Among the most fraught moments in medicine is the arrival of a feverish newborn — a being too young to speak, too fragile to wait, and too vulnerable for guesswork. At a pediatric emergency center in Genoa, Italy, researchers have identified a blood test that reads two molecular signals in opposite directions to determine, with near-perfect accuracy, whether a bacterial or viral infection is at work. The finding, presented at the European Emergency Medicine Congress, offers a potential path away from the reflexive overtreatment that has long defined care for febrile infants — and toward a more
Blood test ratio could distinguish bacterial from viral infections in febrile newborns
Febrile newborns currently undergo unnecessary invasive procedures and antibiotic treatment when viral infections are misidentified as bacterial.
Combining them into a single ratio separated bacterial from non-bacterial infection with very high accuracy
Mark
So the test measures two different molecules and creates a ratio. Why those two molecules specifically?
Mimi
They move in opposite directions depending on the infection type. CD64 rises when bacteria are present, CD169 when viruses are. By comparing them, you get a signal that's much clearer than either one alone.
Luke
But this is a 60-baby study. That's small. The specificity is 96.6 percent—that's excellent—but we don't know if those numbers hold when you test 500 babies across different hospitals with different patient populations.
Mark
What's the practical problem they're trying to solve?
Mimi
Right now, when a newborn comes in with fever, doctors can't tell if it's bacterial or viral. So they treat it as if it might be bacterial—antibiotics, lumbar punctures, hospital admission. Many of those babies only had a virus.
Luke
And the cost of that is real. Unnecessary antibiotics, unnecessary procedures, unnecessary time in the hospital. But also—antibiotic resistance.
Mark
So if this test works, it could change how emergency departments make decisions?
Mimi
Exactly. Instead of guessing and over-treating, they'd have a clear answer from a blood test.
Mark
When could this actually be available in a typical emergency room?
Luke
That's the open question. Right now it requires specialized equipment and trained staff. They're working on a simpler version, but that's still a research goal, not a finished product.
Mimi
The larger trial—DISCERN—will tell us whether the small study's results are real or just luck. That's the next critical step.
O Pulso
- Every febrile newborn under three months is treated as a potential emergency, because the cost of missing a bacterial infection can be a child's life — yet most of these infants have only a virus.
- The current diagnostic gap drives a cascade of harm: unnecessary antibiotics, lumbar punctures, and days of hospitalization for babies who would have recovered on their own.
- A Genoa research team found that measuring the ratio of two immune markers — CD64, which rises in bacterial infection, and CD169, which rises in viral infection — produced a diagnostic score so distinct that bacterial and viral cases barely overlapped.
- In a 60-infant study, the test caught every bacterial infection and misidentified almost none, achieving 100% sensitivity and 96.6% specificity — numbers that, if they hold, could redefine clinical decision-making.
- The test currently depends on flow cytometry equipment unavailable at most bedsides, but the team is racing to build a point-of-care version and is launching a 300–500 baby validation trial called DISCERN.
Among the most fraught moments in medicine is the arrival of a feverish newborn — a being too young to speak, too fragile to wait, and too vulnerable for guesswork. At a pediatric emergency center in Genoa, Italy, researchers have identified a blood test that reads two molecular signals in opposite directions to determine, with near-perfect accuracy, whether a bacterial or viral infection is at work. The finding, presented at the European Emergency Medicine Congress, offers a potential path away from the reflexive overtreatment that has long defined care for febrile infants — and toward a more precise, humane response to one of medicine's most urgent uncertainties.
A fever in a newborn under three months is treated as a medical emergency almost by definition. The infant immune system is still forming, and what resolves easily in an older child can turn catastrophic within hours. When these babies arrive in emergency departments, doctors face a binary choice with asymmetric consequences: treat for bacterial infection and risk unnecessary antibiotics and invasive procedures, or hold back and risk missing something life-threatening. For decades, uncertainty has pushed medicine toward overtreatment.
At the IRCCS Istituto Giannina Gaslini in Genoa, Italy, Dr. Tommaso Bellini and colleagues have been developing a more precise answer. Presenting at the European Emergency Medicine Congress, they described a blood test that measures two immune markers — CD64, found on neutrophils and elevated during bacterial infection, and CD169, found on monocytes and elevated during viral infection — and combines them into a single diagnostic ratio. Because the two markers move in opposite directions depending on the infection type, their ratio produces a score that cleanly separates the two conditions.
In a study of 60 infants, the results were striking. Babies with no infection averaged a ratio of 1.10. Those with confirmed viral infections averaged 0.33. Those with bacterial infections averaged 16.22. The test achieved 100 percent sensitivity and 96.6 percent specificity — meaning it caught every bacterial case and generated very few false alarms. Current standard biomarkers used in emergency settings produce false positives at rates high enough to drive widespread unnecessary treatment; this approach appears to sidestep that problem.
