A High-Throughput NMR Method for Lipoprotein-X Quantification

Lipoprotein X (LP-X) is an abnormal cholesterol-rich lipoprotein particle that accumulates in patients with cholestatic liver disease and familial lecithin-cholesterol acyltransferase deficiency (FLD). Because there are no high-throughput diagnostic tests for its detection, a proton nuclear magnetic resonance (NMR) spectroscopy-based method was developed for use on a clinical NMR analyzer commonly used for the quantification of lipoproteins and other cardiovascular biomarkers. The LP-X assay was linear from 89 to 1615 mg/dL (cholesterol units) and had a functional sensitivity of 44 mg/dL. The intra-assay coefficient of variation (CV) varied between 1.8 and 11.8%, depending on the value of LP-X, whereas the inter-assay CV varied between 1.5 and 15.4%. The assay showed no interference with bilirubin levels up to 317 mg/dL and was also unaffected by hemolysis for hemoglobin values up to 216 mg/dL. Samples were stable when stored for up to 6 days at 4 °C but were not stable when frozen. In a large general population cohort (n = 277,000), LP-X was detected in only 50 subjects. The majority of LP-X positive cases had liver disease (64%), and in seven cases, had genetic FLD (14%). In summary, we describe a new NMR-based assay for LP-X, which can be readily implemented for routine clinical laboratory testing.

Comparison Between the Direct Method and Friedewald's Formula for the Determination of Low-Density Lipoprotein Cholesterol Serum Levels in Horses

This study aimed to compare the use of enzymatic colorimetry and Friedewald's formula for the determination of LDL in horses. A total of 260 samples were used. Direct analysis was performed to determine low-density lipoprotein (LDL), high-density lipoprotein (HDL) cholesterol, triglycerides (TG), and total cholesterol (TC). The LDL level was calculated using the Friedewald equation (LDL= TC-HDL-TG/5). The correlations between the direct LDL analysis and the LDL calculated using the Friedewald formula were determined by Pearson's tests, and the coefficient of determination was also obtained by linear regression using SAS software (P<.05) and the kappa value. The mean value (± standard deviation) of the LDL was 22.12 (±10.34) mg/dL, and that of the result obtained by the Friedewald formula was 19.94 (±19.13) mg/dL. The correlation between the two variants analyzed in this experiment was significant, with a value of P < .001, and values of r = 0.688 and R² = 0.4893. These results are relevant, given that recent studies have demonstrated a correlation between LDL plasma values and equine metabolic syndrome. Previous studies have reported discrepancies between the data obtained using Friedewald's method and the LDL-c values directly determined in humans and animals. The results of the present study suggest that the Friedewald method can be used to estimate the LDL plasma concentration in horses. Nevertheless, the coefficient of determination was not found to be adequate to recommend the Friedewald formula as a replacement for the enzymatic colorimetric method in determining LDL in horses.

Clinical laboratory characteristics of patients with obstructive jaundice accompanied by dyslipidemia

BACKGROUND

Abnormal lipid metabolism manifests as hypercholesterolemia in patients with obstructive jaundice due to lipoprotein X (LpX). Our aim was to explore the clinical laboratory characteristics of patients with obstructive jaundice accompanied by dyslipidemia in a large number of samples.

METHODS

A total of 665 patients with obstructive jaundice were included and categorized into two groups (with/without dyslipidemia) based on the ratio of the sum of HDL-c and LDL-c to total cholesterol [(HDL-c + LDL-c)/TC] with a cut-off value of 0.695. Laboratory liver, kidney, and blood lipid parameters were determined. Cholesterol composition assessment was performed by ultracentrifugation and high-performance liquid chromatography (UC-HPLC), and serum protein profiles were analyzed by capillary electrophoresis.

RESULTS

Liver function in patients with obstructive jaundice accompanied by dyslipidemia was more aggravated than that in patients with simple obstructive jaundice (P < 0.05). The (HDL-c + LDL-c)/TC ratio was negatively correlated with bilirubin levels (P < 0.05). In addition, the difference in ApoB/LDL-c ratios was statistically significant between the obstructive jaundice accompanied by dyslipidemia group and healthy control group (P < 0.05). The LDL-c concentration determined by the UC-HPLC method was more than five times that determined by the enzymatic method (P < 0.05). Bisalbuminemia was found in 43 of 60 patients with obstructive jaundice accompanied by hypercholesterolemia.

