayed AE. The delayed AE reported were food impaction in the stent (n=3), stent migration &/or dysfunction (n=3), bleeding (n=2), cholangitis (n=1), peritonitis (n=1), and unknown AE (n=3). The pooled delayed AE rates were 13.81% (95% CI 8.45-21.76%) (Fig 1C). Nine articles (82 patients) reported intervention rates. The pooled reintervention rates after achieving clinical success were 15.71% (95% CI 9.20-25.51%). I2=0 for all meta-analyses.
Discussion: This is the first meta-analysis to evaluate outcomes of EUS-GBD drainage in MBO. We report a high technical and clinical success come up with a relatively low reintervention and AE rates. There was no heterogeneity in our data. EUS-GBD drainage is a feasible palliative option in MBO in experienced centers.
