HYPOVITAMINOSIS D IN HIV-INFECTED PATIENTS

Abstract

Introduction

Vitamin D (VD) deficiency is highly prevalent worldwide.

Aim

To assess the prevalence of hypovitaminosis D in HIV-positive Romanian patients compared to controls.

Methods

Serum 25OHD concentration was measured in HIV-infected patients and a control sample, matched by age, sex and menopausal status. The 25OHD status was defined as: deficiency < 20 ng/mL (severe deficiency <10 ng/mL), insufficiency 20-30 ng/mL, normal >30 ng/mL.

Results

We evaluated 118 HIV-positive patients (72 males, 46 females), aged 36.9±12.2 years. 98.14% of them were on complex antiviral regimens. The B/C hepatitis coinfection rate was 9.3%. The control sample consisted of 119 subjects, (74 males, 45 women). The median and interquartile range for serum 25OHD concentration in patients was 17.6 (9.7, 26.9) ng/mL and 23.7 (18.4, 27.5) ng/mL in controls (p=0.001). Only 15.96% of HIV-positive cases and 12.71% of controls had normal VD status. The percentage of cases with severe VD deficiency was significantly higher in HIV positive cases (23.52%) compared to HIV-negative controls (4.2%, p=0.001).

Conclusions

Hypovitaminosis D was identified in 84.04% of HIV-infected patients, but the serum 25OHD concentration was not associated with specific HIV-related factors in our sample. Clinical guidelines regarding VD status determination and supplementation in HIV patients are needed.

INTRODUCTION

Vitamin D (VD) plays an essential role in skeletal health. The optimal serum marker to appreciate the VD status of the body is the serum concentration of 25 hydroxyvitamin D (25OHD). Serum levels of 25OHD are highly dependent on the amount of sunlight exposure, dietary intake of VD, skin pigmentation and numerous other factors (1). Low VD status has recently been associated to other negative health outcomes apart from bone-related disorders, such as neoplasia, cardiometabolic diseases including diabetes, autoimmune and infectious diseases (2), although to date most of these associations are not based on strong indisputable evidence. Considering the available evidence, the US Endocrine Society defines optimal VD status as having a serum 25OHD concentration above 30 ng/mL (VD deficiency is defined by a serum 25OHD level of 20 ng/mL or less and VD insufficiency as 21-29 ng/mL) (3).

Several studies have demonstrated very high rates of vitamin D deficiency among HIV-infected patients (4,5). Possible reasons include various disease-related gastrointestinal disturbances leading to decreased VD absorption, comorbidities, less sun exposure (6). Also, the specific antiretroviral treatment can sometimes contribute as some ART drugs induce VD inactivation (7).

However, in the general population worldwide, the prevalence of VD deficiency is also high (8). Some studies have not reported significantly higher prevalences of low VD levels in HIV-infected patients compared to controls (5, 9, 10), while others found that HIV-positive cases have increased prevalence of VD deficiency compared to control groups (11). As mentioned before,in patients on antiretroviral therapy (ART) the risk of VD deficiency appears to be higher, especially in those receiving efavirenz (12).

The prevalence of VD deficiency in Romania has been relatively scarcely studied, but all studies to date demonstrated a very high prevalence of VD deficiency in various groups of subjects (e.g.postmenopausal women with or without osteoporosis) (13,14). Since no study on the prevalence of VD deficiency among HIV positive patients in comparison to healthy controls has been performed in the Romanian population we aim to describe the vitamin D status and compare it between HIV-negative and HIV-infected patients evaluated in a single university department.

MATERIALS AND METHODS

The study group consisted of consecutively recruited adult HIV-positive patients treated and followed-up in the “Prof. Dr. Matei Balş” National Institute for Infectious Diseases, Department of Infectious Diseases of the “Carol Davila” University of Medicine and Pharmacy, Bucharest, Romania. Demographic, clinical/laboratory parameters were extracted from the medical records (CD4 cell count, nadir CD4, status of HBV/HCV co-infection, ART).

The control group was selected from a group of healthy individuals consecutively evaluated in the Endocrinology Department of the “Carol Davila” University of Medicine and Pharmacy. All subjects were informed about the study protocol and gave their written informed consent. All procedures were done in agreement of Helsinki declaration for studies on human subjects. The Institutional Ethics committee from the “C. Davila” University of Medicine and Pharmacy approved the study protocol. The serum VD level in these subjects was evaluated as a part of a larger project aiming to better describe the VD status in the general Romanian population. None of these subjects had taken any VD supplement in the previous 6 months (those not fulfilling this criterion were excluded from the control group). The sample selection for the present study was done in order to fulfil an accurate age, sex and postmenopausal status matching between the study sample and the control group.

