Narrative Review of Systemic Inflammatory Response Mechanisms in Cardiac Surgery and Immunomodulatory Role of Anesthetic Agents

ABSTRACT

Although surgical techniques and perioperative care have made significant advances, perioperative mortality in cardiac surgery remains relatively high. Single- or multiple-organ failure remains the leading cause of postoperative mortality. Systemic inflammatory response syndrome (SIRS) is a common trigger for organ injury or dysfunction in surgical patients. Cardiac surgery involves major surgical dissection, the use of cardiopulmonary bypass (CPB), and frequent blood transfusions. Ischemia-reperfusion injury and contact activation from CPB are among the major triggers for SIRS. Blood transfusion can also induce proinflammatory responses. Here, we review the immunological mechanisms of organ injury and the role of anesthetic regimens in cardiac surgery.

INTRODUCTION

A subset of complex surgeries, such as cardiac surgery, organ transplantation, brain surgery, and abdominal surgery (gastric, pancreatic, spleen, liver), are particularly associated with relatively high incidences of postoperative mortality.[1]

Cardiac surgery is typically a long and complex surgery, involving extensive dissection, ischemic-reperfusion injury, and frequent blood transfusion. Additionally, it is associated with a high risk of developing postoperative dysfunctions of the cardiovascular, respiratory, renal, and central nervous systems.[2-4] The overall operative mortality rate for cardiac surgery was 2.9% in adults[5]; the number ranged from 1.2% for simple procedures to 9.9% for more complicated ones.[3] The overall mortality rate was notably higher in the pediatric population, ranging from 3.4% to 6.9%.[6,7] The mortality rates could be as high as 15% following complex congenital heart surgery.[8] Although surgical techniques, myocardial protection, and perioperative care have improved significantly, the perioperative mortality for complex cardiac surgery remains high, and the incidence of postoperative morbidity is significant; an improved understanding of the possible mechanisms may further improve the outcome.

Systemic inflammatory response syndrome (SIRS) is a common trigger for organ dysfunction/failure in surgical patients.[9] Thus, understanding how SIRS is triggered and affects organ function is essential to seek a potential intervention. Ischemia-reperfusion and contact activation, revolving around the use of cardiopulmonary bypass (CPB), are among the major triggers for SIRS in cardiac surgery. Blood transfusion can also induce proinflammatory responses.[10] Here, we review SIRS, organ injury, and the role of anesthetics in cardiac surgery, focusing on cardiac surgery involving CPB.

METHODS

Database search

We searched electronic databases (PubMed) for this narrative review from January 1, 2000, to September 1, 2022, but we did not exclude commonly referenced articles published before 2000. We used “systemic inflammatory response,” “cardiopulmonary bypass,” and “cardiac surgery” as required primary keywords for the search. We retrieved 813 publications following the initial search. To those keywords, we added either “immunomodulation,” “damage-associated molecular pattern,” “ischemia-reperfusion injury,” “organ injury,” and/or “anesthetics” to further narrow down the literature of choice. We have reviewed basic science and animal model studies, clinical trials, meta-analyses, and systematic and narrative reviews. Priority review was given to studies with more rigorous study designs and those published in higher-quality journals. We excluded case reports from this review. Abstracts from the search were scrutinized and determined for inclusion in the review. Additional searches using more specific terms, such as inflammatory cytokines, immune cells, proteins, and anesthetic agents, were also used to obtain more specific information. Although we tried to include as much information as possible, the fact that this review was a narrative review can be considered a limitation.

