Skip to main content
NCBI home page
PLOS One logoLink to PLOS One
. 2024 Sep 20;19(9):e0309002. doi: 10.1371/journal.pone.0309002

More precise method of low-density lipoprotein cholesterol estimation for tobacco and electronic cigarette smokers: A cross-sectional study

Han-Joon Bae 1,*, Hae Won Jung 1, Seung-Pyo Hong 1
Editor: Shukri AlSaif2
PMCID: PMC11414970  PMID: 39302923

Abstract

Smoking is associated with elevated low-density lipoprotein cholesterol (LDL-C) levels. However, the accuracies of the Friedewald, Sampson, and Martin LDL-C-estimating equations based on smoking status are unclear. We analyzed the accuracy of LDL-C levels estimated using these three equations based on tobacco and electronic cigarette smoking status. Data on LDL-C and other lipid components were obtained from the Korea National Health and Nutrition Examination Survey from January 2009 to December 2021. Direct LDL-C (dLDL-C) levels and smoking data of 12,325 participants were evaluated. Current smokers had higher triglyceride levels than never smokers. Electronic cigarette smokers had higher triglyceride and dLDL-C levels than never smokers. The Martin equation yielded more accurate mean absolute deviations than the other equations for the group with triglyceride levels <400 mg/dL as well as more accurate median absolute deviation values, except for the group with dLDL-C levels <40 mg/dL. Similar estimates were derived from the equations when the triglyceride levels were <150 mg/dL. However, the Martin equation may lead to the overestimation of LDL-C levels. In conclusion, the Martin equation is suitable for triglyceride levels <400 mg/dL regardless of the electronic cigarette/tobacco smoking status; if the triglyceride level is <150 mg, the Friedewald equation could also be considered, regardless of the electronic cigarette/tobacco smoking status.

Introduction

Cardiovascular disease (CVD) is the predominant cause of death and is associated with an increasing number of diseases worldwide [1]. Low-density lipoprotein cholesterol (LDL-C) is a major risk factor for CVD and is the primary target for preventing atherosclerotic cardiovascular events [2–5]. Additionally, the clinical significance of lipid modification is a key factor in atherosclerotic CVD.

Smoking is the most common form of tobacco use worldwide. It is a well-known, significant classical risk factor for CVD and one of the primary causes of death globally [6, 7]. Cigarette smoke is a complex, dynamic chemical mixture composed of multiple compounds. Specifically, it contains more than 4,000 chemicals as well as numerous compounds of variable sizes that constitute particulate matter [6–8]. All forms of cigarette smoke are harmful, and there is no safe level of exposure to it. Additionally, smoke exposure can cause atherogenesis, atherosclerosis, and atherothrombosis [8], and both smoking and smoking cessation have been shown to be associated with changes in cholesterol levels [9, 10].

The best method for measuring LDL-C levels uses ultracentrifuged plasma, but this method is time-consuming and labor-intensive [11]. Therefore, several homogeneous direct methods to measure high-density lipoprotein cholesterol (HDL-C) and LDL-C levels have been developed. These methods use various surfactants, ionic polymers, and other components to measure cholesterol levels in specific classes of lipoproteins in the serum [12]; however, they can be expensive and are not available at all medical centers. Currently, LDL-C is typically calculated using three equations: the Friedewald, Sampson, and Martin equations [13, 14]. However, it is unclear which of these can estimate LDL-C levels in patients more accurately based on their smoking status. Moreover, the use of electronic cigarettes (ECs) has increased considerably recently, but there have been very few studies on LDL-C measurements in this context.

In this study, we examined the accuracy of LDL-C estimation using the three equations based on tobacco cigarette (TC) and EC smoking status.

Materials and methods

Study design and participants

This cross-sectional study utilized data obtained from the Korea National Health and Nutrition Examination Survey (KNHANES) [15] of the South Korean population. Detailed information can be found in a previous study [16]. This study was conducted in accordance with the Declaration of Helsinki and was approved by the institutional review boards of each location (IRB 2018-01-03-P-A, 2018-01-03-2C-A). In the KNHANES, a cohort of 20,388 South Korean adults aged ≥20 years was included between January 2009 and December 2021. Participants with available estimated LDL-C and smoking questionnaire data were included in this study. Individuals with triglyceride (TG) levels >1000 mg/dL and missing cholesterol data were excluded. The detailed inclusion and exclusion criteria are illustrated in S1 Fig.

Lipid sample measurement and lipid level estimation

Blood samples were obtained after a minimum fasting period of 8 h. The samples were analyzed by a homogeneous direct assay using reagents from Sekisui Medical Corporation (Tokyo, Japan) on a Hitachi 7600 automated analyzer (Hitachi, Tokyo, Japan), as previously described [16].

Smoking status

The participants were categorized into three groups according to their smoking status. Current smokers were defined as adults who were actively engaged in cigarette smoking and had a history of smoking cigarettes throughout their lifetime. Former smokers were defined as adults who had a history of smoking >100 cigarettes in their lifetime but had quit smoking at the time of the interview. Never smokers were defined as adults who had either never smoked cigarettes (TC and EC) or had smoked <100 cigarettes throughout their lifetime. EC smokers were defined as adults who were currently engaged in vaping. They shared the defined characteristics of former and current smokers but not of never smokers.

Statistical analyses

Continuous variables are presented as medians and interquartile ranges [IQRs], while categorical variables are presented as numbers and percentages (%). Statistical analysis was conducted by smoking status for each TG level. TG levels were classified into three categories: <150, 150–400, and 400–1000 mg/dL. Overall precision was defined as the ratio of direct LDL-C (dLDL-C) to estimated LDL-C and non-high-density lipoprotein (NHDL) (S2–S4 Figs).

The correlations among the three equations and dLDL-C levels were represented through scatter plots, stratified by each smoking status. Concordance was evaluated using statistical metrics, including mean absolute error (MAE), R2, and root mean square error (RMSE). Residual error was defined as an estimate of the disparity between dLDL-C values and estimated LDL-C values. The mean absolute difference (MAD) and MAE values were computed, and confidence intervals (CIs) for median absolute deviations (MeADs) were established for each equation difference, stratified according to dLDL-C levels. The R package DescTools was used to calculate the CIs for the MeADs of two-sample difference and deviation of the sample. All statistical analyses were executed using R version 4.1.3 (R Foundation for Statistical Computing, Vienna, Austria).

Results

Characteristics of LDL-C stratified by TG levels

In the KNHANES, the dLDL-C levels of 12,325 South Korean individuals (median age, 51 [38–63] years; 54.8% male) were determined. The demographic characteristics and lipid profile of the participants stratified by TG levels and smoking status are shown in S1–S3 Tables.

At each TG level, the current smoker group exhibited higher TG levels (except those with TG levels <150 mg/dL) and dLDL-C levels (except those with TG levels <150 mg/dL) than the never smoker group (except those with TG levels <150 mg/dL). The EC smoker group had higher TG, lower HDL, and higher dLDL-C levels than the never smoker group. No significant differences were observed among the three equations (Friedewald, Sampson, and Martin) for those with TG levels <150 mg/dL. The current smoker group exhibited a lower calculated LDL-C value than the never smoker group when TG levels were ≥150 mg/dL. The positive absolute value of calculated LDL-C was higher in the current smoker group than in the never smoker group. Additionally, the absolute value of LDL-C for the never smoker group with TG levels ≥150 mg/dL determined using the Sampson equation was lower than the values determined using other equations (Friedewald and Martin).

Concordance of direct and estimated LDL-C

The correlations between the estimated LDL-C values and dLDL-C levels, stratified by smoking status, are illustrated in S2 Fig. In the never smoker group, the correlation coefficients were 0.826 (Sampson), 0.818 (Martin), and 0.796 (Friedewald); the corresponding MAEs were 7.06, 7.27, and 8.85. In the former smoker group, the correlation coefficients were 0.919 (Sampson), 0.814 (Martin), and 0.790 (Friedewald), and the corresponding MAEs were 7.55, 7.45, and 11.35. In the current smoker group, the correlation coefficients were 0.811 (Sampson), 0.815 (Martin), and 0.773 (Friedewald); the corresponding MAEs were 8.19, 7.69, and 11.60, respectively. In the EC smoker group, the correlation coefficients were 0.795 (Sampson), 0.797 (Martin), and 0.754 (Friedewald); the corresponding MAEs were 8.40, 7.95 and 11.37.

