Diagnostic accuracy of three dengue NS1 antigen ELISA kits in comparison to chemiluminescence immunoassay.

Abstract

Background objectives: Dengue fever, including its severe forms, remains a major public health concern in endemic regions. Early diagnosis is critical for effective disease management, yet current diagnostic methods vary in sensitivity and accessibility. This study aimed to evaluate the diagnostic performance of three commercially available NS1 ELISA kits against a chemiluminescence immunoassay (CLIA) reference standard.

Methods: A total of 321 archived serum samples, previously tested using the Dengue NS1 Autobio chemiluminescence immunoassay (CLIA), were further analyzed using three NS1 ELISA kits. Among these, 12 samples showed discordant results between at least one ELISA kit and the CLIA and were subsequently analyzed through RT-PCR-based dengue serotyping. The diagnostic performance of the ELISA kits, including sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), and overall agreement, was assessed using MedCalc software. Agreement between tests was further evaluated using kappa (κ) statistics, and comparisons of κ values were conducted using Fleiss's Z-test.

Results: Among the 321 archived serum samples, 105 were initially identified as NS1-reactive and 216 as NS1-negative. RT-PCR confirmed three of the CLIA-positive results as false positives, adjusting the reference count to 102 NS1-reactive and 219 NS1-negative samples. Athenese-Dx ELISA demonstrated the highest sensitivity (98.04%) and agreement with reference (99.07%, κ = 0.978), followed by SD (Sensitivity = 95.10%, Agreement = 97.82%, κ = 0.949) and PanBio (Sensitivity = 94.12%, Agreement = 97.82%, κ = 0.949). Specificity remained high across all ELISA kits (99.07%-99.54%).

Interpretation conclusion: The evaluated ELISA kits demonstrated high specificity and strong agreement with the reference standard, with Athenese-Dx ELISA showing better diagnostic accuracy. NS1-based ELISA remains a valuable tool for dengue diagnosis, particularly in resource-limited settings. Further studies with serotype-specific analysis and larger sample sizes are warranted.