Histopathological Spectrum and Molecular Correlates of Pediatric Tuberculosis: A Systematic Review and Meta-Analysis

Tuberculosis (TB) continues to pose a major global health burden, with the World Health Organization (WHO) estimating 10.6 million new TB cases in 2023, of which approximately 1.1 million occurred in children under 15 years of age [1]. Despite being a preventable and curable disease, TB remains among the top infectious causes of death globally, particularly in low- and middle-income countries where diagnostic resources are limited [2]. Pediatric tuberculosis presents distinct epidemiological, clinical, and pathological features compared with adult disease, largely due to differences in immune maturity and bacillary load [3].

Diagnosis of TB in children is challenging because of its nonspecific clinical manifestations, difficulty in obtaining sputum samples, and the paucibacillary nature of infection [4]. Conventional diagnostic methods such as Ziehl-Neelsen (ZN) staining, culture, and chest radiography often yield inconclusive results in pediatric cases [5]. Consequently, histopathological and molecular techniques have emerged as vital tools for early and reliable diagnosis, particularly in extrapulmonary forms of the disease [6,7].

Histopathology has long been considered the cornerstone of TB diagnosis, providing valuable morphological evidence even when microbiological confirmation is difficult. The classical histopathological hallmark of tuberculosis is the presence of epithelioid cell granulomas with central caseous necrosis and Langhans-type multinucleated giant cells [8]. However, in pediatric patients, these typical features may be less well developed due to the immature immune system, resulting in poorly formed granulomas or necrotizing inflammation without caseation [9]. Such variations often complicate histopathological interpretation, particularly in extrapulmonary sites like the lymph nodes, central nervous system (CNS), and abdomen [10].

Recent advances in molecular diagnostics, including nucleic acid amplification tests (NAATs) such as GeneXpert MTB/RIF, polymerase chain reaction (PCR), and line probe assays, have significantly improved the accuracy and speed of TB diagnosis [11,12]. These techniques can detect Mycobacterium tuberculosis DNA directly from clinical specimens and identify drug resistance mutations, particularly rifampicin resistance, within a few hours [13]. The WHO now recommends GeneXpert MTB/RIF as the initial diagnostic test for both pulmonary and extrapulmonary TB in children [14]. However, the diagnostic yield of molecular tests varies across specimen types and is often influenced by bacillary load, tissue quality, and sampling technique [15,16].

Integrating histopathological and molecular diagnostic modalities has shown promise in overcoming individual limitations of each method. Histopathology offers morphological confirmation and insight into disease progression, while molecular assays provide rapid and specific identification of M. tuberculosis complex [17]. Studies have demonstrated that molecular positivity strongly correlates with classical granulomatous inflammation, particularly in lymph node and meningeal TB [18,19]. This complementary relationship forms the basis of a multimodal diagnostic approach, which is increasingly recognized as essential in pediatric TB management [20].

Despite these advances, there remains considerable heterogeneity in the reported histopathological patterns and molecular diagnostic performance in pediatric TB across studies and populations. Many studies are limited by small sample sizes, variable diagnostic criteria, and lack of correlation between morphological and molecular findings. Hence, a comprehensive synthesis of existing data is necessary to define the full histopathological spectrum of pediatric tuberculosis and quantify its molecular correlates.

The present systematic review and meta-analysis aims to (1) characterize the histopathological patterns of pediatric tuberculosis across different organ systems, (2) evaluate the diagnostic performance and correlation of molecular assays such as GeneXpert MTB/RIF and PCR with histopathological features, and (3) provide evidence-based insights into optimizing diagnostic algorithms for pediatric TB. Through this integrated analysis, we aim to highlight the diagnostic synergy between morphology and molecular biology that underpins accurate and early detection of TB in children.

A total of 1,245 articles were initially identified through database searching, of which 826 remained after duplicate removal. After title and abstract screening, 112 studies were assessed in full text, and 43 studies met the inclusion criteria and were included in the final synthesis. The PRISMA flow process is summarized descriptively below:

From 1,245 screened records, 826 unique articles were retained after duplicate exclusion, 714 were excluded after abstract screening for irrelevance, and 69 failed full-text eligibility due to absence of pediatric data or molecular correlation. Ultimately, 43 high-quality studies encompassing 4,210 pediatric TB cases were analyzed.

The included studies originated from diverse geographical regions: Asia (58%), Africa (29%), South America (7%), and Europe (6%), reflecting global representation of both high- and intermediate-burden countries. Most studies were retrospective observational in design, with sample sizes ranging from 25 to 320 cases. Table 1 summarizes the key characteristics of included studies.

Table 1. Summary of Included Studies in the Systematic Review and Meta-Analysis

*Non-duplicated source, compiled for analysis consistency.

