Replacement of Vitamin D in Patients With Active Tuberculosis

Tuberculosis Tuberculosis (TB) is a global emergency. In 2004, there were an estimated 8.9 million new cases and 1.7 million deaths due to the disease1. In Pakistan, pulmonary tuberculosis (TB) is a major public health problem. The population of Pakistan is 160,943,000 with an estimated prevalence of all cases of tuberculosis at 263/100,000, an incidence of smear positive pulmonary TB of 181/100,000, and a mortality of 34/100,0002.

The management of TB remains a complicated issue even with the availability of effective anti-tuberculous drugs. One important aspect of this issue has been the difficulty clinicians' face in measuring clinical response in these patients, especially in smear-negative pulmonary and extra-pulmonary disease. Christian et al 3. have recently developed and validated a simple clinical score for the evaluation of TB patients. The score is based on eleven clinical variables, which were obtained from the WHO clinical manual for TB. These included self-reported parameters (e.g., cough, hemoptysis, dyspnoea, chest pain and night sweats), and signs comprising of anemic conjunctivae, tachycardia, positive findings at lung auscultation, axillary temperature >37.0˚C, body mass index (BMI) and middle upper arm circumference (MUAC). Each variable contributes 1 point, except for BMI and MUAC which contribute 2 points each, giving a maximum score of 13. Using this, score patients can be divided into 3 severity classes, providing an easy tool for clinical follow up and prognosis.

Vitamin D Vitamin D is normally synthesized in the skin under the influence of sunlight or ingested, It is readily metabo-lized in the liver to form 25-hydroxyvitamin D (25(OH)D), the accepted measure of vitamin D status. 25(OH)D is then further metabolized by the 1 hydroxylase enzyme to its biologically active metabolite, 1 , 25 dihy-droxyvitamin D (1 ,25[OH]2D) 4,5 . Activated vitamin D circulates with a binding protein and enters the target cell to interact with its nuclear receptor. This complex then combines with the retinoic acid X receptor to form a heterodimer, which in turn interacts with the Vitamin D response element on the target gene. By increasing in-testinal calcium and phosphate absorption6, increasing renal calcium reabsorption, and enhancing PTH-mediated bone resorption (via its effect on RANKL), vitamin D has the net effect of increasing the serum cal-cium and phosphate concentrations. Additionally, the vitamin D receptor element is present on multiple genes, the vitamin D receptor in many organs, and 1-alpha hydroxylase activity occurs in extrarenal tissues7. These local tissue effects are responsible for the non-mineral-related effects of vitamin D; such as cell differentiation and proliferation and immune regulation8.

Vitamin D deficiency The two most common causes of vitamin D deficiency remains decreased intake of vitamin D containing food and reduced sun exposure. The major source of vitamin D is exposure to sunlight, anything that diminishes the transmission or penetration of solar UVB into the skin will affect the cutaneous synthesis of vitamin D3. Vita-min D deficiency can therefore occur in people who live without sun exposure (including those whose skin is constantly protected from the sun by protective clothing, indoor life styles and sunscreen use.). In addition, very few foods naturally contain vitamin D and foods that are fortified with with vitamin D are often inadequate to satisfy the vitamin D requirements. Serum levels of less than 20 ng/mL (50 nmol/L) of 25-OH vitamin D are a commonly accepted cutoff for vitamin D deficiency , whereas levels between 20 and 30 ng/mL (50 to 75 nmol/L), are considered insufficiency9,10.

Asymptomatic or minimally symptomatic vitamin D deficiency is increasingly being recognized around the world. In Pakistan, a number of small studies have identified a high prevalence of vitamin D deficiency in pa-tient populations that include those presenting with hip fractures, obstetric cases and those attending ambulatory care clinics. Risk factors include a history of poor nutrition, lack of exposure to sunlight, and a low socioeco-nomic status. Zuberi LM et al11 reported that 92% of outpatients presenting to the AKUH, were vitamin D defi-cient; 62% had severe, 24% moderate and 8% had mild deficiency. Almost half of all these patients (including those with severe deficiency) were asymptomatic. Two separate studies from our center involving healthy, as-ymptomatic volunteers have identified a high prevalence of vitamin D deficiency (unpublished data). Baig MA et al12, in a study on outpatients from two public hospitals in Karachi, identified that 92% patients were vitamin D deficient and that the most severe form of D deficiency was seen in patients with tuberculosis.

Vitamin D Replacement Nutritional deficiency (25OHD <20 ng/ml [50 nmol/L]) requires treatment with 600,000 IU/month or 50,000 IU/week) 13. In this study we plan to give 600,000 IU/month for 2 months. Repeat testing for vitamin D will be performed at 8 weeks. If the levels are <30ng/ml then treatment will be repeated for further 2 months.

