We assess long-term changes in lipid levels in human immunodeficiency disease- (HIV-) infected patients undergoing highly active antiretroviral treatment (HAART) and their association with diabetes mellitus (DM) and thyroid dysfunction. We observed changes in the levels of total cholesterol (TC) and total triglyceride (TG) of 63 HIV-infected patients in the 6 years from starting HAART and analyzed correlations between relevant parameters. TC levels of patients with normal baseline TC levels as well as those diagnosed with DM or impaired fasting glucose (IFG) increased significantly (
Acquired immune deficiency syndrome (AIDS) has been considered to be a chronic disease necessitating long-term management rather than an acute fatal disease that requires highly active antiretroviral therapy (HAART). Patients infected with the human immunodeficiency virus (HIV) can survive for long periods, so non-AIDS diseases account primarily for the disease burden of this patient population.
Dyslipidemia is highly prevalent among HIV-infected patients and may contribute to an increased risk of atherogenesis as well as cardiovascular disease (CVD), which is one of the main causes of death in this population [
The characteristics of dyslipidemia associated with HAART include decreased levels of high-density lipoprotein-cholesterol (HDL-C) and increased levels of low-density lipoprotein-cholesterol (LDL-C) and total cholesterol (TC) [
Endocrine dysfunction (e.g., thyroid dysfunction) is common in HIV-infected patients [
Previously, we undertook cross-sectional studies that did not detail dynamic changes over several years. Consequently, we decided to carry out a longitudinal retrospective study to further observe variations in lipid indices. We hypothesized that the lipid levels of HIV-infected patients undergoing HAART would increase with time and that lipid levels would correlate with glucose metabolism and the levels of thyroid hormones.
All procedures undertaken in studies involving human participants were in accordance with the ethical standards set by Ruhr University Bochum (Bochum, Germany) and/or national research committees and with the
Data for 188 HIV-infected patients undergoing HAART at Saint Josef Hospital, Ruhr University Bochum, from 2000 to 2017, were collected. The final study population was 63 patients.
Patient age, sex, route of transmission, HAART, and the date of diagnosis and the date HAART started were recorded. Levels of TC, triglyceride (TG), glucose, HDL-C, LDL-C, very-low-density lipoprotein-cholesterol (VLDL-C), LDL-C:HDL-C ratio, cluster of differentiation (CD)4+ cell count, and viral load (VL) in the year HAART was started, as well as 2, 4, and 6 years after HAART was started, were collected.
Inclusion criteria were patients: (i) diagnosed with HIV infection and undergoing HAART; (ii) aged >18 years; (iii) continuing HAART.
Exclusion criteria were patients (i) with acute HIV infection; (ii) with severe, life-threatening complications; (iii) who were pregnant; (iv) with autoimmune diseases; (v)with data of TC or TG which is incomplete.
The normal ranges of the indices analyzed were TC, ≤200 mg/dL; TG, ≤150 mg/dL; glucose, 74–106 mg/dL; HDL-C, 33–84 mg/dL; LDL-C, 69–149 mg/dL; LDL-C:HDL-C ratio, 2.5–3.5; thyroid-stimulating hormone (TSH), 0.35–4.94
“Euglycemia” was defined as a glucose level within normal range. “Impaired fasting glucose” (IFG) was denoted as a glucose level of 101–125 mg/dL recorded at least twice. “Diabetes mellitus” (DM) represented a glucose level >125 mg/dL [
VL was determined using a standard reverse transcription-polymerase chain reaction system (COBAS® Amplicor HIV-1 Monitor Test, v1.5; Roche Diagnostic Systems, Branchburg, NJ, USA) with a lower limit of detection of 50 HIV-1 RNA copies/mL. The number of CD4+ cells and activated CD8+ HLA-DR+ T cells was measured by flow cytometry (FACACanto™ II; Becton Dickinson, Franklin Lakes, NJ, USA). Serum levels of TC, TG, HDL-C, LDL-C, and VLDL-C were determined by an automatic biochemical analyzer (AU5800; Beckman Coulter, Brea, CA, USA).
Data analyses were undertaken using SPSS v20 (IBM, Armonk, NY, USA). Numeric variables were expressed as the mean ± standard deviation and qualitative variables as the number of cases (percentages). Repeated-measure tests were used to evaluate trends in lipid levels, after which Student’s
Table
Clinical information for 63 patients before HAART commencement.
