Advertisement

Is the 2020 Sampson equation the best formula for LDL-C estimation?

  • Federica Piani
    Correspondence
    Corresponding author: Department of Medical and Surgical Sciences, University of Bologna, Via Massarenti 9, Bologna, BO 40138, Italy
    Affiliations
    Department of Medical and Surgical Sciences, University of Bologna, Bologna, Italy
    Search for articles by this author
  • Arrigo F.G. Cicero
    Affiliations
    Department of Medical and Surgical Sciences, University of Bologna, Bologna, Italy
    Search for articles by this author
  • Claudio Borghi
    Affiliations
    Department of Medical and Surgical Sciences, University of Bologna, Bologna, Italy
    Search for articles by this author
  • Sergio D’Addato
    Affiliations
    Department of Medical and Surgical Sciences, University of Bologna, Bologna, Italy
    Search for articles by this author
  • on behalf of the BLIP Study group
    Author Footnotes
    1 BLIP Study group: Claudio Borghi, Marilisa Bove, Arrigo F.G. Cicero, Sergio D'Addato, Federica Fogacci, Marina Giovannini, Elisa Grandi, Rita Mancini, Daniela Patrono, Elisabetta Rizzoli, Arianna Rossolini
  • Author Footnotes
    1 BLIP Study group: Claudio Borghi, Marilisa Bove, Arrigo F.G. Cicero, Sergio D'Addato, Federica Fogacci, Marina Giovannini, Elisa Grandi, Rita Mancini, Daniela Patrono, Elisabetta Rizzoli, Arianna Rossolini
Published:September 22, 2020DOI:https://doi.org/10.1016/j.ejim.2020.09.009

