Neurotuberculosis continues to represent a major pediatric health challenge despite significant advancements in tuberculosis (TB) control globally. 1 After HIV/AIDS, TB ranks as the second leading infectious cause of death worldwide. India carries one of the highest TB burdens, contributing approximately 20 percent of new TB cases annually. Childhood TB accounts for nearly 10 to 20 percent of all reported TB cases, and an estimated 8 to 20 percent of TB-related deaths in high-burden regions involve children.2 Within pediatric TB, extrapulmonary forms constitute nearly one-quarter of cases, among which neurotuberculosis is the most severe and life-threatening manifestation. Neurotuberculosis affects the central nervous system (CNS) in approximately 5 to 10 percent of TB cases and occurs more frequently in young children below four to five years of age, primarily due to their immature immunity, which allows easier dissemination of M. tuberculosis from the primary pulmonary focus3. Tuberculous meningitis (TBM) is the most common presentation of neurotuberculosis and is associated with high mortality rates ranging from 15 to 32 percent and significant long-term neurological sequelae in more than half of survivors. In addition to TBM, neurotuberculosis may present as tubercular encephalopathy, characterized by diffuse cerebral edema, demyelination, and microvascular changes leading to rapid onset of neurological deterioration4. Spinal involvement, although rare at less than 1 percent of TB cases, may occur as Pott’s spine or intramedullary tuberculomas. Vertebral destruction and paraspinal abscess formation can cause spinal cord compression leading to paraplegia, particularly in developing countries where diagnosis and treatment may be delayed. The diagnosis of neurotuberculosis is challenging due to its subacute clinical course and the paucibacillary nature of CSF samples, which limits sensitivity of conventional diagnostic methods such as smear microscopy and culture5,6. Culture methods are considered the standard for microbiological confirmation, yet they require several weeks for organism growth and yield limited results in children. Therefore, clinicians frequently rely on a combination of clinical suspicion, CSF findings, and neuroimaging to initiate empirical therapy promptly, given the consequences of delayed treatment. Recent diagnostic innovations have improved early detection.7 The real-time PCR-based GeneXpert MTB/RIF assay enables simultaneous detection of M. tuberculosis DNA and rifampicin resistance within two hours. The World Health Organization has endorsed its use for extrapulmonary TB, including CSF testing, owing to its high specificity and superior sensitivity to smear microscopy. However, data in pediatric neurotuberculosis remain limited8,9. The present study aims to evaluate the diagnostic accuracy of the GeneXpert test in identifying M. tuberculosis in children with suspected neurotuberculosis and correlate the results with CSF biochemical parameters. The findings are expected to guide improved diagnostic strategies and early therapeutic intervention in high-burden settings10. AIM To study the efficacy of geneXpert MTB/RIF assay in cerebrospinal fluid for diagnosis of neurotuberculosis.

The present hospital-based prospective observational study was conducted in the Department of Paediatrics at PBM Hospital, S.P. Medical College, Bikaner, over a duration of one year. All children aged between 6 months and 14 years who were clinically suspected of having neurotuberculosis and whose parents or guardians provided written informed consent were included in the study. Patients who did not provide consent, children younger than 6 months or older than 14 years, and those with immunocompromised status were excluded.

