Cross-sectional study in Canada. 11 HIV-positive adults (6 receiving HAART for 3 y, 5 not receiving any HIV medications).
Vitamin A: HAART (51 ± 5 µg/dL); no HIV medications (66 ± 11 µg/dL)
Mean plasma concentrations of vitamin A and retinol-binding protein were significantly lower (P = 0.03) and higher (P = 0.04), respectively, in those receiving HAART.
Rousseau et al, 2000 (94)
Cross-sectional study in France. 30 HIV-positive adults, mostly injection-drug users (23 receiving HAART for3 y, 7 not receiving HAART).
Vitamin A: total (0.66 ± 1.2 µmol/L); 24 of 30 (80%) deficient (<1.5 µmol/L); concentrations not presented for HAART and non-HAART groups Vitamin E: total (9.24 ± 3.4 mg/L); 10 of 29 (34%) deficient (<6 mg/L); concentrations not presented for HAART and non-HAART groups
Mean plasma concentrations of vitamins A and E were not significantly different between those with a CD4 count < and >250 cells/µL, between those with viral load > and <5000 copies/mL, and between those receiving and not receiving HAART.
Tang et al, 2000 (95)
Cross-sectional study in the United States. 175 HIV-positive injection-drug users (30 receiving HAART, 65 receiving dual- or monotherapy, 80 not receiving any HIV medications).
-Tocopherol2: HAART (1076 ± 468 µg/dL); no HIV medications (778 ± 209 µg/dL) -Carotene3: HAART (1.06 ± 0.01 µg/dL); no HIV medications (0.74 ± 0.02 µg/dL) ß-Carotene2: HAART (8.8 ± 3.1 µg/dL); no HIV medications (5.2 ± 0.9 µg/dL) Vitamin A2: HAART (42.0 ± 11.4 µg/dL); no HIV medications (38.4 ± 6.2 µg/dL) -Tocopherol2: HAART (238 ± 107 µg/dL); no HIV medications (202 ± 59 µg/dL)
Adjusted mean serum concentrations of-tocopherol (P = 0.0008), -carotene (P= 0.05), and ß-carotene (P = 0.02), but not of vitamin A and -tocopherol, were significantly higher in those receiving HAART than in those not taking any HIV medications; no significant differences in adjusted mean serum vitamin concentrations between CD4 cell count categories (<200, 200–499, and 500 cells/µL).
Remacha et al, 2003 (96)
Cross-sectional study in Spain. 126 HIV-positive adults receiving HAART compared with 109 HIV-positive historical control subjects from 1989 to 1992 receiving HAART.
Folate: HAART (1473 ± 1087 mmol/L), 1 of 126 (0.8%) deficient (450 mmol/L); historical control subjects (1057 ± 665 mmol/L), 19 of 109 (17.4%) deficient Vitamin B-12: HAART (402 ± 218 pmol/L), 2 of 126 (1.2%) deficient (150 pmol/L); historical control subjects (330 ± 219 pmol/L), 20 of 109 (18%) deficient
Mean concentrations of red blood cell folate and serum vitamin B-12 were significantly higher in HIV-positive adults receiving HAART than in historical HIV-positive control subjects receiving HAART. Significantly fewer HIV-positive adults receiving HAART than historical control subjects had folate or vitamin B-12 deficiencies.
Woods et al, 2003 (97)
Cross-sectional study from 1995 to 2000 in the United States. 412 HIV-positive adults (615 patient-time intervals in adults receiving HAART, 454 patient-time intervals in adults not receiving HAART).
Median serum concentration of vitamin B-12 was significantly higher at the beginning of each patient-time interval in HIV-positive adults receiving HAART; multivariate analyses were not performed to account for higher intakes of vitamin B-12 (P = 0.0002) in participants receiving HAART.
Longitudinal studies
Look et al, 2001 (98)
Longitudinal study from 1997 to 1998 in Germany. 17 HIV-positive adults studied at baseline and 100 d after HAART initiation.
