HOME
Return to Table of Contents
ADEQUACY OF LIVER BIOPSIES: VARIABILITY ACROSS DIFFERENT METHODS OF SPECIMEN ACQUISITION
R Abdalian, F Siadat, EJ Heathcote, M Guindi
Departments of Medicine & Pathology, University Health Network, Toronto
BACKGROUND: Histological assessment of the liver is considered the gold standard for the diagnosis and staging of disease. Transjugular liver biopsy (TJLB) or percutaneous liver biopsy performed either blind (BPLB) or ultrasound-guided (US) are the main modalities to accomplish this. From a diagnostic perspective, a minimum specimen length of 15 mm and at least 6 portal tracts (PT) are usually deemed adequate for confident histologic assessment.
AIM: To compare adequacy of liver biopsy specimens attained via three methods- percutaneous blind, percutaneous US-guided and transjugular.
METHODS: We evaluated 291 consecutive liver biopsy specimens performed either by the blind percutaneous route using an 15G Jamshidi needle (N=101) or by the US-guided route using an 18G automated cutting needle (N=108), or the transjugular route using an 18G Quickcore needle (N=82). Transplant and targeted mass biopsies were excluded. Differences in specimen adequacy as judged from sample length, width, fragmentation, as well as number of complete PTs (defined as the presence of >3/4 of PT circumference) were sought. REB approval was granted.
RESULTS: BPLB yielded significantly longer (median 23 (6-46) mm) and wider (median 0.96 (0.09-1.18) mm) specimens than US-guided percutaneous biopsies or TJLB (p=0.000 for both). BPLB also yielded a significantly higher number of complete PTs than TJLB but not US-guided biopsies (p=0.000 and p=0.126 respectively). When compared to TJLB, US-guided biopsies were longer (median 16 (4-45) mm) and had a higher number of complete PTs (p=0.031 and p=0.000 respectively). BPLB was more often judged adequate than US- guided biopsies and TJLB when applied to a more liberal (> or =15mm length AND > or =6 PTs) or stringent (> or = 20mm length AND > or =11 PTs) adequacy criteria (p=0.000 for both). Specimen fragmentation, indication for biopsy (diagnosis vs. staging), presence of cirrhosis or biliary disease did not impact sample adequacy by both criteria. Biopsies from the left liver lobe yielded thinner specimens (median 0.70 (0.56-0.85) mm) and were less likely to be adequate (p=0.042). No immediate complications were noted in the US-guided or TJLB groups, whereas one patient in the BPLB group reported pain and suffered a brief episode of hypotension.
CONCLUSION: BPLB yields more adequate specimens when compared to US-guided or TJLB. BPLB and US-guided biopsies may be comparable when performed for diagnosis alone but not for staging of hepatic fibrosis. Differences in technique and diagnostic awareness may account for the observed differences.