CDC tracks highly virulent, multiresistant TB strains

One infects quickly; another resists treatment

While tuberculosis has undergone four consecutive years of decline in the United States, there are still some highly virulent and multidrug-resistant strains in circulation, the Centers for Disease Control and Prevention (CDC) reports.

The CDC recently reported a strikingly virulent, drug-susceptible TB "strain O" capable of infecting even limited, casual contacts of active cases. In addition, the CDC continues to track the national dissemination of multidrug-resistant "strain W" from its origins in nosocomial outbreaks in New York City.

The reports may bolster support for the TB standard proposed by the Occupational Safety and Health Administration during the upcoming hearings and debate on the standard, although CDC officials say the emergence of the strains really do not change any of the currently recommended infection control measures. The TB strains were described at the Infectious Disease Society of America conference in San Francisco. The highly virulent but drug-susceptible strain O caused a large community outbreak in a rural area reportedly along the Kentucky-Tennessee border in 1994-1996.1

With a population of about 10,000 people, the area had reported fewer than one TB case annually over the prior 10 years. Yet during the outbreak period, 21 TB cases (15 culture-positive) were identified, and DNA fingerprints of the 13 available isolates were identical.

To determine extent of transmission, investigation was conducted among close and casual contacts of the cases. Virulence studies using mouse models were also conducted on the outbreak strain of TB. The source case, diagnosed in 1994; "Case B," diagnosed in 1995; and "Case C," diagnosed in 1996, were the sources of infection for the other 18 cases, the CDC determined. Five secondary cases became infected and developed active disease with "only very limited, casual exposure" to the source cases, the CDC reported. Extensive transmission from the three cases to close and casual contacts was also found, as 337 (73%) of 461 contacts had positive TB skin tests.

"We don’t normally see transmission like this in outbreaks," says Sarah Valway, MD, medical epidemiologist in the CDC division of TB elimination. "This one was very clearly different. Typically we see prolonged, close contact. [In the outbreak], we would see causal contact, and there would be transmission."

Mouse studies with the strain revealed an extraordinary rate of growth for the outbreak strain, one that greatly exceeds that seen with other clinical isolates of TB, Valway reports. The data suggest increased transmission was due to increased virulence, rather than environmental factors or patient characteristics.

"There is still a lot more the laboratory people are working on to try to see what is different about this particular strain," Valway tells Hospital Infection Control. "Mycobacterium TB divides [approximately] once every 24 hours. In mouse models, over time, the extent of growth was about 1,000 times greater than other strains."

Nonetheless, Valway says the discovery may be more a reflection of improved laboratory diagnostics and the low background rate of TB in the region rather than the emergence of a new TB strain. The strain has not been detected in any other region.

"I suspect that things like this have happened in the past, but there has never been the laboratory support to be able to look at these things," she says. "If this were happening in a big urban city with a lot of TB, you wouldn’t necessarily notice it."

While posing a threat to causal contacts would certainly merit concern should an undiagnosed case appear in a hospital, Valway notes that patients rapidly respond to drug therapy because the strain is not resistant.

"This is a fully susceptible strain," she says. "Somebody who is in the hospital being treated for TB would respond just as quickly [as with other strains]."

While strain O caused a community outbreak, the highly drug-resistant strain W caused several nosocomial outbreaks in New York in the early 1990s. Mortality rates in the outbreaks exceeded 80%, the CDC reports. The strain is resistant to isoniazid, rifampin, streptomycin, ethambutol, ethionamide, kanamycin, and rifabutin, and has a unique multi-banded DNA fingerprint. Subsequent cases continue to be seen in other areas as people travel and move from New York, she notes. For example, a recent report describes nosocomial transmission of the strain during bronchoscopy procedures at hospital in a South Carolina.2 (See Journal Reviews, p. 190.)

"It is resistant to all first-line drugs and a bunch of second-line," she says. "But there is not that much outside of New York City. There are sporadic cases cropping up across the country. So far, we could trace them all back to New York City or state."

To identify TB cases due to Strain W outside New York state, the CDC reviewed the national TB surveillance database, regional DNA fingerprint laboratories, and its own mycobacteriology laboratory data and epidemiologic reports. As of September, when the data were reported at the IDSA conference, 22 cases have been diagnosed in nine states and Puerto Rico since 1992. Of these cases, 13 were HIV-positive, and three (including one HIV-positive) were health care workers.

Overall, 15 (11 HIV-positive) of the 22 cases died while on anti-TB therapy. The CDC determined that 12 of the 22 cases were exposed to Strain W in New York City prior to their diagnoses. Four of those cases then transmitted TB to eight other people. The source of infection for the remaining two cases has not yet been identified. To aid in continued tracking and containment of the strain, all health departments should notify the CDC of all TB cases resistant to isoniazid, rifampin, streptomycin, and kanamycin.


1. Grady D. New strains of highly contagious tuberculosis identified in the Smokies. NY Times, Sept. 23, 1997:F-3

2. Agerton T, Valway S, Gore B, et al. Transmission of a highly drug-resistant strain (strain W1) of Mycobacterium tuberculosis: Community outbreak and nosocomial transmission via a contaminated bronchoscope. JAMA 1997; 278:1,073-1,077.