Objectives: Identifying the potential sources of bias in the direct measurement of low-density lipoprotein cholesterol (LDL-C) and high-density lipoprotein cholesterol (HDL-C) is important. In this study, we aimed to investigate the effect of remnant cholesterol on LDL-C and HDL-C levels measured using homogenous methods. Methods: We obtained 41 commutable frozen serum samples and measured LDL-C and HDL-C levels. Eight measurement systems were used, and the degree of bias was obtained by comparing with the values obtained using the reference methods. Correlations among remnant cholesterol/LDL-C, remnant cholesterol/HDL-C, and bias were analyzed using Spearman's analysis. Results: In all eight systems, samples with a positive bias >4 % had lower LDL-C levels and higher remnant cholesterol levels, as measured by the reference methods, compared to those with a bias <= 4 %. A significant correlation between remnant cholesterol/LDL-C and a positive bias of LDL-C was observed in six of the eight systems evaluated. For HDL-C bias, three systems showed a positive correlation, and three systems showed a negative correlation. In some systems, LDL-C bias was higher in samples with remnant cholesterol/LDL-C >= 0.25 than in those with remnant cholesterol/LDL-C <0.25. Conclusions: Remnant cholesterol has a potential effect on direct LDL-C and HDL-C measurements, which has been observed when several measurement systems are used. For these systems, manufacturers should improve the methods to reduce the interference of remnant cholesterol.