IMMUNOLOGICAL RESPONSES TO SURGERY

Surgery-mediated inflammation

A surgical incision and dissection trigger both local and central immunological responses. Surgical insults cause direct mechanical injury to cells/tissues leading to cellular disruption and breakdown. These injured cells/tissues release damage-associated molecular pattern molecules (DAMPs) or alarmins, which interact with immune cells and induce inflammation.[11,12] DAMPs are endogenous nuclear, mitochondrial, or cytosolic molecules that are regarded as endogenous danger signals to the host when cells experience stress/injury and release them extracellularly. The role of DAMPs is to alert the host of ongoing tissue damage and initiate the process of tissue repair [Figure 1].[13] Toll-like receptors (TLRs) are one of the major pattern-recognition receptors (PRRs) that recognize DAMPs. Almost all classes of DAMPs can bind to TLRs.[14] A large number of DAMPs bind to TLR2, TLR4, and TLR9. TLR2 is mainly expressed on myeloid cells and endothelial cells,[15,16] with a low level of TLR2 expressed on B cells. TLR4 is expressed in myeloid cells, endothelial cells, and epithelial cells.[15,16] TLR9 is expressed on myeloid cells, B cells, endothelial cells, neurons, and cardiomyocytes.[15-17] DAMPs stimulate these PRRs on immune cells to produce proinflammatory cytokines such as tumor necrosis factor (TNF)-α, interleukin (IL)-1β, IL-6, and IL-8.[18-20] Together with DAMPs, these cytokines activate and recruit neutrophils and monocytes to the sites of tissue injury.[21,22] DAMPs are largely classified into DAMPs from the extracellular matrix (ECM) and intracellular DAMPs [Figure 2].[14] ECM-derived DAMPs are produced by proteolytic cleavage and the release of ECM components upon tissue injury. Proteoglycans (PGs) are ECM-derived DAMPs and include biglycan, hyaluronan, heparin sulfate, and fibrinogen.[23] They mostly bind to TLR2 and TLR4. Cells dying from necrosis or apoptosis release endogenous, intracellular molecules. Intracellular DAMPs are classified into mitochondrial DAMPs, nuclear DAMPs, and cytosolic DAMPs.[14] Mitochondria are the major organelles that release DAMPs at the time of cell death, which include mitochondrial DNA (mtDNA), formyl peptides, and adenosine triphosphate (ATP). mtDNA binds to TLR9, formyl peptides bind to formyl peptide receptor 1 (FPR1), and ATP binds to P2X7. In the animal model, blood mtDNA is significantly elevated immediately after sternotomy, which activates TLR9 that further induces the IL-6 release.[24] Nuclear DAMPs include high-mobility group protein B1 (HMGB1) and histones, both of which bind to TLR2 and TLR4. Cytosolic DAMPs include heat shock proteins (HSPs) and S100 proteins, which also bind to TLR2 and TLR4. Usually, DAMPs remain localized, but they can be detectable in the bloodstream following a significant injury.[25]

Figure 1.

Surgery-mediated inflammation. ACTH, adrenocorticotropic hormone; DAMPs, damaged-associated molecular patterns; HPA, hypothalamic-pituitary-adrenal; IL, interleukin; PRR, pattern recognition receptor; SAM, sympathetic-adrenal-medullary; TNF, tumor necrosis factor (the original figure was created with BioRender.com)

Figure 2.

Damaged-associated molecular patterns (DAMPs) released from damaged cells or tissue injury. ATP, adenosine triphosphate; DAMP, damaged-associated molecular pattern; ECM, extracellular matrix; HMGB1, high mobility group box 1; HSP, heat shock protein; mtDNA, mitochondrial DNA (the original figure was created with BioRender.com)

In addition to the local responses, surgical insult stimulates the hypothalamic-pituitary-adrenal (HPA) axis and the sympathetic-adrenal-medullary (SAM) axis[26] [Figure 1] leading to the systemic secretion of cortisol and catecholamines.[27] Proinflammatory cytokines and physiological stress from tissue injury trigger the brain stem nuclei to deliver the information to the paraventricular nucleus of the hypothalamus where the corticotrophin-releasing hormone (CRH) is then released and further stimulates the corticotropes in the anterior pituitary gland to release the adrenocorticotropic hormone (ACTH). ACTH primarily acts on the adrenal cortex to synthesize glucocorticoids, the most important of which is cortisol, which is subsequently released into systemic circulation.[26] Glucocorticoid receptors are expressed in neutrophils, monocytes, macrophages, T cells, and B cells, and cortisol shifts them to the cells with anti-inflammatory phenotype.[28] Additionally, the sympathetic autonomic nervous system, which is activated by the hypothalamus, also increases the catecholamine production from the adrenal medulla.[27] Catecholamine receptors are found in monocytes, macrophages, natural killer cells,[29] B cells, and T cells, and their stimulation induces anti-inflammatory responses.[30] Anti-inflammatory responses are induced most potently by epinephrine, followed by norepinephrine, and least by cortisol.[31] Anti-inflammatory cytokines response includes the production of anti-inflammatory cytokines such as IL-10 and transforming growth factor (TGF)-β, which then induce a generation of regulatory T cells (Tregs), a subset of cluster of differentiation (CD)4⁺ T cells that have suppressive activity.[32] Other CD4⁺ T subsets include T-helper (Th) 1 cells and Th2 cells. These Tregs also bias CD4⁺ T cells toward anti-inflammatory Th2 cells.[33] In addition, HSPs, chaperone proteins released under stress, amplify regulatory T-cell function.[34,35] Thus, surgically injured tissues demonstrate proinflammatory responses, while leukocytes in the bloodstream are anti-inflammatory and hyporeactive.[36] Adaptation to surgical stress involves coordinating local inflammation with systemic anti-inflammation to allow the concentration of activated phagocytes and other effectors only at the injured local site.[37]

