Introduction An abnormal chest x-ray is considered by the World Health Organisation to have a ‘high sensitivity for pulmonary tuberculosis’ and thus is widely relied upon to screen for active disease.¹ National Institute for Clinical Excellence recommends using chest x-ray (CXR) findings to determine which microbiological samples to collect, which may risk missing smear or culture-positive patients with normal imaging.² Here we assess the yield of respiratory samples for pulmonary TB patients with a normal CXR at diagnosis, and describe the characteristics of culture-positive, normal CXR patients

Methods Health records of all patients entered into the London TB Register for 6 hospitals between 2011 and 2016 with a diagnosis of pulmonary or intra-thoracic nodal tuberculosis were reviewed retrospectively. Patients with a normal CXR and no subsequent abnormal CXR within 6 weeks of starting TB therapy were identified, and radiological and microbiological findings reviewed. Duplicates and denotified patients were excluded.

CT abnormalities included tree-in-bud (11%), consolidation or nodules (46%), cavitation (10%), lymphadenopathy (30%) and pleural effusion (4%). 45% had more than one abnormality.

There was no significant difference in median age or biochemical markers (ESR, CRP, albumin, vitamin D) at diagnosis between the culture-positive and culture-negative groups. 7 culture-negative patients were HIV-positive, compared with 1 culture-positive patient.

Discussion 43% of patients with a normal CXR had respiratory samples which were culture positive for Mycobacterium tuberculosis (M.Tb); as such, we demonstrate the value of obtaining respiratory samples in patients with a suspected diagnosis of pulmonary TB, even when the CXR is normal. Yield of smear on all types of sample was low (4%–5%), however around one third of samples obtained via spontaneous, induced sputum, or BAL, were culture positive.

While CT imaging usually supported the diagnosis, 3 patients in our cohort were found to be culture-positive on sputum samples despite normal CT imaging.

M.Tb culture yield of respiratory samples was lower for patients diagnosed with intra-thoracic nodal TB, however few samples were sent.

References

1. Chest Radiography in Tuberculosis Detection. Summary of current WHO recommendations and guidance on programmatic approaches 2016.http://www.who.int/tb/publications/chest-radiography/en/

2. Diagnosing and Managing Active Tuberculosis. NICE guidance2016. https://www.nice.org.uk/guidance/ng33/chapter/Recommendations#active-tb