Inflammatory markers, iron-related parameters and anemia of inflammation dynamics in acute infections: Insights from comparison of a COVID-19 pneumonia and a sepsis cohort

BACKGROUND: Iron is essential for nearly all organisms, including microbes, as a functional component of many proteins and enzymes involved in vital biochemical functions. In vertebrates, iron metabolism is mostly regulated by the hepatic hormone hepcidin, which plays a role in the so-called “nutritional immunity”, the battle for nutrient metals at the host–pathogen interface. During acute infection, hepcidin production is mainly induced by interleukin (IL)-6, leading to iron sequestration in macrophages and, in turn, iron-restricted erythropoiesis. Both hypoferremia and inflammatory cytokines suppress erythropoiesis, possibly leading to anemia of inflammation (AI). However, host response to invading pathogens is complex, mainly studied in vitro or in animal models, and emerging evidence suggests different patterns of iron homeostasis alteration depending on infection etiology. AIMS: This study aimed to: (1) compare the evolution of inflammatory markers, iron metabolism, and AI in two different cohorts of patients hospitalized for severe acute infections; investigate (2) predictors of in-hospital mortality; (3) mechanisms of hepcidin regulation in vivo; (4) possible predictors of microbial etiology. MATERIALS AND METHODS: A total of 447 patients were enrolled within 48 hours from admission to a medical ward of Verona University Hospital and followed up during hospitalization; 391 were affected by COVID-19 pneumonia and 56 by sepsis. Clinical, biochemical, and iron-related parameters were obtained at defined time points. Cytokine assays were assessed for the COVID-19 pneumonia cohort. Immunological analysis of circulating mononuclear cells was performed for the sepsis cohort, which was followed up after 3 months from discharge. RESULTS: Although the COVID-19 pneumonia patients were older (71 [61-80] vs 65 [53-76] years, p=0.001) and more frequently comorbid (85.2 vs 59.6%, p=0.004) than septic patients, the two cohorts did not differ significantly for disease severity and mortality rate. Baseline anemia was far more frequent in the sepsis compared to the COVID-19 pneumonia cohort (78.6 vs 31.2%, p<0.001), and all sepsis nonsurvivors were anemic. At 3-month follow-up, anemia prevalence was still relevant (38.1%) despite iron biomarkers generally returning to normal values. The baseline biochemical and iron-metabolism profile differed significantly among the two cohorts, with lower C-reactive protein (CRP) and white blood cell (WBC) count and higher ferritin and hepcidin levels observed for COVID-19 pneumonia patients. In COVID-19 pneumonia patients, female sex, increased age and IL-8, and a decrease of PaO2/FiO2 (P/F) ratio and transferrin upon admission significantly predicted mortality. Hemoglobin, CRP, Neutrophil to Lymphocyte ratio, creatinine, erythroferrone, soluble transferrin receptor, IL-1Ra, IL-6, and tumor necrosis factor-α, but not P/F ratio, were significantly associated with basal hepcidin values. In the sepsis cohort, basal hepcidin >160 ng/mL was associated significantly with 30-day mortality (log-rank 4.154, p=0.042). Circulating myeloid precursors were detected across all time points. Their counts were higher in nonsurvivors compared to survivors, in whom a partial, although not full, reduction was observed at 3-month follow-up. Baseline anemia, high CRP and WBC count, and low hepcidin values were significantly associated with a higher risk of sepsis and a lower risk of COVID-19 pneumonia in binary logistic regression. CONCLUSIONS: This study of well-characterized patients highlights a profound alteration of iron homeostasis during acute COVID-19 pneumonia and sepsis, showing important differences possibly reflecting different inflammatory patterns. In-hospital mortality was associated with different factors in the two cohorts. In COVID-19 pneumonia, inflammation and erythropoiesis, rather than the hypoxic drive, seem to upregulate hepcidin at baseline. Specific hematological biomarkers have been identified as possible predictors of infection etiology.