Residual error and MAD

The correlations between differences in the estimated LDL-C and dLDL-C values and TG levels, according to smoking status, are illustrated in Fig 1 (TG levels <400 mg/dL) and S5 Fig (TG levels <1000 mg/dL). In the never smoker group, the coefficients of correlation with TG levels <400 mg/dL were 0.182 (Sampson), -0.142 (Martin), and 0.361 (Friedewald); the corresponding MADs were 6.72, 6.69, and 7.97; and the MeADs with 95% CIs were 7.56 (7.35–7.77), 7.12 (6.92–7.31), and 9.57 (9.29–9.86). In the former smoker group, the coefficients of correlation with TG levels <400 mg/dL were 0.219 (Sampson), -0.129 (Martin), and 0.386 (Friedewald); the corresponding MADs were 6.94, 6.69, and 8.54. In the current smoker group, the coefficients of correlation between LDL-C and TG levels <400 mg/dL were 0.221 (Sampson), -0.088 (Martin), and 0.364 (Friedewald); the corresponding MADs were 7.19, 6.82, and 8.63. In the EC smoker group, the coefficients of the correlation with TG levels <400 mg/dL were 0.235 (Sampson), -0.063 (Martin), and 0.385 (Friedewald); the corresponding MADs were 7.43, 7.17, and 8.93.

Fig 1. Residual error plots for low-density lipoprotein cholesterol (LDL-C) level derived by equations according to smoking status (triglyceride level <400 mg/dL).

Fig 1

Open in a new tab

Differences between the estimates obtained by the Sampson equation (A), Martin equation (B), and Friedewald equation (C) and dLDL-C, stratified by TG level. Residual error was calculated as the difference between dLDL-C and estimated LDL-C values. The dots indicate the individual samples, colored according to smoking status. The color scale indicates individuals. The solid line indicates the trend by the local regression method. MeADs with 95% CIs were calculated by two-sample difference. Cor, correlation coefficient; MAD, mean absolute deviation; MAE, mean absolute error; Median absolute deviation, MeAD; R2, correlation coefficient.

Mean absolute differences stratified by lipid components

The MADs of the three equations according to the dLDL-C levels classified by smoking status are illustrated in Fig 2 (TG levels <150 mg/dL and TG levels <1000 mg/dL). In the never smoker group with TG levels <150 mg/dL, MADs were 8.84, 8.45, and 8.97 for dLDL-C levels <40 mg/dL; 4.91, 4.71, and 5.12 for dLDL-C levels ≥40 and <59 mg/dL; and 4.17, 3.89, and 4.45 for dLDL-C levels ≥60 and <80 mg/dL. In the former smoker group with TG levels <150 mg/dL, MADs were 10.76, 10.20, and 10.69 for dLDL-C levels <40 mg/dL; 5.28, 5.10, and 5.33 for dLDL-C levels ≥40 and <60 mg/dL; and 4.05, 3.78, and 4.31 for dLDL-C levels ≥60 and <80 mg/dL.

Fig 2. Mean absolute deviations of directly measured low-density lipoprotein cholesterol (dLDL-C) stratified by estimated LDL-C.

Fig 2

Open in a new tab

(A) MAD of estimated LDL-C vs. dLDL-C for the never smoker group (TG levels ≥400 and <1000 mg/dL [4]). (B) MAD of estimated LDL-C vs. dLDL-C for the never smoker group (TG levels <400 mg/dL). (C) MAD of estimated LDL-C vs. dLDL-C for the former smoker group (TG levels TG levels ≥400 and <1000 mg/dL). (D) MAD of estimated LDL-C vs. dLDL-C for the former smoker group (TG levels <400 mg/dL). (E) MAD of estimated LDL-C vs. dLDL-C for the current smoker group (TG levels TG levels ≥400 and <1000mg/dL). (F) MAD of estimated LDL-C vs. dLDL-C for the current smoker group (TG levels <400 mg/dL). (G) MAD of estimated LDL-C vs. dLDL-C for the EC smoker group (TG levels TG levels ≥400 and <1000 mg/dL). (H) MAD of estimated LDL-C vs. dLDL-C for the EC smoker group (TG levels <400 mg/dL). HDL, high-density lipoprotein; TG, triglyceride; EC, electronic cigarette. Sampson equation (blue), Martin equation (red), and Friedewald equation (green).

In the current smoker group with TG levels <150 mg/dL, the MADs were 5.82, 5.11, and 6.62 for dLDL-C levels <40 mg/dL; 4.14, 3.99, and 4.60 for dLDL-C levels ≥40 and <60 mg/dL; and 4.15, 3.88, and 4.61 for dLDL-C levels ≥60 and <80 mg/dL. In the EC smoker group with TG levels <150 mg/dL, the MADs were 3.00, 3.13, and 3.82 for dLDL-C levels <40 mg/dL; 4.19, 3.99, and 4.73 for dLDL-C levels ≥40 and <60 mg/dL; and 4.54, 4.22, and 4.82 for dLDL-C levels ≥60 and <80 mg/dL.

Precision of estimated LDL-C stratified by LDL-C

MADs and MeADs with 95% CIs are shown in Table 1 (TG levels <150 mg/dL), S4 Table (TG levels <400 mg/dL), and S5 Table (TG levels <1000 mg/dL).

Table 1. Mean and median absolute deviations with 95% CIs of estimated low-density lipoprotein cholesterol stratified by dLDL-C in the group with TG levels of <150 mg/dL.

Sampson equation Martin equation Friedewald equation
Never smoker
dLDL-C MAD MeAD MAD MeAD MAD MeAD
<40 3.99 -0.96 (-5.85–3.93) 3.75 -2.08 (-6.68–2.53) 4.40 -1.60 (-5.65–2.45)
<70 4.23 -1.63 (-3.35–0.09) 4.07 -1.48 (-3.18–0.21) 4.47 -1.33 (-3.01–0.34)
≥70 and <100 4.31 -0.89 (-1.76 to -0.02) 3.98 0.00 (-0.83–0.83) 4.54 -0.64 (-1.50–0.23)
≥100 and <130 4.92 -1.85 (-2.72 to -0.98) 4.47 -1.48 (-2.34 to -0.63) 5.09 -1.10 (-1.93 to -0.27)
≥130 and <160 5.24 -1.33 (-2.57 to -0.10) 4.83 -0.74 (-1.94–0.46) 5.27 -1.02 (-2.24–0.21)
≥160 6.65 0.00 (-5.55–5.55) 6.46 0.52 (-4.99–6.03) 6.64 0.16 (-5.38–5.71)
Former smoker
dLDL-C MAD MeAD MAD MeAD MAD MeAD
<40 3.95 -6.23 (-22.96–10.51) 2.73 -3.85 (-15.32–7.61) 5.44 -5.45 (-19.62–8.72)
<70 3.90 -1.93 (-8.10–4.25) 3.44 -1.33 (-7.46–4.80) 4.71 -4.27 (-10.85–2.31)
≥70 and <100 4.10 0.74 (-1.09–2.57) 3.72 0.89 (-0.94–2.72) 4.58 1.07 (-0.73–2.88)
≥100 and <130 4.65 -0.74 (-2.51–1.03) 4.37 -0.59 (-2.36–1.17) 4.82 -0.14 (-1.83–1.55)
≥130 and <160 4.49 -0.44 (-2.89–2.00) 4.27 0.22 (-2.16–2.61) 4.53 -0.56 (-3.00–1.89)
≥160 7.97 -1.85 (-11.55–7.84) 7.62 -0.89 (-10.47–8.69) 7.97 -1.76 (-11.44–7.92)
Current smoker
dLDL-C MAD MeAD MAD MeAD MAD MeAD
<40 1.00 -1.48 (-27.64–24.68) 1.55 -2.30 (-30.81–26.22) 1.39 -2.05 (-29.85–25.74)
<70 3.89 -3.71 (-7.77–0.35) 3.73 -3.26 (-7.21–0.69) 4.41 -4.11 (-8.25–0.04)
≥70 and <100 3.64 -0.89 (-4.27–2.49) 3.31 -0.44 (-3.78–2.90) 3.96 -1.21 (-4.63–2.21)
≥100 and <130 4.33 -2.08 (-4.23–0.08) 3.83 -1.04 (-3.09–1.01) 4.64 -1.90 (-4.05–0.25)
≥130 and <160 5.72 -1.93 (-7.18–3.33) 5.07 -0.89 (-6.20–4.42) 5.85 -2.28 (-7.82–3.25)
≥160 8.25 -6.30 (-14.82–2.21) 7.85 -5.26 (-13.37–2.84) 8.14 -5.78 (-14.01–2.45)
Electronic cigarette smoker
dLDL-C MAD MeAD MAD MeAD MAD MeAD
<40
<70 4.65 -7.41 (-15.16–0.33) 4.56 -8.01 (-22.39–6.38) 5.57 -6.69 (-14.09–0.71)
≥70 and <100 4.52 -0.74 (-5.30–3.82) 4.20 -0.30 (-5.10–4.51) 4.69 -2.42 (-7.17–2.33)
≥100 and <130 3.85 0.22 (-3.26–3.71) 3.79 0.74 (-2.89–4.37) 3.96 -0.34 (-3.83–3.15)
≥130 and <160 4.02 3.56 (-4.20–11.32) 3.42 2.37 (-5.32–10.06) 4.23 3.65 (-3.92–11.23)
≥160 7.35 0.22 (-28.27–28.71) 6.97 1.19 (-28.04–30.42) 7.43 0.11 (-28.7–28.92)
Open in a new tab

CI, confidence interval; dLDL-C, direct low-density lipoprotein cholesterol; MAD, mean absolute deviation; MeAD, median absolute deviation. The MeADs with 95% CIs were calculated by two-sample difference.

