Background: Low serum micronutrient levels were common before widespread use of highly active antiretroviral therapy (HAART) and were associated with adverse outcomes. Few data are available on micronutrient levels in subjects taking HAART.
Objective: To determine the prevalence of low serum retinol, α-tocopherol, zinc, and selenium in HIV-infected subjects taking HAART and to assess the association of micronutrient levels with HIV disease status.
Setting: Nutrition for Healthy Living (NFHL) study.
Participants: HIV-infected subjects on HAART.
Methods: Retinol, α-tocopherol, zinc, and selenium were determined in frozen serum samples from 171 men and 117 women. Low serum levels were defined as retinol <30 μg/dL, selenium <85 μg/L, α-tocopherol <500 μg/dL, and zinc <670 μg/L. Association of micronutrient quartiles with CD4 cell count, CD4 count <200 cells/mm3, HIV viral load (VL), and undetectable VL was assessed using adjusted multivariate regression.
Results: Five percent of men and 14% of women had low retinol, 8% of men and 3% of women had low selenium, and 7% of men and no women had low α-tocopherol. Forty percent of men and 36% of women had low zinc, however. Subjects in the upper quartiles of zinc had lower log VL levels than those in the lowest quartile (significant for women). Subjects in the upper quartiles of selenium also tended to have lower VL levels compared with those in the lowest quartile. Surprisingly, women in the upper quartiles of retinol had higher log VLs than those in the lowest quartile. There was no significant association of any micronutrient with CD4 cell count or likelihood of CD4 count <200 cells/mm3. The level of CD4 cell count influenced the association of retinol with log VL in men, however. In men with CD4 counts >350 cells/mm3, those with higher retinol had higher log VLs compared with the lowest quartile, whereas in men with CD4 counts <350, those with higher retinol levels had lower log VLs compared with the lowest quartile.
Conclusions: Low retinol, α-tocopherol, and selenium are uncommon in HIV-infected subjects on HAART. Zinc deficiency remains common, however. Decreased retinol levels in women and in men with CD4 counts >350 cells/mm3 and increased zinc and selenium levels in both genders may be associated with improved virologic control.
From the *Department of Public Health and Family Medicine, Tufts University School of Medicine, Boston, MA; †Department of Epidemiology and Department of Biostatistics, Harvard University School of Public Health, Boston, MA; and ‡Department of Ophthalmology, Johns Hopkins University School of Medicine, Baltimore, MD.
Received for publication December 29, 2005; accepted July 27, 2006.
Supported by National Institutes of Health (NIH) grant 1P01-DK-457-01A2, NIH supplemental grant P01 DK045734-S1, NIH grant P01 DK045734-S2, National Institute on Drug Abuse (NIDA) grant DA11598, NIDA grant DA14501, NIDA grant 1P30DA13868, General Clinical Research Center, Division of Research Resources, and NIH grant M01-RR00054.
Reprints: Clara Jones, MD, Tufts University School of Medicine, 200 Harrison Ave., Posner 4, Boston, MA 02111 (e-mail: firstname.lastname@example.org).
Low levels of serum micronutrients have been reported since early in the HIV epidemic. Many studies linked low levels with worsened HIV disease status and mortality (eg, low retinol,1-3 low α-tocopherol,4 low selenium,5-7 low zinc8). Before widespread use of highly active antiretroviral therapy (HAART), micronutrient deficiencies were often associated with poor dietary intake, increased physiologic stress from uncontrolled HIV, and associated infections. Although some studies suggested that micronutrient supplementation was beneficial (β-carotene plus retinol,9 selenium,10 and mixed multivitamins11,12), other studies found worse outcomes with supplementation (β-carotene plus retinol,13,14 selenium,15 and zinc16,17) or no benefit (retinol18 and β-carotene19). Most subjects in these studies were not taking HAART. There is little information about micronutrient status in HIV-infected persons on HAART in the medical literature. We address the following questions in HIV-infected adults on HAART in the Nutrition for Healthy Living (NFHL) study: (1) the prevalence of low serum levels of retinol, α-tocopherol, zinc, and selenium; (2) whether low levels of these micronutrients are associated with worse HIV disease status; and (3) whether supplementation is associated with better HIV disease status.
