| |
Vitamin D Deficiency in HIV-Infected
and HIV-Uninfected Women in the United States
|
| |
| |
Download the PDF here
JAIDS Journal of Acquired Immune Deficiency Syndromes:
1 July 2011
"it will be important to explore the response to vitamin D supplementation in these women and the impact on a myriad of health outcomes including insulin resistance, cardiovascular disease, bone health, and a variety of mental health outcomes. An understanding of the role that vitamin D deficiency plays in non-AIDS-related morbidities reported in excess among HIV-infected individuals is planned for investigation in WIHS.
In the largest study to date of vitamin D levels in HIV-infected persons in the United States, vitamin D deficiency was highly prevalent among HIV-infected women but was less prevalent than among the HIV-uninfected women. Overall, 66% of the cohort had vitamin D deficiency and 22% had severe deficiency. African American race was the strongest independent predictor of vitamin D deficiency......However, it is important to note that only 15% and 8% of HIV-infected and HIV-uninfected women, respectively, in our cohort had sufficient vitamin D levels compared with 23% of the adult US population.11......Our definition of vitamin D deficiency as levels <20 ng/mL was similar to that in previous NHANES studies of vitamin D and other studies in the literature......Our finding of an association of non-white race with lower vitamin D levels is consistent with other studies,16,17-21 as is our finding of an association of greater BMI with lower vitamin D levels......The association of non-white race likely reflects the increased skin melanin content that interferes with the UV-B required conversion of previtamin D3 to vitamin D3 in the skin. African American and Hispanic women are at particularly high risk of vitamin D deficiency and this may be associated with high BMI values.....visceral adiposity was associated with lower vitamin D levels in a recently published study of African American diabetic patients......Obesity and central fat deposition are thought to lead to lower vitamin D levels due to sequestration of vitamin D in fat cells, leaving lower levels of circulating vitamin D available for hydroxylation to active forms.1,28 Another possible reason may be decreased outdoor activity and thus sun exposure in obese individuals......we found that renal insufficiency was associated with higher vitamin D levels
Among the HIV-infected women, control of HIV viremia and higher CD4 counts were associated with higher vitamin D levels after controlling for other demographic and clinical factors. We found that women currently on a PI-based regimen were significantly less likely to have low vitamin D levels compared with treatment-naive women and women on NNRTI-based regimens. Although women on efavirenz had a higher risk of vitamin D deficiency, it did not reach statistical significance. The mechanism by which efavirenz, an NNRTI, lowers vitamin D levels is through induction of 24-hydroxylase, a CYP450 enzyme, which causes increased catabolism of 25(OH)D3 into the inactive form 24,25(OH)2D.34 Further exploration of individual antiretrovirals especially efavirenz, ritonavir, and tenofovir on vitamin D status is planned within the WIHS."
Adeyemi, Oluwatoyin M MD*; Agniel, Denis MPH*; French, Audrey L MD*; Tien, Phyllis C MD; Weber, Kathleen*; Glesby, Marshall J MD; Villacres, Maria C PhD; Sharma, Anjali MD, MS; Merenstein, Daniel MD**; Golub, Elizabeth T PhD, MPH; Meyer, William PhD; Cohen, Mardge MD*
From the *Department of Medicine, CORE Center, Cook County Hospital, Chicago, IL; Rush University Medical Center; Department of Medicine, University of California, San Francisco, CA; Department of Medicine, Weill Medical College of Cornell University, Ithaca, NY; Department of Pediatrics, University of Southern California, Los Angeles, CA; Department of Medicine, SUNY Downstate Medical Center, Brooklyn, NY; **Department of Medicine, Georgetown University Medical Center, Washington, DC; Johns Hopkins University School of Public Health, Baltimore, MD; and Quest Diagnostics, Baltimore, MD.
Abstract
Background: Vitamin D deficiency is of increasing concern in HIV-infected persons because of its reported association with a number of negative health outcomes that are common in HIV. We undertook this study to determine the prevalence and predictors of vitamin D deficiency among a nationally representative cohort of middle-aged, ethnically diverse, HIV-infected and HIV-uninfected women enrolled in the Women's Interagency HIV Study (WIHS).
Methods: Vitamin D testing was performed by Quest Diagnostics on frozen sera using the liquid chromatography/mass spectroscopy method. Vitamin D deficiency was defined as 25(OH)D ≤20 ng/mL. Comparisons of continuous and categorical characteristics among HIV-infected and HIV-uninfected women were made by Wilcoxon tests and Pearson χ2 tests, respectively.
