Background: The effects of HIV serostatus and combination antiretroviral therapy (cART) on plasma homocysteine (HCY) are uncertain.
Methods: Plasma HCY was assayed in a cross-sectional study of 249 HIV-infected and 127 HIV-uninfected women at the Bronx Women's Interagency HIV Study site.
Results: Mean plasma HCY was 7.42 ± 2.68 in HIV-infected women and 7.18 ± 2.66 μmol/L in HIV-uninfected women (P = 0.40). Hyperhomocysteinemia (defined as HCY >10 μmol/L) was seen in 16.9% and 13.4% of HIV-infected and HIV-uninfected women, respectively (P = 0.45). Among HIV-infected women, cART use was not associated with HCY level. Compared with the lowest quartile, women with HCY in the highest quartile had lower mean serum vitamin B12 and RBC folate levels. In multivariate analysis that did not include micronutrient levels, age, serum creatinine, and lower CD4% were significantly associated with plasma HCY level in HIV-infected women.
Conclusions: Plasma HCY was not associated with HIV serostatus or use of cART in this cross-sectional study. Reduced availability of folate cofactors for HCY remethylation in HIV-infected women with lower folate intake and decreased health status may influence HCY levels.
From the *Department of Medicine, St. Luke's-Roosevelt Hospital Center, Columbia University College of Physicians and Surgeons, New York, NY; †Department of Statistics and Biostatistics, Rutgers University, Piscataway, NJ; ‡Data Solutions, LLC, Bronx, NY; §Department of Epidemiology and Biostatistics, School of Public Health, New York Medical College, Valhalla, NY; ‖Department of Medicine, Montefiore Medical Center and Albert Einstein College of Medicine, Bronx, NY; and ¶Division of International Medicine and Infectious Diseases, Department of Medicine, Weill Cornell Medical College, New York, NY.
Received for publication May 20, 2008; accepted December 17, 2008.
The study is cofunded by the National Cancer Institute, the National Institute on Drug Abuse, and the National Institute on Deafness and Other Communication Disorders. This work was also supported in part by the Center for AIDS Research at the Albert Einstein College of Medicine and Montefiore Medical Center funded by the National Institutes of Health (NIH AI-51519) and by the National Institute of Diabetes and Digestive and Kidney Disease (DK54615). This work was also supported by National Institutes of Health, NIH/NHLBI grant R01 HL083760 (to R.C.K.) and NIH/NIAID grant K24 AI 78884.
Informed consent was obtained from all subjects, and human experimentation guidelines of the US Department of Health and Human Services and those of the authors' institutions were followed in the conduct of this research.
The authors have no commercial or other associations that might pose a conflict of interest.
Data in this article were collected by the Bronx/Manhattan Consortium of the Women's Interagency HIV Study (UO1-AI-35004).
Correspondence to: Dr. Marshall J. Glesby, Division of International Medicine and Infectious Diseases, Department of Medicine, Weill Cornell Medical College, 525 East 68th Street, Box 566, New York, NY 10065 (e-mail: email@example.com).
In the era of combination antiretroviral therapy (cART), epidemiological studies have documented higher rates of coronary heart disease or myocardial infarction among HIV-infected patients relative to the general population.1-5 Traditional risk factors for cardiovascular disease (CVD), such as hypercholesterolemia, hypertriglyceridemia, and low plasma high-density lipoprotein cholesterol,6-9 may account for some of this increased risk. The Data Collection on Adverse Events of Anti-HIV Drugs study found that among HIV-infected individuals, even after adjustment for changes in lipids, there was increased risk of myocardial infarction with protease inhibitor (PI) use.10 HIV infection or cART may also change nontraditional or emerging risk factors for coronary heart disease, including inflammatory markers, clotting factors, apolipoproteins, lipoprotein(a), oxidative stress, nonesterified fatty acids, and homocysteine (HCY).11
Some evidence suggests that the prevalence of mild to moderate hyperhomocysteinemia may be more common in HIV-infected individuals, especially those using cART.12 The pathogenesis of hyperhomocysteinemia in HIV infection, if present, has not clearly been identified. A variety of pathophysiologic processes could contribute, including the effects of HIV infection and replication on host metabolism, antiretroviral drugs on HCY metabolism, and other drugs administered to HIV-infected patients; folate deficiency due to lack of dietary intake; and/or deficiency in vitamins B6 or B12 due to insufficient intake or altered metabolism. Because the determinants of HCY level in HIV-infected individuals have not been well described, we conducted a cross-sectional study to determine the associations of plasma HCY levels with HIV serostatus, cART use, and other host and disease factors.
