Share this article on:

Reduction in Triglyceride Level With N-3 Polyunsaturated Fatty Acids in HIV-Infected Patients Taking Potent Antiretroviral Therapy: A Randomized Prospective Study

De Truchis, Pierre MD*; Kirstetter, Myriam MD; Perier, Antoine MPhil; Meunier, Claire§; Zucman, David MD; Force, Gilles MD; Doll, Jacques MD#; Katlama, Christine MD, PhD**; Rozenbaum, Willy MD††; Masson, Hélène MD‡‡; Gardette, Jean MD§; Melchior, Jean-Claude MD, PhD*and the Maxepa-HIV Group

JAIDS Journal of Acquired Immune Deficiency Syndromes: March 1st, 2007 - Volume 44 - Issue 3 - p 278-285
doi: 10.1097/QAI.0b013e31802c2f3d
Clinical Science

To assess the evolution of triglyceride (TG) levels in HIV-infected patients receiving stable potent antiretroviral therapy treated with N-3 polyunsaturated fatty acids (PUFAs), a prospective double-blind randomized design for a reliable assessment of TG evolution was performed. One hundred twenty-two patients with TG levels >2 g/L and ≤10 g/L after a 4-week diet (baseline TG: 4.5 ± 1.9 g/L) were randomized for 8 weeks to N-3 PUFAs (2 capsules containing 1 g of fish oil 3 times daily, n = 60), or placebo (1 g of paraffin oil capsules, n = 62). An 8-week open-label phase of N-3 PUFAs followed. Evaluation criteria were TG percent change at week 8, percentage of responders (normalization or ≥20% TG decrease), and safety issues. Ten patients with baseline TG levels >10 g/L were not randomized and received N-3 PUFAs as open treatment. The difference (PUFA - placebo) in TG percent change at week 8 was −24.6% (range: −40.9% to −8.4%; P = 0.0033), the median was −25.5% in the PUFA group versus 1% in the placebo group, and mean TG levels at week 8 were 3.4 ± 1.8 g/L and 4.8 ± 3.1 g/L, respectively. TG levels were normalized in 22.4% (PUFA) versus 6.5% (placebo) of patients (P = 0.013) with a ≥20% reduction in 58.6% (PUFA) versus 33.9% (placebo) of patients (P = 0.007). Under the open-label phase of N-3 PUFAs, the decrease in TG levels was sustained at week 16 for patients in the PUFA group (mean TG: 3.4 ± 1.7 g/L), whereas a 21.2% decrease in TG levels occurred for patients in the placebo group (mean TG: 3.3 ± 1.4 g/L). No significant differences were observed between groups in the occurrence of adverse events. The median TG change at week 8 was −43.6% (range: Q1-Q3; 95% CI: −66.5% to −4.6%) for patients with baseline TG levels >10 g/L. The difference in mean total cholesterol between groups (PUFA - placebo) at week 8 was −8.5% (P = 0.0117). This study demonstrated the efficacy of PUFAs to lower elevated TG levels in treated HIV-infected hypertriglyceridemic patients. N-3 PUFAs have a good safety profile.

From the *Hôpital Raymond Poincaré, Garches, France, University Paris-Ile-de-France-Ouest-Versailles; †Hôpital Saint-Antoine, Paris, France; ‡Cardinal Systems, Paris, France; §Innovation Développement Pierre Fabre, Ramonville, France; ∥Hôpital Foch, Suresnes, France; ¶Hôpital Perpétuel Secours, Levallois-Perret, France; #Hôpital A. Mignot, Versailles, France; **Hôpital Pitié-Salpétrière, Paris, France; ††Hôpital Tenon, Paris, France; and ‡‡Hôpital de Poissy-St. Germain, Poissy, France.

Received for publication May 21, 2006; accepted September 20, 2006.

Sponsored by the Institut de Recherche Pierre Fabre.

Reprints: Jean-Claude Melchior, MD, Hôpital Raymond Poincaré, 104, boulevard Raymond Poincaré, 92380 Garches, France, AP-HP, University Paris-Ile-de-France-Ouest-Versailles (e-mail:

The relation between HIV infection and metabolic disorders was established early in the knowledge of the disease;1-6 however, because of the short-term vital prognosis of AIDS, these associated disorders were not considered of major interest. Since 1996, the availability of protease inhibitors (PIs) and the adoption of combination strategies for antiretroviral drugs have dramatically changed the treatment and prognosis of patients infected with HIV. As early as 1997, secondary effects of these antiretroviral combinations were described. Highly active antiretroviral therapy (HAART) itself causes a metabolic syndrome characterized by lipodystrophy, lipoatrophy, dyslipidemia (including hypertriglyceridemia, which was reported to be related to cytokine abnormalities1), and insulin resistance in a high proportion of patients and may be associated with a higher cardiovascular risk than in the HIV-negative population.1,7-12 The risk of developing hyperlipidemia increases with the duration of treatment with PIs.13-17 The etiology of HIV-associated lipodystrophy and metabolic alterations seems to be multifactorial, including hormone and cytokine abnormalities;18-21 HIV drug inhibitory effects on adipocyte differentiation;22 alteration of mitochondrial functions in adipocytes; and altered leptin, adiponectin, and cytokine expression in adipose tissue of patients.22 The results of treatment trials of dyslipidemia in HIV-infected patients are now published.23-27 Hypertriglyceridemia itself has been recognized as an independent cardiovascular risk factor and should be considered in the management of HIV infection.6,28

