Randomized Controlled Trial of Endoscopic Ultrasound‐Guided Liver Biopsy Versus Percutaneous Liver Biopsy for Obtaining Histologically Adequate Specimens

ABSTRACT Aim To compare endoscopic ultrasound‐guided liver biopsy (EUS‐LB) with percutaneous liver biopsy (P‐LB) in terms of histological adequacy, safety, and patient‐reported pain in parenchymal liver disease. Methods In this single‐center randomized controlled trial, 80 patients were assigned to either EUS‐LB ( n = 41) or P‐LB ( n = 39). The primary endpoint was histological adequacy, defined as total specimen length (TSL) ≥ 15 mm and ≥ 6 complete portal tracts (CPTs). The main secondary endpoints were the definitive histological diagnostic rate, specimen characteristics (CPTs, TSL, and longest specimen length), patient‐reported pain, and procedure‐related adverse events (AEs). EUS‐LB employed a 19‐gauge Franseen‐tip needle with wet suction under conscious sedation, whereas P‐LB used a 16‐gauge Tru‐Cut needle under local anesthesia. Results Histological adequacy based on predefined histologic criteria was achieved more frequently with EUS‐LB than with P‐LB (70.7% vs. 10.3%; difference, 60.4%; 95% CI 43.5–77.3). The definitive histological diagnostic rates were similar (92.7% vs. 97.4%). The median TSL was longer with EUS‐LB (20.0 vs. 12.8 mm), whereas ≥ 6 CPTs were more common with P‐LB (97.4% vs. 80.5%). AEs occurred in 2.4% versus 25.6%, driven in the P‐LB arm by procedural stress–related hemodynamic responses (transient grade III hypertension); patient‐reported pain favored EUS‐LB (no pain during the procedure, 95.1% vs. 30.8%). No serious AEs or bleeding events occurred, and none required prolonged hospitalization. Conclusions EUS‐LB met the noninferiority criterion for histologic adequacy and achieved a diagnostic performance comparable to that of P‐LB with lower patient‐reported pain. EUS‐LB may be a reasonable alternative when endoscopic expertise and sedation are available.