Comments on 'Risk factors for drug induced hepatitis with first-line antituberculosis drugs in hospitalized patients of pulmonary tuberculosis'

R Animesh 
 Department of Pulmonary Critical Care and Sleep Medicine, VMMC and Safdarjang Hospital, New Delhi

Correspondence Address:
Dr. R Animesh
Department of Pulmonary Critical Care and Sleep Medicine, VMMC and Safdarjang Hospital, New Delhi

Full Text

Sir,

I read with interest the article titled "Risk factors for drug induced hepatitis with first-line antituberculosis drugs in hospitalized patients of pulmonary tuberculosis" [1] and would like to make the following comments.

Previous anti-tubercular therapy (ATT) has been deemed by the authors to be a risk factor for ATT induced hepatotoxicity. This observation is interesting and claims novelty in that though previous history of isoniazid induced hepatitis has been reported as a potential risk factor [2] for ATT induced hepatotoxicity, previous history of ATT intake has not been similarly incriminated. A closer look at [Table 2] and [Table 3] provided by the authors 1 reveals that the odds ratio (OR) for the development of the drug induced hepatitis (DIH) in case of previous ATT is 0.488 [Table 2] and 0.500 [Table 3] which indicates that previous ATT was not a risk factor for DIH (as reported by the authors) but rather a protective factor for the same. But the reason for it being a protective factor needs further deliberation.As per the data provided on [Table 1] and [Table 2], 1 6.46% (17) of total female patients (263) had DIH as compared to 6.19% (39) of total male patients (630) who had DIH, yet the odds ratio for female developing DIH is computed to be 2.515 (univariate analysis) and 2.463 (multivariate analysis) which seems discrepant.Human immunodeficiency virus (HIV) co-infection has been implicated to increase the risk for DIH with ATT [3] but it was not found to be a risk factor in the present study. One reason which might be responsible for this lack of association is the unavailability of HIV reports in 636 out of 893 (71.2%) patients. As in a vast majority of cases the HIV serostatus was unknown; any attempt to draw out an association (or the lack of it) seems untenable. Also at a time when testing for HIV is strongly recommended in all tuberculosis patients [4] the reasons for the lack of information on the HIV status in more than two-thirds of patients needs to be further scrutinized.

References