Humoral Immune Response During Coronary Artery Bypass Grafting
A Comparison of Limited Approach, “Off-Pump” Technique, and Conventional Cardiopulmonary Bypass
Background—The introduction of limited approaches to the heart and the avoidance of cardiopulmonary bypass (CPB) aim to reduce the invasiveness of CABG by decreasing the systemic release of inflammatory cytokines, such as tumor necrosis factor (TNF)-α, interleukin (IL)-6, and IL-8, as well as the anti-inflammatory agent IL-10. This study compares the humoral immune response in patients undergoing CABG with standard, minimally invasive, and “off-pump” techniques.
Methods and Results—Thirty patients were divided into 3 operative groups: full sternotomy approach plus CPB (group A); full sternotomy approach, off pump (group B); and limited left anterior thoracotomy, off pump (group C). Plasma levels of TNF-α receptors p55 and p75, IL 6, IL-8, and IL-10 were taken at baseline, during CPB, and at 4, 24, and 48 hours and 6 days after surgery. A significant increased release of activated complement factors C5a and C3d, IL-8, and IL-10 was observed in patients subjected to CPB (group A) during the initial period and for a short time after perfusion (P<0.05). TNF-α receptors p55 and p75 showed a prolonged elevation (up to 48 hours) in the CPB group compared with the 2 off-pump groups. IL-6 showed no different release among the 3 surgical groups throughout the entire period. There was no significant difference in any parameter measured in relation to the type of operative approach.
Conclusions—There is an inflammatory, as well as an anti-inflammatory, response during CABG that is related to the general surgical trauma. The release of immune mediators is enhanced by the use of CPB during various perioperative and postoperative phases. The type of operative approach did not influence this immune response.
The use of cardiopulmonary bypass (CPB) is associated with a systemic inflammatory response. Because of the exposure to a large foreign body, systemic endotoxemia occurs and is followed by a release of inflammatory and anti-inflammatory cytokines.1 2 Cytokines such as interleukin (IL)-6, IL-8, and IL-10 as well as tumor necrosis factor (TNF) and the activated complement factors C5a for the alternative pathway and C3d for the classical immune pathway are released. Systemic immune response syndrome, the capillary leakage syndrome, and reperfusion injury of the lungs or the heart are entities reflecting the possible clinical impact of an unphysiological activation of the immune system. Independent of CPB, a general activation of the immune system is observed during any operative procedure as a physiological response to the surgical trauma. Because new “off-pump” techniques have been introduced in CABG, it is now possible to differentiate the influence of CPB and surgical access on different modalities of the immune response. The present study evaluates the humoral immune response by comparing the conventional CABG technique with 2 different approaches for off-pump CABG procedures.
After approval by the local ethics committee, 30 patients with 1-, 2-, or 3-vessel coronary artery disease were subsequently enrolled in the study. Inclusion criteria were as follows: age <75 years, stable angina, and an ejection fraction >35%. Exclusion criteria were as follows: acute myocardial infarction <2 weeks, active or prior history of autoimmune disorders, medication with immune-modulating agents such as steroids or antiphlogistics, elevated white blood cell or C-reactive protein levels, or any history or signs of infectious disease before surgery. All patients underwent nonemergent CABG surgery. Patients in all 3 groups received the same intravenous anesthesia, consisting of sufentanil (Sufenta, Janssen-Cilag) and propofol (Disoprivan, Glaxo Welcome Zeneca) under a standardized protocol. Patients were separated into 3 groups. Patients in group A (with CPB) were subjected to a median sternotomy. In all patients, the internal thoracic artery was harvested as a pedicle. The CPB circuit consisted of roller pumps (Stoekert), a membrane oxygenator (Maxima Forte, Medtronic), an open hard shell reservoir (Medtronic), and a 40-μL arterial filter (Medtronic). Full-dose heparin (300 U/kg) was applied before cannulation of the ascending aorta and the right atrium for installation of CPB. Aprotinin (Trasylol, Bayer) was given at a dosage of 4 000 000 U to all patients of group A before CPB. The distal graft anastomoses were performed on the arrested heart with the use of crystalloid cardioplegia (Bretschneider-HKT, Köhler) and normothermic blood temperature. The proximal anastomoses were performed during reperfusion of the beating heart on CPB. After coming off bypass, the patient was given protamine to neutralize the heparin dosage to 100%.