Median follow-up serum concentrations of vitamin B-6, folate, and methylmalonic acid were not significantly higher than median baseline concentrations; however, baseline concentrations of vitamin B-6, folate, and methlymalonic acid were not significantly different from those of a cohort of HIV-negative healthy control subjects.
Observational studies of minerals in HIV-infected persons receiving highly active antiretroviral therapy (HAART)
Reference
Study design, location, and population
Mineral concentrations
Results and conclusions
Cross-sectional studies
Batterham et al, 2001 (99)
Cross-sectional study in Australia. 48 HIV-positive adults (35 receiving HAART, 13 not receiving any HIV medications).
Selenium: HAART with detectable (>400 copies/mL) viral load (2.16 ± 0.54 µmol/L); HAART with undetectable viral load (2.22 ± 0.93 µmol/L); no HIV medications (2.40 ± 0.83 µmol/L)
Mean serum concentrations of glutathione peroxidase (P = 0.001), lipid peroxidase (P = 0.03), and uric acid (P = 0.009), but not of selenium, were significantly different between HIV-positive persons receiving HAART and those not taking any HIV medications.
Wellinghausen et al, 2000 (100)
Cross-sectional study in Germany. 79 HIV-positive adults (52 receiving HAART; 4 receiving dual- or monotherapy, 23 not receiving any HIV medications).
Zinc: HAART (12.5 ± 2.8 µmol/L), 25% deficient (<10.5 µmol/L); no HIV medications (12.7 ± 2.7 µmol/L), 22% deficient
Zinc concentrations were not significantly different between those receiving and those not receiving HAART.
Longitudinal studies
Rousseau et al, 2000 (94)
Longitudinal study from 1995 to 1998 in France. 44 HIV-positive adults, mostly injection-drug users. At baseline, none were receiving HAART, but 80% were receiving dual-combination therapy. Of 30 participants with follow-up data, 23 were receiving HAART and 7 were not receiving any HIV medications.
Mean serum concentrations of selenium, iron, zinc, and copper did not significantly increase from baseline; however, significantly fewer participants had selenium deficiency and copper overload at follow-up; at follow-up, mean concentrations of selenium, iron, zinc, and copper were not significantly different between those receiving and those not receiving HAART.
Intervention studies of micronutrients in HIV-infected persons receiving highly active antiretroviraltherapy (HAART)
Reference
Study design, population, and inclusion and exclusion criteria
Intervention
Primary outcomes
Major findings
Conclusions
Nonrandomized trials
McComsey et al, 2003 (101)
Nonrandomized, open-label pilot study without placebo control in the United States. 10 HIV-positive adults receiving HAART for 12 mo. 9 had lipoatrophy and 1 had sustained hyperlactemia at enrollment.
Daily vitamin C (1000 mg) and vitamin E (800 IU) and twice-daily N-acetyl cysteine (600 mg) for 24 wk.
Intervention significantly increased fasting glucose and insulin resistance and decreased waist-to-hip ratio compared with placebo; intervention had no significant effect on peripheral fat, lipoatrophy, CD4 cell count, or plasma viral load.
24 wk of a supplement worsened fasting glucose and insulin resistance and did not significantly improve peripheral fat, lipoatrophy, immunologic status, or plasma viral load.
Batterham et al, 2001 (99)
Nonrandomized trial without placebo control in Australia. 66 HIV-positive adults enrolled and 48 completed study (32 receiving HAART, 3 receiving dual therapy, 13 not receiving any HIV medications). Exclusion criteria included taking supplements within 4 wk of enrollment, not clinically stable or with active infection, and change of HIV medication regimen within 6 wk of enrollment.
Daily supplementation with either a low-dose1 or a high- dose2antioxidant regimen for 12 wk.
Intervention significantly increased concentrations of glutathione and glutathione peroxidase from baseline, but had no significant effect on allantoin, uric acid, or plasma viral load; no significant differences between those receiving low-dose and those receiving high-dose regimens.
12 wk of an antioxidant supplement increased oxidative defenses, but did not affect oxidative stress or plasma viral load.
Randomized trials
Jensen-Fangel et al, 2003 (102)
Randomized crossover trial without placebo control in Denmark. 15 HIV-positive adults receiving HAART; all with chronic nelfinavir-associated diarrhea.
