Elsevier

Atherosclerosis

Volume 233, Issue 1, March 2014, Pages 83-90
Atherosclerosis

Clinical impact of direct HDLc and LDLc method bias in hypertriglyceridemia. A simulation study of the EAS-EFLM Collaborative Project Group

https://doi.org/10.1016/j.atherosclerosis.2013.12.016Get rights and content

Highlights

  • Direct HDLc and LDLc measurements show marked bias in hypertriglyceridemic serum.

  • Application of HDL-multipliers to SCORE is unreliable in patients with severe hypertriglyceridemia.

  • LDLc measurements in hypertriglyceridemic serum yield discordant treatment goals.

  • Analytical errors and risk misclassifications are dependent on the method.

Abstract

Background

Despite international standardization programs for LDLc and HDLc measurements, results vary significantly with methods from different manufacturers. We aimed to simulate the impact of analytical error and hypertriglyceridemia on HDLc- and LDLc-based cardiovascular risk classification.

Methods

From the Dutch National EQA-2012 external quality assessment of 200 clinical laboratories, we examined data from normotriglyceridemic (∼1 mmol/l) and hypertriglyceridemic (∼7 mmol/l) serum pools with lipid target values assigned by the Lipid Reference Laboratory in Rotterdam. HDLc and LDLc were measured using direct methods of Abbott, Beckman, Siemens, Roche, Olympus, or Ortho Clinical Diagnostics. We simulated risk reclassification using HDL- and sex-specific SCORE multipliers considering two fictitious moderate-risk patients with initial SCORE 4% (man) and 3% (woman). Classification into high-risk treatment groups (LDLc >2.50 mmol/l) was compared between calculated LDLc and direct LDLc methods.

Results

Overall HDLc measurements in hypertriglyceridemic serum showed negative mean bias of −15%. HDL-multipliers falsely reclassified 70% of women and 43% of men to a high-risk (SCORE >5%) in hypertriglyceridemic serum (P < 0.0001 vs. normotriglyceridemic serum) with method-dependent risk reclassifications. Direct LDLc in hypertriglyceridemic serum showed positive mean bias with Abbott (+16%) and Beckman (+14%) and negative mean bias with Roche (−7%). In hypertriglyceridemic serum, 57% of direct LDLc measurements were above high-risk treatment goal (2.50 mmol/l) vs. 29% of direct LDLc (33% of calculated LDLc) in normotriglyceridemic sera.

Conclusion

LDLc and HDLc measurements are unreliable in severe hypertriglyceridemia, and should be applied with caution in SCORE risk classification and therapeutic strategies.

Introduction

The EAS/ESC guidelines recommend that asymptomatic individuals at high cardiovascular disease (CVD) mortality risk should be identified for statin therapy [1]. For this purpose, risk assessment is performed using the SCORE (Systematic COronary Risk Evaluation) prediction model estimating 10-y risk of CVD mortality, based on gender, age, total cholesterol, systolic blood pressure and smoking status. Recently, the 2011 ESC–EAS guidelines on the management of dyslipidemias have considered the additional impact of high-density lipoprotein cholesterol (HDLc) on CVD risk by displaying 4 separate SCORE charts to 4 different levels of HDLc (mmol/l): 0.8, 1.0, 1.4 and 1.8 [1]. The effects of differing HDLc levels may also be calculated from the classical SCORE using HDL- and sex-specific multipliers according to Descamps et al. [2].

In patients with dyslipidemia, prevention strategies with either lifestyle changes or lipid-lowering agents are primarily targeted by low-density lipoprotein cholesterol (LDLc). The higher the predicted risk, the lower is the recommended LDLc goal and hence the need to initiate statin therapy. The recommended LDLc therapeutic goal is <2.50 mmol/l in high risk individuals (SCORE 5–9%) and <1.80 mmol/l or a 50% reduction in LDLc in very high risk individuals (SCORE ≥10%) [1].