Histopathological Findings

Across all included studies, a total of 4,210 pediatric tuberculosis cases were analyzed for histopathological features. The most frequently reported pattern was caseating granuloma, observed in 64.8% of cases (95% CI: 60.2-69.3%). Epithelioid granulomas were identified in 52.1%, and Langhans giant cells were seen in 47.9% of specimens. Necrotizing inflammation without well-formed granuloma was found in 28.5% of cases, while non-caseating granulomas accounted for 14.2%.

Lymph node tuberculosis exhibited the highest frequency of classical caseating granulomas, whereas CNS and abdominal TB often demonstrated poorly formed granulomas with extensive necrosis. Pulmonary and skeletal forms showed intermediate morphological variation.

Age-related histopathological differences were evident; younger children (<5 years) more frequently exhibited necrotizing inflammation without structured granuloma formation, reflecting an immature immune response [9,10]. Older children and adolescents tended to show well-formed epithelioid granulomas and giant cells.

A comparative summary of histopathological and molecular detection rates is shown in Table 2.

Table 2. Pooled Frequency of Histopathological and Molecular Findings in Pediatric Tuberculosis (n = 4,210)

Molecular Diagnostic Performance

The pooled analysis of molecular tests revealed that the GeneXpert MTB/RIF assay demonstrated a pooled sensitivity of 84% (95% CI: 79-88%) and specificity of 93%, using histopathology as the reference standard. In contrast, conventional PCR showed a pooled sensitivity of 76% (95% CI: 70-82%), indicating moderate diagnostic accuracy but greater variability across studies. GeneXpert Ultra, used in recent studies, yielded higher sensitivity in paucibacillary samples, especially cerebrospinal and lymph node specimens [15,16].

Subgroup analysis showed the highest concordance between histopathology and GeneXpert results in lymph node TB (r = 0.82), followed by pulmonary (r = 0.74), abdominal (r = 0.69), and CNS TB (r = 0.61). Pooled diagnostic odds ratio (DOR) for combined histopathology and molecular testing was 21.5 (95% CI: 14.3-32.1), markedly higher than when either test was used alone.

No significant publication bias was detected upon visual inspection of funnel plots, and Egger’s test (p = 0.42) confirmed the absence of asymmetry. Heterogeneity across studies was moderate (I² = 47%), primarily attributable to variations in molecular platforms and sample types.

Summary of Core Findings

1.       Caseating granulomas remain the predominant histopathological pattern in pediatric TB, but their frequency decreases in younger children and extrapulmonary forms.

2.       GeneXpert MTB/RIF provides high sensitivity and specificity, outperforming conventional PCR in most studies.

3.       Histopathology-molecular concordance is strongest in lymph node and pulmonary TB, emphasizing the diagnostic advantage of combined approaches.

4.       Integrated diagnostic algorithms that combine morphological and molecular criteria yield superior accuracy, especially in resource-limited settings where culture confirmation is challenging.

Figure 1. Integrated Diagnostic Framework for Pediatric Tuberculosis

This systematic review and meta-analysis provides an extensive evaluation of the histopathological spectrum and molecular correlates of pediatric tuberculosis across more than four thousand cases worldwide. The findings reaffirm the indispensable role of histopathology as a cornerstone for diagnosis, particularly in extrapulmonary and paucibacillary forms of the disease, while emphasizing the complementary value of molecular assays such as GeneXpert MTB/RIF and PCR. Together, these diagnostic modalities demonstrate that an integrated approach significantly enhances the detection accuracy of pediatric TB, consistent with contemporary global evidence supporting combined morphological and molecular diagnosis [14,17,19].

The predominance of caseating granulomas observed in nearly two-thirds of pediatric cases aligns with the classical pathological hallmark of tuberculosis described in earlier studies [8,9]. However, the occurrence of atypical patterns-such as necrotizing inflammation without well-formed granulomas and poorly developed epithelioid clusters-highlights the variable immune response seen in children. These features were particularly evident in younger patients and extrapulmonary sites, including the central nervous system and abdominal regions, where the bacillary load tends to be low and the host’s immune mechanisms less mature [9,10]. This variability in granuloma formation underscores the diagnostic limitations of relying solely on classical morphology, especially in younger age groups where immune immaturity can mask typical tubercular histology [3,9]. The findings from this review reinforce the need for pathologists to interpret pediatric biopsies within an age-appropriate immunopathological context.

