OBJECTIVE--To determine differences in coronary risk factors between women and men and their relation to in-hospital mortality associated with coronary artery bypass grafting. DESIGN--Prospective observational study. SETTING--A regional cardiothoracic centre. PATIENTS--482 (362 (75%) men and 120 (25%) women) consecutive patients who had primary isolated coronary artery bypass grafting. RESULTS--The women were on average three years older than the men (63 v 60 years, P < 0.001). Women more frequently had hypertension (47% v 33%, P < 0.01), diabetes mellitus (21% v 10%, P < 0.005), hypothyroidism (9% v 2%, P < 0.003), and a family history of premature coronary heart disease (49% v 31%, P < 0.0006). More of the men were cigarette smokers (67% v 45%, P > 0.00001). Many of the women and men had dyslipidaemia. Postmenopausal women had a higher concentration of serum total cholesterol than men of a comparable age, (7.3 mmol/l v 6.5 mmol/l, P = 0.0002). Although arterial grafts were often used in both sexes, they were more often used in men than in women (91% v 78% respectively, P = 0.0003). In-hospital mortality was 2.1% (1.4% in men and 4.2% in women, P = 0.14). The estimated one year probability of survival in men who had survived 30 days was 0.99 with 95% confidence interval 0.98 to approximately 1 while that for women was 0.97 with 95% confidence interval 0.91 to approximately 1. Univariate analysis showed that preoperative history of diabetes mellitus was a predictor of mortality (P = 0.03). CONCLUSION--There were differences in the incidence and type of risk factors in men and women who had coronary artery bypass grafting. Preoperative diabetes mellitus was a predictor of in-hospital mortality.