OUP user menu

Extensively drug-resistant tuberculosis: current challenges and threats

Amita Jain, Rajesh Mondal
DOI: http://dx.doi.org/10.1111/j.1574-695X.2008.00400.x 145-150 First published online: 1 July 2008


Extensively drug-resistant tuberculosis (XDR-TB) is defined as tuberculosis caused by a Mycobacterium tuberculosis strain that is resistant to at least rifampicin and isoniazid among the first-line antitubercular drugs (multidrug-resistant tuberculosis; MDR-TB) in addition to resistance to any fluroquinolones and at least one of three injectable second-line drugs, namely amikacin, kanamycin and/or capreomycin. Recent studies have described XDR-TB strains from all continents. Worldwide prevalence of XDR-TB is estimated to be c. 6.6% in all the studied countries among multidrug-resistant M. tuberculosis strains. The emergence of XDR-TB strains is a reflection of poor tuberculosis management, and controlling its emergence constitutes an urgent global health reality and a challenge to tuberculosis control activities in all parts of the world, especially in developing countries and those lacking resources and as well as in countries with increasing prevalence of HIV/AIDS.

  • Mycobacterium tuberculosis
  • extensively drug-resistant tuberculosis


Mycobacterium tuberculosis infects one-third of the global population. More than eight million people develop active tuberculosis (TB) every year, and the disease claims two million lives annually (Dye et al., 1999). The emergence of drug resistance in recent years has made an effective control strategy for tuberculosis indispensable. The rapid spread of drug resistance, especially multidrug-resistant tuberculosis (MDR-TB) and currently extensively drug-resistant tuberculosis (XDR-TB), both in new and in previously treated cases, adds urgency to the need for decisive action to develop control measures (WHO, 2006). Efforts to report on XDR-TB status worldwide form an important part of tuberculosis control and management. Here we review the global epidemiology and challenges to the management of XDR-TB worldwide.



Infection with an M. tuberculosis strain that is resistant to the two most commonly used front-line antitubercular drugs, isoniazid and rifampicin, is defined as MDR-TB. Treatment of MDR-TB is resource-intensive, and the therapeutic strategies recommended for high-prevalence areas comprise combinations of second-line drugs that are more expensive, more toxic and less effective than the drugs used in standard therapy (Iseman, 1993; Rajbhandary et al., 2004; Ward et al., 2005; Nathanson et al., 2006). A Green Light Committee of the Stop Tuberculosis partnership was established to facilitate treatment of MDR-TB in resource-limited countries, where most tuberculosis cases occur (Gupta et al., 2002). While ensuring the proper use of second-line drugs in these countries to prevent drug resistance, this committee encountered strains resistant to virtually all second-line drugs, and this has led to the emergence of new terminology in relation to drug-resistant tuberculosis, known as XDR-TB. The worldwide emergence of XDR-TB and a provisional definition for this form of tuberculosis were first reported in November 2005 (Holtz et al., 2005; Shah et al., 2005) and the term XDR-TB was used for the first time in March 2006, in a report jointly published by US Centers for Disease Control and Prevention (CDC) and World Health Organization (WHO) (CDC, 2006). According to this report, XDR-TB was defined as M. tuberculosis that was resistant not only to isoniazid and rifampicin (MDR-TB) but also to at least three of the six classes of second-line antituberculosis drugs (aminoglycosides, polypeptides, fluoroquinolones, thioamides, cycloserine and para-aminosalycilic acid). As the definition is dependent on difficult-to-perform drug susceptibility testing and as some forms of drug resistance are less treatable than others, it was eventually modified at a meeting of the WHO XDR-TB Task Force, held at Geneva (Switzerland) on 10–11 October 2006. The committee gave a widely accepted definition of XDR-TB: ‘resistance to at least RIF and INH among the first line-anti tubercular drugs (MDR-TB) in addition to resistance to any fluroquinolones i.e. ofloxacin, ciprofloxacin and levofloxacin, and at least one of three injectable second line anti tubercular drugs i.e. amikacin, kanamycin and capreomycin’ (CDC, 2006).


MDR-TB strains constitute 1–3% of global tuberculosis (Farmer et al., 1999). According to the WHO, 425 000 new MDR-TB cases occur every year with the highest rates in the former Soviet Union and China, where up to 14% of all new cases are not responding to standard drug treatment (Zignol et al., 2006). MDR-TB has been identified as a significant problem in every regime under WHO surveillance (WHO, 2004). Mismanagement of such cases paves the way for emergence of XDR-TB cases.