The metabolic effects of N-3 polyunsaturated fatty acids (PUFAs) derived from marine sources (so-called “fish oils”) have been demonstrated to reduce fasting and postprandial triglyceride (TG) levels in individuals without HIV infection.29,30 PUFAs are now recommended in Europe (European Medicines Agency [EMEA]) for secondary prophylaxis after myocardial infarction. Moreover, these products are well tolerated and bear a high benefit/risk ratio.31,32 Few data are available on the effect of PUFAs on metabolic abnormalities in HIV patients,33 however, even though recent studies suggest possible efficacy and safety of fish oil supplementation in the treatment of antiretroviral therapy-associated hypertriglyceridemia.34

Maxepa is a fish oil formulation with a high concentration of N-3 fatty acids (18% eicosapentaenoic acid and 12% docosahexaenoic acid) that acts as a TG-lowering agent. A previous retrospective analysis of the use of N-3 PUFAs in hypertriglyceridemic HIV-infected patients demonstrated a trend of this product in treating the HAART-associated hypertriglyceridemia, without impairing virologic control of the disease (unpublished data, Pierre Fabre Laboratory, 1999). The study suggested good clinical and biological tolerability of Maxepa and a decrease in hypertriglyceridemia that has not been achieved previously with diet alone. Thus, it was interesting to test the efficacy of N-3 PUFAs in a prospective, randomized, controlled study in HIV-infected patients with sustained hypertriglyceridemia after diet alone.

Back to Top | Article Outline



Inclusion Criteria

Ambulatory patients of either gender, aged at least 18 years, and presenting with HIV infection treated with stable multiple antiretroviral therapy for at least 2 months were included in the study if the following criteria were fulfilled: (1) known and documented hypertriglyceridemia (≥3 g/L [3.43 mmol/L]) within 6 months before selection and remaining at least at the same level at the selection visit and (2) baseline TG level >2 g/L and <10 g/L after a 4-week TG-lowering diet. Included patients were randomized to 2 treatment groups (placebo + diet or N-3 PUFAs + diet). Patients with a TG level >10 g/L were not randomized for ethical consideration (because hypertriglyceridemia can lead to acute pancreatitis at such a level). It was proposed that they be treated with N-3 PUFAs plus diet on an open basis and separately analyzed.

Back to Top | Article Outline

Exclusion Criteria

Exclusion criteria included known hypersensitivity to the active compound or product excipients, pregnancy, breast-feeding, concomitant treatments with a possible influence on the interpretation of the results (anticoagulant treatment with international normalized ration [INR] >2.5) unless treatment had remained unchanged for 6 months before inclusion, oral antidiabetics and hormonal treatments (only effective oral hormonal contraception for women was tolerated [in fact, there was 1 such patient in each group], but patients receiving testosterone therapy were not included in the study) initiated or modified in the past 3 months, drugs that induce hepatic metabolism (except treatments for HIV infection), and insulin therapy. Lipid-lowering drugs were to have been stopped for more than 1 month before the selection visit. Patients who regularly consumed high levels of alcohol (>20 g/d) were not included in the study. No patients received vitamin E. Patients with an abnormal fasting blood glucose level were not included (glycemia >6.6 mmol/L).

The study was conducted in 12 French hospital units. The protocol was reviewed by the independent Ethics Committees of Saint-Germain-en-Laye, and informed written consent from the patients was obtained before inclusion. This study was conducted and monitored in compliance with the French regulatory guidelines (Loi Huriet), the Independent Community of Health (ICH) E6 guidelines for Good Clinical Practice, and the ethical principles enunciated in the revised Declaration of Helsinki.

Back to Top | Article Outline

Study Design

This was a double-blind, phase 4, randomized, 2-arm, parallel-group study. After a 4-week screening period on the basis of diet alone, eligible patients were randomized to N-3 PUFAs (2 1-g capsules 3 times daily) plus diet or placebo plus diet during an 8-week period (placebo was presented as ochre gelatin capsules similar to PUFA capsules containing 1 g of paraffin oil). This period of treatment was chosen because previous studies have indicated that after 2 months of treatment, no further effects on TG blood levels could be obtained.34 The randomization was stratified by 3 factors: TG level, gender of subject, and center. The diet period was initiated by a dietitian: all subjects received diet counseling in accordance with American Heart Association (AHA) recommendations at the beginning of the study.35 The regimen was adapted for each patient to obtain a diet with 15% to 18% proteins, 45% to 55% carbohydrates, and 30% to 35% lipids (as percentages of daily calorie intake) without alcohol consumption. The diet was explained to each patient by means of a written document, and the goal was to adhere to the diet throughout the study period. The double-blind treatment phase was followed by an 8-week period during which all patients received N-3 PUFAs plus the diet on an open basis. Patients were asked to bring back the empty cluster to calculate the compliance to treatment. The adherence to study drug was assessed globally between visit 2(V2) and visit 4(V4) using the following formula: Compliance (%) = 100 × (number of capsules delivered - number of capsules brought back)/Number of Days Elapsed × 6 (counting the empty clusters brought back). A compliance <75% was considered to be a major protocol violation; these patients were thus excluded from the per protocol (PP) efficacy population.

