📄 medicine

Comparative Diagnostic Accuracy of GeneXpert MTB/RIF and TB-LAMP for Tuberculosis Detection Among Presumptive Tuberculosis Cases in Bayelsa State, Nigeria

This study in Bayelsa State, Nigeria, found that while TB-LAMP offers higher sensitivity for initial screening, GeneXpert MTB/RIF demonstrates superior concordance with culture and the critical ability to detect rifampicin resistance, supporting a sequential diagnostic algorithm where TB-LAMP is used for screening followed by GeneXpert for confirmation.

Original authors: Patrick UMUWE, Kennedy T. Wariso, Jude ILOZUMBA, Mabel Emudiaga-Ohwerhi, Stephen Uchechukwu OKOLI, Chizaram Precious DANIEL, Enifome OGBE, Praise Ongoiyerimo SAIYOU, Miriam Bathnna, Agbaje Aderonke, U

Published 2026-07-01

📖 4 min read☕ Coffee break read

CC BY 4.0

Original authors: Patrick UMUWE, Kennedy T. Wariso, Jude ILOZUMBA, Mabel Emudiaga-Ohwerhi, Stephen Uchechukwu OKOLI, Chizaram Precious DANIEL, Enifome OGBE, Praise Ongoiyerimo SAIYOU, Miriam Bathnna, Agbaje Aderonke, Urhioko Ochuko, Emperor Ubochioma, Aoife Reilly, Zamzud Margaret, Chidimma Lilian Okoye, Williams MAYAH Tisan

Original paper licensed under CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). ✨ This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer

Imagine you are a detective trying to catch a very sneaky criminal: the bacteria that causes Tuberculosis (TB). In the past, finding this criminal was like waiting for a slow-moving snail to leave a trail; you had to grow the bacteria in a lab (a "culture"), which took weeks. If you were in a busy, resource-limited area like Bayelsa State, Nigeria, waiting that long was dangerous.

This paper is about testing two new, high-tech "metal detectors" designed to find the TB criminal much faster. The researchers wanted to see which metal detector was better: the GeneXpert or the TB-LAMP.

Here is the story of their investigation, broken down simply:

The Setting: A Busy Crime Scene

The researchers gathered 392 people who had symptoms like a bad cough, fever, or weight loss. They took a sample of their spit (sputum) and ran it through three different tests at the same time:

The Gold Standard (Culture): The old-school method of growing the bacteria. It's slow but considered the "truth" because it proves the bacteria is alive and growing.
GeneXpert: A high-tech machine that looks for the bacteria's DNA and also checks if the bacteria is resistant to a specific drug (Rifampicin).
TB-LAMP: Another molecular test that is cheaper and simpler, designed for smaller labs.

The Results: Who Caught the Most Criminals?

1. The Gold Standard (Culture)
Out of all the people tested, the slow-growing culture method confirmed that 10.6% actually had TB. This is the baseline we compare the others against.

2. The GeneXpert (The Reliable Specialist)

The Score: It found TB in 11.2% of people.
The Verdict: This result was almost identical to the Gold Standard. When the researchers compared the two, they found no significant difference.
The Analogy: Think of GeneXpert as a highly trained sniffer dog. It rarely misses a real criminal (91.7% sensitivity) and almost never mistakes a harmless dog for a criminal (97.7% specificity). If it barks, you can be very sure the criminal is there. It also has a special superpower: it can tell you if the criminal is wearing a "bulletproof vest" (drug resistance).

3. The TB-LAMP (The Sensitive but Jumpy Scout)

The Score: It found TB in 14.8% of people.
The Verdict: This number was significantly higher than the Gold Standard. The researchers found a real difference here.
The Analogy: Think of TB-LAMP as a very sensitive motion sensor. It is great at catching the criminal (95.1% sensitivity), but it is a bit "jumpy." It sometimes goes off when there is no criminal, just a harmless animal running by.
The "False Alarm" Problem: The study found that about 2.3% of the people TB-LAMP flagged as having TB actually had a different, harmless type of bacteria (called NTM) that looks similar to TB but isn't. Because TB-LAMP is so sensitive, it picked up these look-alikes, leading to "false positives."

The Big Picture: What Did They Learn?

The researchers concluded that these two tools play different roles in the fight against TB:

GeneXpert is the "Confirmatory Tool": Because it agrees so perfectly with the slow Gold Standard and rarely makes mistakes, it is the best tool to say, "Yes, this person definitely has TB, and here is how to treat it." It is the final judge.
TB-LAMP is the "Screening Tool": Because it is fast and catches almost every single case, it is great for a first look, especially in smaller clinics. However, because it sometimes mistakes harmless bacteria for TB, you cannot trust it alone. If TB-LAMP says "Yes," you must send that sample to a GeneXpert machine to double-check and make sure it's not a false alarm.

