Abstract
OBJECTIVE: To explore the interactions between smoking and CYP2A6 genotypes on type 2 diabetes (T2DM) as well as potential pathways for smoking in causing T2DM. DESIGN: Cross-sectional study. METHODS: A total of 1344 smokers with complete data from a community-based T2DM survey in Guangzhou and Zhuhai of China from July 2006 to June 2007 were interviewed with a structured questionnaire about socio-demographic status and daily cigarette consumption. Serum glucose, insulin, and cotinine were measured after an overnight fast. Subjects were genotyped for CYP2A6 and classified, according to genotype, into normal, intermediate, slow, or poor nicotine metabolizers based on prior knowledge of CYP2A6 allele associations with nicotine C-oxidation rate. Abdominal obesity was defined as a waist-to-hip ratio ≥0.90 for males or ≥0.85 for females. Type 2 diabetic patients (n=154) were diagnosed according to WHO 1999 criteria. Chi-square tests, multivariate logistic regression models, and a structural equation model were used in this study. RESULTS: Multivariate analysis indicated that, compared with light smoking, heavy smoking significantly increased the risk of T2DM (odds ratio (OR)=1.75, 95% CI=1.01-3.05). There were significant interactions between heavy smoking and slow CYP2A6 (OR=5.12, 95% CI=1.08-24.23) and poor CYP2A6 metabolizer genotypes (OR=8.54, 95% CI=1.28-57.02) on T2DM. Structural equation modeling indicated that CYP2A6 moderation of smoking quantity risk on T2DM was mediated by the effects on serum cotinine, abdominal obesity, insulin resistance, and insulin secretion. CONCLUSIONS: Heavy smoking was significantly associated with T2DM, and this association was moderated by CYP2A6 genotype and mediated by serum cotinine, abdominal obesity, insulin resistance, and insulin secretion.