SI conversion factors: To convert cholesterol to mmol/L, values were multiplied by 0.0259.

In the never smoker group with TG levels <150 mg/dL and dLDL-C levels <70 mg/dL, MADs and MeADs were as follows: Sampson equation, 4.23 and -1.63 (-3.35–0.09) vs. Martin equation, 4.07 and -1.48 (-3.18–0.21) vs. Friedewald equation, 4.47 and -1.33 (-3.01–0.34). In the former smoker group with TG levels <150 mg/dL and dLDL-C levels <70 mg/dL, MADs and MeADs were as follows: Sampson equation, 3.90 and -1.93 (-8.10–4.25) vs. Martin equation, 3.44 and -1.33 (-7.46–4.80) vs. Friedewald equation, 4.71 and -4.27 (-10.85–2.31).

In the current smoker group with TG levels <150 mg/dL and dLDL-C levels <70 mg/dL, MADs and MeADs were as follows: Sampson equation, 3.89 and -3.71 (-7.77–0.35) vs. Martin equation, 3.73 and -3.26 (-7.21–0.69) vs. Friedewald equation, 4.41 and -4.11 (-8.25–0.04). In the EC smoker group with TG levels <150 mg/dL and dLDL-C levels <70 mg/dL, MADs and MeADs were as follows: Sampson equation, 4.65 and -7.41 (-15.16–0.33) vs. Martin equation, 4.56 and -8.01 (-22.39–6.38) vs. Friedewald equation, 5.57 and -6.69 (-14.09–0.71).

The Friedewald equation demonstrated higher MADs and MeADs than the other equations for TG levels <150 mg/dL, <400 mg/dL, 400–1000 mg/dL, and <1000 mg/dL according to smoking status (Table 1 and S4–S6 Tables). The Martin equation for TG levels <150 mg/dL and 150–400 mg/dL demonstrated lower MADs and MeADs than the other equations. The Sampson equation demonstrated lower MADs and MeADs for TG levels 400–1000 mg/dL and <400 mg/dL and dLDL-C levels <100 mg/dL. In case of LDL-C levels <70 mg/dL and TG levels >400 mg/dL, the MAD of the Sampson equation was lower than those of the other equations without never smokers.

Precision of estimated LDL-C for the group with daily smoker status

The MADs and MeADs with 95% CIs for the current TC and EC smoking status are illustrated in Figs 3 and 4. In the current smoker group with TG levels <400 mg/dL, the MADs, MeADs, and correlation coefficients obtained for the three equations were similar. However, as the TG level increased, the differences in MADs and correlation coefficients also increased for each equation. The accuracies of the Martin and Sampson equations were similar to that of the Friedewald equation. However, in the current smoker group with TG levels >400 mg/dL, the Martin equation exhibited a slight overestimation. The Friedewald equation exhibited a higher difference than the other equations. However, for TG levels <150 mg/dL, all three equations showed acceptable ranges according to smoking burden.

Fig 3. Distribution plots showing low-density lipoprotein cholesterol (LDL-C) levels measured by different equations according to cigarettes per day.

Fig 3

Open in a new tab

Differences between the estimates obtained by the Sampson equation (A), Martin equation (B), and Friedewald equation and direct LDL-C (dLDL-C), stratified by cigarettes per day. The dots indicate the individual samples, colored according to TG levels. The color scale indicates individuals. The solid line indicates the trend by the local regression method. Cor, correlation coefficient; MAD, mean absolute deviation; MAE, mean absolute error; MeAD, median absolute deviation; R2, correlation coefficient; RMSE, root mean square error; TG, triglyceride.

Fig 4. Residual error plots for low-density lipoprotein cholesterol (LDL-C) level derived by equations according to daily electronic cigarette smoking.

Fig 4

Open in a new tab

Differences between the estimates obtained by the Sampson equation (A), Martin equation (B), and Friedewald equation and dLDL-C, stratified by daily electronic cigarette smoking. The residual error is calculated by the difference between direct LDL-C (dLDL-C), and the values are obtained using the lipid equations. The dots indicate the individual samples, colored according to smoking status. The color scale indicates individuals. The solid line indicates the trend by the local regression method. The median absolute deviations with 95% confidence intervals were calculated by two-sample difference. Cor, correlation coefficient; MAD, mean absolute deviation; MAE, mean absolute error; MeAD, median absolute deviation; R2, correlation coefficient.

In the EC smoker group with TG levels <400 mg/dL and >400 mg/dL, the MADs, MeADs, and correlation coefficients for the Sampson and Martin equations were similar. The Friedewald equation exhibited a higher difference than the other equations.

Discussion

In this investigation, we evaluated and compared the precision of LDL-C levels determined by three equations according to smoking status. This study revealed four main findings: 1) current smokers exhibited a higher TG level than never smokers; 2) the Sampson equation exhibited superior precision to other equations for the group with TG levels >400 mg/dL, regardless of the smoking status; 3) the Martin equation exhibited greater precision than the other equations for the group with TG levels <400 mg/dL, irrespective of the smoking status; and 4) the Friedewald equation was not more accurate than the other equations. However, the differences in the MADs, MeADs, and residual errors with and without EC and TC smoking status was not substantial for TG levels <150 mg/dL with all three equations.

Relationship between lipoprotein and smoking

Smoking decreases HDL-C levels, while smoking cessation is linked to increased HDL-C levels as measured in older adults [9, 10]; however, this association is not completely understood. Smoking induces the release of catecholamines, leading to an increase in circulating fatty acids; this process may lead to elevated concentrations of LDL and very low-density lipoprotein (VLDL) as well as decreased concentrations of HDL-C [17]. Smoking additionally decreases the levels of lecithin cholesterol acyltransferase, an enzyme responsible for esterifying free cholesterol and increasing HDL size. Smoking cessation was associated with reductions in TG levels, potentially attributed to the counterbalancing effect of weight gain; nevertheless, no significant changes were observed in LDL-C level, LDL particle concentration, or LDL size [17]. In this study, current smokers had higher LDL-C and TG levels as well as lower HDL levels than never smokers. Former smokers also had higher TG levels than never smokers. The frequency of EC smoking has increased, particularly owing to the transition from TC smoking to reduce exposure to harmful contents of TCs or bridge the cessation of TC smoking. In this study, EC was common among young and middle-aged adults. Limited data are available on the effects of ECs.

Differences in LDL-C and dLDL-C depending on the equation

Higher differences in LDL-C and dLDL-C levels (in terms of the MADs and MAEs) were observed in current smokers than in never and former smokers, while slightly lower differences were observed in never and former smokers. The MADs, MAEs, residual errors, and 95% CIs of MeADs obtained with the Martin equation in never, former, and current smokers were lower than those obtained with the other equations. The errors remained high for LDL levels <40 mg/dL in all formulas, suggesting limitations in their practical application. In the current smoker group, the MADs and MeADs for TG levels <150 mg/dL were lower than those in the never smoker group. For TG levels <150 mg/dL, the Martin equation yielded lower MADs and MeADs than the other equations, except for those with LDL levels <40 mg/dL in the current smoker group.

Differences in LDL-C and dLDL-C in current smokers

Direct LDL-C measurement is the best approach for accurate measurement of lipid profiles; however, direct measurement is not available in primary clinics and is not cost-effective. Indirect measurements are used in clinical practice for various reasons; therefore, a precise method for indirect calculation of LDL-C is needed. The three well-known equations used in this study exhibited inaccuracies at TG levels >400 mg/dL; the Friedewald equation exhibited a higher positive error than the other equations, while the Martin equation exhibited a slightly higher positive error. Furthermore, the Sampson method exhibited a higher likelihood of underestimating LDL-C levels than the Martin equation at TG levels <150 mg/dL. Moreover, achieving an LDL-C level <40 mg/dL in patients with a positive smoking status led to more precise MAD values when using the Martin equation. The Sampson equation demonstrated greater precision in patients for TG levels >400 mg/dL; however, its application is recommended for patients with low LDL levels. In case of LDL-C levels <70 mg/dL and TG levels >400 mg/dL, the MAD of the Sampson equation was lower than those of the other equations without never smokers. In case of LDL-C levels <70 mg/dL and TG levels > 400 mg/dL with never smokers, the Martin equation had a lower MAD than the other equations.

