Evaluation of a homogeneous direct LDL-cholesterol assay in diabetic patients: effect of glycemic control.
Currently used methods for LDL-C include calculations [based on total cholesterol (TC), HDL-cholesterol (HDLC), and TG concentrations], ultracentrifugation, and most recently, direct LDL-C assays. Calculation methods, such at the Friedewald formula (6), are well known to have a prominent negative bias in patients with TG concentrations >4.5 mmol/L (400 mg/dL) (6,7). In addition, because TGs are measured, fasting blood samples are preferred. This can present a problem for certain patient populations, including many people with diabetes. Ultracentrifugation methods are time-consuming and expensive and generally are performed only in reference laboratories.
Direct LDL-C assays have been developed recently and have been shown to provide accurate and precise measurements of LDL-C (8-12). They overcome the TG and fasting limitations of calculation methods, are readily adapted to clinical laboratories, are less expensive than ultracentrifugation methods, and provide greatly improved turnaround time. Direct LDL-C assays make use of either an immunoseparation step or specific detergents to separate LDL particles from other lipoproteins, followed by measurement of cholesterol by conventional enzymatic reactions. The N-geneous[TM] LDL-C assay is a solute-based homogeneous assay that agrees well with ultracentrifugation (11). The purpose of the present study was to evaluate the effect of glycemic control (as gauged by hemoglobin [A.sub.1c] on the accuracy of the N-geneous assay. Additionally, we compared this direct LDL-C assay to density gradient ultracentrifugation (13) and to the Friedewald calculation (6), and assessed its usefulness in determining a calculated VLDL-cholesterol (VLDL-C) concentration.
Overnight fasting plasma samples were obtained from 52 patients. Direct LDL-C was measured using the N-geneous LDL-C assay (Genzyme Diagnostics) performed on the Beckman SYNCHRON LX 20 (Beckman-Coulter). Briefly, this assay involves a two-reagent process. The first reagent is a detergent that solubilizes only non-LDL lipoprotein particles. The solubilized cholesterol is consumed in a non-color-producing reaction by cholesterol esterase and cholesterol oxidase. The second reagent is then added, which solubilizes the remaining LDL particles and links the enzymatic consumption of LDL-C to a chromogenic coupler. The colored product is measured spectrophotometrically.
TC and TG concentrations were measured using enzymatic methods on the SYNCHRON LX System. HDL-C was measured with Beckman-Coulter reagents on the SYNCHRON LX System, which uses a detergent that solubilizes only HDL particles, allowing HDL-C to be measured by a color-producing enzymatic reaction. An estimated LDL-C concentration was obtained using the Friedewald calculation (6): TC - (HDL-C + TG/5). VLDL-C concentrations were estimated using TG/5, and were also calculated using the direct LDL-C measurement in the formula: VLDL-C = TC - (HDL-C + LDL-C). LDL-C and VLDL-C were also measured by the Vertical Auto Profile (VAP) technique (Atherotech) (13). Briefly, VAP uses density gradient ultracentrifugation to separate lipoproteins, with subsequent measurement of cholesterol in each of the resulting density bands. Hemoglobin A, was measured by HPLC (TOSOH A1C2.2) on diabetic patients (41 of the 52 samples) (14, 15). On all 52 plasma samples, correlation studies were performed comparing LDL-C measurements by the N-geneous LDL-C assay with those from the ultracentrifugation (VAP) technique (13) and Friedewald calculation (6).
Linearity of the direct LDL assay was assessed with a patient plasma sample determined to have a LDL-C concentration of 2400 g/L. The following dilutions, in normal saline, were measured: 1:1, 1:2, 1:4, 1:8, 1:16, 1:32, 1:64, and 1:128. Within-run precision studies used a serum-based control (Liquichek Lipid Control level 1; Bio-Rad Laboratories) and the N-geneous LDL-C lyophilized serum calibrator (Genzyme). Between-run precision studies used Bio-Rad Liquichek Lipid Control level 1 and level 2. Within-day precision was assessed on 20 consecutive assays, whereas between-day precision was determined from data obtained in triplicate over 20 separate days. Interference studies were performed on a patient sample with a TG concentration of 28 mmol/L (2490 mg/dL). Direct LDL-C measurements were obtained from the undiluted sample as well as from 1:2, 1:4, 1:8, and 1:16 dilutions of the sample in normal saline.
Statistical analyses and graphing of results were performed using Microsoft Excel 97 SR-2.
For the 52 patients, the ranges and means for TGs, TC, HDL-C, and direct LDL-C were as follows: TGs, range, 1.3-16.3 mmol/L (51-631 mg/dL), mean, 4.7 mmol/L (183 mg/dL); TC, range, 3.0-7.4 mmol/L (116-287 mg/dL), mean, 4.8 mmol/L (186 mg/dL); HDL-C, range, 0.4-1.9 mmol/L (17-72 mg/dL), mean, 1.2 mmol/L (48 mg/dL); direct LDL-C, range, 0.9-5.7 mmol/L (34-220 mg/dL), mean, 2.9 mmol/L (112 mg/dL). The values obtained for LDL-C and VLDL-C with the VAP method were as follows: LDL-C, range, 0.8-6.2 mmol/L (32-239 mg/dL), mean, 3.0 mmol/L (114 mg/dL); VLDL-C, range, 0.2-2.4 mmol/L (6-92 mg/dL), mean, 0.8 mmol/L (31 mg/dL). The hemoglobin [A.sub.1c] for the 41 patients tested was 5.3-10.3%.