Table 1. Age distribution Age groups Number of patients Percentage 0- <5 years 5 22.7% 5-10 years 3 13.7% >10 years 14 63.6% Total 22 100% In this study we recruited 22 cases. The age of cases ranged between 6 months to 13 years. The most commonly involved age group was over 10 years with 63.6% cases, followed by 0 to <5 years with 22.7% cases and 5-10 years with 13.7% cases. Table 2: Nutritional status according to the WHO classification Nutritional Status Number of patients Percentage Normal 11 50.0% Mild to Moderate Malnutrition 9 40.9% Severe Malnutrition 2 9.1% Total 22 100% As far as the nutritional status of our cases was concerned, a majority had normal nutritional status with 50% cases, followed to mild to moderate malnutrition in 40.9% and severe malnutrition in 9.1% cases. Figure 1, 2: Clinical symptomatology and Clinical signs All patients had multiple symptoms. The most common symptom was fever seen in 72.7% cases, followed by vomiting in 63.6%, headacne in 22.7% and neck pain in 4.5% cases. All patients had multiple signs. The most common clinical sign was convulsions seen in 68.2% cases, followed by abnormal behavior in 18.1%, neurological deficit in 18.1%, and loss of consciousness in 4.5% cases. Table 3. BCG vaccination status BCG Vaccination Number of patients Percentage Present 4 18.2% Absent 18 81.8% Total 22 100% Only 18.2% cases had received BCG vaccination, while the remaining 81.8% had not received BCG vaccination so far. Among children older than 3 years, sensory and cerebellar functions were assessed in 18 cases each, with 77.8% showing normal function and 22.2% displaying abnormalities. Signs of raised intracranial pressure were observed in 72.7% of cases, while 27.3% showed no such signs. Table 4 Tubercular meningitis (TBM) staging Stage of TBM Number of patients Percentage I 9 40.9% II 11 50.0% III 2 9.1% Total 22 100% 40.9% had stage I tubercular meningitis, 50% had stage II tubercular meningitis and 9.1% had stage III tubercular meningitis. Table 5. Gene Xpert MTB/RIF assay status Gene Xpert MTB/RIF assay Number of patients Percentage Positive 2 9.1% Negative 20 90.9% Total 22 100% Only 2 cases (9.1%) were Gene Xpert MTB/RIF assay positive, while the remaining 90.9% were negative. On MRI assessment, hydrocephalus was present in 22.7%, meningeal enhancement in 9.1% cases, tuberculoma in 9.1% cases, vasculitis in 9.1%, ventriculitis in 4.5% and encephalitis in 4.5%. While in the remaining 40.9% cases no abnormality on MRI was detected. Table 6. CSF findings versus Gene Xpert MTB/RIF assay status CSF finding Gene Xpert MTB/RIF assay positive (n=2) Xpert MTB/RIF assay negative (n=20) P value Mean±SD Mean±SD Cell count 182±25.7 96.6±55.8 0.001 Lymphocyte % 90±14.1 61.7±42.7 0.08 Protein 107±51.4 156.2±63.9 0.04 Sugar 37.5±16.3 58.9±23.2 0.02 CSF findings revealed that gene-positive patients had significantly higher cell counts and CSF sugar levels but significantly lower protein levels when compared with gene-negative patients. Although lymphocyte percentage was higher in gene-positive cases, the difference was not statistically significant. Table 7.Neurological abnormality versus Gene Xpert MTB/RIF assay status Gene Xpert MTB/RIF assay positive (n=2) Xpert MTB/RIF assay negative (n=20) P value* Neurological signs Present 2 15 0.12 Absent 0 5 While cross tabulating the neurological abnormalities against the status of Gene Xpert MTB/RIF assay, we found that neurological signs, were equally distributed between gene positive and gene negative cases. Table 8.Mantoux test versus Gene Xpert MTB/RIF assay status Mantoux test Gene Xpert MTB/RIF assay positive (n=2) Xpert MTB/RIF assay negative (n=20) P value* Positive 2 6 0.21 Negative 0 14 While cross tabulating the status of Mantoux test with the status of Gene Xpert MTB/RIF assay, we found that statistically there was no difference between gene positive and gene negative cases.

