Association of controlled attenuation parameter (CAP) and HbA1c in patients with fatty liver

Background and aim: Chronic liver diseases and serum glycosylated hemoglobin (HbA1c) levels are linked to one another through the metabolic syndrome. Our aim now was to analyze for the first time the potential association between hepatic steatosis, as determined quantitatively with the controlled attenuation parameter (CAP), and HbA1c.

Patients and methods: At a tertiary referral center in Germany, we evaluated a group of 212 outpatients with hepatic steatosis retrospectively, of whom 93.4% presented with non-alcoholic liver disease. Hepatic steatosis was assessed non-invasively using controlled attenuation parameter (CAP), which quantifies the degree of ultrasound attenuation based on vibration-controlled transient elastography (Fibroscan). Serum HbA1c and liver function tests were measured with standardized clinical chemistry assays. The NAFLD susceptibility variant PNPLA3 p.I148 M was genotyped using Taqman assays.

Results: Overall in this cohort (113 men, median age 52 years), median CAP was 293 dB/m (100 – 400), and 171 (80.7%) patients presented with elevated CAp ≥238 dB/m, indicating marked hepatic steatosis. Median BMI was 30.2 kg/m2 (17.2 – 47.4), median HbA1c was 5.6% (3.7 – 10.4), and serum ALT activities were 45 U/l (9 – 301). The frequency of elevated CAP increased with higher serum HbA1c levels (rs = 0.230, P = 0.001). Patients with both hepatic steatosis and increased HbA1c levels (HbA1c ≥6.0%) displayed significantly (P = 0.001) higher CAP values as compared to those with normal levels (312 vs. 286 dB/m). As compared to non-diabetics, CAP values and HbA1c levels were higher in diabetics (322 vs. 282 dB/m and 6.8 vs. 5.3%, both P < 0.001). In our cohort, 104 patients (49.0%) carried at least one PNPLA3 p. 148 M risk allele. When stratifying for PNPLA3 genotype, the genetic association was maintained for carriers of the risk allele p. 148 M and normal levels of HbA1c (P < 0.001) but not for those with increased levels. Overall, the risk for hepatic steatosis was independently associated with HbA1c, BMI, ALT and age as determined by multivariate linear regression analysis (all p ≤0.013).

Conclusions: Non-invasive risk stratification and follow-up of fatty liver in patients with metabolic syndrome is needed because of potential progression to steatohepatitis. Steatosis as assessed by CAP is associated with HbA1c in non-diabetic individuals, and the combination of these non-invasive markers improves individual risk assessment of patients with chronic liver diseases.