• Users Online: 77
  • Print this page
  • Email this page


 
 
Table of Contents
ORIGINAL ARTICLE
Year : 2020  |  Volume : 8  |  Issue : 2  |  Page : 60-64

Estimation of non-high-density lipoprotein cholesterol and its correlation as a surrogate measure of apolipoprotein B in patients with metabolic syndrome


Department of General Medicine, Bangalore Medical College and Research Institute, Bengaluru, Karnataka, India

Date of Submission18-Sep-2019
Date of Acceptance12-Jan-2020
Date of Web Publication18-Apr-2020

Correspondence Address:
Dr. B Sumana
Department of General Medicine, Bangalore Medical College and Research Institute, Bengaluru - 560 002, Karnataka
India
Login to access the Email id

Source of Support: None, Conflict of Interest: None


DOI: 10.4103/AJIM.AJIM_59_19

Rights and Permissions
  Abstract 


Background: The primary target of therapy for dyslipidemia according to the present lipid guidelines is low-density lipoprotein cholesterol (LDL-C). Its main drawback is it is a derived value and erroneous values may be obtained if triglyceride values are high. Hence, direct measurement of apolipoprotein B (Apo-B), which is present in all atherogenic particles (very-LDL, intermediate-density lipoprotein, and LDL) would be ideal and is a superior indicator of cardiovascular risk. However, current limitations of Apo-B estimation are lack of standardization and lack of technical expertise and cost factor. To overcome these limitations, there is a need to test a particular lipid value from the standard lipid test, which can be used as a surrogate measure of Apo-B. As non-high-density lipoprotein cholesterol (HDL-C) is a measure of the cholesterol content of all Apo-B-containing lipoproteins, it may represent a simple and inexpensive surrogate to Apo-B in select patient subgroups. Hence, this study was undertaken to estimate non-HDL-C and LDL-C from the standard lipid test and also measure Apo-B levels in patients with metabolic syndrome using appropriate statistical methods, correlation and concordance between non-HDL-C versus Apo-B and LDL-C versus Apo-B were analyzed. Methodology: This was a cross-sectional study carried out in Bangalore on 135 patients of metabolic syndrome fulfilling the International Diabetes Federation criteria, from January 2019 to March 2019. Detailed history and examination was undertaken. Fasting blood glucose, Apo-B-100, and lipid profile were obtained in the fasting state venous blood sample. Results: A significant positive correlation was noted between non-HDL-C levels and Apo-B values with a correlation coefficient of 0.883 and area under the curve (AUC) of 0.972. In contrast, the correlation between LDL-C and Apo-B-100 showed a correlation coefficient of 0.815 and AUC of 0.949. Similarly, the concordance/discordance analysis of the same parameters showed concordance across the range of corresponding non-HDL-C and Apo-B quintiles except in the middle quintile where the values were discordant. Concordance was >50% across 80% of the range of values between LDL-C and non-HDL-C. Conclusion: This study showed that there is a good correlation between non-HDL-C levels and Apo-B values with less discordance. Hence, non-HDL-C can be used as a simple and cost-effective alternative (surrogate) to Apo-B estimation.

Keywords: Apolipoprotein B-100, dyslipidemia, metabolic syndrome, non-high-density lipoprotein cholesterol


How to cite this article:
Prakash K G, Sumana B, Shetty AB. Estimation of non-high-density lipoprotein cholesterol and its correlation as a surrogate measure of apolipoprotein B in patients with metabolic syndrome. APIK J Int Med 2020;8:60-4

How to cite this URL:
Prakash K G, Sumana B, Shetty AB. Estimation of non-high-density lipoprotein cholesterol and its correlation as a surrogate measure of apolipoprotein B in patients with metabolic syndrome. APIK J Int Med [serial online] 2020 [cited 2020 Jul 13];8:60-4. Available from: http://www.ajim.in/text.asp?2020/8/2/60/282849




