Cardiopulmonary bypass (CPB) induces an acute systemic inflammatory response that is triggered by the interaction of blood elements with foreign surfaces and air, abnormal rheology and associated trauma. With much of postoperative morbidity attributable to it, the onset and development of inflammation that ensues CPB is regulated by the release of certain acute phase reactants.1–51–51–51–51–5 Several extensive studies have ascertained that the levels of certain cytokines are significantly raised following CPB thereby triggering an acute phase reaction, such as interleukin (IL)-1β, tumour necrosis factor-alpha (TNFα), IL-6, chemokines as IL-8 and monocyte chemoattractant protein-1 and other acute phase reactants, namely, C-reactive protein (CRP), fibrinogen and α2 macroglobulin and IL-10, an important anti-inflammatory cytokine.4–64–64–6
Moreover, surgical trauma also results from the contact of blood with air and tissues other than the endothelium in the mediastinal field, thus leading to an increased inflammatory response. Therefore, the possibility to avoid direct reinfusion of field-aspirated blood into the circulating volume, as well as extracorporeal circuits with a separate cardiotomy reservoir for shed blood, has been investigated in recent years.7–97–97–9 Oxygenators with separate cardiotomy reservoirs for mediastinal shed blood have incorporated mechanisms, if need arises, to process suctioned blood and reinfuse only red blood cells into the circuit, thereby maintaining an acceptable level of haematocrit and finally minimizing the potential amplification of inflammation within the circuit.
We therefore sought to investigate the level of CPB-associated inflammation in patients undergoing valve surgery by comparing the new Admiral oxygenator (which avoids direct reinfusion of mediastinal shed blood), compared with a conventional CPB circuit with an open cardiotomy reservoir.
Materials and methods
The study was a single-center prospective cohort complying with all institutional guidelines approved by the University of Brescia Medical Center. All patients signed an informed consent.
Thirty patients (with an age range among 18 and 80 years) were scheduled to undergo elective isolated aortic valve surgery and therefore assigned into two cohorts according to the type of oxygenator utilized during CPB. The Admiral group had patients on CPB with the new Admiral hollow-fibre membrane oxygenator (n = 15, Group 1, G1). The conventional group included patients on CPB with the Dideco D903 Avant Oxygenator (n = 15, Group 2, G2). Exclusion criteria were urgent/emergent surgery, redo or combined surgery, preoperative steroid therapy, preoperative anaemia (Hb <10 g/dl), chronic renal failure (>2 mg/dl) and documented coagulopathy. Patients’ characteristics, medical history, preoperative, operative and postoperative parameters were prospectively collected as reported in the Results section. In particular, bleeding was defined as requiring pericardial drainage; transfusion requirements as the need of a least 2 units of packed red blood cells.
All patients underwent aortic valve replacement through a minimally invasive approach, that is an upper J-shape ministernotomy (with a split at the third or fourth intercostal space). The ascending aorta was cannulated in all cases with a 24F straight-tip cannula (EZ Glide, Edwards Lifesciences, Irvine, California, USA), while venous drainage was achieved by means of a 25-Fr percutaneous cannula (QuickDraw, Edwards Lifesciences, Irvine) inserted in the femoral vein. Heparin (300 IU/kg) (Hospira, Lake Forest, Illinois, USA) was administrated before CPB. Myocardial protection was performed with intermittent infusion (every 20 min) of cold crystalloid cardioplegia (St. Thomas solution). All patients were scheduled to undergo aortic valve replacement with the same type of tissue valve (Magna Ease, Edwards Lifesciensces, Irvine).
