Liver enzymes, body mass index, and the risk of diabetes

Abstract

INTRODUCTION Prevalence and incidence of diabetes have increased alarmingly worldwide and across all age, sex, and race groups. Obesity is an important risk factor for diabetes. Meanwhile, there have been several reports about the association between liver enzyme levels and diabetes. However, only few studies investigated the association of liver enzymes, obesity, and the incident diabetes. This study was performed to investigate whether the relationship between body mass index (BMI) and the incident diabetes is modified by ALT or AST levels.

METHODS This study used data from the Korean Genome Epidemiology Study_cardiovascular disease association study (KoGES_CAVAS) which enrolled 10,615 participants aged 40 years or older from rural communities between 2005 and 2011. I carried out a prospective analysis of 6,484 participants (2,497 men and 3,987 women) who completed follow-up examinations until 2014. Serum ALT and AST were measured using the enzymatic methods. BMI was analyzed as both a continuous and categorized variable. Diabetes was defined as fasting blood glucose ≥ 126 mg/dL or current treatment for diabetes. To examine the associations of BMI with the incident diabetes according to ALT or AST levels, multiple logistic regression models were used after stratification into the low and high groups based on the median ALT or AST levels.

RESULTS The median follow-up time was 4.5 years, during which 304 participants (4.7%) developed diabetes. In people with high ALT levels, compared with the first BMI quartile, the adjusted odds ratios (ORs) for incident diabetes of the second, third, and fourth BMI quartiles were 1.88 (95% CI, 0.98 – 3.62), 2.24 (1.21 – 4.13), and 3.32 (1.84 – 5.99), respectively (p-trend < 0.001) after adjustment for age, sex, study year, residential area, hypertension, family history of diabetes, smoking status, alcohol intake, exercise, fasting glucose, and ALT, while in people with low ALT levels, high BMI was not independently associated with the incident diabetes. Similarly, in people with high AST levels, the adjusted ORs for incident diabetes of the second, third, and fourth BMI quartiles, were 2.14 (1.13 – 4.08), 2.44 (1.32 – 4.53), and 3.65 (2.02 – 6.60), respectively (p-trend < 0.001), while in people with low AST levels, categorized BMI was not significantly associated with the incident diabetes but the adjusted OR per 1 kg/m2 increase in BMI was 1.10 (1.02 – 1.19).

CONCLUSION These findings suggest that there was a strong association between BMI and incident diabetes among people with high liver enzyme levels, but the association was only modest among those with low liver enzyme levels.