Stabilityof Total Cholesterol,High-Density-LipoproteinCholesterol,and

When human serum must be stored for extended periods before analysis, conditions must be such that samples give results comparable with those for “fresh ” (unstored) sam-ples. In a balanced experiment with statistically pre-deter-mined sample size, we assessed, at three-weekly intervals for 18 weeks, the stability of total cholesterol, high-density-lipoprotein cholesterol, and triglycerides in human serum stored at-15 #{176}C.No significant changes were observed. AddItIonal Keyphrases: variation, sources of triglycerides analysisof variance Studies of the stability of high-density-lipoprotein choles-terol (HDL cholesterol) in frozen serum or plasma have demonstrated increases (1, 2), decreases (3, 4), and no change (3,5, 6) in assayed valuesdetermined with various precipitation and analytical methods after various storage periods. Other investigators have reported that values for samples with low initial HDL cholesterol tend to increase after frozen storage, whereas values for samples with high initial values tend to decrease (7-9). Triglycerides (10) and total cholesterol (3, 10-13) in serum appear to be stable during frozen storage, although results of such experiments may depend on the analytical methods used (11), and perhaps also on the storage temperature. The laboratory of our Epidemiology Unit is currently participating in the international MONICA project (14), which is monitoring trends in cardiovascular disease in several countries. There is concern for comparability be-tween those laboratories assaying fresh samples and those assaying frozen samples in this project. Although several storage experiments at-20 #{176}Chave been reported (2,3,5,7, 8, 15), participation in the MONICA project required investi-gation of storage effects in our laboratory using our meth-ods.We therefore designed an experiment to determine the effects of storage at-15 #{176}Con HDL cholesterol, triglycer-ides, and total cholesterol in human serum (from apparently healthy people) measured with our routine precipitation and analytical methods