Study Population and Data Collection
Nutrition for Healthy Living
The NFHL study is a longitudinal cohort study of nutrition and HIV disease status in HIV-infected adults (18 years of age or older), with study centers in Boston, Massachusetts and Providence, Rhode Island. Study methods have been published previously.20 The study began in 1995 and enrolled 881 participants. At baseline and 6-month intervals thereafter, questionnaires, laboratory studies, bioelectrical impedance analysis (BIA), and anthropometry measurements were obtained. Fasting insulin, fasting glucose, waist and hip circumferences, and blood for micronutrient analyses were obtained beginning in September 2000. The protocol was approved by the Institutional Review Boards at the Tufts-New England Medical Center (NEMC) in Boston and at Miriam Hospital in Providence. Written informed consent was obtained from each participant.
NFHL participants were selected for this study using the following criteria: the participant was taking HAART, was seen in the clinic between September 2000 and November 2003, and had frozen serum for micronutrient analysis. Analyses for men and women were separately funded; an additional requirement for men was a C-reactive protein (CRP) determination available at the same visit. Of 323 male and 138 female participants on HAART seen between September 2000 and November 2003, 171 male and 117 female participants met these criteria. There were no significant differences in race, intravenous drug use (IDU), CD4 cell count, log10 HIV viral load (log VL), or alanine aminotransferase (ALT) levels between included versus nonincluded subjects. Micronutrient analyses for women were conducted at the Johns Hopkins School of Medicine (laboratory of Dr. Richard Semba) between 2001 and 2004; for men, they were conducted in the General Clinical Research Center (GCRC) Core Laboratory at Tufts-NEMC in 2004. Data for men and women are reported separately.
Blood was drawn after ≥5 hours of fasting, and serum was stored at −70°C. Micronutrients are stable at −70°C through several freeze-thaw cycles.21-23
Retinol and α-Tocopherol
For women, retinol and α-tocopherol were measured by high-performance liquid chromatography (HPLC) using a modified method from the Nutrition Laboratory, Inorganic Toxicology and Nutrition Branch, Division of Laboratory Sciences, National Center of Environmental Health, Centers for Disease Control and Prevention (Rosemary Schleicher, PhD, personal communication, 2002).24 For men, retinol and α-tocopherol were quantified using reverse-phase HPLC (ESA EZChrom Elite Chromatography Data System, Model 542 Autosampler, Model 582 Pump, Model 540 MultiWave detector; ESA, Chelmsford, MA) according to Bieri et al.25 Retinol was considered low if levels were <30 μg/dL (<1.0473 μmol/L); this cutoff has been used previously in the HIV literature2 and has been associated with decreased survival. α-Tocopherol was considered low if levels were <500 μg/dL (<11.61 μmol/L). This cutoff has been used in other publications on micronutrients in HIV4 and in the general population.26
Zinc was measured by graphite furnace atomic absorption spectrometry (AAnalyst 600 with Zeeman background correction; Perkin-Elmer, Shelton, CT) in women and by flame atomic absorption spectrometry (AAnalyst 800 Spectrometer with Zeeman-effect background correction; PerkinElmer) in men. Published studies of zinc levels in HIV-infected subjects have used different cutoff levels (ranging from <550 to <750 μg/L), with widely divergent prevalences of deficiency.27 For this study, deficient zinc was defined as a serum level <670 μg/L (<10.25 μmol/L), using the cutoff referenced by Bender and Bender28 for normal plasma zinc (670-1830 μg/L).
Selenium was measured by graphite furnace atomic absorption spectrometry using a Perkin Elmer AAnalyst 600 with Zeeman background correction (women) or a Perkin-Elmer AAnalyst 800 (men). Deficient selenium was defined as a level <85 μg/L (<1.0795 μmol/L), which is a cutoff associated with increased mortality.5,29,30 Because low levels of retinol, α-tocopherol, and selenium were uncommon, we used quartile analysis to assess the association of micronutrients with HIV disease status.
Dietary micronutrient intake was ascertained using a 3-day food record (FR) (85% of men and 62% of women) or 1-day or 2-day dietary recall. Methods for assessing dietary intake have been published elsewhere.31,32 Nutrient calculations were performed using the Nutrition Data System (NDS) software (version 2.92; Nutrition Coordinating Center, University of Minnesota, Minneapolis, MN).33 If the total reported intake of a micronutrient minus intake from food alone was ≥30% of the recommended dietary allowance (RDA), the participant was considered a supplement user for that micronutrient.
