Previous studies suggest that TNFSF4 is a susceptibility gene of atherosclerosis. But case-control association analysis in Swedish population and German population provided inconsistent, even opposite results. In order to explore the relationship between this gene and coronary heart disease (CHD) in Chinese Han population, we collected 498 cases and 509 controls from Qilu hospital of Shandong University and analyzed the association between five single-nucleotide polymorphisms (SNPs) (rs1234314, rs45454293, rs3850641, rs1234313, and rs3861950) of TNFSF4 and CHD. On the basis of using traditional statistical analysis methods based on single SNP and haplotypes, we introduced principal component score-based logistic regression analysis to deal with the data. The results suggested that in Armitage trend test, only rs3861950 was significant, when used the Bonferroni correction, and all of the five SNPs were not statistically significant. In the logistic regression analysis which adjusts the confounding factors, all of the five SNPs were not statistically significant. In haplotype analysis, the frequencies of six haplotypes were significantly different in cases and controls (CTAGT (P=0.0006), CTAAC (P=0.0123), CCAGT (P=0.0004), GTGGT (P=0.0329), GCGAC (P<0.0001), and GCAAC (P=0.0173)). In principal component score-based logistic regression analysis, the first principal component has statistical significance (P=0.0236). These results indicate that TNFSF4 is a susceptibility gene of CHD in Chinese Han population.