Lipids, biomarkers, and subclinical atherosclerosis in treatment-naive HIV patients starting or not starting antiretroviral therapy: Comparison with a healthy control group in a 2-year prospective study

Abstract

Objective: To assess the effect of HIV infection and combined antiretroviral therapy (c-ART) on various proatherogenic biomarkers and lipids and to investigate their relationship with subclinical atherosclerosis in a cohort of treatment-naive HIV-infected patients.

Methods: We performed a prospective, comparative, multicenter study of 2 groups of treatment-naive HIV-infected patients (group A, CD4>500 cells/μL, not starting c-ART; and group B, CD4<500 cells/μL, starting c-ART at baseline) and a healthy control group. Laboratory analyses and carotid ultrasound were performed at baseline and at months 12 and 24. The parameters measured were low-density lipoprotein (LDL) particle phenotype, lipoprotein-associated phospholipase A2 (Lp-PLA2), interleukin-6 (IL-6), high-sensitivity C-reactive protein (hs-CRP), sCD14, sCD163, monocyte chemoattractant protein-1(MCP-1), and asymmetric dimethylarginine (ADMA). A linear mixed model based on patient clusters was used to assess differences in biomarkers between the study groups and over time.

Results: The study population comprised 62 HIV-infected patients (group A, n = 31; group B, n = 31) and 22 controls. Age was 37 (30-43) years, and 81% were men. At baseline, the HIV-infected patients had a worse LDL particle phenotype and higher plasma concentration of sCD14, sCD163, hs-CRP, and LDL-Lp-PLA2 than the controls. At month 12, there was an increase in total cholesterol (p = 0.002), HDL-c (p = 0.003), and Apo A-I (p = 0.049) and a decrease in sCD14 (p = <0.001) and sCD163 (p<0.001), although only in group B. LDL particle size increased in group B at month 24 (p = 0.038). No changes were observed in group A or in the healthy controls. Common carotid intima-media thickness increased in HIV-infected patients at month 24 (Group A p = 0.053; group B p = 0.048). Plasma levels of sCD14, sCD163, and hs-CRP correlated with lipid values.

Conclusions: In treatment-naive HIV-infected patients, initiation of c-ART was associated with an improvement in LDL particle phenotype and inflammatory/immune biomarkers, reaching values similar to those of the controls. HIV infection was associated with progression of carotid intima-media thickness.

Fig 1. Mean adjusted for group, time, and interaction between group and time and 95% CI in lipid parameters from baseline to months 12 and 24.

*Mean adjusted for group, time, and interaction between group B and time. **Mean adjusted for group and time. In TC, the interaction between time and group was statistically significant for group B at 12 months (p = 0.002). In HDL-c and APO A-1, the interaction was statistically significant for group B at 12 months (p = 0.003 and 0.049, respectively) and 24 months (p = 0.001 and 0.008, respectively). Moreover, for all 3 lipid parameters, differences at baseline were statistically significant for group A compared with the control group (p = 0.03, <0.001, and <0.001, respectively) and group B compared with the control group (p = 0.004, <0.001, and <0.001, respectively). In the case of LDL-c and Apo A-I/Apo B, differences at baseline were statistically significant in group B compared with the control group (p = 0.019 and 0.021); baseline differences in group A compared with the control group were statistically significant only for Apo A-I/Apo B (p = 0.017). Abbreviations: TC, total cholesterol; HDL-c, high-density lipoprotein cholesterol.

Fig 2. Mean adjusted for group, time, and interaction between group and time and 95% CI in LDL particle phenotype parameters from baseline to months 12 and 24.

*Mean adjusted for group, time, and interaction between group B and Time. In LDL size and cholesterol content in lb-LDL particles, the interaction between time and group was statistically significant for group B at 24 months (p = 0.038 and 0.009, respectively). Moreover, in both variables, the differences at baseline were statistically significant for group B compared with the control group (p = 0.011 and 0.001, respectively); baseline differences in group A compared with the control group were statistically significant only for LDL size (p = 0.049). Abbreviations: LDL, low-density lipoprotein cholesterol; lb-LDL, large buoyant LDL particles.

Fig 3. Mean adjusted for group, time, and interaction between group and time and 95% CI in plasma biomarkers from baseline to months 12 and 24.

*Mean adjusted for group, time, and interaction between Group B and time. **Mean adjusted for group and time. In sCD14 and LOG(sCD163), the interaction between group and time was statistically significant for group B at 12 months (p<0.001 and <0.001, respectively) and 24 months (p<0.001 and <0.001, respectively). Moreover, differences at baseline were statistically significant in LOG(sCD163) and sCD14 for group A compared with the control group (p = 0.001 and p = 0.038 respectively) and group B compared with the control group (p<0.001 and p<0.001 respectively). In LDL-Lp-PLA2, the interaction between time and group was statistically significant for group B at 24 months (p<0.001); differences at baseline were statistically significant for groups A and B compared with the control group (p = 0.048 and p = 0.01, respectively). In LOG(hs-CRP), the differences at baseline were statistically significant for groups A and B compared with the control group (p = 0.007 and p = 0.002, respectively). Abbreviations: hs-CRP, high-sensitivity C-reactive protein; LDL-Lp-PLA2, lipoprotein-associated phospholipase A2.

Fig 4. Mean adjusted for group, time, and interaction between group and time and 95% CI in common c-IMT from baseline to months 12 and 24.

The interaction between group and time was statistically significant at 24 months for group B (p = 0.048) and almost significant for group A (p = 0.053). The interaction was statistically significant for group B compared with the control group (p = 0.015). Abbreviations: c-IMT, carotid intima-media thickness.