2003 Volume 10 Issue 3 Pages 145-153
Background: Accurate and precise measurements of total cholesterol (TC) and HDL-cholesterol (HDL-C) are necessary for effective diagnosis and treatment of lipid disorders. We studied the impact of TC certification and HDL-C evaluation in Japanese clinical laboratories to standardize their measurements.
Methods: We selected 78 laboratories participated at least twice for TC and 46 laboratories participated twice for HDL-C in the standardization protocols developed by the Cholesterol Reference Method Laboratory Network (CRMLN). We compared the initial and subsequent results using the performance guidelines established by US National Cholesterol Education Program (NCEP).
Results: For TC, mean percentage bias of all participants was - 0.93% and - 0.49% for the initial and second rounds, respectively. Mean within-sample CV was 0.72% and 0.69% for the initial and second rounds, respectively. For HDL-C, mean percentage bias of all participants was - 1.86% and - 0.06% for the initial and second events, respectively. Mean among-run CV was 1.56% and 1.58% for the initial and second events, respectively.
Conclusions: TC accuracy in the second round than the initial round tended to improvement although statistically not significant, however in the five years follow-up, mean absolute percentage bias was reduced over time. HDL-C accuracy was statistically improved in the second event than the initial event. The precision for both TC and HDL-C did not change. This study shows CRMLN protocols contribute effectively to improvement of TC and HDL-C performance.