MATERIALS AND METHODS
The Women's Interagency HIV Study (WIHS) is a prospective cohort study with recruitment of primarily minority women with or at risk of HIV infection in 6 urban sites in the United States (Bronx, Brooklyn, DC, Los Angeles, San Francisco, and Chicago). At the Bronx, New York site, 537 women were enrolled in the study beginning in 1993, with an additional 233 women enrolled in 2001. About 39% of the women are Hispanic-Latina, 52% African American, and the rest white or Asian. At the Bronx WIHS site, HIV-infected and HIV-uninfected women in active follow-up were offered participation in this substudy at visit 24 (April to September 2006). The only exclusion criterion was acute illness at the time of the visit as determined by the interviewer at the WIHS site.
For this substudy, blood samples were collected by strict blood collection protocol with immediate processing and storage at −70°C. HCY was assayed by fluorescence polarization immunoassay with an interassay coefficient of variation of 8.8% (Quest Diagnostics, Baltimore, MD). Red blood cell (RBC) folate and serum vitamin B12 levels were assayed by competitive immunoassay using direct chemiluminescent technology in HIV-infected women in the highest and lowest quartiles of HCY values using 5.5 and 8.6 μmol/L with an interassay coefficient of variation of 5% and 4.8%, respectively, (Quest Diagnostics). Fasting lipid panels, insulin, and glucose values obtained from WIHS visits conducted within the 12 months before the substudy visit were used, and the homeostatic model assessment of insulin resistance was calculated as (fasting insulin μU/mL × fasting glucose mg/dL)/405.
Hyperhomocysteinemia was defined as HCY >10 μm/L. Data were descriptively analyzed using plots, histograms, means, medians, standard deviation, skewness, and kurtosis to identify erroneous values, outliers, and optimal transformations. Due to the presence of extreme outliers, 3 values of serum creatinine greater than 2.0 mg/dL were set to missing and fasting insulin and homeostatic model assessment index were log transformed. Unadjusted comparisons were made using t tests and χ2 tests. Adjusted associations between HCY as a continuous variable and other variables of interest were assessed by multivariate linear regression modeling using forward stepwise variable section with a P value of 0.05 to enter and to remain in the model. Multivariate logistic regression was used to model adjusted associations with hyperhomocysteinemia. Initial analyses adjusted for age; subsequent models assessed the effects of adjusting for variables with significant univariate associations with the outcome variables. Because data on vitamin B12 and RBC folate were only available in 50% of the HIV-infected women by design, these covariates were not included in the multivariate models. A P value <0.05 was considered statistically significant. Statistical analysis was performed using SAS version 9.3 (SAS Institute Inc, Cary, NC).
Two hundred forty-nine HIV-infected and 127 HIV-uninfected women provided specimens for this study. HIV-infected women were older (mean ± standard deviation) (age 45 ± 7.3 vs. 42 ± 9.9 years, P < 0.0001) and had a lower body mass index (29.0 ± 6.7 vs. 30.8 ± 8.0, P = 0.03). The mean CD4 cell count of the HIV-infected women at the substudy visit was 488 ± 324 cells per cubic millimeter, and 54% had HIV RNA levels below the level of quantification (<80 copies/mL). Fifty-five percent of the HIV-infected women had a clinical diagnosis of AIDS at or before the substudy visit.
Mean plasma HCY was 7.42 ± 2.68 and 7.18 ± 2.66 μmol/L in HIV-infected and HIV-uninfected women, respectively (P = 0.4). Hyperhomocysteinemia was present in 16.9% and 13.4% of HIV-infected and HIV-uninfected women, respectively (P = 0.45). Among HIV-infected women, receiving cART was not associated with increased HCY levels (mean 7.32 ± 2.39 vs. 7.46 ± 2.78 μmol/L in women currently taking cART vs. not taking cART, P = 0.70). HCY levels were also not associated with current or ever use of either nucleoside reverse transcriptase inhibitors or nonnucleoside reverse transcriptase inhibitors or with either current use of PIs or trimethoprim-sulfamethoxasole (TMP-SMX) (data not shown). HCY levels were, however, higher in women who had ever used the following: a dideoxynucleoside analog (didanosine, stavudine, or zalcitabine) (mean 7.66 ± 2.81 vs. 6.97 ± 2.37 in ever vs. never users; P = 0.04), TMP-SMX (7.76 ± 2.93 vs. 6.88 ± 2.14; P = 0.007), or a PI (7.60 ± 2.71 μmol/L vs. 6.92 ± 2.56 μmol/L; P = 0.07). But none of these associations remained statistically significant after adjustment for age in bivariate linear regression models.
