TNF-a plays an important role for leukocyte-endothelium interaction and the release of oxygen radicals from adherent polymorphonuclear neutrophils (PMNs) [20]. Reports on the TNF-a response after CPB are conflicting. A significant increase in TNF-a was demonstrated after release of the aortic cross clamp and termination of CPB in some studies [21], while others reported no detectable TNF-a [22]. In a few studies, TNF-a was detectable preoperatively but without further changes in the TNF-a plasma concentration during or after CPB. A similar inconsistency in TNF-a response was found in pigs subjected to CPB [23]. Animal studies have further revealed that administration of small amounts of TNF-a decrease myocardial performance, suggesting that even small TNF-a levels may also contribute to myocardial dysfunction in humans.

IL-6 is a good predictor of clinical outcome [24] and is thought to be related to the extent of tissue injury. Normally, IL-6 is undetectable in peripheral blood in healthy individuals, but has been found preoperatively in some children with congenital heart disease [25]. In general, IL-6 plasma levels increase during and after surgery and CPB and are related to the duration of aortic clamp time [26]. It is still not clear whether it is the CPB procedure, the surgery, or the combination that elicits the IL-6 response as no differences were found between pediatric patients subjected to surgery with or without CPB [27].

In children subjected to cardiac surgery during CPB, an early IL-8 response has been detected with increasing ­concentration in the following hours. The plasma levels were related to the duration of ischemia/reperfusion and total bypass time [21].

The proinflammatory cytokine response in children undergoing cardiac surgery with CPB shows a large variation in the release pattern and plasma concentrations. This is in contrast to the proinflammatory response in adults, which is well defined and temporary. It is still unknown whether the preoperative high cytokine levels in children are of clinical importance, but neonates with clinical signs of capillary leak syndrome after heart surgery showed signs of a preoperative inflammatory state measured as increased plasma levels of elastase and the complement split products C3d/ C3 [28].

IL-10 and IL-1 receptor antagonists (IL-1ra) seem to be the most consistent antiinflammatory cytokines released during and after cardiac surgery with CPB, both in adults and children [28]. The antiinflammatory response is not caused by blood contact with the circuit because none of these cytokines were detectable when blood was drawn through an isolated pediatric CPB circuit [29]. However, when the CPB is connected to an animal, the stress from the CPB procedure itself elicits an IL-10 response when compared to sham-­operated animals [23].

C-reactive protein (CRP) is an acute-phase protein with antiinflammatory properties as it downregulates PMN chemotaxis. Clinically, CRP is often used as an unspecific marker for infections, and its plasma concentration correlates to the extent of the surgical trauma. High CRP levels are unequivocally found in children after heart surgery during CPB with a peak on the first to third postoperative day, and the response is delayed compared with the IL-6 release [30]. It has been suggested that the stress of sternotomy is sufficient to elicit a CRP response as piglets subjected to anesthesia and sternotomy show the same CRP response as piglets exposed to the CPB procedure [23].

In conclusion, the antiinflammatory response in pediatric cardiac patients shows a more clear release pattern compared to the proinflammatory counterpart. Evidence has emerged that the balance between the pro- and antiinflammatory mediators is most important for the outcome of patients experiencing a systemic inflammatory response [19]. It is therefore irrelevant to follow changes in a single cytokine concentration in the patient at present. In general, the cytokine response is characterized by great variability and interindividual differences. In uncomplicated cases, the systemic response is temporary, balanced, and of a few days’ duration in both pediatric and adult patients [31].

Although steroids have been used for years to attenuate post-bypass inflammation, data to support this derive almost entirely from trials in adults with coronary artery disease. Even in adults, steroid use for cardiac surgery is controversial, and data in children are inconclusive.

The actions of steroids are protean with both immediate and delayed effects. The limitation of inflammatory capillary permeability by either by diminishing recruitment of activated white blood cells to vascular beds or inhibiting prostacyclin production and induction of nitric oxide synthase has been shown [32, 33]. Also, steroids appear to increase antiinflammatory and decrease proinflammatory cytokine levels [34]. Corticosteroids upregulate the production of b-adrenergic receptors and decrease reuptake, thus increasing the availability of these receptors to respond to endogenous or exogenous catecholamines. The availability of cytosolic calcium in myocardial and vascular smooth muscle cells, augmenting contractility, may be increased by physiologic doses of corticosteroids.

