Marked Variation in Hepatic Steatosis in Metabolic Dysfunction-Associated Steatotic Liver Disease Over 5 Years: A Community-Based Study Using Controlled Attenuation Parameter.

BACKGROUND: Metabolic dysfunction-associated steatotic liver disease (MASLD) is very common and associated with significant morbidity and mortality due to its potential progression to cirrhosis and liver cancer. Whereas hepatic fat accumulation is the key to MASLD progression, little is known regarding changes in liver fat content over time in the general population. We aimed to investigate changes in liver fat in a longitudinal study of the general population. METHODS: We conducted a longitudinal study involving 195 randomly selected individuals from the general population, evaluated at two time points, 5 years apart. Participants with hepatic steatosis at baseline did not receive any specific treatment. The primary objective was to assess changes in liver fat content, as estimated by controlled attenuation parameter (CAP) using transient elastography. We also examined the frequency of steatosis resolution and development. CAP variability was assessed in two measurements 7 days apart in a cross-sectional study of 101 volunteers, with a mean variability of 9.9%. Steatosis resolution was defined as reduction of CAP > 10% from baseline with a final value < 275 dB/m, while steatosis development was defined as increase in CAP > 10% with a final value ≥ 275 dB/m. RESULTS: Remarkable variations in liver fat content were observed. Among the 88 participants with steatosis at baseline, 34% had resolution of steatosis (CAP decreased from 300 to 237 dB/m; p < 0.001). Resolution was associated with weight loss and reductions in transaminases and gamma-glutamyl transferase levels. In contrast, 29% of the 107 participants without steatosis at baseline developed it during follow-up. Increase in liver stiffness measurement (≥ 8 kPa) was associated only with persistent hepatic steatosis but not with steatosis resolution. CONCLUSIONS: There is marked variation in liver fat content among participants from the general population over a period of 5 years, indicating its dynamic nature. These variations should be considered in epidemiological studies of MASLD.