Higher levels of soluble adhesion molecules have been found before the introduction of highly active antiretroviral therapy (HAART) in HIV-positive compared with HIV-negative patients. To evaluate the impact of HAART on endothelial function we measured the plasmatic concentrations of soluble adhesion molecules in 120 HIV-positive patients. Concentrations of P-selectin and fibrinolytic factors were significantly higher in HAART patients, whereas soluble intercellular adhesion molecule-1 did not differ significantly. HAART may induce an activation of endothelial function both directly and indirectly through lipid alterations.
Elevated levels of soluble cellular adhesion molecules represent early markers of the development of atherosclerosis . In particular, high concentrations of soluble intercellular adhesion molecule (sICAM-1) and P-selectin have been associated with a risk of myocardial infarction or the extent of atherosclerosis [2–4].
Besides soluble adhesion molecules, fibrinolytic factors such as tissue plasminogen activator (t-PA) and plasminogen activator inhibitor (PAI-1) have also been considered to be markers of endothelial dysfunction [5,6].
Endothelial function has been investigated in HIV-positive patients since the beginning of the epidemic in the early 1980s [7,8]. In the past decade, several reports have shown increased levels of soluble adhesion molecules in patients with different stages of HIV infection compared with the HIV-negative general population [9–11]. The possible correlation between HIV-related immunological alterations and the severity of endothelial dysfunction has also been explored [12,13]. The recent introduction of highly active antiretroviral therapy (HAART) as standard therapy for HIV infection has renewed interest in the study of endothelial function in HIV-positive patients [14–16] in view of the HAART-related metabolic abnormalities  and some premature coronary artery disease reported in HIV-positive patients under HAART .
From 1 January to 31 December 2000, all consecutive HIV-positive patients admitted to the outpatient care unit of the Department of Infectious Diseases of the Catholic University Hospital in Rome were screened for participation in the study. In particular, a regular use of HAART was the first criterion for selection. HAART was defined as a combination of two nucleoside reverse transcriptase inhibitors plus either one protease inhibitor (PI) or one non-nucleoside reverse transcriptase inhibitor (NNRTI). ‘Regular HAART treated’ was considered to be a patient who had taken antiretroviral therapy for at least 6 months continuously. Individuals with familial dyslipidaemic disorders or under treatment with hypolipidaemic drugs, active intravenous drug abusers, and patients with acute concomitant infective or inflammatory disease were excluded.
Thirty patients were selected using a simple randomization table out of 585 patients treated with a PI-including regimen initially screened in the study. In the same way, from 315 patients treated with a NNRTI-including regimen, 30 patients were selected. A total number of 60 patients was thus obtained (HAART group). This group was compared with 60 HIV-positive individuals admitted to the same outpatient unit, in the same period, who were never treated with antiretroviral therapies either because therapy was not indicated or because they refused this therapy, or their HIV-positive status was previously unknown (naive group). Written informed consent was obtained from each individual before inclusion in the study.
The two groups were comparable for several variables, although patients in the HAART group had a longer time from HIV diagnosis (39 versus 70 months; P < 0.001) and presented with more advanced HIV diseases (18 versus 55% in the C stage of CDC classification; P < 0.0001). In each group, chronic hepatitis (37/60, 61% versus 28/60, 47%; P = ns) and cardiovascular diseases (10/60, 17% versus 16/60, 27%; P = ns) were the more frequent co-morbidities. With regard to HAART, it was given for an average of 22.7 ± 10.2 months.
Significantly higher values of P-selectin, t-PA and PAI-1 were found in the HAART group compared with the naive group (Table 1). The levels of sICAM-1 were not statistically different. No difference in plasmatic endothelial markers was observed between PI and NNRTI-treated subjects (P = ns).
No difference was also found in the levels of any plasmatic marker of endothelial dysfunction when the HAART and naive groups were stratified in three tertiles of HIV-RNA viral load. In addition, the correlation coefficient between endothelial markers and the HIV-RNA viral load did not demonstrate any significant value.
In a multivariate linear regression model, after adjustment for age and sex, both t-PA and P-selectin showed an independent association with the use of HAART, whereas PAI-1 presented a P value near to significance (P = 0.06). After adjustment for age, sex, cholesterol and triglycerides, t-PA was still significantly associated with the use of HAART (P < 0.01); a P value near to significance was obtained with P-selectin (P = 0.06). Further adjustment for the time from the first HIV-positive test, C stage of HIV infection and HIV-RNA viral load did not modify the results. Any of the multivariate linear regression analysis models presented a statistically significant difference between the two groups for sICAM-1.
Statistically significant higher values of cholesterol, LDL-cholesterol, triglycerides and HDL-cholesterol were found in the HAART group compared with the naive group at univariate analysis. In a multivariate model a statistically significant association was observed between lipid levels and endothelial markers, namely cholesterol with t-PA, P-selectin and sICAM-1 (P = 0.02, P = 0.03, and P < 0.01, respectively) and triglycerides with PAI-1 (P = 0.02).
Our study comparing 60 HAART-treated subjects, exposed for an average of 23 months of treatment, with 60 HIV-positive subjects naive to antiretroviral therapies, clearly showed that the levels of P-selectin, PAI-1 and t-PA were increased by HAART. According to a previous report , the levels of sICAM-1 seemed to have a different characteristic, not being enhanced in our HAART-treated versus naive patients.
We suggest that whereas in naive patients the HIV-RNA viral load plays a principal role in the enhancement of endothelial marker levels, in HAART-treated individuals antiretroviral therapy may itself induce endothelial dysfunction despite a reduced HIV-RNA viral load.
