2 - Associations Between Histopathology, Transient Elastography, and Magnetic Resonance Measures of Steatosis and Fibrosis Among a Pediatric Cohort Undergoing Liver Biopsy

Poster Number: 2
Publication Number: 2.207

Joseph A. Cornett, Columbia University Vagelos College of Physicians and Surgeons, Morrisville, NC, United States; Ivette Partida, Columbia University Vagelos College of Physicians and Surgeons, New York, NY, United States; Ngoc Q. Duong, Columbia University Vagelos College of Physicians and Surgeons, New York, NY, United States; Qin Wang, Columbia University Irving medical Center, New York, NY, United States; WEI SHEN, Columbia University, New York, NY, United States; Jeff Goldsmith, Columbia University Mailman School of Public Health, New York, NY, United States; Helen Remotti, Columbia University Vagelos College of Physicians and Surgeons, New York, NY, United States; Christine K. Lee, Boston Children's Hospital, Boston, MA, United States; Jennifer Woo Baidal, Columbia University Irving Medical Center, New York, NY, United States

Background: Vibration-controlled transient elastography (VTE) and magnetic resonance (MR) are promising non-invasive methods to grade and stage hepatic steatosis and fibrosis, but associations between these measures in pediatric patients are less known.

Objective: Pilot study to examine associations between non-invasive measures of steatosis and fibrosis and with the gold standard of liver biopsy histopathology.

Design/Methods: We sequentially recruited pediatric patients age < 22 years undergoing liver biopsy for clinical evaluation of suspected chronic liver disease in a cross-sectional study. Trained study staff performed VTE and MR imaging (MRI), spectroscopy (MRS), and elastography (MRE). For hepatic steatosis, we compared histopathology grade (S0-S3), VTE controlled attenuation parameter (CAP) in decibels per minute (dB/m), and MRS and MRI in fat fraction (% fat). For hepatic fibrosis, we compared histopathology METAVIR score (F0-F4), VTE liver stiffness (LS) in kilopascals (kPa), and MRE LS (kPA). We used Pearson’s correlation and linear regression models adjusted for age, sex, and race/ethnicity in these comparisons.

Results: Among 16 pediatric patients, mean age was 13 years (SD 4), 38% were female, 50% reported Hispanic/Latino ethnicity, and 50% had body mass index ≥ 95th percentile for age and sex. For steatosis on histopathology, 7 (50%) had S0, 2 (14%) had S1, 3 (21%) had S2, and 2 (14%) had S3. Correlation between CAP and MRS was 0.65 (95% CI: 0.12, 0.89) and between CAP and MRI was 0.66 (0.23, 0.88). Patients with steatosis on histopathology had higher CAP (adjusted mean difference 96 dB/m [95% CI: 8.8, 183.2]), MRS (25% fat [9%, 42%]), and MRI (20% fat [6%, 35%]) measures than those without steatosis. For fibrosis on histopathology, 9 (64%) had F0, 3 (21%) had F1, 2 (14%) had F2, and 0 had F3 or F4. Correlation between VTE LS and MRE LS was 0.48 (-0.1, 0.82). Patients with fibrosis on histopathology had higher MRE LS (adjusted mean difference 0.9 kPA [0.27, 1.5]) but not VTE LS (1.05 kPa [-1.4, 3.5]) than those without fibrosis.Conclusion(s): In this pilot pediatric study, non-invasive measures of hepatic steatosis and fibrosis in children were feasible. Our findings suggest VTE may be useful for non-invasive measurement of steatosis but may not be as precise as MRE measures for lower-stage fibrosis. Larger, longitudinal pediatric studies using noninvasive measures toward understanding and preventing steatosis are warranted.