  Introduction Top


The hallmark of atherogenic dyslipidemia which is prevalent in persons with metabolic syndrome (with or without Type 2 diabetes mellitus [DM]) is elevated triglyceride levels and low levels of high-density lipoprotein cholesterol (HDL-C), while low-density lipoprotein cholesterol (LDL-C) level may only be marginally elevated. This marginally elevated or near-normal LDL-C levels occur because of the presence of very high levels of a special category of LDL particles called small dense LDL particles. These small dense LDL particles carry less cholesterol per molecule, get oxidized easily, have high atherogenic potential, and are more numerous. The high triglyceride and low HDL-C levels in metabolic syndrome are due to the presence of elevated very-LDL (VLDL) and intermediate-density lipoprotein (IDL) molecules which are also atherogenic. A single apolipoprotein B (Apo-B) molecule is present in all these major atherogenic particles of liver origin (VLDL, IDL, and LDL) and lipoprotein (a). Therefore, the measurement of Apo-B provides direct information on the number of atherogenic particles irrespective of their size.[1] However, current limitations for Apo-B estimations are lack of standardization and lack of accessibility with particular reference to cost and availability of expertise. To overcome these limitations, there is a need to test a particular lipid value from the standard lipid test, which can be used as a surrogate measure of Apo-B. As non-HDL-C actually is a measure of the cholesterol content of all Apo-B-carrying lipoproteins, it may represent a simple and inexpensive surrogate to Apo-B measurement, especially in selected patient subgroups, such as type 2 DM and metabolic syndrome.[2]

LDL-C levels are the main therapeutic goal of both diabetic and nondiabetic dyslipidemia, but their concentrations do not represent the whole mass of lipoprotein particles that also include IDL-C and VLDL-C which have also been shown to be atherogenic. In diabetes, small dense LDL particles are also elevated which are not reflected in the routine LDL-C measurement.[3] Apo-B has been associated with an increased risk of cardiovascular disease independently of LDL-C level in patients of metabolic syndrome.

The available evidence suggests the slightly better performance of Apo-B when compared with non-HDL-C, but practical limitations of such a paradigm shift make Apo-B more of an “additional” lipid test rather than an “advanced” lipid test. Although Apo-B looks very promising and may be of benefit in select patients, more definitive evidence is needed on establishing appropriate cutoff points and further delineating its benefits prior to implementation.[4]

Non-HDL-C in contrast demonstrates competitive performance compared with Apo-B and offers several benefits that Apo-B does not, including:

  1. Assessment of non-HDL-C is simple and easy, calculated by subtraction of HDL-C from total cholesterol levels
  2. Non-HDL is more practical and reliable
  3. Established cut points for treatment interventions are available based on LDL-C levels, which remain valid and independent of increasingly discrepant population percentiles
  4. Assessment of non-HDL-C is inexpensive.


These benefits may lead to cost savings with the use of non-HDL-C as a primary target by improving the overall management of CVD risk and minimizing immediate test repetition by physicians for confounded specimen samples. Hence, this study was undertaken to find out whether the values of non-HDL-C which are measured as a part of standard lipid test correlate with the values of Apo-B, which are measured by immunoassay, using appropriate statistical methods.[4]


  Methodology Top


This was a cross-sectional study carried out at a tertiary care center in Bangalore from January 2019 to March 2019. A total of 135 patients fulfilling the International Diabetes Federation criteria for metabolic syndrome for Southeast Asian population were included in the present study,[5] based on random sampling technique. Patients on anti-hypolipidemic agents and known cardio- or cerebrovascular disease were excluded from the study.

Hospital ethical committee clearance was obtained before the initiation of the study. After obtaining written informed consent, patient details such as age and sex, body mass index, abdominal circumference, history of smoking and alcohol consumption, and history of diabetes and hypertension were taken. Plasma blood glucose and all lipid values were obtained in the fasting state.

Blood glucose was estimated by the hexokinase method, lipids were measured using the Roche Cobas Integra 400 analyzer, total cholesterol by enzymatic cholesterol oxidase method, triglycerides by GPO-POD assay, and HDL by direct measurement. LDL-C was estimated using the Friedwald's equation. Apo-B-100 was measured by immunoturbidimetry method.


