LDL-C or apoB as the Best Target for Reducing Coronary Heart Disease

Should Apob be Implemented into Clinical Practice?

Helena Vaverkova


Clin Lipidology. 2011;6(1):35-48. 

In This Article

Correlation Analyses

Ballantyne et al. used the data from the Measuring Effective Reductions in Cholesterol Using Rosuvastatin Therapy (MERCURY) II study (comparing the efficacy and safety of the lipid-lowering agents atorvastatin and simvastatin with rosuvastatin in high-risk subjects with type IIa and IIb hypercholesterolemia) and performed a correlation analysis between LDL-C and apoB and between non-HDL-C and apoB (with apoB as the independent variable) at the baseline (before statin treatment) and at the end of the study (after statin treatment).[60] In untreated patients, the apoB target of 90 mg/dl was approximately equivalent to an LDL-C level of 100 mg/dl and non-HDL-C level of 130 mg/dl according to the regression equation, which is consistent with the existing ATP III lipoprotein guidelines.[1] However, they found that treatment with statins alters the relationship between apoB and LDL-C and non-HDL-C, and as a consequence of reaching an apoB target of less than 90 mg/dl, it was necessary to reduce LDL-C to less than 70 mg/dl (in high TG patients) or less than 80 mg/dl (in lower TG patients) and non-HDL-C to less than 100 mg/dl. These data are consistent with the more aggressive optional cholesterol goals suggested for patients at very high risk of CHD.[1] On the basis of their results, they consider measurement of apoB during statin therapy to be one option for making sure that the number of atherogenic particles is reduced. However, on the basis of the good correlation between non-HDL-C and apoB at baseline and particularly on statin therapy, they recommend measurement of non-HDL-C (without additional cost) as another approach.[60]

Very similar results of calculated LDL-C and non-HDL-C levels corresponding to apoB levels of 90 mg/dl were obtained in another correlation analysis[61] using a database from the IN-CROSS study that compared treatment with ezetimibe/simvastatin 10/20 mg and rosuvastatin 10 mg in 618 high-risk patients not at their LDL-C goal despite previous use of statins.[64] This correlation analysis also suggested that on treatment with ezetimibe/simvastatin 10/20 mg or rosuvastatin 10 mg, more aggressive LDL-C and non-HDL-C targets (i.e., 70 and 100 mg/dl, respectively) should be achieved in order to normalize the concentration of apoB-containing atherogenic lipoproteins.

The results of both aforementioned correlation analyses could be explained by the findings of Cromwell et al..[10] These authors demonstrated that LDL particles (assessed by NMR) are more cholesterol-depleted when LDL concentrations are lower, independent of TGs or LDL particle size. This helps to explain why patients with low LDL-C often have disproportionately higher numbers of LDL particles (and thus apoB).

Charlton-Menys et al. performed another correlation analysis with LDL-C as an independent variable[50] in a database of patients with T2D included in the CARDS.[49] In this study, an LDL-C concentration of 1.8 mmol/l (70 mg/dl) during atorvastatin treatment was equivalent to a non-HDL-C level of 2.59 mmol/l (100 mg/dl) or an apoB level of 0.8 g/l (80 mg/dl). However, at the more conservative LDL-C targets of 2.59 mmol/l (100 mg/dl) and non-HDL-C of 3.37 mmol/l (130 mg/dl), apoB exceeded the value of 0.9 g/l.

Thus, it is evident from the aforementioned correlation analyses[50,60,61] that the results of correlations depend on the population studied, whether subjects are on treatment or not and also probably on which drugs were used and in which doses.

If LDL-C or non-HDL-C were used as the only goals (without apoB), very low target levels of these parameters should be used, according to the results of these correlation analyses, to obtain a reasonable population percentile level of apoB in the majority of patients (LDL-C <70 mg/dl [2nd population percentile] and non-HDL-C of <100 mg/dl [~7th population percentile]). In this situation, a certain subset of the patients would probably be unnecessarily treated to very low LDL-C levels (below the 5th percentile of the population). Thus, it is obvious that in clinical practice apoB cannot be calculated from non-HDL-C in spite of the good correlation between these two parameters. Furthermore, a good correlation between two parameters does not necessarily mean that they represent the same risk and thus a good correlation is of limited clinical value. For example, Sattar et al. demonstrated in the Insulin Resistance Study database that apoB was more closely related to waist circumference, fasting insulin, insulin sensitivity, fasting glucose, 2-h glucose, C-reactive protein, systolic blood pressure, fibrinogen and plasminogen activator inhibitor-1 than non-HDL-C.[65]

In another study of dyslipidemic patients not treated with hypolipidemic drugs (baseline examination), a very good correlation between LDL-C and apoB (r = 0.903; p < 0.001) was found.[66] In spite of this result, the subjects with apoB levels higher than predicted by a regression equation of apoB versus LDL-C had a significantly higher sex- and age-adjusted BMI, waist circumference, insulin, homeostasis model assessment (fasting insulin × fasting glycemia/22.5), C-peptide, proinsulin, plasminogen activator inhibitor-1, sICAM-1, sVCAM-1, tissue plasminogen activator, von Willebrand factor, frequency of metabolic syndrome and lower values of HDL-C, total cholesterol and LDL-C in comparison with subjects with apoB levels lower than predicted by LDL-C (p < 0.001–0.05 for all parameters). Thus, individuals with apoB levels higher than predicted by their LDL-C levels are more insulin resistant and have a more atherogenic risk profile. This means that, at least for dyslipidemic patients with high cardiometabolic risk, apoB is a more appropriate marker of risk than LDL-C.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as: