Evaluation of the AIX1000 Automated Rapid Plasma Reagin Assay in a High-Prevalence Setting

Abstract

BackgroundManually performed nontreponemal assays, such as rapid plasma reagin (RPR), are labor intensive and time consuming. Recently, commercial automated RPR assays gained attention. The aim of this study was to compare the qualitative and quantitative performance of the AIX1000 (RPR-A; Gold Standard Diagnostics) to a manual RPR test (RPR-M; Becton Dickinson Macrovue) within a high-prevalence setting.MethodsA retrospective panel of 223 samples was selected for comparison between RPR-A and RPR-M, including 24 samples from patients with known syphilis stages and 57 samples from 11 patients in follow-up. In addition, 127 samples obtained during routine syphilis diagnosis with RPR-M were analyzed prospectively with AIX1000.ResultsOverall qualitative concordance (percent agreement) between both assays was 92.0% in the retrospective and 89.0% in the prospective panel. Of 32 discordances, 28 were explained by a treated syphilis infection still positive in one assay and already negative in the other. One sample was false positive with RPR-A, 1 infection remained undetected by RPR-M, and 2 remained undetected by RPR-A. A hook effect was apparent on the AIX1000 at RPR-A titers from 1:32 onward; however, no infections were missed. Accepting a ±1 titer difference, quantitative concordance between both assays reached 73.1% and 98.4% for the retrospective and prospective panels, respectively, with an upper limit of reactivity for RPR-A at 1:256.ConclusionsThe AIX1000 showed a similar performance to Macrovue RPR with the exception of a negative deviation for high-titer samples. Within the reverse algorithm used in our high-prevalence setting, AIX1000's main advantage is automation.