BACKGROUND In patients with diabetes, studies investigating the association between nonalcoholic fatty liver disease (NAFLD) and coronary artery calcium score (CACS) have shown conflicting results. The aim of this study was to evaluate the association between NAFLD and CACS in diabetic patients.

METHODS This is the cohort study performed in Seoul National University Hospital Gangnam Healthcare Center. NAFLD was defined as cases with the typical ultrasonographic findings without excessive alcohol consumption, medications causing hepatic steatosis or other chronic liver diseases. CACS was evaluated using the Agatston method. Diabetes was defined as cases with fasting serum glucose ≥ 126 mg/dl, glycated hemoglobin (HbA1c) ≥ 6.5%, or those taking anti-diabetic medications. Multivariate linear regression analyses were performed with use of the interaction term of NAFLD × glycemic level and other confounders of CACS such as age, sex, hypertension, body mass index, waist circumference, HDL cholesterol and triglyceride.

RESULTS A total of 213 participants with diabetes were included in the study. As 77 subjects (36.2%) had CACS 0, causing left sided skewness, CACS was analyzed after log transformation to Ln (CACS + 1). A statistically significant interaction was observed between NAFLD and HbA1c ≥ 7% (P for interaction = 0.014). While NAFLD was not associated with CACS in the group with HbA1c < 7% (P = 0.229), it was significantly associated in the group with HbA1c ≥ 7% (P = 0.010) after adjusting for covariates in multivariate analyses.

CONCLUSIONS This study demonstrated an effect modification of glycemic level on the association between NAFLD and CACS. NAFLD was independently associated with CACS only in diabetes patients with higher HbA1c, after adjustment for confounders.