Aim To evaluate the diagnostic value of serum lipids, lipoprotein (a) and their combination in coronary heart disease (CHD). Methods 546 patients with suspected CHD were enrolled into this study and divided into CHD group (n=354) and non-CHD group (n=192) according to the result of coronary arterography. Their clinical data were analysed by case-control method, and serum lipids were detected by standard method. Results Compared with non-CHD group, the levels of serum total cholesterol, low density lipoprotein cholesterol, lipoprotein (a) and the ratio of total cholesterol to high density lipoprotein cholesterol in CHD group were significantly raised, whereas the level of high density lipoprotein cholesterol were significantly lowered. There was no significant difference in serum triglyceride level and triglyceride/high density lipoprotein cholesterol ratio between two groups. The abnormal rate of total cholesterol, low density lipoprotein cholesterol, total cholesterol/high density lipoprotein cholesterol ratio and lipoprotein (a) in CHD group significantly increased, while the abnormal rate of other lipids didn't show significant difference. The predicative value and likelihood ratio of lipoprotein (a) for CHD was not superior to that of other lipids, but when combined with low density lipoprotein cholesterol, these value would further be elevated. In logistic regression model, lipoprotein (a) wasn't more related to CHD risk than other lipids. Conclusion The diagnostic value of lipoprotein (a) isn't better than other lipids. When combined with low density lipoprotein cholesterol, it may be regarded as a better predicative marker for CHD.