MxA biomarker helps separate viral from bacterial infections in adults

Distinguishing a viral infection from a bacterial infection can determine whether a seriously ill adult receives antibiotics, but conventional laboratory markers do not always provide a clear answer. A prospective observational study of 869 consecutive adult patients found that myxovirus resistance protein A, or MxA, was substantially elevated during viral infections and could serve as a practical biomarker for identifying viral disease in acute care. The results suggest that measuring MxA, particularly alongside markers of systemic inflammation, may help clinicians make more evidence-based decisions about antibiotic treatment.

The clinical problem is especially important in emergency and acute-care settings, where respiratory symptoms can arise from several causes. A patient may have a viral infection, a bacterial infection, or both, yet symptoms such as fever, cough, shortness of breath and abnormal vital signs may overlap. Misidentifying a viral illness as bacterial can expose patients to unnecessary antibiotics, contribute to adverse reactions and accelerate antimicrobial resistance. Conversely, overlooking bacterial infection can delay appropriate treatment. Laboratory evidence is therefore central to the initial assessment, but the available markers do not always produce an unambiguous classification.

MxA is an antiviral protein that is produced in response to type I interferon signaling. This signaling pathway is activated by viral infections, making MxA biologically different from many general markers of illness. C-reactive protein, for example, is a broad indicator of inflammation and can rise in response to both viral and bacterial disease. Procalcitonin tends to increase more strongly in bacterial infection and is used in some clinical settings, but its performance depends on the population, the infection site and the timing of measurement. MxA offers another route to viral identification because it reflects a host response specifically associated with viral replication rather than inflammation alone.

For the study, 869 consecutive adult patients were allocated to groups according to whether they had an infection and what caused it. The researchers evaluated the diagnostic performance of MxA and MxA-based ratios, then compared their results with an optimal statistical model produced through multiple linear regression. They also examined factors that might influence MxA production and developed an algorithm intended to guide antibiotic prescribing in a subgroup of patients with respiratory tract infections. The design reflects a practical question: whether a measurable biological signal can improve classification in routine adult acute care, not merely whether MxA differs between laboratory-defined groups.

MxA was significantly elevated in patients with viral infection and in those with mixed viral and bacterial infections. At an optimal cutoff of 21 micrograms per liter, the study reported a sensitivity of 79.8% and a specificity of 87.1%. The area under the receiver operating characteristic curve was 0.87, indicating that MxA separated viral from bacterial cases with substantial diagnostic accuracy across possible decision thresholds. These findings mean that many patients with viral disease could be identified by a relatively simple concentration measurement, while most patients without the marker had a condition other than a viral infection.

Sensitivity and specificity describe different sides of the same diagnostic problem. A test with high sensitivity is more likely to detect most cases of the target condition, while high specificity means that a positive result is less likely to occur in people without it. The reported values therefore support MxA as a potentially useful rule-out aid for viral infection, although they do not mean that every value can be interpreted independently of the clinical presentation. A false-negative result could occur in a patient with viral infection whose MxA concentration remains below the selected threshold, while a false-positive result could arise when another condition raises MxA. The optimal cutoff is therefore best understood as a value derived from this study population rather than a universal threshold.

The strongest performance emerged when MxA was combined with C-reactive protein and procalcitonin. In this analysis, the combined approach performed comparably to the optimal statistical model, indicating that MxA contributed information that was not fully captured by the more familiar inflammatory markers. Combining tests can be especially useful in ambiguous cases: MxA supplies evidence for viral activity, while CRP and PCT provide complementary information about the host response and the likelihood of bacterial involvement. Rather than replacing clinical assessment, the result supports an integrated interpretation in which viral biology and systemic inflammation are considered together.

The amount of MxA also differed according to viral etiology, showing that the host response is not identical across viral infections. That variability has implications for interpretation. A highly detectable marker may be especially useful for some pathogens and less informative for others, and clinicians should therefore avoid assuming that a single concentration represents every type of viral disease. The study did not report a consistent change in MxA expression associated with comorbidities, suggesting that common accompanying illnesses did not systematically distort the signal in the cohort. Nevertheless, the observational design and the composition of the enrolled population mean that broader studies would be needed to establish how the marker performs in other hospitals, age groups and disease settings.

The respiratory subgroup was especially relevant to the question of antibiotic stewardship. The researchers developed an algorithm to guide prescribing in patients with respiratory tract infections, where the presence of a bacterial pathogen can be difficult to determine from symptoms alone. The findings do not show that MxA alone can replace cultures, molecular testing or clinical judgment, nor do they establish a universal rule for withholding antibiotics. Instead, the proposed algorithm indicates how biomarker results might be integrated into a structured decision process. In that context, a viral MxA signal could contribute to the evidence used to reconsider unnecessary antibiotic therapy, while bacterial indicators could support continued assessment and treatment when warranted.

Antibiotic stewardship is not simply a matter of reducing antibiotic use. It requires ensuring that patients who need treatment receive it at the appropriate time and that unnecessary exposure is minimized. A biomarker that adds specific information about viral infection could help make that balance more precise, particularly when standard markers are discordant or the clinical picture is uncertain. MxA would not settle every diagnostic question, because the reported sensitivity leaves a meaningful proportion of viral infections below the selected cutoff and because bacterial and viral infections can coexist. Its strongest role may therefore be as one component of a broader strategy that combines laboratory findings, pathogen detection, examination, imaging when appropriate and repeated clinical assessment.

The study was supported through the Cooperatio Program for intensive care medicine, infection and immunity at Charles University, with financial support for biochemical and statistical analyses provided by Biovendor. The sponsor was reported to have had no role in study design, data acquisition, analysis, interpretation, manuscript preparation or the decision to submit the paper, and the authors declared no competing interests. The findings are now reported in BMC Infectious Diseases as a prospective observational evaluation of MxA in adult acute care. The 869-patient cohort, the reported area under the curve of 0.87 and the improvement achieved with CRP and PCT together point toward a promising addition to viral diagnostics, while the need for prospective validation and clinical implementation studies keeps the conclusions appropriately bounded.

Subject of Research: Evaluation of MxA as a diagnostic biomarker for distinguishing viral from bacterial infection in acutely ill adults

Article Title: Clinical utility of Myxovirus resistance protein A (MxA) for differentiation between viral and bacterial infection in adult acute care: a prospective observational study

Article References: Muller, J., Rajdl, D., Vimmerova, H., & Matejovic, M. (2026). Clinical utility of Myxovirus resistance protein A (MxA) for differentiation between viral and bacterial infection in adult acute care: a prospective observational study. BMC Infectious Diseases. https://doi.org/10.1186/s12879-026-14385-1

Image Credits: AI Generated

DOI: 10.1186/s12879-026-14385-1

Keywords: MxA, viral infection, bacterial infection, diagnostic biomarker, procalcitonin, C-reactive protein, acute care, antibiotic stewardship, respiratory infection, interferon response, Clinical, utility