Abstract
Introduction:
Immune dysregulation persists in people with HIV (PWH) on antiretroviral therapy (ART) and may lead to accelerated vascular aging and cardiovascular disease (CVD). While delayed time to initiation of ART has been linked to worse cardiovascular outcomes, the effect of ART initiation during acute infection on these outcomes is not well understood.
Methods:
Participants were enrolled from the SEARCH010/RV254 acute HIV (AHI) and HIV-NAT chronic HIV (CHI) cohorts in Thailand. Participants with 6-year follow-up and viral suppression (VL < 50 copies/uL) at follow-up were included. Both unmatched cohorts and age and gender-matched cohorts were analyzed. Demographics, HIV labs, and cardiovascular risk factors from enrollment and 6-year follow-up were obtained from electronic records. Framingham Risk Score (FRS), Vascular age (VA), Vascular age deviation (VAD), and 10-year atherosclerotic cardiovascular disease (ASCVD) risk were calculated from previously published equations. Vascular outcomes in AHI and CHI cohorts were compared, and univariable and multivariable linear regression were used to investigate risk factors associated with worse vascular scores.
Results:
373 AHI participants and 608 CHI participants were identified. AHI participants had younger age, higher prevalence of syphilis, and lower prevalence of prior hepatitis B, tuberculosis, diabetes, and hypertension. Higher CD4 T-cell and lower CD8 T-cell counts were seen in the AHI cohort at enrolment and 6-year follow-up. In all participants, the AHI cohort had a lower median FRS (p<0.001) and VA (p<0.001), but higher VAD (p < 0.001). However, in matched cohorts, no differences were found in FRS-based outcomes. In all participants, higher VAD after 6 years of ART associated with higher BMI (p<0.001) and higher CD4 T-cell count (p<0.001), which persisted in multivariable analysis. When FRS components were analyzed individually, CD4 T-cell count associated only with male sex and cholesterol.
Conclusions
We did not identify differences in FRS-based vascular outcomes at six years in matched cohorts of participants who started ART during AHI vs CHI. We identified a correlation between higher CD4 T-cell count and worse FRS-based vascular outcomes, which may be driven by underlying metabolic risk factors. Further study is needed to confirm these findings and evaluate underlying mechanisms.
Keywords: Framingham Risk Score, Acute HIV, CD4 T-cell, cardiovascular risk, HIV, Vascular Age
Introduction
There are over 39 million people living with HIV (PWH) worldwide, with the greatest burden in low- and middle-income countries in Sub-Saharan Africa and Southeast Asia (1). With an aging population of PWH worldwide, non-communicable complications of HIV such as cardiovascular disease (CVD) have become a significant public health concern (1–3). Years of research confirm that delayed antiretroviral therapy (ART) initiation, ART interruptions, and suboptimal adherence with episodic viremia significantly increase the risk of cardiovascular complications in PWH (4). A lower nadir CD4 T-cell count has also been associated not only with the risk of CVD but also with various conventional vascular risk factors including hypertension (HTN), diabetes mellitus (DM), metabolic syndrome and vascular inflammation in PWH on ART (5–8). However, earlier ART initiation at a CD4 T-cell count beyond 500 cells/μL during chronic HIV infection (CHI) has only been shown to yield marginal benefits in terms of mitigating CVD risk (9). To date, it remains unclear if initiating ART during acute HIV infection (AHI) could lead to significant reduction in CVD risk compared to those who started ART during CHI.
The Framingham risk score (FRS) is a measure of 10-year cardiovascular risk calculated from easily measurable clinical factors including age, sex, cholesterol, and smoking status (10). The validity of FRS has also been independently demonstrated in Asian populations (13).The FRS can also be converted to an equivalent vascular age (VA) or heart age, and vascular age deviation (VAD), which can be utilized as other CVD health outcomes that are easily interpretable by patients and clinicians (11). In addition to higher rates of adverse clinical CVD outcomes, PWH also have higher FRS-based and atherosclerotic cardiovascular disease (ASCVD)-based cardiovascular risk scores compared to people without HIV (PWoH) (12).
The primary objective of this study was to evaluate the potential benefits of very early ART on CVD risk by comparing FRS-based CVD metrics between participants in two Thailand-based, observational HIV cohorts that prospectively enroll PWH during AHI or CHI. The secondary objective was to identify common risk factors associated with worse FRS-based metrics in both cohorts.
Methods
Study Participants
Study participants were from the SEARCH010 (RV254) AHI and the HIV Netherlands Australia Thailand (HIV-NAT) 006 CHI cohort studies conducted in Bangkok, Thailand. All participants provided written informed consent for study enrollment and study procedures. Both studies were approved by the Institutional Review Boards of Chulalongkorn University in Bangkok, Thailand and participating organizations.
RV254 Acute HIV Cohort
The RV254 AHI cohort is an ongoing prospective study that began in April 2009, focusing on long-term outcomes following ART initiated during AHI (NCT00796146 and NCT00796263) (14). In this cohort, AHI is defined as 4th generation enzyme immunoassay (EIA) negative and nucleic acid testing positive (Fiebig stage I/II) or 4th generation EIA positive and 2nd generation EIA negative (Fiebig stage III-V) (15). Participants undergo laboratory blood testing, substance use screening, co-infection screening, and neuropsychiatric assessments during pre-ART assessment (AHI, baseline) and during regular longitudinal follow-up. In RV254, we included all participants who had completed their week 288 (approximately 6-year) follow-up visit, with HIV suppression (plasma HIV RNA <50 copies/uL), and had complete data available to calculate FRS at 6-year visit (described below).
