Quality of Tissue Samples Obtained by Endoscopic Ultrasound-Guided Liver Biopsy: A Randomized, Controlled Clinical Trial

INTRODUCTION:Liver biopsy (LB) remains essential for the diagnosis and staging of parenchymal liver diseases. Endoscopic ultrasound-guided LB (EUS-LB) has emerged as an attractive alternative to percutaneous and transjugular routes. We aimed at comparing the adequacy of samples obtained by EUS-LB with percutaneous LB. METHODS:A single-center, randomized, controlled clinical trial was designed. Patients undergoing LB were randomly assigned to EUS-LB or percutaneous LB groups. EUS-LB was performed with a 19-gauge Franseen core needle through a transduodenal and transgastric route. Percutaneous LB was performed with a 16-gauge Tru-Cut needle. The main outcome was the percentage of adequate samples obtained. Secondary outcomes were the percentage of accurate histologic diagnosis, number of complete portal tracts (CPT), total and longest specimen length (TSL and LSL), sample fragmentation, adverse events, and patients' satisfaction. An adequate specimen was defined as TSL >= 20 mm and including >= 11 CPT. RESULTS:Ninety patients were randomized (44 to EUS-LB and 46 to percutaneous LB) and included in the analysis. The percentage of adequate tissue samples was 32.6% and 70.4% for percutaneous LB and EUS-LB, respectively (P < 0.001). A final histologic diagnosis was provided in all cases but one. TSL was longer after EUS-LB (23.5 vs 17.5 mm, P = 0.01), whereas the number of CPT was similar in both groups. Sample fragmentation occurred more often after EUS-LB (P < 0.001). No differences in adverse events were found. Satisfaction reported with both procedures was high. DISCUSSION:EUS-LB is safe and accurate and may be considered an alternative to percutaneous LB for the evaluation of parenchymal liver diseases.