Limitations

This study has certain limitations. First, dLDL-C levels were measured using an enzymatic method, whereas beta quantification (ultracentrifugation) was conducted in most studies. This method is regarded as the standard measurement technique for LDL-C; however, the lipid data from the KNHANES underwent validation through a standardization program from the United States Centers for Disease Control and Prevention. Second, genetic components are associated with lipid metabolism and lipid profiles; however, these factors could not be explored owing to the absence of genetic data. Third, metabolic components, insulin resistance, and effects of statins were not evaluated. Fourth, this cross-sectional study did not include follow-up measurements of blood samples; therefore, we were unable to assess follow-up data. Fifth, the sample sizes of the high TG level (>500 mg/dL) and low LDL-C level (<40 mg/dL) groups were small. Sixth, the levels of small lipid components such as VLDL, apoprotein A/B, and lipoprotein A were not measured. Further studies are needed with large sample sizes and data on VLDL-cholesterol, apoprotein, and measured lipoprotein A.

Conclusions

This study utilized a nationally representative sample and validated the precision of LDL-C estimation in patients with smoking habits. Our investigation revealed variations in the precision of estimated LDL-C levels according to smoking status. In those with TG levels <400 mg/dL, daily smoking of ECs or TCs did not have a significant impact on LDL estimations. We recommend the application of the Sampson equation for those with TG levels >400 mg/dL and the Martin equation for those with TG levels <400 mg/dL, regardless of smoking status. Using different equations may help minimize the errors in LDL-C measurements.

Supporting information

S1 Fig. Flow diagram of the inclusion and exclusion criteria for participants from the KNHANES (2009–2021).

KNHANES, Korea National Health and Nutrition Examination Survey; LDL-C, low-density lipoprotein cholesterol; dLDL-C, direct low-density lipoprotein cholesterol.

(DOCX)

pone.0309002.s001.docx (69.4KB, docx)
S2 Fig. Correlation of estimated vs. direct low-density lipoprotein cholesterol (LDL-C) levels in patients with smoking status.

A. The Sampson equation vs. direct LDL-C for never smokers. B. The Martin equation vs. direct LDL-C for never smokers. C. The Friedewald equation vs. direct LDL-C for never smokers. D. The Sampson equation vs. direct LDL-C for former smokers. E. The Martin equation vs. direct LDL-C for former smokers. F. The Friedewald equation vs. direct LDL-C for former smokers. G. The Sampson equation vs. direct LDL-C for current smokers. H. The Martin equation vs. direct LDL-C for current smokers. I. The Friedewald equation vs. direct LDL-C for current smokers. J. The Sampson equation vs. direct LDL-C for electronic cigarette smokers. K. The Martin equation vs. direct LDL-C for electronic cigarette smokers. L. The Friedewald equation vs. direct LDL-C for electronic cigarette smokers. Cor, correlation coefficient; MAE, mean absolute error; R2, correlation coefficient; RMSE, root mean square error; TG, triglyceride. The dots indicate the individual samples colored according to TG level. The color scale indicates individuals. SI conversion factors: To convert cholesterol to mmol/L, values were multiplied by 0.0259.

(DOCX)

pone.0309002.s002.docx (2.4MB, docx)
S3 Fig. Mean absolute deviations of directly measured low-density lipoprotein cholesterol (dLDL-C) stratified by estimated LDL-C.

(A) MAD of estimated LDL-C vs. dLDL-C for the never smoker group (TG levels <1000 mg/dL). (B) MAD of estimated LDL-C vs. dLDL-C for the never smoker group (TG levels <150 mg/dL). (C) MAD of estimated LDL-C vs. dLDL-C for the former smoker group (TG levels <1000 mg/dL). (D) MAD of estimated LDL-C vs. dLDL-C for the former smoker group (TG levels <150 mg/dL). (E) MAD of estimated LDL-C vs. dLDL-C for the current smoker group (TG levels <150 mg/dL). (F) MAD of estimated LDL-C vs. dLDL-C for the current smoker group (TG levels <150 mg/dL). (G) MAD of estimated LDL-C vs. dLDL-C for the EC smoker group (TG levels <150 mg/dL). (H) MAD of estimated LDL-C vs. dLDL-C for the EC smoker group (TG levels <150 mg/dL). HDL, high-density lipoprotein; TG, triglyceride; EC, electronic cigarette. Sampson equation (blue), Martin equation (red), and Friedewald equation (green).

(DOCX)

pone.0309002.s003.docx (1.5MB, docx)
S4 Fig. Mean absolute deviations of directly measured low-density lipoprotein cholesterol (dLDL-C) stratified by non-high-density lipoprotein (NHDL).

(A) MAD of estimated LDL-C vs. dLDL-C for the never smoker group (TG levels <1000 mg/dL). (B) MAD of estimated LDL-C vs. dLDL-C for the former smoker group (TG levels <1000 mg/dL). (C) MAD of estimated LDL-C vs. dLDL-C for the current smoker group (TG levels <150 mg/dL). (D) MAD of estimated LDL-C vs. dLDL-C for the EC smoker group (TG levels <150 mg/dL. NHDL, non-high-density lipoprotein; TG, triglyceride; EC, electronic cigarette. Sampson equation (blue), Martin equation (red), and Friedewald equation (green).

(DOCX)

pone.0309002.s004.docx (146.7KB, docx)
S5 Fig. Residual error plots for low-density lipoprotein cholesterol (LDL-C) level estimated using equations according to smoking status (triglyceride levels <1000 mg/dL).

Differences between the estimates obtained by the Sampson equation (A), Martin equation (B), and Friedewald equation (C) and direct LDL-C (dLDL-C), stratified by triglyceride (TG) level. Residual error was calculated by the difference between direct LDL-C (dLDL-C) and the values obtained using the equations. The dots indicate the individual samples, colored according to smoking status. The color scale indicates individuals. The solid line indicates the trend by the local regression method. Cor, correlation coefficient; MAD, mean absolute deviation; MAE, mean absolute error; R2 indicates correlation coefficient.

(DOCX)

pone.0309002.s005.docx (1.5MB, docx)
S1 Table. Characteristics of study population with triglyceride levels of ≥400 mg/dL and <1000 mg/dL.

(DOCX)

pone.0309002.s006.docx (20.5KB, docx)
S2 Table. Characteristics of the study population with triglyceride levels of ≥150 mg/dL and <400 mg/dL.

(DOCX)

pone.0309002.s007.docx (21.6KB, docx)
S3 Table. Characteristics of the study population with triglyceride levels of <150 mg/dL.

(DOCX)

pone.0309002.s008.docx (22.3KB, docx)
S4 Table. Mean and median absolute deviations with 95% confidence intervals of estimated low-density lipoprotein cholesterol stratified by dLDL-C in the group with TG levels of <400 mg/dL.

(DOCX)

pone.0309002.s009.docx (20.3KB, docx)
S5 Table. Mean and median absolute deviations with 95% confidence intervals of estimated low-density lipoprotein cholesterol stratified by dLDL-C in the group with TG levels of <1000 mg/dL.

(DOCX)

pone.0309002.s010.docx (19.9KB, docx)
S6 Table. Mean and median absolute deviations with 95% confidence intervals of estimated low-density lipoprotein cholesterol stratified by dLDL-C in the group with TG levels of >400 mg/dL and <1000 mg/dL.

(DOCX)

pone.0309002.s011.docx (21.6KB, docx)

Acknowledgments

We are grateful to the Korean Ministry of Health and Welfare for providing the data.

Data Availability

KNHANES data are available online (https://knhanes.cdc.go.kr/knhanes/eng/index.do).

Funding Statement

This work was supported by a grant from the Research Institute of Medical Science, Daegu Catholic University, Daegu, South Korea (2022, RD-22-0007). The funders had no role in considering the study design or in the collection, analysis, interpretation of data, writing of the report, or decision to submit the article for publication.

