Butyrylcholinesterase is synthesized in the liver. The serum butyrylcholinesterase level has been cross-sectionally reported to be higher in patients with diabetes, hyperlipidaemia, obesity and fatty liver than in those without them. It is not known whether serum butyrylcholinesterase is associated with the risk of future type 2 diabetes.


A prospective cohort study.


A total of 8470 Japanese men aged 40–55 years without type 2 diabetes at baseline.


Type 2 diabetes was diagnosed if a fasting plasma glucose (FPG) level was ≥7·0 mmol/l, if a HbA1c level was ≥6·5% or if participants were taking oral hypoglycaemic medication or insulin.


During the 42 227 person-years of follow-up, 868 cases had developed type 2 diabetes. Serum butyrylcholinesterase was significantly positively correlated with body mass index (BMI), FPG, alanine aminotransferase (ALT), γ-glutamyltransferase (GGT) and triglycerides (TG), whereas negatively with high-density lipoprotein (HDL) cholesterol. In Cox proportional hazards models, after adjusting for age, BMI, FPG, alcohol consumption, smoking habit, walk to work, regular leisure-time physical activity and family history of diabetes, the highest quartile (398–806 IU/l) of serum butyrylcholinesterase increased the risk of type 2 diabetes compared with the lowest quartile (56–311 IU/l) [hazard ratio (HR) 1·41 (95% confidence interval (CI), 1·14–1·74)]. After further adjusting for ALT and GGT, this association remained [HR 1·40 (95% CI, 1·13–1·73)]. Furthermore, this association was significant independent of TG and HDL cholesterol.


Elevated serum butyrylcholinesterase was independently associated with an increased risk of future type 2 diabetes.