Fasting lipid parameters (TG, total cholesterol, and high-density lipoprotein cholesterol [HDL-C]) were measured at screening, before randomization, and under treatment at the end of each 8-week period. Blood samples were taken at home by a nurse of the study staff between 7:00 and 9:00 AM in the morning after a fasting period of at least 12 hours. All tests were performed by a centralized laboratory accredited for lipid analysis. The primary outcome parameter was the percentage variation in TG levels between the randomization (week 4) and week 12 visits (or last known value under double-blind treatment in case of premature withdrawal): TG% = 100 × (TGLKVW12 − TGBL)/TGBL, where LKVW12 indicates last known value at week 12 and BL indicates baseline. The percentage of responders (TG normalization or at least 20% decrease) was compared between groups. The definition of responders was established on the basis of the fact that a decrease of 20% in TG level was considered to be clinically relevant. The evolution of TG levels during the open period and the global evolution since randomization were also evaluated.

General adverse events were recorded, along with date of onset and resolution, intensity (subjective assessment of mild, moderate, or severe), seriousness and causal relation to study treatment in the investigator's opinion. Vital signs, standard laboratory tests, CD4+ cell count, CD8+ cell count, and viral load were assessed at baseline and at the end of each 8-week period.

The protocol design is presented in Figure 1.



Back to Top | Article Outline

Statistical Considerations

Using the percentage change from baseline of the TG level as the primary endpoint, a type I error rate (α) of 0.05 (2-sided), a power (1 − β) of 80%, and an SD of 40%, 64 patients per group were necessary to identify a 20% difference between treatment groups.

For the stratified randomization, a nondeterministic (75%) dynamic allocation procedure was used,36 stratifying by 3 factors: TG level, gender of subject, and center. A randomization test was performed to validate the randomization algorithm. A total of 10,000 data sets were generated using the same algorithm that had been used to allocate patients to treatment in the trial. The order that patients entered the trial was retained for each rerandomization.

Because this was a superiority trial, the primary analysis was performed on the intent-to-treat (ITT) population (ie, all randomized patients who took at least 1 dose of study medication and for whom a TG level was measured at least once after the first intake of treatment). A general linear model for the primary endpoint was fitted as follows: [(TGDB − TGBL)/TGDB] · 100 = TGBL + Center + Gender + TG (class at baseline) + Treatment + Error. The adjusted least square means (LSM) and the estimated treatment effect were calculated with their corresponding 95% confidence intervals. Statistical robustness was confirmed through prespecified sensitivity analyses. In particular, confirmatory models similar to the primary model were fitted using the log ratio and the raw difference in TG levels between randomization and week 12. Furthermore, a nonparametric ranked analysis of covariance was performed. The same analyses were performed in the PP population (patients in the ITT population who did not present with a major protocol deviation).

Responders were analyzed using Cochran-Mantel-Haenszel methodology. The unadjusted relative risk was estimated, and the adjusted relative risk was calculated independently for each stratification factor. Descriptive analyses were performed for the evolution of TG levels at the end of the double-blind period and the end of the open-label period in relation to baseline as well as at the end of the double-blind period taking into account the effect of the randomization. The distributions of the other lipid parameters (total cholesterol and HDL-C) were explored, and variations were analyzed in the same way as the primary criteria.

The number of patients presenting with adverse events was compared between the 2 groups using a χ2 test or Fisher exact test. Descriptive analyses were used to summarize the incidence rates of events by body system and preferred term. All analyses were performed using SAS software, release 8.1 (SAS Institute, Cary, NC).

Back to Top | Article Outline


Baseline Patient Characteristics

A total of 122 patients were randomized at the end of the screening period (60 to N-3 PUFAs and 62 to placebo), and 10 patients whose TC level was greater than 10 g/L were directed to the open N-3 PUFA arm of the study. Two patients randomized to the PUFA group were excluded from the efficacy analyses: the first because he never started the treatment and the second because a lack of data after randomization.

Baseline patient characteristics are summarized in Table 1. No relevant differences were noted between groups regarding age, gender ratio, history of HIV infection, or history of hypertriglyceridemia. In both groups, nearly 90% of patients were male; more than 10 years had elapsed, on average, since HIV diagnosis; and antiretroviral therapy had begun, on average, approximately 7 years ago. In total, approximately 60% of patients were asymptomatic and 30% presented with established AIDS. The duration of hypertriglyceridemia was longer in the placebo group, although not significantly. A combination of nucleoside reverse transcriptase inhibitors and PIs was the most common antiretroviral therapy (46.6% of patients in the PUFA group and 53.2% in the placebo group). Nucleoside reverse transcriptase inhibitors plus nonnucleoside reverse transcriptase inhibitors were taken by 25.9% and 22.6% of patients, respectively, whereas 17.2% and 17.7% of patients were taking nucleoside transcriptase inhibitors plus nonnucleoside reverse transcriptase inhibitors plus PIs. Ritonavir-boosted PIs were used in all cases of PI regimens.