CONCLUSIONS

In patients with obstructive jaundice, the decreased (HDL-c + LDL-c)/TC ratio may be a novel marker to identify dyslipidemia secondary to LpX. The decreased ratio was associated with poor liver function and indicated disease progression.

Update on the laboratory investigation of dyslipidemias

The role of the clinical laboratory is evolving to provide more information to clinicians to assess cardiovascular disease (CVD) risk and target therapy more effectively. Current routine methods to measure LDL-cholesterol (LDL-C), the Friedewald calculation, ultracentrifugation, electrophoresis and homogeneous direct methods have established limitations. Studies suggest that LDL and HDL size or particle concentration are alternative methods to predict future CVD risk. At this time there is no consensus role for lipoprotein particle or subclasses in CVD risk assessment. LDL and HDL particle concentration are measured by several methods, namely gradient gel electrophoresis, ultracentrifugation-vertical auto profile, nuclear magnetic resonance and ion mobility. It has been suggested that HDL functional assays may be better predictors of CVD risk. To assess the issue of lipoprotein subclasses/particles and HDL function as potential CVD risk markers robust, simple, validated analytical methods are required. In patients with small dense LDL particles, even a perfect measure of LDL-C will not reflect LDL particle concentration. Non-HDL-C is an alternative measurement and includes VLDL and CM remnant cholesterol and LDL-C. However, apolipoprotein B measurement may more accurately reflect LDL particle numbers. Non-fasting lipid measurements have many practical advantages. Defining thresholds for treatment with new measurements of CVD risk remain a challenge. In families with genetic variants, ApoCIII and lipoprotein (a) may be additional risk factors. Recognition of familial causes of dyslipidemias and diagnosis in childhood will result in early treatment. This review discusses the limitations in current laboratory technologies to predict CVD risk and reviews the evidence for emergent approaches using newer biomarkers in clinical practice.

Should apolipoprotein B replace LDL cholesterol as therapeutic targets are lowered?

PURPOSE OF REVIEW

The success of LDL cholesterol (LDL-C) as a predictor of atherosclerotic cardiovascular disease and a therapeutic target is indisputable. Apolipoprotein B (apoB) is a more contemporary and physiologically relevant measure of atherogenic lipoproteins. This report summarizes recent comparisons of apoB and LDL-C as biomarkers of cardiovascular risk.

RECENT FINDINGS

Multiple recent reports have found that LDL-C methods perform poorly at low concentrations (<70 mg/dl). Several meta-analyses from randomized controlled trials and large prospective observational studies have found that apoB and LDL-C provide equivalent information on risk of cardiovascular disease. More innovative analyses have asserted that apoB is a superior indicator of actual risk when apoB and LDL-C disagree.

SUMMARY

ApoB is more analytically robust and standardized biomarker than LDL-C. Large population studies have found that apoB is at worst clinically equivalent to LDL-C and likely superior when disagreement exists. Realistically, many obstacles prevent the wide spread adoption of apoB and for now providers and their patients must weigh the costs and benefits of apoB.

Severe hypercholesterolemia mediated by lipoprotein X in a patient with cholestasis

Lipoprotein X (LpX) is an abnormal lipoprotein associated with cholestasis. It is a significant cause of severe hypercholesterolemia and should always be considered in patients with cholestatic liver disease. This case highlights the significance of LpX as a cause of severe hypercholesterolemia in a patient with cholestasis secondary to a granulomatous hepatitis attributed to tuberculosis. Lipoprotein agarose gel electrophoresis and gradient gel electrophoresis were performed for the detection of LpX. The liver function tests, electrolytes, lipid profile and bile acids were also determined. Anti-tuberculous therapy was initiated and the liver functions improved with normalisation of the lipid profile.

Evaluation of Methods for the Measurement of Low-Density Lipoprotein Cholesterol

A high concentration of low-density lipoprotein cholesterol (LDL-C) in plasma is one of the strongest risk factors for atherosclerotic cardiovascular disease and mortality. The most common approach to determining LDL-C in the clinical laboratory is the Friedewald calculation. There is an increased interest to improve the accuracy of LDL-C estimated by this equation. The expert panel convened by National Cholesterol Education Program has recommended the development of accurate direct methods to measure LDL-C. Several homogeneous and fully automated methods have been introduced in recent years that show improved precision and accuracy over earlier methods, especially the Friedewald calculation. Each of the atherogenic particles in plasma—very-low, intermediate-, and low-density lipoprotein—as well as lipoprotein (a), contain one molecule of apolipoprotein B (apoB) and thus, plasma total concentration of apoB reflects the number of atherogenic particles. Several studies suggested that the measurement of apoB could improve the prediction of risk of coronary artery disease. Thus, in addition to the newly developed direct assays, alternative calculation procedures have been proposed that also take into consideration total serum apoB concentration for the estimation of LDL-C and the presence of small, dense LDL particles. The new generation of homogenous methods for the measurement of LDL-C and the use of serum apoB concentration for the estimation of LDL-C can contribute to the accurate LDL-C determination.

