False-positive TB tests: Is the reagent at fault?

Improper reading of skin tests also a problem

False-positive reactions to a tuberculosis skin testing reagent are causing a stir among employee health departments.

In one month after switching from Aplisol (Parkdale Pharmaceuticals, Rochester, MN) to Tubersol (Pasteur Merieux Connaught USA, Swiftwater, PA), 11 health care workers at the Grady Health System in Atlanta had a new tuberculin skin test (TST) conversion. Chest radiographs showed no evidence of tuberculosis, and retesting with the Tubersol purified protein derivative (PPD) came out negative.1

Employee health and infection control professionals concluded that the false positives were the result of the switch.

In a letter to the Journal of the American Medical Association, Henry Blumberg, MD, an infectious disease specialist at Grady and the Atlanta-based Emory University School of Medicine, and colleagues disputed a recent published study concluding that Aplisol and Tubersol are equivalent.2

"Our experience demonstrates the need for a better and more reliable test for detection of tuberculosis infection and suggests that when a TST is used among a low-risk population . . . the majority of positive results actually may be false-positives," the letter stated.

Questions about false positives with Aplisol have lingered for years. Employee health professionals have reported itching and redness associated with the reagent. But does Aplisol produce an induration that is significantly larger than what Tubersol would produce in the same person?

According to a comparative study conducted by Margaret Villarino, MD, MPH, at the Centers for Disease Control and Prevention, and colleagues at CDC and state health departments, there were no significant differences between the reaction sizes of the two reagents when used in low-risk populations.3

Some tuberculosis experts assert that "false positives" actually stem from improper reading of the skin test reactions and not from a problem with the reagent.

In one study, pediatricians, pediatric academicians, and nurses were asked to read the Mantoux tuberculin reaction of a known tuberculin converter. Only 7% read it correctly.3

"People don’t know how to read them very well at all," says Lee B. Reichman, MD, MPH, executive director of the National Tuberculosis Center at the University of Medicine and Dentistry of New Jersey in Newark. The National Tuberculosis Center provides training to hospitals and health departments across the country. "It’s very disappointing. If people are properly and appropriately trained and experienced, then you get better readings."

Edward Nardell, MD, tuberculosis control officer at the Massachusetts Department of Public Health and associate professor of medicine at Harvard Medical School in Boston, agrees that the skill level of people reading the skin tests may have made a difference in the Villarino study. But he notes that employee health departments must conduct the tests in a time-pressured, real-world environment — not in a controlled study.

"I think you can train a research nurse or physician to tell the difference between a soft induration you get with Aplisol and the harder induration you get with Tubersol, and to ignore the erythema [redness]. You seem to get a lot more [redness] with Aplisol," says Nardell, who also practices at Cambridge Hospital.

"I personally have seen an intern who was told by employee health that she had a positive PPD, having previously tested negative," he says.

"I looked at it and it was pretty impressive, with a lot of redness and some soft induration. I had her retested with Tubersol and there was no reaction at all. I think the majority of occupational health nurses and physicians would call that positive," Nardell adds.

The Massachusetts Department of Health, which provides tuberculin statewide, is switching back to Tubersol, despite the increased cost of that reagent. That change includes the employee health department at Cambridge Hospital.

"We’re in a low-prevalence situation in most of the country for tuberculosis," says Nardell. "The last thing in the world you want is an overly sensitive test."

Reichman suggests that employee health departments using Aplisol focus on proper reading of skin tests. "They should choose an antigen — don’t change the antigen — and make sure it’s read properly," he says. "You can’t compare people unless the antigen stays the same."

References

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

2. Villarino ME, Burman W, Wang YC, et al. Comparable specificity of two commercial tuberculin reagents in persons at low risk for tuberculous infection. JAMA 1999; 281:169-171.

3. Kendig EL Jr., Kirkpatrick BV, Carter WH, et al. Underreading of the tuberculin skin test reaction. Chest 1998; 113:1,175-1,177.