HIV Disease Status
We evaluated CD4 count (cells/mm3), CD4 cell count <200 cells/mm3 (yes/no), log VL, and undetectable VL (yes/no). CD4 cell counts were determined using a specific monoclonal antibody and fluorescence-activated cell sorter (FACS) analysis. HIV RNA was measured with the Roche Amplicor Monitor reverse transcriptase polymerase chain reaction (PCR) assay (Roche Molecular Systems, Somerville, NJ), with a lower detection limit of 400 copies/mL. We assigned an undetectable VL a value of 200 copies/mL for statistical analysis.
Age, gender, race (black, Hispanic, white, or other), housing insecurity, and poverty (defined as personal income <$10,000 per year or household income <150% of the Federal poverty level) were considered.
Years of known HIV infection was considered. All subjects were current HAART users (defined as 2 protease inhibitors [PIs], 2 nucleoside reverse transcriptase inhibitors [NRTIs] with 1 PI, 2 NRTIs with 1 nonnucleoside reverse transcriptase inhibitor [NNRTI], a combination of a PI and an NNRTI with an NRTI, or at least 3 NRTIs).
Body mass index (BMI) was calculated: [weight in kilograms]/[height in square meters].34 Lipodystrophy was self-reported as a marked or moderate increase in abdominal girth, breast size, or buffalo hump or a marked or moderate decrease in facial fat or limb fat during the prior 3 years.
High-Sensitivity C-Reactive Protein
CRP was measured by latex immunoassay (Wako Diagnostics, Richmond, VA) on a Roche Diagnostics (Indianapolis, IN) Hitachi 911 chemistry analyzer. CRP values were available for 168 male and 59 female subjects. A CRP level >10 mg/L suggests an acute-phase response.
Hepatitis B surface antigen (HBsAg) was detected by enzyme immunoassay (EIA) using the Genetic Systems HbsAg EIA 3.0 from Bio-Rad Laboratories (Bio-Rad Laboratories, Redmond, WA).
Antibody to hepatitis B surface antigen (HBsAb) was detected by EIA using the ETI-AB-AUK Plus (DiaSorin, Inc., Stillwater, MN), and antibody to hepatitis B core antigen (HBcAb) was detected by EIA using the EIT-AB-COREK-Plus (DiaSorin, Inc.). Antibody to hepatitis C virus (HCV Ab) was detected by enzyme-linked immunosorbent assay (ELISA) using Ortho HCV, version 3.0 (Ortho-Clinical Diagnostics, Raritan, NJ). Hepatitis C virus (HCV) RNA was detected using the Amplicor HCV Test, version 2.0, from Roche (Roche Diagnostics). A dichotomous variable for liver disease was derived: if HBsAg or HCV RNA was positive and/or ALT was >80, the liver disease score was “yes”; if not, the liver disease score was “no.”
Means and proportions for demographic, laboratory, and other variables were calculated and compared between micronutrient quartiles using the Student t test, Wilcoxon nonparametric test, or χ2 test. Continuous HIV disease status indicators (CD4 cell count and log VL) were analyzed using multivariate linear regression, and dichotomous HIV disease status indicators (CD4 count <200 cells/mm3 and undetectable VL) were analyzed using a log link function for relative risk (RR). Age, race, poverty, years HIV-positive, history of IDU ever, BMI, and liver score were included in all multivariate models for continuous dependent covariates; in addition, CD4 cell count models were adjusted for log VL, and log VL models were adjusted for CD4 cell counts. The models for dichotomous HIV disease status indicators did not allow adjustment for as many covariates because of problems with convergence but were all adjusted for liver score, age, and poverty. We evaluated effect modification by the following covariates: liver score, history of IDU ever, age >43 years, CD4 count >350 cells/mm3 (for log VL models), and detectable VL (for CD4 cell count models) using a chunk test with the likelihood ratio test. Statistical tests were 2-sided; P values were considered significant if <0.05. All analyses used SAS version 9.0 (SAS Institute, Cary, NC).