References

  • 1.GBD 2017 Causes of Death Collaborators. Global, regional, and national age-sex-specific mortality for 282 causes of death in 195 countries and territories, 1980–2017: a systematic analysis for the Global Burden of Disease Study 2017. Lancet. 2018;392: 1736–1788. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 2.Arnett DK, Blumenthal RS, Albert MA, Buroker AB, Goldberger ZD, Hahn EJ, 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. Circulation. 2019;140: e563–e595. doi: 10.1161/CIR.0000000000000677 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 3.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: a report of the American College of Cardiology/American Heart Association Task Force on Clinical Practice Guidelines. Circulation. 2019;139: e1082–e1143. doi: 10.1161/CIR.0000000000000625 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 4.Langlois MR, Nordestgaard BG, Langsted A, Chapman MJ, Aakre KM, Baum H, et al. Quantifying atherogenic lipoproteins for lipid-lowering strategies: consensus-based recommendations from EAS and EFLM. Clin Chem Lab Med. 2020;58: 496–517. doi: 10.1515/cclm-2019-1253 [DOI] [PubMed] [Google Scholar]
  • 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. Eur Heart J. 2020;41: 111–188. doi: 10.1093/eurheartj/ehz455 [DOI] [PubMed] [Google Scholar]
  • 6.Borgerding M, Klus H. Analysis of complex mixtures—cigarette smoke. Exp Toxicol Pathol. 2005;57 Suppl 1: 43–73. doi: 10.1016/j.etp.2005.05.010 [DOI] [PubMed] [Google Scholar]
  • 7.Selya AS, Hesse ND. Time to first cigarette and serum cholesterol levels. Soc Sci Med. 2017;174: 213–219. doi: 10.1016/j.socscimed.2016.12.014 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 8.Messner B, Bernhard D. Smoking and cardiovascular disease: mechanisms of endothelial dysfunction and early atherogenesis. Arterioscler Thromb Vasc Biol. 2014;34: 509–515. doi: 10.1161/ATVBAHA.113.300156 [DOI] [PubMed] [Google Scholar]
  • 9.Gepner AD, Piper ME, Johnson HM, Fiore MC, Baker TB, Stein JH. Effects of smoking and smoking cessation on lipids and lipoproteins: outcomes from a randomized clinical trial. Am Heart J. 2011;161: 145–151. doi: 10.1016/j.ahj.2010.09.023 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 10.He B-M, Zhao S-P, Peng Z-Y. Effects of cigarette smoking on HDL quantity and function: implications for atherosclerosis. J Cell Biochem. 2013;114: 2431–2436. doi: 10.1002/jcb.24581 [DOI] [PubMed] [Google Scholar]
  • 11.Havel RJ, Eder HA, Bragdon JH. The distribution and chemical composition of ultracentrifugally separated lipoproteins in human serum. J Clin Invest. 1955;34: 1345–1353. doi: 10.1172/JCI103182 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12.Miller WG, Myers GL, Sakurabayashi I, Bachmann LM, Caudill SP, Dziekonski A, et al. Seven direct methods for measuring HDL and LDL cholesterol compared with ultracentrifugation reference measurement procedures. Clin Chem. 2010;56: 977–986. doi: 10.1373/clinchem.2009.142810 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13.Erturk Zararsiz G, Bolat S, Cephe A, Kochan N, Yerlitas SI, Dogan HO, et al. Validation of Friedewald, Martin-Hopkins and Sampson low-density lipoprotein cholesterol equations. PLoS One. 2022;17: e0263860. doi: 10.1371/journal.pone.0263860 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 14.Steyn N, Muller Rossouw H, Pillay TS, Martins J. Comparability of calculated LDL-C with directly measured LDL-C in selected paediatric and adult cohorts. Clin Chim Acta. 2022;537: 158–166. doi: 10.1016/j.cca.2022.10.003 [DOI] [PubMed] [Google Scholar]
  • 15.Kweon S, Kim Y, Jang M-J, Kim Y, Kim K, Choi S, et al. Data resource profile: the Korea National Health and Nutrition Examination Survey (KNHANES). Int J Epidemiol. 2014;43: 69–77. doi: 10.1093/ije/dyt228 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16.Bae H-J, Kim S-W, Kim I-S. Comparison of low-density lipoprotein cholesterol estimation methods in individuals with insulin resistance: a cross-sectional study. Clin Chim Acta. 2023;547: 117393. doi: 10.1016/j.cca.2023.117393 [DOI] [PubMed] [Google Scholar]
  • 17.Chelland Campbell S, Moffatt RJ, Stamford BA. Smoking and smoking cessation—the relationship between cardiovascular disease and lipoprotein metabolism: a review. Atherosclerosis. 2008;201: 225–235. doi: 10.1016/j.atherosclerosis.2008.04.046 [DOI] [PubMed] [Google Scholar]

Decision Letter 0

Shukri AlSaif

5 Jun 2024

PONE-D-24-08009More precise method of low-density lipoprotein cholesterol estimation for tobacco and electronic cigarette smokers: a cross-sectional studyPLOS ONE

Dear Dr. Bae,

Thank you for submitting your manuscript to PLOS ONE. After careful consideration, we feel that it has merit but does not fully meet PLOS ONE’s publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process.

Please submit your revised manuscript by Jul 20 2024 11:59PM. If you will need more time than this to complete your revisions, please reply to this message or contact the journal office at plosone@plos.org. When you're ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

Please include the following items when submitting your revised manuscript:

  • A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'.

  • A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.

  • An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter. Guidelines for resubmitting your figure files are available below the reviewer comments at the end of this letter.

If applicable, we recommend that you deposit your laboratory protocols in protocols.io to enhance the reproducibility of your results. Protocols.io assigns your protocol its own identifier (DOI) so that it can be cited independently in the future. For instructions see: https://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols. Additionally, PLOS ONE offers an option for publishing peer-reviewed Lab Protocol articles, which describe protocols hosted on protocols.io. Read more information on sharing protocols at https://plos.org/protocols?utm_medium=editorial-email&utm_source=authorletters&utm_campaign=protocols.

We look forward to receiving your revised manuscript.

Kind regards,

Shukri AlSaif

Academic Editor

PLOS ONE

Journal Requirements:

When submitting your revision, we need you to address these additional requirements.

1. Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at

https://journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and

https://journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf

2. Thank you for uploading your study's underlying data set. Unfortunately, the repository you have noted in your Data Availability statement does not qualify as an acceptable data repository according to PLOS's standards.

At this time, please upload the minimal data set necessary to replicate your study's findings to a stable, public repository (such as figshare or Dryad) and provide us with the relevant URLs, DOIs, or accession numbers that may be used to access these data. For a list of recommended repositories and additional information on PLOS standards for data deposition, please see https://journals.plos.org/plosone/s/recommended-repositories.

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Yes

Reviewer #2: Partly

**********

2. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: Yes

Reviewer #2: Yes

**********

3. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

Reviewer #2: Yes

**********

4. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: Yes

**********

5. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: This work investigated the accuracy of three equations used for LDL-C calculation in smokers. Several points should be addressed before publication.

1. The rationale for this analysis should be elaborated more. Is calculated LDL-C really needed? Why don't you just measure LDL-C directly? If calculated LDL-C is needed, what is the rationale? Whether it is availability problem, technical difficulty of the test and/or cost problem, I think it should be described in the Introduction.

2. Why did you decide that analysis was done according to the smoking status? Was calculated LDL-C less accurate in smokers compared to non-smokers?

3. Related to #2, you described the relationship between smoking and lipid profile in the Discussion section (line 262-277). I think this may be moved to the Introduction, because it helps to justify why the analysis should be done according to the smoking status.

4. What is the novelty of the work? It seems that the Conclusion just confirms that the study finding is consistent with the pre-existing knowledge overall.

Reviewer #2: 1. The tables S1, S2, and S3 should include posthoc test results for significant comparisons. Additionally, significant results should be highlighted in bold to enhance the readability and comprehension of the tables.

2. Given that the comparisons are made within the same sample, adjusted p-values (e.g., using the Benjamini-Hochberg method) should be provided to control for multiple testing.

3. The statement in line 110 should indicate that TG <150 is excluded.

4. The font sizes in the axes and titles of all figures should be increased to improve readability.

5. The definition of overall precision as the ratio of direct LDL-C (dLDL-C) to non-high-density lipoprotein (non-HDL-C) should be supported by a reference or a clear rationale, such as citing relevant studies or providing a logical explanation for why this ratio is an appropriate measure

6. In Figure 2, the LDL-C ranges should be explicitly labeled on the plots for better clarity.

7. Figure 2 currently presents TG <150 mg/dL and TG <1000 mg/dL. It should categorize as TG ≥150–TG <400 mg/dL and TG ≥400–TG <1000 mg/dL. Including TG <150 mg/dL in the <1000 mg/dL category can lead to complexity in comparisons.

8. In line 212, the comment on TG categories is confusing. Table 1 presents data for TG <150 mg/dL, Table S4 for <400 mg/dL, and Table S5 for <1000 mg/dL. However, the text mentions TG levels <150 mg/dL, 150–400 mg/dL, and 400–1000 mg/dL. The TG categories should be clarified consistently throughout the manuscript.

9. The issue of inconsistent TG categories appears in multiple places. The TG categories should be clearly defined and consistently used throughout the manuscript.

10. In Figure 3, the legend should be reordered to present TG <150, 150–400 mg/dL, and 400–1000 mg/dL sequentially to improve the clarity of the plot.

11. The statement in the results about the superiority of the Martin equation in Figure 3 should be moderated. The results suggest that the equations perform similarly.

