Table 2.
Prevalence of low vitamin D in HIV-infected individuals in low- to middle-income countries versus high-income countries
| Country (region) | Reference | Type of study | Patients | % Women | Age (mean or median; SD, IQR) | Results | Remarks |
| Low–middle-income countries (LMICs) | |||||||
| Turkey (EU/CA) | Aydin et al. [50] | Cross-sectional study, June to October 2010 | 96 HIV-infected patients (80.2% on ART) | 18 | 40.1 (range, 20–70) | Patients on ART: 14.3% had 25(OH)D <10 ng/ml, 67.5% had 25(OH) D 10–20 ng/ml; patients without ART: 15.8% had 25(OH)D <10 ng/ml, 73.7% had 25(OH) D 10–20 ng/ml | 25(OH)D levels were low in women with veiled dressing style. No relation between low BMD and 25(OH)D levels. |
| Israel (ME/NA) (Ethiopian origin and Caucasian origin) | Shahar et al. [53] | Cross-sectional study, Summer 2009 | 43 HIV-infected Ethiopians (mean CD4 233 cells/μl, 82% on ART, 20 study participants on PI) | 100 | 35.9 ± 8.2 | 65% had 25(OH)D <10 ng/ml 16.6% had 25(OH)D 10–20 ng/ml | PIs used were LPV/r, invirase/r, IDV. All participants were living in Israel for at least 10 years. Significantly more Ethiopian than Caucasian women covered their face and hands. |
| 32 HIV-infected Caucasians (mean CD4 264 cells/μl, 64% on ART, 21 study participants on PI) | 100 | 34.8 ± 8.7 | 6.25% had 25(OH)D <10 ng/ml 15.6% had 25(OH) D 10–20 ng/ml | ||||
| Batswana (SSA) | Steenhoff et al. [97] | Prospective study, December 2011 to April 2012 | 60 HIV-infected study participants (PI 25%, EFV 33%, NVP 42%, TDF/ NNRTI 13%) | 50 | 19.5 ± 12 | At baseline: 5% had 25(OH)D <20 ng/ml 26.5% had 25(OH)D 20–31 ng/ml. Mean 25(OH)D was 36.5 ng/ml in 4000 IU group and 34.5 ng/ml in 7000 IU group. At 12 weeks: 1.5% had 25(OH)D <20 ng/ml 16.8% had 25(OH)D 20–31 ng/ml. Mean 25(OH)D was 54.8 ng/ml in daily 4000 IU group and 56.5 ng/ml in daily 7000 IU group. | Δ25D was two-fold higher in study participants on EFV or NVP compared to those on PIs. At 6 weeks, both NNRTI regimens resulted in greater Δ25D than those on PIs. Study participants on TDF did not differ in Δ25D from study participants on other regimens. |
| Thailand1 (EA/P) | Chokephaibulkit et al. [98] | Cross-sectional study, October 2010 to February 2011 | 101 perinatally HIV-infected adolescents on ART (NNRTI-based: NVP 30%, EFV 20%, and PI-based: 50%), median CD4 646 cells/μl | 50 | 14.3 (13, 15.7) | Median 25(OH)D was 24.8 ng/ml, 24.7% had 25(OH)D <20 ng/ml, 46.5% had 25(OH)D 20–30 ng/ml | No associations between vitamin D deficiency and BMI, BMD, EFV use, HIV RNA, CD4, or self-reported sunlight exposure were observed. |
| Thailand2 (EA/P) | Avihingsanon et al. [99] | Cross-sectional analysis of cohort, July 2010 to June 2011 | 673 HIV-infected adults (93% on ART; EFV 31%, TDF 79% and 57% of patients had previously used d4T), median CD4 571 cells/μl | 47 | 41.5 (37.2, 47) | 40.6% had 25(OH)D <20 ng/ml, 29.9% had 25(OH)D 20–30 ng/ml | Female sex, age >37 years, and EFV use were independent predictors of hypovitaminosis D. |
| Thailand3 (EA/P) | Aurpibul et al. [100] | Cross-sectional study, March to September 2011 | 80 perinatally HIV-infected children on ART (NVP-based 55%, EFV-based 31%, PI-based 14%), median CD4 784 cell/μl | 56 | 12.2 (9.1, 14.3) | Median 25(OH)D was 33.5 ng/ml, 10% had 25(OH)D <20 ng/ml, 33% had 25(OH)D 21–29 ng/ml | Only geographic location was significantly associated with low vitamin D level. |
| Brazil 1 (LA) | Sales et al. [101] | Cross-sectional study, August 2011 to December 2013 | 32 HIV-infected women (most were on ART but type of ART was not indicated) | 100 | 41.7 | 15.63% had 25(OH)D<10 ng/ml; 65.63% had 25(OH)D 11–29 ng/ml | Factors related to the virus itself and to the use of ART may have contributed for the low vitamin D levels. |
| 66 HIV-infected men | 0 | 39 | 18.75% had 25(OH)D >30 ng/ml; 12.12% had 25(OH)D<10 ng/ml, 71.43% had 25(OH)D 11–29 ng/ml, 15.31% had 25(OH)D >30 ng/ml | ||||
| Brazil 2 (LA) | Canuto et al. [102] | Cross-sectional study, September 2013 | 125 HIV-infected patients (83.2% on ART but type of ART was not indicated) | 51.2 | 40.3 ± 11 | Mean 25(OH)D was 39.3 ng/ml 1.6% had 25(OH)D ≤20 ng/ml 22.4% had 25(OH)D 21–29 ng/ml | Higher 25(OH)D levels were associated with female sex, no use of sunscreen, and previous opportunistic infections. Lower values were associated with the use of ART, overweight and obesity. |
| High-income countries | |||||||