A total of 11 randomized patients discontinued the study before the end of the double-blind period: 5 (8.3%) of 60 in the PUFA group and 6 (9.7%) of 62 in the placebo group.

Back to Top | Article Outline

Efficacy Results

Evolution of Triglyceride Levels at the End of the Double-Blind Period

Descriptive statistics for the evolution of TG levels during treatment are displayed in Table 2. The application of the primary model to test the clinical hypothesis of at least a 20% difference between treatment groups in change from baseline TG levels showed a significant superiority of N-3 PUFAs (LSM difference of 24.6% [95% CI: −40.9 to −8.4]; P = 0.0033; Table 3). The median change in TG level was a decrease of 25.5% in the PUFA group compared with an increase of 1.0% in the placebo group. The preplanned sensitivity models used for confirmatory analysis were supportive of the conclusions reached using the primary model. Treatment effect P values from the prespecified log ratio model (P = 0.0005), the raw difference model (P = 0.0016), and the nonparametric model (P = 0.0012) were similar to that of the primary model. The P value from the primary model can thus be considered to be robust, and the corresponding LSM difference between groups can be accepted as a good estimate of the treatment effect. The PP analysis of the main endpoint gave similar results (LSM difference of 24.7% [range: −42.5 to −7.0]; P = 0.0068). The median change in TG level was a decrease of 26.8% in the PUFA group compared with an increase of 1.6% in the placebo group.





The randomization test performed to validate the dynamic randomization algorithm showed proportions of P values of 0.0556 (556/10,000) for P = 0.05 and 0.0015 (15/10,000) for P = 0.001. These results indicate that the algorithm was sufficiently random for asymptotic statistical methods to be acceptable.

TG levels had returned to normal at the end of the double-blind period for 22.4% of patients in the PUFA group versus only 6.5% of patients in the placebo group, with a relative risk of success of 3.5 with N-3 PUFAs (P = 0.0126). Furthermore, a significantly higher percentage of patients achieved a 20% decrease on N-3 PUFAs compared with placebo (58.6% vs. 33.9%; P = 0.007).

Back to Top | Article Outline

Evolution of Triglyceride Levels at the End of the Study

A total of 111 patients (55 PUFA patients and 56 placebo patients) continued on to the 8-week open period on PUFAs after the double-blind treatment period. In the PUFA group, the TG level remained almost stable at the end the open phase (4.2%). Conversely, the TG level that had remained almost unchanged in the placebo arm during the randomized period decreased after the switch to open PUFAs to a similar extent compared with that obtained in the randomized PUFA arm (−21.2%). The overall evolution of the TG level throughout the study is illustrated in Figure 2.



Back to Top | Article Outline

Evolution of Triglyceride Levels in the Open Maxepa Arm

For the 10 patients with a baseline TG level >10 g/L included in the open PUFA arm, an important decrease in TG level was achieved during the first 8 weeks of treatment: a median (Q1, Q3) percent change of −43.6% (95%CI: −66.5% to −4.6%; mean TG change of −35.6%). The efficacy was sustained in the next 8 weeks, with a global decrease over the 16-week period of −38.7% (95%CI: −62.1% to +2.1%) for 9 patients.

Back to Top | Article Outline

Evolution of Total and High-Density Lipoprotein Cholesterol

At baseline, total cholesterol was 6.2 ± 1.4 mmol/L (PUFA group) versus 6.7 ± 1.8 mmol/L (placebo group), respectively (mean ± SD, difference not significant [NS]). At week 8, total cholesterol was 6.1 ± 1.4 mmol/L (PUFA group) versus 6.9 ± 1.7 mmol/L (placebo group), respectively (difference NS). The total mean cholesterol change was −0.4% in the PUFA group and 5.7% in the placebo group. The application of a similar general linear model to that used for the primary endpoint showed that the difference in mean total cholesterol between groups (PUFA - placebo) at week 8 was −8.5% (P = 0.0117). HDL-C (see Table 2) was comparable between groups as well at baseline (0.98 ± 0.23 mmol/L [PUFA group] vs. 0.98 ± 0.25 mmol/L [placebo group]; difference NS) and at week 8 (1.06 ± 0.26 mmol/L [PUFA group] vs. 1.05 ± 0.27 mmol/L [placebo group]; difference NS). A slight but nonsignificant increase in HDL-C was noted in both groups (6.57% and 8.97%, respectively, in the PUFA and placebo groups), with no significant difference between groups.

Back to Top | Article Outline

Glucose Metabolism

The fasting glucose level was normal for all patients at the time of randomization (<6.6 mmol/L) and remained unchanged at 8 weeks, except for 1 patient in the placebo group who exhibited a slight increase in fasting glycemia (7.2 mmol/L) without other metabolic abnormalities. Measurement of insulin was not assessed in the present study.

