Journal Review

False positives found when skin-testing workers

Researchers cite problems with TST reagent

Blumberg HM, White N, Parrott P, et al. False-positive tuberculin skin test results among health care workers. Research letters. JAMA 2000; 283:279.

The authors report false-positive tuberculin skin test (TST) results associated with one of two commercially available reagents, concluding that when a TST is used among a low-risk population — such as health care workers in an institution that has a well-functioning tuberculosis infection control program — the majority of positive results actually may be false.

Two different commercial tuberculin reagents are available in the United States: Aplisol (Parkdale Pharmaceuticals, Rochester, MI) and Tubersol (Pasteur Mérieux Connaught USA, Swift- water, PA). The authors cite a previous study that concluded either product will correctly classify comparable numbers of people not infected with TB and "the choice of product used for [tuberculin] skin testing has little effect on test performance."

Atlanta skin test contradicts study

The authors of the current paper note, however, that "experience at our institution with false-positive tuberculin skin test (TST) results among a group of health care workers, which was associated with the use of Aplisol, contradicts these conclusions."

They investigated a marked increase in TST conversions among health care workers at the Grady Health System in Atlanta that began in mid-September 1999. Tuberculin skin tests are mandatory for all health care workers every six months unless there is documentation of a previously positive result. Tests are placed and read by the hospital’s employee health service using the Mantoux method; self-reading is not permitted. Two-step testing is required for newly hired employees. A positive TST is defined as induration of 10 mm or greater.

They found that in the three six-month periods before September 1999, baseline conversion rates were as follows: four (0.09%) of 4,670 health care workers tested between January and June 1998; eight (0.2%) of 4,363 tested from July through December 1998; and five (0.1%) of 4,358 tested from January through June 1999. Between Sept. 15 and Oct. 15, 1999, 11 (1.2%) health care workers had a new TST conversion (median, 12 mm induration; range, 10-20 mm) among 914 individuals tested, a marked increase compared with the previous 1.5 years.

"All 11 health care workers were subsequently found to have chest radiographs that showed no evidence of tuberculosis," the authors noted. "Our investigation found that the pharmacy had switched from Tubersol, which had been used exclusively for the prior six years, to Aplisol because of lower price. . . . All 11 health care workers who tested positive with Aplisol were retested with Tubersol PPD and had a negative TST."


Researchers develop more sensitive TB test

Franchi A, Amicosante M, Rovatti E, et al. Evaluation of a Western blot test as a potential screening tool for occupational exposure to Mycobacterium tuberculosis in health care workers. J Occup Environ Med 2000; 42:64-68.

Is there a better way to test health care workers for occupational exposure to tuberculosis? Researchers at the University of Modena in Italy say there is: using a Western blot test to detect an antibody as a marker of exposure to TB.

Currently, skin tests measure the reaction to tuberculin purified protein derivative (PPD) to screen health care workers for risk of TB infection. The researchers developed an M. bovis serological Western blot test as an earlier marker of TB contact.

The antibody test could not be used with BCG-vaccinated workers due to their high reactivity. But among nonvaccinated health care workers, the Western blot test did, in fact, detect exposure earlier than the PPD skin test and with greater sensitivity. For example, the Western blot test identified 95% of workers in the tuberculosis and respiratory diseases division as being sensitized to Mycobacterium tuberculosis, compared with 73% identified by PPD. In the infectious disease division, Western blot identified 59% of workers as sensitized, compared with 41% identified by PPD.

On the downside, the authors noted that the Western blot methodology "would give more limited information about the level or intensity of MTB-exposure than the PPD skin test.

"Overall, this study suggests that the WB test antibody market, as a sensitive indicator of MTB contact among exposed HCWs, might provide, in association with the PPD skin testing, new tools to assess the TB risk in health care facilities with higher accuracy, thus allowing a more timely and appropriate implementation of the environmental and health surveillance measures for the primary prevention and control of TB infection in the workplace," the authors stated.