TB Research

Alterations and predictive potential of haematological parameters in paediatric tuberculosis and non-tuberculous mycobacterial disease.

Michaela Krivošová, Peter Kunč, Jaroslav Fábry, Juraj Mokrý, Matúš Dohál

BMC infectious diseases · 2026-02

Abstract

BACKGROUND: Tuberculosis (TB) and non-tuberculous mycobacterial (NTM) disease represent a significant health challenge in the paediatric population, with difficulties arising from non-specific clinical presentation, low bacterial load, and limited sensitivity of available diagnostic tools. These limitations highlight the need for alternative, easily accessible biomarkers to improve diagnostic accuracy and disease differentiation.

METHODS: This study retrospectively analysed blood cell parameters and their ratios from the health records of paediatric patients admitted to the Clinic of Children Tuberculosis and Respiratory Diseases in Dolný Smokovec, Slovakia between 2020 and 2024. In total, 425 patients were included and further classified according to their diagnosis to patients with TB disease (further divided to bacteriologically positive/negative), latent TB, NTM disease, and controls with documented contact with a TB-positive person/controls with symptoms.

RESULTS: This study showed significantly different counts of platelets (PLT), mean platelet volume (MPV), platelet-to-lymphocyte ratio (PLR), and lymphocyte count in TB patients compared to controls. However, the multivariable model showed insufficient predictive value (AUC = 0.67). In NTM patients, parameters PLT, lymphocytes, and red blood cell count were able to discriminate patients from controls with AUC of 0.84.

CONCLUSIONS: Haematological parameters did not show sufficient discriminative power in paediatric patients with TB disease, while in NTM patients, selected parameters showed promising discrimination, requiring validation in larger cohorts. This study confirmed that PLT and related markers show essential role in both diseases and should be further investigated. A panel of markers could provide greater predictive utility in future.

CLINICAL TRIAL: Not applicable.

SUPPLEMENTARY INFORMATION: The online version contains supplementary material available at 10.1186/s12879-026-12864-z.