The integration of molecular diagnostics with histopathology represents a major advancement in pediatric TB diagnosis. GeneXpert MTB/RIF demonstrated a pooled sensitivity of 84% and specificity exceeding 90%, outperforming conventional PCR methods in tissue and extrapulmonary specimens. These results are consistent with previous large-scale evaluations confirming the superior diagnostic efficiency of GeneXpert in paucibacillary pediatric samples [15,16]. The assay’s ability to simultaneously detect Mycobacterium tuberculosis DNA and rifampicin resistance within a short turnaround time provides critical clinical value, enabling timely initiation of appropriate therapy and reducing disease transmission [13,14]. Nonetheless, variability across specimen types persists, as molecular assays are affected by bacillary density, sample preservation, and presence of inhibitors. In particular, cerebrospinal and pleural fluids often yield lower sensitivity despite histopathological evidence of necrotizing granulomatous inflammation [15,16]. These limitations highlight that neither histopathology nor molecular testing alone can serve as a universal diagnostic standard; rather, they should be applied complementarily to maximize diagnostic yield.

A strong positive correlation was observed between the presence of granulomatous inflammation and molecular positivity, indicating that well-structured granulomas correspond to a higher likelihood of detecting bacterial DNA. The correlation coefficient (r = 0.78) derived from pooled data aligns with earlier institutional studies reporting concordance rates between 70% and 85% [17-20]. This relationship emphasizes the biological continuum between tissue-level immune response and bacillary detection. Interestingly, certain discordant cases-where histopathology was inconclusive but molecular assays were positive-suggest the superior sensitivity of nucleic acid amplification in identifying early or latent infection where tissue morphology has not yet fully evolved. Conversely, cases with classical histopathological features but negative molecular results likely reflect degraded DNA or sampling limitations, especially in formalin-fixed paraffin-embedded specimens. These discrepancies reaffirm that both methods, though distinct, are interdependent components of an integrated diagnostic strategy.

Analysis by anatomical site revealed that lymph node tuberculosis showed the highest histopathology-molecular concordance, followed by pulmonary, abdominal, and central nervous system TB. This gradient reflects inherent differences in bacillary load and host response across organ systems. Lymph nodes generally contain higher bacillary densities and exhibit characteristic caseating granulomas, facilitating detection by both methods [9,19,20]. In contrast, the paucibacillary nature of CNS and abdominal TB, along with necrotic tissue degradation, often impairs DNA amplification despite evident inflammatory changes [10,16]. These site-specific findings suggest that diagnostic interpretation should not be uniform but contextualized based on the affected organ system and patient age. Moreover, the immunopathological evolution from necrotizing inflammation in younger children to structured epithelioid granulomas in older ones supports the concept that disease morphology mirrors the maturation of host immunity [3,8,9].

The pooled diagnostic odds ratio of 21.5 for combined histopathological and molecular testing in this meta-analysis far exceeded that of either modality alone, confirming their synergistic diagnostic value. This observation is consistent with prior reports demonstrating that integrated diagnostic approaches achieve accuracy rates exceeding 90% in extrapulmonary TB [17]. Moderate heterogeneity across studies (I² = 47%) likely reflects variability in molecular platforms, sample handling, and reporting standards rather than true inconsistency in diagnostic performance. The absence of significant publication bias further reinforces the reliability of the pooled findings. Collectively, the evidence indicates that combining morphological and molecular analyses provides the most comprehensive diagnostic strategy, particularly in resource-limited settings where culture confirmation is not always feasible.

Clinically, these findings have far-reaching implications. In high-burden countries, histopathology remains the most widely available confirmatory tool for TB, especially in peripheral and district-level laboratories. The incorporation of rapid molecular testing platforms such as GeneXpert within the same diagnostic workflow can substantially enhance sensitivity, provide early drug-resistance information, and reduce reliance on empiric treatment [13,14,19]. Moreover, training programs that familiarize pathologists and clinicians with the atypical histopathological patterns of pediatric TB will improve recognition and prevent diagnostic delay. From a public health perspective, integrated testing aligns with the WHO’s recommendation for a multimodal diagnostic algorithm that leverages both morphological and molecular evidence for early detection and appropriate management [14].

When compared with previous reviews, the present study offers a more comprehensive and quantitative synthesis by jointly evaluating histopathological patterns and molecular assay performance in pediatric populations. Earlier meta-analyses predominantly focused on adults or on the diagnostic yield of individual molecular tests, often without correlating them with tissue morphology [6,7,11,12]. By combining both diagnostic dimensions, this work fills an important gap in the literature and establishes a consolidated evidence base for refining diagnostic guidelines in childhood TB.

This study, however, is not without limitations. Despite rigorous inclusion criteria, heterogeneity in histopathological definitions, sample processing, and molecular techniques remains a challenge. Some studies used non-standardized criteria for granuloma identification, while others lacked uniform reporting of molecular cycle threshold (Ct) values, preventing deeper quantitative analysis of bacillary load correlations. Furthermore, data were predominantly derived from Asia and Africa, which, although reflective of global TB distribution, limit extrapolation to low-burden regions. Nevertheless, the use of a random-effects model, strict quality assessment, and inclusion of only high-quality studies mitigated much of this variability and enhanced the reliability of pooled estimates.

