Validation of the Martin method to estimate low-density lipoprotein cholesterol concentrations in Japanese populations and a modified method for laboratory information system application

Abstract

Objectives

High concentrations of low-density lipoprotein cholesterol (LDL-C) are a risk factor for cardiovascular disease. We validated the efficacy of the Martin method is useful in the estimation of LDL-C concentrations was validated in Japanese populations and derived a modified Martin method for easy laboratory information system applications.

Methods

We created 3 subject groups, including 2664 health check-up participants registered with the Resource Center for Health Science, 29,806 clinical patients (A) in the Gifu University Hospital, and 113,716 clinical patients (B) in the Fujita Health University Hospital. Each method to estimate serum LDL-C concentrations (Friedewald formula, Martin method and modified Martin method) was validated by correlation analysis with serum LDL-C concentrations measured using a direct method

Results

The correlation coefficients with the direct method in terms of the Friedewald formula, Martin method, and modified Martin method were 0.963, 0.972 and 0.970 in the health check-up participants; 0.946, 0.962 and 0.961 in clinical patients A; and 0.961, 0.979 and 0.978 in clinical patients B, respectively. Concordance ratios with using the direct method in the Friedewald formula, Martin method and modified Martin method were 82.8%, 85.5% and 85.3% in the health check-up participants; 76.4%, 80.5% and 80.2% in clinical patients A; and 76.1%, 82.6% and 82.6% in clinical patients B, respectively.

Conclusion

Our results show that the Martin and modified Martin methods display good performance in terms of the estimation of LDL-C concentrations among triglyceride concentrations of a wide range, and they may thus be useful for estimating LDL-C concentrations.

Keywords

Lipids < analytes analytical systems < laboratory methods

Introduction

High blood cholesterol concentrations are critical risk factors for cardiovascular disease (CVD). In particular, low-density lipoprotein cholesterol (LDL-C) concentrations are adopted for risk assessment and treatment goals in principal guidelines such as those of the European Society of Cardiology/European Atherosclerosis Society (2019 ESC/EAS Guideline) and American College of Cardiology/(2019 ACC/AHA Guideline).^1,2

The beta-quantification (BQ) method, which is the international standard method for serum or plasma LDL-C level measurements,³ requires ultracentrifugation and is not suitable for routine clinical assays. Moreover, regulations allow the use of LDL-C concentrations estimated by the Friedewald formula only for patients with triglyceride (TG) values of <400 mg/dL. Furthermore, previous reports have demonstrated that LDL-C values estimated by the Friedewald formula in patients with TG values of <400 mg/dL deviated from those measured by the BQ or direct methods.^4–14

Several studies have reported unique methods for estimating LDL-C developed for the improvement of consistency with the LDL-C concentrations established by the BQ method.^4–8 Martin reported a new method to estimate LDL-C concentrations with variable TG to very-low-density lipoprotein cholesterol (VLDL-C) ratios by TG concentrations and non–high-density lipoprotein cholesterol (HDL-C) concentrations.¹⁵ The Martin method was revealed to be a superior approach to the Friedewald formula by several studies in estimating LDL-C concentrations,^16–19 whereas only a few reports have demonstrated its use in Japanese populations .^20,21 Previous studies involving Japanese participants have been validated using under fasting conditions but have not been confirmed in populations who participated without fasting, including clinical patients. Furthermore, the Martin method is not suitable for use in the laboratory information system (LIS) because it requires selection of TG concentrations to VLDL-C ratio from 180 patterns by TG and non–HDL-C concentrations.¹⁵

This study aimed to revalidate the suitability of the Martin method when applied to samples obtained from Japanese participants under fasting conditions who underwent health check-ups, and samples obtained from clinical patients under fasting and non-fasting conditions analysed using two different protocols for LDL-C measurement. We also sought to validate the utility of a novel, modified Martin method, designed for LIS-related applications.

