SummaryObjective 25-Hydroxyvitamin D [25(OH)D] levels after recovery from tuberculosis (TB) may reflect pre-morbid levels and therefore provide insight into pathogenesis. We assessed 25(OH)D levels after recovery from TB disease, and compared to levels in persons without TB disease. Methods Case-control study. Cases were persons who had recovered from culture-confirmed Mycobacterium tuberculosis disease. Controls were persons without TB disease. Total 25(OH)D was measured from stored plasma specimens using liquid chromatography-mass spectrometry. Results 29 persons with prior TB disease and 36 controls were included. Median 25(OH)D levels were 24.7 ng/mL (IQR, 18.3-34.1) in prior TB disease, and 33.6 ng/mL (IQR, 26.2-42.4) in controls (Mann-Whitney; P = 0.01). Multivariable linear regression analysis showed that black race (adjusted mean difference [β] = -8.3 ng/mL; 95% CI -14.5, -2.2; P < 0.01), enrollment in winter (β = -10.4 ng/mL; 95% CI -17.0, -3.8; P < 0.01) and prior TB disease (β = -5.8 ng/mL; 95% CI -11.4, -0.3; P = 0.05) were associated with lower 25(OH)D levels. Conclusions Persons who had recovered from TB disease had lower 25(OH)D levels compared to controls without TB disease, after adjusting for important confounders. Larger, longitudinal studies are needed to further characterize the possible role of low 25(OH)D in the pathogenesis of TB disease and TB recurrence after recovery. © 2013 Elsevier Ltd. All rights reserved.