Increased Hemoglobin A1c Correlates with Increased Liver Stiffness in a Mexican-American Population

Introduction: Liver stiffness measured by transient elastography (TE) predicts liver fibrosis progression and risk of hepatocellular carcinoma. Studies have shown associations between liver stiffness and diabetes, insulin resistance, and obesity, independent of steatosis. In this study, we aimed to determine whether glucose control (hemoglobin A1c) was associated with liver stiffness in a sample of Mexican Americans with high rates of liver disease in south Texas.

Methods: Liver stiffness was measured in 159 participants in the community-based Cameron County Hispanic Cohort study in Brownsville, Texas. Univariable and multivariable linear regression was used to determine associations between HbA1c and log-transformed liver stiffness (log kPa) after adjusting for demographic and clinical covariates. Results are presented as Z-standardized β and p-values.

Results: One hundred fifty-nine participants underwent TE, and 6 were excluded due to unreliable results. Overall, 25% had diabetes and 44% were obese, with median age of 52 years. In univariable analysis, liver stiffness was associated with log HbA1c (β = 0.24, p = 0.0027), controlled attenuation parameter (CAP; β = 0.22, p = 0.0033), waist circumference (β = 0.30, p < 0.0001), and with log insulin (β = 0.16, p = 0.0574). In multivariable analysis, log HbA1c was robustly associated (β = 0.18, p = 0.0260) with liver stiffness after adjusting for age, sex, CAP, waist circumference, and insulin levels.

Discussion: This study presents evidence of a relationship between glucose control and liver stiffness, independent of steatosis, in high-risk Mexican Americans sampled from a population-based study in the U.S. Our results suggest that HbA1c may be an important tool for prevention and management of liver disease. Longitudinal follow-up studies are needed to determine if glucose control over time modulates risk of liver-related morbidities.