WHO recommendations for multidrug-resistant tuberculosis.

2234 www.thelancet.com Vol 388 November 5, 2016 3 Casas EC. Experiences with short MDR-TB regimen in unstable settings. The 46th Union World Conference on Lung Health; Cape Town, South Africa; Dec 2–6, 2015. http://capetown. worldlunghealth.org/programme/ programme-by-type/sponsored-satellitesymposia/pdf/12-Sponsored-satellitesymposium.pdf (accessed June 18, 2016). 4 Casas E, Gashu T, Greig J, et al. 9-month short-course MDR-TB treatment in HIVand non-HIV-co-infected patients in Uzbekistan and Swaziland: interim outcomes of two prospective studies. The 46th Union World Conference on Lung Health; Cape Town, South Africa; Dec 2–6, 2015. http://capetown. worldlunghealth.org/programme/ programme-by-type/e-poster/pdf/EP-03-04Dec.pdf (accessed June 18, 2016). 5 Lalor MK, Greig J, Allamuratova S, et al. Risk factors associated with default from multiand extensively drug-resistant tuberculosis treatment, Uzbekistan: a retrospective cohort analysis. PLoS One 2013; 8: e78364. 6 WHO. Global tuberculosis report 2015. Geneva: WHO, 2015. http://apps.who.int/iris/ bitstream/10665 /191102/1/9789241565059 _eng.pdf (accessed June 18, 2016). the potential benefi t for some patients with multidrug-resistant (MDR) tuberculosis. We wish to address their concern regarding applicability in countries of the former Soviet Union. Interim data (1-year, relapse-free outcomes due in 2017) from our observational study of the shortened regimen in Karakalpakstan, Uzbekistan, contributed to the meta-analysis on which WHO based their recommendations. The Ministry of Health of Karakalpakstan, in partnership with Médecins sans Frontières, chose to pilot the shortened regimen in parallel with the line-probe assay for second-line drugs. This decision was made on the basis of the challenges of maintaining quality patient-centred care when scaling up a 20-month regimen: successful outcomes dropped to 62%; and loss to follow-up increased to 20%. When considering whether to incorporate some or all of the WHO recommendations, national programme managers must balance the benefit of reducing treatment duration for some patients against the extra resources required to identify them. While some settings have high levels of resistance, there remain patients within these contexts who are likely to benefi t from shorter regimens. Keeping patients at the centre of these decisions is key. With global successful treatment outcomes of just 50% for MDR tuberculosis, perhaps the focus should be on who might benefit from the shorter regimen rather than who might not.