AIM: To evaluate whether diabetic patients differ from non-diabetic patients when referred for coronary angiography regarding previous history, indication for and findings at coronary angiography, use of medication, exercise test results and mortality. METHODS: Data were prospectively collected on patients referred for consideration of coronary revascularization to seven of the eight public Swedish heart centers that performed approximately 92% of all bypass operations in Sweden in 1994. RESULTS: 2762 patients were included of whom 406 (15%) had a history of diabetes mellitus. There was no difference in age or sex in the two groups. Chronic stable angina was the most common indication (73% in both groups) and only 3% were admitted due to silent ischemia. Diabetic patients had more severe symptoms (Canadian Cardiovascular Society III-IV) than non-diabetic patients (66% vs. 58%, p<0.01). They more frequently used ACE-inhibitors (33% vs. 19%, p<0.0001) and calcium channel blockers (47% vs. 40%, p<0.01) and more often had a diagnosis of arterial hypertension than non-diabetic patients (50% vs. 33%, p<0.0001). Diabetic patients more often had depressed myocardial function (EF<35%); 12% and 8%, respectively (p<0.01), and more extensive coronary artery disease (left main/3-VD; 48% vs. 37%, p<0.001). The mortality during the subsequent 21 months was 7.9% among diabetic patients and 3.6% among non-diabetic patients (p<0.001). CONCLUSION: Among patients being referred for coronary angiography in Sweden, 15% were patients with a history of diabetes. They differed from patients without such a history by more often having severe symptoms and a higher prevalence of left main/triple vessel disease. Coronary angiography may thus be underused in diabetic patients with chest pain.