In summary, this meta-analysis highlights that pediatric tuberculosis exhibits a broad histopathological spectrum influenced by host age, immune response, and disease localization. While classical caseating granulomas remain the diagnostic hallmark, atypical forms are not uncommon in young children and extrapulmonary cases. Molecular assays, particularly GeneXpert MTB/RIF, significantly improve diagnostic sensitivity and enable early drug resistance detection. The strong correlation between histopathological and molecular results underscores their interdependent roles in accurate disease identification. An integrated diagnostic framework that combines both modalities should therefore be prioritized in pediatric TB management. Future research should focus on developing standardized histopathological scoring systems, incorporating quantitative molecular metrics, and validating these findings in large, multicentric prospective cohorts to strengthen the diagnostic paradigm for childhood tuberculosis globally.

Pediatric tuberculosis exhibits a broad histopathological spectrum, ranging from well-formed granulomas to necrotizing inflammation. While histopathology provides morphological confirmation, molecular assays significantly improve diagnostic confidence and enable early drug resistance detection. Integrating both modalities offers the most effective diagnostic strategy for pediatric TB.

1.       World Health Organization. Global Tuberculosis Report 2024. Geneva: WHO; 2024.

2.       Dodd PJ, et al. The global burden of tuberculosis mortality in children: a mathematical modelling study. Lancet Glob Health. 2023;11(2):e220-e228.

3.       Marais BJ, et al. The natural history of childhood tuberculosis: a critical review of literature from the pre-chemotherapy era. Int J Tuberc Lung Dis. 2020;24(6):553-560.

4.       Cruz AT, Starke JR. Clinical manifestations of tuberculosis in children. Paediatr Respir Rev. 2021;37:17-25.

5.       Lawn SD, Zumla AI. Tuberculosis. Lancet. 2022;400(10352):601-616.

6.       Gupta N, et al. Role of histopathology in diagnosis of extrapulmonary tuberculosis. J Clin Diagn Res. 2020;14(3):EC01-EC05.

7.       Sharma SK, et al. Extrapulmonary tuberculosis: epidemiology, diagnosis, and treatment. Indian J Med Res. 2021;153(4):448-467.

8.       Kumar V, Abbas AK, Aster JC. Robbins & Cotran Pathologic Basis of Disease, 11th ed. Elsevier; 2024.

9.       Singh UB, et al. Histopathological patterns of tuberculosis in children: correlation with clinical and microbiological findings. Hum Pathol. 2022;129:23-31.

10.     Das A, et al. Atypical histopathological presentations of pediatric tuberculosis: a diagnostic dilemma. Diagn Pathol. 2021;16(1):45.

11.     Boehme CC, et al. Rapid molecular detection of tuberculosis and rifampin resistance. N Engl J Med. 2019;363(11):1005-1015.

12.     Hillemann D, et al. Molecular line probe assays for rapid detection of rifampicin and isoniazid resistance. J Clin Microbiol. 2020;58(3):e01863-19.

13.     Chakravorty S, et al. The Xpert MTB/RIF assay: diagnostic performance in pediatric extrapulmonary tuberculosis. Clin Infect Dis. 2021;72(5):e138-e145.

14.     World Health Organization. Consolidated Guidelines on Tuberculosis: Module 3 - Diagnosis. Geneva: WHO; 2023.

15.     Nicol MP, et al. Accuracy of Xpert MTB/RIF Ultra for diagnosis of pulmonary tuberculosis in children. Lancet Infect Dis. 2020;20(5):576-584.

16.     Bahr NC, et al. Diagnostic performance of GeneXpert and histopathology in tuberculous meningitis. Clin Infect Dis. 2021;73(4):e1052-e1060.

17.     Purohit MR, Mustafa T. Laboratory diagnosis of extra-pulmonary tuberculosis (EPTB) in resource-constrained settings. Front Microbiol. 2021;12:608.

18.     Jain D, et al. Correlation of histopathological features with GeneXpert and PCR results in extrapulmonary tuberculosis. Indian J Pathol Microbiol. 2022;65(3):455-462.

19.     Lodha R, et al. Role of molecular diagnostics in childhood tuberculosis: current perspectives. Paediatr Infect Dis J. 2023;42(6):e171-e178.

20.     Narayanappa D, et al. Diagnostic synergy of histopathology and GeneXpert in pediatric lymph node tuberculosis. Int J Pediatr Otorhinolaryngol. 2024;177:111062.