Miliary tuberculosis disease complicated by Pott's abscess in an infant: Seven-year follow-up.

Sir,

I have two comments on the interesting case report by Bayhan et al.[1]

First, I presume that the devastating clinical course in the case in question with Pott's abscess that left residual kyphosis can not be attributed solely to miliary tuberculosis (TB) unless there is a concomitant underlying condition. It is noteworthy that immunocompromised individuals are generally more prone to have co-infections as well as protracted clinical courses of these infections. Though Bacille Calmette-Guérin (BCG) vaccine coverage in children in Turkey has been reported to range between 94 and 99%,[2] TB is continuously reported. On the other hand, the available data have revealed that there was an upward trend in human immunodeficiency virus (HIV) infection incidence in the last decade in Turkey, and pediatric HIV infection was reported to constitute 1% of the total cases between 2007 and 2011.[3] Despite no studies on the prevalence of pediatric TB in HIV-infected patients are yet present in Turkey, the recently published multicenter, retrospective, observational study on the prevalence, characteristics, management, and outcome of pulmonary TB in Asian HIV-infected children in the TREAT Asia Pediatric HIV Observational Database (TApHOD) has shown that over a 13-year period, the period prevalence of pulmonary TB in HIV pediatric patients was 17.1% (range 5.7-33.0% per country).[4] I, therefore, presume that underlying HIV infection ought to be considered in the case in question and CD4 count and viral load estimation must be contemplated by Bayhan et al.[1]

Secondly, Bayhan et al. addressed that tuberculin skin test (TST) was negative in the studied patient. Moreover, early morning gastric aspirate and urine were negative for acid-fast bacilli (AFB). The authors preliminarily diagnosed the patient with miliary TB and Pott's disease, and began administering antituberculous treatment on day 11 of hospitalization. Tru-Cut biopsy for Pott's abscess was performed on day 14 of hospitalization, and AFB analysis and mycobacterial culture of the biopsy specimen was positive, whereas polymerase chain reactions (PCRs) of biopsy specimen were negative.[1] I presume that the delay in the diagnosis of TB in the case in question might be related to the diagnostic protocol employed in the patient. It is well-known that TST has many limitations, including difficulty in administration and interpretation, the need for a return visit by the patient, and false-positive results caused by significant cross-reaction with Mycobacterium bovis incorporated in BCG vaccines and many non-tuberculous mycobacteria. Interferon-γ release assays (IGRAs) are blood tests that measure T-lymphocyte release of interferon-γ after stimulation by antigens specific for Mycobacterium tuberculosis. Because these antigens are not found on M. bovis-BCG or most non-tuberculous mycobacteria, IGRAs are more specific and rapid tests than the TST, yielding fewer false-positive results.[5] Hence, IGRAs are more preferred worldwide for the diagnosis of TB infection even in HIV-infected patients as well as it would significantly decrease unnecessary preventive treatment in healthy unexposed BCG-vaccinated children. I wonder why Bayhan et al.[1] did not consider IGRAs in the diagnostic algorithm in their studied patient.