Sensitivity and specificity of a rapid, point-of-care assay distinguishing bacterial from viral infections in acute care settings

Differentiating between bacterial and viral infections is essential in the emergency department (ED) to guide appropriate antibiotic treatment. Recent assays, such as the Food and Drug Administration (FDA)-cleared MeMed BV assay, can help with this distinction. The MeMed BV assay measures three host immune proteins—Tumor necrosis factor (TNF)-related apoptosis-induced ligand, Interferon (IFN)-γ-induced protein 10, and C-reactive protein—in blood samples. The results are combined into a score from 0 to 100, where a higher score indicates a higher likelihood of a bacterial infection and a lower score suggests a viral infection or other non-bacterial causes. The study evaluates the feasibility of using the MeMed BV assay-generated score to help clinicians decide whether to order blood cultures in emergency departments, as well as assessing the accuracy of the MeMed BV assay in distinguishing between bacterial and viral infections in ED patients. The assay’s performance was also compared with standard biomarker tests and white blood cell (WBC) counts for predicting bacterial infections. This study included 203 illness episodes, of which a total of 108 were associated with bacterial infection (25 had a positive blood culture; 107 had a non-blood sample positive for a bacterial pathogen) and 18 were associated with a viral pathogen. Combined, the MeMed BV assay’s “high bacterial likelihood” and “moderate bacterial likelihood” scores yielded a sensitivity and specificity of 87% and 32%, respectively, in correctly identifying any bacterial infection; the scores had a sensitivity and specificity of 95% and 25%, respectively, in correctly identifying bacteremia. The future of differentiating between bacterial and viral infections is moving toward integrated diagnostic methods. To address antibiotic resistance, future diagnostics will leverage host immune response biomarkers, advanced molecular panels, and machine learning techniques. This will enable rapid and accurate results at the point of care.

IMPORTANCE

Distinguishing between bacterial and viral infections in healthcare settings can be challenging, as patients often present with similar symptoms and clinical features. In recent years, several new biomarkers have emerged as promising targets for algorithm-based tests designed to differentiate between these types of infections. An increasing number of FDA-approved assays that utilize these biomarkers have entered the market, employing various algorithms. These tests are being used in different clinical settings, such as emergency rooms, to assist clinicians in making accurate diagnoses and prescribing antibiotics appropriately. It is essential for clinicians to correctly order and interpret the results of these assays to fully realize their potential. Our study demonstrates that when such an assay is not used within the FDA-approved context, it does not achieve its full effectiveness.