HIV-NAT 006 Chronic HIV Cohort
The HIV-NAT 006 HIV cohort is a prospective study following participants enrolled with CHI. Chronic HIV is defined as 4th generation EIA positive and 2nd generation EIA positive, or 4th generation EIA positive and known exposure > 2 months from testing. Briefly, participants are followed up every 6–12 months and undergo laboratory blood testing, substance use screening and co-infection screening (16, 17). Participants are initiated on treatment at baseline visit and remain on suppressive ART. In HIV-NAT 006, we included participants who had a plasma HIV RNA during baseline visit, a minimum of 5 years of follow-up, and complete data available to calculate FRS at a selected visit with HIV suppression occurring between 5–7 years post-enrollment.
Data Collection
Participants’ information was extracted from the respective databases of RV254 and HIV-NAT 006. Specifically, HIV-related parameters (plasma HIV RNA, CD4 and CD8 T-cell counts), body mass index (BMI) and factors for FRS calculation were collected at cohort enrollment (baseline) for both cohorts, during the week 288 (6-year) visit in RV254, and during the selected visit occurring between 5–7 years (6-year) post-enrollment in HIV-NAT006. As different self-reported questionnaires were used in the two cohorts, cigarette smoking status (smoker/nonsmoker) and prior illicit drug use (ever used: yes/no) were recorded as binary factors. Exposure to various antiretroviral agents, defined by any usage over 3 months, were recorded. History of concurrent infections, including tuberculosis (TB), hepatitis C (determined by Anti-HCV antibody), hepatitis B (determined by hepatitis B surface antigen), and syphilis (determined by treponemal specific tests or Venereal Disease Research Laboratory, VDRL) were identified based on adverse event reports and laboratory test results recorded between the baseline and 6-year visit. Hepatitis C status was further stratified into active or treated/resolved based on positive or negative HCV RNA detection.
Framingham Risk Score Based Vascular Outcomes
FRS was calculated using age, sex, lipid profile, systolic blood pressure, smoking, diabetes mellitus (DM), and antihypertensive usage status as described elsewhere (10). FRS was converted using a standardized calculator to 10-year estimated risk of adverse vascular event outcomes (10). Vascular age (VA), or heart age, is a term used to quantify an individual’s vascular health based on their FRS. VA is determined by matching an individual’s 10-year cardiovascular risk % with that of a person sharing the same 10-year cardiovascular risk % but with no vascular risk factors. For example, if a 40 year old man has a 10-year cardiovascular risk of 5.6%, his VA would be 45 years because a 45 year old man with no cardiovascular risk factors has an estimated 10-year cardiovascular risk of 5.6% (11). Individuals with more vascular risk factors and worse control over these factors will have a higher FRS and VA, resulting in a deviation from their actual age, denoted as Vascular Age Deviation (VAD, calculated as vascular age minus actual age). Conversely, in individuals with no vascular risk factors, VA equals actual age, and VAD is 0. VAD allows a comparison of vascular health between individuals irrespective to their chronological age. Higher VAD indicates worse vascular health. Of note, a VA of 30 was set as the lowest limit in the 2008 FRS report. Because of this, to avoid overestimating VA for participants who were younger than 30 with no vascular risk factors at the 6-year visit, their actual age was assigned as their VA. We also calculated the 10-year ASCVD risk score using the pooled cohort equation (18).
Cohort Matching
An age and sex matched combined cohort was created from the unmatched cohort using sensitivity analysis with a 1:1 ratio of AHI to CHI participants. Age matching was performed using a 5-year age band. Details of the matched cohorts are shown in Supplementary Table 1.
Statistical Analysis
We compared FRS-based 10-year CVD risk, VA, VAD and Pooled Cohort Equation-based 10-year ASCVD risk between AHI and CHI participants in both unmatched and matched combined cohort. Continuous variables are presented in median (interquartile range [IQR]), and categorical variables as frequencies and percentages. Continuous variables were compared using student’s t tests or Wilcoxon rank-sum tests, whereas categorical variables were compared using chi-square or Fisher’s exact test. Univariable and multivariable linear regression were used to investigate the risk factors associated with increased FRS, VA, VAD and 10-year ASCVD risks. The variables in the univariate models with p-value < 0.2 were included in the multivariable model. All analyses were performed using Stata version 17.0 (StataCorp, College Station, Texas 77,845 USA).
Results
Demographics and HIV Parameters (Table 1)
Table 1.
