Research

August 10, 2011

TB antibody detection tests fail to diagnose tuberculosis accurately

Commercially available serological tests fail to accurately diagnose active tuberculosis (TB) and they are not as cost effective as other recommended TB tests, according to two papers published Aug. 9 in the Public Library of Science (PLoS) Medicine.

People wrongly diagnosed as TB-free may not get treated and may develop more severe illness, as well as continue to spread the disease.  For a person who actually does not have TB, a false-positive result could lead to unnecessary treatment, possible side effects, delay in obtaining the correct diagnosis and a costly waste of medical resources.

“Misdiagnosis remains a major obstacle for control of the TB epidemic,” said Dr. Karen Steingart, affiliate professor of health services at the UW School of Public Health, “despite great advances made in controlling TB over the past decade. Unfortunately, for active TB, we do not have a simple, accurate, point-of-care test, as we have for infections like HIV or malaria.” Steingart led one of the studies.

Serological tests have never been recommended by any international guideline committee. In fact, the International Standards for TB Care discourages their use. Nevertheless, millions of these tests are performed each year in many parts of the world, especially in areas with weak regulatory mechanisms for diagnostics. A recent survey shows that serological tests for TB are being sold in Afghanistan, Bangladesh, Brazil, Cambodia, China, India, Indonesia, Kenya, Myanmar, Nigeria, Pakistan, Philippines, Russia, South Africa, Thailand, Uganda and Vietnam.

Based on the evidence in the PLoS Medicine papers and expert opinion, the World Health Organization (WHO) has concluded that present commercial serological tests provide inconsistent and imprecise estimates of sensitivity and specificity.

Sensitivity is the percentage of patients with a positive serological test result among patients with TB confirmed by a reference standard; a high sensitivity means that the test detects most patients with TB. Specificity is the percentage of patients with a negative serological test result among people without TB; a high specificity means the test gives few false-positive diagnoses.

WHO has issued a policy statement strongly recommending that these tests not be used with either adults or children to diagnose active pulmonary TB or extrapulmonary TB, which occurs when TB affects parts of the body other than the lungs. This is the first time WHO has released a negative policy asking health officials worldwide not to use a particular type of test or procedure.

At the same time, WHO clearly does not want to restrict innovative research for better antibody detection tests for active TB, because accurate, rapid tests eventually could be developed.

“The WHO policy encourages future research on new or improved antibody detection tests and point-of-care tests for TB,” Steingart said.

An evaluation performed by the United Nations Special Programme for Research and Training in Tropical Disease (TDR) found that none of the commercially available serological tests investigated were good enough to replace conventional tests for TB, such as sputum microscopy, nor did they help as an extra check to rule out TB.

In their paper, “Commercial Serological Tests for the Diagnosis of Active Pulmonary and Extrapulmonary Tuberculosis: An Updated Systematic Review and Meta-Analysis,” Steingart and her colleagues appraised the methodological quality of individual studies and rated the overall quality of the body of evidence as very low.

For pulmonary TB, the team identified 67 studies, including 32 studies from low- and middle-income countries. They found sensitivity and specificity from individual studies to be erratic. Sensitivity values ranged from 0 to 100 percent, and specificity from 31 to 100 percent.

The most widely used TB test showed sensitivity of 76 percent and specificity of 92 percent in patients whose sputum smears were positive for TB, and sensitivity of 59 percent and specificity of 91 percent in patients whose sputum smears were negative.

For extrapulmonary TB, the team identified 25 studies, including 10 from low- and middle-income countries. They once again noted the erratic sensitivity and specificity values from individual studies, with sensitivity values ranging from 0 to 100 percent, and specificity, 59 to 100 percent.

Steingart commented, “The new WHO policy recommendation against using commercial serological tests for the diagnosis of active TB is a major step.  It is now up to countries with a high burden of TB to implement this policy by issuing regulations and educating doctors, laboratory personnel and consumers to prevent continued use of inferior and harmful diagnostic tests.”

Dr. Madkukar Pai of McGill University in Montreal was the senior author of both papers. He is from the Respiratory, Epidemiology and Clinical Research Unit at the Montreal Chest Institute. Dr. David Dowdy of Johns Hopkins University in Baltimore was the first author on the second paper describing a cost-effectiveness analysis of serological testing versus other diagnostic techniques in India.  Dowdys study showed that these tests result in more secondary infections and false-positive diagnoses than sputum smear microscopy, while increasing per-patient costs.

In addition to Steingart and Pai, researchers involved in  the systemic review and meta-analysis were Laura Flores and Philip C. Hopewell of the University of California, San Francisco and its Curry International Tuberculosis Center; Nandini Dendukuri and Ian Schiller of McGill University; Suman Laal of New York University Langone Medical Center and the Veterans Affairs Medical Center in  New York City; and Andrew Ramsay of the UNICEF/UNDP/World Bank/WHO Special Programme  for Research and Training in Tropical Diseases (TDR), World Health Organization, Geneva, Switzerland.

The systematic review was commissioned by the World Health Organization with funding provided by USAID through a grant administered by TDR.

The researchers declare no competing financial interests.