Materials and methods

Study populations

This retrospective study was conducted using three lipid profile databases of 2664 health check-up participants in the Resource Center for Health Science, 29,806 patients in the Gifu University Hospital (clinical patients A), and 113,716 patients in the Fujita Health University Hospital (clinical patients B), measuring TG, total cholesterol (TC), HDL-C, and LDL-C by the direct method (LDL-C-D). Enrolled participants were categorized into two groups to analyse protocols involving different principle reagents for LDL-C-D measurement. Table 1 presents the study subject characteristics. Sera from health check-up participants were collected during fasting; sera from the patients in the other groups were collected at any other time regardless of fasting. The serum TG, TC, HDL-C and LDL-C-D concentrations were measured immediately after blood sampling in both groups. This study was approved by the ethics boards of the Gifu University (2021-A059) and the Fujita Health University (FM21-093). Informed consent was obtained by provision of an opt-out option on each institution’s website.

Table 1.

Characteristics of health check-up participants and clinical patients.

	Health check-up participants	Clinical patients A	Clinical patients B
Sex; Male/Female	1586/1078	16466/13343	56109/57607
Age (years)	48 (41–55)	65 (54–73)	65 (49–74)
Triglyceride (mg/dL)	86 (61–127)	116 (82–166)	109 (76–159)
Total cholesterol (mg/dL)	196 (175–219)	116 (82–166)	187 (161–214)
HDL-C (mg/dL)	55 (46–65)	52 (42–64)	57 (46–70)
LDL-C; direct method (mg/dL)	120 (100–142)	106 (86–126)	106 (85–129)
Non–HDL-C (mg/dL)	139 (118–163)	133 (111–156)	126 (103–152)

Note: HDL-C, high-density cholesterol; LDL-C, low-density cholesterol; Non–HDL-C, TC – HDL-C Data were expressed as median (quartile range).

Lipid measurement and calculation

Serum TG level measurements were performed by employing the free-glycerol elimination lipoprotein-lipase and glycerol kinase method using ‘Cholestest TG’ (SEKISUI Medical Co., Inc.) in health check-up participants and clinical patients A and ‘DETAMINAR L TG Ⅱ’ (Hitachi Chemical Diagnostics Systems Co., Ltd.) in clinical patients B. Serum TC level measurements were performed by employing the cholesterol esterase- and cholesterol oxidase-related method using ‘Cholestest CHO’ (SEKISUI Medical Co., Inc.) in health check-up participants and clinical patients A and ‘DETAMINAR L TC Ⅱ’ (Hitachi Chemical Diagnostics Systems Co., Ltd.) in clinical patients B. Serum HDL-C and LDL-C concentrations were measured by employing the homogeneous enzymatic assay method using ‘Cholestest N HDL’ and ‘Cholestest LDL’ in health check-up participants and clinical patients A and ‘MetaboLead HDL-C’ and ‘MetaboLead LDL-C’ (Hitachi Chemical Diagnostics Systems Co., Ltd.) in clinical patients B. The LDL-C concentrations were estimated by the Friedewald formula (LDL-C-F; [TC] − [HDL-C] − [TG]/5) and the Martin method (LDL-C-M; [TC] − [HDL-C] − [TG]/variable factor).²² The TG to VLDL-C ratios were calculated by (TG)/([TC] − [HDL-C] − [LDL-C-D]).

Derivation of the modified Martin’s LDL-C estimation method for LIS applications

We derived the formulae by modifying the Martin method (LDL-C-mM) to estimate the LDL-C values for LIS applications. We applied regression models by the least square method to estimate TG to VLDL-C ratios using the 360-cell table of the median TG to VLDL-C ratios by non–HDL-C and TG, as reported by Martin SS. We created the method applying 6-type equations for TG <400 mg/dL and 6-type factors for TG>400mg/ml, depending on the non–HDL-C level (Table 2). If TG values are ≥400 mg/dL, the Martin and modified Martin methods use the same formulae.

Table 2.

The factors for TG to VLDL‐C ratios derived by modified Martin method.

		TG, mg/dL
		7–400	>400
Non–HDL-C, mg/dL	<100	0.020* (TG) + 3.38	11.9
	100–129	0.016* (TG) + 3.40	10.0
	130–159	0.014* (TG) + 3.41	8.8
	160–189	0.012* (TG) + 3.39	8.1
	190–219	0.010* (TG) + 3.42	7.5
	≥220	0.009* (TG) + 3.41	6.7

Note: LDL-C, low-density lipoprotein cholesterol; Non–HDL-C, (total cholesterol) – (high-density lipoprotein cholesterol); TG, triglyceride.

