i read with interest the recent article in the new england journal of medicine by dick menzies et al, entitled, “four months of rifampin or nine months of isoniazid for latent tuberculosis in adults.” the two regimens were found to be equivalent.
a few thoughts are below.
this is a nice clinical trial which mirrors our recent clinical experience of preferring rifampin over isoniazid.
when will we have good data for a short course regimen? four months of rifampin is too long, and i’m not convinced that rifapentine based regimens are ready for prime time. hopefully i am wrong.
1/4 of the people in the world have latent tuberculosis infection. if we were to give rifampin to all of them, what would happen in terms of drug resistance? i ask this question because we believe there are a number of patients with subclinical tuberculosis who have fibrotic scar or granuloma on chest imaging, and if they were to have bronchoscopy or thoracic surgical resection, would likely grow tuberculosis. the question is when you see nodule or fibrotic scar on chest ct, what size is necessary to trigger our index of suspicion for tb.
i will plan to write more about this on a future blog post about artificial intelligence and ct scanning for mycobacterial disease (tuberculosis and non-tuberculous mycobacteria i.e. m. abscessus).