Vitamin D and activated vitamin D in tuberculosis in equatorial Malaysia: a prospective clinical study

Abstract

Background: Vitamin D deficiency (low plasma 25-hydroxyvitamin D [25D] concentration) is often reported in tuberculosis. Adjunctive vitamin D has been tested for its potential to improve treatment outcomes, but has proven largely ineffective. To better understand vitamin D in tuberculosis, we investigated determinants of 25D and its immunologically active form, 1,25-dihydroxyvitamin D (1,25D), their inter-relationship in tuberculosis, longitudinal changes and association with outcome. Methods: In a prospective observational study of adults with smear-positive pulmonary tuberculosis in Sabah, Malaysia, we measured serial 25D, 1,25D, vitamin D-binding protein (VDBP), albumin, calcium, parathyroid hormone, chest x-ray, week 8 sputum smear/culture and end-of-treatment outcome. Healthy control subjects were enrolled for comparison. Results: 1,25D was elevated in 172 adults with tuberculosis (mean 229.0 pmol/L, 95% confidence interval: 215.4 - 242.6) compared with 95 controls (153.9, 138.4-169.4, p < 0.001), directly proportional to radiological severity (p < 0.001), and fell rapidly within one week of treatment commencement. Tuberculosis patients with higher baseline 1,25D achieved significantly higher percentage weight gain over time, including when controlling for baseline weight, however persistently elevated 1,25D was associated with worse residual x-ray changes and lower end-oftreatment BMI. 1,25D was inversely associated with PTH (p < 0.001), consistent with the extra-renal origin of the 1,25D. 25D did not differ between tuberculosis patients (mean 63.9 nmol/L, 95% CI: 60.6 - 67.3) and controls (61.3, 57.2- 65.3, p = 0.24), and was unassociated with outcomes. Among tuberculosis patients in multivariable analyses, sex, age and VDBP were associated with 25D, and age and albumin with 1,25D. 1,25-dihydroxyvitamin was not significantly asscociated with 25D. Vitamin D deficiency <25 nmol/L was uncommon, occurring in only five TB patients; 1,25D was elevated in three of them. Conclusions: In an equatorial setting, high extra-renal production of 1,25D was seen in tuberculosis, including in individuals with 25D in the deficient range; however, severe 25D deficiency was uncommon. Baseline elevation of 1,25D, a marker of macrophage activation, was associated with better weight gain but persistent elevation of 1,25D was associated with worse radiological and BMI outcomes. 1,25D warrants testing in larger datasets including TB patients less responsive to treatment, such as multi-drug resistant TB, to test its utility as a marker of tuberculosis severity and treatment response.