In linear regression analysis, significant univariate predictors of HCY level included age, serum creatinine, CD4 percentage, and ever having taken TMP-SMX (Table 1). After adjustment for age, TMP-SMX use was no longer a significant predictor. In a multivariate model, age, serum creatinine, and CD4% each remained significantly associated with plasma HCY level (Table 1). Multivariate logistic regression models found that age [odds ratio (OR) = 2.66 per 10 years, 95% confidence interval (CI): 1.53 to 4.64], serum creatinine (OR = 1.45 per mg/dL, 95% CI: 0.98 to 2.16), CD4 percentage (OR = 0.83 per 5% increase, 95% CI: 0.70 to 0.98), and ever having taken TMP-SMX (OR = 2.67, 95% CI: 1.07 to 6.67) were independent predictors of hyperhomocysteinemia (HCY >10 umol/L).
Serum vitamin B12 and RBC folate levels were measured in HIV-infected women in the highest and lowest quartiles of the HCY distribution (HCY <5.5 μmol/L and >8.6 μmol/L). Serum levels of both vitamin B12 and RBC folate were each lower in women in the highest vs. the lowest HCY quartile (487 ± 215 pg/mL vs. 617 ± 364 pg/mL; P = 0.02 and 353 ± 139 nmol/L vs. 432 ± 114 nmol/L; P < 0.001, respectively). In logistic regression models, these associations remained significant after adjustment for age, CD4, count, and creatinine level (data not shown).
In this cross-sectional study, we did not find an association of either HIV serostatus or cART with HCY concentrations in women. Although higher HCY levels were associated with ever using a dideoxynucleoside analog or TMP-SMX, the associations did not remain significant after adjustment for age. In multivariate analysis, age, serum creatinine, and lower CD4 percentage were each independently associated with higher HCY levels. We also observed strong inverse associations of both RBC folate and vitamin B12 with HCY levels.
Our results help clarify the previously published conflicting data on the potential relationships between plasma HCY concentrations and both HIV infection and cART. In an evaluation of 73 participants from the Swiss HIV Cohort Study, a significantly higher mean total plasma HCY concentration was observed in HIV-infected men receiving PI-based cART compared with healthy controls (9.1 vs. 7.8 μmol/L, P = 0.02).12 The subgroup of 26 HIV-infected women on PI regimens, however, did not have statistically different HCY levels compared with healthy controls (8.25 vs. 7.4 μmol/L, P = 0.19).12 The HIV-infected subjects were older than the controls in this study, and the analysis was not adjusted for age, which we have shown to be an important potential confounder. Furthermore, the potential contribution of cART vs. HIV disease severity could not be assessed given the design. In contrast, our results are similar to those of Bongiovanni et al,13 who measured plasma HCY levels in 83 treatment-naive HIV-infected subjects, 161 on cART, and 54 healthy controls. Although mean HCY levels were higher in the subjects on cART compared with the other 2 groups in an unadjusted analysis, only older age and low serum folate concentration (and not patient group) were associated with hyperhomocysteinemia in a multivariate analysis. In another cross-sectional study of 98 patients who were on stable cART regimens for at least 6 months, duration of cART and PI exposure were not significantly associated with elevated HCY level.14 Taken together, these data suggest that hyperhomocysteinemia is not a direct effect of HIV or cART.
Elevated plasma HCY is known to occur in association with nutritional deficiencies of vitamin B12 and folate in the general population.15,16 Two prior studies found that serum folate concentrations were independently associated with hyperhomocysteinemia in HIV-infected patients.13,14 Folate or vitamin B12 insufficiency due to poor intake or malabsorption could be a function of or marker of severity of HIV disease. In our study, both RBC folate-a more robust measure of folate stores than serum levels-and vitamin B12 levels remained lower in women with HCY in the highest vs. lowest quartile even after adjustment for CD4 cell count, suggesting that HIV disease severity is not the primary underlying cause of the micronutrient deficiencies.
The observed associations in our study of HCY level with RBC folate and vitamin B12, if causal, suggest a potential to lower HCY levels by inexpensive and feasible methods such as folate supplementation and therefore to decrease CVD risk in HIV-infected populations. However, lack of association between HCY level and HIV infection or cART in HIV-infected subjects makes hyperhomocysteinemia an unlikely explanation for increased risk of CVD in the HIV-infected population. In addition, 2 recent randomized clinical trials of HCY lowering that have shown vitamin supplementation can effectively lower HCY levels have nevertheless failed to identify a beneficial effect on CVD outcomes in general populations.17,18
Plasma HCY is known to be elevated in chronic renal insufficiency.19 In our analyses, we found a positive relationship between plasma HCY and serum creatinine in HIV-infected women with creatinine values in the 0.2-1.5 mg/dL range in multivariate analysis. These findings are consistent with data from the general population and diabetics showing a strong inverse relationship between plasma HCY and glomerular filtration rate, even within the normal range of the latter parameter.20
This study is limited by its cross sectional design and therefore no inferences on temporal relationship or causation between different variables can be made. Another potential limitation is the lack of dietary intake data on micronutrients of interest such as on folate, vitamin B6, and vitamin B12, though we found no association between multivitamin use and plasma HCY. Although serum levels of these nutrients have been shown to be reliable surrogates of dietary intake in prior studies of general populations,15 data in this regard are limited in HIV-infected populations. It is possible that unmeasured metabolic parameters associated with HIV infection or its treatment may also contribute to low folate or B12 levels. We were also not able to determine if HCY level was associated with future cardiovascular events as too few events have occurred during subsequent follow-up of the cohort.
