Multidrug-resistant tuberculosis outbreak in an Italian prison: tolerance of pyrazinamide plus levofloxacin prophylaxis and serial interferon gamma release assays.

The optimal treatment for latent tuberculosis infection (LTBI) in subjects exposed to multidrug-resistant (MDR) tuberculosis (TB) remains unclear, and the change in response of the QuantiFERON-TB Gold In-Tube (QTB-IT) test during and after treatment is unknown. Between May 2010 and August 2010, 39 prisoners at the 'Casa Circondariale' of Modena, Italy, were exposed to a patient with active pulmonary MDR TB. All contacts were tested with the tuberculin skin test and QTB-IT. Upon exclusion of active TB, subjects positive to both tests were offered 6 months' treatment with pyrazinamide (PZA) and levofloxacin (LVX). QTB-IT testing was repeated at 3 and 6 months after initial testing in all subjects who were offered LTBI treatment. Seventeen (43.5%) of 39 subjects tested positive to both tuberculin skin test and QTB-IT test, and 12 (70.5%) agreed to receive therapy with PZA and LVX at standard doses. Only five (41.6%) of 12 subjects completed 6 months' treatment. Reasons for discontinuation were asymptomatic hepatitis, gastritis and diarrhoea. The QTB-IT values decreased in all subjects who completed the treatment, in two (33%) of six of those who received treatment for less than 3 months and in one (50%) of two patients who discontinued therapy after 3 months. The QTB-IT test results never turned negative. Despite the small number of subjects, the study confirmed that PZA plus LVX is a poorly tolerated option for MDR LTBI treatment. We observed a large degree of variation in the results of the QTB-IT test results among participants. The study confirmed that the interferon gamma release assay is not a reliable tool for monitoring the treatment of MDR LTBI in clinical practice.