WHO recommendations on shorter treatment of multidrug-resistant tuberculosis.

2486 www.thelancet.com Vol 387 June 18, 2016 Tuberculosis is now the world’s commonest cause of death from infectious disease. The ominous spread of multidrug-resistant (MDR) and extensively drugresistant (XDR) tuberculosis, and the scarce treatment options available, are priority global health issues. With a global estimate in 2014 of 450 000 cases of MDR and XDR tuberculosis causing 150 000 deaths, and the continuing spread, the WHO End TB Strategy highlights the threat that MDR and XDR tuberculosis pose to global public health security. Poor treatment success of available treatment regimens and evolution of Mycobacterium tuberculosis strains with resistance patterns beyond XDR tuberculosis pose major management challenges. Available treatment regimens have poor effi cacy, are toxic and expensive, and require lengthy treatment, compromising patient adherence and therapeutic drug monitoring. Furthermore, health-care providers fi nd it diffi cult to design eff ective regimens because of inadequate laboratory facilities for tuberculosis drug susceptibility testing (DST). The recent release by WHO of new recommendations aimed at speeding up tuberculosis DST using a rapid molecular MTBDRsl test and use of shorter MDR tuberculosis treatment regimens is a welcome, long-awaited development. The recommendations highlight the advantages of the new regimen (4–6 months of kanamycin acid, moxifl oxacin, protionamide, clofazimine, pyrazinamide and high-dose isoniazid and ethambutol followed by 5 months of moxifl oxacin, clofazimine, pyrazinamide, and ethambutol), and provide a fact sheet with the necessary explanations. The shorter duration (9 months) and its low cost (<US$1000) will go some way in improving the current dismal status quo. The regimen is recommended only for MDR tuberculosis cases not previously treated with, or resistant to, second-line anti-tuberculosis drugs. Although the new guidelines are welcomed, we have concerns about whether the shorter treatment regimen is likely to be eff ective in all geographical settings. The WHO recommendations are based on a multicentre study of 1200 patients, not geographically representative of all MDR and XDR tuberculosis endemic regions, and they might not be applicable in specifi c hotspots for MDR and XDR tuberculosis, such as countries of the former Soviet Union in which there are numerous previously treated cases. Thus, a cautious decision-making approach, based on DST, is essential. However, the short regimens have produced excellent outcomes under operational research conditions in some settings in which they benefi ted from preexisting knowledge of the local epidemiology of drug resistance and availability of rapid MTBDRsl testing to ensure DST for the key drugs composing the treatment regimen. Under these conditions, the WHOrecommended short MDR tuberculosis regimen could be very useful for some patients, as treatment duration is substantially reduced. Data show that the short regimen, previously called the Bangladesh regimen, might not be theoretically eff ective in certain settings. Our International Carbapenems Study Group recently did a multicentre, retrospective cohort study of 348 patients recruited in several clinical centres specialised in the management of MDR and XDR tuberculosis cases and located in eight countries in Europe and three countries in South America. Adults with a culture-confi rmed diagnosis of MDR tuberculosis were assessed according to meropenem-containing and imipenem-containing, WHO recommendations on shorter treatment of multidrug-resistant tuberculosis