Demographics of Acute and Chronic HIV Cohorts at enrollment and 6-year follow-up
| Characteristics | Baseline | 6 years | ||||
|---|---|---|---|---|---|---|
| AHI N=369 | CHI N=608 | p-value | AHI N=369 | CHI N=608 | p-value | |
| Age, years | 26 (23, 31) | 33 (28, 38.5) | <0.001 | 32 (28, 37) | 38 (33, 44) | <0.001 |
| Male, n (%) | 357 (96.75) | 402 (66.12) | <0.001 | |||
| CD4 T-cell count, cells/mm3 | 371 (266, 495) | 232 (127, 335) | <0.001 | 681 (558, 855) | 578 (428, 747) | <0.001 |
| CD8 T-cell count, cells/mm3 | 515 (338, 882) | 838 (592, 1145) | <0.001 | 629 (468, 797) | 728 (566, 944) | <0.001 |
| CD4/CD8 ratio | 0.71 (0.43, 1.01) | 0.26 (0.14, 0.39) | <0.001 | 1.11 (0.91, 1.37) | 0.78 (0.56, 1.05) | <0.001 |
| HIV-1 RNA, log10 copies/mL | 5.90 (5.34, 6.72) | 4.82 (4.36, 5.18) | <0.001 | |||
| Hepatitis B co-infection, n (%) | 16 (4.34) | 97/539 (18.00) | <0.001 | |||
| Hepatitis C co-infection, n (%) | 25 (6.78) | 26/547 (4.75) | 0.191 | |||
| Positive syphilis antibody, n (%) | 195 (52.85) | 23/151 (15.23) | <0.001 | |||
| Tuberculosis, n (%) | 10 (2.71) | 64 (10.53) | <0.001 | |||
| Hypertension, n (%) | 7 (1.90) | 16 (2.63) | 0.52 | 17 (4.61) | 53 (8.72) | 0.016 |
| Diabetes mellitus, n (%) | 1 (0.27) | 10 (1.64) | 0.060 | 3 (0.81) | 26 (4.28) | 0.001 |
| Smoking, n (%) | 36 (9.76) | 113 (18.59) | 0.60 | 48 (13.01) | 113 (18.59) | 0.023 |
| Ever drugs used, n (%) | 29 (7.92) | 20 (4.62) | 0.052 | |||
| Total cholesterol, mg/dL | 172 (151, 196) | 176 (155, 200) | 0.061 | 200 (175, 230) | 195 (172, 221) | 0.014 |
| HDL cholesterol, mg/dL | 38 (32, 46) | 44 (36, 53) | <0.001 | 49 (42, 57) | 48 (41, 57) | 0.22 |
| Systolic BP, mmHg | 117 (108, 125) | 118 (110, 125) | 0.57 | 121 (112, 130) | 118 (110, 128) | 0.12 |
| Diastolic BP, mmHg | 74 (67, 81) | 74 (70, 80) | 0.15 | 76 (71, 83) | 75 (69, 83) | 0.23 |
| Statin used, n (%) | 8 (2.17) | 49 (8.06) | <0.001 | |||
Characteristics of study participants at baseline and 6-year follow-up. Data presented as median (IQR) for continuous variables and as n (%) for categorical variables. BP: blood pressure; HDL: high-density lipoprotein
The study included 977 participants, with 369 from the RV254 cohort and 608 from the HIV-NAT 006 cohort, all of whom met the inclusion criteria described above. At baseline, AHI participants were younger than CHI participants (median age 26 [IQR 23–31] vs. 33 [IQR 28–39] years, p<0.001) and had a higher percentage of males (96% vs 66%, p<0.001). AHI participants had higher baseline plasma HIV RNA levels (median 5.90 [IQR 5.34–6.72] vs 4.82 [IQR 4.36–5.18] log10 copies/mL, p<0.001) and baseline CD4 T-cell counts (median 371 [IQR 266–495]) vs. 232 [IQR 127–335] cells/mm3, p<0.001) than CHI participants. The difference in CD4 T-cell count remained significant at 6-year follow-up (median 681 [IQR 558–855] vs. 578 [IQR 428–747] cells/mm3, p<0.001). Additional results are shown in Table 1.
Co-infections and Substance Use
AHI participants had a lower prevalence of HBV co-infection (4% vs. 18%, p<0.001) and history of TB infection (3% vs. 11%, p<0.001) compared to CHI participants. In contrast, the prevalence of a positive syphilis serology was significantly higher in the AHI cohort (53% vs. 15% p<0.001). There was no significant difference in HCV co-infection or reported drug use between the two cohorts.
Cardiovascular Risk Factors
The CHI cohort had higher prevalence of DM (4% vs. 1%, p=0.001) and HTN (9% vs. 5%, p=0.016) than the AHI cohort at 6 years. With more frequent usage of lipid-lowering agents (statins) in the CHI cohort (8% vs. 2%, p<0.001), CHI participants had slightly lower total cholesterol levels than AHI participants at 6-year follow-up (195 [IQR 172–221] vs. 200 [IQR 175–230] mg/dL, p=0.014). The prevalence of smoking was comparable between cohorts at baseline, but higher in the CHI cohort at 6-year (18% vs. 13%, p=0.023).
Antiretroviral Agent Exposure
At 6 years, the two cohorts demonstrated significantly different exposures to various antiretroviral drug classes. Compared to CHI participants, AHI participants had a lower frequency of protease inhibitor (PI) exposure (29% vs. 45%, p<0.001), but a higher prevalence of exposure to non-nucleoside reverse transcriptase inhibitors (NNRTIs) (99% vs. 73%, p<0.001), abacavir (ABC) (82% vs. 5%, p<0.001), and integrase strand transfer inhibitors (INSTIs) (95% vs. 12%, p<0.001).
Framingham Risk Score Related Outcomes (Table 2)
Table 2.
Vascular Outcomes at 6-year follow-up in all participants
| Characteristics | 6-year follow-up | ||
|---|---|---|---|
| AHI N=369 | CHI N=608 | p-value | |
| FRS | 2 (0, 5) | 4 (1, 7) | <0.001 |
| FRS-based 10-yr CVD risk (%) | 2.3 (1.6, 3.9) | 2.8 (1.6, 5.6) | 0.001 |
| FRS-based 10-yr CVD risk’ by category, n (%) | <0.001 | ||
| Low risk <10% | 354 (95.9) | 545 (89.6) | |
| Intermediate 10–20% | 6 (1.6) | 44 (7.2) | |
| High 20% or more | 9 (2.4) | 19 (3.1) | |
| Vascular age, years | 34 (30, 40) | 38 (31, 45) | <0.001 |
| Vascular age deviation, years | 1 (−1, 6) | 0 (−5, 6) | <0.001 |
| ASCVD* (Pooled cohort equation) | 0.38 (0.17, 1.09) | 0.83 (0.36, 2.15) | <0.001 |
| 10-year ASCVD risk score, (% category) | 0.031 | ||
| <7.5% | 357 (96.8) | 569 (93.6) | |
| ≥7.5% | 12 (3.3) | 39 (6.4) | |
Cardiovascular risk assessment and comparison between acute and chronic HIV cohorts at the 6-year follow-up. Data presented as median (IQR) for continuous variables and as n (%) for categorical variables.
