Single-spin density-gradient ultracentrifugation (DUC) has proven to be a reproducible method for detection of low-density-lipoprotein (LDL) heterogeneity. Recently another method has been described for this: gradient gel electrophoresis (GGE) of serum, a method that might be more suitable for screening. To gain insight into the relationship of GGE to DUC and into their reproducibility, we determined LDL heterogeneity by DUC and GGE in 41 healthy individuals. In 90.2% (n = 37) of the subjects, the number of LDL subfractions found by both methods agreed. In addition, the density and the relative migration distance of the predominant LDL subfraction observed with the respective methods showed a strong correlation (Pearson correlation, r = 0.85, P less than 0.0001). Although it was not possible to compare for all aspects of LDL heterogeneity, these data suggest that GGE is a valid method of analysis for LDL heterogeneity. In screening programs, it may be necessary to store samples. Therefore, we studied in 24 sera the influence of storage at -80 degrees C for one, four, and 12 weeks on the LDL subfraction distribution detected by each method. LDL heterogeneity was maintained during storage under these conditions.