Variables | Mean (±SD) or number (%) |
| |
Age (years) | 40.47 ± 9.40 |
Sex | |
Male | 52 (82.5%) |
Female | 11 (17.5%) |
HIV transmission category | |
Male homosexual | 42 (66.7%) |
Heterosexual | 16 (25.4%) |
Unknown | 5 (7.9%) |
HAART regimen | |
2 NRTI + PI | 14 (22.2%) |
2 NRTI + NNRTI | 35 (55.6%) |
2 NRTI + II | 3 (4.8%) |
Other | 11 (17.5%) |
Duration of diagnosis (years) | 2.43 ± 3.79 |
Baseline TC level | |
Normal | 50 (79.4%) |
Increased | 13 (20.6%) |
Baseline TG level | |
Normal | 41 (65.1%) |
Increased | 22 (34.9%) |
Thyroid function | |
Normal | 45 (71.4%) |
Abnormal | 18 (28.6%) |
Sub-hypothyroidism | 5 (7.9%) |
Overt hypothyroidism | 11 (17.5%) |
Overt hyperthyroidism | 2 (3.2%) |
Glucose metabolism | |
Euglycemia | 15 (23.8%) |
IFG | 34 (54.0%) |
DM | 14 (22.2%) |
HBV or HCV co-infection | |
neither | 48 (76.2%) |
HBV | 13 (20.6%) |
HCV | 2 (3.2%) |
HIV: human inmunodeficiency virus; HAART: highly active antiretroviral therapy; NRTI: nucleoside reverse transcriptase inhibitor; PI: protease inhibitor; NNRTI: non-nucleoside reverse transcriptase inhibitor; II: integrase inhibitor; TC: total cholesterol; TG: total triglyceride; IFG: impaired fasting glucose; DM: diabetes mellitus; HBV: hepatitis-B virus; HCV: hepatitis-C virus.
Patients were divided into two groups according to their baseline TC level. Fifty patients had a baseline TC level within normal range (normal group) and 13 had a baseline TC level higher than the upper limit of normal (hypercholesterolemia group).
There were significant changes in the TC levels of the two groups with increasing time (P < 0.05) and the TC levels of the hypercholesterolemia group were significantly higher than those of the normal group (P < 0.05) over the whole 6 years (ESM Table
Similarly, we grouped patients according to their glucose metabolism. The TC levels of patients with different glucose metabolism changed significantly over time (P < 0.05) (ESM Table
TC levels did not show a significant difference in groups with different thyroid function (P > 0.05).
Patients were divided into two groups according to their baseline TG level. Forty-one patients had a normal baseline level (normal group) and 22 had a TG level higher than the upper limit of normal (hypertriglyceridemia group).
The TG levels of the hyperglyceridemia group were significantly higher than those of the normal group (P < 0.05). The TG levels (in mg/dL) of the hyperglyceridemia group were higher than that of the normal group for each year since HAART commencement (220.27 ± 86.68
There were no significant differences among patients with different glucose metabolism and thyroid function (P > 0.05).
Changes in CD4+ cell counts and activated CD8+ HLA-DR T-cell counts were analyzed to assess therapeutic effects and patient immunity (ESM Table
CD4+ cell counts (per
Correlations between indices before patients started HAART were done to analyze relationship of HIV infection and TC levels, as HAART may change the pattern of patients’ lipid metabolism and VL levels. Levels of TC, HDL-C, and LDL-C correlated negatively with VL (R1 = −0.511, R2 = −0.516, R3 = −0.396, and P < 0.05) (Table
Correlation between relevant parameters before HAART.
r | P | |
| ||
TC & TG | 0.339 | 0.015 |
TC & HDL-C | 0.456 | 0.019 |
TC & LDL-C | 0.856 | 0.000 |
TC & lgVL | -0.511 | 0.000 |
TC & CD4 | 0.499 | 0.000 |
TG & VLDL-C | 0.693 | 0.000 |
HDL-C & lgVL | -0.516 | 0.007 |
VLDL-C & CD4 | 0.621 | 0.001 |
LDL-C & TSH | 0.555 | 0.021 |
LDL-C & lgVL | -0.396 | 0.045 |
FT3 & CD4 | 0.596 | 0.002 |
LgVL & CD4 | -0.512 | 0.000 |
HAART: highly active antiretroviral treatment; TC: total cholesterol; TG: total triglyceride; HDL-C: high-density lipoprotein-cholesterol; LDL-C: low-density lipoprotein-cholesterol; VL: viral load; VLDL-C: very-low-density lipoprotein-cholesterol; TSH: thyroid-stimulating hormone; FT3: free triiodothyronine.
