EUS-guided biopsy using 19-gauge Franseen needle versus percutaneous ultrasound-guided biopsy using 18-gauge full-core cutting needle for parenchymal liver disease: A randomized controlled, non-inferiority trial
In brief
EUS-guided liver biopsy achieves adequate tissue in 98% vs 64% percutaneous
In a randomized trial of 90 patients, EUS-guided biopsy met the predefined adequacy criteria in 98% of cases, far surpassing the 64% success rate of standard percutaneous biopsy and delivering longer cores and more portal tracts. Minor adverse events were more common with the endoscopic approach, but pain was low and patient satisfaction remained high.
- Journal
- Gastrointestinal endoscopy (Q1)
- Published
- 11 August 2026
- Study design
- Randomized controlled trial
- Evidence level
- Level 1, High (CEBM 1b)
- Authors
- Ishank Johri, Vikram Bhatia, Amar Mukund, Archana Rastogi, Chaggan Bihari, Srajit Singh, et al.
- PMID
- 42580638
- DOI
- 10.1016/j.gie.2026.07.046
Why clinicians should know about it
- Picked for Radiology, Radiation Oncology, Nuclear Medicine, Medical Physics and Imaging (top studies of the week, 16 August 2026): Randomized trial comparing EUS‑guided vs percutaneous liver biopsy
Abstract
BACKGROUND AND AIMS: There is limited data comparing endoscopic ultrasound-guided liver biopsy (EUS-LB) with percutaneous ultrasound-guided liver biopsy (PC-LB) regarding tissue adequacy, histologic yield, adverse events, and patient satisfaction. METHODS: In this single-center, parallel-group, non-inferiority trial, we randomized participants 1:1 to EUS-LB (19-G Franseen-tip biopsy needle; modified wet-heparin suction; left and/or right lobe) or PC-LB (18-G BioPince™ full-core needle; right lobe). The primary endpoint was specimen adequacy, defined as total specimen length (TSL) ≥15 mm and complete portal tracts (CPT) ≥8. Non-inferiority was assessed using absolute risk difference with a prespecified margin of -10%. RESULTS: Ninety participants were randomized. In intention-to-treat analysis, the primary endpoint was achieved in 44/45(97.8%) of EUS-LB and 29/45(64.4%) of PC-LB group (risk difference +33.3%; 95%CI: +17.9% to +48.1%), meeting non-inferiority and demonstrating superiority (p<0.001). With EUS-LB and PC-LB, the median TSL was 6.9cm versus 2.0cm and the median CPT number was 42 versus 9, respectively (both p<0.001). Even a single pass of EUS-LB outperformed PC-LB, with respect to sample adequacy, TSL, total CPT, and biopsy area (all p<0.001). PC-LB specimens were less fragmented (median 1 vs 13; p<0.001) and had greater median width (747.8μm vs. 667.2μm; p<0.001). Adverse events occurred more frequently with EUS-LB (26.7% vs. 4.4%; p=0.007); all but one were minor. Post-procedural pain was minimal, and satisfaction ratings were high in both groups. CONCLUSIONS: EUS-LB was non-inferior and superior to PC-LB for the prespecified adequacy endpoint, yielding substantially greater specimen length and CPT counts, with more frequent but predominantly minor adverse events. TRIAL REGISTRATION: CTRI/2023/10/058529; NCT06047327.
Abstract as published, via PubMed.
For healthcare professionals. The summary is generated by AI from the published abstract, and the evidence level is assigned automatically from the study design on the Oxford CEBM hierarchy. Neither is medical advice. Read the full paper before changing practice.