Gradual increases in dyslipidemia among male patients with human immunodeciency virus primarily treated with tenofovir plus lamivudine plus efavirenz: a 3-year follow-up study

Introduction: Since the development of antiretroviral therapy (ART) with TDF plus 3TC plus EFV, this specic regimen has not been studied enough with long-term lipid and uric monitoring. Methods: A prospective follow-up cohort study was performed. Sixty-one treatment-naive male patients with HIV were divided into three groups based on their baseline CD4+ T cell count (26, 12, and 23 patients in the <200, 200 to 350, and >350 groups, respectively). The lipid and purine metabolism parameters of the patients over 144 weeks were analyzed. Result: TG, TC, LDL-c and HDL-c levels all gradually increased over 144 weeks, but the increases in TC levels and HDL-c levels were signicant (P=0.001, 0.000, respectively). Moreover, the percentages of hypercholesterolemia, hyper LDL cholesterolemia and hypertriglyceridemia all showed gradual and nonsignicant increases; the percentage of low HDL cholesterolemia showed a gradual and signicant decrease (P=0.0007). Furthermore, the lower the baseline CD4+ T lymphocyte counts were, the higher the TG levels were and the lower the TC, LDL-c and HDL-c levels were. However, only baseline LDL-c levels differed signicantly between the three groups (P=0.0457). Although the UA level and the percentages of hyperuricemia gradually increased over 144 weeks, there was no signicant difference between the different follow-up time point groups or between the three CD4+ T cell count groups (all P>0.05). Further analyses revealed that the main factors contributing to lipid metabolism were age, anthropometric parameters and follow-up weeks, and virus load was the main factor contributing to uric acid levels. Conclusions: These ndings provide a reference for clinicians to monitor lipid metabolism parameters closely during long-term ART with the TDF plus 3TC plus EFV regimen. percentages. One-way ANOVA was used to compare metabolism parameters from baseline to 144 weeks, and a paired t-test was used to compare metabolism parameters between baseline and some follow-up time points. The Kruskal-Wallis H (K) test for K independent samples was used to compare the percentage of dyslipidemia and hyperuricemia from baseline to 144 weeks. The Mann-Whitney test for two independent samples was used to compare the percentage of dyslipidemia and hyperuricemia between baseline and some follow-up time points. One-way ANOVA was used to compare metabolism parameters between the three different CD4+ T cell count groups at the same time point. Two-way ANOVA was used to compare metabolism parameters between the three different CD4+ T cell count groups from baseline to 144 weeks. A p value <0.05 was considered statistically signicant. 3-year follow-up cohort study was the rst to report the use of long-term lipid and uric monitoring to assess the ecacy of the TDF+3TC+EFV regimen, baseline CD4+ T cell count, age, anthropometric parameters, and immunological and virological indicators of lipid and purine metabolic parameters in male patients with HIV undergoing primary treatment. The results showed that in patients treated with the TDF+3TC+EFV regimen for 3 years, TC, LDL-c and HDL-c levels gradually increased, especially TC and HDL-c levels. TG levels rst gradually decreased and then gradually increased. Moreover, the percentages of hypercholesterolemia, hyper-low-density lipoprotein cholesterolemia and hypertriglyceridemia all gradually increased, while the percentage of hypo-high-density lipoprotein cholesterolemia gradually decreased. This shows that in the early stage of ART treatment, disordered lipid metabolism was improved, especially hypo-high-density lipoprotein cholesterolemia. However, along with prolonged ART, the proportion of hyperlipidemia and hypertriglyceridemia gradually increased, especially hyper-low-density lipoprotein cholesterolemia and hypertriglyceridemia after 48 weeks and hypercholesterolemia after 120 weeks, which ranged from 9.84%, 8.2%, and 24.59% at baseline to 21.43%, 21.43%, and 50.00% at 144 weeks, respectively. According to Spearman correlation analysis and based on