Plasma samples from patients and controls were collected and the serum 25OHD concentration was measured with 25-hydroxy vitamin D (25OHD) by electrochemiluminescence (Cobas e601, Roche Diagnostics, Indianapolis, IN, USA). The 25OHD status was defined as: deficiency <20 ng/mL (severe deficiency <10 ng/mL), insufficiency 20-30 ng/mL, normal >30 ng/mL. Most samples were collected during spring and summer of 2016, in both groups.

The study was approved by the Ethical Committee of our institution and all patients signed an informed consent before entering the study.

For the statistical analysis, SPSS version 17.0 was used (SPSS Inc, Chicago, IL, USA). Categorical and continuous variables were analyzed using Pearson’s Chi square and the Mann Whitney two-sided test, respectively. Variables included: age, sex, serum 25OHD levels, HIV status. Specifically for the HIV-infected cohort, we studied additional variables such as: duration (years) elapsed since HIV diagnosis, number of years of ART treatment, nadir and current CD4 cell count, ART regimen.

RESULTS

We evaluated 118 HIV-positive patients, of which 72 males and 46 females, aged 17-70 years-old (mean±SD 36.9±12.2 years). The 46 females were mostly premenopausal (38 cases) with only 8 postmenopausal HIV-positive patients.

The HIV positive patients had been treated with antiretroviral treatment for a mean period of 4 years. At the time of the study 98.14% of the patients (106 cases) were on complex ART regimens: containing nucleoside reverse transcriptase inhibitors (NRTI) 79.31% (92 cases); non-nucleoside reverse transcriptase inhibitors (NNRTI) 34.48 % (40 cases, of which 13 on efavirenz), protease inhibitors (PI) 68.96% (80 cases), integrase inhibitors (INSTI) 19.8% (23 cases). The most recent CD4 T cell mean count was 554.85±374.10. The mean nadir CD4 count was 269.08±262.2.

Of the HIV positive patients, 11 (9.3%) were coinfected with HBV or HCV. Hepatitis B virus coinfection was present in 4 cases (3.3%), hepatitis C virus in 6 cases (5.08%) and both in one case (0.84%).

The control sample consisted of 119 subjects, 74 males and 45 women, matched by age, sex and menopausal status (Table 1).

Table 1.

The main characteristics of the study and control group (*- p<0.05)

	HIV-positive cases		HIV-negative subjects		Statistical analysis
Sex distribution (M/F)	72/46		74/45		p=0.279, OR=1.3, 95%CI: 0.8-2.3
Menopausal status in women (pre/post)	38/8		37/8		p=0.320
	Median (IQR)	Range	Median (IQR)	Range
Age (years)	35 (26, 44)	17-70	34 (28, 45)	21-67	p=0.627
25OHD serum concentration (ng/mL)	17.6 (9.7, 26.9)	3.14-57.53	23.7 (18.4, 27.5)	5.6-51.2	p=0.001*

The median and interquartile range (IQR) for serum 25OHD concentration in HIV-positive patients was 17.6 (9.7, 26.9) ng/mL while in the control cases it was significantly higher: 23.7 (18.4, 27.5) ng/mL (p=0.001, r=-0.2) (Table 1 and Fig. 1).

Figure 1.

Distribution of serum 25OHD concentration in the study group and the control group.

The distribution of VD status categories (as defined in the Methods section) in the patients and control group is depicted in Figure 2. Only 15.96% of HIV positive cases and 12.71% of controls had normal VD status (not statistically significant). The percentage of cases with severe VD deficiency was significantly higher in HIV positive cases (23.52%) compared to HIV-negative controls (4.2%, p=0.001) but no significant differences were noted for the other categories.

Figure 2.

The VD status in the study group compared with the control group.

In HIV positive patients the serum 25OHD level showed no significant correlation with gender, age, menopausal status in women, number of years of previous ART treatment, nadir or current CD4 cell count (results not shown).

However, 25OHD serum levels were significantly higher in patients under 30 years-old (21.3±20.07 ng/mL) compared to those over 30 years-old (15.2±15.5ng/mL; p=0.019).

In the study group 13 patients (9 males and 4 females) were currently treated with efavirenz. In these cases the serum 25OHD level was not significantly different from that of patients not taking the drug.

DISCUSSION

The prevalence of HIV infection is increasing worldwide and due to the advent of effective ART and increased awareness the survival of HIV-positive patients has improved (15). Currently most patients are offered combination regimens including NRTIs as well as INSTI or NNRTIs or PIs, according to the most recent guidelines (16).

