Objective To formulate an evidence-based treatment for a patient with pulmonary tuberculosis combined with tuberculous meningitis and tuberculous pericarditis. Methods According to the principles of evidencebased clinical practice, we searched The Cochrane Library (Issue 2, 2008), Ovid-Reviews (1991 to 2008), MEDLINE (1950 to 2008), and http://www.guideline.org. to identify the best evidence for treating a patient with pulmonary tuberculosis combined with tuberculous meningitis and tuberculous pericarditis. Results Nine guidelines, 2 systematic reviews, and 11 randomized controlled trials were included. The evidence showed that corticosteroids could help reduce the risk of death and disabling residual neurological deficiencies in patients with tuberculous meningitis. After adjusting for age and gender, the overall death rate of patients with tuberculous pericarditis was significantly reduced by prednisolone (P=0.044), as well as the risk of death from pericarditis (P=0.004). But for patients with pulmonary tuberculosis, there was still a controversy about the use of corticosteroids. Given the evidence, the patient’s clinical conditions, and his preferences, dexamethasone was used for the boy in question. After 7 weeks of treatment, his cerebrospinal fluid returned to normal and pericardial effusion disappeared. Conclusion Corticosteroids should be recommended in HIV-negative people with tuberculous meningitis or/and tuberculous pericarditis. The difference in the effectiveness of various corticosteroids such as dexamethasone, prednisolone, or methylprednisolone and the optimal duration of corticosteroid therapy is still unknown.