Table 1 shows demographics of the 171 men and 117 women. Thirty-six percent of men and 69% of women were poor. Although the mean duration of known HIV infection was >9 years, mean CD4 cell counts were well maintained, and most participants had undetectable VLs. Although most participants denied current IDU, 41% of women and 27% of men had a prior history of ever using injection drugs.
The CRP level was available in 168 men and 59 women. The median CRP level was 1.6 mg/L in men and 1.3 mg/L in women. The mean CRP level was not significantly different across quartiles for any micronutrient. Although 26.8% of men and 35.5% of women had a CRP level >3, only 4% of men and 12.5% of women had a CRP level >10. In subjects with a CRP level >10, the CRP level ranged from 11.3 to 49.8 mg/L (men) and from 10.4 to 16.8 mg/L (women). The mean serum micronutrient levels were not significantly different in subjects with a CRP level >10 compared with subjects with a CRP level <10 (data not shown). Thirty-five percent of men and 27% of women had evidence of liver disease (active viral hepatitis or elevated ALT).
Prevalence of Low Serum Micronutrients
Table 1 shows micronutrient levels by gender. With the exceptions of zinc (for which 40% of men and 36% of women were deficient) and retinol (for which 14% of women were deficient), there was a low prevalence of micronutrient deficiency.
Tables 2A and B show the median and interquartile range (IQR) within each micronutrient quartile by gender. The median retinol level in the lowest quartile was within the nondeficient range for men but was in the deficient range for women. Median zinc levels were in the deficient range in quartile 1 for both genders and in quartile 2 for men. Although quartile 4 generally had the highest prevalence of supplement users for each micronutrient, the difference between quartiles was significant only for α-tocopherol in both genders and was borderline (P = 0.07) for selenium in men (data not shown).
Association of Serum Micronutrient Levels With HIV Disease Status
Tables 2A and B show multivariate associations of micronutrient quartiles with HIV disease status. No micronutrient was associated with CD4 cell count-related disease status (total CD4 count or percentage with CD4 count <200 cells/mm3) in multivariate models.
In men, only zinc showed a trend for difference in disease status by quartiles. Log VL was lower among men in the upper 3 zinc quartiles compared with men in the lowest quartile; this difference approached significance for zinc in quartile 3. In women, log VL was higher among those in the upper 3 retinol quartiles compared with those in the lowest quartile (suggesting that higher serum retinol levels were associated with worse virologic control); this was significant for retinol quartiles 2 and 4. Log VL was lower among women in selenium quartiles 2, 3, and 4 compared with those in the lowest quartile (significant for selenium quartile 3), and log VL was significantly lower among women in zinc quartiles 2 through 4 compared with quartile 1. α-Tocopherol quartiles were not significantly associated with log VL in women.
In men, there were no significant differences in the risk of an undetectable VL in quartiles 2, 3, and 4 compared with quartile 1 for any micronutrient. In women, there was no significant difference in the risk of an undetectable VL for the α-tocopherol or zinc quartiles compared with quartile 1. Women in selenium quartiles 2 through 4 had a higher likelihood of undetectable VL compared with those in quartile 1; this was significant for quartile 3. Women in retinol quartiles 2, 3, and 4 were less likely to have an undetectable VL compared with those in quartile 1; this was significant for quartiles 2 and 4.
Association of Serum Levels of Micronutrient With Reported Dietary and Supplement Intake
We looked at the association of total micronutrient intake (dietary and supplemental) with serum levels. Total intake of α-tocopherol was significantly associated with serum α-tocopherol (P < 0.001) for both genders, and total selenium intake was associated with serum selenium in women (P = 0.03) and men (P = 0.0003). No association between total intake and serum level was observed for retinol or zinc (as expected, because levels are homeostatically controlled). The prevalence of retinol or zinc supplementation did not differ significantly by quartile in either gender.
Association of Supplement Use With HIV Disease Status
Supplement use was not significantly associated with HIV disease status for any micronutrient.
Association of Zinc Intake (Dietary Plus Supplemental) With HIV Disease Status
Because of epidemiologic data associating dietary intake of zinc with worsened outcomes,13,14,16,17 we assessed whether total intake of zinc was associated with HIV disease status. In men, increased total zinc intake was significantly negatively associated with log VL in multivariate models adjusted for race, poverty, years HIV-positive, liver score, and CD4 cell count, but the effect was quite small (−0.007; P = 0.02). There was no significant association of zinc intake with HIV disease status in women.