Statistics

We estimated the suitability of LDL-C evaluations using Spearman’s rank correlation coefficient, Deming regression analysis,²³ and Bland–Altman analysis.²⁴ The concordance of the LDL-C strata between each method was assessed using the Kappa coefficient method. All statistical analyses were performed using JMP 13.2.1 and R 4.1.0.^25,26

Results

The distribution of TG to VLDL-C ratios

Figure 1 shows the distribution of TG to VLDL-C ratios calculated by LDL-C-D in health check-up participants, as well as in clinical patients A and B. In health check-up participants and clinical patients A and B, the median (25^th–75^th percentile) of TG to VLDL-C ratios was 5.50 (4.00–6.78), 4.71 (3.73–5.89) and 6.22 (5.10–7.61), respectively. Because the Friedewald formula estimates LDL-C by fixing the TG to VLDL-C ratio at 5.00, variations in the TG to VLDL-C ratio are sources of error in the LDL-C estimation. The TG to VLDL-C ratio contributions less than 4.50, causing an underestimation compared with LDL-C-F, were 34.1%, 44.2% and 13.8%, respectively. Similarly, the TG to VLDL-C ratio contributions over 5.50, causing an overestimation compared with LDL-C-F, were 46.4%, 32.1% and 65.9%, respectively.

Figure 1.

TG to VLDL-C ratio distributions calculated by TG/(TC – HDL-C – LDL-C-D) in health check-up participants (a), clinical patients A (b), and B (c).

Correlations and biases between LDL-C-D and estimated LDL-C

We performed correlation and Bland–Altman analyses of LDL-C-F, LDL-C-M, and LDL-C-mM with LDL-C-D as reference for health check-up participants, clinical patients A, and clinical patients B, respectively. In all groups, we observed higher Spearman’s rank correlation coefficients (ρ) between LDL-C-D and LDL-C-M than between LDL-C-D and LDL-C-F. The same trend was observed for TG <400 mg/dL with similar results for LDL-C-M and LDL-C-mM. In TG ≥400 mg/dL, representing the same results of LDL-C-M and LDL-C-mM, Spearman’s rank correlation coefficients for LDL-C-M were slightly inferior to those for LDL-C-F in health check-up participants and clinical patients A, whereas the correlation coefficients for LDL-C-M and LDL-C-mM were slightly superior in clinical patients B. In all subject groups, the differences between LDL-C-M and LDL-C-D were smaller than those between LDL-C-F and LDL-C-D. A wide variance could be detected in several cases for TG ≥400 mg/dL when LDL-C-D was <100 mg/dL. LDL-C-M and LDL-C-mM limits of agreement (LoA), calculated by the mean ±2 × standard deviation of biases between estimated LDL-C and LDL-C-D concentrations, converged more than those of LDL-C-F in all subject groups (Table 3).

Table 3.

The performance on method of LDL-C estimation reference to the direct method.