  Statistical Analysis Top


The results were analyzed using SPSS version 22 software of IBM. Four statistical methods were used to see the agreement between non-HDL-C and Apo-B-100 (1) Pearson's correlation test, (2) area under the curve (AUC), and (3) concordance–discordance analysis. “P” (Probability that the result is true) ≥0.05 was considered as statistically significant after assuming all the rules of statistical tests.


  Results Top


The average age of the patient population was 51 ± 10.8 years, with the majority of them being male (57%). There was a significant positive correlation between non-HDL-C and Apo-B-100, with correlation coefficient being 0.883 and AUC being 0.972. The concordance/discordance analysis of plasma Apo-B and non-HDL-C showed good concordance across the range of non-HDL-C and Apo-B quintiles, except in the middle quintile where the values were discordant. Concordance was greater at the extremes. However, concordance was >50% across 80% of the range of values. In contrast, the correlation between LDL-C and Apo-B-100 showed a correlation coefficient of 0.815, AUC of 0.949, and discordance across 80% of the range of values.

Patient characteristics are as shown in [Table 1].
Table 1: Characteristics of the patients in the study

Click here to view


The average age of the patients in the study was 51 ± 10.8 years. Seventy percent of them were in the age group of 40–60 years [Figure 1]. In the study, 57.7% were males and 42.3% were females.
Figure 1: Bar diagram showing age distribution of subjects

Click here to view


Pearson correlation between apolipoprotein B and non-high-density lipoprotein and low-density lipoprotein cholesterol

In the present study, there was a significant positive correlation between non-HDL-C and Apo-B-100 with a correlation coefficient of 0.883. It was found that with each increase in a limit of Apo-B-100, there was an increase in the corresponding limit of non-HDL-C linearly and vice versa. Similarly, there was also a positive correlation between Apo-B and LDL-C, but the correlation coefficient was lesser at 0.815 [Table 2].
Table 2: Correlation between apolipoprotein-B-100 and non-high-density lipoprotein and apolipoprotein B and low-density lipoprotein cholesterol

Click here to view


This is also well reflected in the scatter diagram where the values were less dispersed around the slope between non-HDL-C and Apo-B when compared to the slope between LDL-C ands Apo-B [Figure 2].
Figure 2: Scatter plot showing positive correlation between non-high-density lipoprotein cholesterol and APO-B-100 (a) and low-density lipoprotein cholesterol an APO-B-100 (b)

Click here to view


Assessment of the relation between apolipoprotein B and non-high-density lipoprotein cholesterol and low-density lipoprotein cholesterol using receiver operator curve analysis

To relate Apo-B with non-HDL-C, another statistical method was utilized, namely receiver operator curve (ROC). The AUC for the relation between Apo-B and non-HDL-C was 0.972 and AUC between Apo-B and LDL-C was 0.949, as shown in [Figure 3]. The ROC curve correlates the validity of the new test when compared with the gold standard, ROC value reaching 1 indicating perfect correlation. The ROC between Apo-B and non-HDL was reaching unity and the peak of the curve was almost up to the upper left corner, indicating an excellent correlation with the gold standard test, namely Apo-B.
Figure 3: Receiver operator characteristic curve showing validity of non-high-density lipoprotein cholesterol with respect to APO-B-100 (a) and low-density lipoprotein cholesterol with APO-B-100 (b)

Click here to view


Concordance–discordance analysis between apolipoprotein B and non-high-density lipoprotein cholesterol and apolipoprotein B and low-density lipoprotein cholesterol

Another statistical method of analysis was used to assess relation between Apo-B and non-HDL-C, namely the concordance–discordance analysis. In this method, the non-HDL-C and Apo-B values are arranged in an ascending order, and each value in non-HDL-C is compared with every other value in the Group-Apo-B. Thus, this method checks the agreement between Apo-B and non-HDL-C across the range of distribution of lipid values and is a better method of estimation of surrogacy.

