Characterizing Cytokine Profile for Enhancement of Early Prediction of Burn-Induced Sepsis

Authors:
Anna Dvorkin, Diana Tedesco, Fadi Khalaf, Marc Jeschke, Maria Fernanda Hutter, Zachary Ricciuti

Body of Abstract:
Background: The early identification and initiation of treatment for sepsis following burn injury remains a major issue in burn patients. The lack of predictive biomarkers drives unacceptably high mortality and morbidity. Sepsis, which occurs approximately 7-10 days following a burn injury, is the leading cause of death in burn patients. Despite advances in burn care and improvements in the survival rate of burn patients, many individuals continue to die from burn-induced sepsis as we lack the necessary physiological insights required to enhance early identification. Moreover, comprehensive temporal characterizations of relevant clinical biomarkers, such as cytokines, could enable more effective early diagnosis and thus improve outcomes for patients. Furthermore, the goal of this study was to characterize the cytokine profile of septic and non-septic burn patients across multiple timepoints.
Methods: Patients admitted to a regional burn center over 15 years were included based on the following criteria: adults (>18 years) with thermal injuries affecting >5% of total body surface area (TBSA). Patients were categorized into two groups based on the presence or absence of sepsis during hospitalization. Circulating cytokines were detected in the serum of burn patients collected at the following timepoints post burn: 0-3, 4-9, 10-16, 17-30, and 31+ days. Pairwise comparisons were done between septic and non-septic patients at each timepoint. Wilcoxon rank sum test was used to determine significance and Benjamin-Hochberg test to further correct for multiple comparisons.
Results: A total of 341 patients were included. Patients with sepsis had a significant elevation of the anti-inflammatory cytokine G-CSF at timepoints 2 (p<0.05), 3 (p<0.001), 4 (p<0.05), and 5 (p<0.05). IFN-α2, possessing both pro and anti-inflammatory properties, was significantly elevated within septic patients at timepoints 1(p<0.05), 3 (p<0.05), and 4 (p<0.05). Both G-CSF and IFN-α2 demonstrated their strongest fold change within timepoints 5 (3.15) and 4 (2.15), respectively. The proinflammatory cytokine IL-15 was significantly elevated in patients with sepsis at timepoints 1 (p<0.05), 2 (p<0.001), and 3 (p<0.05) as well as the proinflammatory TNF-α which was significantly elevated at timepoints 1 (p<0.05), 2 (p<0.05), 3 (p<0.05), and 4 (p<0.05). Interestingly, IL-15 and TNF-α demonstrated a reduction in fold change between their first and last significant timepoint, decreasing from 2.42 and 2.14, respectively to 1.72 and 1.45, respectively. The proinflammatory IFN-γ was significantly elevated at timepoints 1 (p<0.05), 2 (p<0.05), and 4 (p<0.05) displaying the greatest fold changes of all measured cytokines at 8.26, 2.72, and 4.39, respectively. Conclusion: This study provides time-dependent characterization of cytokine profiles between septic and non-septic burn patients, identifying 5 cytokines which may potentially be used as effective diagnostic criteria for burn-induced sepsis