In summary, this cross-sectional study of women with and at risk of HIV infection found no association between HIV serostatus or cART with HCY levels. In HIV-infected women, RBC folate, vitamin B12, and CD4% were significantly associated with HCY level, which may in part be explained by reduced availability of folate cofactors for HCY remethylation in women with lower folate intake secondary to the decreased health status that occurs with lower CD4 counts.
The authors thank the study participants and staff of the Bronx WIHS site, especially Esther Robison, PhD, and William Meyer, PhD, and Larry Hirsch of Quest Diagnostics for their assistance.
1. Klein D, Hurley LB, Quesenberry CP Jr, et al. Do protease inhibitors increase the risk for coronary heart disease in patients with HIV-1 infection? J Acquir Immune Defic Syndr
2. Currier JS, Taylor A, Boyd F, et al. Coronary heart disease in HIV-infected individuals. J Acquir Immune Defic Syndr
3. Mary-Krause M, Cotte L, Simon A, et al. Increased risk of myocardial infarction with duration of protease inhibitor therapy in HIV-infected men. AIDS
4. Obel N, Thomsen HF, Kronborg G, et al. Ischemic heart disease in HIV-infected and HIV-uninfected individuals: a population-based cohort study. Clin Infect Dis
5. Triant VA, Lee H, Hadigan C, et al. Increased acute myocardial infarction rates and cardiovascular risk factors among patients with human immunodeficiency virus disease. J Clin Endocrinol Metab
6. Shor-Posner G, Baum MK. Nutritional alterations in HIV-1 seropositive and seronegative drug users. Nutrition
7. Constans J, Pellegrin JL, Peuchant E, et al. Plasma lipids in HIV-infected patients: a prospective study in 95 patients. Eur J Clin Invest
8. Grunfeld C, Kotler DP, Hamadeh R, et al. Hypertriglyceridemia in the acquired immunodeficiency syndrome. Am J Med
9. Grunfeld C, Pang M, Doerrler W, et al. Lipids, lipoproteins, triglyceride clearance, and cytokines in human immunodeficiency virus infection and the acquired immunodeficiency syndrome. J Clin Endocrinol Metab
10. Friis-Moller N, Reiss P, Sabin CA, et al. Class of antiretroviral drugs and the risk of myocardial infarction. N Engl J Med
11. Crook M. The basis and management of metabolic abnormalities associated with cardiovascular risk in human immunodeficiency virus infection and its treatment. Ann Clin Biochem
12. Bernasconi E, Uhr M, Magenta L, et al. Homocysteinaemia in HIV-infected patients treated with highly active antiretroviral therapy. AIDS
13. Bongiovanni M, Casana M, Pisacreta M, et al. Predictive factors of hyperhomocysteinemia in HIV-positive patients. J Acquir Immune Defic Syndr
14. Uccelli MC, Torti C, Lapadula G, et al. Influence of folate serum concentration on plasma homocysteine levels in HIV-positive patients exposed to protease inhibitors undergoing HAART. Ann Nutr Metab
15. Fung TT, Rimm EB, Spiegelman D, et al. Association between dietary patterns and plasma biomarkers of obesity and cardiovascular disease risk. Am J Clin Nutr
16. Verhoef P, Stampfer MJ, Buring JE, et al. Homocysteine metabolism and risk of myocardial infarction: relation with vitamins B6, B12, and folate. Am J Epidemiol
17. Bonaa KH, Njolstad I, Ueland PM, et al. Homocysteine lowering and cardiovascular events after acute myocardial infarction. N Engl J Med
18. Lonn E, Yusuf S, Arnold MJ, et al. Homocysteine lowering with folic acid and B vitamins in vascular disease. N Engl J Med
19. Nerbass FB, Draibe SA, Feiten SF, et al. Homocysteine and its determinants in nondialyzed chronic kidney disease patients. J Am Diet Assoc
20. Friedman AN, Hunsicker LG, Selhub J, et al. Total plasma homocysteine and arteriosclerotic outcomes in type 2 diabetes with nephropathy. J Am Soc Nephrol