Xpert MTB/RIF assay more sensitive than smear microscopy for diagnosing tuberculosis

1. A systematic review and meta-analysis of studies comparing the Xpert MTB/RIF assay to smear microscopy for the diagnosis of tuberculosis in children found that Xpert was much more sensitive at diagnosis than smear microscopy, with both tests having very high specificity.

2. Xpert was much less sensitive, however, than the reference standard of culture, but a negative Xpert test did not rule out a diagnosis of tuberculosis. 

Evidence Rating Level: 1 (Excellent)

Study Rundown: The global burden of childhood tuberculosis is underestimated due to difficulties in diagnosis. Smear microscopy is the most used method of diagnosing tuberculosis, especially in poorer countries without access to laboratories, however, it is difficult to collect adequate sputum from children and has low sensitivity. This study conducted a systematic review and meta-analysis of the sensitivity and specificity of the Xpert MTB/RIF assay, a rapid PCR test developed to improve methods of tuberculosis diagnosis, compared to microscopy, against a reference standard of culture in the diagnosis of tuberculosis in children.

15 studies were analyzed, ranging from those done in low-income countries to high-income countries, consisting of nearly 5000 specimens from 3,640 children who were examined for pulmonary tuberculosis. The Xpert assay was shown to diagnose tuberculosis with similar efficacy in expectorated and induced sputum samples compared to gastric lavage samples, but was around 40% more sensitive than smear microscopy in detecting tuberculosis in each sample types. However, compared to the reference standard of culture, Xpert still remained less sensitive in diagnosis, but a negative Xpert test did not rule out a diagnosis of tuberculosis. Limitations of this study included that the analyzed studies examined very different populations of children mostly treated as inpatients versus outpatients and that culture is an imperfect reference standard especially in children due to variations in specimen quality, age, or disease severity. Strengths of this study were the inclusions of additional analyses for age, HIV status of children, and comparing smear test status.

This study was funded by the WHO and Global TB Program of Texas Children’s Hospital

Click to read the study in The Lancet Respiratory Medicine

Relevant Reading: Assessment of the Xpert MTB/RIF assay for diagnosis of tuberculosis with gastric lavage aspirates in children in sub-Saharan Africa: a prospective descriptive study

In-Depth [systematic review and meta-analysis]: 15 studies containing 4,768 respiratory specimens from 3,640 children were analyzed in this study (median 265 specimens per study). 12% (420 of 3,640) of all children had cultures positive for tuberculosis and 11% (406 of 3,640) were positive with Xpert. 44% (668 of 1576) of children with negative cultures were started on empiric anti-tuberculosis therapy, while only 2% (10 of 668) tested positive with Xpert. The sensitivities of Xpert varied considerably between studies, while the specificities were consistently much higher, ranging from 93-100%, with similar results for microscopy. The sensitivities of Xpert were similar between sputum samples (62%, 95% Credible Interval [CI] 51-73%) and gastric lavage samples (66%, 51-81%), but were much more sensitive than microscopy (sputum: 26%, 14-39%; gastric lavage: 22%, 12-35%). Specificities remained high for each, with greater than 98% for Xpert and greater than 99% for microscopy for each sample type. Additionally, six studies were used for rifampicin resistance testing, comparing Xpert to culture. 11 (4.6%) of 240 children were resistant by culture and 10 (4.2%) were resistant by Xpert, with four children determined negative for resistance who had indeterminate Xpert results. The sensitivity of Xpert for detecting rifampicin resistance was 86% (95% CI 53-98) and the specificity was 98% (94-100).

Image: PD

©2015 2 Minute Medicine, Inc. All rights reserved. No works may be reproduced without expressed written consent from 2 Minute Medicine, Inc. Inquire about licensing here. No article should be construed as medical advice and is not intended as such by the authors or by 2 Minute Medicine, Inc.