	ρ*	Slope†	Intercept†	95% LoA‡
Health check-up participants
All cases (N = 2664)
LDL-C-F	.963	0.97	2.0	−18.8, 15.6
LDL-C-M	.972	0.94	5.9	−15.3, 13.4
LDL-C-mM	.970	0.93	7.3	−14.5, 14.0
TG <400 mg/dL (N = 2638)
LDL-C-F	.964	0.95	4.9	−16.5, 13.9
LDL-C-M	.974	0.94	5.8	−14.4, 11.7
LDL-C-mM	.972	0.93	7.2	−14.7, 12.5
TG ≥400 mg/dL (N = 26)
LDL-C-F	0.864	1.68	−99.2	−92.4, 32.4
LDL-C-M and LDL-C-mM	0.842	1.12	1.1	−25.4, 51.9
Clinical patients A
All cases (N = 29806)
LDL-C-F	0.946	1.04	−4.2	−21.6, 22.5
LDL-C-M	0.962	1.00	4.1	−13.2, 20.6
LDL-C-mM	0.961	0.99	4.3	−13.6, 20.8
TG <400 mg/dL (N = 29323)
LDL-C-F	0.950	1.02	−0.7	−17.5, 19.3
LDL-C-M	0.967	0.99	4.6	−11.5, 18.5
LDL-C-mM	0.965	0.99	4.8	−11.9, 18.7
TG ≥400 mg/dL (N = 483)
LDL-C-F	0.812	1.56	−81.9	−108.2, 57.6
LDL-C-M and LDL-C-mM	0.807	1.18	−1.8	−42.3, 75.2
Clinical patients B
All cases (N = 113716)
LDL-C-F	0.961	1.05	−11.3	−31.9, 20.9
LDL-C-M	0.979	1.02	−4.6	−24.9, 19.9
LDL-C-mM	0.978	1.02	−4.2	−24.9, 20.0
TG <400 mg/dL (N = 111633)
LDL-C-F	0.962	1.01	−6.3	−23.3, 13.4
LDL-C-M	0.972	1.00	−2.3	−18.3, 12.9
LDL-C-mM	0.972	0.99	−1.9	−18.4, 13.1
TG ≥400 mg/dL (N = 2083)
LDL-C-F	0.817	2.55	−205.4	−163.7, 95.4
LDL-C-M and LDL-C-mM	0.826	2.18	−123.4	−111.1, 124.7

*: Spearman’s rank correlation coefficient. †: Estimated by Deming regression analysis. ‡: Limit of agreement estimated by Bland–Altman analysis. TG; triglyceride, LDL-C-D, serum low-density lipoprotein cholesterol concentrations measured by direct method; LDL-C-F, serum low-density lipoprotein cholesterol concentrations estimated by Friedewald formula; LDL-C-M, serum low-density lipoprotein cholesterol concentrations estimated by Martin method; serum low-density lipoprotein cholesterol concentrations estimated by novel modified Martin method.

LDL-C concentrations strata for intervention strategies on ESC/EAS guidelines

The overall concordance ratios with ESC/EAS guideline classification by LDL-C-D in health check-up participants were 82.8%, 85.5% and 85.3% in LDL-C-F, LDL-C-M and LDL-C-mM, respectively. Moreover, those in clinical patients A were 76.4%, 80.5% and 80.2% and those in clinical patients B were 76.1%, 82.6% and 82.6%, respectively. The concordance ratios for LDL-C-F, LDL-C-M and LDL-C-mM among subjects with TG ˂400 mg/dL were 83.2%, 85.7% and 85.6% in health check-up participants, 77.0%, 81.1% and 81.0% in clinical patients A, and 77.0%, 83.0% and 83.0% for clinical patients B, respectively. The concordance ratios for LDL-C-F and LDL-C-M among subjects with TG 400 mg/dL or higher were 38.4% and 53.8% in health check-up participants, 39.5% and 42.0% in clinical patients A, and 27.6% and 59.1% in clinical patients B. In all subject groups, Kappa coefficients between LDL-C-D and LDL-C-M or LDL-C-mM were higher than those between LDL-C-D and LDL-C-F (Figure 2).

Figure 2.

Concordance of LDL low-density lipoprotein-C concentrations strata for intervention strategies on ESC/EAS guidelines in health check-up participants (a), as well as clinical patients A (b) and B (c) referring to LDL-C-D low-density lipoprotein-C measured using the direct method.

Discussion

The present study evaluated the Martin method for serum LDL-C level estimation by correlation with serum LDL-C concentrations measured by the direct method in health check-up participants and clinical patients. Furthermore, we derived a novel modified Martin method, which could be applied to LIS, and evaluated its performance. This study has several limitations. First, the reference method of this study is the direct method and not the gold standard BQ method. Second, we used only reagents manufactured by SEKISUI Medical Co., Inc. and Hitachi Chemical Diagnostics Systems Co., Ltd. in both subject groups; thus, a potential difference in reagent variety was not addressed. However, we found that both LDL-C-M and LDL-C-mM correlated better with LDL-C-D than LDL-C-F, and higher concordance in the LDL-C strata could be observed on the ESC/EAS guideline.

