Association between anti

We examined the risk of myocardial infarction associated with glucose-lowering therapy among diabetes patients with and without obstructive coronary artery disease. A cohort of patients with type 1 or type 2 diabetes (n？=？12,030), who underwent coronary angiography from 2004 to 2012, were stratified by presence of obstructive (any stenosis ？50%) coronary artery disease and by type of diabetes treatment: diet, non-insulin treatment and insulin (±oral anti-diabetics). The primary endpoint was myocardial infarction. Adjusted hazard ratios were calculated using diet-treated patients without coronary artery disease as reference. In patients without coronary artery disease, risk of myocardial infarction was similar in patients treated with non-insulin medication (adjusted hazard ratio 0.70, 95% confidence interval 0.27–1.81) and insulin (adjusted hazard ratio 0.76, 95% confidence interval 0.27–2.08) as compared to diet only. In patients with coronary artery disease, the risk of myocardial infarction was higher than in the reference group and an incremental risk was observed being lowest in patients treated with diet (adjusted hazard ratio 3.79, 95% confidence interval 1.61–8.88), followed by non-insulin medication (adjusted hazard ratio 5.42, 95% confidence interval 2.40–12.22), and highest in insulin-treated patients (adjusted hazard ratio 7.91, 95% confidence interval 3.51–17.82). The presence of obstructive coronary artery disease defines the risk of myocardial infarction in diabetes patients. Glucose-lowering therapy, in particular insulin, was associated with risk of myocardial infarction only in the presence of coronary artery disease