We tested for effect modification by liver disease, history of IDU ever, age >43 years versus <43 years (the approximate median age), and CD4 cell count >350 cells/mm3 versus <350 cells/mm3 (for the analysis in which log VL was the outcome).
There seemed to be a difference in the relation of retinol with log VL in men with CD4 cell counts >350 cells/mm3 versus <350 cells/mm3. In men with CD4 counts >350 cells/mm3, increased retinol was associated with increased log VL compared with quartile 1; the difference was most marked for quartile 2. In men with CD4 counts ≤350 cells/mm3, increased retinol was associated with lower log VL compared with quartile 1; the difference was least marked for quartile 3. Women with CD4 counts >350 cells/mm3 had higher log VL in quartiles 2 and 4 (compared with quartile 1). Women with CD4 counts ≤350 cells/mm3 had higher log VL in quartiles 3 and 4 (compared with quartile 1) but lower log VL in quartile 2. It is unclear whether these results represent true biologic differences or chance findings.
Our data show that low serum retinol in men and low α-tocopherol and low selenium in both genders are less common now compared with published reports from the pre-HAART era. Prior studies reported a 10% to 39% prevalence of low retinol,2,4,26,35-38 12% to 27% prevalence of low α-tocopherol levels,4 and 77% prevalence of low selenium levels35,36 in HIV-infected persons not taking HAART.39 Zinc deficiency remains quite common.
We found no associations between serum micronutrient levels and CD4 cell counts. We did find that subjects in the upper quartiles of serum zinc tended to have lower log VL levels than those in the lowest quartile. Furthermore, women in the upper quartiles of selenium also tended to have lower log VL levels compared with those in the lowest quartiles. In contrast, women in the upper quartiles of serum retinol had higher log VL levels than those in the lowest quartile. In women, the higher VL with retinol quartiles was seen in those with CD4 counts >350 cells/mm3 (except quartile 3) and in those with CD4 counts ≤350 cells/mm3 (except quartile 2). In men, those with CD4 counts >350 cells/mm3 had a higher VL with increased retinol quartiles, whereas those with CD4 counts ≤350 cells/mm3 had a lower VL with increased retinol quartile.
There are few published data on micronutrient levels in HIV-infected persons taking HAART. Tang et al40 reported on 175 HIV-infected drug users. Although mean retinol and α-tocopherol levels were lower in drug users compared with National Health and Nutrition Examination Survey III (NHANES III) averages, α-tocopherol was significantly higher in those on PI combination therapy in comparison with those on no antiretroviral therapy, monotherapy, or combination therapy without a PI. Retinol was not different by treatment category. Rousseau et al39 assessed micronutrient levels in 44 HIV-infected persons in 1995 (pre-HAART therapy) and in 30 of the same persons in 1998 when 77% were on HAART. In 1995, 77% were selenium deficient (<60 μg/L) and 23% were zinc deficient. In 1998 (after HAART), only 10% were selenium deficient but 27% were zinc deficient. Retinol and α-tocopherol were only measured in 1998; 80% of these HIV-infected persons were vitamin A deficient (<1.5 μmol/L or 42.8 μg/dL), whereas 34% were vitamin E deficient (<600 μg/dL). Seventy-seven percent were intravenous drug users.
