REmoval of cytokines during CArdiac surgery (RECCAS): a randomised controlled trial

Abstract

Background: Cardiopulmonary bypass (CPB) triggers marked cytokine release often followed by a systemic inflammatory response syndrome, associated with adverse postoperative outcomes. This trial investigates the intraoperative use of haemoadsorption (HA) during cardiac surgery with CPB to assess its impact on postoperative systemic inflammatory response.

Methods: In this prospective randomised controlled trial (ethics approval no. 5094-14DRKS00007928), patients (> 65 years) undergoing elective on-pump cardiac surgery were randomised to intraoperative HA (CytoSorb) during CPB or standard care without HA. Primary outcome was the difference in mean interleukin (IL)-6 serum concentrations between groups on intensive care unit (ICU) admission. The secondary outcomes included various clinical and biochemical endpoints. Statistical methods included paired and unpaired t-tests, Wilcoxon, Mann-Whitney U-tests, and chi-square tests.

Results: Thirty-eight patients were allocated to receive either intraoperative HA (n = 19) or standard care (n = 19). The primary outcome, IL-6 levels on ICU admission, did not differ between the study group and controls (214.4 ± 328.8 vs. 155.8 ± 159.6 pg/ml, p = 0.511). During surgery pre- versus post-adsorber IL-2, IL-6, IL-8, IL-10, heparan sulfate and myoglobin post- levels were reduced. Furthermore, IL-6 levels did not differ between the study groups on day 1 and 2 in the ICU. While sequential organ failure assessment scores, lactate levels, and C-reactive protein and procalcitonin (PCT) showed no statistically significant differences. Regarding haemodynamic stability in the treatment group the cardiac index (3.2 ± 0.7 vs. 2.47 ± 0.47 l/min/m², p = 0.012) on ICU day 2 increased, and lower fluid requirements as well as decreased fibrinogen requirement were observed. Need for renal replacement therapy did not differ though a shorter duration was observed in the treatment group. Time on ventilator, respiratory parameters, infectious complications, delirium scores, ICU and hospital lengths of stay, and mortality did not differ between groups.

Conclusion: HA did not reduce the IL-6 level on ICU admission or afterwards. Even though HA reduced cytokine load during cardiac surgery in the treatment group. There were no significant differences between groups in the postoperative course of other cytokine concentrations, organ dysfunction, ICU and hospital lengths of stay and mortality rates. Trial registration prospectively DRKS00007928 and published under: Baumann A, Buchwald D, Annecke T, Hellmich M, Zahn PK, Hohn A. RECCAS - REmoval of Cytokines during Cardiac Surgery: study protocol for a randomised controlled trial.

Trials: 2016;17: 137.

Keywords: Cardiac Surgery; Cardiopulmonary bypass; Cytokines; Haemoadsorption; Heart–lung-machine; Inflammation.

Fig. 1

Flow Diagram of Study Design. Consolidated Standards of Reporting Trials (CONSORT) flow diagram of the study design depicting recruiting and dropouts of the REmoval of Cytokines during CArdiac Surgery [2] trial. N = 38. Abbreviations C3a, complement component 3a; CPB, cardiopulmonary bypass; HA, haemoadsorption, AIDS, acquired immunodeficiency syndrome; ARF score, acute renal failure score; ASA, American Society of Anesthesiologists; CBP, cardiopulmonary bypass; CD4, cluster of differentiation 4, Cl, cardiac output; EF, ejection fraction, EuroSCORE, European System for Cardiac Operative Risk Evaluation; GEDI, Global End-Diastolic Volume Index, GFR, glomerular filtration rate; HA, haemoadsorption; ICU, intensive care unit; IL, interleukin, MAP, mean arterial pressure; NuDESC, Nursing Delirium Screening Scale; NYHA, New York Heart Association; PCT, procalcitonin; RRT, renal replacement therapy, SOFA, sequential organ failure assessment; TNF-a, tumor necrosis factor alpha; PaO₂/FiO₂, ratio of arterial oxygen partial pressure to fractional inspired oxygen