      Keywords

      Dear Editor,
      More than 4 million deaths in Europe are caused by cardiovascular disease (CVD) each year [
      • Mach F
      • Baigent C
      • Catapano AL
      • Koskinas KC
      • Casula M
      • Badimon L
      • et al.
      2019 ESC/EAS Guidelines for the management of dyslipidaemias: lipid modification to reduce cardiovascular risk.
      ]. Low-density lipoprotein-cholesterol (LDL-C) has been demonstrated to be one of the major determinants of atherogenesis and CVD burden, resulting the most important modifiable risk factor for CVD prevention strategies [
      • Mundi S
      • Massaro M
      • Scoditti E
      • Carluccio MA
      • van Hinsbergh VWM
      • Iruela-Arispe ML
      • et al.
      Endothelial permeability, LDL deposition, and cardiovascular risk factors-a review.
      ]. According to the last European Guidelines, both direct and indirect assessment of LDL-C can be used in clinical practice [
      • Mach F
      • Baigent C
      • Catapano AL
      • Koskinas KC
      • Casula M
      • Badimon L
      • et al.
      2019 ESC/EAS Guidelines for the management of dyslipidaemias: lipid modification to reduce cardiovascular risk.
      ]. However, the direct measurement techniques are time consuming, expensive, not fully standardized, and not worldwide available. Thus, the most used method for LDL-C quantification continues to be the 1972 old, landmark Friedewald formula (FF): LDL-C = Total cholesterol – High-Density Lipoprotein cholesterol (HDL-C) – Triglycerides/5 [
      • Mach F
      • Baigent C
      • Catapano AL
      • Koskinas KC
      • Casula M
      • Badimon L
      • et al.
      2019 ESC/EAS Guidelines for the management of dyslipidaemias: lipid modification to reduce cardiovascular risk.
      ,
      • Friedewald WT
      • Levy RI
      • Fredrickson DS.
      Estimation of the Concentration of Low-Density Lipoprotein Cholesterol in Plasma, Without Use of the Preparative Ultracentrifuge.
      ]. FF is based on the two assumptions that, in chylomicron absence, most plasma TG are contained in triglyceride-rich Very-Low Density Lipoproteins (VLDL), and the ratio of TG mass to that of VLDL is relatively constant and roughly equal to 5:1 in normal subjects as well as patients with all types of hyperlipoproteinemia, excluding the rare Type III [
      • Friedewald WT
      • Levy RI
      • Fredrickson DS.
      Estimation of the Concentration of Low-Density Lipoprotein Cholesterol in Plasma, Without Use of the Preparative Ultracentrifuge.
      ]. FF results in reasonably approximated LDL-C estimations, however it suffers from some limitations: it is not applicable for patient with chylomicronemia, overestimates LDL-C values in patients with Type III hypercholesterolemia and leads to LDL-C underestimation in patients with plasma TG ≥ 400 mg/dl [
      • Friedewald WT
      • Levy RI
      • Fredrickson DS.
      Estimation of the Concentration of Low-Density Lipoprotein Cholesterol in Plasma, Without Use of the Preparative Ultracentrifuge.
      ,
      • Agrawal M
      • Spencer HJ
      • Faas FH.
      Method of LDL Cholesterol Measurement Influences Classification of LDL Cholesterol Treatment Goals.
      ]. Moreover, LDL-C underestimation has also been reported for subjects with low TG levels [
      • Tighe DA
      • Ockene IS
      • Reed G
      • Nicolosi R.
      Calculated low density lipoprotein cholesterol levels frequently underestimate directly measured low density lipoprotein cholesterol determinations in patients with serum triglyceride levels ≤4.52 mmol/l: An analysis comparing the LipiDirect® magnetic LDL assay with the Friedewald calculation.
      ]. For this reason, different research groups have developed new equations for LDL-C estimation during the last decades [
      • Sampson M
      • Ling C
      • Sun Q
      • Harb R
      • Ashmaig M
      • Warnick R
      • et al.
      A New Equation for Calculation of Low-Density Lipoprotein Cholesterol in Patients With Normolipidemia and/or Hypertriglyceridemia.
      ,
      • Vujovic A
      • Kotur-Stevuljevic J
      • Spasic S
      • Bujisic N
      • Martinovic J
      • Vujovic M
      • et al.
      Evaluation of different formulas for LDL-C calculation.
      ,
      • Martin SS
      • Blaha MJ
      • Elshazly MB
      • Toth PP
      • Kwiterovich PO
      • Blumenthal RS
      • et al.
      Comparison of a novel method vs the Friedewald equation for estimating low-density lipoprotein cholesterol levels from the standard lipid profile.
      ]. Although promising, the efficacy and potential clinical role of these new formulae have not been thoroughly validated and while one formula should be preferred to another is still being debating.
      We report here a comparison between directly dosed (D-LDL) and calculated (C-LDL) LDL-C in an Italian population of 114,774 adults. We retrospectively analyzed lipid samples of adults with more than 18 years and performed the following statistical analyses: One-way ANOVA, Spearman correlation, Mean Absolute Deviations (MAD), and Concordance in classification between C-LDL and D-LDL through cross-tabulations by LDL-C and TG categories.
      Lipid profiles were anonymously collected by the Laboratorio Metropolitano Unico (LUM) of Bologna and linked to age and sex. TC concentration was measured using Colorimetric Enzymatic CHOD-PAP assays; HDLc level was dosed by Enzymatic colorimetric Immunoinhibition; LDL-C concentration was tested by Colorimetric Enzymatic Selective elimination assay; TG were dosed using Colorimetric Enzymatic GPO-PAP assay. The LDL-C equations compared were: FF, Sampson formula (SF), Martin formula (MF), and Vujovic formula (VF) [
      • Friedewald WT
      • Levy RI
      • Fredrickson DS.
      Estimation of the Concentration of Low-Density Lipoprotein Cholesterol in Plasma, Without Use of the Preparative Ultracentrifuge.
      ,
      • Sampson M
      • Ling C
      • Sun Q
      • Harb R
      • Ashmaig M
      • Warnick R
      • et al.
      A New Equation for Calculation of Low-Density Lipoprotein Cholesterol in Patients With Normolipidemia and/or Hypertriglyceridemia.
      ,
      • Vujovic A
      • Kotur-Stevuljevic J
      • Spasic S
      • Bujisic N
      • Martinovic J