Lipid profiling of lipoprotein X: Implications for dyslipidemia in cholestasis

Lipoprotein X (Lp-X) is an abnormal lipoprotein that may typically be formed in intra- and extrahepatic cholestasis and potentially interfere with lipid analysis in the routine lab. To gain insight into lipid class and species composition, Lp-X, LDL and HDL from cholestatic and control serum samples were subjected to mass spectrometric analysis including phospholipids (PL), sphingolipids, free cholesterol (FC), cholesteryl esters (CE) and bile acids. Our analysis of Lp-X revealed a content of 46% FC, 49% PL with 34% phosphatidylcholine (PC) as main PL component. The lipid species pattern of Lp-X showed remarkable high fractions of mono-unsaturated species including PC 32:1 and PC 34:1 and phosphatidylethanolamine (PE) 32:1 and 34:1. LDL and HDL lipid composition in the same specimens strongly reflected the lipid composition of Lp-X with increased PC 32:1, PC 34:1, PE 32:1, PE 34:1 and FC accompanied by decreased CE compared to controls. Comparison of Lp-X and biliary lipid composition clearly indicates that Lp-X does not originate from a sole release of bile lipids. Moreover, these data present evidence for increased hepatic fatty acid and PL synthesis which may represent a reaction to high hepatic FC level observed during cholestasis.

Premature myocardial infarction is strongly associated with increased levels of remnant cholesterol

BACKGROUND

Remnant cholesterol has been defined as the cholesterol present in triglyceride-rich remnant lipoproteins. Elevated levels of remnant cholesterol have been associated with increased cardiovascular risk. Acute myocardial infarction (AMI) in very young individuals (≤40 years) represents a rare disease with a typical risk factor profile and a lipid phenotype that is characterized by a predominance of elevated triglyceride-rich lipoproteins.

OBJECTIVE

The aim of this study was to investigate the role of remnant cholesterol in premature AMI.

METHODS

We prospectively enrolled 302 patients into our multicenter case-control study comprising 102 consecutive myocardial infarction survivors (≤40 years) and 200 hospital controls. Myocardial infarction patients were frequency matched for age, gender, and center. Remnant cholesterol was calculated from standard lipid parameters.

RESULTS

Remnant cholesterol was 1.7-fold higher in premature AMI patients compared with controls (61.1 ± 36.8 vs 35.8 ± 16.8 mg/dL; P < .001). Remnant cholesterol was the lipid fraction most strongly associated with premature myocardial infarction (odds ratio 3.87; 95% confidence interval 2.26-6.64; P < .001) for an increase of 1-standard deviation. This observation was independent from clinical risk factors and plasma lipid levels.

CONCLUSIONS

Remnant cholesterol is strongly associated with premature myocardial infarction, can be easily calculated, and might serve as a new potent risk marker in this young patient population.

A multicentre analysis of four low-density lipoprotein cholesterol direct assays in samples with extreme high-density lipoprotein cholesterol concentrations

BACKGROUND

Although LDL-C has been traditionally estimated using the Friedewald formula (FF), several direct homogeneous assays have been developed to overcome the limitations of this formula and the complicated manual procedure required in the reference method. However, several differences have been reported between these assays in certain situations.

METHODS

Two groups of 105 samples with extreme low and high HDL-C concentrations were processed, employing four different instruments and with the reagents for total cholesterol, triglycerides, HDL-C and LDL-C provided by the distinct manufacturers.

RESULTS

Statistical tests indicated important differences between HDL-C and LDL-C homogeneous methods. Poor correlation, significant bias and high discrepancy in cardiovascular disease risk classification were observed for LDL-C direct assays in the low HDL-C group, whereas better results were obtained when comparing LDL-C levels estimated with the FF. In contrast, three of the four instruments generated LDL-C direct results with a good agreement in the high HDL-C group, even though an appreciable misclassification percentage in risk categories must be taken into account.

