Abstract
Background Sexually transmitted infections (STIs) and bacterial vaginosis (BV) are highly prevalent risk factors for HIV infection, partly driven by genital inflammation. We have developed a novel true point-of-care inflammation lateral flow assay called the Genital InFlammation Test (GIFT). GIFT detects the presence of elevated genital inflammatory cytokines to screen for asymptomatic STIs/BV cases that are missed in settings where syndromic management is practiced. Novel point-of-care tests (like GIFT) must benchmark their performance against the guidelines specified by the WHO’s REASSURED criteria, which includes being equipment-free. This study compared the performance of human eye reading of the GIFT device to an automated device reader. Methods Bio-banked lateral vaginal wall swabs collected from 10 women were tested for the presence of inflammation biomarker bands (IL-1alpha, IL-1beta) on GIFT devices, which were visually assessed by two independent researchers for band intensity (using G-scores between 0–10). Visual inspection of G-score band intensity was benchmarked against cytokine concentrations measured by commercial ELISA. G-score visual assessments were compared with band intensity measurements using an automated device reader (AXXIN-AX-2X-S™). Results Significant concordance was observed between visual (G-scores) for IL-1alpha and IL-1beta and automated reader intensities (rho=0.96, pConclusion This laboratory evaluation of the GIFT device performance, assessed visually or using an automated reader suggests that the reader may offer a sensitivity benefit over visual reading at lower vaginal cytokine concentrations, while accuracy will be comparable at higher concentrations.