The logic behind the index is deceptively simple. C-reactive protein, an acute-phase protein produced by the liver, rises rapidly in response to systemic inflammation and is one of the most widely used markers of inflammatory activity in clinical practice. Ferritin, the intracellular iron-storage protein, is also an acute-phase reactant, and in severe COVID-19 it can reach extraordinary concentrations, partly reflecting macrophage activation and the cytokine-driven disarray that characterizes critical disease. Albumin, by contrast, falls during inflammation and malnutrition, a phenomenon known as transient capillary leak combined with suppressed hepatic synthesis and, in prolonged illness, genuine protein-energy depletion. By placing the two inflammatory markers in the numerator and the nutritional marker in the denominator, the HNI is intended to capture, in a single value, the simultaneous burden of inflammation and the erosion of nutritional reserve.

The study itself was an observational cross-sectional analysis of 60 patients with confirmed COVID-19, stratified into three groups according to the level of care they required: outpatients treated without hospital admission, patients managed on general wards, and patients admitted to the intensive care unit. The researchers measured a broad panel of laboratory parameters, including standard inflammatory markers, a set of cytokines, and β2-microglobulin, a small protein component of the major histocompatibility complex class I molecule that is shed into the serum during heightened immune cell turnover. The retrospective design relied on stored serum samples collected during routine clinical practice, and the protocol received ethics approval from the Sakarya University Faculty of Medicine Health Sciences Scientific Research Ethics Committee, with the informed consent requirement waived because only archived specimens were used.

The central finding was a clear gradient across the three care levels. Intensive care patients exhibited significantly higher concentrations of inflammatory markers, cytokines, β2-microglobulin, and the HNI itself compared with ward patients and outpatients. This pattern is consistent with the established understanding of severe COVID-19 as a state of escalating immune activation, in which circulating interleukin-6, interleukin-10, and tumor necrosis factor-alpha rise in parallel with acute-phase proteins, and in which rapid lymphocyte turnover releases β2-microglobulin into the circulation. The fact that a composite index derived from three routine laboratory tests tracked this gradient as faithfully as the more specialized cytokine assays is the study’s most practically interesting observation.

The correlation analysis sharpened that picture considerably. HNI showed a very strong correlation with ferritin, with a correlation coefficient of 0.945, a figure so high that it raises an immediate statistical question: when a variable is part of the formula that defines an index, the two will inevitably move together, a phenomenon known as mathematical coupling. The authors themselves flag this concern explicitly, noting that the strong association between HNI and ferritin should be interpreted cautiously precisely because ferritin sits in the numerator of the index. Beyond ferritin, HNI showed moderate correlations with the measured cytokines, suggesting that the index does carry information about the broader inflammatory milieu rather than merely echoing a single component of its own formula.

Diagnostic performance was assessed using receiver operating characteristic analysis, a standard technique for evaluating how well a continuous marker discriminates between two groups, in this case presumably severe and non-severe disease. The area under the curve for HNI reached 0.953, a value that, on its face, would indicate near-perfect discrimination. An AUC of 1.0 represents flawless classification and 0.5 represents performance no better than chance, so 0.953 would place HNI among the most accurate severity markers reported for COVID-19. Yet the authors are careful to temper enthusiasm, emphasizing that this finding should be interpreted cautiously given the small sample size of 60 patients and the potential for mathematical coupling between the index and its constituent measurements.

That caution is well founded and reflects a broader lesson in biomarker research. Composite indices derived from routine laboratory values have proliferated during and after the pandemic, ranging from the neutrophil-to-lymphocyte ratio to more elaborate immunonutritional scores combining multiple cell counts and serum proteins. Their appeal is obvious: they are cheap, they require no specialized assays, and they can be calculated from blood tests already ordered for clinical reasons. Their peril is equally obvious: combining correlated variables can inflate apparent performance, small samples can produce optimistic estimates that shrink dramatically on validation, and the arithmetic of an index can create associations that do not exist in any single underlying measurement. The authors of the present study appear acutely aware of these pitfalls, describing HNI as an exploratory marker rather than a validated clinical tool.

Perhaps the most intellectually honest element of the paper is its conclusion about what HNI actually represents. Rather than claiming the index as an independent biomarker with novel predictive power, the authors conclude that it appears to reflect the combined burden of inflammation and nutritional status, and that its clinical utility beyond the individual parameters from which it is built remains uncertain. This framing matters. A composite index earns its place in clinical practice only if it adds information that its components do not already provide, either by improving risk stratification, by simplifying decision-making, or by capturing an interaction between inflammation and nutrition that neither dimension reveals alone. Whether HNI meets any of those tests cannot be determined from a 60-patient cross-sectional study; it requires validation in larger, ideally prospective cohorts with outcomes such as mortality, mechanical ventilation, and length of stay as endpoints.

The inclusion of β2-microglobulin in the analysis adds a distinctive dimension to the work. As a component of MHC class I molecules expressed on nearly all nucleated cells, β2-microglobulin is released during increased cellular turnover and immune activation, and elevated serum levels have been reported in severe infections, in autoimmune disease, and in conditions involving lymphocyte proliferation. Its significant elevation in the intensive care group of this COVID-19 cohort aligns with the picture of profound immune activation in critical disease, and its correlation with the composite index suggests that HNI, despite its simple formula, moves in step with markers of deep immunological disturbance. Whether β2-microglobulin itself offers prognostic value beyond established markers remains a question for larger studies, but its presence here underscores the study’s attempt to connect a bedside index to the underlying biology of the cytokine-driven inflammatory response.

The study, led by Gülsüm Kaya of Yalova University with colleagues including corresponding author Cengiz Karacaer of Sakarya University, was conducted without external funding and published as open access research on 29 September 2026. Its contribution is best understood as a carefully hedged proof of concept: evidence that a single arithmetic combination of C-reactive protein, ferritin, and albumin can track the severity gradient of COVID-19 and correlate with both cytokine levels and β2-microglobulin, offered with explicit acknowledgment of the mathematical and statistical caveats that such composite measures demand. For clinicians, the message is that routine laboratory data already contain much of the severity signal that specialized assays reveal, but that no new index should be adopted into practice until it has demonstrated, in adequately powered validation studies, that it adds genuine value beyond the numbers already on the chart. For researchers, the work provides a template for how composite biomarkers should be evaluated: with correlation analyses that acknowledge coupling, with ROC analyses interpreted in light of sample size, and with conclusions that distinguish what an index measures from what it independently predicts.

Subject of Research: A combined inflammatory and nutritional biomarker index for assessing COVID-19 severity

Article Title: Hyperinflammatory nutritional index as an exploratory combined inflammatory and nutritional marker: associations with cytokines and β2-microglobulin in COVID-19

Article References: Kaya, G., Karacaer, C., Sunu, C., Ergenç, H., & Öğütlü, A. (2026). Hyperinflammatory nutritional index as an exploratory combined inflammatory and nutritional marker: associations with cytokines and β2-microglobulin in COVID-19. BMC Infectious Diseases, 26(1), Article 1738. https://doi.org/10.1186/s12879-026-14525-7

Image Credits: AI Generated

DOI: 10.1186/s12879-026-14525-7

Keywords: COVID-19, hyperinflammatory nutritional index, C-reactive protein, ferritin, albumin, cytokines, β2-microglobulin, inflammation, nutritional status, biomarkers, disease severity, immunonutrition