XDR-TB in the United States

Abstract & Commentary

By Stan Deresinski, MD, FACP, Clinical Professor of Medicine, Stanford, Associate Chief of Infectious Diseases, Santa Clara Valley Medical Center, is Editor for Infectious Disease Alert.

Source: Shah NS, et al. Extensively drug-resistant tuberculosis in the United States, 1993-2007. JAMA. 2008; 300:2153-2160.

Synopsis: In contrast to some regions of the world, the incidence of XDR-TB in the United States has decreased and remained at a very low level.

CDC investigators analyzed 15 years of surveillance data of culture-confirmed cases of tuberculosis from the 50 states and the District of Columbia, identifying 201,399 with isoniazid- and rifampin-susceptibility results. Of these, 3,379 (1%-7%) were resistant to both drugs and, thus, categorized as MDR-TB. Among the MDR-TB isolates, sufficient susceptibility testing was available to further categorize 2,087; 83 (0.04% of the total TB cases with drug susceptibility results) were XDR-TB (ie, they were additionally resistant to a fluoroquinolone and to a second-line injectable drug).

The largest annual number of cases of XDR-TB, 18, were reported at the beginning of the surveillance period in 1993, but a median of only 3.5 cases per year were reported over the last decade of the study. During 1993-1997, 25 of 40 (62%) patients with XDR-TB were known to be HIV-infected; this proportion decreased to six of 43 (14%) during 1998-2007. Just over half were US-born, and 40% were Hispanic. Three of the patients with XDR-TB were health care workers.

The median time to culture conversion among the XDR-TB-infected patients was 183 days (range, 103-344 days), a significantly longer time than for MDR-TB patients, or those infected with susceptible strains. Only 44% of XDR-TB patients, however, completed treatment, and 35% died, a death rate more than twice that of MDR-TB patients and more than six times that of patients infected with susceptible strains. Seventy percent (21 of 30) of HIV-infected patients died, while only 9% (2 of 22) without HIV coinfection died.


These data demonstrate that the number of cases of XDR-TB in the United States is small and diminishing, with only a few cases identified annually. This, of course, is not the case in many places in the world. Attention was drawn to the problem of XDR-TB in 2006 with the report of an outbreak in South Africa involving 53 patients, all of whom died.1 Of note, all whose serostatus was known were HIV infected, indicating a link between these two epidemics, one also observed in the US experience. This experience was a harbinger of what was to follow. In February 2008, WHO reported that MDR-TB, a way station on the path to XDR-TB, was present in 5.5% of all cases of TB in the 81 countries surveyed, with China and countries of the former Soviet Union having the highest rates. For example, 22.3% of TB cases in Baku, Azerbaijan, were MDR-TB. Limited data were available from Africa, where only six countries were represented. It was estimated that 490,000 MDR-TB cases emerge annually and account for more than 110,000 deaths.

At least one case of XDR-TB was identified in each of the 45 countries and, once again, the former Soviet Union was among the leaders, with the proportion of MDR-TB that were XDR-TB ranging from 4% in Armenia to 24% in Estonia. WHO estimates that approximately 40,000 cases of XDR-TB will emerge annually in the world.

Many of the patients in the report by Shah et al had acquired drug resistance (ie, initial isolates were generally drug susceptible, while one or more subsequent isolates were XDR-TB). It is generally assumed that this phenomenon is the consequence of poorly chosen regimens or poor patient compliance. A recent publication from a group in South Africa, however, found that, in all 17 patients (all HIV infected) for whom adequate data were available, the development of MDR-TB or XDR-TB was the result of exogenous reinfection.3 This circumstance is, fortunately, not likely to apply to the United States, with its lower prevalence of MDR-TB and XDR-TB and its generally superior isolation facilities.


  1. Gandhi NR, et al. Extensively drug-resistant tuberculosis as a cause of death in patients co-infected with tuberculosis and HIV in a rural area of South Africa. Lancet. 2006;368:1575-1580.
  2. http://www.who.int/tb/features_archive/drs_factsheet.pdf
  3. Andrews JR, et al. Exogenous reinfection as a cause of multidrug-resistant and extensively drug-resistant tuberculosis in rural South Africa. J Infect Dis. 2008;198:1582-1589.

See also:

  1. Jassai M, Bishai WR. Extensively drug-resistant tuberculosis. Lancet Infect Dis. 2008 Nov 4 [Epub ahead of print].