Lp(a) and conventional risk profiles predict the severity of coronary stenosis in high-risk hospital-based patients

To explore predictive power of Lp(a), of conventional lipoprotein profiles and their carrier proteins, and of biometric measurements, for the presence and severity of angiographically documented coronary disease in high-risk patients, and to compare risk profiles in men and women.

We determined coronary artery disease (CAD) risk factors in 1308 Australian Caucasian patients (313 women and 995 men) aged < or =65 years who consecutively underwent coronary angiography.

In univariate analyses of the risk factors, lipid profiles, Lp(a), cigarette smoking, diabetes, hypertension and obesity were all higher in men and women with CAD and changed significantly with the number of significantly diseased vessels (> or =50% luminal obstruction). When stepwise logistic regression analysis was applied, age (OR 1.06, 95% CI: 1.04-1.09), TC/HDL-C (OR 1.29, 95% CI: 1.15-1.45), male gender (OR 2.64, 95% CI: 1.67-4.16), hyperLp(a) (> or =300 mg/L) (OR 2.09, 95% CI: 1.42-3.07), lifetime smoking dose (OR 1.02, 95% CI: 1.01-1.03), diabetes (OR 2.19, 95% CI: 1.14-4.18) and waist/hip ratio (OR 14.53, 95% CI: 1.21-174.90) were predictive of the disease. Both Lp(a) levels and percentage of hyperLp(a) increased linearly with the number of significantly diseased vessels. When the analyses were conducted in men and women separately, hyperLp(a), TC/HDL-C, lifetime smoking dose and age remained as significant predictors in both groups but the waist/hip ratio was only predictive in women.

As Lp(a) is an independent predictor of the occurrence and extent of coronary stenosis and relevant to treatment options, we suggest that it should be measured routinely in the coronary risk profile assessment of high-risk patients.