Patients enrolled in the study were selected to undergo surgery either with a conventional oxygenator (Dideco D903 Avant; Arvada, Colorado, USA) or with a novel oxygenator (Admiral; Eurosets, Medolla, Italy), which allows for separate collection of mediastinal shed blood into a separate chamber; prior to reinfusion in the patient, if need arises, a convenient lever allows for cardiotomy blood to be directed to an autotransfusion line and being processed before retransfusion into the circulating volume. Both systems have a phosphorylcoline coating on the inner surface of the oxygenator membrane. In all instances, a S3 Stockerton heart-lung machine (Sorin, Arvada, Colorado, USA) was utilized. Volumes and composition of the priming solution with an electrolytic solution added with sodium gluconate (Fresenius Kabi, Verona, Italy) and mannitol were similar in both groups as reported in the results. Mild hypothermia (33°C) was utilized in all cases. No corticosteroids have been utilized among the anaesthetic protocol in the current series of patients.
In order to assess the degree of inflammation in either groups, we analyzed the following parameters pre, intra and postoperatively at different time points as reported below:
1. White blood cells (WBCs), platelets count and fibrinogen: preoperatively, and postoperatively after 1, 18, 36, 48, 72, 96 h.
2. CRP: preoperatively, and postoperatively after 1, 18, 36, 48 and 72 h
3. D-dimer: preoperatively and postoperatively after 1, 18 and 36 h
4. IL-6, IL-10, TNF-α: preoperatively, intraoperatively (1 h after institution of CPB), and postoperatively after 1, 24, 48 h
Haematologic parameters (WBC, platelet count, CRP, D-dimer, fibrinogen) were analyzed as per routine laboratory methods. IL-6, IL-10, TNF-α were assessed by means of ELISA with a sensitivity of less than 2 pg/ml (Thermo Scientific, Epsom, UK). Peripheral venous blood samples were collected into sterile EDTA tubes (Becton Dickinson, San Jose, California, USA) and immersed into melting ice, and then centrifuged within 5 min at 1600g for 10 min. Plasma storage in multiple aliquots was achieved at −80°C until the time of analysis and thawed once.
Data are reported as mean ± standard deviation and analyzed using nonparametric Mann–Whitney U-test or Wilcoxon test as appropriate for continuous variables, while Fisher exact test was used for discrete variables. A P value of less than 0.05 was considered statistically significant. All statistical analyses were performed using SPSS 13.0 for Windows (SPSS, Chicago, Illinois, USA).
Baseline patient characteristics and perioperative data were similar between the two groups as summarized in Tables 1 and 2Tables 1 and 2. All patients well tolerated the surgical procedure, in particular with regard to the CPB techniques, and survived without significant complications related to the study. Preoperative demographics did not vary in terms of age (G1 = 68.2 ± 10.4 years vs. G2 = 70.9 ± 10.1 years, P = NS) and EuroSCORE (G1 = 6.86 ± 2.4 vs. G2 = 6.23 ± 1.2, P = NS) among the groups. Intraoperatively, the durations of CPB (G1 = 104 ± 30.7 min vs. G2 = 98.9 ± 17.4 min, P = NS) and aortic cross-clamping time (G1 = 73.3 ± 17.5 min vs. G2 = 73.9 ± 19.1 min, P = NS) and the incidence of perioperative complications did not differ significantly between the two groups (Table 3). Of note, no clinically evident stroke occurred in the current study population.
The inflammatory response following CPB was significantly decreased in the Admiral group compared with the open-cardiotomy group, as assessed by means of IL-6, TNF-α, D-dimer, CRP and leukocytosis levels. A similar pattern was seen with all markers that generally rose during or right after the procedure and peaked postoperatively after which they tended to subside after 48 h.
The level of IL-6 was comparable preoperatively between the two groups, although it was significantly raised during CPB for the open cardiotomy group (G1 = 11.8 ± 12.5 pg/ml vs. G2 = 26.5 ± 24.9 pg/ml, P = 0.02), and remained so until 24 h postoperatively. On postoperative day 2 following the surgical procedure, IL-6 returned to similar levels in both groups (Fig. 1a).