The high prevalence of low serum zinc in our study is consistent with that in pre-HAART studies. Baum et al29 found that 56% of HIV-positive drug users had zinc deficiency (<750 μg/L). In our cohort, an acute-phase response41 is an unlikely cause of low serum zinc. Our participants had few active infections, and serum zinc levels were not associated with CRP. Low zinc has been associated with chronic diarrhea, and 16% of men and 12% of women reported current diarrhea (≥30 days) at the time of their visit. There was no significant difference in the percentage with diarrhea by zinc quartile, however. In our cohort, 25% of women and 34% of men took supplemental zinc, but zinc intake was not associated with serum zinc levels. The association of lower log VL with higher serum zinc in our study could reflect improvement of immune status and improved virologic control because of higher zinc. Alternatively, it could simply reflect reduced viral sequestration of zinc (because zinc is an integral part of the HIV RNA polymerase) in those with better virologic control. We found no adverse association of zinc intake with HIV disease status, in contrast to the findings of Tang et al.16
Selenium deficiency was uncommon. There seemed to be a lower log VL and a higher likelihood of an undetectable VL in the upper 3 selenium quartiles compared with the lowest quartile for both genders, but this was not significant for most quartiles. There was no association of serum selenium levels and CD4 cell counts, in agreement with the findings of Rousseau et al39 in their post-HAART data. Kupka et al7 found no association of selenium and CD4 cell count at baseline in their longitudinal observational study but did find a weak association in the first few years of follow-up. Batterham et al42 found no effect of antioxidant supplementation (including selenium) with CD4 cell count or VL. A similar rationale for low selenium in HIV can be postulated as with serum zinc (ie, that higher viral replication results in lower selenium levels because of viral utilization and sequestration of selenium in HIV selenoproteins43). Improved virologic control with HAART could thus result in improved selenium levels.
We were surprised that higher serum retinol levels in women were associated with higher log VL and lower likelihood of an undetectable VL compared with the lowest quartile. Our study also suggests that the level of CD4 cell count influences the association of retinol with log VL in men. Other studies have raised concern that supplemental vitamin A (in HIV-infected Tanzanian women14) or β-carotene (in cardiovascular disease trials)44 might have adverse effects. In addition, Tang et al16 found a U-shaped relation between vitamin A intake and progression to AIDS in the Multicenter AIDS Cohort Study (MACS). Semba et al18 and Humphrey et al45 found no effect of vitamin A supplementation on CD4 count or VL in placebo-controlled randomized trials of injection drug users and HIV-infected women, respectively, however. Jiamton et al11 found no effect of commercial multivitamin supplementation (including 3000 μg of vitamin A and 6 mg of β-carotene) on CD4 count or VL in a placebo-controlled trial in Thailand, and Batterham et al42 found no difference in log VL with multiple antioxidant supplementation (including vitamin A). Rousseau et al39 did not find any difference in vitamin A levels in subjects with a VL <5000 or >5000 copies/mL. Of note, a study in Kenyan men receiving treatment for genital ulcers found a higher risk of HIV seroconversion in men with higher retinol (>20 μg/dL) compared with those with lower retinol,46 suggesting an adverse effect of higher retinol. Other studies are required to assess the significance of these findings.
α-Tocopherol deficiency was uncommon in our study, and serum α-tocopherol was not associated with any HIV disease status. In a longitudinal study of men in the MACS (before widespread use of HAART), Tang et al4 found no association of serum vitamin E with CD4 count decline <200 cells/mm3.
There are several limitations to this study. Most importantly, this is a cross-sectional analysis, and a finding of association does not necessarily imply causation. There are currently few data in the literature addressing micronutrient levels in a HAART-using population, however. Cross-sectional studies can be quite helpful in identifying associations for additional study. Second, we cannot ascribe the lower prevalence of low serum micronutrients directly to HAART because we have no comparison group of HIV-infected persons not on HAART. Most of our participants entered the study on HAART. Comparisons between those on HAART and not on HAART could be confounded by a multitude of other factors (eg, those not on HAART are somewhat different demographically and are more likely to have other conditions, such as a history of IDU and liver disease, that could have an impact on micronutrient levels) because this is an observational study and HAART use was not randomly assigned.
We also recognize that serum levels may not be the best measure of total body stores for micronutrients, especially when serum levels are tightly regulated across a range of dietary intakes.18,47 Factors governing the set point of serum levels are incompletely known. We adjusted for active liver disease, making confounding by liver disease less likely. In addition, for those persons for whom CRP levels were available, CRP was not associated with serum micronutrient levels, and inclusion of CRP levels in final models did not change interpretation of the association of micronutrients with outcomes.
Most of our participants had adequate serum levels of retinol, α-tocopherol, and selenium. Low serum zinc was common. Decreased serum retinol levels in women and in men with CD4 counts >350 cells/mm3 as well as increased serum zinc levels in both genders were associated with improved virologic control. Although micronutrient supplement use was relatively common, it was not significantly associated with improved HIV disease status. There was no evidence that a higher dietary intake of zinc was harmful.