ASCVD risk was calculated using the American Heart Association and American College of Cardiology 2013 Pooled Cohort Equation risk calculator (18). FRS: Framingham Risk Score. ASCVD:10-year atherosclerotic cardiovascular disease.
At 6-year follow-up, the AHI cohort had a lower median FRS (2 [IQR 0–5] vs. 4 [IQR 1–7], p<0.001), VA (34 [IQR 30–40] vs. 38 [IQR 31–45], p<0.001) and ASCVD risk score (0.38, IQR 0.17–1.09 vs 0.83 IQR 0.36–2.15 p<0.001). More participants in the AHI cohort had a low 10-year CVD risk (<10%) compared to those in the CHI cohort (96% vs. 90%, p<0.001). Similarly, 97% of AHI participants had a low 10-year ASCVD risk (<7.5%) using the pooled cohort equation, compared to 94% of CHI participants (p=0.031). However, participants in the AHI cohort had a slightly higher VAD when compared to those in the CHI cohort (1 [IQR −1 to 6] vs. 0 [IQR −5 to 6], p<0.001). Exclusion of statin users in both cohorts did not alter the statistical outcomes.
Framingham-risk Score Based Outcomes in the Matched Cohort
In the age- and sex-matched cohort with 264 AHI and 264 CHI participants, AHI participants continued to have higher CD4 T-cell counts and CD4/CD8 ratio, lower CD8 T-cell counts, a higher frequency of syphilis but a lower frequency of HBV infection compared to CHI participants, similar to the unmatched cohort. However, frequencies of DM and HTN were no longer significantly different between AHI and CHI participants (Supplementary Table 2). Additionally, there was no difference in FRS-based 10-year CVD risk, ASCVD risk, VA, or VAD between AHI and CHI participants in the matched cohort (Supplementary Table 3). In short, these findings indicate that FRS-based metrics between the two cohorts became comparable following matching by sex and age. Similar to the unmatched cohort, exclusion of statin users in the matched cohort did not alter the statistical outcomes.
Factors Associated with VAD in the Unmatched Combined Cohort (Table 3)
Table 3.
Factors associated with VAD at 6-year follow-up in all participants (N=977)
| Univariable | Multivariable model | |||||
|---|---|---|---|---|---|---|
| Coef (95%CI), p-value | Model 1 | Model 2 | Model 3 | Model 4 | Model 5 | |
| Age at week 288≥40 years (vs. <40) | 1.25 (0.19, 2.31), p=0.021 | 1.74 (0.75, 2.74), p=0.001 | 1.81 (0.72, 2.90), p=0.001 | 1.77 (0.68, 2.86), p=0.001 | 1.80 (0.71, 2.89), p=0.001 | 1.79 (0.70, 2.88), p=0.001 |
| Sex (Male vs. female) | 6.19 (5.06, 7.32), p<0.001 | 6.85 (5.71, 7.99), p<0.001 | 7.58 (6.22, 8.94), p<0.001 | 7.49 (6.15, 8.83), p<0.001 | 7.39 (6.03, 8.75), p<0.001 | 7.40 (6.05, 8.74), p<0.001 |
| BMI | 0.70 (0.57, 0.83), p<0.001 | 0.69 (0.57, 0.81), p<0.001 | 0.56 (0.43, 0.69), p<0.001 | 0.57 (0.44, 0.70), p<0.001 | 0.56 (0.43, 0.70), p<0.001 | 0.57 (0.43, 0.70), p<0.001 |
| CHI cohort (vs. AHI cohort) | −1.45 (−2.47, −0.43), p=0.005 | 0.56 (−0.45, 1.57), p=0.278 | 0.22 (−0.85, 1.29), p=0.69 | −0.47 (−2.05, 1.11), p=0.561 | 0.39 (−0.77, 1.54), p=0.510 | 0.20 (−1.59, 1.99), p=0.824 |
| Drugs used (yes vs. no) | 1.81 (−0.39, 4.01), p=0.107 | 1.44 (−0.54, 3.43), p=0.154 | 1.47 (−0.52, 3.46), p=0.146 | 1.47 (−0.52, 3.47), p=0.148 | 1.49 (−0.51, 3.48), p=0.144 | |
| Change in CD4 from baseline (per 100-cells increase) | 0.31 (0.09, 0.53), p=0.006 | 0.24 (0.03, 0.46), p=0.025 | 0.27 (0.05, 0.48), p=0.015 | 0.25 (0.04, 0.46), p=0.022 | 0.25 (0.04, 0.46), p=0.022 | |
| CD4:CD8 ratio at week 288 (per 0.1-unit increase) | 0.05 (−0.06, 0.16), p=0.404 | |||||
| HBV (yes vs. no) | 0.24 (−1.31, 1.78), p=0.764 | |||||
| HCV (yes vs. no) | −0.07 (−2.29, 2.15), p=0.950 | |||||
| PI exposure (yes vs. no) | −0.79 (−1.81, 0.23), p=0.129 | 0.81 (−0.20, 1.83), p=0.117 | ||||
| NNRTI exposure (yes vs. no) | 1.50 (0.19, 2.81), p=0.024 | |||||
| INSTI exposure (yes vs. no) | 1.61 (0.61, 2.61), p=0.002 | |||||
| ABC exposure (yes vs. no) | 1.13 (0.08, 2.18), p=0.034 | |||||
| Duration on PI, year | N/A | |||||
| Duration on NNRTI, year | P=0.041 | |||||
| 0 | Ref | Ref | ||||
| >0 – 1.9 | 2.04 (0.42, 3.65), p=0.013 | 0.23 (−1.55, 2.00), p=0.800 | ||||
| ≥2 | 1.38 (0.04, 2.73), p=0.044 | 0.07 (−1.38, 1.51), p=0.930 | ||||
| Duration on INSTI, year | 0.0062 | |||||
| 0 | Ref | Ref | ||||
| >0 – 1.9 | 1.78 (−0.04, 3.6), p=0.056 | 0.14 (−2.02, 2.29), p=0.899 | ||||