Introduction
In recent years, the numbers of patients with human immunode ciency virus (HIV) and acquired immune de ciency syndrome (AIDS) have sharply increased. By the end of 2019, approximately 38 million people worldwide lived with HIV, and 33 million people died of HIV-related diseases [1]. At the end of October 2019, a total of 958,000 patients in China were living with HIV [2,3]. At the end of September 2018, a total of 262,000 patients died of HIV-related diseases in China [2,3].
The most effective treatment for AIDS is antiretroviral therapy, which can prolong life expectancy and improve quality of life [4,5].
When a patient's CD4+ T cell count reaches more than 350/mm3 and the viral load reaches undetectable levels within the rst year of starting treatment, AIDS patients are predicted to have a normal life expectancy [4,5]. The cumulative survival rates of AIDS patients have increased markedly [6][7][8]. As of 2017, 20.9 million patients with HIV had received antiretroviral treatment worldwide.
However, metabolic abnormalities, cardiovascular risk factors, and osteoporosis have become important factors affecting the prognosis and quality of life of AIDS patients [9][10][11][12].
HIV infection itself and antiretroviral therapy (ART) treatment drugs can cause dyslipidemia. However, as a rst-line ART program launched since the National Twelfth Five-Year Plan in China, the tenofovir (TDF) plus lamivudine (3TC) plus efavirenz (EFV) regimen has a weaker effect on lipid metabolism. Our previous study showed that newly diagnosed male AIDS patients had decreased total cholesterol (TC) levels, uric acid (UA) levels [13,14], and high-density lipoprotein cholesterol (HDL-c) levels as well as increased triglyceride (TG) levels, especially patients with CD4+ counts <200/µl. The dyslipidemia and decreased UA levels gradually returned to normal 4 weeks after initial ART with the TDF plus 3TC plus EFV regimen [13,14]. This speci c regimen has not been studied enough using long-term lipid and uric monitoring, which is the focus of the current study.

Study population
A prospective cohort study was conducted on sixty-one male patients with HIV who were treatment-naive at the Public and Health Clinic Centre of Chengdu from October 1, 2012, to December 31, 2017. Among them, the rst 50 cases have been described in the literature [13][14][15].
The following exclusion criteria were used in this study: patients with acute infections; patients with opportunistic infections or AIDS-related malignant tumors at the time of enrollment; patients with opportunistic infection occurring within 3 months before enrollment and still in unstable condition within 2 weeks before enrollment; patients with hemoglobin <9g/dL, white blood cell count <2000/µL, neutrophil count <1000/µL, platelet count <75000/µL, serum creatinine > 1. 5-fold upper limit of the normal value   (ULN), aspartate aminotransferase/alanine aminotransferase/alkaline phosphatase >3-fold ULN, total bilirubin >2-fold ULN, serum   creatine phosphokinase > 2-fold ULN, or creatinine clearance rate <60 mL/min; women who were pregnant or lactating; current drug   users; patients with severe mental or neurological diseases; patients with a history of alcoholism; and patients with severe   digestive tract ulcers. AIDS, dyslipidemia and hyperacidemia diagnostic criteria were applied according to the corresponding guidelines. [16][17][18].
The participants were divided into three groups according to their baseline CD4 T cell counts: there were 26, 12, and 23 patients in the <200, from 200 to 350 and >350 cells/µl groups, respectively.

Measurement of anthropometric parameters
The subjects fasted overnight for at least 12 hours. At 8:00 am the next day after emptying stool and urine, anthropometric parameters, including height, body weight (BW), body fat weight, body nonfat weight, body mass index (BMI) and body fat percentage, were measured by specially trained researchers using a body fat measuring instrument.

Detection of laboratory indicators
The subjects fasted overnight for at least 12 hours. At 8:00 am the next day, the venous blood of those patients was drawn to measure TC, LDL-c, HDL-c, TG, UA, HIV viral nucleic acid (HIVRNA), and T lymphocyte subsets.
TC, TG, HDL-c, LDL-c and UA levels were measured by the enzymatic method of an automatic biochemical analyzer purchased from Zhejiang Eastern European Biological Products Company. HIVRNA was detected by uorescent quantitative PCR; T lymphocyte subsets (including CD3+ count, CD4+ count, CD8+ count, CD3+%, CD4+%, CD8+%) were measured by ow cytometry using a Beckman ow cytometer.
Databases were established according to the needs of the research by two researchers simultaneously collecting and entering data. Approximately 30% of the data were randomly selected by the researchers to assess data integrity, authenticity, and accuracy.

Patient and public involvement
Patients and the public were involved in the development of the research question or in the design of the study. Patients received oral and written information about this study; however, they were not involved in the recruitment and implementation of the study. In addition, the burden of the intervention was assessed by the patients themselves. After signing an informed consent form by the participants, they were assessed for eligibility prior to data collection.