1. Semba R, Caiaffa W, Graham N, et al. Vitamin A deficiency and wasting as predictors of mortality in human immunodeficiency virus-infected injection drug users. J Infect Dis
2. Semba R, Graham N, Caiaffa W, et al. Increased mortality associated with vitamin A deficiency during human immunodeficiency virus type 1 infection. Arch Intern Med
3. Semba R, Miotti P, Chiphangwi J, et al. Infant mortality and maternal vitamin A deficiency during human immunodeficiency virus infection. Clin Infect Dis
4. Tang A, Graham N, Semba R, et al. Association between serum vitamin A and E levels and HIV-1 disease progression. AIDS
5. Baum M, Shor-Posner G, Lai S, et al. High risk of HIV-related mortality is associated with selenium deficiency. J Acquir Immune Defic Syndr Hum Retrovirol
6. Campa A, Shor-Posner G, Indacochea F, et al. Mortality risk in selenium-deficient HIV-positive children. J Acquir Immune Defic Syndr
7. Kupka R, Msamanga G, Spiegelman D, et al. Selenium status is associated with accelerated HIV disease progression among HIV-1-infected pregnant women in Tanzania. J Nutr
8. Lai H, Lai S, Shor-Posner G, et al. Plasma zinc, copper, copper:zinc ratio, and survival in a cohort of HIV-1-infected homosexual men. J Acquir Immune Defic Syndr
9. Fawzi W, Mbise R, Hertzmark E, et al. A randomized trial of vitamin A supplements in relation to mortality among human immunodeficiency virus-infected and uninfected children in Tanzania. Pediatr Infect Dis J
10. Burbano X, Miguez-Burbano M, McCollister K, et al. Impact of a selenium chemoprevention clinical trial on hospital admissions of HIV-infected participants. HIV Clin Trials
11. Jiamton S, Pepin J, Suttent R, et al. A randomized trial of the impact of multiple micronutrient supplementation on mortality among HIV-infected individuals living in Bangkok. AIDS
12. Shor-Posner G, Lecusay R, Miguez M, et al. Psychological burden in the era of HAART: impact of selenium therapy. Int J Psychiatry Med
13. Fawzi W, Msamanga G, Hunter D, et al. Randomized trial of vitamin supplements in relation to transmission of HIV-1 through breastfeeding and early child mortality. AIDS
14. Fawzi W, Msamanga G, Spiegelman D, et al. A randomized trial of multivitamin supplements and HIV disease progression and mortality. N Engl J Med
15. McClelland R, Baeten J, Overbaugh J, et al. Micronutrient supplementation increases genital tract shedding of HIV-1 in women: results of a randomized trial. J Acquir Immune Defic Syndr
16. Tang A, Graham N, Kirby A, et al. Dietary micronutrient intake and risk of progression to acquired immunodeficiency syndrome (AIDS) in human immunodeficiency virus type 1 (HIV-1)-infected homosexual men. Am J Epidemiol
17. Fawzi W, Villamor E, Msamanga G, et al. Trial of zinc supplements in relation to pregnancy outcomes, hematologic indicators, and T cell counts among HIV-1-infected women in Tanzania. Am J Clin Nutr
18. Semba R, Lyles C, Margolick J, et al. Vitamin A supplementation and human immunodeficiency virus load in injection drug users. J Infect Dis
19. Coodley G, Coodley M, Lusk R, et al. Beta-carotene in HIV infection: an extended evaluation. AIDS
20. Knox T, Spiegelman D, Skinner S, et al. Diarrhea and abnormalities of gastrointestinal function in a cohort of men and women with HIV infection. Am J Gastroenterol
21. Comstock G, Alberg A, Helzlsouer K. Reported effects of long-term freezer storage on concentrations of retinol, beta-carotene, and alpha-tocopherol in serum or plasma summarized. Clin Chem
22. Gunter E, Driskell W, Yeager P. Stability of vitamin E in long-term stored serum. Clin Chim Acta
23. Hsing A, Comstock G, Polk B. Effect of repeated freezing and thawing on vitamins and hormones in serum. Clin Chem
24. Sowell A, Huff D, Yeager P, et al. Simultaneous determination of retinal, α-tocopherol, lutein/zeaxanthin, β-cryptoxanthin, lycopene, trans-β-carotene, and four retinyl esters in serum by reverse-phase high performance liquid chromatography with multiwavelength detection. Clin Chem
25. Bieri J, Tolliver T, Catignani G. Simultaneous determination of alpha-tocopherol and retinol in plasma or red cells by high pressure liquid chromatography. Am J Clin Nutr
26. Baum M, Shor-Posner G, Ying L, et al. Micronutrients and HIV-1 disease progression. AIDS
27. Siberry G, Ruff A, Black R. Zinc and human immunodeficiency virus infection. Nutr Res
28. Bender D, Bender A. Vitamin A and carotenes. In: Nutrition: A Reference Handbook
. New York: Oxford University Press; 1997:228-244.