IL-10 did follow other inflammatory markers in rising at CPB and then gradually declining over the postoperative period. However, there was no statistically significant difference in the levels between the two groups at any point in time, and the Admiral curve mirrored the open cardiotomy group curve (Fig. 1b).
Similarly, TNF-α levels continuously rose from the onset of CPB, with a steeper rise in the open cardiotomy group. The difference became significant at 1 h postoperatively (G1 = 29.1 ± 28.7 pg/ml vs. G2 = 45.5 ± 23.6 pg/ml, P = 0.03). Following the same pattern as IL-6, the levels dropped after that and were comparable between the two groups by 24 h postoperatively (Fig. 1c).
The levels of CRP were slower to rise once CPB was initiated, only becoming steeper after the first postoperative hour. In contrast to other markers suggesting more inflammation in the open-cardiotomy group, there was a faster rise of levels in the Admiral one that was significantly higher at 18 h postoperatively (G1 = 169.1 ± 164.8 mg/l vs. G2 = 57.1 ± 39.3 mg/l, P = 0.04). Although not significant after that point, they remained higher than those of the open cardiotomy group until 72 h postoperatively when they were numerically similar (Fig. 1d).
The level of leukocytosis followed the general pattern of rising immediately postoperatively and then declining gradually. There was reduced leukocytosis in the Admiral group, although it did not reach statistical significance at any time points (Fig. 2a).
Platelets count did not significantly differ among the two groups throughout the early postoperative period (Fig. 2b).
The D-dimer levels rose significantly higher immediately postoperatively in the open cardiotomy group (at 1-h postop, G1 = 1332.3 ± 953.88 ng/ml vs. G2 = 2791.9 ± 1740.7 ng/ml, P = 0.02). After the first postoperative hour, they declined and assumed similar levels in both groups, slightly falling more to be significantly less in the open cardiotomy group after 48 h (G1 = 1117.0 ± 501.9 ng/ml vs. G2 = 580.4 ± 272.4 ng/ml, P = 0.01) (Fig. 2c).
The baseline levels of fibrinogen were significantly higher in the Admiral group (G1 = 417.3 ± 91.8 mg/dl vs. G2 = 333.7 ± 62.2 mg/dl, P = 0.02). Both levels fell considerably after bypass as reflected in the 1-h postoperative values (G1 = 277.3 ± 111.1 mg/dl vs. G2 = 241.7 ± 65.9 mg/dl). The levels started rising thereafter and continued to do so till 48 h postoperatively after which they progressively plateaued (Fig. 2d).
The constant contact with air and tissues other than the endothelium significantly increases the inflammatory response during CPB. With increased understanding of the need to avoid direct reinfusion of field-aspirated blood into the circulating volume, there has also been an increasing interest in developing extracorporeal circuits with a separate cardiotomy reservoir for shed blood, as avoidance of its reintroduction into the circulating volume might yield incremental benefits. Moreover, during the past few years, there has also been a vivid interest towards the potential benefits of additional pharmacological strategies (e.g. with the use of ulinastatin) to reduce the inflammatory profile during the perioperative period, albeit with controversial results.10,1110,11
The pathophysiologic disturbance continues well into the postoperative period and is again reflected by the levels of the inflammatory markers, as confirmed by several studies. The measurement of inflammation in our study followed the established pattern that is associated with CPB.3,93,9 The levels of inflammatory markers generally rose following the onset of CPB and then gradually declined within the postoperative period. This archetype pattern was exemplified by our measurements of the proinflammatory cytokines as IL-6 and TNF-α. The increased levels of IL-6 have been noted in response to many major surgeries as well as CPB.9,129,12 Moreover, stimulating the production and release of acute phase reactants such as CRP from the liver has a considerable role to play in post-CPB inflammation. The open cardiotomy group had significantly increased levels of such acute inflammatory markers at the onset of CPB, although such difference was less evident thereafter. TNF-α is another potent proinflammatory cytokine that was significantly raised in the open cardiotomy group in the immediate postoperative period. The plasma concentrations of these early cytokines fall in the early postoperative period due to early degradation of these molecules.
