Table 1.

Current knowledge and research gaps for childhood tuberculous meningitis.

Research Area	Current Knowledge	Research Gaps
Pathogenesis	Novel biomarker and host genotype studies offer new insight into TBM pathophysiology Young age is a major determinant of progression from TB infection to TBM BCG vaccination and TB preventive therapy for children exposed to TB cases are important preventative strategies	Further understanding of the mechanism of TB dissemination from lungs to CSF and CSF invasion Further understanding of host and pathogen factors that determine why some children develop TBM
Epidemiology	More than half of all children with TB globally are undiagnosed or unreported Infants have an up to 20% risk of developing TBM following TB infection Advanced disease stage at diagnosis is associated with high mortality and morbidity	The number of children with TBM are currently unknown, and modelling studies are needed to provide estimates of the burden, morbidity, and mortality in children globally Operational research can help to identify and mitigate the impact of drivers behind diagnostic delays and missed opportunities for prevention Improve mechanisms for reporting of TBM in all age groups to national TB surveillance programs
Diagnostics	TBM can be diagnosed with reasonable confidence with clinical, laboratory, and neuroimaging findings MRI is superior to CT imaging for children being evaluated for TBM, both from a diagnostic perspective but also to delineate pathological and prognostic features M. tuberculosis detection remains the ‘gold standard’ diagnostic test but is limited by poor sensitivity Biomarkers have potential to improve our ability to discriminate children with TBM from children with other causes for their symptoms and signs	Establish validated clinical case definitions in adults and children taking into account different settings stages of TBM and HIV-status Investigate the ability of MRI CSF flow imaging and thin slice CT to differentiate communicating and non-communicating hydrocephalus Investigate utility of other modalities (e.g., 18F-PET/CT) to identify early small infarctions missed with conventional imaging Develop new adequately sensitive, accessible, and rapid diagnostic tests, especially at point of care to allow prompt diagnosis Further investigate the role of non- or less invasive TB testing (e.g., serum, urine, or saliva) in diagnosing TBM in patients where CSF is difficult or cannot be obtained Further investigate the utility of new omic technology, transcriptional and metabolomic biomarkers in diagnosing TBM from other non-TBM CNS infections, including in various populations, stages of TBM, and HIV-status
Treatment	Current recommended doses of TB drugs to treat TBM do not reach optimal CSF levels Higher doses of rifampicin are required to penetrate the blood CSF barrier High CSF concentrations of isoniazid are associated with improved survival To date, the only adjuvant therapy proven to reduce mortality in TBM is corticosteroids for up to 8 weeks Neurocritical care in severe or complicated disease should be part of routine management of TBM	What is the optimal dose, drug regimen, and duration of TB drugs to adequately treat TBM? Is shortened therapy non-inferior to the standard 12 months? Should patients who are fast acetylators receive different doses of isoniazid than slow acetylators? Besides rifampicin, isoniazid, and pyrazinamide, which first or second-line TB drug is the most optimal 4th drug in a drug-susceptible TBM regimen? When is the optimal time to start TB treatment in HIV-positive children? Do the new second line TB drugs (bedaquiline, delamanid, and pretomanid) have a role in treating drug-resistant TBM? Can new host-direct therapy (e.g., high-dose aspirin, thalidomide, and monoclonal-antibodies such as TNF-alpha inhibitors) reduce mortality or neurodisability in children with TBM? What is the optimal management for CNS complications in TBM? (e.g., hydrocephalus, tuberculomas, and paradoxical reactions) What is the optimal supportive and critical care in low-resourced settings to improve mortality and morbidity?
Long term morbidity	Morbidity and mortality from TBM remain unacceptably high, even if treated Reasons for poor outcome are multifactorial; however, the most important predictor of poor outcome is diagnostic and treatment delay Although the array of long-term sequelae is broad, the most common long-term outcomes are physical and developmental disability Neurocognitive deficits can occur without physical disability and have important psychosocial and educational consequences for children, especially those with immature brains	Establish validated and culturally appropriate tools to assess neurodevelopment and function in children with TBM Standardisation of patient follow up and strengthened surveillance to include physical, neurocognitive, and neurodevelopmental assessments Evaluation of whether optimised anti-TB therapy and host-directed therapy could improve long-term neurodevelopmental and neurocognitive outcomes across various paediatric age ranges Evaluation and characterisation of early intervention and targeted neurorehabilitation services to improve long-term outcomes

TBM: tuberculous meningitis; TB: tuberculosis; BCG: bacille Calmette–Guérin; CSF: cerebrospinal fluid; MRI: magnetic resonance imaging; CT: computerised tomography; PET: positron emissions tomography; CNS: central nervous system; TNF: tumour necrosis factor.