The results obtained from the N-geneous direct LDL-C assay demonstrated acceptable correlation with the reference (VAP) LDL-C measurements: [R.sup.2] = 0.92; y-intercept = 0.06 mmol/L (2.5 mg/dL); slope = 0.96; [S.sub.y|x] = 0.24 mmol/L (9.1 mg/dL). A small negative bias was identified at lower hemoglobin [A.sub.1c] concentrations, but no significant bias was identified with hemoglobin [A.sub.1c] up to 10.3% (Fig. 1A). As expected, the Friedewald LDL-C showed a prominent negative bias at TG concentrations >4.5 mmol/L (400 mg/dL), whereas the direct LDL-C measurement showed minimal bias throughout the range of patient TG concentrations (Fig. 1B).
The estimated VLDL-C (TG/5) compared with the VAP VLDL-C demonstrated an increasingly positive bias (up to 45%) with increasing TG concentrations, whereas VLDL-C calculated using the measured direct LDL-C, HDL-C, and TC concentrations showed little bias when compared with VAP VLDL-C over the same TG range (data not shown). Correlation between the calculated VLDL-C and the VAP VLDL-C was fair: [R.sup.2] = 0.77; slope = 0.93; y-intercept = 0.03 mmol/L (1.1 mg/dL); [S.sub.y|x] = 0.23 mmol/L (8.9 mg/dL).
[FIGURE 1 OMITTED]
On the basis of the results of serial dilutions of 6.2 mmol/L (240 mg/dL) LDL-C, the N-geneous assay demonstrated good linearity for LDL-C concentrations of 0.8-6.2 mmol/L (30-240 mg/dL). Acceptable recoveries (91.1-99.6%) were obtained throughout this range. Within-day precision data obtained using the Bio-Rad level 1 control showed a mean of 1.6 mmol/L (61 mg/dL) with a CV of 1.4%, whereas the Genzyme calibrator sample yielded a mean of 2.5 mmol/L (95 mg/dL) with a CV of 1.3%. Between-day precision on the Bio-Rad level 1 control demonstrated a mean of 1.6 mmol/L (60 mg/dL) with a CV of 1.5%, whereas the Bio-Rad level 2 control showed a mean of 3.3 mmol/L (125 mg/dL) and a CV of 1.4%. No interference with the direct LDL-C assay was observed at TG concentrations of 1.8-14.1 mmol/L (156-1245 mg/dL). Throughout this range, the maximum deviation of any individual measurement from the mean was <7%. However, we observed a negative bias of 78% at a TG concentration of 28.1 mmol/L (2490 mg/dL).
Methods for the direct measurement of LDL-C are now readily available to hospital and commercial laboratories. These assays involve either immunoseparation or detergent separation of LDL-C from other lipoprotein-associated cholesterol. As a whole, these assays have been well studied and compared to the indirect calculation and ultracentrifugation techniques of determining LDL-C concentration (8-12), and have generally shown good correlation with both methods. In the present study, we found good correlation of the N-geneous direct LDL-C assay with the Friedewald calculation at TG concentrations <2.3 mmol/L (200 mg/dL; data not shown) and with the VAP density gradient ultracentrifugation technique. We also demonstrated a prominent negative bias in the Friedewald calculation, but not in the direct LDL-C method, compared with the VAP LDL-C at TG concentrations >4.5 mmol/L (400 mg/dL). No interference with the N-geneous assay was observed at TG concentrations up to 14.1 mmol/L (1245 mg/dL), a point at which the TG concentrations would override the LDL-C in clinical importance because of the risk of acute pancreatitis. The assay showed good linearity from LDL-C concentrations of 0.8-6.2 mmol/L (30-240 mg/dL), and both within-day and day-to-day precision were excellent. These findings are consistent with a previous report evaluating the N-geneous LDL-C assay (11).
Situations where a direct LDL-C measurement appear to be preferable to a calculated LDL-C value include patients with TGs >4.5 mmol/L (400 mg/dL) and patients who are unable to fast (6, 7). Often persons with diabetes will fall into one or both of these categories and will likely represent a large group on whom direct LDL-C assays will be used. Persons with diabetes often have increases in altered forms of LDL particles, such as glycated LDLs and oxidized LDLs (4,5). It was therefore important to assess what effect, if any, glycemic control has on the ability of a direct LDL assay to accurately measure LDL-C. Our data show that the N-geneous direct LDL assay shows no significant bias associated with increasing hemoglobin Al, up to 10.3% as measured by HPLC.