Back to Top | Article Outline

Safety Results

The incidence of treatment-emergent adverse events during the randomized double-blind study period was not more frequent in the PUFA group: 19 (32.2%) of 59 patients versus 27 (43.5%) of 62 of patients in the placebo group (NS). The most common adverse events were minor gastrointestinal disorders without vomiting, which did not lead to stopping the treatment, except for 1 patient in the placebo group (11.9% of patients on PUFA compared with 14.5% patients in the placebo group; P = 0.790), and infections and/or infestations (11.9% of patients on N-3 PUFAs compared with 16.1% of patients in the placebo group; P = 0.604). The proportion of patients experiencing events that were thought by the investigator to be possibly or probably related to study drug was not significantly different between groups (11.9% in the PUFA group vs. 14.5% in the placebo group; P = 0.790). Two patients experienced serious adverse events on N-3 PUFAs (renal colic and urinary calculus, which were considered to be related to indinavir), and 3 patients in the placebo group experienced serious adverse events (diarrhea, an episode of acute alcohol abuse, and surgical treatment of condylomas, which were not considered to be related to the study treatment).

No relevant variation was observed in either group for vital signs during the study. Transaminases rose to at least twice the upper limit of normal (ULN) for 2 patients from each group at the end of the double-blind period. No clinically significant change was otherwise observed for laboratory parameters. The CD4+ cell count, CD8+ cell count, and viral load remained stable throughout the study.

Back to Top | Article Outline


The present study is the first randomized double-blind study assessing the effect of N-3 PUFAs on hypertriglyceridemia of HAART-treated patients.

The primary efficacy analysis testing a difference of at least 20% between treatment groups was significant for the ITT and PP populations (PUFA - placebo: −24.6% [95% CI: −40.9% to −8.4%]; P = 0.0033 for ITT, and PUFA - placebo: −24.7% [95% CI: −42.5% to −7%]; P = 0.0068 for PP). Nevertheless, the primary model (which included baseline TG levels) gave a good estimate of the difference between treatment groups (PUFA - placebo). Furthermore, this estimate was well supported by all the preplanned sensitivity analyses, which gave similar P values for the log ratio, raw differences analyses, and the nonparametric analysis of covariance for the difference between treatment groups.

The median percent change in TG level over the 8-week double-blind period was −25.5% (95% CI: −44% to +1%; range: Q1-Q3) for N-3 PUFAs and 1.0% (95% CI: −30% to +31%; range: Q1-Q3) for placebo in the ITT population. Because the distribution of the percent change in TG level is asymmetric (right-skewed) by nature, the median percent change more accurately represents the central distribution of the evolution of TG levels in each group.

The secondary criteria based on percentages of responders were also significantly different in favor of N-3 PUFAs. A further indication of the efficacy of N-3 PUFAs can be clearly seen in the second phase of the trial (the open phase, where all patients were assigned to N-3 PUFAs). In this phase, there was a marked decrease in TG level in patients who were assigned placebo at the time of randomization and a sustained TG level at week 16 in patients who had been assigned N-3 PUFAs. Despite comparable levels in total cholesterol at entry and week 8 between groups, the difference in mean total cholesterol variation (−5.8%) indicates that the use of N-3 PUFAs was associated with a decrease in total cholesterol. Conversely, although HDL-C was comparable between groups, a slight (NS) increase was observed at week 8. These results ruled out a hypothetic elevation of low-density lipoprotein cholesterol (LDL-C) on N-3 PUFAs, as also demonstrated by others.34,37 The slight increase in HDL-C was not significant and could not be considered or analyzed. From a metabolic point of view, a decrease in very low-density lipoprotein (VLDL) TG is usually associated with an increase in HDL-C. This effect may contribute to the cardiovascular benefit of lowering TG levels reported in the literature.28 The present study was not designed to test this hypothesis in the particular situation of HIV-infected patients. In light of the results of the present study, the effect of decreasing TG levels with PUFAs is clear in HIV-infected patients. The results should not be overinterpreted, however, and the cardiovascular hypothetic benefit of this result in this situation remains unknown and needs further long-term studies.

For the 10 patients with a baseline TG level >10 g/L included in the open PUFA arm, the important decrease in TG level achieved at week 8 and sustained at week 16 further demonstrated the high activity of N-3 PUFAs on hypertriglyceridemia. Even if the TG level was not normalized for all patients, the possible risks (eg, acute pancreatitis) related to extremely high levels of TG >10 g/L were probably reduced by the treatment. Overall, the study showed good clinical and biologic tolerability of N-3 PUFAs. The most common adverse events were minor gastrointestinal disorders, particularly moderate diarrhea, which could be related, in part, to the toxicity of the HIV treatment. The paraffin oil used in the placebo group could also have played a role in this kind of adverse event, however.

The CD4+ cell count, CD8+ cell count, and viral load measurements were stable throughout the study, suggesting the lack of effect of N-3 PUFAs on HIV disease itself in patients without any change in their HIV treatment during the study.

The efficacy of N-3 PUFAs was reported in hyperlipidemic HIV-negative patients receiving the same dose as in the present study (6 g/d).30 The decrease in TG levels was 28% in patients with type IIb hyperlipemia and 41% in patients with type IV hyperlipemia.