Results: One thousand seven hundred seventy-eight women (1268 HIV positive) were studied. Sixty-three percent had vitamin D deficiency (60% HIV positive vs. 72% HIV negative; P < 0.001). Multivariable predictors of vitamin D deficiency were being African American (adjusted odds ratio 3.02), Hispanic (adjusted odds ratio 1.40), body mass index (adjusted odds ratio 1.43), age (adjusted odds ratio 0.84), HIV positive (adjusted odds ratio 0.76), glomerular filtration rate <90·mL-1·min-1 (adjusted odds ratio 0.94), and WIHS sites Los Angeles (adjusted odds ratio 0.66) and Chicago (adjusted odds ratio 0.63). In the HIV-positive women, multivariate predictors were undetectable HIV RNA (adjusted odds ratio 0.69), CD4 50-200 cells per cubic millimeter (adjusted odds ratio 1.60), CD4 <50 cells per cubic millimeter (adjusted odds ratio 1.94), and recent protease inhibitor use (adjusted odds ratio 0.67).
Conclusions: In this study of more than 1700 women in the United States, most women with or without HIV infection had low vitamin D levels and African American women had the highest rates of vitamin D deficiency. An understanding of the role that vitamin D deficiency plays in non-AIDS-related morbidities is planned for investigation in WIHS.
INTRODUCTION
Vitamin D deficiency is of increasing concern in HIV-infected persons because of its reported association with a number of negative health outcomes that are common in HIV. In addition to its known association with osteoporosis, low serum levels of 25-hydroxyvitamin D [25(OH)D] in studies in the general population have been linked to hypertension, diabetes, metabolic syndrome, cardiovascular disease,1-7 and certain solid organ malignancies.8,9 Low serum levels of 25(OH)D may also partly explain racial differences in disease outcomes. A recent US study found that the higher risk of cardiovascular mortality observed in black vs. white was attenuated by adjusting for serum 25(OH)D levels.10
Although the evidence supporting the role of vitamin D on a variety of health outcomes continues to emerge, vitamin D deficiency remains common in the US population.11,12 In the most recent National Health and Nutrition Examination Survey (NHANES) study report, only 23% of the adult US population had sufficient vitamin D levels.11 Vitamin D deficiency can be caused by several factors, including decrease in skin synthesis of previtamin D with skin pigmentation and aging, low sunlight exposure, obesity, and insufficient vitamin D intake. In HIV-infected persons, certain antiretroviral medications may be associated with vitamin D deficiency due to their potential effect on vitamin D metabolism.13-15 The reported prevalence of vitamin D deficiency among HIV-infected adolescents and adults has varied widely, ranging from 45% to 87% in studies conducted in Europe, Australia, and the United States.16-22 To date, studies on vitamin D in HIV-infected individuals have included primarily white men16-21 or have focused on young adults.22
We undertook this study to determine the prevalence and predictors of vitamin D deficiency among a nationally representative cohort of middle-aged, ethnically diverse, HIV-infected and HIV-uninfected women enrolled in the Women's Interagency HIV Study (WIHS).
RESULTS
Table 1 shows the demographic and clinical characteristics of the 1778 women included in the analysis by HIV status. HIV-infected women were older, more likely to be HCV seropositive, and less likely to be current smokers or currently employed compared with HIV-uninfected women (P < 0.001). Median vitamin D levels were higher in HIV-infected (16, interquartile range 10-25) than in HIV-uninfected women (14, interquartile range 9-20), both in the vitamin D deficiency (25(OH)D < 20 ng/mL) range. The proportion of women with vitamin D deficiency was also lower in HIV-infected than HIV-uninfected women (60% vs. 72%; P < 0.001). (Figure 1). Overall, 63% of women had vitamin D deficiency and 22% had severe vitamin D deficiency [25(OH)D < 10 ng/mL]. Only 13% of the entire cohort had sufficient vitamin D levels (>30 ng/mL). There were 128 (7%) women with detectable 25(OH)2D levels (implying measurable levels from oral supplementation) and this did not differ significantly across sites (data not shown). HIV-infected women were more likely to have detectable vitamin D2 levels, odds ratio (OR) of 1.57 [95% confidence interval (CI): 1.03 to 2.46].Older age was associated with detectable vitamin D2 levels, with an unadjusted OR of 1.96 (95% CI: 1.53 to 2.51) comparing the 75th age percentile 50 years of age with the 25th age percentile 37 years of age. We also observed a correlation between age and vitamin D2 level (r = 0.07, P = 0.005).