In the present study, we also compared estimated and calculated VLDL-C concentrations to those measured via the VAP procedure. As expected, estimating VLDL-C by dividing the TG concentration by 5 showed a positive bias that became more prominent as the TG concentration increased. Calculating VLDL-C yielded little average bias when compared with VAP VLDL-C; however, a few individual calculated values showed larger biases, up to 150%. These large biases were most pronounced at lower VLDL-C concentrations, where they would be less clinically significant. Additionally, there was fair correlation between the calculated and VAP VLDL-C values. The exact significance of VLDL-C as a clinically important risk factor for atherosclerosis remains controversial, and calculated VLDL-C may be inaccurate in the presence of lipoproteins such as intermediate-density lipoproteins and lipoprotein(A), for example. Nevertheless, a calculated VLDL-C value may be useful in clinical and research settings.
In conclusion, the N-geneous direct LDL-C assay shows good correlation with the VAP ultracentrifugation method of measuring LDL-C and does not appear to be affected by glycemic control or TGs >4.5 mmol/L (400 mg/dL), supporting its usefulness in diabetic patients.
(1.) National Cholesterol Education Program. Second report of the Expert Panel on Detection, Evaluation, and Treatment of High Blood Cholesterol in Adults (Adult Treatment Panel II). Circulation 1994;89:1329-445.
(2.) Sacks FM, Pfeffer MA, Moye LA, Rouleau JL, Rutherford JD, Cole TG, et al. The effect of pravastatin on coronary events after myocardial infarction in patients with average cholesterol levels. New Engl J Med 1996;335: 1001-9.
(3.) Scandinavian Simvastatin Survival Study Group. Randomized trial of cholesterol lowering in 4444 patients with coronary heart disease: the Scandinavian Simvastatin Survival Study (4S). Lancet 1994;344:1383-9.
(4.) Lyons TJ. Lipoprotein glycation and its metabolic consequences. Diabetes 1992;41(Suppl 2):67-73.
(5.) Lyons TJ. Oxidized low density lipoproteins: a role in the pathogenesis of atherosclerosis in diabetes. Diabet Med 1991;8:411-9.
(6.) Friedewald WT, Levy RI, Fredrickson DS. Estimation of the concentration of low-density lipoprotein cholesterol in plasma, without use of the preparative ultracentrifuge. Clin Chem 1972;18:499-502.
(7.) McNamara JR, Cohn JS, Wilson PWF, Schaefer EJ. Calculated values for low-density cholesterol in the assessment of lipid abnormalities and coronary disease risk. Clin Chem 1990;36:36-42.
(8.) McNamara JR, Cole TG, Contois JH, Ferguson CA, Ordovas JM, Schaefer EJ. Immunoseparation method for measuring low-density lipoprotein cholesterol directly from serum evaluated. Clin Chem 1995;41:232-40.
(9.) Whiting MJ, Shephard MDS, Tallis GA. Measurement of plasma LDL cholesterol in patients with diabetes. Diabetes Care 1997;20:12-4.
(10.) Hirany S, Li D, Jialal I. Amore valid measurement of low-density lipoprotein cholesterol in diabetic patients. Am J Med 1997;102:48-53.
(11.) Rifai N, Iannotti E, DeAngelis K, Law T. Analytical and clinical performance of a homogeneous enzymatic LDL-cholesterol assay compared with the ultracentrifugation-dextran sulfate-[Mg.sup.2+] method. Clin Chem 1998;44:1242-50.
(12.) Nauck M, Graziani MS, Bruton D, Cobbaert C, Cole TG, Lefevre F, et al. Analytical and clinical performance of a detergent-based homogeneous LDL-cholesterol assay: a multicenter evaluation. Clin Chem 2000;46:506-14.
(13.) Chung BH, Segrest JP, Ray MJ, Brunzell JD, Hokanson JE, Krauss RM, et al. Single vertical spin density gradient ultracentrifugation. Methods Enzymol 1986;128:181-209.
(14.) Khuu HM, Robinson CA, Goolsby K, Hardy RW, Konrad RJ. Evaluation of a fully automated high-performance liquid chromatography assay for hemoglobin A1c. Arch Pathol Lab Med 1999;123:763-7.
(15.) Khuu HM, Robinson CA, Brissie RM, Konrad RJ. Postmortem diagnosis of unsuspected diabetes established by determination of decedent's hemoglobin A1c level. J Forensic Sci 1999;44:643-6.
Brian D. Ragland, Robert J. Konrad, Carolyn Chaffin, C. Andrew Robinson, and Robert W. Hardy * (Department of Pathology, University of Alabama at Birmingham, Birmingham, AL 35294; * address correspondence to this author at: The University of Alabama at Birmingham, Department of Pathology, LHRB 573, 1530 3rd Ave. 5, Birmingham, AL 35294-0007; fax 205-975-9927, e-mail email@example.com)
|Printer friendly Cite/link Email Feedback|
|Title Annotation:||Technical Briefs|
|Author:||Ragland, Brian D.; Konrad, Robert J.; Chaffin, Carolyn; Robinson, C. Andrew; Hardy, Robert W.|
|Date:||Nov 1, 2000|
|Previous Article:||Stability of cannabinoids in hair samples exposed to sunlight.|
|Next Article:||Evaluation of DNA fragment sizing and quantification by the Agilent 2100 Bioanalyzer.|