| ≥2 | 1.59 (0.52, 2.66), p=0.004 | −0.21 (−2.02, 1.60), p=0.819 | ||||
| Duration on ABC, year | 0.025 | |||||
| 0 | Ref | Ref | ||||
| >0 – 1.9 | −0.18 (−2.06, 1.71), p=0.855 | −2.35 (−4.38, −0.33), p=0.023 | ||||
| ≥2 | 1.54 (0.4, 2.69), p=0.008 | −0.86 (−2.52, 0.79), p=0.305 | ||||
Data expressed as Coefficient (95% CI), p-value. Model 1 was adjusted for age, sex, BMI and cohort. Model 2 was adjusted for age, sex, BMI, cohort, drugs used, change in CD4 from baseline, and exposure to PI. Model 3 was adjusted for age, sex, BMI, cohort, drugs used, change in CD4 from baseline, and duration on ABC. Model 4 was adjusted for age, sex, BMI, cohort, drugs used, change in CD4 from baseline, and duration on NNRTI. Model 5 was adjusted for age, sex, BMI, cohort, drugs used, change in CD4 from baseline, and duration on INSTI. ABC: abacavir; BMI: body mass index; HBV: hepatitis B; HCV: hepatitis C; INSTI: integrase strand transfer inhibitor; NNRTI; non-nucleoside reverse transcriptase inhibitors; TB: tuberculosis.
As VAD is a measure of vascular health independent of chronological age, this measure was chosen as the dependent variable to identify potential HIV-related and unrelated factors associated with vascular risk at the 6-year follow-up. In univariable analysis, factors associated with higher VAD at the 6-year follow-up included older age (p<0.021), gender (p<0.001), higher BMI (p<0.001), greater increase of CD4 T-cell count from baseline (p=0.006), and exposure to NNRTI (p=0.024), INSTI (p=0.002) and ABC (p=0.034). Factors with a p-value <0.2 in univariable analysis were included in multivariable analysis. Multivariable analysis demonstrated that age, sex, higher BMI, and higher change in CD4 T-cell count from baseline were consistently associated with VAD across various statistical models designed to address collinearity in antiretroviral agent exposure in the two cohorts. Similar results were seen in association analysis with FRS and VA (Supplementary Tables 4 and 5).
Association between CD4 T-cell count, VAD, cohort, and smoking
The association between CD4 T-cell count and VAD was further evaluated within each cohort. In univariable analysis, higher CD4 T-cell count and greater increase in CD4 T-cell count from baseline were associated with higher VAD in both the AHI cohort (p<0.001, p=0.014) and CHI cohort (p=0.007, p=0.055) individually. However, in the multivariable analysis, the association between CD4 T-cell count and VAD persisted only in the AHI but not the CHI cohort (Supplementary Table 6).
As an association between CD4 T-cell count and cigarette smoking has been reported in the past (19), we further explored the association between CD4 T-cell count and VAD stratified by smoking status in the unmatched combined cohort as well in each AHI and CHI cohort. In the unmatched combined cohort, VAD was not associated with change in CD4 T-cell count (p=0.063) in non-smokers or in smokers (p=0.055) (Figure 1A). In the AHI cohort, change in CD4 T-cell count was associated with VAD in smokers (p=0.009) but not non-smokers (p=0.140). In contrast, in the CHI cohort change in CD4 T-cell count was not associated with VAD in either non-smokers (p=0.110) or smokers (p=0.521) (Figure 1B).
Figure 1: Correlation plot (Change in CD4 T-cell count from baseline and VAD) stratified by smoking status and Cohort.
A 6-year visit VAD (y-axis) plotted against Change in CD4 T-cell count from baseline (x-axis) for nonsmokers (pink, p=0.063) and smokers (blue, p=0.055) in all study participants. B. 6-year visit VAD (y-axis) plotted against Change in CD4 T-cell count from baseline (x-axis) for nonsmokers (pink) and smokers (blue). Left plot shows only AHI participants (p=0.009 for smokers, p=0.140 for nonsmokers). Right plot shows only chronic HIV participants (p=0.521 for smokers, p=0.110 for nonsmokers).
Association between change in CD4 T-cell Count and Framingham Risk Score Components
Additional data exploration was performed to determine if CD4 T-cell count was associated with specific FRS components in the combined cohort. In univariable analysis, change in CD4 T-cell count was associated with male gender (p=0.002) and higher total cholesterol at 6 years (p=0.001) (Table 4).
Table 4.
