In the Journals

Xpert TB test showed limited effect in high-resource, low-incidence settings

Show Citation

January 24, 2014

The Xpert test for tuberculosis demonstrated lower sensitivity for less extensive disease and, as a result, may have limited effect in high-resource, low-incidence settings, researchers reported in Clinical Infectious Diseases.

According to Claudia Denkinger, MD, PhD, of McGill University and McGill International TB Cenre in Montreal, in the developing world, the Xpert MTB/RIF assay (Cepheid) has a detection rate of close to 90% of all pulmonary TB on expectorated sputum. However, its performance in the developed world has not been studied previously and currently, only data from evaluation for FDA approval report in the package insert are available.

Claudia Denkinger, MD, PhD 

Claudia Denkinger

Xpert MTB/RIF resulted in only marginally improved detection and time to diagnosis over standard smear- and culture-based diagnostic algorithms when it is done on unprocessed, induced sputum in a high-resource, ambulatory, tertiary-care setting where clinician expertise is high and patients present early in their disease,” Denkinger told Infectious Disease News.

Denkinger and colleagues evaluated the accuracy of the Xpert test among 502 consecutive patients who presented with possible active TB from October 2011 to May 2013. Most of the patients (74%) had an abnormal chest X-ray and only 18% had symptoms that suggested active TB. Twenty-five of the patients had culture-confirmed active TB. Eleven were smear-positive, but only seven of those had TB.

Among the 25 cases, the Xpert test had a sensitivity of 46% and a specificity of 99.8%. The sensitivity was higher in smear-positive patients (86%) compared with smear-negative patients (29%). Two of the isolates were identified as rifampin-resistant on the Xpert test, but only one was confirmed resistant by culture. The other was confirmed to have a resistant mutation on sequencing.

Most of the culture-positive but Xpert-negative patients had minimal disease. Eleven of the patients had no symptoms on presentation. The researchers also observed a longer time to culture positivity among those who were Xpert-negative but culture-positive: 28 days vs. 14 days for Xpert-positive, culture-positive patients.

“Considerations about the implementation of Xpert should keep the already existing infrastructure in mind and consider the added benefit of a test,” Denkinger said. “While our study showed limited benefit in a tertiary center with high resources, Xpert may still have an important role in settings where smear microscopy is not done on site, other nucleic acid amplification tests for TB are not available and caregivers are less experienced in diagnosing TB.”– Emily Shafer

Claudia Denkinger, MD, PhD, can be reached at: McGill University, McGill International TB Centre, 1020 Pine Ave West, Montreal, QC H3A 1A2, Canada; email: Claudia.denkinger@mail.mcgill.ca.

Disclosure: The researchers report no relevant financial disclosures.