An equally potent anti-inflammatory cytokine, IL-10 inhibits the synthesis and downregulates the actions of the proinflammatory cytokines, IL-6 and TNF-α. Bical et al.13 recently demonstrated significantly lower levels of IL-10 after CPB with a miniaturized extracorporeal circuit (MECC) when compared with a standard one in a study group of 20 patients. However, the IL-10 levels in our study remained similar throughout in both groups. The discrepancy could simply be a result of the fact that IL-10, which is released in response to the early effectors of inflammation, is an even more indirect estimate of the process.
CRP levels were significantly higher in the Admiral group after the first postoperative hour than those in the open cardiotomy group at that time. With a standard deviation almost mirroring the mean value, its statistical significance can be doubted. Bical et al.13 showed a comparable CRP levels in both groups. If adjusted for the eccentric spike, the CRP levels in our study are validated by studies comparing on-pump and off-pump coronary artery bypass grafting (CABG) wherein it was concluded that the nondifferent elevation suggests it is not activated by CPB as by the surgery itself.13,1413,14
The sudden fall of fibrinogen levels in both groups of this study at the onset of CPB merely reflect the activation of the coagulation cascade that is responsible for the thrombotic and bleeding complications of CPB. It is in part initiated by the interaction of surface-adsorbed fibrinogen (which undergoes conformational changes) with activated platelets and mediators of the contact pathway.15 Both adsorption onto biocompatible surfaces of the extracorporeal circuit and cleavage by thrombin contribute to its consumption and decline. The significantly higher baseline levels in the Admiral group result in a relatively higher (although not significant) level in that group throughout the measurements. But the fact that the gradients of the fibrinogen-time curve were comparable at all times is more significant from a speculative standpoint, implying a similar level of coagulation despite a considerable difference in baseline conditions. A side product of both coagulation and fibrinolysis, D-dimer, is a useful marker, as it is directly associated with the intensity of both processes.6,166,16 Its levels have been determined to increase during extracorporeal perfusion, indicating ongoing thrombin production, fibrin formation and fibrinolysis.6 In this study, the levels were significantly higher during the immediate postoperative period in the open cardiotomy group but remained similar thereafter.
Leukocytes are well documented major players in the CPB-associated inflammatory reaction as well.2,4,172,4,172,4,17 Responding to the barrage of early proinflammatory stimuli, leukocyte activation, the central cell type in this reaction, leads to further recruitment, activation and damaging cytotoxic effects. Ascione et al.3 showed significantly higher leukocytosis in patients undergoing cardiac surgery after CPB than without. Although not statistically significant in our patients, there was reduced leukocytosis in the Admiral group. A similar pattern of leukocytosis was noted by Ohata et al.,18 who compared miniaturized and conventional CPB systems.
As shown in the above results, the Admiral group fares significantly better than the open cardiotomy group in terms of the level of inflammation (as shown by IL-6, TNF-α, D-dimer and leukocyte counts) following CPB. The absence of any considerable difference in the postoperative outcomes of the two groups could be attributed to the low risk and small sample size of the study groups.
This study is not a prospective randomized evaluation, albeit the main characteristics of the patients enrolled in the study as well as the type of surgeries performed accounted for a comparison among homogenous groups. A better understanding of the effect of these two oxygenators on clinical outcomes could have been achieved by using higher risk patients. The low-risk profile of patients in this study accounted for a better tolerance of the increased level of inflammation, developing virtually no complications that could have been statistically assessed. The small sample size contributed to this effect as well, and also caused some statistical quirks as in the case of CRP and fibrinogen levels. Another major limitation of the study was that the measured levels of inflammatory cytokines and acute phase proteins were not corrected for hemodilution.