The Friedewald formula for LDL-C estimation also has its limitations because it should be used in patients with TG values of ˂400 mg/dL and fasting samples. Previous studies showed that LDL-C values estimated using the Friedewald formula deviated from those obtained by the EQ and the direct methods. Various alternative methods have been established to estimate LDC-C and showed better correlations with the BQ method or direct method than the Friedewald formula.^4–8 Among them, the Martin method is an innovative and highly reliable method because it uses variable TG to VLDL-C ratios obtained by large-scale lipid profile datasets measured by the BQ method.¹⁵ The benefits of this method have been demonstrated in several previous studies that have shown it to be highly compatible with the BQ method and the direct methods.^16–19 Moreover, the Martin method requires choosing one of the 180-pattern of factors from the TG and non–HDL-C values, which is not practical for LIS application. We modified the Martin method into the 12-pattern formula based on the TG and non–HDL-C values to simplify LIS implementation. The 12-pattern formula could be simpler to implement in LIS than the 180-pattern formula. The performance of this approach may vary with different study populations, whereas only few studies have reported Japanese subject-related results. Similar to previous studies, our results showed that the Martin method correlates better with the direct method than the Friedewald formula.^16–19 This study yielded better correlations and more convergent LoA in LDL-C-M and LDL-C-mM compared with LDL-C-F not only in the health check-up participant and clinical patient groups but also in clinical patients with different reagents. Because sera from clinical patients were sampled any time regardless of fasting, LDL-C-M and LDL-C-mM might be superior to estimate LDL-C concentrations with or without fasting. However, all estimates tended to be higher than LDL-C-D with TG ≥400 mg/dL. Previous reports showed that the two methods have superior correlation, but that errors occur via differences in reactivity in CM, VLDL and IDL fragments.²⁷ Our results indicated that clinical patients A and B using different reagents had comparable correlation coefficients, especially in TG <400 mg/dL patients. However, there were several cases with estimated LDL-C values by LDL-C-F, LDL-C-M and LDL-C-mM substantially higher than LDL-C-D measurement values in clinical patient B with LDL-C-D < 100 mg/dL. These cases were presumably affected by the presence of non–LDL-C enriched in TG such as VLDL-C, which cannot be included in the measurement of the direct method in clinical patients B. In addition to the difference in the subject populations, this might be caused by the measurement reagent differences for the direct method, representing the reference method.²⁸ For TG ≥400 mg/dL, LDL-C-M and LDL-C-mM might still be better than F, although this aspect might still require further improvement.

The Friedewald formula reportedly underestimates the CVD risk.^10,29 We obtained similar results in this study. Concordance ratios of LDL-C level strata for intervention strategies on ESC/EAS guideline with the direct Martin method in all subject groups were higher than those by the Friedewald formula, with improvement of concordance ratios by reduction of the underestimation. Therefore, the Martin method could be more useful to assess the CVD risk than the Friedewald formula. Furthermore, this study shows that not only the former but also the modified Martin method derived by us for LIS application yield better correlation and concordance of LDL-C strata on ESC/EAS guideline with the direct method than the Friedewald formula. LDL-C-mM could be potentially applied for LIS easier than the former Martin method as it relies on selecting the formula from 12 patterns by TG and non–HDL-C concentrations. Therefore, our results indicate that not only the Martin method but also the modified Martin method are useful for estimating serum LDL-C concentrations, especially for LIS-related applications as a pre-treatment examination.

Conclusion

We derived the modified Martin method for simple LIS application and the validation of the Martin and the modified Martin methods in health check-up participants and clinical patients. Both the Martin and the modified Martin methods resulted in better correlations with the direct method than the Friedewald formula, hence these methods are valid for LDL-C concentrations estimation. In particular, the modified Martin method could be potentially useful for LDL-C estimation on LIS-related applications.