CONCLUSIONS

Our results indicate that extreme low or high HDL-C levels can represent a non-previously described source of variation between commercially available LDL-C homogeneous assays.

Comparison of two methods of estimation of low density lipoprotein cholesterol, the direct versus friedewald estimation

Current recommendations of the Adult Treatment Panel and Adolescents Treatment Panel of National Cholesterol Education Program make the low-density lipoprotein cholesterol (LDL-C) levels in serum the basis of classification and management of hypercholesterolemia. A number of direct homogenous assays based on surfactant/solubility principles have evolved in the recent past. This has made LDL-C estimation less cumbersome than the earlier used methods. Here we compared one of the direct homogenous assays with the widely used Friedewald's method of estimation of LDL-C to see the differences and correlation. We used direct homogenous assay kit to estimate serum LDL-C and high-density lipoprotein cholesterol (HDL-C). Serum Triglyceride (TG) and Total Cholesterol (TC) was estimated and using Friedewald's formula LDL-C was calculated. The LDL-C level obtained by both methods in 893 fasting serum samples were compared. The statistical methods used were paired t-test and Pearson's correlation.There was significant difference in the mean LDL-C levels obtained by the two methods at the TG levels <200 mg/dl (p<0.02) and TC levels >150 mg% (p<0.001). The correlation coefficient (r) between Friedewald's and direct assay estimation was 0.88. Friedewald's method classified 23.5 % of patients as high cardiac risk whereas there were 17.58% by direct assay.Both had good correlation even though the serum triglyceride and total cholesterol levels affect the difference in LDL-C estimated by both methods. Taking into account the cost and performance, Friedewald's method is as good or even better for classifying and managing patients.

Reliability of low-density lipoprotein cholesterol, non-high-density lipoprotein cholesterol, and apolipoprotein B measurement

There is little understanding of the reliability of laboratory measurements among clinicians. Low-density lipoprotein cholesterol (LDL-C) measurement is the cornerstone of cardiovascular risk assessment and prevention, but it is fraught with error. Therefore, we have reviewed issues related to accuracy and precision for the measurement of LDL-C and the related markers non-high-density lipoprotein cholesterol (non-HDL-C) and apolipoprotein B. Despite the widespread belief that LDL-C is standardized and reproducible, available data suggest that results can vary significantly as the result of methods from different manufacturers. Similar problems with direct HDL-C assays raise concerns about the reliability of non-HDL-C measurement. The root cause of method-specific bias relates to the ambiguity in the definition of both LDL and HDL, and the heterogeneity of LDL and HDL particle size and composition. Apolipoprotein B appears to provide a more reliable alternative, but assays for it have not been as rigorously tested as direct LDL-C and HDL-C assays.

Ten-year evaluation of homogeneous low-density lipoprotein cholesterol methods developed by Japanese manufacturers. Application of the Centers for Disease Control and Prevention/Cholesterol Reference Method Laboratory Network lipid standardization protocol

AIM

The risk index for atherosclerotic cardiovascular diseases in the Japanese metabolic syndrome-focused health checkup program was changed from total cholesterol (TC) to low-density lipoprotein cholesterol (LDL-C). We discuss the validity of this change with respect to standardization.

METHODS

The beta-quantification procedure of the Centers for Disease Control and Prevention (CDC) uses the LDL-C reference value as a target. Clinical laboratories and commercial manufacturers use homogeneous LDL-C methods for standardization. (A) For clinical laboratories, LDL-C in 648 samples requested from 108 hospitals was analyzed. (B) Manufacturers participated in the CDC/Cholesterol Reference Method Laboratory Network LDL-C standardization protocol. The standardization was conducted with a performance follow-up for the 10-year period from 1998 to 2008 at 2-year intervals, 6 times.

RESULTS

(A) In clinical laboratories, acceptable LDL-C levels within ±4% of the CDC's criteria remained 70.4%, 456 of 648 subjects. Negative maximum bias deviating from the LDL-C target value was -35.8%, -52.5 mg/dL, and positive maximum bias was +24.5%, +32.3 mg/dL. (B) For manufacturers, the standardization achievement rate of the analytical reagent/instrument/calibrator system in the last four standardizations from 2002 to 2008 remained on average 66.6%, far lower than the level required.

