Questions raised about TB blood tests
Expert judgment needed in some cases
Nine years after the Food and Drug Administration approved the first blood test to detect latent tuberculosis infection, hospitals are still struggling to determine how to use the tests or whether to use them at all.
At presstime for HEH, the Centers for Disease Control and Prevention was poised to release a new guideline for assessing the results of the interferon-gamma release assay (blood test). The agency is likely to acknowledge some difficulties in interpreting the test, particularly among low-risk individuals who are near the cut-off for a positive blood test.
The Food and Drug Administration has approved three TB blood tests that are currently available: QuantiFERON-TB Gold and QuantiFERON-TB Gold In-Tube by Cellestis of Valencia, CA, and T-SPOT TB by Oxford Immunotec of Marlborough, MA.
In the QuantiFERON tests, whole blood is mixed with antigens synthetic peptides that represent specific M. tuberculosis proteins that are not in any BCG vaccine strain and are not in most non-tuberculous mycobacteria. The test measures the release of interferon-gamma, a protein produced by white blood cells in response to the antigens. The T-SPOT test measures the number of white blood cells that produced interferon-gamma.
"Most people who are sensitized to these antigens due to prior Mtb infection will produce interferon-gamma and be positive, but those without prior infection will not be sensitized and will be negative by these tests," explains Gerald Mazurek, MD, medical officer and epidemiologist in CDC's Division of TB Elimination.
"In certain populations, there's clear evidence that there's improved specificity for IGRAs, but there's some lingering questions about sensitivity compared to the tuberculin skin test," he says.
Many hospitals have used the blood test to confirm a positive skin test, although previous CDC guidelines did not recommend using both tests. When an individual who is low-risk for tuberculosis has a positive skin test, many experts now believe "it's reasonable to confirm or refute it with another test," says Mazurek.
"If [the blood test] is negative, it doesn't mean you can completely discount the positive skin test, but it makes it less likely that the person is infected with tuberculosis," he says. "If you have a positive skin test and a negative IGRA, one might consider following [the employee] and repeating the test to see if they become consistently positive."
One caveat: "Extremely immune-compromised" individuals should be considered infected if they are positive by either test, says Mazurek. Further evaluation is necessary to determine if treatment is appropriate, he says.
The blood test is most compelling when used with employees who have been vaccinated with BCG. Because the blood test is more specific, BCG will not trigger a positive result. Employees with a history of BCG may have more faith in the blood test and may be more likely to initiate treatment for latent tuberculosis infection, Mazurek says.
Concern about false positives
Hospitals have had variable experiences with the new blood tests. The OSF Saint Francis Medical Center in Peoria, IL, wanted to move away from tuberculin skin test screening, which requires two visits for placement and reading of the results. Yet the first screening of about 6,500 employees produced 287 positives. Of those, 123 had previously been TST-positive, but that still left 164 new positives.
"That's quite a high number for us. We would never expect to see that many new TST responders," says William S. Scott, MD, MPH, medical director of the hospital's Center for Occupational Health and clinical assistant professor at the University of Illinois College of Medicine-Peoria.
The hospital retested the positive employees, and 66 of them reverted to normal. "That's a reversion rate of almost 50%. That's very concerning," says Scott. "It's the flip of a coin."
Scott suspects that the problem occurs with low-risk individuals whose results are near the cut-off. If the cut-off were raised from .35 to 1 international units per milliliter of blood, the conversion rate would be within an expected range, he says.
In fact, Mazurek says it's reasonable to consider risk or potential TB exposure when interpreting blood test results, just as the evaluation of induration in a skin test may include risk factors.
Despite the investment in the QuantiFERON test, Saint Francis Medical Center returned to skin testing and now uses the blood test as a confirmatory test or with BCG-vaccinated individuals. "It makes sense to use it in a targeted fashion rather than as a massive screening tool," Scott says.
At the Cleveland (OH) Clinic, a switch to QuantiFERON-TB Gold In-Tube initially was plagued with a high number of indeterminate results. The problem was with lab technique, says infectious disease physician Cyndee Miranda, MD. With help from the manufacturer and an internal review and training, the lab process improved and indeterminate rate is now low, she says.
There is still a concern about new converters who are near the cut-off and don't have a known exposure or risk factors, she says. "We're not sure they're really positive," she says. Some tests will be repeated or the individuals may be monitored rather than treated, she says.
The Marshfield (WI) Clinic now uses QuantiFERON-TB Gold In-Tube and finds fewer false positives than with the skin test, says Bruce Cunha, RN, MS, COHN-S, manager of employee health and safety. Per CDC recommendations, the clinic limits its annual screening program because it is in a community that is low-risk for TB.
Still, in some cases, infectious disease physicians make a judgment call, he says. "The closer you are to the cutoff value, the greater the potential that an employee with have an odd reading," he says. "But we see odd readings with the skin test, too."
The bottom line, says Mazurek: "It's not a perfect test. It took 100 years to refine the skin test and it's still not perfect. I think we've accomplished a lot with the blood tests. At some point we'll have a better understanding of the tests and how and when to use them."