There is a direct relationship between serum LDLc and incidence of CVD. Similarly, there is a strong inverse association between HDLc and CVD, although recent Mendelian randomization studies found no causal relationship between genetically decreased or increased HDLc and the risk of myocardial infarction [3], [4]. However, our concern about including HDLc and LDLc in risk estimation models relates to the potential for analytical error due to imprecision and bias of the lipid measurements. Despite the widespread belief that the calculation or measurement of LDLc or HDLc is standardized and reproducible, results can vary significantly with methods from different manufacturers. In the previous century, the earliest measurements involved ultracentrifugation and precipitation for isolation of LDL and HDL [5]. In the late 1990s, “homogeneous” or “direct” LDLc and HDLc methods have been introduced in the clinical laboratories and largely replaced the older assays [6], [7], [8]. Direct LDLc and HDLc methods are commercially available as ready-to-use reagents, enabling full automation of the measurements, however their bias (deviation from “true” value) is a major point of concern. Discrepant results have been reported among the various direct methods, particularly in hypertriglyceridemic and dyslipidemic samples [9], [10], [11], [12], [13]. This is also evident from large-scale accuracy-based quality surveys organized across different laboratories [14]. Problems with direct HDLc assays also raise concerns about the reliability of calculated LDLc and non-HDLc treatment goals [12]. Poor reliability of these methods relate to the heterogeneity of both LDL and HDL particles [11], [12].

In this study, we aimed to illustrate the potential impact of analytical errors in current LDLc and HDLc measurements on making clinical decisions. A simulation is used here to explore potential CVD risk misclassifications as defined by the SCORE model. Misclassification may occur if a true lipid concentration is within a desirable range, but the reported lipid value is in a high-risk range, or if a true lipid concentration is in a high-risk range but the reported lipid value is in a desirable range [15]. These misclassifications represent a clinically relevant issue because they reflect the practically difficult situation with treatment options: to avoid unnecessary treatment of a patient whose lipid concentration is in a desirable risk category, or failure to treat a patient whose lipid concentration is in a high-risk category, and to distinguish between ‘moderate’ and ‘high-risk’ categories when lipid values are near a cutpoint [15]. Misclassification as defined here is of greatest concern because of its potential impact on the patient and healthcare economics. Using data of the Dutch National EQA-2012 external quality assessment of clinical laboratories, representing all LDLc and HDLc reagent systems used in The Netherlands, we simulated the effects of analytical error and hypertriglyceridemia on HDL-adjusted SCORES and concordance of treatment goals.

Section snippets

Samples

The Dutch external quality assessment (EQA) organizer, the Stichting Kwaliteitsbewaking Medische Laboratoriumdiagnostiek (SKML), runs an accuracy-based EQA scheme for clinical chemistry analytes including lipids and apolipoproteins. Quality of the Dutch EQA program has been described previously [16], [17]. Briefly, serum pools are prepared in an ISO 13485:2003 certified production facility according to CLSI C37-A protocol [18] and value-assigned for total cholesterol, LDLc and HDLc with CDC

HDL-adjusted SCORES

Median HDLc concentrations measured by the various methods did not show any major differences in the NTG serum (Table 2A). However, inter-laboratory imprecision caused a variability of HDLc concentrations reported among the participating laboratories. The application of HDL-multipliers to a fictitious woman (SCORE 3%) or man (SCORE 4%) yielded high-risk categories (SCORE >5%) in a minority of all HDLc measurements (3% and 1%, respectively). According to the HDLc target value measured in the

Discussion

In this study on a Dutch EQA database, representing all LDLc and HDLc methods used in the clinical laboratories in The Netherlands and most other countries, we observed marked deviations from the analytical target value (bias) with measurements in HTG sera. These biases exceeded far beyond the U.S. National Cholesterol Education Program (NCEP) recommendations for LDLc (≤4%) and HDLc (≤5%) [21], [22]. We simulated that application of HDL-multipliers of SCORE may yield different risk

Conclusion

As pointed out in this simulation study and previously by Warnick, Nauck, and Rifai [6], “Laboratories supporting lipid clinics with a high proportion of specimens with atypical lipoproteins could observe discrepant results on certain specimens that might confound treatment decisions.” It is vitally important for clinical laboratories to consider assay reliability and specificity when choosing methods, particularly in dyslipidemic samples. Additionally, more efforts are needed to address

Acknowledgments

We thank Alberico Catapano (EAS President), John Chapman (EAS Past-president), and Sverre Sandberg (EFLM Chair of Scientific Division) for establishing the EAS-EFLM Collaborative Project, intended to produce laboratory testing recommendations and guidelines for management of dyslipidemia and prevention of CVD.

References (50)

  • A. Varbo et al.

    Remnant cholesterol as a causal risk factor for ischemic heart disease

    J Am Coll Cardiol

    (2013)
  • A.D. Sniderman et al.

    Discordance analysis of apolipoprotein B and non-high density lipoprotein cholesterol as markers of cardiovascular risk in the INTERHEART study

    Atherosclerosis

    (2012)
  • L. Masana et al.

