ETD

• biomarker coagulativi
• infezioni del torrente ematico
• sepsi

Background: The modern definition of sepsis emphasizes the concept of “dysregulated host response to infection.” While various biomarkers of infection and organ dysfunction are currently well defined, biomarkers capable of adequately reflecting the concept of dysregulation of the response to infection have not been identified; considering the crucial role played by immunothrombosis in the regular and physiological response of the host to infection, biomarkers capable of reflecting sepsis-induced coagulopathy have the potential to represent an effective model for studying sepsis.
Aim: The aim of this pilot study is to evaluate whether alterations in various coagulation biomarkers occur differently in models of bloodstream infection by Gram+ and Gram- bacteria. Furthermore, the study aims to evaluate the possible correlation of biomarkers with the clinical outcome of 30-day mortality.
Results: A total of 49 patients with BSI were included in the study, forty-six had monomicrobial BSI, two had Gram-negative polymicrobial BSI, and one patient had Gram-positive polymicrobial BSI. Thirty-seven out of 49 (75.5%) infections were caused by Gram-negative bacteria and 12 out of 49 (24.5%) by Gram-positive bacteria. Among the most common Gram- species were E. coli (26.5%), Klebsiella pneumoniae (14.3%), and P. mirabilis (10.2%), while the most frequently involved Gram + bacterium was S. aureus (10.2%).
In the group of patients with Gram - BSI, compared to the group with Gram + infections, D-dimer levels tended to be higher and PC, PS, and AT levels slightly lower, although the difference was not statistically significant. PT and aPTT were not significantly different in the two groups. Platelet counts were significantly lower in the Gram-negative BSI group than in the Gram-positive group (165 vs. 299 x103/mmc, p=0.002).
D-dimer values were lower in patients who died within 30 days than in survivors (2366 ± 1418 vs. 3254 ± 4220 ng/mL). No coagulation biomarker showed a statistically significant association with 30-day mortality.
Extremely high D-dimer values were found in the K. pneumoniae infection model, particularly compared to the values found in E. coli (6200 ± 8827 vs. 2308 ± 1953 ng/mL), as well as a significantly lower platelet count in cases of BSI caused by E. coli compared to other etiologies, values that are close to statistical significance (126 ± 64 vs. 203 ± 123 x103/mmc, p=0.058).
Furthermore, compared to “other Enterobacterales,” lower values were observed for E. coli for D-dimer (2308±1953 vs. 4293±6378 ng/ml), PC (76 vs. 91%), PS (67 vs. 75%), AT (80 vs 88%), higher values of PT (15.2 vs 13.7 seconds) and aPTT (31.9 vs 28.5), and a significantly higher percentage of patients with coagulopathy identified with SIC score and DIC using mISTH. Despite the clear trend of difference, no statistically significant differences were found.
Conclusions: Although our pilot study did not allow us to identify statistically significant differences for some parameters due to the small sample size, it confirms the important role that coagulation biomarkers can play in sepsis and how sepsis-induced coagulopathy differs significantly depending on the different pathogens, with significant differences even within the Gram+ and Gram- groups.
In addition to the pathophysiological interest of this type of study, which deserves further investigation using appropriate methodology and, if possible, a prospective design, it opens up prospects for specific antithrombotic therapies targeted according to different infection models, in line with precision medicine.