Patients of group B (off-pump group) were subjected to a median sternotomy, and harvest of the internal thoracic artery was performed as described above. Heparin was administered at a dosage of 100 U/kg to achieve an activated clotting time of >250 seconds. The distal anastomoses were completed with the use of mechanical stabilizers for immobilization of the myocardial surface at the site of the target coronary artery (CTS, CardioThoracic Systems, or Octopus, Medtronic). To obtain a bloodless field, the coronary artery was temporarily occluded proximal and distal to the anastomotic site with a 4-0 monofilament suture snared over a piece of pericardium. The anastomoses were completed with the same technique as described for group A. The proximal anastomoses were performed during tangential clamping of the ascending aorta. Heparin was neutralized with protamine to 80%. Group C patients underwent a minimally invasive direct coronary artery bypass (MIDCAB) procedure through an anterolateral (7- to 10-cm) thoracotomy at the left fourth intercostal space. Harvest of the internal thoracic artery was performed under direct vision by use of a retractor (ThoraLift, US Surgical). Heparin was administered at a dosage of 100 U/kg to achieve an activated clotting time of >250 seconds. After a partial opening of the pericardium, the left anterior descending coronary artery was identified. Stabilization and temporary vascular occlusion were achieved as described above. The anastomosis was performed by using a running suture technique (8-0 Prolene, Ethicon). Protamine was applied to neutralize 80% of the heparin.
Blood samples were taken at 9 time points for each patient as follows: sample 1, the day before surgery; sample 2, 10 minutes after introduction of anesthesia; sample 3, 10 minutes after the sternotomy; sample 4, 10 minutes after the onset of CPB (group A only); sample 5, 10 minutes after reperfusion of all grafts; sample 6, in the intensive care unit (ICU) 4 hours after surgery; sample 7, postoperative day 1; sample 8, postoperative day 2; and sample 9, postoperative day 6. All blood samples were immediately evaluated in the laboratory and spun at 2000g for 15 minutes; the plasma was separated, frozen at −80°C, and batched. TNF receptor p55, TNF receptor p75, and IL-6 were measured by use of the Cobas Core Automat (Hoffmann La Roche) according to the instructions of the manufacturers. Complement C5a was measured by enzyme immunoassay (Behring). Complement C3d was measured with double-zone rocket immunoelectrophoresis with the use of 1% (wt/vol) agarose (Indubios A37) containing anti-C3c and anti-C3d immunoglobulins (Binding Site) at 4.5 V/m for 18 hours. IL-8 and IL-10 were measured by ELISA.
Data are presented as the arithmetic mean±SD and the arithmetic mean±SEM (Tables); SEM was used for graphs (Figures). Means were compared by Student t test for paired samples and ANOVA. All values were examined by repeated measures; significance was assumed at a value of P<0.05.
Preoperative, perioperative, and postoperative clinical data of the patients are listed in Tables 1 to 3⇓⇓⇓ for each group, respectively. The perioperative and postoperative course and recovery were uneventful for all patients. Complements C5a and C3d were at normal levels before surgery, after the induction of anesthesia, and after thoracotomy for all surgical groups. C5a increased significantly after the onset of CPB to 1.37±0.27 μg/L in group A but was also elevated in groups B and C for sample 5 (reperfusion) to 0.84±0.26 and 0.97±0.29 μg/L, respectively. The values returned to normal levels at 4 hours for groups B and C. In contrast, the elevated levels for group A were sustained at 4 hours and at 24 hours after surgery (Figure 1⇓). The difference between group A and groups B and C was significant (P<0.05). Complement C3d showed a significant increase in group A after the onset of CPB, with a maximum of 8.65±0.96 mg/L at 4 hours after surgery, but it returned to normal levels afterward. Group B also showed a significant increase, with a maximum of 6.15±0.81 mg/L at 4 hours after surgery, whereas group C showed no significant increase in any sample (Figure 2⇓). There was only one statistically significant difference when comparing group A with group C; this difference was for the 4-hour postoperative sample (P<0.05).
TNF receptor p55 (Figure 3⇓) and TNF receptor p75 (Figure 4⇓) showed a significant increase after reperfusion for groups A and B but not for group C. For group B, this sample had a maximum value of 3.61±0.93 ng/L. The maximum value was reached 4 hours after surgery in group A, with 4.44±1.38 ng/L. In contrast, the values in group B remained statistically higher than group A values for the entire period of hospitalization.