A total of 22 cases were recruited for the study, with the majority of cases being over the age of 5 years (63.6%) and the mean age of the cases was 6.4±3.4 years. As far as the nutritional status of our cases was concerned, a majority had normal nutritional status with 50% cases, followed by mild to moderate malnutrition in 40.9% and severe malnutrition in 9.1% cases. Similar reports on malnutrition and pediatric neurotuberculosis have been made by Banga et al.,11 Jyothy et al.12 and Bhatia et al.13 The study also evaluated the clinical-laboratory profile of neurotuberculosis in children. The results showed that the most common presenting symptoms were fever and excessive crying, which were present in all cases. Other common symptoms included loss of consciousness, neck pain, abnormal behavior, neurological deficit, headache, convulsions, and vomiting. All patients had fever and excessive crying as clinical features, and the majority had other clinical features such as loss of consciousness (95.5%), neck pain (90.9%), abnormal behavior (81.8%), neurological deficit (81.8%), headache (77.3%), convulsions (68.2%), and vomiting (63.6%).In a similar study by Banga et al.11 the most frequent symptoms reported in patients with tubercular meningitis were fever (68%), headache (59%), and vomiting (54%). Seizures were reported in 37.5% of cases, altered sensorium in 32%, and focal deficits in 19.6%. Only 5.4% of cases had excessive irritability. Cough was present in 23.2% of patients. Our study is the first of its kind to provide the information on sensory examination and a detailed cerebellar function in patients with tubercular meningitis. The sensory examination showed that pain sensation was normal in 77.3% of patients, temperature in 63.6%, crude touch in 54.5%, light touch in 59.1% and pressure in 59.1%. In terms of cerebellar function, all patients had normal gait, and nystagmus and incoordination were absent in all cases except for 1. Additionally, all patients had a Glasgow Coma Scale (GCS) score above 10, with a mean GCS score of 14.2±1.1. On examination for signs of raised intracranial pressure, none of the patients had diplopia, but 4.5% had ataxia, 18.2% had headache, and 72.7% had vomiting. Furthermore, only 18.2% of the cases had received BCG vaccination, while the remaining 81.8% had not. Vomiting is the most common clinical feature of raised ICP according to Kaur et al.and Israni et al.14,15 Additionally, 40.9% of the patients had stage I tubercular meningitis, 50% had stage II, and 9.1% had stage III. In accordance with our findings, Israni et al.also found that out of the 29 children in the study, 62% had stage III illness at the time of their presentation. The study found that stage III illness, low Glasgow Coma Scale (GCS) scores, and elevated intracranial pressure (ICP) were strong predictors of death. Results from cerebrospinal fluid analysis and neuroimaging did not predict death. The study also found that TBM in children is associated with certain characteristic symptoms, and the most significant factors that led to a poor outcome were stage III illness, poor GCS, lack of Bacillus Calmette-Guérin (BCG) immunization at birth, and elevated ICP. The study also performed MRI examination and found that meningeal enhancement was present in 9.1% of cases, tuberculoma in 9.1%, hydrocephalus in 22.7%, vasculitis in 9.1%, ventriculitis in 4.5% and encephalitis in 4.5%. None of the cases had infarct or basal exudates. While in the study done by Singh et al.16 the prevalence of MRI findings was higher. The finding that only 2 out of 22 cases (9.1%) were positive for the Gene Xpert MTB/RIF assay suggests that the test may have a low rate of sensitivity in detecting Mycobacterium tuberculosis (MTB) in patients with tubercular meningitis (TBM). However, the specificity is 99.5%.58 The gene Xpert test has a positive rate between 8-25%. In their study, Jyothy et al.found 21.4% positivity of geneXpert in TBM. There are several possible reasons for the low rate of positivity in this study. Firstly, the sensitivity of the Gene Xpert MTB/RIF assay is dependent on the number of bacilli present in the specimen. In TBM, the bacilli may be present in low numbers, making it harder to detect them using the assay. Secondly, the assay is primarily designed to detect MTB in respiratory samples, which may have different characteristics compared to cerebrospinal fluid (CSF) samples. It may not be suitable to use it on CSF samples directly. Additionally, the specimen may be degraded or contaminated. Finally, other type of Meningitis caused by different agents may also have similar presentation with TBM and may make diagnosis difficult. Finally, the study also compared the CSF findings, neurological findings and Mantoux test status between gene-positive and gene-negative patients. The study found that Gene positive cases had a significantly higher cell count than negative (p value-0.001) but there was no difference between the two groups in lymphocyte % (p value-0.08) . Gene positive cases had a significantly lower protein level (p value-0.04) and higher sugar level in CSF (p value-0.02) , and only hydrocephalus was significantly associated with positive Gene Xpert MTB/RIF. Jyothy et al.12 also found CSF results to be significantly superior in gene positive cases. All other findings, namely, meningeal enhancement, tuberculoma, vasculitis, infarcts, encephalitis and ventriculitis were equally distributed between gene positive and gene negative cases. neurological signs, namely, Kernig sign, Brudzinski sign and neck rigidity were equally distributed between gene positive and gene negative cases. None of the studies done so far have compared these findings between gene positive and gene negative cases, making our study’s observations unique and a benchmark for future studies.