12. The findings related to Figure 4 should be included in the manuscript to provide a complete interpretation of the results.

13. The performance of the equations in the category of LDL-C <70 mg/dL and TG >400 mg/dL should be evaluated. These results could provide valuable insights into the literature.

14. A more detailed discussion comparing the results with existing literature should be provided. Comparing these findings with studies conducted in different populations would help to understand the consistency and differences in the results.

**********

6. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: No

**********

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email PLOS at figures@plos.org. Please note that Supporting Information files do not need this step.

PLoS One. 2024 Sep 20;19(9):e0309002. doi: 10.1371/journal.pone.0309002.r002

Author response to Decision Letter 0

17 Jul 2024

Thank you for your thoughtful and thorough review of our manuscript. We have made every effort to fulfil the reviewers’ specific requests. We have provided a point-by-point response to all comments from the editorial committee and reviewers below and also indicated our revisions to the manuscript in response to each comment.

Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: This work investigated the accuracy of three equations used for LDL-C calculation in smokers. Several points should be addressed before publication.

1. The rationale for this analysis should be elaborated more. Is calculated LDL-C really needed? Why don't you just measure LDL-C directly? If calculated LDL-C is needed, what is the rationale? Whether it is availability problem, technical difficulty of the test and/or cost problem, I think it should be described in the Introduction.

▶Response: We appreciate the reviewers’ comments. Accordingly, the following text has been added to the Introduction (3rd paragraph):

“The best method for measuring LDL-C levels uses ultracentrifuged plasma, but this method is time-consuming and labor-intensive [11]. Therefore, several homogeneous direct methods to measure high-density lipoprotein cholesterol (HDL-C) and LDL-C levels have been developed. These methods use various surfactants, ionic polymers, and other components to measure cholesterol levels in specific classes of lipoproteins in the serum [12]; however, they can be expensive and are not available at all medical centers.”

2. Why did you decide that analysis was done according to the smoking status? Was calculated LDL-C less accurate in smokers compared to non-smokers?

▶Response: Thank you for your comment. We have added the following information to the Introduction (second paragraph) to clarify this aspect:

“…, and both smoking and smoking cessation have been shown to be associated with changes in cholesterol levels [9, 10].”

3. Related to #2, you described the relationship between smoking and lipid profile in the Discussion section (line 262-277). I think this may be moved to the Introduction, because it helps to justify why the analysis should be done according to the smoking status.

▶Response: Thank you for your comment, which we agree with. As mentioned in the response to the previous comment, we have now referred to this aspect in the Introduction (third paragraph).

“and both smoking and smoking cessation have been shown to be associated with changes in cholesterol levels [9, 10].”

4. What is the novelty of the work? It seems that the Conclusion just confirms that the study finding is consistent with the pre-existing knowledge overall.

▶Response: Thank you for pointing this out. We agree with your comment, and research on LDL validation has indeed been conducted in several countries and research institutes. However, research on validation according to smoking status, particularly for users of electronic cigarettes, is lacking. Accordingly, we have added the following text to the Introduction (third paragraph):

“Moreover, the use of electronic cigarettes (ECs) has increased considerably recently, but there have been very few studies on LDL-C measurements in this context.”

Reviewer #2:

1. The tables S1, S2, and S3 should include posthoc test results for significant comparisons. Additionally, significant results should be highlighted in bold to enhance the readability and comprehension of the tables.

▶Response: Thank you for your helpful comments. Accordingly, we have highlighted the significant results in bold in Tables S1, S2, and S3, and added adjusted p-values derived using the Benjamini–Hochberg post-hoc method to the tables.

2. Given that the comparisons are made within the same sample, adjusted p-values (e.g., using the Benjamini-Hochberg method) should be provided to control for multiple testing.

▶Response: Thank you for pointing this out. As mentioned in the response to the previous comment, the required changes have now been made to Tables S1, S2, and S3.

S1 Table. Characteristics of the study population with triglyceride levels of ≥400 mg/dL and <1000 mg/dL

Smoke status

Adjusted p-values (Benjamini–Hochberg method) Never Never Never Former Former Current

Former Current Electronic Current Electronic Electronic

Age, years 0.037 0.001 <0.001 <0.001 <0.001 0.001

Male, % <0.001 <0.001 <0.001 0.101 0.658 0.244

Body mass index, kg/m2 0.602 0.602 0.768 0.602 0.602 0.602

Laboratory data 　 　 　 　 　

Fasting glucose, mg/dL 0.088 0.600 0.108 0.137 0.030 0.088

Blood urea nitrogen, mg/dL <0.001 0.691 0.691 <0.001 0.016 0.691

Creatinine, mg/dL <0.001 <0.001 <0.001 0.002 0.350 0.523

GFR_EPI 0.004 0.224 0.168 <0.001 <0.001 0.224

Cholesterol 　 　 　 　 　

Total cholesterol, mg/dL 0.998 0.998 0.998 0.998 0.998 0.998

Triglyceride, mg/dL 0.808 <0.001 0.492 0.006 0.530 0.492

HDL-C, mg/dL 0.971 0.971 0.971 0.971 0.971 0.971

Direct LDL-C, mg/dL 0.983 0.642 0.144 0.642 0.144 0.158

Non-HDL-C, mg/dL 0.978 0.978 0.978 0.978 0.978 0.978

Sampson, mg/dL 0.416 0.054 0.977 0.416 0.690 0.416

Martin, mg/dL 0.387 0.066 0.917 0.387 0.661 0.387

Friedewald, mg/dL 0.366 0.018 0.902 0.366 0.707 0.366

Positive absolute value Martin, mg/dL 0.877 0.006 0.030 0.006 0.024 0.364

Positive absolute value Sampson, mg/dL 0.217 <0.001 0.080 0.003 0.204 0.993

Positive absolute value Friedewald, mg/dL 0.181 <0.001 0.056 <0.001 0.181 0.740

S2 Table. Characteristics of the study population with triglyceride levels of ≥150 mg/dL and <400 mg/dL

Smoke status

adjusted p-values (Benjamini–Hochberg method) Never Never Never Former Former Current

Former Current Electronic Current Electronic Electronic

Age, years 0.001 <0.001 <0.001 <0.001 <0.001 <0.001

Male, % <0.001 <0.001 <0.001 0.021 0.901 0.5436

Body mass index, kg/m2 0.048 0.888 0.174 0.066 0.634 0.174

Laboratory data 　 　 　 　 　 　

Fasting glucose, mg/dL <0.001 0.020 0.413 <0.001 <0.001 0.084

Blood urea nitrogen, mg/dL <0.001 0.517 0.396 <0.001 <0.001 0.478

Creatinine, mg/dL <0.001 <0.001 <0.001 <0.001 0.014 0.853

GFR_EPI <0.001 0.018 0.035 <0.001 <0.001 <0.001

Cholesterol 　 　 　 　 　 　

Total cholesterol, mg/dL 0.006 0.201 0.709 0.201 0.201 0.378

Triglyceride, mg/dL <0.001 <0.001 <0.001 0.008 0.030 0.338

HDL-C, mg/dL 0.042 <0.001 0.219 0.219 0.719 0.719

Direct LDL-C, mg/dL 0.012 0.329 0.195 0.006 0.018 0.329

Non-HDL-C, mg/dL 0.024 0.547 0.518 0.100 0.100 0.500

Sampson, mg/dL <0.001 0.009 0.681 0.378 0.378 0.538

Martin, mg/dL <0.001 0.090 0.934 0.224 0.224 0.541

Friedewald, mg/dL <0.001 0.003 0.541 0.461 0.461 0.541

Positive absolute value Martin, mg/dL 0.605 <0.001 0.003 <0.001 0.002 0.149

Positive absolute value Sampson, mg/dL 0.003 <0.001 0.968 0.581 0.248 0.140

Positive absolute value Friedewald, mg/dL <0.001 <0.001 <0.001 <0.001 0.004 0.810

S3 Table. Characteristics of the study population with triglyceride levels of <150 mg/dL

Smoke status

Adjusted p-values (Benjamini–Hochberg method) Never Never Never Former Former Current