Association of Change in CD4 T-cell count from baseline and individual components of FRS
| Univariable | ||||
|---|---|---|---|---|
| coef | 95%CI | p-value | ||
| Age group | ||||
| <30 | Ref | |||
| 31–39 | 0.09 | (−0.27, 0.45) | 0.616 | |
| 40–49 | 0.39 | (−0.01, 0.79) | 0.057 | |
| ≥50 | 0.48 | (−0.10, 1.07) | 0.107 | |
| Male (vs. female) | −0.55 | (−0.88, −0.21) | 0.002 | |
| DM | 0.56 | (−0.28, 1.39) | 0.19 | |
| Wk288 cholesterol | 0.01 | (0.002, 0.01) | 0.001 | |
| Wk288 systolic<120 (vs.≥120) | −0.02 | (−0.30, 0.27) | 0.911 | |
| Wk288 HDL | −0.01 | (−0.02, 0.01) | 0.311 | |
| Smoking | 0.22 | (−0.16, 0.60) | 0.260 | |
Discussion
In this study, we evaluated FRS-based vascular outcomes in two large HIV cohorts conducted in Bangkok, Thailand: one initiated ART during AHI and the other initiated ART during CHI. This unique study setting allowed us to examine whether ART initiated during AHI reduced CVD risk as measured by FRS vascular outcomes. Other strengths of this study include a large sample size, a unified duration of ART, the inclusion of age- and sex-matched comparisons, and the inclusion of essential non-HIV variables including ART exposures and co-infections.
Our study confirmed the clear immunological benefits of very early ART initiation in HIV infection, as shown by the higher CD4 T-cell count and CD4/CD8 ratio at 6-year follow-up in the AHI cohort. At 6-year follow-up, the median age of the AHI and CHI cohorts were 32 and 38, with a respective FRS-based 10-year CVD risk of 2.3% and 2.8%, representing a low 10-year CVD risk (<10%). In a recent study that examined the FRS-based 10-year CVD risk of over 300,000 healthy Royal Thai Army personnel aged 30–60 from 2017 to 2021, the mean 10-year CVD risk was 3.1–3.5% among the 30–34 years old personnel and 5.3–5.5% among the 35–39 years old personnel (20). While the higher 10-year CVD risk in the Thai Army study could be partly driven by the higher frequency of smoking (~30% vs. 13% in RV254 and 19% in HIV-NAT 006), our findings highlight that PWH can achieve a comparable level of conventional CVD risk factor control as PWoH.
To examine whether very early ART initiation mitigates CVD risk, we compared FRS-based metrics between AHI and CHI participants at 6-year follow-up. While we identified several differences between cohorts in FRS-based CVD parameters, all differences disappeared when matching cohorts for age and gender. Thus, despite better immunological outcomes, our findings do not support that very early ART offers clear benefit in reducing the occurrence or the control of conventional CVD risk factors. However, it is important to note that while few prior studies have examined the effect of ART initiated as early as AHI (as in this work), a plethora of prior research demonstrates that early ART initiation and ART adherence improve clinical CVD outcomes in PWH (21, 22). To reconcile this with our work, we note that the outcomes in this study are limited to the CVD components utilized in calculation of the FRS, and thus do not take into account immunological or other pathophysiological factors of HIV infection and early ART initiation that may affect clinical vascular outcomes. For example, a recent large-scale analysis from the Veterans Aging Cohort Study (VACS) reported that T-helper type 17 cells, senescent cells, and CD4+ T effector memory cells re-expressing CD45RA were significantly associated with incident CVD in PWH but not PWoH (23). Future studies are needed to evaluate the associations between CD4 T-cell subsets, specific CVD and metabolic risk factors, and cardiovascular outcomes in PWH.
Additionally, while excluding participants on statins did not affect the results shown in this work, the recently published REPRIEVE trial demonstrated that initiation of pitavastatin significantly reduced vascular events in PWH of low and moderate CVD risk, a benefit that could not be explained by LDL reduction alone (24). This work supports the presence of HIV-specific mechanisms contributing to CVD. However, the trial only included PWH above the age of 40, and additional work is needed to identify if even younger PWH (such as in these cohorts) may benefit from statin use.
The secondary goal of this study was to utilize FRS-based VAD to identify early predictors of worse quantifiable vascular health scores in PWH. In multivariable analysis including all participants, higher BMI and greater change in CD4+ T-cell count were independently associated with VAD at the 6-year visit. The association with higher BMI is unsurprising, as obesity is known to increase CVD risk. However, the association between higher VAD and greater increase in CD4 T-cell count from baseline and higher CD4 T-cell count at 6 years is counter-intuitive, as a higher CD4 T-cell count is associated with better immunologic recovery. When stratified by cohort and smoking status, an association between CD4 T-cell count and VAD only persisted in smokers in the AHI cohort, indicating smoking may be a contributing factor to this association, as has been reported previously (19). However, other studies have also reported associations between higher CD4 T-cell counts and higher VA or VAD (25) (11), as well as associations between higher CD4 T-cell counts and metabolic syndrome in CHI cohorts (26–28). Of note, most of these studies were performed in PWH who were older and on a longer duration of ART than in the cohorts studied here. Thus, it remains unclear if the association of higher CD4 T-cell count and VAD is driven by one or multiple underlying specific CVD risk factors.
This work has several limitations. The outcomes presented here are based on young Thai PWH, which may limit generalizability. Additionally, prevalence of vascular outcomes at follow-up was relatively low which reduces statistical power and ART regimens differed somewhat between groups. We also acknowledge the limitations of FRS as a predictor of 10-year CVD in PWH given it is calculated based on a limited number of conventional CVD risk factors as discussed above. However, the use of FRS and therefore FRS-based VAD, allowed us to explore risk factors that were associated with CVD outcomes in the combined cohort independent of the individual cohorts’ slightly different median age. We also acknowledge the limitation of the 2008 FRS score in evaluating participants under the age of 30. Of note, because we assigned participants under the age of 30 to have VA equal to their actual age, we repeated all analyses including only participants over the age of 30 and the significant associations discussed here remained unchanged (Supplementary Table 7). This indicates this was not a significant limitation or bias in our analysis. Finally, in future work we also hope to more thoroughly evaluate the effect of duration of co-infections on vascular outcomes.