The Proposed Strategy

The paper suggests a "two-step" strategy for places like Bayelsa:

Step 1: Use the TB-LAMP first to quickly scan everyone. It's like using a wide net to catch everything.
Step 2: If the net catches something, run it through the GeneXpert to confirm it's actually the TB criminal and not a harmless look-alike.

This approach uses the speed of the first tool and the accuracy of the second to find more patients without wasting money on treating people who don't have the disease.

Technical Summary: Comparative Diagnostic Accuracy of GeneXpert MTB/RIF and TB-LAMP in Bayelsa State, Nigeria

Problem Statement
Tuberculosis (TB) remains a leading cause of mortality globally, with a disproportionate burden in resource-limited settings where diagnostic delays contribute to treatment failure and drug resistance. While mycobacterial culture is the diagnostic reference standard, its prolonged turnaround time and infrastructural requirements limit its utility in high-burden regions. Although Nucleic Acid Amplification Tests (NAATs) offer rapid alternatives, there is a need for context-specific evaluations of their comparative accuracy and concordance with culture in diverse epidemiological settings, particularly in Nigeria. Misdiagnosis carries severe consequences, including suboptimal treatment, the emergence of drug-resistant strains, and unnecessary drug toxicity from false positives.

Methodology
The study employed a cross-sectional diagnostic accuracy design conducted between January and June 2025 in Bayelsa State, Nigeria.

Participants: 392 individuals (aged 6–80) presenting with TB-suggestive symptoms (e.g., persistent cough, fever, weight loss) were recruited from 40 Directly Observed Therapy (DOT) sites across eight health facilities.
Reference Standard: All participants provided a single sputum specimen processed via solid mycobacterial culture on Löwenstein–Jensen (LJ) medium. Isolates were differentiated from non-tuberculous mycobacteria (NTM) using the SD Bioline TB Ag MPT64 Rapid Test.
Index Tests: The same sputum samples were tested in parallel using:
1. GeneXpert MTB/RIF: A closed-cartridge NAAT system targeting the rpoB gene for MTBC detection and rifampicin resistance.
2. TB-LAMP: A loop-mediated isothermal amplification assay targeting the gyrB gene and IS6110 insertion sequence, performed using the HumaLoopT instrument.
Analysis: Diagnostic performance (sensitivity, specificity, PPV, NPV) was calculated against culture. Concordance between tests was assessed using McNemar's chi-square test for paired proportions (Stata v16.0).

Key Results

Positivity Rates: Culture positivity was 10.6% (41/392). GeneXpert detected 11.2% (44/392), while TB-LAMP detected 14.8% (58/392).
Concordance with Culture:
- GeneXpert: Showed no statistically significant discordance with culture (McNemar $\chi^2$ = 1.80, $p$ = 0.18), indicating strong agreement.
- TB-LAMP: Showed significant discordance with culture, detecting substantially more positive cases (McNemar $\chi^2$ = 12.57, $p$ = 0.0004).
Diagnostic Performance Metrics (vs. Culture):
- GeneXpert: Sensitivity 91.7%, Specificity 97.7%, PPV 97.6%, NPV 99.0%.
- TB-LAMP: Sensitivity 95.1%, Specificity 94.5%, PPV 68.4%, NPV 99.1%.
False Positives: Notably, 2.3% of TB-LAMP-positive results were identified as NTM, contributing to its lower Positive Predictive Value (PPV) compared to GeneXpert.

Key Contributions
The study provides empirical evidence comparing the diagnostic accuracy of two molecular platforms against the culture reference standard in a high-burden Nigerian setting. It quantifies the trade-off between the higher sensitivity of TB-LAMP and its lower specificity due to NTM cross-reactivity. Furthermore, it validates the high concordance of GeneXpert with culture, reinforcing its reliability as a standalone confirmatory test.

Significance and Claims
The paper concludes that GeneXpert MTB/RIF demonstrates high diagnostic accuracy and strong concordance with culture, supporting its role as a primary confirmatory tool, particularly for suspected multidrug-resistant TB (MDR-TB) due to its simultaneous rifampicin resistance detection.

Conversely, TB-LAMP is characterized as a rapid, low-cost screening option suitable for peripheral and second-tier laboratories where infrastructure is limited. However, due to its lower specificity and the risk of false positives from NTM, the authors claim it requires confirmatory testing to rule out NTM infection and assess drug resistance.

The authors propose a sequential diagnostic algorithm: utilizing TB-LAMP for initial screening followed by GeneXpert for the confirmation of reactive specimens. This strategy is claimed to potentially enhance case detection rates while optimizing resource utilization and minimizing misdiagnosis in resource-limited settings.

The Setting: A Busy Crime Scene

The Results: Who Caught the Most Criminals?

The Big Picture: What Did They Learn?

The Proposed Strategy

Technical Summary: Comparative Diagnostic Accuracy of GeneXpert MTB/RIF and TB-LAMP in Bayelsa State, Nigeria

More like this