Former Current Electronic Current Electronic Electronic

Age, years <0.001 <0.001 0.026 <0.001 <0.001 <0.001

Male, % <0.001 <0.001 <0.001 0.2964 0.2385 0.49

Body mass index, kg/m2 <0.001 0.008 0.074 <0.001 0.253 0.454

Laboratory data 　 　 　 　 　 　

Fasting glucose, mg/dL <0.001 0.021 0.302 <0.001 0.012 0.959

Blood urea nitrogen, mg/dL <0.001 <0.001 0.475 <0.001 <0.001 0.014

Creatinine, mg/dL <0.001 <0.001 <0.001 <0.001 0.024 0.845

GFR_EPI <0.001 <0.001 0.700 <0.001 <0.001 <0.001

Cholesterol 　 　 　 　 　 　

Total cholesterol, mg/dL 0.904 0.369 0.904 0.369 0.904 0.904

Triglyceride, mg/dL <0.001 <0.001 0.004 0.611 0.857 0.857

HDL-C, mg/dL <0.001 <0.001 0.044 0.410 0.730 0.882

Direct LDL-C, mg/dL 0.555 0.779 0.863 0.546 0.906 0.779

Non-HDL-C, mg/dL 0.036 0.843 0.835 0.099 0.835 0.835

Sampson, mg/dL 0.297 0.340 0.886 0.120 0.678 0.784

Martin, mg/dL 0.159 0.548 0.972 0.126 0.692 0.797

Friedewald, mg/dL 0.282 0.282 0.818 0.096 0.656 0.784

Positive absolute value Martin, mg/dL 0.552 0.030 0.500 0.234 0.552 0.887

Positive absolute value Sampson, mg/dL 0.597 0.363 0.402 0.402 0.402 0.363

Positive absolute value Friedewald, mg/dL 0.600 0.600 0.805 0.988 0.600 0.600

3. The statement in line 110 should indicate that TG <150 is excluded.

▶Response: Thank you for pointing this out. Accordingly, the corresponding text has been modified as follows:

“At each TG level, the current smoker group exhibited higher TG levels (except those with TG levels <150 mg/dL) and dLDL-C levels (except those with TG levels <150 mg/dL) than the never smoker group (except those with TG levels <150 mg/dL).”

4. The font sizes in the axes and titles of all figures should be increased to improve readability.

▶Response: Thank you for this helpful comment. Accordingly, we have increased the font sizes of all axes and titles in the figures.

5. The definition of overall precision as the ratio of direct LDL-C (dLDL-C) to non-high-density lipoprotein (non-HDL-C) should be supported by a reference or a clear rationale, such as citing relevant studies or providing a logical explanation for why this ratio is an appropriate measure

▶Response: Thank you for your insightful comments.

Non-HDL-C, calculated by subtracting HDL-C from total cholesterol, includes all particles that cause cardiovascular disease, namely, LDL-C, very low-density lipoprotein (VLDL), intermediate density lipoprotein, and lipoprotein A cholesterol. Non-HDL-C does not distinguish between LDL-C and remnant cholesterol and is considered a comprehensive target for managing cardiovascular risk. Clinically, LDL-C levels are related to non-HDL-C values in most cases; however, some patients have significantly higher non-HDL-C levels than LDL-C levels.

1. Nordestgaard B.G. Quantifying atherogenic lipoproteins for lipid-lowering strategies: consensus-based recommendations from EAS and EFLM. Atherosclerosis. 2020; 294: 46-61

2. Nordestgaard B.G. Triglyceride-rich lipoproteins and atherosclerotic cardiovascular disease: new insights from epidemiology, genetics, and biology. Circ Res. 2016; 118: 547-563

Accordingly, the corresponding text has been modified as follows: Method statistical analyses 1st pararagh

“Overall precision was defined as the ratio of direct LDL-C (dLDL-C) to estimated LDL-C and non-high-density lipoprotein (NHDL). (S2 Fig, S3 Fig and S4 Fig).”

6. In Figure 2, the LDL-C ranges should be explicitly labeled on the plots for better clarity.

▶Response: Thank you for pointing this out. We have now added labels for the LDL-C ranges and the axis indexes.

7. Figure 2 currently presents TG <150 mg/dL and TG <1000 mg/dL. It should categorize as TG ≥150–TG <400 mg/dL and TG ≥400–TG <1000 mg/dL. Including TG <150 mg/dL in the <1000 mg/dL category can lead to complexity in comparisons.

▶Response: Thank you for your insightful comment. To illustrate the LDL-C error ranges for TG <150 mg/dL and the more commonly used TG <1000 mg/dL, Figure 2 depicts the TG <150 and TG <1000 mg/dL ranges. Additional data for the TG ≥150 and <400 mg/dL and TG ≥400 and <1000 mg/dL groups have now been included in the supplementary material.

8. In line 212, the comment on TG categories is confusing. Table 1 presents data for TG <150 mg/dL, Table S4 for <400 mg/dL, and Table S5 for <1000 mg/dL. However, the text mentions TG levels <150 mg/dL, 150–400 mg/dL, and 400–1000 mg/dL. The TG categories should be clarified consistently throughout the manuscript.

▶Response: Thank you for your comment. In real-world applications, we divided the groups to show the error when TG levels are below 400 mg/dL and TG levels are 1000 mg/dL. In the real world, we classified the difference between TGs below 400 and 1000 differently from the existing categories to explain the variation by LDL level. It has been modified like this.

The Friedewald equation demonstrated higher MADs and MeADs than the other equations for TG levels <150 mg/dL, <400 mg/dL, <1000 mg/d, and 400-1000 mg/dL according to smoking status (Tables 1, S4,S5 and S6).

9. The issue of inconsistent TG categories appears in multiple places. The TG categories should be clearly defined and consistently used throughout the manuscript.

▶Response: Thank you for your comment. As mentioned in the response to the previous comment, the relevant data has been added and changes have now been made to the text.

10. In Figure 3, the legend should be reordered to present TG <150, 150–400 mg/dL, and 400–1000 mg/dL sequentially to improve the clarity of the plot.

▶Response: Thank you for your comment. We have changed the sequence of the TG groups as follows:

11. The statement in the results about the superiority of the Martin equation in Figure 3 should be moderated. The results suggest that the equations perform similarly.

▶Response: Thank you for your comment. Accordingly, we have changed the text as follows (Results - Precision of estimated LDL-C for the group with daily smoker status):

“The accuracies of the Martin and Sampson equations were similar to that of the Friedewald equation.”

12. The findings related to Figure 4 should be included in the manuscript to provide a complete interpretation of the results.

▶Response: Thank you for your comment. Accordingly, we have added the following text to the manuscript (Results - Precision of estimated LDL-C for the group with daily smoker status):

“In the EC smoker group with TG levels <400 mg/dL and >400 mg/dL, the MADs, MeADs, and correlation coefficients for the Sampson and Martin equations were similar. The Friedewald equation exhibited a higher difference than the other equations.”

13. The performance of the equations in the category of LDL-C <70 mg/dL and TG >400 mg/dL should be evaluated. These results could provide valuable insights into the literature.

▶Response: Thank you for your valuable suggestions. Accordingly, we have provided the corresponding data in S6 Table and added the following text to the Discussion (Differences in LDL-C and dLDL-C in current smokers):

“In case of LDL-C levels <70 mg/dL and TG levels >400 mg/dL, the MAD of the Sampson equation was lower than those of the other equations without never smokers. In case of LDL-C levels <70 mg/dL and TG levels > 400 mg/dL with never smokers, the Martin equation had a lower MAD than the other equations.”

S6 Table. Mean and median absolute deviations with 95% confidence intervals of estimated low-density lipoprotein cholesterol stratified by dLDL-C in the group with TG levels >400 mg/dL and <1000 mg/dL

Sampson equation Martin equation Friedewald equation

MAD MeAD MAD MeAD MAD MeAD

Never smoker

9.81 -2.52 (-11.16 to 6.12) 9.49 -1.78 (-10.29 to 6.74) 14.11 -10.06 (-20.09 to -0.04)

Former smoker

9.91 -5.34 (-9.82 to -0.85) 10.36 -7.12 (-11.96 to -2.27) 16.99 -15.21 (-21.71 to -8.72)

Current smoker

11.87 -9.34 (-13.88 to -4.80) 12.45 -7.56 (-11.66 to -3.47) 20.23 -17.49 (-23.57 to -11.42)

Electronic smoker

12.83 -15.12 (-39.4 to 9.16) 13.00 -13.71 (-37.40 to 9.97) 17.05 -19.45 (-50.38 to 11.47)

14. A more detailed discussion comparing the results with existing literature should be provided. Comparing these findings with studies conducted in different populations would help to understand the consistency and differences in the results

▶Response: Thank you for your comment. However, although numerous reports on LDL validation have been published recently, it is difficult to make comparisons due to insufficient literature on validation according to smoking status. In fact, the lack of data in this context was the main motivation behind designing and conducting this study.

Attachment

Submitted filename: comment_EC-PLOS_-1.docx

pone.0309002.s012.docx (102.6KB, docx)

Decision Letter 1

Shukri AlSaif

5 Aug 2024

More precise method of low-density lipoprotein cholesterol estimation for tobacco and electronic cigarette smokers: a cross-sectional study

PONE-D-24-08009R1

Dear Dr. Bae,

We’re pleased to inform you that your manuscript has been judged scientifically suitable for publication and will be formally accepted for publication once it meets all outstanding technical requirements.

