72 Chapter 4 In this study there was no association between hyperferritinemia and the causative pathogen of CAP, neither for specific pathogens or for bacterial and viral causes in general. Iron is an essential micronutrient for almost all living organisms, including bacteria, and several pathogens have developed mechanisms to extract iron from ferritin [24, 25]. Epidemiological studies have suggested an association between conditions with iron overload and an increased frequency of infections with a variety of microbes [26, 27], including Still’s disease [28] and hemochromatosis [29]. The vast majority of patients included in our study did not have a comorbidity linked with hyperferritinemia. As such, the current data do not provide insight into the association between hyperferritinemia and increased susceptibility to pneumonia. Likewise, whether strongly elevated ferritin levels induced by infection facilitate microbial multiplication remains to be established. Hyperferritinemia modestly associated with the severity of disease on hospital admission; only in the analysis using a cut-off of 500 ng/ml the PSI score was higher in patients with high ferritin concentrations, while the MEWS and qSOFA scores were not different between groups. Nevertheless, most markers reflecting systemic inflammation were higher in patients with hyperferritinemia including sCD163 (like ferritin considered a marker for macrophage activation) [1, 30]. The plasma concentrations of MPO and proteinase-3 (neutrophil products derived from azurophilic granules) were higher in patients with hyperferritinemia, while NGAL (a neutrophil protein derived from secondary granules) was not; notably, neutrophil numbers did not differ between groups stratified according to ferritin levels, suggesting that hyperferritinemia may be a marker of neutrophil activation relating to azurophilic granule release. We measured a large set of endothelial cell biomarkers [31, 32] pointing at enhanced activation (higher sVCAM-1, von Willebrand factor, s-thrombomodulin and TFPI levels), a more disturbed glycocalyx integrity (higher syndecan levels) and a reduced endothelial barrier function (higher angiopietin-2) in patients with hyperferritinemia. Endothelial cell dysfunction has been suggested as a potential target in the treatment of sepsis [31, 32]; the current data suggest that hyperferritinemia may assist in identifying patients who might benefit from vasculoprotective agents. D-dimer concentrations were higher in patients with hyperferritinemia, pointing at enhanced coagulation activation. The higher circulating levels of the endogenous anticoagulant proteins s-thrombomodulin and TFPI in the presence of hyperferritinemia likely reflect increased shedding from the activated vascular endothelium, which is expected to result in a loss of their cell-associated anticoagulant properties and to stimulate a procoagulant phenotype [31, 33]. Recently, the term MALS was introduced for a combination of clinical and laboratory features characterized by fever, hepatosplenomegaly, hepatobiliary dysfunction, coagulopathy, pancytopenia and hypertriglyceridemia [1, 10]. Very high ferritin concentrations (>4420 ng/ml, i.e., eight to 17-fold higher than the cut-off values used in the present study) were proposed as a surrogate biomarker for MALS with 97% specificity and 98% negative predictive value [1, 10]. The incidence of MALS in critically ill sepsis patients was reported to vary between 3 and 4%, and presence of MALS was independently associated with early mortality. Our group recently found a MALS incidence of 10%, defined by extreme hyperferritinemia (>4420 ng/ml), in patients with CAP admitted to the ICU [4]. MALS in critically ill CAP patients was associated with increased mortality and exaggerated systemic inflammation, cytokine release, endothelial cell activation and coagulation activation [4], resembling the host response
RkJQdWJsaXNoZXIy MTk4NDMw