Clinical laboratory evaluation of the Hologic Panther Aptima BV and CV/TV assays for the diagnosis of vaginitis in Dunedin, Aotearoa New Zealand

• Published in: Microbiology spectrum 2024-11, p.e0127424
• Main Authors: Elvy, Juliet, Carter, Katelyn, Paterson, Jenna, Smith, Megan, Parslow, Gayleen, Ussher, James E
• Format: Article
• Language: English
• Subjects: Clinical Microbiology; Research Article
• ISSN: 2165-0497; 2165-0497
• DOI: 10.1128/spectrum.01274-24

Vaginitis presentations are common, but traditional diagnostic methods are imperfect. Molecular methods for bacterial vaginosis (BV) and vulvovaginal candidiasis (CV) are increasingly available but not commonly utilized in Aotearoa New Zealand. We evaluated the Hologic Aptima BV and CV/ (TV) assays against our current methods (Gram stain, yeast culture, and Hologic Aptima TV assay) and performed a retrospective BV clinical audit. The BV Aptima assay performed well with high sensitivity (97.5%) and specificity (96.3%) when the indeterminate BV category was excluded. BV indeterminate samples were almost evenly split between positive and negative results when tested on the Aptima BV assay. BV Gram stain interpretation was error prone, with 20% of samples discordant on duplicate examination. Although the Aptima CV assay was highly sensitive, it lacked specificity compared with Gram stain (83.5%) but was similar to culture (91.2%). Our BV clinical audit showed that patients with a BV indeterminate result were less likely to be treated for BV than those with a positive result, meaning more women may be treated for BV if this assay were implemented. Overall, implementation may improve laboratory workflow and consistency of reporting, but cost may be a barrier. The clinical impact of changing methods needs to be considered.IMPORTANCEIn this paper, we evaluate the performance of the Aptima molecular assays against current Gram stain and culture methods, as well as a clinical audit to determine the potential clinical impact of implementation. Although molecular methods are increasingly used in other countries, New Zealand has not yet adopted this approach. Importantly, we found Gram stain for bacterial vaginosis (BV) to be error prone, with 20% of Gram stain results discordant on repeat examination. We show the potential for molecular methods to increase BV diagnoses and improve reproducibility and consistency of reporting which, according to our clinical audit results, would lead to more women being treated for this dysbiosis condition overall.