Conclusions
In this study, we evaluated FRS-based vascular outcomes at six years in a matched cohort of participants who started ART during AHI compared to a CHI cohort and did not find significant differences. However, additional research is needed to evaluate more nuanced and pathophysiologic markers of vascular disease, such as blood biomarkers, during AHI and after early ART compared to CHI cohorts. We also identified a correlation o higher CD4 T-cell count and greater change in CD4 T-cell count and worse FRS-based vascular outcomes, which warrants further study to confirm these findings and evaluate potential underlying mechanisms.
Supplementary Material
Acknowledgements
We would like to thank the study participants who committed so much of their time for this study. We also would like to thank all research clinicians and staffs from the HIV-NAT and SEARCH010/RV254 study teams for their strong commitment and contributions to the study.
Funding:
The RV254/SEARCH 010 is supported by cooperative agreements (W81XWH-18–2-0040) between the Henry M. Jackson Foundation for the Advancement of Military Medicine, Inc., and the U.S. Department of Defense (DOD) and by an intramural grant from the Thai Red Cross AIDS Research Centre and, in part, by the Division of AIDS, National Institute of Allergy and Infectious Diseases, National Institute of Health (DAIDS, NIAID, NIH) (grant AAI21058–001-01000). Antiretroviral therapy for RV254/SEARCH 010 participants was supported by the Thai Government Pharmaceutical Organization, Gilead Sciences, Merck and ViiV Healthcare. The HIV-NAT 006 cohort is supported by intramural fund of HIV-NAT, Thai Red Cross AIDS research Centre, Thailand and antiretroviral therapy is supported by the National HIV program (NAP), National Health Security Office (NHSO), Social Security Office and Ministry of Public Health (MOPH), Thailand. The authors were additionally supported by grants from the National Institutes of Mental Health (NIMH) (grants R01MH132356 and R01MH130197), and additional funds from NIMH.
List of abbreviations
- ABC
Abacavir
- AHI
Acute HIV
- ART
Antiretroviral therapy
- ASCVD
Atherosclerotic cardiovascular disease
- BMI
Body mass index
- CHI
Chronic HIV
- CVD
Cardiovascular Disease
- DM
Diabetes mellitus
- EIA
Enzyme immunoassay
- FRS
Framingham Risk Score
- HBV
Hepatitis B
- HCV
Hepatitis C
- HDL
High-density lipoprotein
- HTN
Hypertension
- INSTI
integrase strand transfer inhibitor
- NNRTI
non-nucleoside reverse transcriptase inhibitors
- PI
Protease inhibitor
- PWH
People with HIV
- PWoH
People without HIV
- TB
Tuberculosis
- VA
Vascular Age
- VAD
Vascular Age Deviation
Footnotes
Conflict of Interests
K.H has no competing interests.
W.M.H has no competing interests.
T.A has a C2F scholarship from Chulalongkorn University, Thailand
C.S has no competing interests.
R.K has no competing interests.
A.A received research grants from Gilead Sciences, MSD and ViiV healthcare/GSK
S.S has received research grants from the NIH
P.C has no competing interests.
Disclaimer
Material has been reviewed by the Walter Reed Army Institute of Research. There is no objection to its presentation and/or publication. The opinions or assertions contained herein are the private views of the author, and are not to be construed as official, or as reflecting true views of the Department of the Army or the Department of Defense, the National Institutes of Health, or the Henry M. Jackson Foundation for the Advancement of Military Medicine, Inc. The investigators have adhered to the policies for protection of human subjects as prescribed in AR-70–25.
Data Availability Statement
The data that support the findings of this study are available from the corresponding author upon reasonable request.
References
- 1.Shah ASV, Stelzle D, Lee KK, Beck EJ, Alam S, Clifford S, et al. Global Burden of Atherosclerotic Cardiovascular Disease in People Living With HIV: Systematic Review and Meta-Analysis. Circulation. 2018;138(11):1100–12. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Hsue PY, Waters DD. Time to Recognize HIV Infection as a Major Cardiovascular Risk Factor. Circulation. 2018;138(11):1113–5. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Ovbiagele B, Nath A. Increasing incidence of ischemic stroke in patients with HIV infection. Neurology. 2011;76(5):444–50. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.El-Sadr WM, Lundgren J, Neaton JD, Gordin F, Abrams D, Arduino RC, et al. CD4+ count-guided interruption of antiretroviral treatment. N Engl J Med. 2006;355(22):2283–96. [DOI] [PubMed] [Google Scholar]
- 5.Mogadam E, King K, Shriner K, Chu K, Sondergaard A, Young K, et al. The association of nadir CD4-T cell count and endothelial dysfunction in a healthy HIV cohort without major cardiovascular risk factors. SAGE Open Med. 2020;8:2050312120924892. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Ho JE, Scherzer R, Hecht FM, Maka K, Selby V, Martin JN, et al. The association of CD4+ T-cell counts and cardiovascular risk in treated HIV disease. AIDS. 2012;26(9):1115–20. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Lichtenstein KA, Armon C, Buchacz K, Chmiel JS, Buckner K, Tedaldi EM, et al. Low CD4+ T cell count is a risk factor for cardiovascular disease events in the HIV outpatient study. Clin Infect Dis. 2010;51(4):435–47. [DOI] [PubMed] [Google Scholar]