In summary, the study reports on the use of diagnostic tool Xpert-MTB/RIF assay for the diagnosis of TBM in children. The results showed that the assay has a low positivity rate. However, since cases with ventriculitis and encephalitis had positive gene assay it may be a promising alternative tool to the conventional diagnostic tools in diagnosing pediatric TBM. However, it is important to consider other factors such as the clinical probability of TBM when interpreting test results. Additionally, the study suggests the need for further research to confirm the findings and contribute to the development of pediatric TBM screening and treatment guidelines.

1. Tuberculosis. World Health Organisation. Fact sheet No. 104. Published March 2014. http://www.who.int/mediacentre/factsheets/fs104/en/. 2. Bernaerts A, Vanhoenacker FM, Parizel PM, et al. Tuberculosis of the central nervous system: overview of neuroradiological Findings. EurRadiol. 2003;13(8):1876-1890. 3. Andronikou, Wieselthaler N, Smith B, et al. Value of early follow-up CT in pediatric tuberculous meningitis. PediatrRadiol.2005;35:1092-1099. 4. Andronikou, Savvas, Wieselthaler, Nicky. Modern imaging of tuberculosis in children: thoracic, central nervous system and abdominal tuberculosis. PediatrRadiol. 2004;34:861-875. 5. Torok ME. Tuberculous meningitis: advances in diagnosis and treatment. Br Med Bull. 2015;113:117-131. 6. Principi N, Esposito S. Diagnosis and therapy of tuberculous meningitis in children. Tuberculosis. 2012;92:377-383. 7. Garg RK. Tuberculous meningitis. ActaNeurol Scand. 2010;122(2):75-90. 8. Cameron LH, Starke RJ. Tuberculosis (mycobacterium tuberculosis).In Kliegman RM,editor. Nelson textbook of pediatrics, 21st edition. Philadelphia:Elsevier;2019.p.1578-9. 9. Donald P R, Schaaf H S, Schoeman J F. Tuberculous meningitis and miliary tuberculosis: the Rich focus revisited. J Infect. 2005;50:193–195. 10. Lammie GA, Hewlett RH, Schoeman JF, Donald PR. Tuberculous cerebrovascular disease: a review. J Infect. 2009;59:156–166. 11. Banga S, Azad C, Gupta R, Sawal N, Mahajan V, Chander J, Guglani V. Changing Clinicoradiologic Spectrum of Intracranial Neurotuberculosis in Children: A Cross-sectional Study. J Child Neurol. 2020;35(13):879-888. 12. A J, Ratageri VH, Illalu S, Fattepur SR, Wari PK. The Utility of CSF Xpert MTB/RIF in Diagnosis of Tubercular Meningitis in Children. Indian J Pediatr. 2019;86(12):1089-1093. 13. Bhatia R, Dayal R, Jindal S, Agarwal D, Goyal A. GeneXpert for diagnosis of tubercular meningitis. Indian J Pediatr. 2016;83:1353–5. 14. Kaur H, Sharma K, Modi M, et al. Prospective Analysis of 55 Cases of Tuberculosis Meningitis (TBM) in North India. J ClinDiagn Res. 2015;9(1):DC15-DC19. doi:10.7860/JCDR/2015/11456.5454 15. Israni AV, Dave DA, Mandal A, et al. Tubercular meningitis in children: Clinical, pathological, and radiological profile and factors associated with mortality. J Neurosci Rural Pract. 2016;7(3):400-404. 16. Singh P, Paliwal VK, Neyaz Z, Srivastava AK, Verma R, Mohan S. Clinical and magnetic resonance imaging characteristics of tubercular ventriculitis: an under-recognized complication of tubercular meningitis. J Neurol Sci. 2014 Jul 15;342(1-2):137-40.