Background: Corticosteroid exerts anti-inflammatory action and can prevent tissue
damage resulting from various causes. Studies have shown that corticosteroids may prevent the
damaging effect of tuberculosis (TB) in various organs, but there is no published meta-analysis
specifically looking for the effect of corticosteroid in endobronchial TB.
Objective: To synthesize the evidence regarding the usefulness of corticosteroid in endo-bronchial
Methods: A comprehensive search was performed of the major electronic databases till 30th November
2018. Randomized trials comparing treatment with corticosteroid as an adjunct to antitubercular
drugs (ATT) versus placebo/no treatment in endobronchial TB were included. Three
authors independently applied eligibility criteria, assessed the studies for methodological quality,
and extracted data. The review is registered at PROSPERO database [CRD42016047063].
Results: Out of 525 search results, 4 trials including 205 patients (151 children) were eligible for
inclusion. Oral prednisolone was used in various dose schedules. Rifampicin containing ATT regimen
was used in 3 trials. The bronchoscopy findings showed no significant improvement at 1
month (effect size could not be calculated due to 0 event in the intervention group, p = 0.05), 2
months (RR 1.26, 95% CI 0.89 to 1.8), and at completion of ATT (RR 0.63, 95% CI 0.1 to 4.14) in
steroid-treated group compared to the control group. The need for repeat bronchoscopy was significantly
decreased in the steroid group (RR 0.13, 95% CI 0.02 to 0.9). Among the adverse events, the
infection rate was significantly lesser in the steroid group (RR 0.53, 95% CI 0.29 to 0.97); but other
adverse events (mortality, hypertension, and abdominal distension) showed no significant difference
between the two groups. The GRADE evidence generated was of very low quality.
Conclusion: The present meta-analysis showed that oral steroid does not help patients with endobronchial
tuberculosis. However, the quality of evidence was very low. Future trials with robust
design and a larger sample size would be required to provide any firm recommendation regarding
the use of oral prednisolone in endobronchial tuberculosis.