CONCLUSIONS

The standardization achievement rate of homogeneous LDL-C methods was much low-er than that of TC. TC should still be used as a risk index for atherosclerotic cardiovascular diseases. The standardization achievement rate of homogeneous LDL-C should be maintained at 100%, at least using samples with normal lipoprotein profiles. The accuracy and specificity of LDL-C should be further improved before practical and clinical use.

Multicentric evaluation of the homogeneous LDL-cholesterol Plus assay: comparison with beta-quantification and Friedewald formula

OBJECTIVES

To evaluate the analytical and clinical performance of a new version of the LDL-C Plus assay and compare it with the beta-quantification (BQ) method in a multicenter study.

DESIGN AND METHODS

Direct LDL-C was measured in 169 fresh pooled samples and in 830 clinical samples with known LDL-C by BQ. The reactivity of lipoproteins and the effect of hemoglobin, bilirubin and chylomicrons (CM) were studied.

RESULTS

Direct LDL-C total imprecision was <2.2%; inaccuracy <+/-2.5% (unaffected by triglycerides up to 9.5 mmol/L); and total error 9.8%. Direct assay reacted with 95%, 50% and 25% of the cholesterol in the LDL, intermediate (IDL) and VLDL fractions, respectively. A significant association was observed with BQ. Icteric samples showed a negative bias and the effect of CM was variable. A positive bias was observed when VLDL-cholesterol/triglyceride ratio was >0.57.

CONCLUSIONS

LDL-C Plus assay represents a valid alternative to BQ for clinical laboratories.

Interference in a homogeneous assay for low-density lipoprotein cholesterol by lipoprotein X

BACKGROUND

Homogeneous assays for cholesterol in low-density lipoprotein (LDL) are currently in wide use for guideline-based diagnosis and monitoring of dyslipaemic or coronary conditions. In some sera from patients with impaired liver function, we measured implausibly low LDL concentrations using a sugar compound-based assay [LDL-cholesterol (LDL-C), Roche Diagnostics]. We investigated whether an interfering factor, possibly associated with cholestasis, is consistently responsible for this disturbance.

METHODS

We compared results of the LDL-C assay in samples with implausible (n=158) and plausible (n=65) LDL concentrations with those of another assay based on two selective detergents (LDLD, Beckman Coulter) and with sequential density ultracentrifugation. We measured total bilirubin, triglycerides, bile acids and lipoprotein X (Lp X) concentrations in samples with the described disturbance and examined the effect of bile salt addition to normal samples.

RESULTS

The LDL-C assay was negatively biased compared to the LDLD assay (bias -0.63 mmol/L) and sequential density ultracentrifugation (bias -0.85 mmol/L) in samples with an implausible lipoprotein profile, but showed good method agreement in all other samples. The bile acid concentration did not correlate with the LDL bias, and addition of bile acids showed no interference with the LDL-C assay. The Lp X concentration correlated with the bias between the LDL-C and LDLD assays (R=0.66, p<0.0001); there was no interference with the LDLD assay, even at high Lp X concentrations.

CONCLUSIONS

We conclude that the LDL-C assay is subject to interference by Lp X and can provide grossly negatively biased results in cholestatic conditions. In such patients, LDL measurement with an assay based on a different method should be performed.

Improved Specificity of a New Homogeneous Assay for LDL-Cholesterol in Serum with Abnormal Lipoproteins

Background: Although a homogeneous assay for serum LDL-cholesterol (LDL-C) has become a routine clinical procedure, problems remain in assay performance characteristics.

Methods: We examined the performance of a recently developed automated homogeneous assay (New-Daiichi assay) for serum LDL-C and compared the results with those obtained by the current homogeneous method (Denka-Seiken assay) or by ultracentrifugation as a control.

Results: The New-Daiichi assay showed satisfactory basic performance characteristics such as reproducibility, linearity, and stability. There was no interference in the assay by various substances examined. The LDL-C values obtained with this method correlated well with those obtained by ultracentrifugation. In samples from patients with obstructive jaundice, both methods detected cholesterol from abnormal lipoproteins (such as lipoprotein-X and -Y), but the New-Daiichi assay was less reactive and more specific for LDL-C.

Conclusion: The new method has improved performance for the accurate measurement of LDL-C in clinical practice.

Pseudohyponatraemia in a patient with obstructive jaundice

INTRODUCTION

Pseudohyponatraemia is uncommonly associated with severe hypercholesterolaemia. Severe hypercholesterolaemia encountered in obstructive jaundice due to an abnormal lipoprotein, lipoprotein X (LpX), may result in pseudohyponatraemia.

