September 13, 2010

New Test Detects TB in Less than 2 Hours

Microscope image of yellow-stained TB bacteria Smear microscopy, the most widely used diagnostic test for TB, reveals yellow-glowing TB bacteria in a patient's sputum. Scientists have developed a new automated TB test that's faster and more accurate than current diagnostic testsRonald W. Smithwick, Centers for Disease Control and Prevention

Scientists have developed an automated test that can rapidly and accurately detect tuberculosis and drug-resistant TB bacteria in patients. The finding could pave the way for earlier diagnosis and more targeted treatment of this disease.

TB kills about 1.8 million people each year, mainly in developing countries, and drug-resistant TB is a growing threat. Early detection and treatment helps stop the disease from spreading, but current diagnostic tests have many shortcomings. The most widely used test, called smear microscopy, misses more than half of TB cases and can’t determine whether the bacteria are drug resistant. A more sensitive test involves growing bacterial cultures. It can spot drug resistance but may take up to 6 weeks to get results. Both tests require assessment by trained staff.

To develop a faster and easier-to-use test, researchers led by Dr. David Alland of the University of Medicine and Dentistry of New Jersey collaborated with Cepheid, a diagnostics company, to create a DNA-based test called Xpert MTB/RIF. The test detects the TB-causing bacterium Mycobacterium tuberculosis and also resistance to rifampin (RIF), one of the most common treatments. RIF resistance is a good indicator of multidrug resistance. Drug-resistant TB requires different treatment than drug-susceptible TB.

Development of the MTB/RIF test has been supported for more than 8 years by NIH’s National Institute of Allergy and Infectious Diseases (NIAID). To use the test, a technician adds a small sample of a patient’s sputum to a plastic test cartridge and loads it into the machine. The instrument then automatically performs a series of steps that ultimately leads to an analysis of DNA from bacteria in the sample. A computerized printout reports the presence of TB bacteria and whether or not the bacteria are resistant to RIF.

As described in the September 1, 2010, issue of the New England Journal of Medicine, the researchers assessed the performance of the new automated test on 1,730 patients with suspected TB in 4 countries. Each patient provided 3 sputum specimens. The samples were assessed by MTB/RIF and by conventional smear microscopy and bacterial culture tests. The clinical trial was led and supported by the Foundation for Innovative New Diagnostics, a Swiss-based nonprofit organization.

The new automated test successfully identified 98% of all confirmed TB cases and 98% of patients with RIF-resistant bacteria in less than 2 hours. In addition, a single MTB/RIF analysis detected TB in over 72% of patients who did not appear to have TB according to smear microscopy but who were later found to have TB in culture tests. When the automated test was repeated, the sensitivity increased by about 13%. When the test was run a third time, it detected about 90% of TB cases that were missed by smear microscopy.

The scientists note that the MTB/RIF test makes it possible to detect TB and drug resistance in a single clinic visit and perhaps begin treatment immediately, a significant advantage in developing countries. “The test also indicates rapidly whether difficult-to-treat drug-resistant forms are present,” says Alland. “This is a major advance over other rapid TB detection methods, which are complex, labor-intensive, and technically challenging.”

—by Vicki Contie

Related Links