Sharpen focus on TB blood tests

Scores near cutoff may be false positive

Using a blood test to screen health care workers for tuberculosis can cut your false positives by two-thirds, but it is critical to evaluate the numerical result on the test, according to members of a national TB testing task force.

While the Food and Drug Administration approved the QuantiFERON-TB Gold test with a cutoff value of 0.35, 1.1 is actually the most statistically significant trigger for recommending treatment for latent TB infection (LTBI) of health care workers, says Wendy Thanassi, MD, MA, chief of occupational health for the Palo Alto (CA) VA Health Care System.

Thanassi and colleagues at the University of Illinois at Chicago and the Cleveland Clinic have pooled data from 2,500 hospital workers to investigate the best treatment protocol for interferon gamma release assays (IGRAs). The Centers for Disease Control and Prevention is considering their input. (The T-Spot test, also FDA-approved, has similar issues surrounding its cut point.)

“The greatest variability in any result is right around the cutoff. Somebody is going to fall just to the left and just to the right,” says Thanassi, who is scheduled to speak on the topic at the annual conference of the Association of Occupational Health Professionals in Healthcare (AOHP), Oct. 3-6 in Las Vegas. “That’s, in fact, what we’re seeing in QuantiFERON.”

To avoid false positives, it is imperative to test only with an IGRA — and not to use the blood tests to “confirm” a TB skin test, says Thanassi. CDC does not recommend using both tests, but does allow for some confirmatory use.1

“If you do a TB skin test and 72 hours later you do a blood test, the blood test will be falsely positive. That [effect] will last up to six months,” she cautions.

The TB skin test has its own flaws, notes David Marder, MD, MPH, director of the University Health Services at the University of Illinois Hospital and Health Sciences System in Chicago, who is also speaking at the AOHP conference session.

“[Hospitals] think they’re saving money [by using the TST],” he says. “In the long run, they’re not. They’re missing some true positives because you’ll get false negatives with the skin test.”

They’re also getting many false positives, says Marder. In the first year that he implemented QuantiFERON-TB Gold, his positives dropped by two-thirds. QuantiFERON is especially useful in testing foreign-born employees who have been vaccinated with BCG because it is more specific.

First, do no harm

Testing hundreds of health care workers annually for latent TB infection has always been challenging because it is inherently a group that is at low risk of infection.

CDC guidelines call for hospitals to conduct a risk assessment and to restrict their annual TB screening to health care workers at “medium” risk of exposure. Employees in low-risk settings receive a baseline TB test at hire.2

Health care workers with a positive TB skin test have often been reluctant to take the standard treatment for LTBI, which was isoniazid for nine months. A new, 12-week regimen of directly observed therapy with isoniazid and rifapentine may improve treatment for LTBI, CDC says.3

But Thanassi and Marder caution that overtreating health care workers for LTBI can expose them to uncomfortable side effects and serious health risks, including liver damage. To “do no harm,” Marder prefers to monitor employees closely and retest them more frequently if their TB blood tests are above 0.35 but below 1.0.

“Even the people up to almost 1 we believe are low-positive,” he says. “They probably have been exposed, but I’m not convinced they’re conversions and need to be treated.”

To make things even more complicated, the blood test has resulted in some people moving from negative to positive and then, over time, back to negative.

“When we repeated the tests on people [with results] between 0.35 and 1.0, a lot of the people reverted to negative,” he says. “Maybe it was a transient infection. I’m not sure I even understand what latent TB is anymore.

“We think it also happens with a skin test, but when you had a positive, we never retested them.

There have been randomized trials where they retested skin tests and some of them reverted to negative. My guess is that this is a phenomenon that occurs immunologically. And we’re in the process of understanding why this is,” he says.

Treat all positives after exposure

Thanassi also takes a cautious approach to low-positives. In its most recent guidelines on the IGRAs, CDC emphasizes the importance of clinical judgment and says, “For healthy persons who have a low risk for both infection and progression, discounting an isolated positive result as a false positive is reasonable.”

“This is an outstanding test, vitally important for the world of tuberculosis,” says Thanassi. “For this low risk, low prevalence, serially tested population, it has to be looked at differently and with a relatively critical eye.”

“If they have no risky travel and no known contact and they’re under 1.1, I just retest them the next year. If they are over 1.1, I would call them in for a retest right away. If they stay in the gray zone I would retest them every year,” she says.

“If they went up and were high, I would send them for infectious disease consultation for consideration of treatment [for LTBI],” she says.

There is an important caveat: In the case of a known exposure, even a positive result near the test cutoff would trigger treatment, says Marder.

“I don’t want to take a chance in those settings. The cost would be too high if we were wrong,” he says. “Your QuantiFERON [result] can go up slowly. Maybe I just caught it at the spot where it’s converting and it’s just going up.”

Through further research, experts hope to learn more about how to interpret the TB blood test results. But Marder says that the current confusion doesn’t reflect poorly on the test itself.

“This is clinical medicine,” he says. “You have a new test. Now you have new information. We have to figure out what it means and what we’re going to offer to people [in treatment].”


1. Centers for Disease Control and Prevention. Updated guidelines for using Interferon Gamma Release Assays to detect Mycobacterium tuberculosis infection - United States, 2010. MMWR 2010; 59(RR-5):1-25.

2. Centers for Disease Control and Prevention. Guidelines for Preventing the Transmission of Mycobacterium tuberculosis in Health-Care Settings, 2005. MMWR 2005; 54(No. RR-17):1-121.

3. Centers for Disease Control and Prevention. Recommendations for use of an Isoniazid-Rifapentine regimen with direct observation to treat latent Mycobacterium tuberculosis infection. MMWR 2011; 60;1650-1653.