      • Vujovic M
      • et al.
      Evaluation of different formulas for LDL-C calculation.
      ,
      • Martin SS
      • Blaha MJ
      • Elshazly MB
      • Toth PP
      • Kwiterovich PO
      • Blumenthal RS
      • et al.
      Comparison of a novel method vs the Friedewald equation for estimating low-density lipoprotein cholesterol levels from the standard lipid profile.
      ].
      SF represents the most recent LDL-C equation and is based on D-LDL data of 8,656 American adults with a high frequency of hypertriglyceridemia [
      • Sampson M
      • Ling C
      • Sun Q
      • Harb R
      • Ashmaig M
      • Warnick R
      • et al.
      A New Equation for Calculation of Low-Density Lipoprotein Cholesterol in Patients With Normolipidemia and/or Hypertriglyceridemia.
      ]. The resulting formula was: C-LDL = TC/0.948 - HDLc/0.971 – [TG/8.59 + (TG*non-HDLc)/2140 – TG2/16100] – 9.44. Compared to parallel beta-quantification estimates, SF resulted to fit better than both FF and MF. However, no external validating studies have been performed so far. Conversely, MF has been validated in different studies and its use for LDL-C estimation has been also recommended by the last American Guidelines [
      • Grundy SM
      • Stone NJ
      • Bailey AL
      • Beam C
      • Birtcher KK
      • Blumenthal RS
      • et al.
      2018 AHA/ACC/AACVPR/AAPA/ABC/ACPM/ADA/AGS/APhA/ASPC/NLA/PCNA Guideline on the Management of Blood Cholesterol: Executive Summary: A Report of the American College of Cardiology/American Heart Association Task Force on Clinical Practice Guidelines.
      ]. Martin et al developed their formula with a novel calculation method: a Vertical Auto Profile test based on a density-based separation procedure by ultracentrifugation [
      • Martin SS
      • Blaha MJ
      • Elshazly MB
      • Toth PP
      • Kwiterovich PO
      • Blumenthal RS
      • et al.
      Comparison of a novel method vs the Friedewald equation for estimating low-density lipoprotein cholesterol levels from the standard lipid profile.
      ]. A significant novelty was that TG/VLDL-C ratio was not expressed as a fixed term but as a variable depending on both TG and non–HDL-C levels of every sample analyzed. The final equation resulted: C-LDL = non-HDL-C – TG/X, where X was the adjustable variable. This method, referred to as Martin/Hopkins algorithm, was derived using the lipid profiles of 1,350,908 people living in the United States. Finally, we incorporated to our study the equation by Vujovic and colleagues: C-LDL=TC - HDLc - TG/6.85 [
      • Vujovic A
      • Kotur-Stevuljevic J
      • Spasic S
      • Bujisic N
      • Martinovic J
      • Vujovic M
      • et al.
      Evaluation of different formulas for LDL-C calculation.
      ]. VF was derived from a Serbian population with TG values <400 mg/dl and has also been found to be more accurate than FF.
      Due to its large size, it is reasonable to assume that our population included healthy individuals as well as subjects affected by hyperlipidemia, diabetes, or other metabolic disorders. The main results of our analyses are reported in Table 1. Overall, our results showed that SF had the best accuracy, followed by VF and MF. In particular, SF had the highest concordance rate between its C-LDL estimates and the respective D-LDL. This is especially important because the misclassification of the grade of hypercholesterolemia could lead to delays in the treatment with the potential development or progression of CVD. According to the last ESC/EAS guidelines, LDL-C lowering treatment should be initiated in people with D-LDL ≥ 100 mg/dl and moderate cardiovascular (CV) risk and in people with D-LDL > 115 mg/dl and low CV risk [
      • Mach F
      • Baigent C
      • Catapano AL
      • Koskinas KC
      • Casula M
      • Badimon L
      • et al.
      2019 ESC/EAS Guidelines for the management of dyslipidaemias: lipid modification to reduce cardiovascular risk.
      ]. In our sample, 72,993 lipid panels (63.6%) showed a value of D-LDL > 115 mg/dl, underlining the extent of the population potentially eligible for lipid-lowering therapies and the importance of a precise estimation of LDL-C. In fact, under-treatment caused by the under-estimation of LDL-C could play a crucial role in the failure of primary CVD prevention strategies. Our data (Table 1) showed that SF and VF had the lowest underestimation rates in both the hypertriglyceridemic (TG ≥ 250 mg/dl) and normotriglyceridemic groups. In other words, if we used SF or VF, we could prevent D-LDL underestimation of 10,000 more people than using FF. It appears evident how this could potentially affect the correct management of lipid-lowering therapies and the success of CVD prevention.
      Table 1Principal results of the comparison between Friedewald, Martin, Sampson, and Vujovic LDL-C formulae.
      TOTALTG ≥ 250 mg/dl
      CorrelationMADConcordance rateUnderestimation rateCorrelationMADConcordance rateUnderestimation rate
      Friedwaldr = 0.949, p<0.0019.7 ± 9.676.8% (88161/114774)20.6% (23655/114774)r = 0.931, p<0.00134.2 ± 22.649.7% (2517/5061)34.3% (1735/5061)
      Martinr = 0.973, p<0.0017.7 ± 8.383.8% (92483/114774)18.1% (20744/114774)r = 0.964, p<0.00112.2 ± 27.173.1% (3701/5061)26.8% (1354/5061)
      Sampsonr = 0.976, p<0.0017.7 ± 6.485.7% (98310/114774)10.7% (12290/114774)r = 0.952, p<0.00115.6 ± 15.985.8% (4344/5061)13.2% (666/5061)
      Vujovicr = 0.968, p<0.0018.0 ± 7.785.1% (97711/114774)10.4% (11992/114774)r = 0.958, p<0.00123.3 ± 19.468.8% (3480/5061)31% (1571/5061)
      For the concordance rate between C-LDL and D-LDL, C-LDL values were grouped by TG level (< 250 and ≥ 250 mg/dl) and by the 5 LDL-C targets suggested by the last European guidelines (≤55, 56-69, 70-99, 100-115, >115 mg/dl). MAD values are expressed as mean ± standard deviation. Abbreviations: r=correlation coefficient; MAD= Mean Absolute Deviations.
      In conclusion, although not yet validated SF appears promising to improve LDL-C estimation accuracy and thus CVD prevention strategies efficacy. Moreover, this new equation is freely available to download and could be easily implemented by laboratories. Further studies are warranted to validate this new formula but seems inevitable that FF time will soon be over.