29. Baum M, Campa A, Lai S, et al. Zinc status in human immunodeficiency virus type 1 infection and illicit drug use. Clin Infect Dis
30. Baum M, Shor-Posner G. Micronutrient status in relationship to mortality in HIV-1 disease. Nutr Rev
31. Hendricks K, Dong K, Tang A, et al. High-fiber diet in HIV-positive men is associated with lower risk of developing fat deposition. Am J Clin Nutr
32. Woods M, Spiegelman D, Knox T, et al. Nutrient intake and body weight in a large HIV cohort that includes women and minorities. J Am Diet Assoc
33. Schakel S, Sievert Y, Buzzard I. Sources of data for developing and maintaining a nutrient database. J Am Diet Assoc
34. National Institutes of Health, National Heart, Lung and Blood Institute. Clinical Guidelines on the Identification, Evaluation, and Treatment of Overweight and Obesity in Adults
. NIH Publication No. 98-4083, Bethesda, MD, 1998.
35. Bogden J, Baker H, Frank O, et al. Micronutrient status and human immunodeficiency virus (HIV) infection. Ann NY Acad Sci
36. Beach R, Mantero-Atienza E, Shor-Posner G, et al. Specific nutrient abnormalities in asymptomatic HIV-1 infection. AIDS
37. Visser M, Maartens G, Kossew G, et al. Plasma vitamin A and zinc levels in HIV-infected adults in Cape Town, South Africa. Br J Nutr
38. Read J, Bethel J, Harris D, et al. Serum vitamin A concentrations in a North American cohort of human immunodeficiency virus type 1 infected children. Ped Infect Dis J
39. Rousseau M, Molines C, Moreau J, et al. Influence of highly active antiretroviral therapy on micronutrient profiles in HIV-infected patients. Ann Nutr Metab
40. Tang A, Smit E, Semba R, et al. Improved antioxidant status among HIV-infected injecting drug users on potent antiretroviral therapy. J Acquir Immune Defic Syndr
41. Duggan C, MacLeod W, Krebs N, et al. Plasma zinc concentrations are depressed during the acute phase response in children with Falciparum malaria. J Nutr
42. Batterham M, Gold J, Naidoo D, et al. A preliminary open label dose comparison using an antioxidant regimen to determine the effect on viral load and oxidative stress in men with HIV/AIDS. Eur J Clin Nutr
43. Taylor E, Cox A, Zhao L, et al. Nutrition, HIV, and drug abuse: the molecular basis of a unique role for selenium. J Acquir Immune Defic Syndr
44. Vivekananthan D, Penn M, Sapp S, et al. Use of antioxidant vitamins for the prevention of cardiovascular disease: meta-analysis of randomised trials. Lancet
45. Humphrey J, Quinn T, Fine D, et al. Short-term effects of large-dose vitamin A supplementation on viral load and immune response in HIV-infected women. J Acquir Immune Defic Syndr
46. MacDonald K, Malonza I, Chen D, et al. Vitamin A and risk of HIV-1 seroconversion among Kenyan men with genital ulcers. AIDS
47. Food and Nutrition Board, National Research Council. Dietary reference intakes for vitamin A, vitamin K, arsenic, boron, chromium, copper, iodine, iron, manganese, molybdenum, nickel, silicon, vanadium, and Zinc. Washington, DC: National Academy Press, 2002.
Keywords:© 2006 Lippincott Williams & Wilkins, Inc.
retinol; alpha-tocopherol; zinc; selenium; HIV; Nutrition for Healthy Living Cohort; micronutrients; CD4 cell count; log viral load