Accuracy of Tuberculosis Infection Diagnosis through IP-10-Based Assays for Immune Detection of Present Mycobacterium tuberculosis: A Cross-Sectional Evaluation
Stoichita, A.; Burecu, M.; Nita, C.; Teleaga, C.; Radu, A. D.; Mihai, M.; Mahler, B.; Ibraim, E.
Show abstract
BackgroundReliable detection of latent Mycobacterium tuberculosis (Mtb) infection (LTBI) remains challenging, particularly in TB contacts and immunocompromised individuals, where interferon-{gamma} release assays (IGRAs) demonstrate variable sensitivity. IP-10, a chemokine produced at substantially higher concentrations than IFN-{gamma}, represents a promising immune marker. This study aimed to evaluate the diagnostic performance of two IP-10 based assays RIDA(R)QUICK TB (lateral flow) and RIDASCREEN(R) TB (ELISA), by comparison with QuantiFERON-TB Gold Plus (QFT-Plus) assay or a composite reference standard. MethodsA cross-sectional diagnostic accuracy study enrolled 99 adults: 49 with culture-confirmed active pulmonary TB, 30 close TB contacts and 20 individuals with autoimmune disease, in Bucharest, Romania. All participants underwent RIDA Quick, RIDA Screen and QFT-Plus testing. Indeterminate results for all assays were reclassified using a composite reference standard. ResultsAgainst culture in active TB cases, RIDA(R)QUICK TB demonstrated a sensitivity of 85.7% (95% CI: 72.8-94.1) and PPV of 97.7%, while RIDA(R)SCREEN TB achieved 91.8% sensitivity (95% CI: 80.4-97.7) and 97.8% PPV. Specificity and NPV could not be reliably estimated due to the near-absence of true-negative individuals. Agreement with QFT-Plus was moderate to good ({kappa}=0.47-0.93).ROC analysis performed against QFT-Plus as a comparator demonstrated good immunological discrimination for RIDA(R)QUICK TB (AUC = 0.828) and RIDA(R)SCREEN TB (AUC = 0.767), reflecting concordance with QFT-Plus rather than diagnostic accuracy against confirmed infection. ConclusionIP-10 based assays demonstrated higher sensitivity than QFT-Plus and excellent PPV across bacteriological standard, supporting their use as complementary tools for LTBI detection. Larger, more heterogeneous cohorts are needed to accurately define specificity and operational integration.
Matching journals
The top 6 journals account for 50% of the predicted probability mass.