Only a few studies have investigated the efficacy of fish oil in HIV-infected patients. A short open study on moderate hypertriglyceridemia (2-5 g/L) in HIV-infected patients was reported by Manfredi et al,33 comparing 1 group of patients with diet and exercise, 1 group receiving treatment with fibrate, and 1 group receiving ω-3 fatty acids (2 g of fish oil per day). The decreases in mean triglyceridemia were only 5.6% and 15.8% at 3 and 6 months, respectively, but the initial level was only 3 g/L and the dose administrated was only 2 g/d. Another open-label randomized study was conducted in 52 patients by Wohl et al.34 The mean decrease in triglyceridemia was 25% at week 4, which represents results comparable with ours. The results of our study are emphasized by the efficacy of the treatment in the second open period in both groups. Moreover, the efficacy of the treatment in the group with severe hypertriglyceridemia >10 g/L is a positive and notable result of our study.

Thus, the prescription of N-3 PUFAs in the treatment of HAART-related hypertriglyceridemia showed a favorable safety profile, and no issues regarding the control of the viral disease or the hepatic tolerance appeared during the study.

The results of this study are comparable to the results obtained with other lipid-lowering drugs on TG levels and confirm the retrospective data obtained previously.38-40 Decreases of 17%, 25%, and 27% were obtained by different treatments with fibrates at 3, 6, and 9 months, respectively.33 Among the other drugs tested in this specific field, rosiglitazone unexpectedly caused significant increases in serum TG and cholesterol concentrations.41,42 Metformin seems to have some beneficial impact on the insulin-resistance syndrome existing in these patients.43 The results of the present study cannot emphasize a direct beneficial effect on the metabolic syndrome often associated with HIV infection.

The study performed here included one of the largest populations followed for lipid-lowering drug efficacy and safety testing. Our results suggest that N-3 PUFAs have a sustained effect on HAART-linked hypertriglyceridemia, without any specific safety concerns. The beneficial effects of PUFAs are of importance regarding the absence of drug interaction recognized with these compounds, even if only 22.4% of the patients presented with complete normalization of their hypertriglyceridemia. The efficacy of N-3 PUFAs is recognized as secondary prevention in patients with myocardial infarction.44 Because of the high benefit/risk ratio of the product, N-3 PUFAs could be proposed as a possible treatment in patients with isolated or predominant hypertriglyceridemia in the context of HAART to decrease the higher level of TGs. The place of PUFAs in the armamentarium of treatment of metabolic disorders in HIV-infected patients (eg, statins, fibrates, glitazones) needs to be further investigated with future prospective studies, however. No general recommendations can be proposed, despite the positive results of the present study. Therapies to reduce or prevent cardiovascular risk factors (eg, hypertriglyceridemia)45 in these patients are useful, however.46,47

Treatment with PUFAs is a good way to reduce some risks of hypertriglyceridemia such as sudden death in patients with known or unknown coronary artery disease,44,48 even if a recent study suggests that this may not necessarily be the case in patients with high risk of ventricular tachyarrhythmia.49 Improvement of TG levels in the context of a metabolic syndrome and reducing the risk of acute pancreatitis in case of a high level of hypertriglyceridemia (>10 g/L) may be considered as beneficial effects in the context of HIV infection. Recent studies have shown that the beneficial effect on cardiovascular disease occurs not only by reducing the rate of sudden death in secondary prevention but by reducing cardiovascular ischemic events by 30%.44 Further studies should be useful to assess the beneficial effects of PUFAs that might be expected in the HIV population, which is at high risk of cardiac ischemic disease.46,47

Back to Top | Article Outline


The authors are indebted to the following physicians for their help in recruiting and following up patients: H. Berthe, P. de Truchis, and J-C. Melchior (Hôpital R. Poincaré); G. Force (Hôpital Perpétuel Secours); F. Boccara (Hôpital St. Antoine); C. Katlama (Hôpital Pitié-Salpétrière); M. Kirstetter (Hôpital St. Antoine); W. Rozenbaum (Hôpital Tenon); A. Simon (Hôpital Pitié Salpétrière); V. Perronne (Hôpital F. Quesnay); H. Masson (Hôpital Touladjian); D. Zucman (Hôpital Foch); P. Massip (Hôpital Purpan); and J. Doll (Hôpital A. Mignot).

The organization and management of this study were monitored by Pr. J-C. Melchior, the Principal Investigator. The methodology and monitoring were monitored by Cardinal Systems, Paris. Patients were affiliated with the following associations: Action-Traitement, Actif-Santé, and TRT 5.

