25-Hydroxyvitamin D levels in newly diagnosed rifampicin resistant and non-resistant pulmonary tuberculosis patients and their household contacts in Peshawar, Pakistan: a comparative cross-sectional study

Introduction: Tuberculosis (TB) is an emerging epidemic globally and drug resistance is on the rise. Low vitamin D has been associated with active TB, but this association in rifampicin resistant TB remain uncertain. Objectives: To determine serum 25-hydroxyvitamin D levels in newly diagnosed rifampicin resistant tuberculosis (RR-TB) patients, non-resistant pulmonary tuberculosis (NR-TB) patients and their household contacts (HC) in Peshawar, Pakistan. Methodology: A comparative cross-sectional study was conducted across multiple healthcare facilities in Peshawar, Pakistan. Serum samples from sputum smear/ GeneXpert positive NR-TB patients (n=29), RR-TB patients (n=26) and their household contacts (n=27) were obtained. Serum 25(OH)-vitamin D levels were determined by ELISA. Data were analyzed using SPSS version 25 with statistical significance set at p < 0.05. Results: Median age of HC, NR-TB patients and RR-TB patients were 31.89 years, 32.72 years and 33.0 years, respectively. Smoking history was positive in 11.1% of HC, 3.4% of NR-TB patients and 15.4% RR-TB patients. Median BMI of HC, NR-TB and RR-TB patients was 21.515 kg/m2, 20.85 kg/m2 and 20.373 kg/m2, respectively. About 3.7% of HC and 13.8% of TB patients were underweight. According to study predefined thresholds, vitamin D deficiency was higher in RR-TB patients (84.6%) than in NR-TB patients (75.9%) and HC (59.3%), respectively. Median serum 25(OH)-D levels of RR-TB patients (13.14 ng/ml) was significantly lower than NR-TB patients (22.35 ng/ml) and HC (29.93 ng/ml). After Bonferroni adjustment, RR-TB versus HC comparison remained statistically significant (adjusted p<0.05), whereas RR-TB versus NR-TB (adjusted p=0.12) and NR-TB versus HC (adjusted p=0.16) were non-significant. Conclusion: Lower serum 25(OH)D were observed among TB patients with lowest levels in RR-TB patients, and the RR-TB versus HC comparison remained statistically significant after multiple-comparison adjustment. Because this cross-sectional study cannot establish temporality or causality, the findings should be interpreted as associations and require confirmation in larger longitudinal and interventional studies.