Supplemental Material

Supplemental Material - Validation of the Martin method to estimate low-density lipoprotein cholesterol concentrations in Japanese populations and a modified method for laboratory information system application

Supplemental Material for Validation of the Martin method to estimate low-density lipoprotein cholesterol concentrations in Japanese populations and a modified method for laboratory information system application by Hidekazu Ishida, Yasuko Yamamoto, Midori Saito, Yuya Ishihara, Takashi Fujita, Mariko Ishida, Yohei Kato, Yuzuru Nohisa, Hidetoshi Matsunami, Masao Takemura, Tadayoshi Hata, Hiroyasu Ito and Kuniaki Saito in Annals of Clinical Biochemistry

Supplemental Material

Supplemental Material - Validation of the Martin method to estimate low-density lipoprotein cholesterol concentrations in Japanese populations and a modified method for laboratory information system application

Supplemental Material

Supplemental Material - Validation of the Martin method to estimate low-density lipoprotein cholesterol concentrations in Japanese populations and a modified method for laboratory information system application

Footnotes

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This research was supported by Fujita Health University Graduate School.

Ethical approval

Gifu University Hospital Clinical Research Ethics Board (2021-A059) and the Fujita Health University Research Ethics Board (FM21-093) approved this study.

Guarantor

KS.

Contributorship

HI, YY and TM researched literature and conceived the study. HI gained ethical approval, data analysis, and wrote the first draft of the manuscript. HI, YY and TM involved in the investigation. MS, YI, TF, MI, YK, YN, HM, MT and TH involved in the supply of clinical laboratory data. TH, HI and KS involved as supervision and project administration. All authors reviewed and edited the manuscript and approved the final version of the manuscript.

ORCID iDs

Hidekazu Ishida

Yasuko Yamamoto

Supplemental material

Supplemental material for this article is available online.

References

Mach

Baigent

Catapano

, et al. 2019 ESC/EAS Guidelines for the management of dyslipidaemias: lipid modification to reduce cardiovascular risk. Eur Heart J 2020; 41: 111–188. doi:10.1093/eurheartj/ehz455

Arnett

Blumenthal

Albert

, et al. 2019 ACC/AHA guideline on the primary prevention of cardiovascular disease: executive summary: a report of the american college of cardiology/american heart association task force on clinical practice guidelines. J Am Coll Cardiol 2019; 74: 1376–1414. doi:10.1016/j.jacc.2019.03.009

Bachorik

Ross

. National cholesterol education Program recommendations for measurement of low-density lipoprotein cholesterol: executive summary. The national cholesterol education program working group on lipoprotein measurement. Clin Chem 1995; 41: 1414–1420

Hattori

Suzuki

Tsushima

, et al. Development of approximate formula for LDL-chol, LDL-apo B and LDL-chol/LDL-apo B as indices of hyperapobetalipoproteinemia and small dense LDL. Atherosclerosis 1998; 138: 289–299. doi:10.1016/s0021-9150(98)00034-3

Anandaraja

Narang

Godeswar

, et al. Low-density lipoprotein cholesterol estimation by a new formula in Indian population. Int J Cardiol 2005; 102: 117–120. doi:10.1016/j.ijcard.2004.05.009

Ahmadi

Boroumand

Gohari-Moghaddam

, et al. The impact of low serum triglyceride on LDL-cholesterol estimation. Arch Iran Med 2008; 11: 318–321

Chen

Zhang

Pan

, et al. A modified formula for calculating low-density lipoprotein cholesterol values. Lipids Health Dis 2010; 9: 52. doi:10.1186/1476-511X-9-52

de Cordova

CMM

de Cordova

. A new accurate, simple formula for LDL-cholesterol estimation based on directly measured blood lipids from a large cohort. Ann Clin Biochem Int J Lab Med 2012; 50: 13–19. doi:10.1258/acb.2012.011259

Choi

Shim

J-S

Lee

, et al. Comparison of formulas for calculating Low-density lipoprotein cholesterol in general population and high-risk patients with cardiovascular disease. Korean Circ J 2016; 46: 688. doi:10.4070/kcj.2016.46.5.688

10.

Bansal

Kaur

. Does friedewald formula underestimate the risk of ischemic heart disease? Indian J Clin Biochem 2014; 29: 496–500. doi:10.1007/s12291-013-0392-2

11.

Krishnaveni

. Assessing the validity of Friedewald’s formula and Anandraja’s formula for serum L DL -Cholesterol calculation. J Clin Diagn Res 2015; 9:12: BC01–BC04. doi:10.7860/jcdr/2015/16850.6870

12.