Back to Top | Article Outline


1. Capron L, Kim YU, Laurian A, et al. Atheroembolism in HIV-positive individuals. Lancet. 1992;340:1032-1040.
2. Feingold KR, Soued M, Serio MK, et al. Multiple cytokines stimulate hepatic lipid synthesis in vivo. Endocrinology. 1989;125:267-274.
3. Grunfeld C, Kotler DP, Hamadeh R, et al. Hypertriglyceridemia in the acquired immunodeficiency syndrome. Am J Med. 1989;86:27-31.
4. Hellerstein MC, Grunfeld C, Wu K, et al. Increased de novo lipogenesis in human immunodeficiency virus infection. J Clin Endocrinol Metab. 1993;76:559-565.
5. Mildvan D, Machado SG, Willets I, et al. Endogenous interferon and triglyceride concentrations to assess response to zidovudine in AIDS and advanced AIDS-related complex. Lancet. 1992;339:453-456.
6. Peck MD, Mantero-Atienza E, Miguez-Burbano MJ, et al. The esterified plasma Fatty Acid Profile is altered in early HIV-1 infection. Lipids. 1993;28:593-597.
7. Csaszar A. Hypertriglyceridemia, the coronary heart disease risk marker “solved.” Acta Physiol Hung. 2005;92:109-120.
8. Balasubramanyam A, Sekhar RV, Jahoor F, et al. Pathophysiology of dyslipidemia and increased cardiovascular risk in HIV lipodystrophy: a model of systemic steatosis. Curr Opin Lipidol. 2004;15:59-67.
9. Friis-Moeller N, Weber R, Reiss P, et al. Cardiovascular disease risk factors in HIV patients association with antiretroviral therapy. Results from the DAD study. AIDS. 2003;17:1179-1193.
10. Depairon M, Chessex S, Sudre PH, et al. Premature atherosclerosis in HIV-infected individuals. Focus on protease inhibitor therapy. AIDS. 2001;15(3):329-334.
11. Holmberg SD, Moorman AC, Greenberg AE, et al. Trends in rates of myocardial infarction among patients with HIV (2). N Engl J Med. 2004;350:730-732.
12. Nolan D. Metabolic complications associated with HIV protease inhibitor therapy. Drugs. 2003;63:2555-2574.
13. Calza L, Manfredi R, Chiodo F. Dyslipidaemia associated with antiretroviral therapy in HIV infected patients. J Antimicrob Chemother. 2004;53:10-14.
14. Barbaro G, Di Lorenzo G, Cirelli A, et al. An open label prospective, observational study of the incidence of coronary artery disease in patients with HIV infection receiving highly active antiretroviral therapy. Clin Ther. 2003;25:2405-2418.
15. Mary-Krause M, Cotte L, Simon A, et al. Increased risk of myocardial infarction with duration of protease inhibitor therapy in HIV infected men. AIDS. 2003;17:2479-2486.
16. Calza L, Manfredi R, Chiodo F. Hyperlipidaemia in patients with HIV 1 infection receiving highly active antiretroviral therapy: epidemiology atherogenesis, clinical course and management. Int J Antimicrob Agents. 2003;22:89-99.
17. Salomon J, de Truchis P, Melchior JC. Nutrition and HIV infection. Br J Nutr. 2002;87(Suppl 1):S111-S119.
18. Christeff N, de Truchis P, Melchior JC, et al. Longitudinal evolution of HIV-1-associated lipodystrophy is correlated to serum cortisol: DHEA ratio and IFN-α. Eur J Clin Invest. 2002;32:775-784.
19. Christeff N, Melchior J-C, De Truchis P, et al. Lipodystrophy defined by a clinical score in HIV-infected men on highly active antiretroviral therapy: correlation between dyslipidaemia and steroid hormone alteration. AIDS. 1999;13:2251-2260.
20. Ledru E, Christeff N, Patey O, et al. Alteration of tumor necrosis factor-alpha T-cell homeostasis following potent antiretroviral therapy: contribution to the development of human immunodeficiency virus-associated lipodystrophy syndrome. Blood. 2000;95:3191-3198.
21. Christeff N, Melchior J-C, De Truchis P, et al. Increased serum interferon alpha in HIV-1 associated lipodystrophy syndrome. Eur J Clin Invest. 2002;32:43-50.
22. Gougeon ML, Penicaud L, Fromenty B, et al. Adipocytes targets and actors in the pathogenesis of HIV associated lipodystrophy and metabolic alterations. Antivir Ther. 2004;9:161-177.
23. Dubé MP, Stein JH, Aberg JA, et al. Guidelines for the evaluation and management of dyslipidemia in human immunodeficiency virus (HIV)-infected adults receiving antiretroviral therapy. Clin Infect Dis. 2003;37:613-627.
24. Schambelan M, Benson CA, Carr A, et al. Management of metabolic complications associated with antiretroviral therapy for HIV-1 infection: recommendations of an International AIDS Society-USA Panel. J Acquir Immune Defic Syndr. 2002;31:257-275.
25. Chuck SK, Penzak SR. Risk benefit of HMG CoA reductase inhibitors in the treatment of HIV protease inhibitor related hyperlipidaemia. Expert Opin Drug Saf. 2002;1:5-17.
26. Badiou S, Merle De Boever C, Dupuy AM, et al. Fenofibrate improves the atherogenic lipid profile and enhances LDL resistance to oxidation in HIV positive adults. Atherosclerosis. 2004;172:273-279.