The interest in the VD status of HIV-positive patients has recently increased, since VD has been viewed as an immunomodulatory and antiinfective agent. VD deficiency may associate macrophage dysfunction, low CD4, decreased immune activity and infections progression (including that of HIV) (17,18). Low VD status in HIV-positive patients has been associated with earlier onset osteopenia and osteoporosis (19-21) and potentially with a faster progression of HIV infection, even in those on successful ART (22, 23). Also vertebral fractures occur frequently in HIV positive patients and are associated with low vitamin D status (24).

Vitamin D deficiency is commonly encountered in the general population worldwide (8) and also in HIV-infected persons (10, 25) but rates vary across studies (possibly due to different cut-offs used, heterogeneous populations, different geographical location). Prevalences of 58-86% for VD deficiency have been reported in European HIV-infected patients (7, 26-27) and 70% in a study performed in the United States. (9) In our study 84.04% of the HIV-infected patients had hypovitaminosis D. The levels were higher in younger patients compared to those over 30 years-old. This is in line with data observed in the general population showing a decline of 25OHD levels with age (28).

However, data regarding an increased prevalence of VD deficiency in HIV-positive subjects compared to controls are conflicting (9, 11). For example in the Study to Understand the Natural History of HIV and AIDS in the Era of Effective Therapy (SUN Study) HIV-infected patients actually had slightly lower prevalence of VD deficiency compared to the matched sample from the general population (9).

According to the studies performed to date, the prevalence of VD deficiency in Romania is very high, 83-91% (13, 14).

There is also a very significant seasonal variation in serum VD levels in Romania (29); its influence on our results was minimised by sampling all subjects during spring and summer, the proportion of subjects in the study and control groups evaluated in each month being similar.

In the current study, the prevalence of low VD status among controls was 87.29%, not significantly higher than that observed among HIV-infected patients, 84.04%. However, in our study on a homogeneous Caucasian population, HIV-positive patients had a higher prevalence of severe VD deficiency and lower mean serum concentrations of 25OHD compared to age- and sex-matched healthy individuals.

These findings are important because extremely low serum VD levels could be associated with deleterious consequences in an already fragile population. VD can modulate the immune response and deficient levels have been correlated, among numerous other consequences, with increased susceptibility to infection (30).

Although it is not universally accepted that serum levels of VD are lower in HIV infection than in the general population there are certain possible reasons supporting the validity of our conclusions. These may include dietary deficiency (worsened by possible HIV-related gastrointestinal disturbances e.g., opportunistic infections, malabsorption), lack of sun exposure (possibly more than in the general population as the numerous comorbidities usually lead to a more sedentary lifestyle) (31). Some ART drugs have also been associated with alterations in vitamin D levels. For instance, efavirenz induces enzymes responsible for the inactivation of VD and its use has been linked to lower 25OHD serum levels (7). In our study the impact of specific antiretroviral drugs was assessed but due to the wide variability of drugs used according to the current guidelines (16), the number of patients taking a particular drug was relatively small. This might explain why there was no significant difference between the serum 25OHD levels in HIV positive patients receiving efavirenz compared to those not treated with this drug. The impact of the different ART drugs on the VD status needs to be further studied on larger patient groups in our country, focusing on the former pediatric cohort of HIV infected patients, infected in their early childhood, and which are now adults with ages between 25 and 29 years old, and with a long exposure to ART.

Overall HIV infected patients appear to be at increased risk of developing VD deficiency and particularly severe VD deficiency, as our data suggest. The mechanisms underlying low VD in HIV infection are complex (increased prevalence of risk factors, comorbid conditions, certain HIV-associated factors) (31).

Optimal detection and management protocols in these patients are needed. Clinical trial data assessing the optimal replacement dose in HIV-infected patients are not available. Target levels recommended in general population are therefore to be used until more specific data are released for this particular population. Thus a serum 25OHD level above 30 ng/mL is recommended (requiring at least 1500-2000 IU of VD supplements) (3).

In conclusion, HIV-infected patients have lower serum 25OHD levels compared to controls and are at increased risk of severe vitamin D deficiency and, potentially, its consequences. Despite the lack of clear HIV-specific clinical guidelines, the determination of the VD status and/or routine VD supplementation at least in certain high-risk HIV positive cases might be useful.

Conflict of interest

The authors declare that they have no conflict of interest.

Acknowledgement

This work was supported by the grant “Phenotypic skeletal and extraskeletal features of vitamin D deficiency in HIV / AIDS-infected patients” – received by Catalina Poiana MD, PhD, in the frame of the “Paraclinic Surrogate Investigations in Evaluation and Treatment of Patients with Osteo-Renal Metabolic Disorders” competition. Authority Contractor: European Academy of HIV / AIDS and Infectious Diseases, Contract no. 5597 / 07.12.2013