Endoscopic ultrasound-guided parenchymal liver biopsy: a systematic review and meta-analysis.

BACKGROUND Endoscopic ultrasonography (EUS)-guided liver biopsy is a novel technique to obtain adequate liver samples for diagnosis of liver parenchymal diseases. There are studies that have evaluated the feasibility and safety of EUS-guided parenchymal liver biopsy (EUS-LB), however, factors that can influence specimen quality are yet to be determined. Our aim was to determine the diagnostic accuracy of EUS-LB and evaluate factors associated with specimen quality. METHODS We performed a detailed search of PubMed/MEDLINE and Web of Science™ databases to identify studies in which results of EUS-guided liver parenchymal biopsies were reported published up to July 2020. A random effects model was used to estimate pooled values (mean ± SE) for total specimen length (TSL) and complete portal tracts (CPT). Subgroup analyses were applied to find out the procedural factors associated with better specimen quality using Cochran's Q test. A total of 10 meta-analyses were done focusing on international studies. Total of 1326 patients who underwent EUS-LB. EUS-LBs performed for suspicion of parenchymal liver disease. Pooled mean values for TSL and CPT with subgroup analyses. RESULTS Twenty-three studies with a total of 1326 patients were included in our meta-analysis. Overall pooled mean TSL and CPT were 45.3 ± 4.6 mm and 15.8 ± 1.5, respectively. In subgroup analysis, core biopsy needles proved to better in terms of CPT than fine-needle aspiration needles (18.4 vs 10.99, p = 0.003). FNB with slow-pull or suction technique provided a similar TSL (44.3 vs 53.9 mm, p = 0.40), however, slow-pull technique was better in terms of CPT (30 vs 14.6, p < 0.001). Heterogeneity was present among the studies. Another limitation is the low number randomized control trials. CONCLUSION EUS-guided parenchymal liver biopsy is a good alternative to other methods of liver sampling. Using FNB needles with a slow-pull technique can provide better results.