The CDC and WHO undertook surveillance to estimate worldwide XDR-TB prevalence during 2000–2004, based on data from 25 supranational reference laboratories (SRLs) on six continents that collaborate with national reference laboratories (NRLs) to increase culture and drug susceptibility testing (DST) capacity and provide quality control for global surveys to assess antituberculosis drug resistance. They used a standardized reporting form, and requested anonymous, individual-level data from all reference laboratories on all isolates tested for susceptibility to at least three second-line drug (SLD) classes, during 2000–2004. SRLs receive varying proportions of isolates from countries for surveillance, diagnosis and quality assurance. Hence, to complement the SRL survey, additional population-based data were analysed from the United States National Tuberculosis Surveillance System, which contains data on all reported tuberculosis cases during 1993–2004, and Latvia's national MDR-TB registry from the 2000–2002 cohort of MDR-TB patients. Limitations to this surveillance were: (1) not all SRLs tested for susceptibility to SLDs, (2) certain laboratories test for only one or two SLDs, (3) laboratories used different (but generally accepted) media and methods to test for SLD susceptibility and (4) SLD-susceptibility data from SRLs are based on a convenience sample and are not population-based, with one exception: South Korea. Based on the surveillance data, worldwide occurrence of XDR-TB was reported. Reported overall worldwide prevalence of XDR-TB is 6.6% among all MDR-TB isolates; reported XDR-TB prevalence in industrialized countries (e.g. USA, UK, Ireland, Germany, France, Belgium, Spain, Japan and Australia) is 6.5%; and reported XDR-TB prevalence in Russia and Eastern Europe (e.g. Republic of Georgia, Czech Republic, Armenia and Azerbaijan) is around 13.6%. Prevalence of XDR-TB from the Asiatic region (e.g. Bangladesh, Indonesia, Thailand, Papua New Guinea and East Timor), Africa and the Middle East was not well defined in the report owing to a small number of tuberculosis cases taken in the study in comparison with other studied nations. However, reported XDR prevalence is around 1.5% in Asia and only 0.6% in Africa and the Middle East. The Republic of Korea has reported the highest number of XDR-TB cases, these representing 15.4% of all MDR-TB patients (Shah et al., 2007).

A detailed description of XDR-TB findings and preliminary data from the US National TB Surveillance System was published in 2006 (CDC, 2006). These data indicated that 74 tuberculosis cases reported during 1993–2004 met the case definition for XDR-TB. A recent report from Germany and Italy reported 10.3% and 14.3% XDR-TB isolates, respectively, among 83 and 43 MDR-TB strains. These patients had a fivefold higher risk for death and longer hospitalization with longer treatment durations, revealing a highly significant association between XDR-TB status and mortality risk (Migliori et al., 2007a, b). A recent study reported that prevalence of XDR-TB in France was 4% of tested multidrug-resistant strains (Bouvet, 2007). A recent study from Iran reported 12 (10.9%) XDR strains from 113 MDR-TB strains (Masjedi et al., 2006). In Hong Kong, nine out of the 75 MDR-TB strains (12%) had extensive drug resistance with simultaneous resistance to ethionamide, amikacin, ofloxacin and cycloserine (Kam & Yip, 2004). From India, 7.4% of XDR-TB cases were recognized from 68 MDR-TB strains during a preliminary study made by us recently (Mondal & Jain, 2007). Although this figure was based upon a small number of MDR-TB from a North Indian setting, this clearly indicates that the problem of XDR-TB exists, and the true extent may be much higher than the reported figure, given that the annual risk of tuberculosis and prevalence of acquired MDR-TB and tuberculosis with HIV is increasing in India (Narain & Lo, 2004). The emergence of XDR-TB in settings such as India, South Africa, China, South Korea, Ethopia and Estonia, where tuberculosis control programmes have been unable to monitor treatment regimens for MDR-TB adequately due to huge population sizes and high annual tuberculosis cases, is a cause of great concern.