Within one week, you’ll receive an e-mail detailing the required amendments. When these have been addressed, you’ll receive a formal acceptance letter and your manuscript will be scheduled for publication.

An invoice will be generated when your article is formally accepted. Please note, if your institution has a publishing partnership with PLOS and your article meets the relevant criteria, all or part of your publication costs will be covered. Please make sure your user information is up-to-date by logging into Editorial Manager at Editorial Manager® and clicking the ‘Update My Information' link at the top of the page. If you have any questions relating to publication charges, please contact our Author Billing department directly at authorbilling@plos.org.

If your institution or institutions have a press office, please notify them about your upcoming paper to help maximize its impact. If they’ll be preparing press materials, please inform our press team as soon as possible -- no later than 48 hours after receiving the formal acceptance. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org.

Kind regards,

Shukri AlSaif

Academic Editor

PLOS ONE

Review Comments to the Author

Reviewer #1: Comments from the reviewer has been properly addressed and the manuscript has been improved.

Thanks for your effort.

Reviewer #2: I would like to express my gratitude to the authors for their attention to and implementation of the suggested corrections.

Acceptance letter

Shukri AlSaif

12 Sep 2024

PONE-D-24-08009R1

PLOS ONE

Dear Dr. Bae,

I'm pleased to inform you that your manuscript has been deemed suitable for publication in PLOS ONE. Congratulations! Your manuscript is now being handed over to our production team.

At this stage, our production department will prepare your paper for publication. This includes ensuring the following:

* All references, tables, and figures are properly cited

* All relevant supporting information is included in the manuscript submission,

* There are no issues that prevent the paper from being properly typeset

If revisions are needed, the production department will contact you directly to resolve them. If no revisions are needed, you will receive an email when the publication date has been set. At this time, we do not offer pre-publication proofs to authors during production of the accepted work. Please keep in mind that we are working through a large volume of accepted articles, so please give us a few weeks to review your paper and let you know the next and final steps.

Lastly, if your institution or institutions have a press office, please let them know about your upcoming paper now to help maximize its impact. If they'll be preparing press materials, please inform our press team within the next 48 hours. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org.

If we can help with anything else, please email us at customercare@plos.org.

Thank you for submitting your work to PLOS ONE and supporting open access.

Kind regards,

PLOS ONE Editorial Office Staff

on behalf of

Dr. Shukri AlSaif

Academic Editor

PLOS ONE

Associated Data

    This section collects any data citations, data availability statements, or supplementary materials included in this article.

    Supplementary Materials

    S1 Fig. Flow diagram of the inclusion and exclusion criteria for participants from the KNHANES (2009–2021).

    KNHANES, Korea National Health and Nutrition Examination Survey; LDL-C, low-density lipoprotein cholesterol; dLDL-C, direct low-density lipoprotein cholesterol.

    (DOCX)

    pone.0309002.s001.docx (69.4KB, docx)
    S2 Fig. Correlation of estimated vs. direct low-density lipoprotein cholesterol (LDL-C) levels in patients with smoking status.

    A. The Sampson equation vs. direct LDL-C for never smokers. B. The Martin equation vs. direct LDL-C for never smokers. C. The Friedewald equation vs. direct LDL-C for never smokers. D. The Sampson equation vs. direct LDL-C for former smokers. E. The Martin equation vs. direct LDL-C for former smokers. F. The Friedewald equation vs. direct LDL-C for former smokers. G. The Sampson equation vs. direct LDL-C for current smokers. H. The Martin equation vs. direct LDL-C for current smokers. I. The Friedewald equation vs. direct LDL-C for current smokers. J. The Sampson equation vs. direct LDL-C for electronic cigarette smokers. K. The Martin equation vs. direct LDL-C for electronic cigarette smokers. L. The Friedewald equation vs. direct LDL-C for electronic cigarette smokers. Cor, correlation coefficient; MAE, mean absolute error; R2, correlation coefficient; RMSE, root mean square error; TG, triglyceride. The dots indicate the individual samples colored according to TG level. The color scale indicates individuals. SI conversion factors: To convert cholesterol to mmol/L, values were multiplied by 0.0259.

    (DOCX)

    pone.0309002.s002.docx (2.4MB, docx)
    S3 Fig. Mean absolute deviations of directly measured low-density lipoprotein cholesterol (dLDL-C) stratified by estimated LDL-C.

    (A) MAD of estimated LDL-C vs. dLDL-C for the never smoker group (TG levels <1000 mg/dL). (B) MAD of estimated LDL-C vs. dLDL-C for the never smoker group (TG levels <150 mg/dL). (C) MAD of estimated LDL-C vs. dLDL-C for the former smoker group (TG levels <1000 mg/dL). (D) MAD of estimated LDL-C vs. dLDL-C for the former smoker group (TG levels <150 mg/dL). (E) MAD of estimated LDL-C vs. dLDL-C for the current smoker group (TG levels <150 mg/dL). (F) MAD of estimated LDL-C vs. dLDL-C for the current smoker group (TG levels <150 mg/dL). (G) MAD of estimated LDL-C vs. dLDL-C for the EC smoker group (TG levels <150 mg/dL). (H) MAD of estimated LDL-C vs. dLDL-C for the EC smoker group (TG levels <150 mg/dL). HDL, high-density lipoprotein; TG, triglyceride; EC, electronic cigarette. Sampson equation (blue), Martin equation (red), and Friedewald equation (green).

    (DOCX)

    pone.0309002.s003.docx (1.5MB, docx)
    S4 Fig. Mean absolute deviations of directly measured low-density lipoprotein cholesterol (dLDL-C) stratified by non-high-density lipoprotein (NHDL).

    (A) MAD of estimated LDL-C vs. dLDL-C for the never smoker group (TG levels <1000 mg/dL). (B) MAD of estimated LDL-C vs. dLDL-C for the former smoker group (TG levels <1000 mg/dL). (C) MAD of estimated LDL-C vs. dLDL-C for the current smoker group (TG levels <150 mg/dL). (D) MAD of estimated LDL-C vs. dLDL-C for the EC smoker group (TG levels <150 mg/dL. NHDL, non-high-density lipoprotein; TG, triglyceride; EC, electronic cigarette. Sampson equation (blue), Martin equation (red), and Friedewald equation (green).

    (DOCX)

    pone.0309002.s004.docx (146.7KB, docx)
    S5 Fig. Residual error plots for low-density lipoprotein cholesterol (LDL-C) level estimated using equations according to smoking status (triglyceride levels <1000 mg/dL).

    Differences between the estimates obtained by the Sampson equation (A), Martin equation (B), and Friedewald equation (C) and direct LDL-C (dLDL-C), stratified by triglyceride (TG) level. Residual error was calculated by the difference between direct LDL-C (dLDL-C) and the values obtained using the equations. The dots indicate the individual samples, colored according to smoking status. The color scale indicates individuals. The solid line indicates the trend by the local regression method. Cor, correlation coefficient; MAD, mean absolute deviation; MAE, mean absolute error; R2 indicates correlation coefficient.

    (DOCX)

    pone.0309002.s005.docx (1.5MB, docx)
    S1 Table. Characteristics of study population with triglyceride levels of ≥400 mg/dL and <1000 mg/dL.

    (DOCX)

    pone.0309002.s006.docx (20.5KB, docx)
    S2 Table. Characteristics of the study population with triglyceride levels of ≥150 mg/dL and <400 mg/dL.

    (DOCX)

    pone.0309002.s007.docx (21.6KB, docx)
    S3 Table. Characteristics of the study population with triglyceride levels of <150 mg/dL.

    (DOCX)

    pone.0309002.s008.docx (22.3KB, docx)
    S4 Table. Mean and median absolute deviations with 95% confidence intervals of estimated low-density lipoprotein cholesterol stratified by dLDL-C in the group with TG levels of <400 mg/dL.

    (DOCX)

    pone.0309002.s009.docx (20.3KB, docx)
    S5 Table. Mean and median absolute deviations with 95% confidence intervals of estimated low-density lipoprotein cholesterol stratified by dLDL-C in the group with TG levels of <1000 mg/dL.

    (DOCX)

    pone.0309002.s010.docx (19.9KB, docx)
    S6 Table. Mean and median absolute deviations with 95% confidence intervals of estimated low-density lipoprotein cholesterol stratified by dLDL-C in the group with TG levels of >400 mg/dL and <1000 mg/dL.

    (DOCX)

    pone.0309002.s011.docx (21.6KB, docx)
    Attachment

    Submitted filename: comment_EC-PLOS_-1.docx

    pone.0309002.s012.docx (102.6KB, docx)

    Data Availability Statement

    KNHANES data are available online (https://knhanes.cdc.go.kr/knhanes/eng/index.do).

    Articles from PLOS ONE are provided here courtesy of PLOS

    RESOURCES