AIMS: To assess the prevalence of microalbuminuria (MA) and elevated serum lipoprotein (a) (Lp (a)) concentration, and their association with coronary artery disease (CAD) and other conventional cardiovascular risk factors in non-diabetic patients attending a lipid clinic. METHODS: Clinical details were obtained from 96 consecutive non-diabetic patients from whom a fasting blood sample was taken to measure serum lipid, lipoprotein, apolipoprotein and plasma glucose, urea, and electrolyte concentrations. The urine albumin/creatinine ratio (Ua/Uc) was estimated from a random clinic sample. RESULTS: Of the patients, 26% had MA (defined as a Ua/Uc > 2.2 mg/mumol), 38% had an elevated Lp (a) concentration (defined as > 0.4 g/l), 36% were hypertensive (blood pressure > 160/95) or were taking antihypertensive medication, and 25% had established CAD defined on clinical criteria. In men the Ua/Uc ratio was highly associated with age, plasma low density lipoprotein cholesterol, and triglyceride concentrations. In women there was no association between the Ua/Uc ratio and variables examined. Lp (a) concentration was not associated with variables examined in either sex. In multiple logistic regression analysis adjusted for age and sex, serum Lp (a) concentration, diastolic blood pressure and treatment of hyperlipidaemia were highly associated with CAD. MA was not, however, associated with CAD. CONCLUSIONS: MA is common in a lipid clinic and is more likely to be found among older male patients with hyperlipidaemia. However, in contrast with Lp (a) concentrations, MA is not a risk factor for CAD in this high risk population. Lp (a) concentration may be a useful tool in the lipid clinic, but there does not seem to be a justification for measuring the Ua/Uc ratio, at least in non-diabetic subjects.