CPB-mediated inflammation

CPB is an integral part of cardiac surgery. Dr. John H Gibbon performed the first successful cardiac surgery using CPB on May 6, 1953.[38] CPB has been undergoing continuous modification since then to mitigate CPB-related complications, but CPB is an inherently unnatural process. CPB evokes significant inflammatory processes primarily via the three mechanisms: 1) Contact activation, 2) Ischemia-reperfusion injury, and 3) Endotoxemia [Figure 3].

Figure 3.

Cardiopulmonary bypass (CPB) mediated inflammation. ATP, adenosine triphosphate; NADPH, nicotinamide adenine dinucleotide phosphate; NOS, nitric oxide synthase; NOX, NADPH oxidase; ROS, reactive oxygen species (the original figure was created with BioRender.com)

Contact activation ensues as a result of blood coming into contact with non-endothelial surfaces. Thus, contact activation-mediated inflammation starts soon after the initiation of CPB. A number of mediators are activated in the process, including complements and intrinsic/extrinsic coagulation factors.[39,40] The details on the mechanism of contact activation by CPB can be sought in other reviews.[41] These various mediators contribute to the development of proinflammatory responses in the systemic circulation, which is opposed to the anti-inflammatory direction induced by the HPA and SAM axis activation. Systemic proinflammatory responses activate a number of cells, including neutrophils, endothelial cells, and platelets in the circulation as a result.

During CPB, the utilization of aortic cross-clamp results in a period of cardiac ischemia. The use of circulatory arrest can also cause hypoperfusion to other peripheral organs. Ischemia results in tissue hypoxia.[42] The reperfusion of the previously ischemic tissues with oxygen induces the production of reactive oxygen species (ROS) and necrotic cell death, which is considered a major pathology of ischemia-reperfusion injury.[43] ROS is produced by several mechanisms. During tissue ischemia, ATP is converted to hypoxanthine. When oxygen is resupplied to tissues containing xanthine oxidase, hypoxanthine reacts with oxygen for ROS production. Xanthine oxidase is highly expressed in the intestine and the liver but not in the heart.[44] NADPH oxidase (NOX) is also implicated in ROS production. Ischemic tissues enhance the expression of NOX.[45] The reintroduction of oxygen contributes to ROS production via NOX. The previously ischemic tissues release neutrophil chemoattractants and increase adhesion molecule expression on the vascular bed.[46] Recruited neutrophils play a major role in the degree of ischemia-reperfusion injury as the depletion of neutrophils or neutropenia significantly attenuated ischemia-reperfusion injury.[47] Although NOX is expressed ubiquitously, the level of NOX expression is lower in non-phagocytic cells compared to phagocytes such as neutrophils. Thus, it is intuitive that neutrophils serve as a source of ROS in postischemic tissues. Nitric oxide (NO), on the other hand, is protective in ischemia-reperfusion injury via its anti-oxidant effect (neutralization of ROS) and anti-inflammatory effect (inhibition of neutrophil adhesion/emigration).[48] Arginine is necessary for the synthesis of NO by nitric oxide synthase.[49] After ischemia-reperfusion, arginase activity significantly increases and consumes arginine. This limits the availability of NO at the ischemic site, contributing to worsening ischemia-reperfusion injury.