Predictors of Vitamin D Deficiency
In univariate analysis including the entire cohort, non-white race, younger age, HIV-uninfected status, HCV-uninfected status, current smoking, a higher BMI, normal renal function, being employed, and WIHS site were associated with vitamin D deficiency (Table 2). In multivariable analysis, being of black or Hispanic race (compared with white race) was associated with higher odds of vitamin D deficiency, as was a greater BMI and normal renal function. On the other hand, HIV infection, older age, and being from the LA or Chicago WIHS site (compared with the Bronx site) were associated with lower odds of vitamin D deficiency. When vitamin D was studied continuously in the entire cohort, we found similar associations, although the association of HIV infection with higher vitamin D levels did not reach significance (+0.06 log vitamin D increase; 95% CI: -0.01 to +0.13).
Treatment-Related Factors Among HIV-Infected Women
We studied the HIV-related factors associated with vitamin D deficiency in HIV-infected participants only in a model that controlled for all the variables presented in Table 2. Undetectable HIV RNA was associated with a lower odds of vitamin D deficiency and a lower CD4 count was associated with a higher odds of vitamin D deficiency after adjustment for demographic and clinical factors (Table 3). Compared with no antiretroviral therapy use, recent PI use was associated with lower odds of vitamin D deficiency, whereas recent NNRTI use was associated with higher odds of vitamin D deficiency, although the association did not reach statistical significance. Because efavirenz has been linked to low vitamin D levels,13-15,21 we conducted a separate analysis of efavirenz and vitamin D status. When compared with participants not on antiretroviral therapy, the lowest risk of vitamin D deficiency was in women on non-efavirenz-containing regimens; efavirenz-/tenofovir-; OR 0.49 (0.35-0.68), P < 0.001, and efavirenz-/tenofovir+; OR 0.42 (0.31-0.57).
DISCUSSION
In the largest study to date of vitamin D levels in HIV-infected persons in the United States, vitamin D deficiency was highly prevalent among HIV-infected women but was less prevalent than among the HIV-uninfected women. Overall, 66% of the cohort had vitamin D deficiency and 22% had severe deficiency. African American race was the strongest independent predictor of vitamin D deficiency. Similar to the recently published Study to Understand the Natural History of HIV/AIDS in the Era of Effective Therapy (SUN) study,21 HIV-infected women in our study were less likely to have vitamin D deficiency than HIV-uninfected women. HIV-infected women in the WIHS may be more engaged in primary care (and thus have a greater knowledge regarding vitamin D); a higher proportion had detectable 25(OH)2D levels and had a lower prevalence of risk factors associated with vitamin D deficiency than their HIV-uninfected counterparts, who were enrolled based on having similar risk behaviors as HIV-infected women. However, it is important to note that only 15% and 8% of HIV-infected and HIV-uninfected women, respectively, in our cohort had sufficient vitamin D levels compared with 23% of the adult US population.11
Our prevalence of 60% with vitamin D deficiency in HIV-infected women is consistent with the 45%-87% of vitamin D deficiency reported in other studies conducted in HIV-infected individuals.16-22 The wide range of reported deficiency rates likely reflects differences in chosen cut points for deficiency (<10, <15, <20 to <30 ng/mL) and differences in geographic location, season of the year, and demographics of the participants. Our definition of vitamin D deficiency as levels <20 ng/mL was similar to that in previous NHANES studies of vitamin D and other studies in the literature.
Our finding of an association of non-white race with lower vitamin D levels is consistent with other studies,16,17-21 as is our finding of an association of greater BMI with lower vitamin D levels.19,21,26-30 The association of non-white race likely reflects the increased skin melanin content that interferes with the UV-B required conversion of previtamin D3 to vitamin D3 in the skin. African American and Hispanic women are at particularly high risk of vitamin D deficiency and this may be associated with high BMI values. In agreement with this notion, visceral adiposity was associated with lower vitamin D levels in a recently published study of African American diabetic patients.27,28 Obesity and central fat deposition are thought to lead to lower vitamin D levels due to sequestration of vitamin D in fat cells, leaving lower levels of circulating vitamin D available for hydroxylation to active forms.1,28 Another possible reason may be decreased outdoor activity and thus sun exposure in obese individuals.30 The interplay between these factors and race is likely synergistic,particularly in the WIHS, and needs study.