The findings of our study demonstrate that the Admiral is a better oxygenator than conventional open circuit oxygenators in terms of inflammatory response in patients undergoing valve surgery, although no considerable difference was detected in terms of peri-operative outcomes. Avoidance of direct reinfusion of mediastinal shed blood may yield potentially incremental benefits in patients with higher surgical risk or associated comorbities and further investigation in such subset of patients should be warranted.
No funding to be disclosed.
1. Asimakopoulos G, Taylor KM. Effects of cardiopulmonary bypass on leukocyte and endothelial adhesion molecules. Ann Thorac Surg
2. Vohra HA, Whistance R, Modi A, Ohri SK. The inflammatory response to miniaturised extracorporeal circulation: a review of the literature. Mediators Inflamm
3. Ascione R, Lloyd CT, Underwood MJ, Lotto AA, Pitsis AA, Angelini GD. Inflammatory response after coronary revascularization with or without cardiopulmonary bypass. Ann Thorac Surg
4. Paparella D, Yau TM, Young E. Cardiopulmonary bypass induced inflammation: pathophysiology and treatment. An update. Eur J Cardiothorac Surg
5. Seghaye M, Duchateau J, Bruniaux J, et al. Interleukin-10 release related to cardiopulmonary bypass in infants undergoing cardiac operations. J Thorac Cardiovasc Surg
6. Wi HJ. Cohn LH. Extracorporeal circulation: the response of humoral and cellular elements of blood to extracorporeal circulation. Cardiac surgery in the adult
. New York: McGraw-Hill; 2008. 370–389.
7. DeSomer F, Van BY, Caes F, et al. Phosphorylcholine coating offers natural platelet preservation during cardiopulmonary bypass. Perfusion
8. Khosravi A, Skrabal CA, Westphal B, et al. Evaluation of coated oxygenators in cardiopulmonary bypass systems and their impact on neurocognitive function. Perfusion
9. Fromes Y, Gaillard D, Ponzio O, et al. Reduction of the inflammatory response following coronary bypass grafting with total minimal extracorporeal circulation. Eur J Cardiothorac Surg
10. Song J, Park J, Kim JY, et al. Effect of ulinastatin on perioperative organ function and systemic inflammatory reaction during cardiac surgery: a randomized double-blinded study. Korean J Anesthesiol
11. Chen TT, Jiandong-Liu, Wang G, Jiang SL, Li LB, Gao CQ. Combined treatment of ulinastatin and tranexamic acid provides beneficial effects by inhibiting inflammatory and fibrinolytic response in patients undergoing heart valve replacement surgery. Heart Surg Forum
12. Torre-Amione G, Kapadia S, Lee J, Bies RD, Lebovitz R, Mann DL. Expression and functional significance of tumor necrosis factor receptors in human myocardium. Circulation
13. Bical OM, Fromes Y, Gaillard D, et al. Comparison of the inflammatory response between miniaturized and standard CPB circuits in aortic valve surgery. Eur J Cardiothorac Surg
14. Franke A, Lante W, Fackeldey V, et al. Pro-inflammatory cytokines after different kinds of cardio-thoracic surgical procedures: is what we see what we know? Eur J Cardiothorac Surg
15. Lindon J, McManama G, Kushner L, Merrill EW, Salzman EW. Does the conformation of adsorbed fibrinogen dictate platelet interactions with artificial surfaces? Blood
16. Gram J, Janetzko T, Jespersen J, Bruhn HD. Enhanced effective fibrinolysis following the neutralization of heparin in open heart surgery increases the risk of postsurgical bleeding. Thromb Haemost
17. Butler J, Parker D, Pillai R, Westaby S, Shale DJ, Rocker GM. Effect of cardiopulmonary bypass on systemic release of neutrophil elastase and tumor necrosis factor. J Thorac Cardiovasc Surg
18. Ohata T, Mitsuno M, Yamamura M, et al. Minimal cardiopulmonary bypass attenuates neutrophil activation and cytokine release in coronary artery bypass grafting. J Artif Organs