Skip Navigation LinksHome > January 2010 - Volume 53 - Issue 1 > Friedewald Equation Underestimates Low-Denisty Lipoprotein E...
JAIDS Journal of Acquired Immune Deficiency Syndromes:
doi: 10.1097/QAI.0b013e3181c0b05c
Letters to the Editor

Friedewald Equation Underestimates Low-Denisty Lipoprotein Elevations for Patients With High Triglyceride Levels in the ARTEMIS and TITAN Trials

Nelson, Mark MD*; DeMasi, Ralph PhD†; Moecklinghoff, Christiane MD‡; Hill, Andrew M PhD§‖

Free Access
Article Outline
Collapse Box

Author Information

*St. Stephens Centre, Chelsea and Westminster Hospital, London, United Kingdom; †Tibotec Research and Development, Yardley, PA.; ‡Janssen-Cilag, Mechelen, Belgium; §University of Liverpool, Liverpool, United Kingdom and ‖Tibotec Research and Development, Mechelen, Belgium

To the Editors:

The standard method of indirectly calculating low density lipoprotein (LDL) levels, using the Friedewald equation, may underestimate the number of patients with grade 3 elevations of LDL in the ARTEMIS (TMC114-C211) and TITAN (TMC114-C214) trials. This was a particular problem for patients treated with lopinavir/ritonavir, which raises triglycerides above levels, where the Friedewald equation can normally be used.

LDL is an important marker of cardiovascular risk and can be measured either directly or indirectly. Current National Cholesterol Education Program (NCEP) treatment guidelines recommend maintaining LDL levels at least below 130 mg/dL (<3.4 mmol/L), and preferably below 100 mg/dL (<2.6 mmol/L), for those with a high risk of cardiovascular disease.1 In HIV clinical trials, LDL elevations are normally reported by the AIDS Clinical Trials Group grading scale.2 Grade 3 elevations of LDL (above 190 mg/dL or 4.9 mmol/L) are considered high enough to justify starting cholesterol lowering drugs, even for patients with a low (<1%) 10-year risk of cardiovascular disease.1

Direct measurement of LDL cholesterol, using ultracentrifugation and precipitation, (known as “beta quantification”) is cumbersome and time consuming and requires expensive instrumentation and trained personnel.3 LDL cholesterol can also be calculated indirectly using the Friedewald equation4:

This equation assumes that virtually all plasma triglyceride is carried on VLDL and that the cholesterol to triglyceride ratio of VLDL is constant at around 5:1. The Friedewald equation is not valid if triglyceride levels are above 400 mg/dL (>4.52 mmol/L). For patient samples with these high triglyceride levels, the LDL value is not normally reported.4 Therefore, if drugs raise triglyceride levels, use of the Friedewald equation may miss important elevations in LDL cholesterol. In previous clinical trials, lopinavir/ritonavir has been shown to elevate triglyceride levels more than other antiretrovirals.5-8

Equation (Uncited)
Equation (Uncited)
Image Tools

The ARTEMIS trial evaluated lopinavir/ritonavir versus darunavir/ritonavir 800/100 mg OD in treatment-naive patients, in combination with tenofovir/emtricitabine.7 The TITAN trial evaluated lopinavir/ritonavir versus darunavir/ritonavir 600/100 mg twice a day in treatment-experienced patients with HIV-1 RNA >1000 copies per milliliter, in combination with optimized nucleoside reverse transcriptase inhibitor/nonnucleoside reverse transcriptase inhibitor combinations.8 In both trials, fasting lipid data were collected at baseline and through 48-96 weeks of randomized treatment. LDL was calculated using the Friedewald equation, and then either (1) excluding LDL data from patients with triglyceride levels above 400 mg/dL or (2) setting triglycerides equal to 400 mg/dL for those with higher levels.

Results from this analysis are shown in Table 1: the percentage of patients with grade 3 elevations in LDL was analyzed by treatment arm and time in the ARTEMIS and TITAN trials.

Table 1
Table 1
Image Tools

In ARTEMIS, the darunavir/ritonavir arm led to significantly lower mean triglyceride levels at week 48 (142 mg/dL or 1.6 mmol/L) and remained below the NCEP guidelines for intervention. In the lopinavir/ritonavir arm, mean triglycerides rose to 195 mg/dL (2.2 mmol/L) at Week 48, which was above the NCEP levels for intervention. There were 179 patient visits in the lopinavir/ritonavir arm with no LDL calculated owing to high triglycerides versus 48 such patient visits in the darunavir/ritonavir arm. In the lopinavir/ritonavir arm, patients with missing LDL levels at week 48 had mean total cholesterol of 6.0 mmol/L, versus 4.8 mmol/L for those with available LDL levels.

