Reference standardization and analytical performance of a liquid homogeneous high-density lipoprotein cholesterol method compared with chemical precipitation method

Citation
P. Halloran et al., Reference standardization and analytical performance of a liquid homogeneous high-density lipoprotein cholesterol method compared with chemical precipitation method, ARCH PATH L, 123(4), 1999, pp. 317-326
Citations number
26
Categorie Soggetti
Research/Laboratory Medicine & Medical Tecnology","Medical Research Diagnosis & Treatment
Journal title
ARCHIVES OF PATHOLOGY & LABORATORY MEDICINE
ISSN journal
00039985 → ACNP
Volume
123
Issue
4
Year of publication
1999
Pages
317 - 326
Database
ISI
SICI code
0003-9985(199904)123:4<317:RSAAPO>2.0.ZU;2-B
Abstract
Background.-The use of high-density lipoprotein cholesterol (HDL-C) levels as a risk factor for coronary heart disease necessitates an accurate and pr ecise method for measuring HDL-C. The Centers for Disease Control and Preve ntion HDL-C reference method (RM) and designated comparison method (DCM) ar e time-consuming, expensive, and impractical for routine clinical use. We e valuated the Liquid N-geneous (LN-gen) HDL-C assay (Genzyme Diagnostics, Ca mbridge, Mass) to determine if this homogeneous reagent meets the National Cholesterol Education Program requirements for HDL-C evaluation. Design.-Accuracy of the LN-gen HDL-C assay was compared in combination with phosphotungstic acid (PTA) precipitation with DCM HDL-C for normotriglycer idemic serum specimens (triglycerides <2.0 g/L) and with RM HDL-C for speci mens with triglycerides levels greater than or equal to 2.0 g/L. Setting.-Genzyme Diagnostics (with RM and DCM assayed by Pacific BioMetrics Inc, Seattle, Wash) and the Lipid Reference Laboratory of the University H ospital Rotterdam, The Netherlands. Results.-Linear regression to DCM (n = 90) was (LN-gen = 1.015 DCM + 0.01 g /L, r = 0.993, SE = 0.015 g/L) and (PTA = 1.004 DCM - 0.017 g/L, r = 0.980, SE = 0.025 g/L), with a mean percent bias to DCM of 3.3% and -2.8% for LN- gen and PTA, respectively. The comparison with RM (n = 69) showed an increa sed mean bias for PTA (-5.8%) as compared with LN-gen (1.5%). The correlati on and regression equations were (LN-gen = 1.020 RM - 0.002 g/L, r = 0.985, SE = 0.017 g/L) and (PTA = 1.042 RM - 0.032 g/L, r = 0.984, SE = 0.018 g/L ). The precision of LN-gen was confirmed at <2.1% coefficient of variation, and the total error was calculated to be less than or equal to 7.7% for bo th normotriglyceride and elevated triglyceride specimens at HDL-C decision points of 0.35 g/L and 0.60 g/L. Conclusions.-The LN-gen HDL-C assay offers a cost-effective convenient meth od for meeting the 1998 precision, bias, and total error recommendations of the National Cholesterol Education Program.