Twice-daily calcium carbonate (1350 mg) f or 14 d. A subset of 6 patients additionally treated with twice-daily calcium gluconate (2950 mg) and an extra 300 mg calcium carbonate.
Clinical improvement of diarrhea
Intervention had no significant effect on clinical measurements of diarrhea.
14 d of a calcium carbonate supplement did not clinically improve nelfinavir-associated diarrhea.
Spada et al, 2002 (103); De Souza et al, 2005 (104)
Randomized, placebo-controlled trial in Brazil. 29 HIV-positive adults with CD4 count <500 cells/µL. 26 initiated HAART and 3 initiated dual-combination therapy at study enrollment.
Intervention had no significant effect on CD4 cell count, CD4: CD8 cells, or plasma viral load as compared with placebo; intervention significantly increased lymphocyte viability compared with placebo.
6 mo of vitamin E supplementation improved lymphocyte viability, but did not affect immune cell count or plasma viral load.
Jaruga et al, 2002 (105)
Randomized, placebo-controlled trial in Poland. 30 HIV-positive adults receiving HAART.
Daily vitamin A (5000 IU), vitamin C (50 mg), and vitamin E (100 IU) for 6 mo.
Intervention significantly increased concentrations of catalase and superoxide dismutase and significantly lowered thiobarbituric acid–reactive substances; the CD4 cell count increased in the intervention group from baseline, whereas the mean CD4 count of the placebo group decreased, but the difference was not statistically significant.
6 mo of an antioxidant multivitamin supplement significantly increased antioxidant defenses, significantly reduced oxidative stress, and possibly improved immunologic status.
Burbano et al, 2002 (106)
Randomized placebo-controlled trial in the United States. 186 HIV-positive injection-drug users (85 receiving HAART, 39 receiving dual- or monodrug therapy, 52 not receiving any HIV medications).
Daily selenium (200 µg) for 2 y.
CD4 cell count, hospital admissions
Significantly fewer participants in the intervention group than in the placebo group had a decrease in CD4 cell count of >50 cells/µL during the study; intervention significantly reduced hospital admissions because of opportunistic infections and other HIV-related conditions; in multivariate analyses,3the placebo group had a 2.4 greater risk of hospitalization (P = 0.01).
2 y of a selenium supplement decreased large reductions in CD4 cell count and reduced the risk of hospitalization.
Kaiser et al, 2006 (107)
Randomized placebo-controlled trial in the United States. 40 HIV-positive adults receiving HAART.
Micronutrient supplementation twice daily for 12 wk.4
Intervention significantly increased absolute CD4 cell count (P = 0.03) and mean change in CD4 cell count from baseline (P= 0.01) and had no significant effects on fasting glucose, insulin, lipids, or plasma viral load.
12 wk of micronutrient supplementation increased CD4 cell count.
1 Included 5450 IU ß-carotene, 250 mg vitamin C, 100 IU vitamin E, 100 µg Se, 50 mg coenzyme Q10, 10 mg thiamine, 25 mg vitamin B-6, 55 mg pantothenic acid, 250 µg folate, 50 µg vitamin B-12, and 5 mg Zn.
2 Included 21800 IU ß-carotene, 1000 mg vitamin C, 400 IU vitamin E, 200 µg Se, 200 mg coenzyme Q10, 40 mg thiamine, 100 mg vitamin B-6, 220 mg pantothenic acid, 1000 µg folate, 200 µg vitamin B-12, and 20 mg Zn.
3 Multivariate analysis adjusted for HAART (yes or no), use of other HIV medications (yes or no), age >50 y, CD4 cell count at baseline, and plasma viral load >10000 copies/mL at baseline.
3 y, 5 not receiving any HIV medications).
3 y, 7 not receiving HAART).
-Tocopherol2: HAART (1076 ± 468 µg/dL); no HIV medications (778 ± 209 µg/dL) 
-Tocopherol2: HAART (238 ± 107 µg/dL); no HIV medications (202 ± 59 µg/dL)