Background: Metabolic dysfunction-associated steatotic liver disease (MASLD) affects over 25% of the U.S. adult population. MASLD can progress to fibrosis, cirrhosis and hepatocellular carcinoma if left untreated. Therefore, MASLD fibrosis screening is encouraged for at risk populations. Prior studies have shown that noninvasive testing with Vibration Control Transient Elastography (VCTE-FibroScan) is beneficial however, less accurate in obese subjects. While it is known that an elevated BMI may overestimate fibrosis on VCTE, there remains limited information regarding the BMI and discordant number of stages when compared to liver biopsies.
Methods: We performed a retrospective analysis of 245 MASLD adult patients who underwent FibroScan evaluation and a liver biopsy. Medical charts were manually reviewed for demographics, BMI, metabolic co-morbidities, CAP and LSM values. Inclusion criteria had to CAP score of 250+, 10 Liver stiffness measurements (LSM) values with IQR/med <30%, and a liver biopsy with a minimum of 10 portal tracts. FibroScan was completed with either M or XL probe depending on BMI and machine suggestions. FibroScan stages of fibrosis were defined by liver stiffness measures (kPa) < 5 (stage 0), 5-7.99 (stage 1) 8-9.99 (stage 2), 10-13.99 (stage 3) and 14+ (stage 4). BMI was determined within number of discordant stages between FibroScan and biopsy. Data was entered into SPSS v25.
Results: Mean participant age 51±13yo, BMI 39.6±8 kg/m², white 88%, female 69%, diabetic 54%, hypertensive 61% and dyslipidemic 54%. Average ALT and AST 58±46 IU/L and 47±34 IU/L, respectively. 89% of subjects were obese (class 1- 19%, class 2- 23%, class 3- 47%). Advanced fibrosis was noted in 45% of our cohort. The stage of fibrosis by FibroScan and liver biopsy were identical in 37% of patients. Discordant stages 1, 2, 3 and 4 were 32%, 14%, 11%, and 6%, respectively. Figure 1 details BMI average±95%CI for each discordant stage and figure 2 compares BMI with 2 or more stage discrepancies to none or 1 stage. 39% of subjects with BMI ≥40 kg/m² had 2-4 discordant stages compared BMI <40 kg/m²with 0-1 in 21% (P=0.003); Odds Ratio 2.12 (95% CI: 1.21-3.69), P=0.008. In 13 subjects that FibroScan overestimated liver fibrosis by 4 stages, all but 1 subject had BMI 40+. Interestingly, the prevalence of advanced fibrosis was lower in Class 3 (37%) compared to Class 0-2 (51%, P=0.04). Age and metabolic co-morbidities were not associated with discordant stages of fibrosis.
Conclusions: Average BMI increases with stages of discordance by FibroScan. Class 3 obese individuals are more likely to have a greater number of discordant stages. FibroScan’s over-estimation of fibrosis in Class 3 obese individuals supports a second non-invasive test, such as ELF, before considering a liver biopsy.

Gary Abrams

Appearances