Methylprednisolone is able to reliably (and beneficially) alter the balance of proinflammatory and antiinflammatory mediators in the blood of patients subjected to CPB, indicating that the drug decreases the SIRS associated with CPB. Specific hemodynamic benefits (increased CI, decreased SVR) seem to be associated with use of the drug in this setting, yet these alterations may increase the need for postoperative IV hemodynamic agents (vasoconstrictors, etc.). Pulmonary compliance seems not to be influenced by steroids. Increased P(A-a)O₂ (perhaps by increasing pulmonary shunt) and early postoperative tracheal extubation may be hindered for undetermined reasons. Several studies have used different doses of MP and dexamethasone. There is no general agreement on the type of preparation and dose of the steroid. Most of the studies are concerned with adult patients. Varan et al. [35] demonstrated that the increase in cytokine levels and CRP was not significantly different between high and low doses. Niazi et al. [36] demonstrated an increase in the cardiac index during and after CPB in patients undergoing coronary artery bypass surgery that received 30 mg/kg of MP. In this study, after an initial increase in the cardiac index, it gradually decreased in the post-bypass period. Jansen et al. [37] reported normothermia, higher blood pressure levels without supportive treatment, and a shorter stay in the ­intensive care unit for dexamethasone-treated patients (receiving 1 mg/kg) compared to nontreated patients. Another study showed that two doses of 30 mg/kg MP before CPB and before declamping of the aorta suppressed the production of IL-8 and 6 [38]. The authors also found a higher postsurgical cardiac index in the MP-treated group. In the Varan study, the increase in IL-6 and 8 levels after surgery was not significantly different in low- and high-dose MP groups. The cumulative inotropic support required, the duration of mechanical ventilation, the stay in intensive care unit, urine volume, blood loss, and peak core temperature were not significantly different between the two groups. Several studies have demonstrated suppression of levels of interleukins. Jorens et al. [39] showed that MP pretreatment of 30 mg/kg failed to prevent IL-8-mediated pulmonary neutrophil infiltration after CPB, although an increase in serum IL-8 levels was less pronounced in MP-treated than in nonsteroid-treated patients. Butler et al. [40] investigated the levels of cytokines during cardiopulmonary bypass and the effects of intraoperative MP at a dose of 10 mg/kg versus no steroids in the pediatric age group. Clinical and hemodynamic consequences were not mentioned. The IL-6 level was elevated above baseline, peaking earlier in the nonsteroid group. Both IL-6 and CRP levels at 24 h postoperatively were higher in the nonsteroid group.

Cardiopulmonary bypass has also been shown to play a role in the development of pulmonary dysfunction after open-heart surgery. Increased protein leakage as early as 10 min after the onset of CPB in patients with clinical signs of capillary leak syndrome has also been demonstrated by Seghaye et al. [41]. Corticosteroids in several doses have been used with the hope of preventing pulmonary dysfunction after CPB surgery. Varan et al. [35] did not observe any difference in oxygenation parameters for any patients in the two groups, indicated by similar ratios of PaO₂/FiO₂ in the early postoperative period.

They concluded that CPB surgery initiates an SIR and high-dose MP is not superior to low-dose MP in suppressing this reaction. In this study, a low-dose treatment with MP (2 mg/kg) is preferable to a high-dose treatment considering the possible side effects, although none were observed. Children who received steroids prior to CPB had fewer febrile episodes, improved respiratory gas exchange, and better renal function, and they required less supplemental fluid postoperatively than did controls. As a result, the group given dexamethasone required fewer days of mechanical ventilatory support and was discharged from the ICU sooner.

Gessler et al. [42] evaluated the effects of steroids on SIRS by measuring, among other parameters, IL-8 and the total neutrophil count (TNC), as well as some clinical parameters. In this study, administration of steroids did not show a significant impact on the clinical outcome and the degree of the inflammatory response following cardiac surgery. They concluded that the lack of suppression of the inflammatory reaction may have been due to the dose and timing of steroid administration, the lower inflammatory reaction in patients with shorter time on CPB and less severe operative trauma, or an age older than 3 months at the time of surgery.

Although not measuring the amount of SIRS, some publications have shown beneficial effects of steroid administration by measuring other parameters. Checchia et al. [43] showed in 2003 that steroids reduced the postoperative troponin levels in a pediatric population, indicating better myocardial preservation. The Group from Toronto has shown clinically improved outcomes after the use of steroids in a study of high-risk pediatric cardiac surgery [44].

Additional well-designed (prospective, randomized, double-blind, placebo-controlled) clinical investigations (with large numbers of patients and tightly controlled perioperative management) involving corticosteroids and patients undergoing cardiac surgery with CPB need to be done. Whether or not suppression of the SIRS associated with CPB with corticosteroids (or any other drug/technique) is clinically desirable and beneficial remains to be determined.

