EBUS-TBNA in a paediatric population in London

Tuberculosis (TB) is a leading cause of death and morbidity worldwide. Of the 10 million incidence cases of TB estimated in 2017, 10% were in children. The diagnosis of TB in the paediatric population remains challenging due to the paucibacillary nature of the disease. Culture confirmation is low for both pulmonary and extrapulmonary disease; 30.3% and 19.4% respectively resulting in empiric treatment for presumed TB. Mediastinal TB is a common presentation of TB in children but difficult to access for sampling. A retrospective case review of all children (<16 years) with mediastinal/hilar lymphadenopathy who underwent an EBUS-TBNA between January 2010-2018 with a potential diagnosis of TB in a London TB centre were reviewed. There were 8 cases in total all performed under sedation with no reported complications. 75% cases showing granulomas on rapid on-site histological evaluation. All 8 cases were confirmed to have mycobacterium tuberculosis from EBUS samples. 7/8 cases were culture positive with a mean time for culture results at 13.3 days. 4/7 culture positive cases were smear positive. TB PCR picked up 1 further case where microscopy and culture remained negative. 1 case had multidrug resistant TB which was immediately identified on TB PCR allowing early initiation of correct drug therapy. In our cohort we show EBUS-TBNA is a safe and effective way of investigating TB lymphadenitis in children. It is essential that in a high resource setting we approach childhood TB with a standardised diagnostic approach, and utilise EBUS-TBNA as a diagnostic modality. Samples should be sent for drug sensitivity testing including point of care molecular assays so as to identify drug resistant TB in a timely fashion.