The MeMed BV is a protein-based assay that uses host protein levels to differentiate bacterial from viral infection.
The chemiluminescence immunoassay uses machine learning to integrate the blood levels of TRAIL, IP-10, and CRP, providing a score from zero to 100. In the zero to 34 range, the infection is more likely to be viral. A score of 66 to 100 is more likely bacterial infection, and a score of 35 to 65 is considered equivocal.
The test is indicated for patients presenting not only to the ED but also urgent care with suspected bacterial or viral infection, with symptoms occurring for less than seven days. It’s cleared for adult and pediatric patients, making it something of an outlier in sepsis diagnostics. “As someone who helps with our pediatric hospital and oversees part of the acute care testing program, we often don’t have studies in pediatric patients. There is a litany of great studies in pediatric patients with this marker,” he said.
A prospective study of the test’s diagnostic accuracy enrolled adults with symptoms of fever and lower respiratory tract infection presenting to several EDs (Halabi S, et al. Clin Microbiol Infect. 2023;29[9]:1159–1165). Reference standard diagnosis (bacterial/viral/indeterminate) was based on three experts independently reviewing comprehensive patient data, including follow-up data. The assay generated three results: viral infection or other nonbacterial condition, equivocal, or bacterial infection including co-infection. Performance was assessed against the reference standard with the indeterminate reference standard and equivocal MeMed BV cases removed.
Of 490 enrolled patients, 415 met eligibility criteria. The reference standard classified 104 patients with bacterial infection, 210 with viral, and 101 as indeterminate. The test was equivocal in 9.6 percent of cases (30/314). Excluding the indeterminate reference standard diagnoses and equivocal results from the BV test, the test’s sensitivity for bacterial infection was 98.1 percent. Specificity and negative predictive value were 88.4 percent and 98.8 percent, respectively. “Patients who had bacterial infections had scores that were quite a bit higher relative to the viral infections,” Dr. Farnsworth said. “It does seem to separate out the different groups quite well, with very few of these patients with bacterial infection down in the viral range and vice versa.”
An ongoing randomized controlled trial is evaluating this assay at multiple sites. The pilot phase of the trial enrolled adult ED and urgent care patients presenting with symptoms of lower respiratory tract infection for whom physicians considered antibiotic treatment (Singer AJ, et al. Acad Emerg Med. 2025;32[9]:975–984). Patients were randomized to an interventional arm, in which clinicians were given MMBV results, and to standard care (physicians did not have MMBV results). The primary aim was to assess antibiotic prescription rate in the standard care arm. The study population included 214 patients. “They excluded patients who were admitted to the hospital, so we’re looking at a not-necessarily acute but at a less sick population,” Dr. Farnsworth said.
Overall, antibiotic prescription rates were 30 percent in the standard care arm and 24 percent in the MMBV intervention arm. In the MMBV arm, 78 percent of patients with bacterial scores were prescribed antibiotics, compared with 41 percent in the standard care arm. Among patients with viral MMBV scores, 12 percent were prescribed antibiotics, compared with 25 percent in the standard care arm. There was no increase in ED or urgent care return visits or hospitalizations in the standard care arm versus the MMBV arm.
“It will be interesting to see how this trial plays out in the long term,” Dr. Farnsworth said.
The takeaways: The MMBV assay has good sensitivity and specificity for bacterial infection and can be performed at the point of care and on commonly used immunoassay platforms. It can’t differentiate between viral and other nonbacterial etiologies. “It really tells you that if there is infection present, is it more likely bacterial or viral.” In some studies, patients with equivocal results were excluded, which does not reflect real-world practice. And it cannot predict which patients will become septic. “But it can be used in managing septic patients,” he noted.