The sensitivity of an interferon-γ release assay in microbiologically confirmed pediatric tuberculosis.

UNLABELLED: This study aimed at determining the sensitivity of a whole blood interferon-γ release assay (IGRA) among children with microbiologically confirmed tuberculosis in a high-burden country. Children with a diagnosis of tuberculosis based on clinical and radiographic assessment were tested with an IGRA in addition to microbiologic examination of appropriate specimens for acid-fast bacilli, mycobacterial rRNA, and observation for growth of Mycobacterium tuberculosis on appropriate culture media. Of the 405 children with a clinical diagnosis of tuberculosis, 91 (22.5 %) had microbiologically confirmed tuberculosis, of whom 81 were tested with an IGRA. A positive result was obtained in 43 (sensitivity 53.1 %, 95 % confidence interval 42.3 to 63.6 %), uninfluenced by age, sex, or disease manifestation. CONCLUSIONS: The sensitivity of a whole blood interferon-γ release assay in microbiologically confirmed pediatric tuberculosis was low. An IGRA cannot, thus, be used as rule-in test, but it might be useful to rule in tuberculosis among children in whom tuberculosis is notoriously difficult to confirm microbiologically.

Non-invasive diagnosis of lung tuberculosis in children by single voxel ¹H-magnetic resonance spectroscopy.

UNLABELLED: Our previous study showed that (1)H-magnetic resonance spectroscopy ((1)H-MRS) can detect lipid peaks characteristic for Mycobacterium tuberculosis infection in cerebral lesions of young children; therefore, we aimed to extend and validate the application of (1)H-MRS for the diagnosis of active pulmonary tuberculosis lesions in three adolescent patients. Here, we document lipid peaks characteristic for M. tuberculosis infection by (1)H-MRS from lung tissue surrounding lung cavities of two patients whose sputum samples were positive for acid-fast bacilli by microscopy and positive for M. tuberculosis by genetic testing, indicating active tuberculosis. A similar lipid peak was found also in the pleural effusion of a third patient with concurrent lung cavity compatible with active tuberculosis. However, in a patient with a pyogenic pulmonary abscess, (1)H-MRS of the drained pus displayed different characteristic peaks but no lipid peak at all. CONCLUSION: Our findings further validate (1)H-MRS as a rapid, non-invasive, and specific diagnostic tool for active tuberculosis in children with microbiologically documented infection outside the central nervous system, specifically in the lungs.

The diagnosis of brain tuberculoma by (1)H-magnetic resonance spectroscopy.

UNLABELLED: Toddlers are more prone to develop severe and extrapulmonary tuberculosis (TB) than older children. This is partially explained by differences in the immune response. Early and specific diagnosis is essential to start adequate treatment, especially if the central nervous system (CNS) is involved. The lack of sputum production and inherent dangers or impossibility of sampling CNS lesions may delay diagnosis. In addition, the magnetic resonance imaging (MRI) features of TB abscesses are non-specific and may mimic abscesses of other infectious etiology. (1)H-magnetic resonance spectroscopy ((1)H-MRS) may increase specificity of diagnosis by identifying lipids within the lesions that are considered characteristic for TB. Therefore, we studied four children with presumable CNS-TB with MRI and (1)H-MRS. In vivo and in vitro (1)H-MRS showed elevated lipid peaks within the TB lesions. CONCLUSION: (1)H-MRS allows to non-invasively identifying TB with high specificity and may allow early installment of targeted antimicrobial treatment.