It is possible that sICAM-1 levels in HIV-positive patients reached a maximum plasmatic level that could not be further enhanced by HAART. This peculiar sICAM-1 characteristic in comparison with P-selectin, t-PA and PAI-1 could be explained by a greater sensitivity of this molecule to HIV-related factors.
The loss of association between P-selectin and PAI-1 with HAART, after adjustment for cholesterol and triglycerides, allow us to hypothesize that lipid alterations secondary to HAART could contribute, with an indirect effect, to the enhancement of both PAI-1 and P-selectin. Lipid alterations and fat redistribution in HIV-positive patients under HAART therapy have been widely described . In addition, Stein et al.  recently demonstrated that atherogenic lipoprotein associated with PI contribute to impaired endothelial function.
We suggest that HAART itself might have a direct effect on endothelial dysfunction. Additional experimental data are required to confirm this possibility. Our findings, based on the independence of t-PA from all tested variables at multivariate analysis, support the possible direct effect of HAART at least on the increase of t-PA.
In summary, the results of this study indicate that HIV-positive patients under HAART present significantly higher levels of soluble markers of endothelial dysfunction, such as the adhesion molecules P-selectin, t-PA and PAI-1 compared with naive patients.
We propose that HAART acts on endothelial function both as a direct stimulating factor of, for instance, t-PA expression and indirectly through lipid alterations, as indicated by our data for P-selectin and PAI-1. Both mechanisms may enhance the HIV effect on endothelial function and lipid profile (Fig. 1).
1.Ridker PM, Hennekens CH, Roitman-Jhonson B, Stampfer MJ, Allen J. Plasma concentration of soluble intercellular adhesion molecule 1 and risk of future myocardial infarction in apparently healthy men.Lancet
2.Davies MJ, Gordon JL, Gearing AJ, Pigott R, Woolf N, Katz D, Kyriakopoulos A. The expression of the adhesion molecules ICAM-1, VCAM-1, PECAM and E-selectin in human atherosclerosis.J Pathol
3.Blankenberg S, Rupprecht H, Bickel C, Peetz D, Hafner G, Tiret L, Meyer J. Circulating cell adhesion molecules and death in patients with coronary artery disease.Circulation
4.Ridker PM, Buring JE, Rifai N. Soluble P-selectin and the risk of future cardiovascular events.Circulation
5.Ridker PM, Vaughan DE, Stampfer MJ, Manson JE, Hennekens CH. Endogenous tissue-type plasminogen activator and risk of myocardial infarction.Lancet
6.Hamsten A, de Faire U, Walldius G, Dahlen G, Szamosi A, Landou C, et al
. Plasminogen activator inhibitor in plasma: risk factor for recurrent myocardial infarction.Lancet
7.Teitel JM, Shore A, Read SE, Schiarone A. Immune function of vascular endothelial cells is impaired by HIV.J Infect Dis
8.Lafeuillade A, Alessi MC, Poizot-Martin I, Boyer-Neumann C, Zandotti C, Quilichini R, et al
. Endothelial cell dysfunction in HIV infection.J Acquired Immune Defic Syndr
9.Puppo F, Brenci S, Scudeletti M, Lanza L, Bosco O, Indiveri F. Elevated serum levels of circulating intercellular adhesion molecule-1 in HIV infection.AIDS
10.Seigneur M, Constans J, Blann A, Renard M, Pellegrin JL, Amiral J, et al
. Soluble adhesion molecules and endothelial cell damage in HIV infected patients.Thromb Haemost
11.Aukrust P, Bjornsen S, Lunden B, Otterdal K, Ng EC, Ameln W, et al
. Persistently elevated levels of von Willebrand factor antigen in HIV infection.Thromb Haemost
12.Sipsas N, Sfikakis PP, Sfikakis P, Choremi H, Kordossis T. Serum concentrations of soluble intercellular adhesion molecule-1 and progress towards disease in patients infected with HIV.J Infect
13.Galea P, Vermont-Desroches C, Le Contel C, Wijdenes J, Chermann JC. Circulating cell adhesion molecules in HIV-1 infected patients as indicator markers for AIDS progression.Res Immunol
14.Mastroianni CM, Lichtner M, Mengoni F, D'Agostino C, d'Ettorre G, Forcina G, et al
. Changes in circulating levels of soluble cell adhesion molecules following highly active antiretroviral treatment of HIV-1-infected patients.Clin Immunol
15.Stein JH, Klein MA, Bellehumeur JL, McBride PE, Wiebe DA, Otuos JD, Sosman JM. Use of human immunodeficiency virus-1 protease inhibitors is associated with atherogenic lipoprotein changes and endothelial dysfunction.Circulation
16.Hadigan C, Meigs JB, Rabe J, D'Agostino RB, Wilson PW, Lipinska I, et al
. Increased PAI-1 and tPA antigen levels are reduced with metformin therapy in HIV-infected patients with fat redistribution and insulin resistance.J Clin Endocrinol Metab
17.Carr A, Samaras K, Burton S, Law M, Freund J, Chisholm DJ, Cooper DA. A syndrome of peripheral lipodystrophy, hyperlipidemia, an insuline resistance in patients receiving HIV protease inhibitors.AIDS
18.Depairon M, Chessex S, Sudre P, Rodondi N, Doser N, Chave J-P, et al
. Premature atherosclerosis in HIV-infected individuals – focus on protease inhibitor therapy.AIDS