Platelets are known to be activated during CPB because of contact with the foreign surfaces of the extra-corporeal circuit and also because of numerous other factors, such as hypothermia, shear forces, use of exogenous drugs, and release of endogenous chemicals [45]. Platelets are also activated by surgical trauma where the surgical incision, via activation of tissue factor and subsequently factor VII and factor X, may at least partly explain this process [46]. Activated platelets express reorganized surface molecules, such as glycoprotein IIb/IIIa, which forms a fibrinogen-binding complex. Simultaneously, there is a movement of cytoplasmic granules toward the cell surface. These granules fuse with the cell membrane and extrude their contents. This is the secretion method of platelet factor-4, β-TG, and von Willebrand factor. At the same time, those molecules that previously were an integral part of the granule membrane, such as P-selectin, now will be expressed on the surface of the activated platelets [45]. Extracorporeal circuits are manufactured from synthetic materials, and there is a material-derived platelet activation dependent on glycoprotein IIb/IIIa receptors [47]. P-selectin is upregulated by several mechanisms. One is through thrombin, which is redundant because of formation during CPB [48]; another is through newly generated cytokines that stimulate platelets [49]. Both the prothrombotic and the proinflammatory mechanisms occurring during CPB might be attributed to the release of soluble CD 40 ligand (sCD40L) by platelets [50]. Platelets activated during CPB form conjugates both between themselves and with leukocytes. P-selectin is expressed by activated platelets, which contribute to leukocyte conjugate formation by binding P-selectin glycoprotein. Activated platelets use this adhesion pathway to stimulate conjoined monocytes, thus leading to secretion of the proinflammatory cytokines IL-1β, IL-8, and MCP-1. P-selectin also induces tissue factor expression and fibrin deposition by monocytes, thus contributing to the evolution of thrombus. Endothelial cells express the adhesion molecule CD40 and activated platelets express the complementary CD40 ligand on their surface. CD40 ligand is structurally related to TNF-α and induces endothelium to secrete chemokines and express further adhesion molecules. Substantial secretion of IL-8 and MCP-1 was noted on platelets binding to endothelium [51].

Assessment of platelet activation can be done my means of flow cytometry and the analysis of granule proteins (such as β-TG) after degranulation. Thrombospondin, an extracellular matrix protein, has also been investigated, but β-TG seems superior to thrombospondin as a marker for platelet activation in vivo [52].

Along with CPB came hypothermia, representing another milestone in the history of open-heart surgery. Hypothermia was used as a tool for lowering the metabolic needs of selected regional beds and/or the whole body. Although hypothermia was introduced in the 1950s by Dr. Wilfred Gordon Bigelow, allowing open-heart operations to be performed in a bloodless field after interruption of blood flow, it was not until the 1960s that hypothermia begun to be used in conjunction with CPB, ushering in the era of modern-day open-heart surgery.

There are a number of proposed clinical applications of hypothermia, including traumatic brain injury. During cardiac surgery with cardiopulmonary bypass (CPB), deep hypothermia is used to protect immature organs from ischemia when surgery requires complete arrest of the systemic circulation. CPB itself, because of the contact of blood with the large non-endothelial surfaces and air/blood interface, triggers the whole body inflammatory response. This often leads to capillary leakage, edema, organ dysfunction, and SIRS. This might explain why in this condition prolonged inflammation together with an acute systemic inflammatory response is one of the major correlates with the adverse clinical outcomes associated with CPB and a major cause of postoperative morbidity [31, 53, 54]. Hypothermia has been suggested to play a protective role in reducing the acute inflammation, but there are scant data demonstrating a beneficial effect of hypothermic CPB [55]. Data focusing on cellular and molecular effects of hypothermia are limited.

The inflammatory response that accompanies open-heart surgery is multifactorial, in both adults and children [56]. The mismatches between the foreign surfaces in the heart-lung machine and blood vessel surfaces are extreme in children. Therefore, specially designed heart-lung machines with small priming volumes and surfaces have been developed [57]. The surgical trauma, anesthesia, and deviation from normal organ perfusion are other important factors causing inflammatory activation [58]. It has also been suggested that the degree of hypothermia may influence the inflammatory response during CPB. Deep hypothermia and circulatory arrest have been shown to trigger less-pronounced inflammatory response than low-flow CPB in newborns, as assessed by measurements of IL-6, IL-8, and C3a [59]. This phenomenon may be the result of a shorter CPB time, but a protective effect of hypothermia per se may also be part of the explanation [60, 61]. However, in a randomized study [62], designed to further elucidate this issue, researchers did not find any significantly attenuated inflammatory response in the moderate hypothermia group. In fact, this group showed an enhanced IL-8 response and an attenuated IL-10 response during CPB, potentially suggesting an inflammatory net effect.

Diestel et al. [61] used endothelial cells as they play a pivotal role during activation of the inflammatory cascade by expression of cytokines. A special cell culture model was used to exclusively study the effects of hypothermia and drug treatment on the inflammatory response of endothelial cells. Using time-temperature settings analogous to clinical application during CPB in children, they measured the concentrations of IL-6, IL-8, and MCP-1. They concluded that hypothermia upregulated IL-6 and TNF-a and that MP-pretreatment attenuated this response.

