Extensively drug-resistant tuberculosis (XDR-TB) is present in all regions and poses serious challenges for public health and clinical management. Laboratory diagnosis is difficult and little evidence exists to guide clinicians in treating people with XDR-TB effectively. To summarise the available data on diagnosis and treatment, the current authors performed a systematic review on 13 recent studies of the epidemiology and clinical management of XDR-TB. Studies that met inclusion criteria were reviewed, in order to assess methodology, treatment regimens and treatment outcomes. Meta-analysis of currently available data is not possible because of inconsistent definitions and methodologies. Data show that XDR-TB can be successfully treated in up to 65% of patients, particularly those who are not co-infected with HIV. However, treatment duration is longer and outcomes are in general poorer than for non-XDR TB patients. To strengthen the evidence for extensively drug-resistant tuberculosis diagnosis, treatment and prevention, future studies should: 1) be prospective in design; 2) adopt standardised, internationally accepted definitions; 3) use quality-assured laboratory testing for all first- and second-line drugs; and 4) collect data on an agreed-upon set of standard variables, allowing for comparisons across studies. Early diagnosis and aggressive management of extensively drug-resistant tuberculosis provide the best chance of positive outcome, but prevention is still paramount.