<![CDATA[Introduction: Childhood tuberculosis (TB) remains a significant global health challenge, with diagnostic difficulties compounded by nonspecific clinical presentations and limitations of available tests. The Mantoux tuberculin skin test (TST) has been a cornerstone of TB diagnosis for decades, yet its accuracy in children remains controversial due to factors including BCG vaccination, age-related immune responses, and comorbidities. This systematic review aims to comprehensively evaluate the relationship between Mantoux test sensitivity and the diagnosis of childhood tuberculosis. Methods: A systematic review was conducted following PRISMA guidelines. We screened studies based on predefined criteria including pediatric population (0-18 years), TST evaluation against reference standards, and reporting of diagnostic performance data. Seventy-five studies comprising systematic reviews, meta-analyses, diagnostic accuracy studies, cohort studies, and cross-sectional investigations were included. Data extraction encompassed study characteristics, TST protocols, reference standards, diagnostic performance metrics, comparison with IGRAs, population risk factors, and clinical context. Results: Pooled TST sensitivity for active TB in children ranged from 64% to 86% across major meta-analyses, with the highest values (82-86%) observed in microbiologically confirmed cases in high-income countries and lower values (64-74%) in low-income settings or clinically diagnosed cases. Specificity ranged from 49% in BCG-vaccinated populations to 92% in high-income settings without BCG. For latent TB infection, TST sensitivity was estimated at 82% with specificity of 98% in non-BCG and 82% in BCG-vaccinated children. Head-to-head comparisons showed IGRAs demonstrate superior specificity (85-98% vs. 56-92%), particularly in BCG-vaccinated populations, while sensitivity comparisons varied by context. Age <5 years, HIV infection, malnutrition, and helminth co-infection significantly reduced TST sensitivity. Novel antigen-specific skin tests (C-Tb, ECST, Diaskintest) showed IGRA-like specificity (94-97%) unaffected by BCG. Discussion: TST performance in children is highly context-dependent, with BCG vaccination status, age, geographic setting, and comorbidities serving as critical determinants. The apparent paradox of variable TST-IGRA comparisons is explained by population differences: TST may be more sensitive in high-burden settings where case detection is paramount, while IGRAs offer superior specificity in low-incidence, BCG-vaccinated populations. The diagnostic gap in children under 5 years—where both tests perform poorly—represents a critical challenge. Sequential testing strategies and novel skin tests offer promising alternatives. Conclusion: The Mantoux test remains a valuable tool for childhood TB diagnosis, but its limitations must be understood contextually. No single test is universally superior; test selection should be guided by local epidemiology, BCG vaccination policies, and available resources. Future research should focus on validating novel skin tests in pediatric populations and developing integrated diagnostic algorithms for young and immunocompromised children.]]>