Qing and colleagues [63] investigated the cell mechanisms by which hypothermia could ameliorate inflammation using a pig model of CPB and sham cardiac operation. In half of the animals (n  =  6), the core temperature was maintained at 28 °C during CPB and in the other half at 37 °C. Measurements made in liver samples obtained before CPB and 6 h after corroborated previous findings indicating that hypothermia increases intrahepatic concentrations of IL-10 while decreasing TNF-a. This effect was associated with less hepatic cell necrosis but without effects on apoptosis. They further documented increased activation of the signal transducer and activator of the transcription (STAT)-3 pathway and increased expression of the suppressor of cytokine signaling (SOCS)-3. These two signaling events are important because they have been associated with antiinflammatory effects in settings other than hypothermia [64]. Activation of STAT-3 is known to increase IL-10, and both IL-10 and activated STAT-3 can increase expression of SOCS-3 with subsequent decreases in TNF-a. These data demonstrate that moderate hypothermia during CPB is associated with activation of the Jak/STAT pathway, leading to the expression of IL-10, which in turn upregulates SOCS-3 and finally attenuates TNF-a production. This antiinflammatory shift in the cytokine balance is associated with liver protection.

Hypothermia has become the subject of scientific and clinical interest as growing evidence points to therapeutic effects not only during but also after an ischemic event. A widely publicized study has shown improved survival and neurologic recovery when hypothermia is instituted in patients who remain comatose after resuscitation from cardiac arrest [65]. Hypothermia also has been shown to be effective in selected subsets of patients presenting with stroke [66], head trauma [67], and myocardial infarction [68]. Thus, there is considerable interest in understanding the mechanisms by which hypothermia works and especially to determine whether benefits may involve mechanisms other than reduction in metabolic activity. Recent studies have in fact shown that hypothermia can activate cell- protecting pathways. For example, hypothermia can induce expression of heat shock protein 70 [69] and cause a shift in the inflammatory cascade during CPB, reducing proinflammatory mediators and increasing the antiinflammatory cytokine IL-10 [70].

A recent study by Stocker et al. [71] has addressed an important and previously unresolved clinical question regarding the most appropriate temperature during pediatric CPB. This study has shown that moderate hypothermia at 24 °C does not offer any advantages over mild hypothermia at 34 °C during pediatric CPB for repair of congenital heart disease in terms of the postoperative clinical course and severity of SIRS or markers of organ injury. Moreover, there was a tendency toward a shorter duration of mechanical ventilation with mild hypothermia. The depth of hypothermia during CPB did not influence the risk of postoperative bleeding, blood product transfusion requirements, or infection. As expected, there was a trend toward a shorter duration of CPB in the mildly hypothermic group, although this was related to the shorter duration of the rewarming period.

When considering the biochemical and cellular manifestations of inflammation, CPB resulted in a marked acute phase reaction in all children, but this was not influenced by temperature. In keeping with previous observations, CPB also resulted in activation of the receptor pathways and in deactivation of circulating monocytes. No influence by the temperature during CPB was noted. Of particular interest is the fact that TNF-a was not influenced by CPB in either group. In this study, no difference in any clinical or biochemical markers of end-organ injury between the two study groups was found. As would be expected, there was postoperative deterioration in lung and renal function, but again, the CPB temperature did not influence this. A transient microalbuminuria early after bypass indicated a significant capillary leak, although independent of bypass temperature.

Chemokines such as IL-8 and MCP-1 are, respectively, potent activators of neutrophils and monocytes [72]. Activation of neutrophils results in degranulation with increased release of myeloperoxidase (MPO) that together with increased production of reactive oxygen species may lead to tissue damage and are important contributors to inflammation. A long CPB time has previously been associated with increased levels of IL-8 and MPO, potentially reflecting enhanced activation of neutrophils [70]. The results in this study further support such a notion by showing raised IL-8 and MPO levels accompanied by increased leukocyte counts in patients with long CPB times. A similar pattern was observed in patients with long aortic cross-clamp times. Both long CPB time and long aortic cross-clamp time may induce enhanced oxidative stress, possibly contributing to the enhanced IL-8 and MCP-1. Eggum et al. [62] showed in a randomized study that duration of CPB will be longer in those with moderate compared with those with mild hypothermia in that it takes longer to cool and rewarm patients from 25 °C, rendering it difficult to evaluate the relative importance of each factor. Nevertheless, although those with moderate hypothermia showed some trends for a higher degree of inflammation than those with mild hypothermia during CPB, the differences were rather modest. In this study, the aortic cross-clamp time and time on CBP were associated with increased chemokine levels and leukocyte activation, underscoring that these procedures should be as short as possible to avoid an excessive inflammatory response and possible adverse clinical effects.