Portal de Periódicos da CAPES

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

2020; Elsevier BV; Volume: 83; Linguagem: Inglês

10.1016/j.ejim.2020.09.009

1879-0828

Federica Piani, Arrigo F.G. Cicero, Claudio Borghi, Sergio D’Addato,

Atherosclerosis and Cardiovascular Diseases

Dear Editor, More than 4 million deaths in Europe are caused by cardiovascular disease (CVD) each year [[1]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: 111-188Crossref PubMed Scopus (2549) Google Scholar]. 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 [[2]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: 35-52Crossref PubMed Scopus (129) Google Scholar]. According to the last European Guidelines, both direct and indirect assessment of LDL-C can be used in clinical practice [[1]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: 111-188Crossref PubMed Scopus (2549) Google Scholar]. 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 [1Mach 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: 111-188Crossref PubMed Scopus (2549) Google Scholar, 3Friedewald 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: 499-502Crossref PubMed Scopus (62) Google Scholar]. 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 [[3]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: 499-502Crossref PubMed Scopus (62) Google Scholar]. 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 [3Friedewald 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: 499-502Crossref PubMed Scopus (62) Google Scholar, 4Agrawal M Spencer HJ Faas FH. Method of LDL Cholesterol Measurement Influences Classification of LDL Cholesterol Treatment Goals.Journal of Investigative Medicine. 2010; 58: 945-949Crossref PubMed Scopus (21) Google Scholar]. Moreover, LDL-C underestimation has also been reported for subjects with low TG levels [[5]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: 236-242Crossref PubMed Scopus (35) Google Scholar]. For this reason, different research groups have developed new equations for LDL-C estimation during the last decades [6Sampson 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: 1-9Crossref Scopus (91) Google Scholar, 7Vujovic 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: 27Crossref PubMed Scopus (57) Google Scholar, 8Martin 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: 2061-2068Crossref PubMed Scopus (353) Google Scholar]. 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) [3Friedewald 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: 499-502Crossref PubMed Scopus (62) Google Scholar, 6Sampson 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: 1-9Crossref Scopus (91) Google Scholar, 7Vujovic 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: 27Crossref PubMed Scopus (57) Google Scholar, 8Martin 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: 2061-2068Crossref PubMed Scopus (353) Google Scholar]. 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 [[6]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: 1-9Crossref Scopus (91) Google Scholar]. 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 [[9]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: 3168-3209Crossref PubMed Scopus (590) Google Scholar]. 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 [[8]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: 2061-2068Crossref PubMed Scopus (353) Google Scholar]. 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 [[7]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: 27Crossref PubMed Scopus (57) Google Scholar]. VF was derived from a Serbian population with TG values 115 mg/dl and low CV risk [[1]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: 111-188Crossref PubMed Scopus (2549) Google Scholar]. 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/dlCorrelationMADConcordance rateUnderestimation rateCorrelationMADConcordance rateUnderestimation rateFriedwaldr = 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. Open table in a new tab 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. 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. 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. The authors wish to thank Daniela Patrono, Rita Mancini and Eric Ramazzotti for the support in collecting the data.