Group C showed only a slight increase of TNF receptor p55 at 4 and 24 hours after surgery, with 2.6±0.47 and 2.57±0.31 ng/L, respectively. Significant differences (P<0.05) between the groups could be evaluated at 4 hours (group A versus groups B and C) and at 48 hours (group A versus group B).
A similar response was observed for TNF receptor p75, with increased values for group A starting during reperfusion, with a maximum at 4 hours after surgery and a continuous elevation for the whole postoperative period (Figure 4⇑). Compared with group A, groups B and C showed a statistically significant elevation at the same time period (P<0.05). IL-6 showed the first increase during reperfusion with a strong but short maximum elevation for the 4-hour postoperative sample. This elevation was at similar values for all 3 groups (817.81±187.96, 749.43±173.31, and 773.68± 213.06 pg/mL for groups A, B, and C, respectively). The values returned to almost normal levels after 24 hours in all groups (Figure 5⇓).
A similar reaction could be observed for IL-8 with a short but strong elevation for the 4-hour postoperative sample in all groups, but IL-8 was most elevated in group A (78.06±17.14 pg/mL) versus group B (50.03±15.11 pg/mL) and group C (29.05±7.80 pg/mL). The difference was significant between group A and group C (P<0.05). A statistical difference was found at 24 hours after surgery between group A and groups B and C and at 48 hours between group A and group C (P<0.05) (Figure 6⇓).
For IL-10, a strong increase was observed for group A during reperfusion, for the 4-hour and the 24-hour samples. The value during reperfusion in group A was significantly higher than the values in groups B and C (103.70±45.14 versus 11.48±4.56 versus 17.48±6.79 pg/mL) (P<0.05). After 48 hours, the values for all groups returned to normal (Figure 7⇓).
The immune response associated with cardiac surgery is thought to be dominantly influenced by the use of CPB. Nevertheless, there is little literature available to compare the effect of CPB with the unique immune response due to the surgical trauma alone.3 4 Because different off-pump approaches have been introduced for CABG,5 6 7 there are only a few clinical trials available that compare the immune response in patients undergoing conventional CABG plus CPB with off-pump CABG. Furthermore, the influence of the extent of the approach to the heart on the immune response has not been clarified yet, inasmuch as Chaudhary et al8 could show a significant difference in the release of inflammatory mediators when laparoscopic and traditional cholecystectomy are compared. Apart from a publication by De Paulis et al,9 who compared PTCA, MIDCAB, and conventional single CABG procedures regarding troponin I, creatine kinase, and C-reactive protein, the present study evaluated for the first time the standard sternotomy approach and off-pump surgery with a limited access approach to the heart by focusing on the humoral immune response, such as complement cascade activation by C5a and C3d, the release of TNF receptors p55 and p75, inflammatory interleukins, such as IL-6 and IL-8, and the anti-inflammatory mediator IL-10.
Complement activation of the alternative pathway is reflected by an increase of C5a, which is part of the acute-phase immune response. This first immune response is accompanied by the release of TNF-α IL-1, and IL-6. Our results show that this response is unique for patients in all 3 surgical groups, thus reflecting, in part, the surgical trauma. As demonstrated in the MIDCAB group, the extent of reaction is not related to the size of the incision or splitting the sternum. Nevertheless, when CPB is used, this acute-phase response is significantly augmented throughout a defined perioperative period. Furthermore, the early increase of C5a 10 minutes after the onset of CPB demonstrates a potential triggering mechanism through contact with the artificial surface of the oxygenator tubes and filters. As pointed out by Vertrees et al,10 activation of the complement cascade and the increase of inflammatory immune regulators, such as IL-6, IL-8, and TNF-α, cause a release of neutrophil leukocytes from the bone marrow as part of the acute-phase reaction. Furthermore, the increase of C5a is accompanied by a decrease of monocytes, which was clearly demonstrated by our previously reported results.11 This phenomenon is explained by an increased adhesion of monocytes to the endothelial surface influenced by C5a.12 Burrows et al13 showed an increased release of premature monocytes from the bone marrow as a secondary reaction to the acute-phase response. There is a decrease of activated (HLA-DR–positive) monocytes, whereas the total number of monocytes increased 4 hours after surgery and onward.14 This reaction is pronounced in CPB patients compared with patients subjected to the off-pump technique. However, the immune response to the surgical trauma was demonstrated by the same increase of IL-6 in all surgical groups.