Comorbidity with tuberculosis and HIV/AIDS affects around 11 million people and resulted in the death of nearly 200 000 in 2005. Yet, <0.5% of HIV-positive people were screened for tuberculosis in that year. A recent study from India by Singh (2008) of 54 full-blown AIDS patients suspected of having HIV–TB coinfection reported a high mortality rate among those with XDR-TB. Of the 54 patients studied, 12 (22.2%) were MDR-TB cases, among which were four (33.3%) XDR-TB cases. All four patients in whom XDR-TB was isolated died within 2.6 months of diagnosis. The first report of XDR-TB with HIV came from Kwazulu Natal (KZN), South Africa (Gandhi et al., 2006). Of 536 tuberculosis patients at the Church of Scotland Hospital, KZN, which serves a rural area with high HIV infection rates, 221 were found to have MDR; of these, 53 were diagnosed with XDR-TB. Fifty-two of these patients died, most within 25 days. Of the 53 patients, 44 were tested for HIV and all 44 were HIV-positive. In KZN, two healthcare workers died from XDR-TB with HIV. Spoligotyping results of 46 XDR-TB isolates demonstrated that 85% of the strains belonged to the KZN family and 15% to the Beijing family. A number of cases in KZN resulted from in-hospital infection. Acquisition of extensive drug resistance appears to be the primary mechanism for the XDR-TB epidemic in South Africa, and an estimated 63–75% of cases developed XDR-TB through acquisition (Mlambo et al., 2008). A large number of XDR-TB patients in KZN were infected with the same strain of M. tuberculosis (F15/LAM4/KZN) (Pillay & Sturm, 2007).

Challenges and threats in management

Mismanagement of tuberculosis cases has played a major role in the emergence of drug-resistant tuberculosis, mainly owing to poor treatment methodologies (Uplekar & Shepard, 1991; Prasad et al., 2002; Nathanson et al., 2006). The erratic use of SLDs (and their poor quality; Prasad & Garg, 2007) and factors linked to poor control practices, e.g. lack of measures to ensure adherence to treatment protocols and the treatment of tuberculosis in unmonitored private-sector hospitals, have played a major role in the appearance of XDR-TB cases in developing countries. Uplekar (2003) reported on the low quality of some tuberculosis management in the private sector in some parts of world. It is vital that clinicians caring for tuberculosis patients are aware of the possibility of drug resistance and have access to laboratories that can provide early and accurate diagnosis so that effective treatment is provided as soon as possible. Effective treatment of tuberculosis cases requires good quality front-line drugs and all six classes of SLDs available to clinicians who have expertise in treating drug-resistant cases, especially MDR-TB. A report from South Korea revealed that XDR-TB was significantly associated with the cumulative duration of previous treatment received with second-line tuberculosis among patients in a tertiary-care tuberculosis hospital (Jeon et al., 2008). The inexorable rise of drug-resistant strains (one in ten new infections is resistant to at least one antituberculosis drug) and the worrying spread of XDR-TB threaten to undermine tuberculosis control efforts. Treatment of MDR-TB cases is difficult, complicated, very costly, challenging and requires experienced skills, together with good quality of second line antituberculosis drugs, a high standard of microbiology laboratories as well as proper management of patient care with standardized, empirical and individual approaches (Iseman, 1993; Farmer et al., 1999; Gupta et al., 2001). Treating MDR-TB with SLDs may cure >65% of patients and prevent ongoing transmission (Mukherjee et al., 2004; Van Deum et al., 2004; Leimane et al., 2005). However, most of the evidence relating to the successful management of MDR-TB has been generated from high-income countries where treatment is provided in referral hospitals (Espinal & Dye, 2005).

Most XDR-TB cases seem to have appeared among HIV-positive individuals, and the epidemic of XDR-TB among those living with HIV/AIDS in all parts of world is increasing rapidly and represents a serious challenge to researchers, epidemiologists, clinicians and policy-makers. Confirmed cases of XDR-TB from KZN, South Africa, are of particular significance given the high mortality rate in those who are HIV-positive. Dr Mario Raviglione, WHO Stop TB Department Director, said ‘this is a wake-up call’. There are problems in the management of tuberculosis. HIV fuels XDR-TB. Once someone is infected with the bacterium there is a 5–10% lifetime risk of developing tuberculosis; the risk is 5–15% annually with those who are HIV-positive. The incidence of tuberculosis is decreasing or stable in all regions of the world except for Africa, where it is on the increase, with HIV fuelling tuberculosis. Coordination of tuberculosis and HIV programmes and interventions will be needed. Dr Karin Weyer, tuberculosis Research Director at the South African Medical Research Council, warns: ‘We are afraid that this outbreak of XDR-TB might be the tip of the iceberg, as we haven't really looked properly elsewhere. There are higher prevalence rates in pockets of eastern Europe and South-East Asia but we are particularly worried in South Africa given our HIV problem, because of the rapid spread of XDR-TB amongst HIV patients and their rapid death’ (Wise, 2006). This threat is also worrisome in eastern European countries where 53% of the projected number of MDR-TB cases will be treated and where HIV infection is spreading rapidly (Stop TB, 2006). The appearance of XDR-TB among HIV-infected individuals is very concerning, as rapid progression of infection towards disease and the high potential of spread among immunosuppressed individuals greatly accelerate the consequences of poor control practices and management of drug-resistant tuberculosis. However, the presence of XDR-TB is independent of poor prognostic factors in non-HIV-infected patients with MDR-TB (Kim et al., 2007). These epidemiological trends have resulted in a change to our global strategy to call for ‘universal access to high-quality diagnosis and patient-centred treatment’, including specialized treatment for MDR-TB and for tuberculosis/HIV coinfection.