Meeusen

Lueke

Jaffe

, et al. Validation of a proposed novel equation for estimating LDL cholesterol. Clin Chem 2014; 60: 1519–1523. doi:10.1373/clinchem.2014.227710

13.

Mora

Rifai

Buring

, et al. Comparison of LDL cholesterol concentrations by friedewald calculation and direct measurement in relation to cardiovascular events in 27 331 women. Clin Chem 2009; 55: 888–894. doi:10.1373/clinchem.2008.117929

14.

Sibal

Neely

RDG

Jones

, et al. Friedewald equation underestimates low-density lipoprotein cholesterol at low concentrations in young people with and without Type 1 diabetes. Diabetic Med 2010; 27: 37–45. doi:10.1111/j.1464-5491.2009.02888.x

15.

Martin

S. S.

Blaha

M. J.

Elshazly

M. B.

, 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. doi:10.1001/jama.2013.280532

16.

Martin

S. S.

Blaha

M. J.

Elshazly

M. B.

, et al. Friedewald-Estimated versus directly measured low-density lipoprotein cholesterol and treatment implications. J Am Coll Cardiol 2013; 62: 732–739. doi:10.1016/j.jacc.2013.01.079

17.

Mehta

Reyes-Rodríguez

Bello-Chavolla

, et al. Performance of LDL-C calculated with Martin’s formula compared to the Friedewald equation in familial combined hyperlipidemia. Atherosclerosis 2018; 277: 204–210. doi:10.1016/j.atherosclerosis.2018.06.868

18.

Petridou

Anagnostopoulos

. Validation of the novel Martin method for LDL cholesterol estimation. Clin Chim Acta 2019; 496: 68–75. doi:10.1016/j.cca.2019.06.023

19.

Cartier

L-J

St-Coeur

Robin

, et al. Impact of the Martin/Hopkins modified equation for estimating LDL-C on lipid target attainment in a high risk patient population. Clin Biochem 2020; 76: 35–37. doi:10.1016/j.clinbiochem.2019.12.002

20.

Sonoda

Takumi

Miyata

, et al. Validity of a novel method for estimating low-density lipoprotein cholesterol levels in cardiovascular disease patients treated with Statins. J Atheroscler Thromb 2018; 25: 643–652. doi:10.5551/jat.44396

21.

Ikezaki

Fisher

Lim

, et al. Direct versus calculated LDL Cholesterol and C-Reactive protein in cardiovascular disease risk assessment in the Framingham offspring study. Clin Chem 2019; 65: 1102–1114. doi:10.1373/clinchem.2019.304600

22.

University JH . LDL Calculator JAMA, 2013; 310: 2061–8. https://www.ldlcalculator.com

23.

Deming

. Opportunities in Mathematical statistics, with special reference to sampling and quality control. Science 1943; 97: 209–214. doi:10.1126/science.97.2514.209

24.

Martin Bland

Altman

. Statistical methods for assessing agreement between two methods of clinical measurement. The Lancet 1986; 327: 307–310

25.

Schuetzenmeister

Manuilova

Model

. Mcr: Method Comparison Regression; 2021. https://CRAN.R-project.org/package=mcr

26.

Team

RCR

. A Language and Environment for Statistical Computing; 2021. https://www.R-project.org/

27.

Yano

Matsunaga

Harada

, et al. Comparison of two homogeneous LDL-Cholesterol Assays using fresh hypertriglyceridemic serum and quantitative ultracentrifugation fractions. J Atheroscler Thromb 2019; 26: 979–988. doi:10.5551/jat.47191

28.

Miida

Nishimura

Okamura

, et al. A multicenter study on the precision and accuracy of homogeneous assays for LDL-cholesterol: comparison with a beta-quantification method using fresh serum obtained from non-diseased and diseased subjects. Atherosclerosis 2012; 225: 208–215. doi:10.1016/j.atherosclerosis.2012.08.022

29.

Takeuchi

Nemoto

K-I

Takahashi

, et al. Comparison of cardiovascular disease risk associated with 3 lipid measures in Japanese adults. J Clin Lipidol 2014; 8: 501–509. doi:10.1016/j.jacl.2014.06.005

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.07 MB

0.23 MB

0.27 MB