In the TITAN trial, the darunavir/ritonavir arm also led to significantly lower triglyceride levels at week 48 (221 mg/dL or 2.5 mmol/L) versus lopinavir/ritonavir (283 mg/dL or 3.2 mmol/L).

In both the ARTEMIS and TITAN trial, imputing triglycerides of 400 mg/dL (4.51 mmol/L) for patients with higher triglyceride levels resulted in more grade 3 elevations in LDL being identified compared with using the normal Friedewald equation (Table 1). The percentage of patients with grade 3 elevations in LDL was higher in both arms when calculated with the modified method but particularly in the lopinavir/ritonavir arm.

In summary, the Friedewald equation, used to calculate LDL in the ARTEMIS and TITAN trials, underestimated the number of patients with LDL elevations because LDL was not reported for patient samples with triglyceride levels above 400 mg/dL. A significant percentage of these patients may have LDL levels high enough to need intervention with lipid-lowering drugs. Analysis of the AIDS Clinical Trials Group 5087 trial has also showed that the Friedewald equation underestimates the actual levels of LDL for HIV-infected individuals with high triglyceride levels.9 In both the TITAN and ARTEMIS trials, the Friedewald equation missed fewer grade 3 LDL elevations for the darunavir/ritonavir arm, which raised triglycerides less. Assuming a value of 400 mg/dL for triglyceride levels above this level allows the identification of more patients with LDL grade 3 elevations. This is an approximation, and probably overestimates the LDL level, because the triglyceride fraction of the Friedewald equation has been set to a lower level than is measured. For patients with triglycerides above 400 mg/dL (4.51 mmol/L), the best alternative is direct measurement of LDL using beta quantification.

Mark Nelson, MD*

Ralph DeMasi, PhD†

Christiane Moecklinghoff, MD‡

Andrew M. Hill, PhD§∥

*St. Stephens Centre, Chelsea and Westminster Hospital, London, United Kingdom

†Department of Statistics, Tibotec Research and Development, Yardley, PA.

‡Clinical Research, Janssen-Cilag, Mechelen, Belgium

§Pharmacology Research Laboratories, University of Liverpool, Liverpool, United Kingdom

∥Tibotec Research and Development, Mechelen, Belgium

Back to Top | Article Outline


1. Grundy S, Cleeman J, Merz N, et al. Implications of recent clinical trials of the National Cholesterol Education Program Adult Treatment Panel III Guidelines. Circulation. 2004;110:227-239. Erratum in Circulation. 110:763.

2. ACTG 2007. AIDS Clinical Trials Group. Division of AIDS table for grading the severity of adult and paediatric adverse events. Available at: Accessed March 2007.

3. Nauck M, Warnick R, Rifai N. Methods for measurement of LDL-cholesterol: a critical assessment of direct measurement by homogenous assays versus calculation. Clin Chem. 2002;48:236-254.

4. Friedewald WT, Levy RL, Fredrickson D. Estimation of the concentration of low-density lipoprotein cholesterol in plasma, without use of the preparative ultracentrifuge. Clin Chem. 1972;18:499-502.

5. Molina J, Andrade-Villanueva J, Echevarria J, et al. Once-daily atazanavir/ritonavir versus twice-daily lopinavir/ritonavir, each in combination with tenofovir/emtricitabine, for management of antiretroviral-naïve HIV-1 infected patients: 48 week efficacy and safety results of the CASTLE study. Lancet. 2008;372:646-651.

6. Walmsley S, Bernstein B, King M, et al. Lopinavir-ritonavir versus nelfinavir for the initial treatment of HIV-1 infection. N Engl J Med. 2002;346:2039-2046.

7. Ortiz R, DeJesus E, Khanlou H, et al. Efficacy and safety of once-daily darunavir/ritonavir versus lopinavir-ritonavir in treatment naïve HIV-1 infected patients at Week 48. AIDS. 2008;22:1389-1397.

8. Madruga J, Berger D, McMurchie M, et al. Efficacy and safety of darunavir-ritonavir compared with that of lopinavir-ritonavir at 48 weeks in treatment-experienced, HIV-infected patients in TITAN: a randomized, controlled Phase III trial. Lancet. 2007;370:49-58.

9. Evans S, Fichtenbaum C, Aberg J, for the A5087 Study Team. Comparison of direct and indirect measurement of LDL-C in HIV-infected individuals: ACTG 5087. HIV Clin Trials. 2007;8:45-52.

© 2010 Lippincott Williams & Wilkins, Inc.