Controlling community- and hospital-acquired infections among patients (e.g. tuberculosis and TB-HIV) and healthcare workers is of importance (Migliori et al., 2007a, b). Strengthening basic tuberculosis programmes and infection control measures is crucial for preventing the selective pressure and environments in which resistant strains are transmitted from person to person. Additionally, MDR-TB programmes that rely on quality assured and internationally recommended treatment regimens according to WHO guidelines must be scaled up and strengthened to prevent further SLD resistance and spread of XDR-TB.

The Green Light Committee of the Stop Tuberculosis partnership provides a global mechanism to help affected countries to achieve these steps. The main priority interventions that will be needed for XDR management is the strengthening of tuberculosis control (through sound implementation of the Stop tuberculosis strategy and systematic application of treatment in both the public and the private sectors, with a focus on laboratory capacities and infection control), improvement of SLD management based on the new WHO guidelines (Mukherjee et al., 2004), and promotion of research and development of new diagnostics, vaccines and drugs against tuberculosis (Hopewell et al., 2006; Matteelli et al., 2007).

A study by Huong (2006) from Vietnam suggested that, from a public health perspective, treatment of patients with MDR-TB with SLDs might not be necessary to prevent its spread and that directly observed therapy short course (DOTS), the internationally recommended standardized management strategy for tuberculosis, alone may suffice in some settings (DeRiemer et al., 2005). In the absence of a specific public health treatment programme, there was no apparent increase in the prevalence of MDR-TB among untreated patients between surveys conducted in 1996 and 2001 in Vietnam. However, a good DOTS programme should reduce acquired drug resistance generated by erratic, unsupervised therapy and by an unreliable drug regimen. There is substantial evidence that treatment based on isoniazid and rifampicin will not cure or substantially improve tuberculosis in patients whose infecting organisms are already resistant to those drugs (Espinal et al., 2000). Moreover, there is no evidence that ineffective treatment can reduce the transmission of MDR-TB strains. However, most patients who are promptly and properly treated for MDR-TB with SLDs can be cured even in poor settings (Gupta et al., 2001; Nathanson et al., 2006).


Resistance to antituberculosis drugs has been noted since the drugs were first introduced, and occasionally outbreaks of drug-resistant tuberculosis have been reported worldwide. However, recent outbreaks of XDR-TB have differed considerably from previous outbreaks of drug-resistant tuberculosis and even MDR-TB outbreaks. The global threat of XDR-TB has great significance for public health. Its very existence is a reflection of weaknesses in tuberculosis management. Controlling the tuberculosis epidemic and preventing XDR-TB necessitate that healthcare providers diagnose tuberculosis early and initiate effective therapy promptly. Tuberculosis patients should then successfully complete the therapy with high-quality antitubercular drugs for the appropriate duration. Specifically, directly observed therapy should be used more widely. Initial treatment regimens should be used that prevent the development of drug-resistant tuberculosis, and treatment activities should be coordinated between public health departments and other hospitals that provide facilities and care for patients with tuberculosis. XDR-TB will prove to be far more serious due to its virulence. All evidence suggests that XDR-TB reflects a failure to implement the measures recommended in the WHO's Stop Tuberculosis Strategy. This strategy emphasizes expanding high-quality DOTS programmes, surveillance of drug-resistant tuberculosis and HIV associated with tuberculosis, strengthening healthcare systems and primary care services, good clinical practices, empowering patients and communities to improve health, and enabling and promoting research (newer drugs and vaccines for tuberculosis) with advanced microbiology laboratories for early and accurate diagnosis, as well drug susceptibility testing for first- and second-line antitubercular drugs.

‘The WHO emphasizes that good tuberculosis control prevents the emergence of drug resistance in the first place and that the proper treatment of multi-drug resistant tuberculosis prevents the emergence of XDR-TB.’


  • Editor: Willem van Leeuwen


View Abstract