Incidence of arterial injuries detected by arteriography following percutaneous right-lobe ultrasound-guided core liver biopsies in human subjects

PURPOSE: To determine the incidence and significance of arterial injuries detected by angiography subsequent to ultrasound-guided random core liver biopsies in normal healthy adults. MATERIALS AND METHODS: Retrospective analysis of 55 potential living related liver donors who underwent an ultrasound-guided random liver biopsy and a visceral angiogram was performed (January, 1999 to May, 2002). All liver biopsy samples (obtained by 2-3 18-gauge needle passes) were re-evaluated prospectively by a transplant pathologist for adequacy (defined: ≥5 complete portal triads). Subjects who underwent angiograms before the biopsy or >7 days after the biopsy were excluded from the arterial injury evaluation. Angiograms were reviewed by two angiographers. Arterial injuries were identified and classified by consensus into contusions, active bleeding, arterial-venous fistulae, and pseudoaneurysms. RESULTS: Mean needle pass was 2.1. No major complications were encountered. All samples were deemed pathologically adequate. Forty-eight potential donors were included for the arterial injury evaluation. Three arterial injuries (two arterioportal fistulae, 4.2%) were found in 48 angiograms (6.3%). None of the three injuries required intervention. CONCLUSION: The incidence of arterioportal fistulae following core liver biopsies has not changed over the past three decades despite improvement in biopsy needle technology, reduction of needle caliber, and the use of image guidance.