| USA 1 (NA) | Schwartz, Moore et al. [103] | Cross-sectional study, October 2009 to January 2010 | 507 HIV-negative study participants | 100 | 41.3 (33.6, 48.7) | Median 25(OH)D was 14 ng/ml; 72% had 25(OH)D <20 ng/ml; 18% had 25(OH)D 20–30 ng/ml; Median 25(OH)D was 14 ng/ml; 70% had 25(OH)D <20 ng/ml; 20% had 25(OH)D 20–30 ng/ml; Median 25(OH)D was 17 ng/ml; 57% had 25(OH)D <20 ng/ml; 24% had 25(OH)D 20–30 ng/ml | Vitamin D levels were lower if ART included efavirenz (15 vs. 19 ng/ml, P < 0.001). |
| 358 HIV-positive ART naive patients | 100 | 42.9 (36.3, 49.6) | |||||
| 893 HIV-positive patients on ART (PI 61%, NRTI 98%, NNRTI 26%) | 100 | 44.9 (39.3, 50.7) | |||||
| USA 2 (NA) | Hidron et al. [104] | Cross-sectional study, 2007–2010 | 933 HIV-infected patients (82% on ART; TDF/EFV 31.6%, TDF without EFV 29.9%, no TDF 20.6%) | 2.5 | 50 (range, 24–86) | Median 25(OH)D was 19 ng/ml 53.2% had 25(OH)D <20 ng/ml | Risk factors for vitamin D deficiency in HIV-positive patients included black race, winter season and higher GFR, increasing age and TDF use. |
| 5355 HIV-negative study participants | 13.1 | 63 (22–97) | Median 25(OH)D was 24 ng/ml, 38.5% had 25(OH)D <20 ng/ml | ||||
| USA 3 (NA) | Lake et al. [105] | Cross-sectional analysis of prospective study, June 2010 to April 2011 | 122 HIV-infected patients on ART (PI 34%, NNRTI 58%, raltegravir 17%, TDF 80%, ABC 29%), mean CD4 520 cells/μl | 5 | 49 (41, 55) | Median 25(OH)D was 20 ng/ml, 67.2% had 25(OH)D <30 ng/ml | After 12 weeks of vitamin D supplementation (vitamin D3 50000 IU twice weekly for 5 weeks, then 2000 IU daily), 81% of insufficient persons achieved 25OHD ≥30 ng/ml. 25OHD repletion rates were comparable between HIV-positive patients and controls. |
| Australia (EA/P) | Klassen et al. [106] | Cross-sectional study, January 2008 to December 2012 | 997 HIV-infected patients (66% on ART; EFV 24%, NNRTI ± PI 25%, no NNRTI/PI 17%) | 12 | 41 (32,48) | Mean 25(OH)D was 24.8 ng/ml; 40% had 25(OH)D <20 ng/ml; 71% had 25(OH)D <30 ng/ml | Men, Caucasian country of origin, summer/autumn, total cholesterol to HDL ratio >5 and HIV infection were associated with vitamin D deficiency. |
| May 2009 to April 2010 | 3653 HIV-uninfected individuals | 53 | 50 (39, 61) | Mean 25(OH)D was 27.6 ng/ml; 22% had 25(OH)D <20 ng/ml; 63% had 25(OH)D <30 ng/ml | |||
| United Kingdom (EU/CA) | Gedela et al. [107] | Cross-sectional study, January 2008 to December 2009 | 253 HIV-infected ART-naive study participants (64.4% were white and 35.6% were black or other ethnicity) with median CD4 450 cells/μl | 18 | 36 (range, 16–75) | 12.6% had 25(OH)D ≤10 ng/ml; 58.5% had 25(OH)D ≤20 ng/ml | Vitamin D deficiency was common among ART naive patients, with those of nonwhite ethnicity at highest risk; no association was found with CD4 cell count, HIV viral load, and HIV clinical staging. |
| Belgium (EU/CA) | Theodorou et al. [108] | Retrospective study, December 2005 to March 2011 | 2044 HIV-infected study participants (73.4% on ART; EFV 15.8%, 2NRTI/NNRTI 23.2%, 2NRTI/PI 35.9%, second line 14.2%) | 41.5 | 43 (range, 20–85) | Median 25(OH)D was 13.8 ng/ml, 32.4% had 25(OH)D <10 ng/ml, 89.2% had 25(OH)D <30 ng/ml | 25(OH)D <30 ng/ml is associated with general factors (female sex, winter season) and specific factors related to HIV (duration of treatment, second line treatments with multiple and complex combinations of ART). 25(OH)D <10 ng/ml is associated with a low CD4 cell count, a higher CDC stage and EFV therapy. |
| Spain (EU/CA) | Bañón et al. [109] | Prospective study, 2012 | 365 HIV-infected patients (98% on ART; TDF/FTC 77%, the remaining on ABC/3TC, EFV 33%) | 24 | 44 (range, 22–75) | At baseline: 15% had 25(OH)D <10 ng/ml; 48% had 25(OH)D 10–19.9 ng/ml; 26% had 25(OH)D 20–29.9 ng/ml | After calcidiol supplementation (oral monthly dose of 16 000 IU), 25(OH)D levels increased in comparison with nonsupplemented patients (+16.4 vs. + 3.2 ng/ml; P < 0.01). |
Data shown in the table includes published articles and abstracts related to prevalence of hypovitaminosis D in HIV-infected adolescents or adults from RLS in 2014 and 2015 plus articles of special interest (+) from 2013. However, for RRS, only the articles published in mid-2014 to 2015 and had more than 100 HIV-infected participants were included. HICs, high-income countries; LMICs, low- to middle-income countries; SSA, sub-Saharan Africa.