Association between Liver Biopsy Length and the Diagnosis of Non-Alcoholic Steatohepatitis
Michael R Peterson, Cynthia E Behling, Niraj S Patel, Amber N Mackey, Carolyn S Hansen, Rohit Loomba. UC San Diego Health System, San Diego, CA; Sharp Memorial Hospital, San Diego, CA
Background: Adequacy assessment is a critical part of the evaluation of liver biopsies performed for non-focal liver diseases. A small number of studies using chronic viral hepatitis as a model have examined the minimum amount of liver that must be examined to ensure an accurate diagnosis. No study has yet examined adequacy criteria in NASH, a disease in which the patterns of liver injury and fibrosis vary from chronic viral hepatitis. In this study, we examined the association between biopsy length and the diagnosis of steatohepatitis.
Design: Thirty 18G liver biopsies of at least 2.5 cm length performed for clinical suspicion of NASH were identified. Removable tape was used to mask the biopsies (1 H&E and 1 trichrome) so that 0.5 cm increments were added to each review, resulting in biopsies of 0.5 cm, 1.0 cm, 1.5 cm, and 2.0 cm lengths, and full length. The biopsies were randomized before each length was scored and reviewed without knowledge of prior score. Biopsies were scored as “not steatohepatitis”, “steatosis with borderline features of steatohepatitis”, and “definitely steatohepatitis” using the NASH-CRN system.
Results: There was an association between biopsy length and the probability that a biopsy would be diagnosed as “definitely steatohepatitis.” 55% of the full-length biopsies were diagnosed as “definitely steatohepatitis”, as compared to 37% of 0.5 cm, 42% of 1.0 cm, 45% of 1.5 cm, and 47% of 2.0 cm biopsies. This difference was statistically significant when compared to the full-length biopsy for the 0.5 and 1.0 cm biopsies (p=0.004 and p=0.037, respectively). When the diagnoses of the shortened biopsies were compared to the full-length, 48% showed no discordant diagnoses at any length (i.e. the same diagnosis was rendered at every biopsy length). 27% of the 0.5 cm biopsies, 38% of the 1.0 cm biopsies, 28% of the 1.5 cm biopsies, and 18% of the 2.0 cm biopsies had diagnoses that differed from the full-length biopsy.
Conclusions: Short liver biopsies (0.5 and 1.0 cm) are more likely to be diagnosed as “not steatohepatitis” or “steatosis with borderline features of steatohepatitis” relative to biopsies of 1.5 cm length or longer. 2.0 cm biopsy length resulted in the fewest discordant diagnoses as compared to the full-length biopsy. We recommend that biopsies performed for clinical suspicion of NASH should be of at least 2.0 cm length. Future analyses will focus on the effects of biopsy length on individual NAS components and fibrosis.
Tuesday, March 5, 2013 1:00 PM
Poster Session IV # 145, Tuesday Afternoon