The test's limitation is practical: it currently requires flow cytometry equipment and trained staff, making it unsuitable for most emergency departments without specialist infrastructure. Bellini's team is working to convert it into a faster, low-cost point-of-care format. An expanded cohort of 75 infants is already underway, and a larger multicenter validation study — DISCERN — will assess the test in 300 to 500 babies.
Beyond individual patients, the stakes extend to public health. Antibiotic overuse accelerates bacterial resistance, eroding the effectiveness of drugs that medicine depends on broadly. A reliable test that tells clinicians which newborns truly need antibiotics — and which do not — would protect both the child in front of them and the broader ecosystem of treatment. The numbers from Genoa are early, but they point toward a future where one of medicine's most anxiety-laden decisions becomes, at last, a little less uncertain.
A fever in a baby under three months old is a medical emergency. The immune system is still forming, and what might be a manageable illness in an older child can turn catastrophic in hours. When parents bring a febrile newborn to the emergency department, doctors face an urgent diagnostic puzzle: Is this a bacterial infection that demands immediate antibiotics, or a viral one that will resolve with time and supportive care? The stakes of guessing wrong are high. Misidentify a virus as bacterial, and the infant may receive unnecessary antibiotics, undergo invasive procedures like lumbar punctures, and spend days in the hospital. Miss a real bacterial infection, and the consequences are far worse.
At the pediatric emergency department of IRCCS Istituto Giannina Gaslini in Genoa, Italy, Dr. Tommaso Bellini and his colleagues have been working on a solution. They presented findings this month at the European Emergency Medicine Congress showing that a simple blood test—one that measures the ratio of two molecules in the bloodstream—can distinguish between bacterial and viral infections in newborns with remarkable accuracy. The test compares levels of CD64, a marker found on neutrophils that rises in response to bacterial infection, against CD169, a marker on monocytes that rises in response to viral infection. By combining these two measurements into a single ratio, the researchers found they could separate sick babies into three clear groups.
The study involved 60 infants: some brought in with fever, others with non-infectious conditions serving as controls. Blood samples were analyzed using flow cytometry, an advanced laboratory technique that measures individual cell characteristics. Babies with no infection had an average ratio score of 1.10. Those confirmed to have viral infections averaged 0.33. Those with bacterial infections averaged 16.22. The separation was stark. The test achieved 100 percent sensitivity—it caught every single case of bacterial infection—and 96.6 percent specificity, meaning false alarms were rare. In practical terms, this means a doctor could rely on the test to tell them which babies actually need antibiotics and which do not.
Dr. Bellini explained the logic: the two markers move in opposite directions depending on the infection type. During a viral infection, one molecule is elevated while the other drops. During a bacterial infection, the pattern reverses. "Combining them into a single ratio separated bacterial from non-bacterial infection with very high accuracy," he said. The current standard biomarkers used in emergency departments can suggest infection, but they generate false positives at rates high enough to drive unnecessary treatment. This new approach appears to solve that problem.
There is a catch. The test currently requires flow cytometry equipment and trained laboratory staff. It is not a bedside test. It is not widely available outside specialist labs. Running the analysis takes time—time that matters when a newborn is running a fever. Bellini's team is now working to translate the test into a faster, simpler format that could be deployed in emergency departments without specialized infrastructure. The research is continuing with an expanded cohort of 75 babies, and a larger validation study called DISCERN is beginning, which will assess the test's accuracy in 300 to 500 infants across multiple centers.
Dr. Maša Sorić, head of emergency medicine at University Hospital Merkur in Zagreb and not involved in the research, sees the potential. "For parents, taking a very young baby with a fever into the emergency department can be a very worrying time," she said. Current practice requires treating every case as potentially serious until proven otherwise, which means invasive testing, antibiotics, and prolonged hospitalization for many infants who turn out to have only a virus. If these findings hold up in larger trials, the impact could be substantial: fewer unnecessary procedures, fewer unnecessary antibiotics, and faster answers for frightened parents.
There is also a public health dimension. Antibiotic overuse drives bacterial resistance, which threatens the effectiveness of these drugs for everyone. If doctors can confidively identify which babies truly need antibiotics and which do not, they can reserve these medications for cases where they actually work. The test is still in early stages, but the initial numbers suggest it could change how emergency departments approach one of their most anxiety-laden decisions.
Citações Notáveis
In very young babies with fever, distinguishing serious bacterial infection from self-limiting viral illness is a major diagnostic dilemma. Our standard biomarkers can help show when there might be a bacterial infection, but they also create 'false positives', which drives overtreatment.— Dr. Tommaso Bellini, pediatric emergency physician
If these results can be reinforced in a larger trial, they could mean that we have a much better way of knowing which babies probably have a bacterial infection that needs antibiotics, and which babies probably do not.— Dr. Maša Sorić, head of emergency medicine at University Hospital Merkur
Quer a matéria completa?
Leia o original em EurekAlert! Science News Releases ↗