Diagnostic accuracy of the TrueNat™ MTB plus assay for detecting pulmonary tuberculosis in adults

Abstract

Tuberculosis (TB), caused by Mycobacterium tuberculosis remains a global threat, affecting 10.8 million people and causes 1.3 million deaths annually. Over 2.5 million cases go undiagnosed partly due to current diagnostic limitations. In particular, smear microscopy is less sensitive, culture is slow and prone to contamination, and the rapid Xpert® MTB/RIF Ultra (Ultra) needs advanced infrastructure. This study assessed the diagnostic accuracy of TrueNat™ MTB Plus assay (TrueNat), a portable, WHO-endorsed point-of-care tool, compared to Ultra, smear microscopy, and Löwenstein–Jensen (LJ) culture. This cross-sectional study enrolled 260 consenting adult participants (≥18 years) with presumptive TB in northern Tanzania. Participants’ sputum samples were tested for M. tuberculosis using smear microscopy, LJ culture, Ultra and TrueNat. TrueNat performance was assessed using sensitivity, specificity, predictive values and area under the curve (AUC) against the standard-of-care and a composite reference standard. Age and body-mass-index were summarised using median and interquatile range (IQR). Categorical variables were reported as proportions. Multivariate logistic regression identified TB predictors (p<0.05). Data analysis and visualization were conducted using R. Among 260 participants, 165 (63.5%) were male, with a median age of 46.0 years (IQR: 35.5–57.3); 15 (5.8%) were HIVpositive, and 52 (20%) had undernutrition. TB was detected in 109 (41.9%) participants by at least one test. TrueNat showed sensitivity/specificity of 34.4%/94.7% vs. smear microscopy, 98.9%/95.3% vs. culture, and 86.2%/95.2% vs. Ultra. The AUC values were 0.75, 0.92, 0.96, and 0.91 compared to smear microscopy, LJ culture, Ultra, and a composite reference method, respectively. Participants reporting weight loss were 2.84 times more likely (95% CI: 1.68–4.84, p<0.001) to test positive for TB by TrueNat. The TrueNat compared favorably to both Ultra and culture in terms of diagnostic accuracy, offering the added benefits of faster results and greater suitability for resource-limited settings. These findings suggest that TrueNat is a promising test for rapid TB detection in low-resource settings, warranting further studies to evaluate its implementation and impact on patient management.