The result of linear mixed-effect model was made to explore factors influence TC levels in the six years since patients received HAART (Table
Model-based estimated factors correlated with TC levels.
estimate | Standard error | P value | |
| |||
Sex ( male vs. female) | 4.44 | 14.17 | 0.755 |
IFG vs. euglycemia | 27.84 | 11.09 | 0.014 |
DM vs. euglycemia | 45.41 | 11.80 | 0.000 |
age | 0.45 | 0.49 | 0.355 |
nevirapine | 24.79 | 11.07 | 0.028 |
lgVL | 39.10 | 22.12 | 0.077 |
CD4 | 0.004 | 0.001 | 0.000 |
TC: total cholesterol; IFG: impaired fasting glucose; DM: diabetes mellitus; VL: viral load.
In general, in the present study, the VL of most patients was below the limit of detection and CD4+ cell counts increased continually after treatment. These observations suggest that a favorable clinical curative effect had been elicited.
Changes in lipid metabolism after HAART were observed in our study. However, the data for the levels of HDL-C, LDL-C, VLDL-C, and LDL-C:HDL-C ratio were incomplete. Hence, changes in the levels of TC and TG over time were analyzed. We found that several patients with baseline levels of TC and TG within normal ranges could suffer hypercholesterolemia or hypertriglyceridemia over 6 years of HAART. The levels of TC and TG of these patients increased remarkably overall. We assumed that physicians monitored the levels of TC and TG upon acceptance of HAART irrespective of whether patients had hypercholesterolemia or hypertriglyceridemia at that time.
The TC levels of patients with hyperglycemia (IFG and DM) also increased significantly in the first 4 years from HAART commencement. The interactions between disturbances in glucose metabolism and lipid metabolism are complicated. One mainstream hypothesis is that insulin resistance can lead to dyslipidemia [
We observed that the LDL level was correlated positively with the TSH level before HAART commencement, although there were no remarkable differences among patients with different thyroid function. According to Yingyun Gong and his colleagues, TSH elevated concentration of LDL-c by upregulating hepatic PCSK9 expression [
We noted significant changes in levels of TC and TG from the year HAART started to 2 years after HAART commencement in patients with normal baseline levels of TC and TG, as well as those diagnosed with IFG or DM. However, there were no significant changes from 2 years of HAART commencement, such as between year 2 and year 4. We suggest that disorders in lipid metabolism occur mainly in the early years after HAART commencement.
Levels of TC, HDL-C, and LDL-C were correlated negatively with VL before HAART commencement in the present study. HIV seroconversion was ascribed to reductions in levels of TC, HDL-C and LDL-C according to Multicenter AIDS Cohort Study [
Long-term HAART might result in dyslipidemia. Nevirapine was found to be positively correlated with TC levels in this study while there was no statistical significance between tenofovir and emtricitabin with TC levels. However, Alex Marzel et al. concluded that HAART containing tenofovir was negatively correlated with TC levels [
The main strength of our study was that it was a 6-year follow-up study with complete repeat measurement of the levels of TC and TG as well as HIV viral load. Moreover, with continuous and good compliance, therapeutic effect was satisfactory.
There were three main limitations in our study. First, we found the TC level 6 years after HAART commencement to be lower than that at 4 years, but it increased persistently in the first 4 years from HAART commencement. This phenomenon may have been because the TC level benefited from HIV clearance, an increase in the CD4+ cell count with improvement of disease, and supplementation with thyroid hormones, or it was temporary, in which case the TC level could increase again. Therefore, following up TC levels for a longer time would be a good strategy. Second, the number of patients diagnosed with subclinical hypothyroidism or hyperthyroidism was few. As a result, enlarging the sample size is crucial to increase the reliability of our data. Finally, levels of TSH, FT3, FT4, HDL-C, LDL-C, VLDL-C, and the LDL-C:HDL-C ratio were not detected regularly, so we could not observe changes in these indices.
HIV-infected patients tend to suffer from dyslipidemia, especially those with hyperglycemia and thyroid dysfunction. It may be necessary to prophylactically treat this population with antilipidemic medicine as even patients with normal TC are inclined to dyslipidemia.
The SPSS data used to support the findings of this study are available from the corresponding author upon request.
All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards.
The authors declare that there are no conflicts of interest.
The authors would like to thank the native English speaking scientists of Elixigen Company (Huntington Beach, California) for editing their manuscript. This work was supported by the National Science and Technology Major Project during the 13th Five-Year Plan of China (Grant no. 2017ZX10105001-005).
See Table 1a-3 in the Supplementary Material for comprehensive image analysis.