27. Rao A, D'Amico S, Balasubramanyam A, et al. Fenofibrate is effective in treating hypertriglyceridemia associated with HIV lipodystrophy. Am J Med Sci. 2004;327:315-318.
28. Underwood PM. Cardiovascular risk, the metabolic syndrome, and the hypertriglyceridemic waist. Curr Opin Lipidol. 2004;15:495-497.
29. GISSI-Prevenzione Investigators. Dietary supplementation with n-3 polyunsaturated fatty acids and vitamin E after myocardial infarction: results of the GISSI-Prevenzione trial. Lancet. 1999;354:447-454.
30. Simons LA, Hickie JB, Balasubramaniam S. On the effects of dietary n-3 fatty acids (Maxepa) on plasma lipids and lipoproteins in patients with hyperlipidemia. Atherosclerosis. 1985;54:75-88.
31. Burr ML, Gilbert JF, Holliday RM, et al. Effects of changes in fat, fish, and fibre intakes on death and myocardial reinfarction, diet and reinfarction trial dart. Lancet. 1989;30:757-761.
32. Singh RB, Niaz MA, Sharma JP, et al. Randomized, double-blind, placebo controlled trial of fish oil and mustard oil in patients with suspected acute myocardial infarction: the Indian experiment on infarct survival-4. Cardiovasc Drugs Ther. 1997;11:485-491.
33. Manfredi R, Calza L, Chiodo F. Polyunsaturated ethyl esters of n-3 fatty acids in HIV-infected patients with moderate hypertriglyceridemia: comparison with dietary and lifestyle changes, and fibrate therapy. J Acquir Immune Defic Syndr. 2004;36:878-880.
34. Wohl DA, Tien H-C, Buzby M, et al. Randomized study of the safety and efficacy of fish oil (omega-3 fatty acid) supplementation with dietary and exercise counseling for the treatment of antiretroviral therapy-associated hypertriglyceridemia. Clin Infect Dis. 2005;41:1498-1504.
35. Krauss RM, Eckell RH, Howard DB, et al. American Heart Association dietary guidelines, recommendations 2000: a statement for healthcare professionals from the Nutrition Committee of the AHA. Circulation. 2000;102:2284-2299.
36. Pocock SJ, Simon R. Sequential treatment assignment with balancing for prognostic factors in the controlled clinical trial. Biometrics. 1975;31:103-115.
37. Sirtori CR, Crepaldi G, Manzato E, et al. One year treatment with ethyl esters on n-3 fatty acids in patients with hypertriglyceridemia and glucose intolerance: reduced triglyceridemia, total cholesterol and increased HDL, without glycemic alterations. Atherosclerosis. 1998;137:419-427.
38. Zimetbaum P, Frishman WH, Kahn S. Effects of gemfibrozil and other fibric acid derivatives on blood lipids and lipoproteins. J Clin Pharmacol. 1991;31:25-37.
39. Superko HR, McGovern ME, Raul E, et al. Differential effect of two nicotinic acid preparations on low-density lipoprotein subclass distribution in patients classified as low-density lipoprotein pattern A, B or I. Am J Cardiol. 2004;94:588-594.
40. Aberg JA, Zackin RA, Brobst SW, et al. A randomized trial of the efficacy and safety of fenofibrate versus pravastatine in HIV-infected subjects with lipid abnormalities: AIDS Clinical Trials group Study 5087. AIDS Res Hum Retroviruses. 2005;21:757-767.
41. Hadigan C, Yawetz S, Thomas A, et al. Metabolic effects of rosiglitazone in HIV lipodystrophy: a randomized, controlled trial. Ann Intern Med. 2004;140:786-794.
42. Sutinen J, Haekkinen AM, Westerbacka J, et al. Rosiglitazone in the treatment of HAART associated lipodystrophy a randomized double blind placebo controlled study. Antivir Ther. 2003;8:199-207.
43. Driscoll SD, Meininger GE, Lareau MT, et al. Effects of exercise training and metformin on body composition and cardiovascular indices in HIV infected patients. AIDS. 2004;18:465-473.
44. Harper CR, Jacobson TA. Usefulness of omega 3 fatty acids and the prevention of coronary heart disease. Am J Cardiol. 2005;96:1521-1529.
45. Hokanson JE, Austin MA. Plasma triglyceride level is a risk factor for cardiovascular disease independent of high lipoprotein cholesterol level: a meta-analysis of population based prospective studies. J Cardiovasc Risk. 1996;3:213-219.
46. D:A:D: Study Group. Combination antiretroviral therapy and the risk of myocardial infarction. N Engl J Med. 2003;349:1993-2003.
47. D:A:D: Study Group. Combination antiretroviral therapy and the risk of myocardial infarction. N Engl J Med. 2004;350:955.
48. Marchioli R, Levantesi G, Macchia A, et al. Antiarrhythmic mechanisms of n-3 PUFA and the results of the GISSI-Prevenzione trial. J Membr Biol. 2005;206:117-128.
49. Brouver IA, Zock PL, Camm AJ, et al. Effect of fish oil on ventricular tachyarrhythmia and death in patients with implantable cardioverter defibrillators: the Study on Omega-3 Fatty Acids and Ventricular Arrhythmia (SOFA) Randomized Trial. JAMA. 2006;295:2613-2619.

antiretroviral therapy; fish oil; HIV; hypertriglyceridemia; lipodystrophy; n-3 polyunsaturated fatty acids

© 2007 Lippincott Williams & Wilkins, Inc.