Statistical methods
The Statistical Package for the Social Sciences software version 17.0 (IBM Inc., Armonk, NY, USA) and GraphPad Prism 8 (GraphPad) software were used for statistical analysis. TC, LDL-c, HDL-c, TG and UA levels normally distributed, and statistical analysis was conducted directly. Nonnormally distributed HIVRNA levels were subjected to natural logarithmic transformation before statistical analysis. Quantitative data were expressed as χ±SD, and categorical data were expressed as rates or percentages. One-way ANOVA was used to compare metabolism parameters from baseline to 144 weeks, and a paired t-test was used to compare metabolism parameters between baseline and some follow-up time points. The Kruskal-Wallis H (K) test for K independent samples was used to compare the percentage of dyslipidemia and hyperuricemia from baseline to 144 weeks. The Mann-Whitney test for two independent samples was used to compare the percentage of dyslipidemia and hyperuricemia between baseline and some follow-up time points. One-way ANOVA was used to compare metabolism parameters between the three different CD4+ T cell count groups at the same time point. Two-way ANOVA was used to compare metabolism parameters between the three different CD4+ T cell count groups from baseline to 144 weeks. A p value <0.05 was considered statistically signi cant.

Ethical consideration
The study was approved by the hospital ethics committee of the Public and Health Clinic Centre of Chengdu. All patients gave written informed consent.

Baseline conditions
Sixty-one treatment-naive male patients with HIV in the Public and Health Clinic Centre of Chengdu from October 1, 2012, to December 31, 2017, were divided into three groups according to their baseline CD4+ T cell count: there were 26, 12, and 23 patients in the <200, 200 to 350, and >350 cell/µl groups, respectively. There were 42 cases of infection through homosexual contact, 13 cases of infection through heterosexual contact and 5 cases involving both types of sexual contact. The general information, baseline immunity and virological indicators, and lipid metabolism parameters of 61 patients are shown in Table 1.

Effectiveness of ART with the TDF plus 3TC plus EFV regimen
In 61 patients, the average CD4+ T cell count ( Figure 1A) gradually increased from 319.80 cell/µl at baseline to 464.85 cell/µl at 96 weeks after ART, and the average viral load ( Figure 1B) decreased rapidly from 49846.32 IU/ml at baseline to undetectable levels (measured by a high-precision detection method) at 72 weeks. The percentage of viral load reaching undetectable levels ( Figure   1C) ranged from 21.31% at 12 weeks to 100.00% at 72 weeks.
Long-term dynamic changes in anthropometric parameters after treatment with TDF+3TC+EFV The body weight ( TC, LDL-c, HDL-c, TG and UA levels ( Figure 3A, 3B, 3C, 3D and 5A) all gradually increased with prolonged ART, but the increases were small, and only the increases in TC levels ( Figure 3A) and HDL-c levels ( Figure 3C) Figure 3A) and signi cant differences in HDL-c levels at 48, 96 and 144 weeks ( Figure 3C) Figure 3B), TG ( Figure 3D) or UA levels ( Figure 5A) between different time points compared with baseline (all P>0.05).
The percentages of hypercholesterolemia ( Figure 4A), hyper-low-density lipoprotein cholesterolemia ( Figure 4B) and hypertriglyceridemia ( Figure 4D) Figure 4C) gradually decreased with prolonged ART, and a signi cant difference was found from baseline to 144 weeks ( Figure 4C) Figure 5B) slightly decreased with the extension of the HAART treatment time, but this decrease was not signi cant (P>0.05).

Effect of baseline CD4+ T cell count on lipid and purine metabolic parameters after treatment with TDF+3TC+EFV
The lower the CD4+ T cell count at baseline was, the higher the TG levels ( Figure 6D) and the lower the TC (Figure 6A), LDL-c ( Figure 6B), HDL-c ( Figure 6C) and UA levels were ( Figure 7); moreover, these changes were maintained throughout the follow-up period after ART treatment. However, there was no signi cant difference in the change from baseline to 96 weeks between the three different CD4+ T cell count groups (all P>0.05). The difference of the LDL-c levels ( Figure 6B) at baseline was signi cant between the three different CD4+ T cell count groups (2.22 vs. 2.70, 2.67 mmol/L, F=3.256, P=0.0457). TC levels ( Figure 6A) and HDL-c levels ( Figure 6C) all gradually increased along with prolonged ART regardless of the CD4+ T cell count at baseline.

