Increase of lipid plasma concentrations under protease inhibitor-containing regimens are not related to modifications of the tumour necrosis factor system

Bonnet, Fabricea; Savès, Marianneb; Droz, Cécileb; Peuchant, Evelynec; Chêne, Genevièveb; Beylot, Jacquesa; Morlat, Philippea

Author Information

aService de Médecine Interne et Maladies Infectieuses, Hôpital Saint-André, 1 rue Jean Burguet, 33075 Bordeaux Cedex, France; bINSERM U 330, Epidémiologie, Santé Publique et Développement, Université Victor Segalen Bordeaux 2, 146 rue Léo-Saignat, 33076 Bordeaux Cedex, France; and cLaboratoire de Biochimie, Hôpital Saint-André, 1 rue Jean Burguet, 33075 Bordeaux Cedex, France.

Received: 6 October 2000; accepted: 30 January 2001.

Article Outline

Hypertriglyceridaemia and hypocholesterolaemia have been reported in several infectious diseases [1]. In HIV-infected patients, they have been shown to be related to immunological activation status, and have been considered to be potential prognostic markers of HIV infection [2]. In the era before highly active antiretroviral therapy, it was shown that changes in the concentration of plasma lipids were related to modifications of cytokine production, particularly to activation of the tumour necrosis factor (TNF) system [3].

We and others have shown that the use of protease-inhibitor (PI)-containing regimens is associated with an increase in plasma lipid levels and atherogenic risk, although the mechanism of this dyslipidaemia remains to be identified [4,5]. One of the hypotheses is that changes in plasma lipids were related to modifications of immune status and the TNF system. Therefore, we studied the evolutions of several usual lipid markers in PI-treated patients and TNF system activation, as measured by the plasma levels of its soluble receptor type II (p75) sTNFRII during one year of follow-up.

Fifty-one PI-naive HIV-1-infected adults (37 men and 14 women, median age 38 years: 25–75th percentile: 35–46) were consecutively included in a one year prospective cohort from the day of initiation of a PI-containing regimen (M0). The baseline median CD4 T lymphocyte count was 138/μl (79–202), and median plasma HIV RNA (branched DNA Chiron 2.0) 4 log10 copies/ml (3.4–4.9). Treatment regimens combined one (n = 5) or two (n = 46) nucleoside reverse transcriptase inhibitors and one PI: indinavir (n = 30), saquinavir (n = 15), ritonavir (n = 4) or nelfinavir (n = 2). Plasma measurements were always performed in the morning, usually in non-fasting conditions. All patients remained on PI therapy during the follow-up and did not receive any specific drug or dietary interventions for lowering lipid concentrations.

At month 12 (M12), the median level of CD4 cell count was 224 cells/μl (P < 0.0001 compared with M0), of HIV RNA 3.1 log10 copies/ml (P < 0.0001 compared with M0). Concomitantly, a significant increase in plasma lipid levels was observed as M0/M12 concentrations were: cholesterolaemia, 4.1/4.9 mmol/l (P = 0.0001); triglyceridaemia, 1.2/1.6 mmol/l (P = 0.0001); LDL cholesterolaemia 2.7/3.3 mmol/l (P = 0.0001). A significant positive correlation was observed between M0 and M12 between the evolution of cholesterolaemia and LDL cholesterolaemia (r = 0.79, P < 0.05) (data not shown). Conversely, no significant variation was observed for HDL cholesterolaemia: 0.8/0.8 (P = 0.44) and sTNFRII: 8.9/8.1 ng/ml (P = 0.16). We did not identify a significant correlation between the evolution of sTNFRII and cholesterolaemia (P = 0.19) or triglyceridaemia (P = 0.44) (data not shown) between M0 and M12.

Moreover, there was no significant sTNFRII plasma level difference between subgroups of patients with high or low levels of cholesterolaemia or triglyceridaemia at M6 or M12 (Table 1).

Our data show a significant increase in total and LDL cholesterolaemia and of triglyceridaemia under PI therapy. No correlation between the plasma level of sTNFRII and lipids was shown as it was before the era of PI therapy. These PI-associated lipid disturbances thus seem to be independent of the TNF system activation status. Our results suggest that they are not related to the change in cytokine production, as during the natural evolution of HIV infection. They may rather be caused by a pharmacological impact of PI-containing regimens. If, in the past, immune activation in HIV infection was associated with dyslipidaemia, these metabolic disturbances are not now corrected by the decrease in immune activity but are instead enhanced by PI. These results emphasize the need for a survey of plasma lipid profiles in patients treated with PI.

Fabrice Bonneta

Marianne Savèsb

Cécile Drozb

Evelyne Peuchantc

Geneviève Chêneb

Jacques Beylota

Philippe Morlata

Back to Top | Article Outline


1. Gallin JI, Kaye D, O'Leary WM. Serum lipids in infection. N Engl J Med 1969, 281: 1081 –1076.
2. Constans J, Pellegrin JL, Peuchant E. et al. Plasma lipids in HIV-infected patients: a prospective study in 95 patients. Eur J Clin Invest 1994, 24: 416 –420.
3. Zangerle R, Sarcletti M, Gallati H, Reibnegger G, Wachter H, Fuchs D. Decreased plasma concentrations of HDL cholesterol in HIV-infected individuals are associated with immune activation. J Acquir Immune Defic Syndr 1994, 7: 1149 –1156.
4. Bonnet F, Savès M, Droz C, Peuchant E, Chêne G, Beylot J, Morlat P. Increase of atherogenic plasma profile in HIV-infected patientstreated with protease inhibitor-containing regimens. J Acquir Immune Defic Syndr 2000, 67: 199 –200.
5. Périard D, Telenti A, Sudre P. et al. Atherogenic dyslipidemia in HIV-infected individuals treated with protease inhibitors. Circulation 1999, 100: 700 –705.
© 2001 Lippincott Williams & Wilkins, Inc.