Endotoxin is also called a lipopolysaccharide and a component of gram-negative bacteria. Endotoxemia in patients undergoing cardiac surgery involving CPB has been widely recognized.[50,51] Although the source of endotoxin has been speculated, including gut translocation, the pathogenesis is not fully elucidated. In addition, its level is not consistent during cardiac surgery in the literature.[52] Recent studies also reported that endotoxin levels were found less frequently than previously reported.[53]

Injured tissues release DAMPs when tissues and cells undergo significant stress and death, which further increases tissue damage.[54] So far, the study of DAMPs in the perioperative setting is rather limited. Despite that a number of DAMPs exist, mtDNA and HMGB1 are the main DAMPs that have been studied in surgical settings until the present. Blood mtDNA levels were elevated soon after CPB.[55,56] HMGB1 was detected after unclamping the aortic cross-clamp.[57] Certainly, the data is in line with the fact that cardiac surgery is a high risk of developing organ injury and mortality. Blood transfusion is often indicated in cardiac surgery. A blood transfusion may also contribute to DAMPs load. A high amount of mtDNA was detected in fresh frozen plasma (FFP) and platelets.[58] Much less mtDNA was detected in packed RBCs (pRBCs), probably because pRBCs were leukocyte-reduced, and mammalian RBCs lack mitochondria. An additional source of DAMP production is the use of cardiotomy suction.[59] S100B level increases with cardiotomy suction use. It is important to delineate what contributes to the production of DAMPs in cardiac surgery for a potential modification and improvement of our clinical care.

Several therapeutic approaches have been introduced to mitigate the inflammatory responses, such as ultrafiltration, miniaturized extracorporeal circulation (MECC), hemoadsorption device, and heparin-coated CPB. Ultrafiltration is beneficial in removing excess water and low-molecular-weight substances, and reduction of blood transfusion[60]; however, its role in the reduction of inflammatory mediators was inconsistent.[61,62] The positive postoperative clinical outcomes were not apparent.[63,64] MECC typically has a shorter tubing system that aims to reduce blood contact surfaces, and in turn, decreases inflammatory reaction. The levels of TNF-α, monocyte chemotactic protein-1, and ROS were lower in the MECC group compared with standard extracorporeal circulation.[65,66] The utilization of MECC also resulted in lower proinflammatory cytokine production than that of off-pump coronary bypass graft (OPCABG) surgery.[67] However, the improvement in clinical outcomes has not been elucidated according to the recent literature.[66,67] Hemoadsorption is proposed to eliminate inflammatory cytokines using biocompatible, highly porous polymer cartridges. A recent meta-analysis reported a significant reduction in 30-day mortality and intensive care unit stay in patients undergoing non-elective cardiac surgery using hemadsorption,[68] although the use of adsorbent did not show a significant reduction in proinflammatory cytokines.[68-70] Use of a heparin-coated CPB circuit was associated with promising clinical outcomes,[71] its role in immunomodulation was demonstrated only in a small clinical study.[72] Nevertheless, factors other than inflammatory responses, for example, surgical techniques, can play a significant role in morbidity and mortality, so it will be critical to have well-matched studies to fully elucidate the role of these interventions.

Neutrophils and organ injury

Overall, a number of mechanisms exist to facilitate proinflammatory responses in patients undergoing cardiac surgery on CPB, particularly at the systemic level. Among various components involved in the responses, neutrophils are considered to play a major role in organ injury. Neutrophils are activated by various mechanisms such as complements, cytokines, and coagulation factors at local tissues and at a systemic level during cardiac surgery. In addition, DAMPs act on neutrophils. DAMPs induce neutrophil extracellular traps (NETs), leading to tissue injury and organ dysfunction.[73,74] Thus, the crosstalk between neutrophils and NETs is one of the major underlying mechanisms of organ injury. In addition to neutrophils, DAMPs can certainly contribute to organ dysfunction via another pathway. HMGB1 inhibits protein C and upregulates tissue factor (TF) expression.[74] These events induce micro-thrombosis and tissue injuries, which subsequently release de novo DAMPs to aggravate local tissue damage.