The contributing factors of lipid and purine metabolic parameters
According to Spearman correlation analysis, nonalcoholic fatty liver disease, age, body weight, BMI, body fat weight, and body fat percentage were all positively correlated with TC, TG and LDL-c levels; body nonfat weight was positively correlated with LDL-c levels; and follow-up duration was positively correlated with TC and HDL-c levels. In contrast, nonalcoholic fatty liver disease, body weight, BMI, body fat weight, body fat percentage and body nonfat weight were all negatively correlated with HDL-c levels ( Table  2). Based on multiple stepwise regression analysis, the factors that affected TC levels were body weight, age, body nonfat weight and follow-up duration; the factors that affected TG levels were body weight, body nonfat weight, BMI and age; the factors that affected HDL-c levels were BMI, CD3+CD4+ count and body weight; and the factors that affected LDL-c levels were BMI and followup duration (Table 3).
In addition, nonalcoholic fatty liver disease, body weight, body nonfat weight, CD3+CD4+ cell count and virus load were all positively correlated with UA levels, while age was negatively correlated with UA levels ( Table 2). Based on multiple stepwise regression analysis, only virus load was a contributing factor for UA (Table 3).
In this prospective cohort study, we aimed to identify the long-term dynamic characteristics of lipid and purine metabolism and to identify in uencing factors after ART with the TDF+3TC+EFV regimen. To our knowledge, this prospective 3-year follow-up cohort study was the rst to report the use of long-term lipid and uric monitoring to assess the e cacy of the TDF+3TC+EFV regimen, baseline CD4+ T cell count, age, anthropometric parameters, and immunological and virological indicators of lipid and purine metabolic parameters in male patients with HIV undergoing primary treatment. The results showed that in patients treated with the TDF+3TC+EFV regimen for 3 years, TC, LDL-c and HDL-c levels gradually increased, especially TC and HDL-c levels. TG levels rst gradually decreased and then gradually increased. Moreover, the percentages of hypercholesterolemia, hyper-low-density lipoprotein cholesterolemia and hypertriglyceridemia all gradually increased, while the percentage of hypo-high-density lipoprotein cholesterolemia gradually decreased. This shows that in the early stage of ART treatment, disordered lipid metabolism was improved, especially hypo-high-density lipoprotein cholesterolemia. However, along with prolonged ART, the proportion of hyperlipidemia and hypertriglyceridemia gradually increased, especially hyper-low-density lipoprotein cholesterolemia and hypertriglyceridemia after 48 weeks and hypercholesterolemia after 120 weeks, which ranged from 9.84%, 8.2%, and 24.59% at baseline to 21.43%, 21.43%, and 50.00% at 144 weeks, respectively. According to Spearman correlation analysis and based on multiple stepwise regression analysis, the contributing factors of lipid and purine metabolic parameters included age, anthropometric parameters, and immunological and virological indicators.
This alternate cause of early lipid and purine metabolism may be partially due to the changes in appetite and weight gain among patients who contracted AIDS after ART treatment. However, this study found that within 48 weeks after ART, all anthropometric parameters of those patients, including body weight, body fat weight, lean body weight, body fat percentage and BMI, did not increase. Body weight increased by 2.81 kg, and body fat weight increased by 2.74 kg from 48 to 144 weeks, but lean body weight did not increase over the 144 weeks.
This study was a cohort study with a follow-up time of 144 weeks. The proportion of patients with abnormal lipid metabolism during the whole study period was less than that reported in the literature. The reasons may be related to the younger age of patients in this cohort, the duration of follow-up was not too long, and the TDF+3TC+EFV regimen had little effect on metabolism.
It is still necessary to expand the sample size, increase the number of female patients, extend the follow-up time and increase other ART regimens for further discussion.
A previous study found that a high CD4+ cell count was a risk factor for hypertriglyceridemia, while a CD4+ cell count less than 200 copies/mm 3 increased the risk of hypercholesterolemia [29]. Regardless of whether the initial treatment regimen was based on D4T, the risk of hyperlipidemia in HIV/AIDS patients aged 50 and above was signi cantly higher than that in young HIV/AIDS patients aged under 40 [33]. In this prospective 3-year follow-up cohort study, the long-term effect of baseline CD4+ T lymphocyte count on lipid parameters was that the lower the baseline CD4+ T lymphocyte count was, the higher the TG levels were and the lower the TC, LDL-c and HDL-c levels were. After comparisons between different time points over the 144 weeks and comparisons with baseline levels, no signi cant difference in UA levels was found. The lower the baseline CD4+ T cell count was, the higher the UA levels were, and these changes were maintained throughout the follow-up period after ART treatment; however, there was no statistical signi cance in the change from baseline to 144 weeks between the three different CD4+ T cell count groups. That is, the TDF+3TC+EFV regimen and CD4+ T cell count at baseline had no long-term dynamic effects on purine metabolism.

Conclusions
These ndings provide a reference for the clinical use of long-term lipid and uric monitoring during long-term ART with the TDF plus 3TC plus EFV regimen.

Declarations
Ethics approval and consent to participate The study was approved by the hospital ethics committee of the Public and Health Clinic Centre of Chengdu. All patients gave written informed consent.

Consent for publication
All authors read and approved the nal manuscript, and gave written informed consent for publication.
Availability of data and material All data, models, or code generated or used during the study are available from the rst author by request: Dafeng Liu, E-mail: ldf312@126.com

Competing interests
The authors declare that they have no con icts of interest to this study.