The concordance/discordance analysis of plasma Apo-B and non-HDL-C is as shown in [Table 3]. Overall, across the range of non-HDL-C and Apo-B quintiles, except in the middle quintile, the values were concordant. Concordance was greater at the extremes, being 93% at the first quintile and 62% at the fifth quintile. The concordance was >50% across 80% of the range of values.
Table 3: Concordance analysis by quintiles of apolipoprotein B and non-high-density lipoprotein cholesterol levels in the study subjects

Click here to view


The concordance–discordance analysis was also performed between Apo-B and LDL-C. Apo-B and LDL were concordant only in the first quintile, i.e., 98% concordance [Table 4]. There was discordance in the four successive quintiles, indicating as the dyslipidemia increases small dense LDL increases which is not estimated by the routine LDL measurement.[6]
Table 4: Concordance analysis by quintiles of low-density lipoprotein and apolipoprotein B-100 levels

Click here to view



  Discussion Top


The metabolic syndrome consists of a constellation of metabolic abnormalities that confer increased risk of cardiovascular disease. Apo-B is a superior estimate of this cardiovascular disease over and above LDL-C and triglycerides since it measures all atherogenic particles including VLDL, IDL, and LDL including small dense LDL-C which is not measured by the traditional methods of LDL-C measurement.[7]

This was a cross-sectional study undertaken in search of a valid surrogate marker for Apo-B since it is a relatively new, expensive, and not widely available measurement of atherogenic risk. On the other hand, non-HDL-C can be readily calculated from the routinely tested standard lipid profile.[8]

The relationship between non-HDL-C and Apo-B-100 was analyzed using three statistical methods – Pearson's correlation, ROC analysis, and concordance–discordance ratio.

In this study, Pearson correlation applied between non-HDL-C and Apo-B showed a significant positive correlation between the two with r = 0.883 and P = <0.001. In a study by Sniderman et al.[9] done on 2103 men without coronary artery disease at the baseline, there was a similar high correlation between non-HDL-C and Apo-B-100 with r = 0.87 and P < 0.001.

In another study by Hermans et al.[10] on 45 diabetic patients, the Pearson correlation between non-HDL-C and Apo-B was r = 0.94 and P < 0.001 [Table 5].
Table 5: Statistical correlation

Click here to view


ROC analysis of non-HDL-C with Apo-B in this study showed AUC as 0.972. In a study by Zhang et al.,[11] in China, 5486 consecutive subjects underwent testing for non-HDL-C and Apo-B; ROC analysis between the two parameters showed AUC as 0.918 [Table 6].
Table 6: Area under curve (AUC)

Click here to view


Concordance/discordance analysis between non-HDL-C and Apo-B in this study showed >50% concordance across 80% of the quintiles. Similar results were observed in a study by Sniderman et al.,[9] with more concordance being observed in the extremes of data [Table 7].
Table 7: Concordance ratio

Click here to view


To the best our knowledge, this is a first study of its kind in the Indian population to check the validity of non-HDL-C as a valid surrogate marker of Apo-B in and to use both correlation and concordance–discordance to check the agreement between the two atherogenic indices.

All the current national and international dyslipidemia management guidelines are LDL centric. This study shows that LDL-C is a poor predictor of atherogenic risk at higher levels of dyslipidemia, especially in patients with metabolic syndrome and diabetes mellitus. It is time to start using Apo-B or its surrogate marker, atleast as an additional target in the management of dyslipidemia if not as the primary goal.[12]


  Conclusion Top


This study shows that non-HDL-C can be used as a valid surrogate marker of Apo-B-100 in patients with metabolic syndrome. Apo-B is a proven superior atherogenic risk indicator than LDL-C and non-HDL-C can be used as its surrogate providing a simple and cost-effective alternative.

Financial support and sponsorship

We would thank API Karnataka Chapter for the financial support of this study.

Conflicts of interest

There are no conflicts of interest.



