EUS vs. ERCP-guided primary drainage of inoperable distal malignant biliary obstruction: systematic review and meta-analysis of randomized controlled trials.

BACKGROUND AND AIMS: Several randomized controlled trials (RCTs) compared endoscopic ultrasound-guided biliary drainage (EUS-BD) to endoscopic retrograde cholangiopancreatography (ERCP) as first-line interventions in distal malignant biliary obstruction (DMBO). We assessed the efficacy and safety of these two approaches. METHODS: A PubMed/Medline, Embase and Cochrane databases bibliographic search until 01/12/2023 was performed to identify RCTs comparing EUS-BD to ERCP for primary biliary drainage in inoperable patients with DMBO. Primary outcome was technical success. Secondary outcomes were clinical success, adverse events (AEs), mean procedure time, 1-year stent patency, and overall survival. Relative risk (RR) with 95% confidence interval (CI) were calculated using random-effect model. RESULTS: Five studies (519 patients) were included. The RR for pooled technical success in EUS-BD was 1.06 ([0.96-1.17]; P=0.27) and 1.02 [0.97-1.08]; P=0.45) for clinical success. 1-year stent patency was similar among the two groups (RR 1.15; [0.94-1.42], P=0.17), with lower reintervention in the EUS-BD group (RR 0.58; [0.37-0.9]; P=0.01). The RR of AEs rate was 0.85 [0.49-1.46]; P=0.55) and severe AEs of 0.97 [0.10-0.17]; P=0.98). On subgroup analysis, EUS-lumen apposing metal stents (LAMS) outperformed ERCP in term of technical success (RR 1.17; [1.01-1.35]; P=0.03). Procedure time was lower in EUS-BD (standardized mean difference -2.36 minutes; [-2.68 to -2.05]; P<0.001). CONCLUSIONS: EUS-BD showed a statistically significant lower re-intervention rate compared to ERCP, but with similar technical success rate, stent patency, clinical success rate and safety profile, while in the subgroup of EUS-LAMS, the technical success was better than ERCP Keywords: distal; biliary obstruction; efficacy; safety.

Endoscopic ultrasound-guided fine-needle biopsy with or without macroscopic on-site evaluation: a randomized controlled noninferiority trial.

BACKGROUND : The advantage of using the macroscopic on-site evaluation (MOSE) technique during endoscopic ultrasound-guided fine-needle biopsy (EUS-FNB) performed with 22G Franseen needles has not been investigated. We aimed to compare EUS-FNB with MOSE vs. EUS-FNB performed with three needle passes. METHODS : This randomized trial involved 10 Italian referral centers. Consecutive patients referred for EUS-FNB of pancreatic or nonpancreatic solid lesions were included in the study and randomized to the two groups. MOSE was performed by gross visualization of the collected material by the endoscopists and considered adequate when a white/yellowish aggregate core longer than 10 mm was retrieved. The primary outcome was diagnostic accuracy. Secondary outcomes were specimen adequacy, number of needle passes, and safety. RESULTS : 370 patients with 234 pancreatic lesions (63.2 %) and 136 nonpancreatic lesions (36.8 %) were randomized (190 EUS-FNB with MOSE and 180 with standard EUS-FNB). No statistically significant differences were found between EUS-FNB with MOSE and conventional EUS-FNB in terms of diagnostic accuracy (90.0 % [95 %CI 84.8 %-93.9 %] vs. 87.8 % [95 %CI 82.1 %-92.2 %]; P = 0.49), sample adequacy (93.1 % [95 %CI 88.6 %-96.3 %] vs. 95.5 % [95 %CI 91.4 %-98 %]; P = 0.31), and rate of adverse events (2.6 % vs. 1.1 %; P = 0.28). The median number of passes was significantly lower in the EUS-FNB with MOSE group (1 vs. 3; P < 0.001). CONCLUSIONS : The accuracy of EUS-FNB with MOSE is noninferior to that of EUS-FNB with three needle passes. MOSE reliably assesses sample adequacy and reduces the number of needle passes required to obtain the diagnosis with a 22G Franseen needle.