      Author contributions

      F.P. and S.D. conceived the ideas; BLIP study group analyzed the data; F.P., A.F.G.C., and C.B. led the writing.

      Declaration of Competing Interest

      The authors reported no funding sources or conflicts of interest for this study.
      The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

      Acknowledgements

      The authors wish to thank Daniela Patrono, Rita Mancini and Eric Ramazzotti for the support in collecting the data.

      References

        • Mach F
        • Baigent C
        • Catapano AL
        • Koskinas KC
        • Casula M
        • Badimon L
        • et al.
        2019 ESC/EAS Guidelines for the management of dyslipidaemias: lipid modification to reduce cardiovascular risk.
        Eur Heart J. 2020; 41: 111-188
        • Mundi S
        • Massaro M
        • Scoditti E
        • Carluccio MA
        • van Hinsbergh VWM
        • Iruela-Arispe ML
        • et al.
        Endothelial permeability, LDL deposition, and cardiovascular risk factors-a review.
        Cardiovasc Res. 2018; 114: 35-52
        • Friedewald WT
        • Levy RI
        • Fredrickson DS.
        Estimation of the Concentration of Low-Density Lipoprotein Cholesterol in Plasma, Without Use of the Preparative Ultracentrifuge.
        Clinical Chemistry. 1972; 18: 499-502
        • Agrawal M
        • Spencer HJ
        • Faas FH.
        Method of LDL Cholesterol Measurement Influences Classification of LDL Cholesterol Treatment Goals.
        Journal of Investigative Medicine. 2010; 58: 945-949
        • Tighe DA
        • Ockene IS
        • Reed G
        • Nicolosi R.
        Calculated low density lipoprotein cholesterol levels frequently underestimate directly measured low density lipoprotein cholesterol determinations in patients with serum triglyceride levels ≤4.52 mmol/l: An analysis comparing the LipiDirect® magnetic LDL assay with the Friedewald calculation.
        Clinica Chimica Acta. 2006; 365: 236-242
        • Sampson M
        • Ling C
        • Sun Q
        • Harb R
        • Ashmaig M
        • Warnick R
        • et al.
        A New Equation for Calculation of Low-Density Lipoprotein Cholesterol in Patients With Normolipidemia and/or Hypertriglyceridemia.
        JAMA Cardiol. 2020; 5: 1-9
        • Vujovic A
        • Kotur-Stevuljevic J
        • Spasic S
        • Bujisic N
        • Martinovic J
        • Vujovic M
        • et al.
        Evaluation of different formulas for LDL-C calculation.
        Lipids Health Dis. 2010; 9: 27
        • Martin SS
        • Blaha MJ
        • Elshazly MB
        • Toth PP
        • Kwiterovich PO
        • Blumenthal RS
        • et al.
        Comparison of a novel method vs the Friedewald equation for estimating low-density lipoprotein cholesterol levels from the standard lipid profile.
        JAMA. 2013; 310: 2061-2068
        • Grundy SM
        • Stone NJ
        • Bailey AL
        • Beam C
        • Birtcher KK
        • Blumenthal RS
        • et al.
        2018 AHA/ACC/AACVPR/AAPA/ABC/ACPM/ADA/AGS/APhA/ASPC/NLA/PCNA Guideline on the Management of Blood Cholesterol: Executive Summary: A Report of the American College of Cardiology/American Heart Association Task Force on Clinical Practice Guidelines.
        J Am Coll Cardiol. 2019; 73: 3168-3209