    Substituting non-HDL cholesterol with LDL as a guide for lipid-lowering therapy increases the number of patients with indication for therapy

    Atherosclerosis

    (2013)
  • J.G. Robinson et al.

    Meta-analysis of comparison of effectiveness of lowering apolipoprotein B versus low-density lipoprotein cholesterol and non high-density lipoprotein cholesterol for cardiovascular risk reduction in randomized trials

    Am J Cardiol

    (2012)
  • S. Mora et al.

    On-treatment non-high-density lipoprotein cholesterol, apolipoprotein B, triglycerides, and lipid ratios in relation to residual vascular risk after treatment with potent statin therapy: JUPITER (justification for the use of statins in prevention: an intervention trial evaluating rosuvastatin)

    J Am Coll Cardiol

    (2012)
  • M.H. Davidson et al.

    Clinical utility of inflammatory markers and advanced lipoprotein testing: advice from an expert panel of lipid specialists

    J Clin Lipidol

    (2011)
  • D. De Bacquer et al.

    Predictive ability of the SCORE Belgium risk chart for cardiovascular mortality

    Int J Cardiol

    (2010)
  • S. Hirayama et al.

    Small dense LDL: an emerging risk factor for cardiovascular disease

    Clin Chim Acta

    (2012)
  • C.L. Haase et al.

    LCAT, HDL cholesterol and ischemic cardiovascular disease: a Mendelian randomization study of HDL cholesterol in 54,500 individuals

    J Clin Endocrinol Metab

    (2012)
  • G.R. Warnick et al.

    Evolution of methods for measurement of HDL-cholesterol: from ultracentrifugation to homogeneous assays

    Clin Chem

    (2001)
  • M. Nauck et al.

    Methods for measurement of LDL-cholesterol: a critical assessment of direct measurement by homogeneous assays versus calculation

    Clin Chem

    (2002)
  • W.G. Miller et al.

    Seven direct methods for measuring HDL and LDL cholesterol compared with ultracentrifugation reference measurement procedures

    Clin Chem

    (2010)
  • C. Cobbaert et al.

    Reference standardization and triglyceride interference of a new homogeneous HDL-cholesterol assay compared with a former chemical precipitation assay

    Clin Chem

    (1998)
  • H.W. Vesper et al.

    A message from the laboratory community to the National Cholesterol Education Program Adult Treatment Panel IV

    Clin Chem

    (2012)
  • P. Caudill et al.

    Assessment of current National Cholesterol Education Program guidelines for total cholesterol, triglycerides, HDL-cholesterol, and LDL-cholesterol measurements

    Clin Chem

    (1998)
  • Cited by (54)

    • Low-density lipoprotein cholesterol and non-high-density lipoprotein cholesterol measurement in Familial Dysbetalipoproteinemia

      2023, Clinica Chimica Acta
      Citation Excerpt :

      In this study HDL-C was measured using the Siemens Atellica HDL-C assay. Since there is a clear difference between methods and manufacturers for HDL-C assays in dyslipidemic patients, the results of this study should not be considered representative to all HDL-C assays. [21,22]. Second, patient samples were analyzed after different freezing periods.

    • Diabetic dyslipidaemia

      2021, Practical Laboratory Medicine
    • How well do laboratories adhere to recommended guidelines for dyslipidaemia management in Europe? The CArdiac MARker Guideline Uptake in Europe (CAMARGUE) study

      2020, Clinica Chimica Acta
      Citation Excerpt :

      The direct methods also have limitations. External quality assessments of laboratories show important between-method and between-laboratory biases for different manufacturers’ LDLC and HDLC assays [17]. These differences are due to non-specific measurement of other lipoproteins especially in dyslipidaemic samples where atypical lipoproteins are present [18].

    • Quantifying atherogenic lipoproteins for lipid-lowering strategies: Consensus-based recommendations from EAS and EFLM

      2020, Atherosclerosis
      Citation Excerpt :

      Most discordances in dyslipidemic samples are observed at lower concentration ranges of LDLC (<1.8 mmol/L) and HDLC (<1.0 mmol/L) [42]. These errors result in misclassifications with respect to ASCVD risk assessment depending on the type of assay, as observed in accuracy-based external quality assessment surveys of hypertriglyceridemic samples organized across different laboratories [43]. The biases noted in dHDLC measurements affect the calculations of LDLC and non-HDLC, since HDLC is used in the calculations [43].

    View all citing articles on Scopus
    View full text