CASE REPORT

We report a case of falsely low sodium measurements in a patient with severe hypercholesterolaemia due to obstructive liver disease. The pathophysiology, complications and analytical effects of LpX are briefly discussed.

CONCLUSION

The possibility of pseudohyponatraemia should be considered in severely hypercholesterolaemic samples.

Falsely low LDL-cholesterol concentrations and artifactual undetectable HDL-cholesterol measured by direct methods in a patient with monoclonal paraprotein

INTRODUCTION

Multiple myeloma is a malignant immunoproliferative disorder with lipoprotein abnormalities. We report a case of falsely low concentrations of LDL-cholesterol (LDL-C) and artifactural undetectable HDL-cholesterol (HDL-C) as measured with direct methods in a patient of multiple myeloma with IgGkappa monoclonal gammapathy and significant hyperlipidemia.

CASE REPORT

The patient had HDL-C and LDL-C concentrations in the 0.63-0.71 mmol/l and 2.22-2.36 mmol/l ranges, respectively, as measured by a traditional semi-quantitative electrophoresis method. The observation of falsely low concentrations of LDL-C and artifactural undetectable HDL-C might result in the mismanagement of patients of multiple myeloma with monoclonal gammapathy, because the LDL-C and HDL-C concentrations are positive and negative risk factors of cardiovascular diseases.

CONCLUSIONS

Care must be taken when using the homogenous method for direct measurement of LDL-cholesterol and HDL-cholesterol in patients of multiple myeloma with monoclonal paraprotein.

The effect of VLDL particles on the accuracy of a direct LDL-cholesterol method in type 2 diabetic patients

OBJECTIVE

To assess the accuracy of the direct method LDL-c Plus, in type 2 diabetic patients.LDL-c Plus was measured in 64 consecutive samples of type 2 diabetic patients and compared with betaquantification (BQ), Friedewald's and an alternative formula.

METHODS

LDL-c Plus was also measured in the VLDL (d<1.006 Kg/L) fraction of these samples and in total serum and the VLDL fraction of a phenotype III patient, before and after diluting it with saline or VLDL from normolipidemic subjects.

RESULTS

LDL-c Plus showed a significant, constant bias (-8.5 +/- 5.6%) against BQ which correlated with VLDL-cholesterol/total triglyceride ratio (r = 0.760, p < 0.0005); bias decreased to zero when the ratio increased. In the VLDL fraction of the diabetic patients and the phenotype III patient LDL-c Plus measured 20.7 +/- 11.6% and 56.2% of the cholesterol, respectively. Dilution with saline did not alter the latter percentage, whereas dilution with normolipidemic VLDL reduced it showing that LDL-c Plus recognized cholesterol-enriched particles in the d<1.006 Kg/L. Friedewald's formula also showed a significant, constant bias (-3.1 +/- 6.4%) against BQ, whereas the alternative formula did not (0.5 +/- 6.1%). Both calculations classified patients better than LDL-c Plus did at NCEP cut-off points.

CONCLUSIONS

In type 2 diabetic patients, LDL-c Plus underestimates LDL-c but measures cholesterol associated to IDL particles in the d<1.006 Kg/L fraction. Although LDLc-Plus might be a better cardiovascular risk estimator when well standardized, at the moment, it does not seem to be superior to calculations.

Differential Reactivity of Two Homogeneous LDL-Cholesterol Methods to LDL and VLDL Subfractions, as Demonstrated by Ultracentrifugation and HPLC

Background: The analytical and clinical performance of homogeneous LDL-cholesterol assays has been reported, but their reactions with subfractions of LDL and VLDL have not been described in detail.

Methods: We evaluated reaction selectivity of two homogeneous LDL-cholesterol assays, LDLk (Kyowa Medex) and LDLd (Daiichi Pure Chemical), with ultracentrifugally isolated VLDL and LDL subfractions to identify the lipoprotein particles from which the cholesterol recognized by these assays originates.