Fig. 2

IL-6 cytokine dynamic after HA. Boxplots depicting the levels of cytokine interleukin (IL)-6 before surgery (baseline) and during the course of intensive care unit (ICU) treatment (admission, day 1 (d1), and day 2 (d2) at ICU) comparing the control group (white bars, n = 19) to the treatment group (dark grey bars, n = 19) receiving haemoadsorption (HA) during cardiothoracic surgery with cardiopulmonary bypass (CPB). Variables are presented as boxplots (10th–90th percentile, median, plus mean values)

Fig. 3

Cytokine IL-2, IL-8, IL-10, TNF-alpha dynamic after HA. Boxplots depicting the levels of cytokine IL-2 (A), IL-8 (B), IL-10 (C) and TNF-alpha (D) before surgery (baseline) and during the course of intensive care unit (ICU) treatment (admission, day 1 (d1), and day 2 (d2) at ICU) comparing the control group (white bars, n = 19) to the treatment group (dark grey bars, n = 19) receiving haemoadsorption (HA) during cardiothoracic surgery with cardiopulmonary bypass (CPB). Variables are presented as boxplots (10th–90th percentile, median, plus mean values). Statistical analysis: Mann–Whitney-U-test of raw data after normality testing (Kolmogorov–Smirnov). *p < 0.05. N = 38

Fig. 4

Plasma protein dynamic after HA. Boxplots depicting the levels of free haemoglobin (A), myoglobin (B), hyaluronan (C) and syndecan (D) before surgery (baseline) and during the course of intensive care unit (ICU) treatment (admission, day 1 (d1), and day 2 (d2) at ICU) comparing the control group (white bars, n = 19) to the treatment group (dark grey bars, n = 19) receiving haemoadsorption (HA) during cardiothoracic surgery with cardiopulmonary bypass (CPB). Variables are presented as boxplots (10th–90th percentile, median, plus mean values). Statistical analysis: Mann–Whitney-U-test of raw data after normality testing (Kolmogorov–Smirnov). *p < 0.05. N = 38

Fig. 5

Clinical outcomes after HA. Bar graphs depicting Cardiac Index (CI) (A), duration of dobutamine need in hours until d2 (control: n = 6, treatment group n = 10) (B), duration of adrenaline need in hours until d3 (control: n = 2, treatment group n = 2) (C), need for crystalloid fluid replacement up to d1 (D), duration of renal-replacement-therapy (RRT) (control: n = 3, treatment group n = 3) (E), number of patients needing fibrinogen (black) (F) during the course of intensive care unit (ICU) treatment (admission, day 1 (d1), and day 2 (d2) at ICU) comparing controls (white bars, n = 19) and treatment group (dark grey bars) having received haemoadsorption (HA) during cardiothoracic surgery with cardiopulmonary bypass (CPB). All variables are presented as mean ± standard deviation (SD), except panel F presenting number of patients and percentage. N = 39. Statistical analysis: A–C t-test, D chi-square

Fig.6

Pre- versus post-adsorber cytokine serum concentrations under HA during surgery and clearance. Boxplots illustrating the levels of cytokines, IL-6 (B), IL-8 (C), IL-10 (D), TNF-alpha (E)), fibrinogen (F), free haemoglobin (G), and myoglobin (H) pre- and post-adsorber (dark grey) haemoadsorption (HA) during cardiothoracic surgery with cardiopulmonary bypass (CPB) at 10, 30, and 60 min compared to baseline (preoperative, white). (I) Clearance of IL-6, IL-8, and IL-10 at 10, 30, and 60 min after CPB initiation. Individual clearance trajectories demonstrate positive clearance values for IL-6 and IL-8 at least at one timepoint. For IL-6, three patients, for IL-8, two patients, initially exhibited negative clearance values. Data is based on n = 19 participants from the treatment group only. All variables are presented as boxplots (10th–90th percentile, median, plus mean values). Statistical analysis for (A–H) Wilcoxon test if normality was not met (Kolmogorov–Smirnov) or paired t-test for fibrinogen and TNF-α. n = 19, *p < 0.05, ** < 0.01, *** < 0.001