Volume 15 Issue 06, June 2025
Impact factor: 2019: 4.679 2020: 5.015 2021: 5.436, 2022: 5.242, 2023:
6.995, 2024 7.75
http://www.internationaljournal.co.in/index.php/jasass
567
PEDIATRIC TUBERCULOSIS: DIAGNOSTIC CHALLENGES AND FUTURE
STRATEGIES
5rd year student, Faculty of Pediatrics Samarkand State Medical University
Tulayev Bobur Zoyir ugli
3rd-year student, Faculty of General Medicine Samarkand State Medical University
Pardayeva Azizabonu Ulug'bek kizi
4th-year students, Faculty of General Medicine
Samarkand State Medical University
Asadbek G'aniyev Ulug'bekovich
Eshmuradov Quyoshbek Sanjar ugli
4th-year student, Faculty of Pediatrics Samarkand State Medical University
Beknazarov Shahobiddin Fazliddin ugli
Abstract:
This comprehensive review delves into the persistent diagnostic barriers inherent in
pediatric tuberculosis (TB), especially in high-burden, resource-limited countries such as
Uzbekistan. Highlighting the clinical ambiguity caused by paucibacillary disease and the
difficulty in specimen collection, the review covers conventional methods—including clinical
scoring, chest radiography, tuberculin skin test (TST), interferon-gamma release assays (IGRAs),
GeneXpert MTB/RIF and Ultra—as well as innovative advances like stool-based assays, blood-
based host-response tests (e.g., Xpert MTB‑HR), and AI-enhanced chest X-ray interpretation.
We also discuss digital decision-support systems, active case-finding, and contact-tracing
strategies. Country-specific data underscore significant underdiagnosis, with up to 58% of cases
missed in children aged 0–4 . The review advocates for integrated diagnostic algorithms
supported by modern tools and stronger health systems, with research priorities focused on cost-
effectiveness, technology adaptation, and tailored pediatric solutions.
Keywords:
pediatric tuberculosis, paucibacillary, GeneXpert MTB/RIF Ultra,Xpert MTB-HR
host-response, stool‑based Xpert, AI‑based chest X‑ray, clinical scoring algorithms ,active case
finding.,digital decision support, Uzbekistan burden
Introduction:
Tuberculosis in children remains a leading cause of mortality, with approximately
1.2 million pediatric cases and 226,000 deaths globally in 2023 . Paucibacillary infection, non-
specific symptoms like fever and cough, and limited laboratory capacity make diagnosis
challenging . In Uzbekistan (~57/100 000 incidence; ~3,200 pediatric cases in 2020), high
MDR‑TB prevalence exacerbates detection issues . Underdetection is especially acute in
children under 5, accounting for ~58% of missed cases .
Volume 15 Issue 06, June 2025
Impact factor: 2019: 4.679 2020: 5.015 2021: 5.436, 2022: 5.242, 2023:
6.995, 2024 7.75
http://www.internationaljournal.co.in/index.php/jasass
568
2. Methods
Literature search: PubMed, WHO, MDPI, Scopus, Google Scholar, ResearchGate; English-
language studies from the past decade, focusing on pediatric TB diagnosis in resource-limited
settings.
Inclusion criteria: Diagnostic performance, field implementation, technological innovations,
modeling of missed cases.
Data triangulation: Supplemented by WHO reports and country-specific program data.
Results
3.1 Core Diagnostic Challenges
Low bacillary burden & sample issues: Microbiological confirmation in children <30%, even
with induced sputum .Symptom overlap: Fever, weight loss, cough often mirror other pediatric
illnesses .
Limited infrastructure: Centralized labs/platforms (GeneXpert, culture) inaccessible in many
rural areas .
3.2 Traditional Diagnostics Revisited
Clinical algorithms: Provide structured guidance but vary widely in sensitivity and specificity;
risk of both under‑ and over‑diagnosis .TST and IGRA: Helpful for LTBI detection; insufficient
to confirm active disease .Chest X-ray: Readings subjective and require radiologist training;
implementation aided by CAD/Ai tools .
Xpert MTB/RIF Ultra & Urine LAM: Ultra improves sensitivity (~80–89%) in children; LAM
suitable for HIV-positive children but variable sensitivity (13–93%) .
3.3 Emerging Technologies
Stool‑based Xpert: Easier to collect; implemented in Vietnam/Tanzania, accounting for ~37% of
pediatric TB testing .
Host-response cartridge (Xpert MTB‑HR): Detects 3-gene blood signature; AUC 0.85–0.89,
sensitivity ~60–90%; performs better in confirmed vs unconfirmed cases .AI-enhanced CXR:
Self-supervised ViT models achieved AUC ~0.70 in pediatric TB detection . Commercial CAD
tools (e.g., qXR) outperform human radiologists in adults, likely adaptable for children with
further training .
Decision-support systems (CDSS): Pilot systems in Philippines use block‑chain and rule‑based
algorithms to assist frontline health workers .
3.4 Uzbekistan-Specific Data
Volume 15 Issue 06, June 2025
Impact factor: 2019: 4.679 2020: 5.015 2021: 5.436, 2022: 5.242, 2023:
6.995, 2024 7.75
http://www.internationaljournal.co.in/index.php/jasass
569
MDR‑TB burden ranks Uzbekistan high in region; national survey launched mid‑2023 .
Contact tracing remains low (<5 contacts/case), delaying pediatric detection .Digital learning
and AI-supported CXR initiatives piloted in Tashkent provinces—evaluation ongoing.
Discussion
:Diagnostic gap: Globally <50% pediatric TB cases detected; in Uzbekistan lower
still, especially among <5 age group .Multimodal strategy: Combining stool-Xpert, blood host-
assays, AI‑supported CXR reading, and decision‑support can triangulate results and compensate
for individual limitations.
Implementation barriers: Infrastructure deficits, cost constraints, lab referral complexities,
training needs, and data-sharing hurdles.Health system interventions: Scale up GeneXpert Ultra
and MTB‑HR tools; train community workers in sample collection (stool, capillary blood);
expand contact tracing via mobile-health platforms.Research agenda: Validate cost-effectiveness
and diagnostic value of integrated algorithms, specifically for Uzbekistan’s epidemiology and
health systems.
Conclusion:
Pediatric TB diagnosis in Uzbekistan needs urgent prioritization. Multi-biological,
multi-tech diagnostic cascades—a combination of stool Ultra, blood host-response assays, AI-
based CXR, and CDSS—should be piloted. Health system strengthening in lab access, training,
and digital infrastructure is equally vital. Future studies should focus on algorithm performance,
economic viability, and adaptation to local workflows.
References
1. Comprehensive diagnostics review. Microorganisms. 2025
2. Cepheid MTB‑HR diagnostic accuracy in children. Lancet Infect Dis. 2023
3. Three-gene host response evaluation. JPIDS. May 2025
4. Stool-based diagnostic scale-up. IDDS Fact Sheet, 2024
5. Urine LAM review in children. BMC Pediatrics. 2024
6. AI for pediatric CXR TB. arXiv. 2024
7. Host-response assay overview. CID. 2021
8. Missed opportunities in pediatric TB. Scoping review 2024
9. ViT self-supervised CXR detection. arXiv. 2024
