Diagnostic Approaches for Tuberculosis Detection

Tuberculosis (TB) diagnosis has evolved from reliance on clinical presentation and microscopy towards molecular and imaging technologies that offer greater sensitivity, specificity and speed. Traditional methods such as sputum smear microscopy and culture remain cornerstones, particularly in resource-limited settings, but they suffer from low sensitivity in paucibacillary and HIV-associated disease and prolonged turnaround times. Nucleic acid amplification tests (NAATs), including cartridge-based real-time PCR, have reduced detection times to under two hours while simultaneously screening for rifampicin resistance. Loop-mediated isothermal amplification (LAMP) has been introduced as a lower-cost molecular option suited to decentralised laboratories. Advances in metagenomic sequencing permit comprehensive pathogen detection and resistance profiling directly from clinical specimens, although they currently require sophisticated infrastructure. Digital chest radiography, increasingly paired with artificial-intelligence algorithms, offers non-invasive triage and case-finding in high-burden communities. Emerging point-of-care biosensors, including CRISPR-based fluorescence assays, hold promise for rapid, equipment-lean detection. The integration of these modalities into tiered diagnostic networks aims to achieve early case identification, guide effective treatment and interrupt transmission at a population level.

Research from Nature Portfolio

Recent studies have demonstrated the application of CRISPR-Cas systems to tuberculosis diagnosis, employing guide-RNA-directed detection of Mycobacterium tuberculosis complex DNA in sputum samples in under 60 minutes. These assays have achieved sensitivities approaching those of culture and maintain specificity in the presence of non-tuberculous mycobacteria. Complementing molecular advances, metagenomic next-generation sequencing directly from sputum has been shown to detect TB bacilli and simultaneously profile drug-resistance mutations, reducing the need for separate phenotypic assays. In the imaging domain, deep-learning networks trained on large, geographically diverse chest radiograph datasets have recently achieved triage accuracies above 90%, enabling automated flagging of presumptive cases for confirmatory testing. Collectively, these developments underscore a shift towards rapid, multiplexed and data-driven diagnostic paradigms.

Research from all publishers

A large cluster-randomised trial evaluated targeted universal testing interventions in South African primary care clinics. By offering sputum testing irrespective of symptoms to defined high-risk groups, the study observed a relative increase of 17% in monthly TB case detection compared with symptom-directed strategies, highlighting the limitations of symptom screening alone. A systematic review and meta-analysis of loop-mediated isothermal amplification for pulmonary TB diagnosis examined over 4,700 patients across multiple settings. It found LAMP to have significantly higher sensitivity than smear microscopy and comparable specificity to cartridge-based NAATs, suggesting its suitability as an alternative in laboratories that lack PCR infrastructure. These findings reinforce the need for diversified diagnostic algorithms that combine imaging, molecular and targeted screening approaches to capture subclinical and smear-negative disease.

Diagnostic Approaches for Tuberculosis Detection publication trend

The graph below shows the total number of articles in diagnostic approaches for tuberculosis detection across all publications each year (not limited to Nature Index journals).

Technical terms

Sputum smear microscopy: Examination of stained expectorated sputum under a microscope to identify acid-fast bacilli characteristic of Mycobacterium tuberculosis.

Nucleic acid amplification test (NAAT): Molecular assay, often PCR-based, that amplifies and detects specific TB DNA sequences and commonly assesses rifampicin resistance.

Loop-mediated isothermal amplification (LAMP): Isothermal nucleic acid amplification method using a set of primers and a strand-displacing polymerase to detect TB DNA without thermal cycling.

Metagenomic sequencing: Untargeted next-generation sequencing approach that identifies all microbial DNA in a sample, enabling pathogen discovery and resistance profiling.

Artificial-intelligence radiography: Deep-learning-based analysis of digital chest X-rays to detect radiological features of TB and prioritise cases for further testing.

References

1. Evaluating systematic targeted universal testing for tuberculosis in primary care clinics of South Africa: A cluster-randomized trial (The TUTT Trial). PLOS Medicine (2023).
2. Diagnostic accuracy of TB-LAMP for pulmonary tuberculosis: a systematic review and meta-analysis. BMC Infectious Diseases (2019).
3. Screening for HIV-Associated Tuberculosis and Rifampicin Resistance before Antiretroviral Therapy Using the Xpert MTB/RIF Assay: A Prospective Study. PLOS Medicine (2011).