We found an association between increasing age and higher vitamin D levels, which is contrary to what is known in the general population, where the elderly have lower vitamin D levels.11,12 A possible reason may be that the majority of older women in WIHS were below the age categorized as elderly in the general population. In our study, the median age of the >50-years group was 55.6 years, a decade lower than the >65 years definition of elderly in the general population. In the NHANES study, although there was a trend toward lower vitamin D levels with increasing age for non-Hispanic whites and Mexican Americans, this inverse association between age and vitamin D levels was not seen in non-Hispanic blacks,12 who are the majority of our study participants. We also found an association between increasing age and detectable 25(OH)2 levels, which could indicate higher rates of vitamin D supplementation in older WIHS women due to the perceived increased risk of bone disease.
Our multicenter study also allowed us to examine geographical differences in vitamin D levels. In temperate latitudes near sea level, cutaneous production of vitamin D almost ceases in winter. In the Reaching for Excellence in Adolescent Health (REACH) cohort study,22 latitude in the United States was an independent predictor of vitamin D levels. We found a significant association between US city of residence and vitamin D levels, with Chicago and LA women having the highest mean vitamin D levels and lowest rates of deficiency. Surprisingly, we found that women in San Francisco had lower levels of vitamin D, whereas women in LA had higher levels compared with the Bronx site after controlling for other factors. San Francisco is in the northern part of the state and thus may have lower UV-B rays year-round. A variety of unmeasured factors including varying degrees of sun exposure, use of sunblock, use of artificial light and tanning beds, dietary differences in consumption of vitamin D-rich foods, and supplemental vitamin D intake could explain differences in vitamin D levels across sites.
In our study, we found that renal insufficiency was associated with higher vitamin D levels. This is contrary to many reports that in chronic kidney disease, vitamin D levels are decreased,31-33 but consistent with findings from the large SUN study.21 We did not measure levels of 1,25(OH)2D, the active form of vitamin D, so we do not speculate on the clinical significance of this finding.
Among the HIV-infected women, control of HIV viremia and higher CD4 counts were associated with higher vitamin D levels after controlling for other demographic and clinical factors. We found that women currently on a PI-based regimen were significantly less likely to have low vitamin D levels compared with treatment-naive women and women on NNRTI-based regimens. Although women on efavirenz had a higher risk of vitamin D deficiency, it did not reach statistical significance. The mechanism by which efavirenz, an NNRTI, lowers vitamin D levels is through induction of 24-hydroxylase, a CYP450 enzyme, which causes increased catabolism of 25(OH)D3 into the inactive form 24,25(OH)2D.34 Further exploration of individual antiretrovirals especially efavirenz, ritonavir, and tenofovir on vitamin D status is planned within the WIHS.
The limitations of the present study need to be addressed. First, vitamin D testing was performed on stored samples. However, this has been done routinely in many large epidemiologic studies of vitamin D, and results are valid if samples are stored at -70°C until testing as was done here. Second, we did not have data from all sites on some of the most prominent factors in determining vitamin D level, including sun exposure, sunscreen use, and dietary vitamin D intake, and could not factor these into our analysis. Third, the samples tested were collected over a 6-month period when women came in for visits; however, they were all fall/winter samples, representing the same season in all the women. The strengths of our study include the large sample size of HIV-infected women across the United States, an HIV-uninfected control group, and the analysis of vitamin D during the same period in all women. Finally, although women in this study do not represent the US population, they are reflective of the epidemiology of HIV in women in the United States, the largest growing demographic group of the present US HIV epidemic.
In conclusion, in this study of more than 1700 women in the United States, we found that most women with and without HIV infection had low vitamin D levels and African American women had the highest rates of vitamin D deficiency. Although there are no formal guidelines on testing for and supplementation of vitamin D in HIV-infected women, it will be important to explore the response to vitamin D supplementation in these women and the impact on a myriad of health outcomes including insulin resistance, cardiovascular disease, bone health, and a variety of mental health outcomes. An understanding of the role that vitamin D deficiency plays in non-AIDS-related morbidities reported in excess among HIV-infected individuals is planned for investigation in WIHS.
| |
| |
| |
|
|
|