IL-6 plays an important role during the acute-phase response. It stimulates the release of immune-competent proteins from the liverlike C-reactive protein and, together with TNF-α and IL-1, causes activation of T cells. IL-10 plays an anti-inflammatory role by suppressing of T-cell activity, especially the activity of cytotoxic T cells and TH1 cells, and thus leads to a reduction of IL-2 synthesis as described by Markewitz et al.15 The present study clearly demonstrates that the augmented inflammatory response is paralleled by an increase of IL-10 in the CPB group; thus, the inflammatory and anti-inflammatory responses are balanced in a physiological pathway.16 IL-10 was much less increased during surgery without the use of CPB. Apart from the acute-phase response, which is due to the surgical trauma and may be enhanced and triggered by the use of CPB, both myocardial ischemia and cardioplegic arrest could have an additional impact on the inflammatory response.17 18 Karube et al19 described the influence of myocardial ischemia on the release of IL-8 and TNF. In the present study, IL-8 and TNF receptors p55 and p75 showed a substantial and sustained increase that started 4 hours after surgery. This could be related to the reperfusion after general ischemia of the cardioplegic-arrested heart. In a study not yet published, we found that the release of troponin I is highly related to the use of cardioplegic arrest and was not present in the off-pump groups. IL-8 has an important influence on the chemotactile activation of T cells and the endothelial barrier function of the endothelium. Reperfusion injury may play an important role in the immune response, resulting in neutrophil interstitial pulmonary infiltration or capillary leakage as described by Finn et al.20 Nawas et al21 found a correlation between patients with diabetes mellitus and an increased coronary IL-8 level. However, the number of patients in each group with diabetes in the present study was too small to confirm these findings. Cold crystalloid cardioplegia may play an important role in the immune response too, as described by Wan et al22 With this in mind, the significantly enhanced reaction of TNF receptors p55 and p75 would not be related to the use of CPB alone but would rather reflect the reperfusion injury after cardioplegic arrest. This is similar to the immune response after myocardial infarction demonstrated by Cain et al.23
However, some of the significantly augmented immune response during and after CPB may be attenuated by the use of aprotinin, which was routinely applied in group A.24 Because of the complex nature of the immune response, the presented data may be discussed under various perspectives. The present study clearly shows the unique immune response, which is due to the general surgical trauma caused by CABG independent of access and use of CPB. There are some preoperative or perioperative variables that may have different effects on the immune response, such as blood transfusion or the presence of diabetes mellitus. Most of the patients in groups A and B but only a few patients in group C (the MIDCAB group) had concomitant diabetes. The few patients with perioperative blood transfusion did not show a significant different immune response when they were compared with those patients undergoing no transfusion in the same surgical group. The length of CPB may have also an important impact on the immune response, but the one patient in group A who had a CPB time >134 minutes did not show an enhanced immune response compared with those with shorter perfusion times. In contrast to Chaudhary et al,8 who could demonstrate a relationship between the size of the skin incision and the immune response, we were not able to show a statistically significant difference in this regard between the sternotomy and limited thoracotomy approach.
In summary, the influence of CPB and cardioplegic arrest on the immune response is demonstrated by a significantly augmented increase of most humoral parameters of inflammation. The complement activation by C5a and C3d seems to be triggered by the extracorporeal circulation, inasmuch as these were significantly elevated only for a well-defined perioperative period in the CPB group. In addition, this inflammatory response was balanced by a parallel increase (to the same extent) of the anti-inflammatory IL-10 in the CPB group compared with the off-pump group. TNF receptors p55 and p75 and IL-8 may be additionally influenced by cardioplegic cardiac arrest followed by a different degree of reperfusion injury. However, the clinical impact of these findings remains unclear and could not be related to a different outcome or even to an obvious clinical benefit for the patients of one surgical group in the present study. As described in Table 3⇑, postoperative ventilation, a stay in the ICU, and a stay in the hospital were not statistically different among the 3 groups. All patients showed an uneventful postoperative coarse. Thus, more detailed trials will be necessary to demonstrate a relationship between the different immune response and clinical features after CPB versus off-pump CABG.
- Copyright © 2000 by American Heart Association
Kirklin JK, Westaby S, Blackstone EH, et al. Complement and the damaging effects of cardiopulmonary bypass. J Thorac Cardiovasc Surg. 1983;86:845–857.
Utley JR. The immune response and cardiopulmonary bypass. In: Gravlee GP, Davis RF, Utley JR, eds. Principles and Practice of Cardiopulmonary Bypass. Baltimore, Md: Williams & Wilkins; 1993:249–266.