- 8.Manner IW, Trøseid M, Oektedalen O, Baekken M, Os I. Low nadir CD4 cell count predicts sustained hypertension in HIV-infected individuals. J Clin Hypertens (Greenwich). 2013;15(2):101–6. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Lundgren JD, Babiker AG, Gordin F, Emery S, Grund B, Sharma S, et al. Initiation of Antiretroviral Therapy in Early Asymptomatic HIV Infection. N Engl J Med. 2015;373(9):795–807. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.D’Agostino RB, Vasan RS, Pencina MJ, Wolf PA, Cobain M, Massaro JM, et al. General cardiovascular risk profile for use in primary care: the Framingham Heart Study. Circulation. 2008;117(6):743–53. [DOI] [PubMed] [Google Scholar]
- 11.Davies TL, Gompels M, Johnston S, Bovill B, May MT. Mind the gap: difference between Framingham heart age and real age increases with age in HIV-positive individuals-a clinical cohort study. BMJ Open. 2013;3(10):e003245. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Msoka T, Rogath J, Van Guilder G, Kapanda G, Smulders Y, Tutu van Furth M, et al. Comparison of Predicted Cardiovascular Risk Profiles by Different CVD Risk-Scoring Algorithms between HIV-1-Infected and Uninfected Adults: A Cross-Sectional Study in Tanzania. HIV AIDS (Auckl). 2021;13:605–15. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Barzi F, Patel A, Gu D, Sritara P, Lam TH, Rodgers A, et al. Cardiovascular risk prediction tools for populations in Asia. J Epidemiol Community Health. 2007;61(2):115–21. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Ananworanich J, Sacdalan CP, Pinyakorn S, Chomont N, de Souza M, Luekasemsuk T, et al. Virological and immunological characteristics of HIV-infected individuals at the earliest stage of infection. J Virus Erad. 2016;2(1):43–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.De Souza MS, Phanuphak N, Pinyakorn S, Trichavaroj R, Pattanachaiwit S, Chomchey N, et al. Impact of nucleic acid testing relative to antigen/antibody combination immunoassay on the detection of acute HIV infection. AIDS. 2015;29(7):793–800. [DOI] [PubMed] [Google Scholar]
- 16.Han WM, Apornpong T, Su Lwin HM, Thammapiwan S, Bonrungsirisap J, Gatechompol S, et al. Non-alcoholic Fatty Liver Disease and Non-alcoholic Steatohepatitis with liver fibrosis as predictors of new-onset Diabetes Mellitus in People living with HIV: A Longitudinal Cohort Study. Clin Infect Dis. 2023. [DOI] [PubMed] [Google Scholar]
- 17.Tumkosit M, Han WM, Tankittiwat K, Chattranukulchai P, Siwamogsatham S, Apornpong T, et al. Higher epicardial fat in older adults living with HIV with viral suppression and relationship with liver steatosis, coronary calcium and cardiometabolic risks. AIDS. 2022;36(8):1073–81. [DOI] [PubMed] [Google Scholar]
- 18.Goff DC, Lloyd-Jones DM, Bennett G, Coady S, D’Agostino RB, Gibbons R, et al. 2013 ACC/AHA guideline on the assessment of cardiovascular risk: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines. J Am Coll Cardiol. 2014;63(25 Pt B):2935–59. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Royce RA, Winkelstein W. HIV infection, cigarette smoking and CD4+ T-lymphocyte counts: preliminary results from the San Francisco Men’s Health Study. AIDS. 1990;4(4):327–33. [PubMed] [Google Scholar]
- 20.Sakboonyarat B, Poovieng J, Jongcherdchootrakul K, Srisawat P, Hatthachote P, Mungthin M, et al. Rising trends in the predicted 10-year risk of cardiovascular diseases among Royal Thai Army personnel from 2017 to 2021. Sci Rep. 2023;13(1):1476. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Jarwani B Cardiovascular Disease and Antiretroviral Therapy. J Glob Infect Dis. 2019;11(3):91–2. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Feinstein MJ, Hsue PY, Benjamin LA, Bloomfield GS, Currier JS, Freiberg MS, et al. Characteristics, Prevention, and Management of Cardiovascular Disease in People Living With HIV: A Scientific Statement From the American Heart Association. Circulation. 2019;140(2):e98–e124. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Kundu S, Freiberg MS, Tracy RP, So-Armah KA, Koethe JR, Duncan MS, et al. Circulating T Cells and Cardiovascular Risk in People With and Without HIV Infection. J Am Coll Cardiol. 2022;80(17):1633–44. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Grinspoon SK, Fitch KV, Zanni MV, Fichtenbaum CJ, Umbleja T, Aberg JA, et al. Pitavastatin to Prevent Cardiovascular Disease in HIV Infection. N Engl J Med. 2023;389(8):687–99. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Guaraldi G, Zona S, Alexopoulos N, Orlando G, Carli F, Ligabue G, et al. Coronary aging in HIV-infected patients. Clin Infect Dis. 2009;49(11):1756–62. [DOI] [PubMed] [Google Scholar]
- 26.Nguyen KA, Peer N, de Villiers A, Mukasa B, Matsha TE, Mills EJ, et al. Metabolic Syndrome in People Living with Human Immunodeficiency Virus: An Assessment of the Prevalence and the Agreement between Diagnostic Criteria. Int J Endocrinol. 2017;2017:1613657. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Wu PY, Hung CC, Liu WC, Hsieh CY, Sun HY, Lu CL, et al. Metabolic syndrome among HIV-infected Taiwanese patients in the era of highly active antiretroviral therapy: prevalence and associated factors. J Antimicrob Chemother. 2012;67(4):1001–9. [DOI] [PubMed] [Google Scholar]
- 28.Taramasso L, Bonfanti P, Ricci E, Maggi P, Orofino G, Squillace N, et al. Metabolic syndrome and body weight in people living with HIV infection: analysis of differences observed in three different cohort studies over a decade. HIV Med. 2022;23(1):70–9. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
The data that support the findings of this study are available from the corresponding author upon reasonable request.