RESUMO
Colorimetric reverse transcription loop-mediated isothermal amplification (RT-LAMP) is a potential and relatively simple rapid diagnostics method for COVID-19 detection. This study aims to evaluate and optimize the RT-LAMP performance on saliva specimens based on a commercially available kit.Modifications on an established protocol (Protocol A) were used, including Proteinase K supplementation (Protocol B); pre-treatment using nuclease-free water and proteinase K (Protocol C); Saliva cooling (Protocol D); saliva dilution after pre-treatment (Protocol E); lastly a combination of saliva cooling and dilution (Protocol F). Protocol performances were evaluated by comparing success rates (SR), diagnostic accuracy (DA), sensitivity, specificity, and predictive values. Additionally, a correlation between the Ct value by RT-qPCR and RT-LAMP performance was analyzed.. A total of 106 specimens were used in this study. Protocols B and C showed 100% unreadable results, therefore were paused. Protocol F showed the highest SR (87.65%) compared to other protocols, with a slight compromise to DA (81.69%), sensitivity (57.14%), specificity (97.67%), PPV (94.12%), and NPV (77.78%). In the sub-analysis of the low Ct value group (Ct < 30), Protocol F demonstrated a higher success rate (86.57%) compared to protocol A (64.18%); increased 3.08% sensitivity and 2.42% NPV; comparable DA; minor reduction in specificity (A = 100%; F = 97.67%) and PPV (A = 100%; F = 92.31%). A combination of saliva cooling-dilution substantially increased the tested kit's success rate, despite a slight decrease in specificity and PPV. Findings confirmed the saliva cooling-dilution procedure was beneficial to the test's SR, sensitivity, and NPV in the low Ct value group. Supplementary Information: The online version contains supplementary material available at 10.1007/s13337-024-00870-1.