Comparison of various lipid, lipoprotein, and bilirubin combinations as risk factors for predicting coronary artery disease

Studies were performed to determine if serum bilirubin, when combined with various lipid and lipoprotein risk factors, enhances our ability to predict coronary artery disease (CAD). This hypothesis was tested in a retrospective study of 644 middle-aged males who had undergone coronary angiography. The traditional risk factors of cholesterol, high density lipoprotein cholesterol (HDL-C), cholesterol/HDL-C ratios, triglycerides, age, cigarette smoking, and systolic blood pressure were tested by discriminant analysis, as were various cholesterol/bilirubin, cholesterol/(HDL-C + bilirubin), and low-density lipoprotein cholesterol (LDL-C)/ (HDL-C + bilirubin) ratios. Each of these bilirubin-containing ratios was found to be an independent risk predictor when tested with the traditional risk factors. When the LDL-C/(HDL-C + bilirubin) ratio was included with the traditional risk predictors, it improved the prediction of severe CAD from 28.4 to 35.3% and the overall correct classification of CAD from 68.3 to 71.1%. When the 75th percentile was used as a cut-point, the diagnostic sensitivities obtained with cholesterol/(HDL-C + bilirubin) ratios (52.1%) and LDL-C/(HDL-C + bilirubin) ratios (51.7%) were better than those obtained with cholesterol/HDL-C ratios (40.4%) (P = 0.033 and 0.048, respectively). LDL-C/(HDL-C + bilirubin) ratios also improved the prediction of severe CAD over those obtained with LDL-C/HDL-C ratios (43.4%); however, the changes were not statistically significant (P = 0.096). If confirmed in other populations, serum bilirubin may be combined with LDL-C/HDL-C ratios, cholesterol/HDL-C ratios, cholesterol, or with various apolipoproteins to improve the prediction of CAD. (C) 2000 Elsevier Science Ireland Ltd. All rights reserved.