Results: The LDLd (y) and LDLk (x) methods correlated highly for whole serum samples: y = 0.986x − 39.5 mg/L (r = 0.966; n = 34). In isolated VLDL, the LDLk and the LDLd methods recovered 17.3% and 23.8% of cholesterol, respectively; but correlation analysis revealed differential reactivity to small and large VLDL particles. For the isolated LDL subfraction of density 1.019–1.040 kg/L, the LDLd method had significantly higher reactivity (95.6–98.7%) than the LDLk (88.4–92.0%). Both methods, however, demonstrated poor recovery (∼50%) for the 1.050–1.063 kg/L fraction, indicating incomplete reactivity with small, dense LDL. Reactivity with lipoprotein(a) was better (71.2–90.8%) for both methods than with small LDL. For intermediate-density lipoprotein (IDL), there was no significant difference in recovery between the two methods (71.7% for LDLk and 68.9% for LDLd), but the LDLk method appeared to be more sensitive to IDL particle size.

Conclusions: The two homogeneous assays for LDL-cholesterol demonstrate only partial reactivity to small, dense LDL and nonspecific reactions to VLDL particles. Modification will be required in the homogeneous methods to obtain LDL-cholesterol values equivalent to those obtained by ultracentrifugation.

Performance of Four Homogeneous Direct Methods for LDL-Cholesterol

Background: Homogeneous LDL-cholesterol methods from Genzyme, Reference Diagnostics, Roche, and Sigma were evaluated for precision, accuracy, and specificity for LDL in the presence of abnormal lipoproteins.

Methods: Each homogeneous method was performed by a Roche/Hitachi 911 according to the vendors’ instructions, and the results were compared with the β-quantification reference method. We measured precision over 20 days using quality-control and frozen serum specimens. Sera from 100 study participants, including 60 with hyperlipidemias, were assayed by each method. Accuracy was evaluated from regression and total error analysis. Specificity was evaluated from the bias (as a percentage) vs concentration of triglycerides.

Results: The total CV was &lt;2% for all methods. Regression slope and intercept (with 95% confidence intervals) were as follows: Genzyme, 0.955 (0.92 to 0.99) and 30.3 (−12 to 73) mg/L; Reference Diagnostics, 0.975 (0.93 to 1.02) and −8 (−63 to 47) mg/L; Roche, 1.067 (1.02 to 1.11) and −101 (−161 to −42) mg/L; and Sigma, 0.964 (0.91 to 1.02) and 164 (89 to 239) mg/L. The percentages of individual results with &gt;12% bias were as follows: Genzyme, 8.0%; Reference Diagnostics, 11.0%; Roche, 10.0%; and Sigma, 30.0%. Total error calculated from mean systematic bias and all-sources random bias was as follows: Genzyme, 12.6%; Reference Diagnostics, 16.5%; Roche, 41.6%; and Sigma, 38.3%. Slopes of bias (as a percentage) vs triglycerides were P &lt;0.001 for all methods except the Roche method, which was P = 0.094.

Conclusions: The evaluated methods show nonspecificity toward abnormal lipoproteins, thus compromising their ability to satisfy the National Cholesterol Education Program goal for a total error of &lt;12%. These homogeneous LDL-cholesterol results do not improve on the performance of LDL-cholesterol calculated by the Friedewald equation at triglyceride concentrations &lt;4000 mg/L.

Methods for Measurement of LDL-Cholesterol: A Critical Assessment of Direct Measurement by Homogeneous Assays versus Calculation

Background: Because LDL-cholesterol (LDL-C) is a modifiable risk for coronary heart disease, its routine measurement is recommended in the evaluation and management of hypercholesterolemia. We critically examine here the new homogeneous assays for direct determination of LDL-C.

Approach: This review relies on published studies and data of the authors using research and routine methods for LDL-C determination. We review experience with methods from their earlier use in lipid research laboratories through the transition to routine clinical testing and the recent development of homogeneous assays. We focus on comparative evaluations and characterizations and the performance of the assays.

Content: Homogeneous assays seem to be able to meet current National Cholesterol Education Program (NCEP) requirements for LDL-C testing for precision (CV &lt;4%) and accuracy (bias &lt;4%), when samples collected from nonfasting individuals are used. In addition, all five currently available assays have been certified by the Cholesterol Reference Methods Laboratory Network. The homogeneous methods also appear to better classify individuals into NCEP cutpoints than the Friedewald calculation. However, the limited evaluations to date raise questions about their reliability and specificity, especially in samples with atypical lipoproteins.

Conclusions: Available evidence supports recommending the homogeneous assays for LDL-C to supplement the Friedewald calculation in those cases where the calculation is known to be unreliable, e.g., triglycerides &gt;4000 mg/L. Before the homogeneous assays can be confidently recommended to replace the calculation in routine practice, more evaluation is needed.