Park SK, Brody JI, Wallace HA, et al. Immunosuppressive effect of surgery. Lancet. 1971;1:53–55.
Howard R. Effect of burn injury, mechanical trauma, and operation on immune defense. Surg Clin North Am. 1980;59:199–207.
Buffolo E, Silva De, Andrade JCS, et al. Coronary artery bypass grafting without cardiopulmonary bypass. Ann Thorac Surg. 1996;61:63–66.
Calafiore AM, Giammarco GD, Teodori G, et al. Left anterior descending coronary artery grafting via left anterior small thoracotomy without cardiopulmonary bypass. Ann Thorac Surg. 1996;61:1658–1665.
Diegeler A, Falk V, Walther T, et al. Minimally invasive coronary artery bypass without extracorporeal circulation. N Engl J Med. 1997;336:1454–1455. Letter.
Chaudhary D, Verma GR, Gupta R, et al. Comparative evaluation of the inflammatory mediators in patients undergoing laparoscopic versus conventional cholecystectomy. Aust N Z J Surg. 1999;69:369–372.
De Paulis R, Colagronde L, Seddio F, et al. Minor myocardial damage and inflammatory response after three different procedures of left anterior descending artery revascularization. G Ital Cardiol.. 1999;29:375–382.
Vertrees R, Tao W, Kramer GC, et al. Tumor necrosis factor monoclonal antibody prevents alteration in leucocyte population during cardiopulmonary bypass. ASAIO J. 1994;40:M554–M559.
Diegeler A, Tarnok A, Rauch T, et al. Changes of leukocyte subsets in coronary artery bypass surgery using cardiopulmonary bypass versus ‘off-pump’ techniques. Thorac Cardiovasc Surg. 1998;46:327–332.
Moat NE, Shore DF, Evans TW. Organ dysfunction and cardiopulmonary bypass: the role of complement and complement regulatory proteins. Eur J Cardiothorac Surg. 1993;7:563–573.
Burrows FA, Steele RW, Marmer DJ, et al. Influence of operations with cardiopulmonary bypass on polymorphonuclear leukocyte function in infants. J Thorac Cardiovasc Surg. 1987;93:253–260.
Rinder CS, Mathew JP, Rinder H, et al. Lymphocyte and monocyte subset changes during cardiopulmonary bypass: effects of aging and gender. J Lab Clin Med. 1997;129:592–602.
Markewitz A, Faist E, Lang S, et al. An imbalance in T-helper cell subsets alters immune response after cardiac surgery. Eur J Cardiothorac Surg. 1996;10:61–67.
Mosmann TR. Interleukin 10. In: Thomson A, ed. The Cytokine Handbook. London, UK: Academic Press; 1994:231–232.
Mullane K. Neutrophil and endothelial changes in reperfusion injury. Trends Cardiovasc Med. 1991;1:282–289.
Butler J, Rocker GM, Westaby S. Inflammatory response to cardioplegic solutions: evaluation by tissue cultures. Ann Thorac Surg.. 1993;55:552–559.
Karube N, Adachi R, Ichikawa Y, et al. Measurement of cytokine levels by coronary sinus blood sampling during cardiac surgery with cardiopulmonary bypass. ASAIO J. 1996;42:M787–M791.
Finn A, Morgan BP, Rebuck N, et al. Effects of inhibition of complement activation using recombinant soluble complement receptor 1 on neutrophil CD11B/CD18 and L-selectin expression and release of interleukin-8 and elastase in simulated cardiopulmonary bypass. J Thorac Cardiovasc Surg. 1996;111:451–459.
Nawas SI, Doherty JC, Vigneswaran WT, et al. Cardiopulmonary bypass increases coronary IL8 in diabetic patients without evidence of reperfusion injury. J Surg Res. 1999;84:46–50.
Wan S, Izzat MM, Lee TW, et al. Avoiding cardiopulmonary bypass in multivessel CABG reduces cytokine response and myocardial injury. Ann Thorac Surg. 1999;68:52–56.
Cain BS, Harken AH, Meldrum DR. Therapeutic strategies to reduce TNF-alpha mediated cardiac contractile depression following ischemia and reperfusion. J Mol Cell Cardiol. 1999;31:931–947.
Peters DC, Noble S. Aprotinin. an update of its pharmacology and therapeutic use in open heart surgery and coronary artery bypass surgery. Drugs. 1999;57:233–240.