 
  References Top

1.
Sniderman AD. Non-HDL cholesterol versus apolipoprotein B in diabetic dyslipoproteinemia: Alternatives and surrogates versus the real thing. Diabetes Care 2003;26:2207-8.  Back to cited text no. 1
    
2.
Expert Panel on Detection, Evaluation, and Treatment of High Blood Cholesterol in Adults. Executive summary of the third report of the national cholesterol education program (NCEP) expert panel on detection, evaluation, and treatment of high blood cholesterol in adults (Adult Treatment Panel III). JAMA 2001;285:2486-97.  Back to cited text no. 2
    
3.
Hermans MP, Ahn SA, Rousseau MF. The atherogenic dyslipidemia ratio [log(TG)/HDL-C] is associated with residual vascular risk, beta-cell function loss and microangiopathy in type 2 diabetes females. Lipids Health Dis 2012;11:132.  Back to cited text no. 3
    
4.
Ramjee V, Sperling LS, Jacobson TA. Non-high-density lipoprotein cholesterol versus apolipoprotein B in cardiovascular risk stratification: Do the math. J Am Coll Cardiol 2011;58:457-63.  Back to cited text no. 4
    
5.
Zimmet P, Alberti KG, Serrano Ríos M. A new international diabetes federation worldwide definition of the metabolic syndrome: The rationale and the results. Rev Esp Cardiol 2005;58:1371-6.  Back to cited text no. 5
    
6.
Zabeen S, Rahman M, Mustaf T, Eusufzai N, Shermin S. Non-HDL Cholesterol and Type 2 Diabetes Mellitus. Anwer Khan Modern Med Coll J 2012;3:15-8.  Back to cited text no. 6
    
7.
Kim BJ, Hwang ST, Sung KC, Kim BS, Kang JH, Lee MH, et al. Comparison of the relationships between serum apolipoprotein B and serum lipid distributions. Clin Chem 2005;51:2257-63.  Back to cited text no. 7
    
8.
Sniderman AD, Hogue JC, Bergeron J, Gagné C, Couture P. Non-HDL cholesterol and apoB in dyslipidaemia. Clin Sci (Lond) 2008;114:149-55.  Back to cited text no. 8
    
9.
Sniderman AD, St. Pierre AC, Cantin B, Dagenais GR, Després JP, Lamarche B. Concordance/discordance between plasma apolipoprotein B levels and the cholesterol indexes of atherosclerotic risk. Am J Cardiol 2003;91:1173-7.  Back to cited text no. 9
    
10.
Hermans MP, Sacks FM, Ahn SA, Rousseau MF. Non-HDL-cholesterol as valid surrogate to apolipoprotein B100 measurement in diabetes: Discriminant Ratio and unbiased equivalence. Cardiovasc Diabetol 2011;10:20.  Back to cited text no. 10
    
11.
Zhang GM, Goyal H, Zhang GM, Ma XB, Zhou YT. The “bad” cholesterol can predict abnormal apolipoprotein B levels in a large unselected outpatient cohort. Oncotarget 2018;9:8011-5.  Back to cited text no. 11
    
12.
Eckel RH, Cornier MA. Update on the NCEP ATP-III emerging cardiometabolic risk factors. BMC Med 2014;12:115.  Back to cited text no. 12
    


    Figures

  [Figure 1], [Figure 2], [Figure 3]
 
 
    Tables

  [Table 1], [Table 2], [Table 3], [Table 4], [Table 5], [Table 6], [Table 7]



 

Top
 
  Search
 
    Similar in PUBMED
   Search Pubmed for
   Search in Google Scholar for
 Related articles
    Access Statistics
    Email Alert *
    Add to My List *
* Registration required (free)  

 
  In this article
Abstract
Introduction
Methodology
Statistical Analysis
Results
Discussion
Conclusion
References
Article Figures
Article Tables

 Article Access Statistics
    Viewed676    
    Printed25    
    Emailed0    
    PDF Downloaded69    
    Comments [Add]    

Recommend this journal