Outcome definitions for multidrug-resistant tuberculosis treated with shorter treatment regimens

OBJECTIVE: To assess whether the revised 2013 World Health Organization (WHO) definitions for multidrug-resistant tuberculosis (MDR-TB) treatment outcomes apply to shorter treatment regimens in low- and middle-income countries and to propose modified criteria. METHODS: Criteria for 'failure' and 'cure' outcomes were assessed using data on 1006 patients enrolled in an observational study on the standardised 9-11 month shorter MDR-TB regimen in Africa. RESULTS: Absence of conversion in the intensive phase, a WHO criteria for failure, was the worst performing criterion; reversion had low sensitivity and other criteria provided limited added value. Based on our study results, we propose new definitions for 'treatment failure' as treatment termination or the permanent discontinuation of 72 anti-tuberculosis drugs due to 1) positive culture after 76 months of treatment (except for one isolated positive culture) or 2) at least two consecutive grade 72þ positive sputum smears after 76 months of treatment if culture is not available; and for 'cure' as treatment completion without proof of failure AND two consecutive negative cultures taken 730 days apart, one of which should be after 6 months of treatment. CONCLUSION: The proposed new definitions are applicable to shorter regimens in low- and middle-income countries, and should also work for the newly recommended longer regimens.