OBJECTIVEA Meta-analysis was performed to assess the association of defective hepatic cytochrome P450 2A6 (CYP2A6) gene with smoking behaviors.

METHODSAll eligible studies published up to 2014 were searched out from PubMed, China National Knowledge Internet (CNKI), ISI Web of knowledge (ISI), vip citation databases (VIP), Chinese BioMedical Literature (CBM) and Elsevier Science Direct, searching words were "smok*","nicotine dependence","CYP2A6","cytochrome P450 2A6","polymorphism","mut*"and"varia*", while 436 articles were concluded. Meta-analysis was performed using Statal 3.1.

RESULTSA total of ten studies were finally included. We didn't find a significant effect of defective CYP2A6 gene on smoking initiation (fixed effect model (FEM): OR = 0.90, 95%CI: 0.78-1.03, I(2) = 25.8%), smoking persistence (random effect model (REM): OR = 0.85, 95%CI: 0.59-1.23, I(2) = 66.3%) and smoking cessation (REM: OR = 0.89, 95%CI: 0.57-1.40, I(2) = 67.1%). But it showed a significant protective effect of CYP2A6*4 on smoking initiation (FEM: OR = 0.78, 95%CI: 0.61-0.99, I(2) = 28.2%), smoking persistence (FEM: OR = 0.53, 95%CI: 0.36-0.77, I(2) = 41.0%) and smoking cessation (REM: OR = 0.49, 95%CI: 0.31-0.80, I(2) = 0.0%).

CONCLUSIONSThis Meta-analysis suggested that there was not a protective effect of defective CYP2A6 gene against smoking behaviors. But smokers with whole CYP2A6 gene deletion would be less likely to start smoking, less smoking persistence and more likely to quit smoking successful than smokers with wild CYP2A6 gene.