Vitamin D replacement is generally very safe. Toxicity, with the development of hypercalciuria and hypercal-cemia occurs only at 25OHD levels above 88 ng/ml (220 nmol/L) 14,15 . In 1997, the National Academy of Sci-ences defined the Safe Upper Limit for vitamin D as 2000 U/day14 but newer data however indicate that higher doses are safe at least over a several-month period16.

The benefits of vitamin D replacement occur gradually; in one study, skeletal effects were seen upto 10 months after a 5 week course of therapy17.

Vitamin D and Tuberculosis Clinical studies suggest that vitamin D enhances antimycobacterial immunity, and that deficiency is associated with susceptibility to active disease18,19. High doses of vitamin D were widely used to treat active TB in the pre-antibiotic era. More recently, case-control studies have demonstrated that a vegetarian diet (low in vitamin D) is an independent risk factor for active TB in South Asians18 and that patients with TB who are of Gujarati Hindu ethnic origin have significantly higher rates of vitamin D deficiency than ethnically matched tuberculin-positive TB contacts. Similarly, Gibney et al, documented a gradation of vitamin D levels in African immigrants to Aus-tralia, with patients with latent TB having lower levels as compared to controls, while patients with active or past history of active TB had levels lower than those with latent infection20. A recent systematic review and meta-analysis 21explored the association between low serum vitamin D and risk of active tuberculosis and con-cluded that patients with tuberculosis have lower serum levels of vitamin D than healthy controls when matched for sex, age, ethnicity, diet and geographical location.

1 ,25(OH)2D has no direct antimycobacterial action, but it does induce antituberculous activity in vitro in both monocytes and macrophages22.Several mechanisms of action have been proposed. Exogenous 1 ,25(OH)2D in-duces a superoxide burst23 and enhances phagolysosome fusion in Mycobacterium tuberculosis-infected macrophages; both phenomena are mediated by phosphatidylinositol 3-kinase, suggesting that this response is initiated by binding membrane vitamin D receptor (VDR) 24.1 ,25(OH)2D also modulates immune responses by binding nuclear VDR, where it up-regulates protective innate host responses, including induction of nitric oxide synthase25.Recently, 25(OH)D has also been shown to support messenger RNA induction of the antimicrobial peptide cathelicidin LL-37, which possesses antituberculous activity26.

Although the association of vitamin D deficiency and tuberculosis has been documented in epidemiological and clinical studies, a causal role is yet to be verified. Interestingly isoniazid and rifampicin therapy has been shown to decrease vitamin D levels, raising the question of whether low vitamin D levels could be a consequence of disease27. Additionally, the affect of replacing vitamin D in the prevention and treatment of tuberculosis has not been studied in a systematic manner. There is, therefore, a need for a randomized controlled trial to evaluate the effect of vitamin D supplementation on antimycobacterial host response and clinical recovery.

Immune responses to Mycobacterium tuberculosis Restriction of infection by M. tuberculosis is dependent on effective macrophage activation which is facilitated by the recruitment of leucocytes to the site of infection. The protective role of CD4+, IFN-γ producing T cells has been confirmed in both acute and chronic mycobacterial infections28. Macrophage TNFα plays a critical role in granuloma formation and in restriction of disease related pathology29. TNF-α and IFN-γ activate granu-loma formation by C-C family of chemokines, CCL2, CCL3, CCL4, CCL5 and the C-X-C chemokine family; CXCL8, CXCL9, CXCL10 and CXCL1230 .

Chemokines play a critical role in determining both activation and migration patterns of circulating monocytes in the blood. CXCL9 is an early predictive marker for IFNγ secreting cells, and is found to be increased in re-sponse to M. tuberculosis antigen stimulation. CCL5 and IFNγ-induced CXCL10 also contribute to the granu-lomatous response31.

Previous studies have shown the utility of mycobacterial antigens such as, from RD1 antigens early secreted and activated T cell (ESAT)6 in investigating specific immune responses in individuals infected with M. tuber-culosis. Specific immune profiles have been shown to be present in TB patients which are coordinate with their disease status32,33.

A study of household contacts of TB patients showed that the dynamics of cytokine response to mycobacterial antigens driven by IL10 was central to protection against, and the development of M. tuberculosis infection, and that a peak inflammatory cytokine response at 6 months was present in those individuals who did not develop disease34 .

We have shown that circulating CXCL9 levels in TB patients differ according to severity of disease35. In addi-tion, mycobacterial antigen induced CXCL9 responses correlate with IFNγ, and that these are greater in local-ised as compared with disseminated disease36. A recent study has also shown the relevance of using both CCL2 (MCP-2) and CXCL10 (IP-10) as biomarkers for tuberculosis37,38 . In this study we hope to determine immune parameters in patients prior to and post-vitamin D treatment.