Identification of Predictive Early Biomarkers for Sterile-SIRS after Cardiovascular Surgery

Abstract

Systemic inflammatory response syndrome (SIRS) is a common complication after cardiovascular surgery that in severe cases can lead to multiple organ dysfunction syndrome and even death. We therefore set out to identify reliable early biomarkers for SIRS in a prospective small patient study for timely intervention. 21 Patients scheduled for planned cardiovascular surgery were recruited in the study, monitored for signs of SIRS and blood samples were taken to investigate biomarkers at pre-assigned time points: day of admission, start of surgery, end of surgery, days 1, 2, 3, 5 and 8 post surgery. Stored plasma and cryopreserved blood samples were analyzed for cytokine expression (IL1β, IL2, IL6, IL8, IL10, TNFα, IFNγ), other pro-inflammatory markers (sCD163, sTREM-1, ESM-1) and response to endotoxin. Acute phase proteins CRP, PCT and pro-inflammatory cytokines IL6 and IL8 were significantly increased (p<0.001) at the end of surgery in all patients but could not distinguish between groups. Normalization of samples revealed significant increases in IL1β changes (p<0.05) and decreased responses to endotoxin (p<0.01) in the SIRS group at the end of surgery. Soluble TREM-1 plasma concentrations were significantly increased in patients with SIRS (p<0.01). This small scale patient study could show that common sepsis markers PCT, CRP, IL6 and TNFα had low predictive value for early diagnosis of SIRS after cardiovascular surgery. A combination of normalized IL1β plasma levels, responses to endotoxin and soluble TREM-1 plasma concentrations at the end of surgery are predictive markers of SIRS development in this small scale study and could act as an indicator for starting early therapeutic interventions.

Fig 1. SIRS parameters of selected study patients.

(A) Body temperature, (B) heart rate, (C) breathing rate, (D) PaCO₂ and (E) leukocyte counts are presented as means ± SEM per study group. Dotted lines indicate threshold values to pathological levels, shaded areas between dotted lines indicate the physiological ranges. * p < 0.05 Kruskal-Wallis with Dunn’s multiple comparison post-hoc test for selected data pairs.

Fig 2. Normalized changes in IL1β and IL10 plasma cytokine levels have potential predictive value for early recognition of SIRS.

Shown are fold changes ± SEM in plasma concentrations of (A) IL1β and (B) IL10 per patient, normalized to the day of admission, over the collected time points. * p < 0.05; ** p < 0.01; *** p < 0.001 Kruskal-Wallis with Dunn’s multiple comparison post-hoc test for selected data pairs. $ p < 0.05 Mann Whitney U-test.

Fig 3. Normalized plasma levels of endocan and cfDNA rise significantly slower whereas soluble TREM-1 plasma levels are upregulated at early time points in SIRS patients.

(A) Fold changes in endocan (ESM-1) ± SEM plasma levels of individual patients normalized to the corresponding plasma levels at admission. The fold increase is significantly lower in SIRS group than in control group ** p < 0.01 Kruskal-Wallis with Dunn’s multiple comparison post-hoc test for selected data pairs; $$ p < 0.01 Mann Whitney U-test. (B) Individual plasma levels of cfDNA ± SEM (expressed as fold changes over admission value) tend to increase during surgery but the rise is lower in SIRS than in the control group. * p < 0.05 Kruskal Wallis with Dunn’s multiple comparison post hoc test for selected pairs. (C) Soluble human triggering receptor expressed on myeloid cells-1 (sTREM-1) ± SEM plasma levels are significantly increased in SIRS patients at the end of surgery and day 2 after surgery compared to the control group. * p < 0.05; ** p < 0.01 Kruskal-Wallis with Dunn’s multiple comparison post hoc test for selected data pairs. A trend to an increased expression at the end of surgery within the SIRS group is also seen. p = 0.0556 Mann Whitney U-test.

Fig 4. Monocyte activation test (MAT) revealed reduced responsiveness in SIRS patients.

Cryopreserved blood of each patient at individual time points was stimulated with reference endotoxin (0.5 EU/ml) corresponding to 0.04 EU/ml in the actual incubation for 8 h at 37°C 5% CO₂. (A) Individual stimulation responses (shown as IL1β concentrations ± SEM) of diluted cryopreserved patient blood to an LPS stimulus show a tendency towards decreased responses (p = 0.1508) at the end of surgery but increased responses one day after surgery (p = 0.0952) in the SIRS group; Mann Whitney U-test. (B) When the responses to the LPS stimulus are normalized to the responses on the admission day, there is a significant decrease at the end of surgery obvious in the SIRS but not in the control group. ** p < 0.01 Kruskal-Wallis with Dunn’s multiple comparison post hoc test for selected data pairs.