Methods for measurement of LDL-cholesterol: A critical assessment of direct measurement versus calculation in the treatment goals according to European guidelines (2007)
Abstract
The aim of this paper is to describe the differences in values ??of LDL-cholesterol which arise when using two different methods of measuring LDL-cholesterol: the direct method and indirect method by calculation according to Friedewald formula, in real world laboratories. Methods: A?multicenter international prospective cross-sectional study.
Each of thirteen laboratories from Slovakia and the Czech Republic provided a?set of data from at least 200 patients in its normal operation, in which the measured values ??of total cholesterol, triglycerides, HDL-cholesterol and LDL-cholesterol were measured by direct method (LDL-D). For each patient the value of LDL-cholesterol was grossed up using the Friedewald indirect method (LDL-F).
We compared the consistency in classification of patients into groups according to the treatment goals of LDL-cholesterol, using the direct or indirect method of assessment. According to European guidelines on cardiovascular disease prevention, the Fourth joint task force (2007) the treatment goals ??of LDL-cholesterol are, in various clinical situations, 2, 2,5 and 3 mmol/l.
Results: The ratio of average values ??of LDL-F compared to the average values ??of LDL-D is in the range 91.9% to 130.3%. In 6 laboratories the average LDL-F is lower than the average LDL-D: in 7 laboratories the average LDL-F is higher than the average LDL-D.
The Pearson correlation coefficient r is between 0.6575 (95% CI 0,6430 - 0,6699) to 0.9795 (95% CI 0,9667 - 0,9894). The correlation coefficient in 10 laboratories was above 0.9 and under 0.9 other in the 3 laboratories.
For each laboratory the number of concordant or discordant inclusions in the appropriate category were calculated from the goals of LDL-cholesterol measured by both methods. The number of discordant results is in the range of 14 to 57%.
If we consider LDL-D as a?standard method, the LDL-F values obtained put patients at higher levels of risk ranging from 0 to 55%, to lower levels of risk ranging from 1,5 to 31.5%. Conclusion: The differences in values obtained among real world laboratories are enormous.
Our results support efforts to replace the determination of LDL-cholesterol with other laboratory parameters.