Anesthetics-mediated immunomodulation

Perioperative anesthetics are shown to be immunomodulatory.[75] The effect of volatile anesthetics on neutrophils has been well described in vitro, in vivo, and ex vivo.[76-78] Neutrophil recruitment is a critical step for neutrophils to manifest their functions. Neutrophils use an array of adhesion molecules for this purpose. Integrins are adhesion molecule families involved in many different biological processes. Among 24 integrins identified so far, β2 integrins, also called leukocyte integrins, are exclusively expressed in leukocytes. β2 integrins consist of the following four members: CD11a/CD18 (αLβ2, leukocyte function-associated antigen 1, LFA-1), CD11b/CD18 (αMβ2, macrophage-1 antigen, Mac-1), CD11c/CD18 (αXβ2), and CD11d/CD18 (αDβ2). Among them, LFA-1 and Mac-1 are highly expressed in neutrophils. LFA-1 and Mac-1 were shown to be inhibited by volatile anesthetics via direct interaction, while intravenous anesthetics did not.[79-83] Volatile anesthetic exposure attenuated neutrophil recruitment in vivo.[76,77,84] These indicate that volatile anesthetics may be favorable to attenuate neutrophil-mediated injury by attenuating their recruitment.

Neutrophils also have a number of TLRs. Volatile anesthetics directly bound to TLR4 and attenuated its activation.[85] In contrast, most intravenous anesthetics did not affect TLR4 function. Propofol inhibited TLR4 function only at very high concentrations. Similarly, TLR2 was directly inhibited by the volatile anesthetic, not by intravenous anesthetics.[86] These results suggest that volatile anesthetics could attenuate DAMP-mediated neutrophil activation, which would help reduce DAMP-induced tissue injury. In contrast, volatile anesthetics activated TLR9, while intravenous anesthetics did not.[87] The profiles of DAMPs may be important to truly understand the role of anesthetics in DAMP-mediated organ injury. So far, no study has determined if the type of anesthetics affects the amounts of DAMPs in the systemic circulation. Certainly, it is unlikely that the amounts of DAMPs released by primary surgical trauma and/or transfused blood products will be altered by anesthetics. However, tissue injury by ischemia-reperfusion injury or tissue injury by DAMPs-mediated micro-thrombosis can possibly be modulated. The role of TLRs, particularly TLR2 and TLR4, in ischemia-reperfusion injury is well recognized.[88] Thus, current data may suggest that volatile anesthetics potentially attenuate DAMP-mediated organ dysfunction, particularly through TLR2 and TLR4.

Volatile anesthetics vs intravenous anesthetics in cardiac surgical patients

As described above, a number of factors are responsible for organ dysfunction and mortality in the setting of cardiac surgery. Volatile anesthetics are shown to protect against ischemia-reperfusion injury in experimental settings.[89,90] These exciting data led to a number of clinical studies examining the role of anesthetic regimens in cardiac surgery, including meta-analysis.

Earlier studies have used various volatile anesthetic exposure regimens. Some investigators started volatile anesthetics before CPB initiation, some started after the release of the aortic cross-clamp, and some used them during the entire case. De Hert et al. examined the timing of volatile anesthetic exposure. They found that the exposure to volatile anesthetics over the entire duration of the case was associated with better outcomes compared to exposure only from just before the CPB, or after the reperfusion.[91]

Overall, a number of prospective studies supported the benefit of volatile anesthetics in cardiac surgery. Likhvantsev et al.[92] randomly assigned sevoflurane anesthesia or propofol-based total intravenous anesthesia in 868 patients undergoing coronary artery bypass graft (CABG) surgery with CPB. Patients who received sevoflurane anesthesia were associated with a shorter length of hospital stay (10 days vs 14 days), a reduction in cardiac troponin T release (0.18 ng/mL vs 0.57 ng/mL), N-terminal pro-brain natriuretic peptide (pro-BNP) release (633 pg/mL vs 878 pg/mL), and mortality at 1-year follow up (17.8% vs 24.8%). De Hert et al.[93] randomly assigned a volatile anesthetic regimen with either desflurane or sevoflurane, or a total intravenous anesthesia (TIVA) regimen to 414 patients undergoing CABG on CPB. They found that the one-year mortality was 12.3% in the TIVA group, while 3.3% in the sevoflurane group, and 6.7% in the desflurane group.

Landoni et al. performed meta-analyses involving patients undergoing surgery under volatile or intravenous anesthetics. The analyses showed that volatile anesthetics were associated with a significant reduction in myocardial infarction and mortality.[94,95] Based on these results, this study group conducted a large randomized control study involving 5,400 patients undergoing CABG on CPB either under volatile anesthetics or intravenous anesthetics. With surprise, there was no difference in the outcome between the groups.[96] However, propofol and other intravenous anesthetics were largely used in the volatile anesthetic group; 89% of patients in the volatile anesthetic group received intravenous anesthetics for induction, and 59% of them received intravenous anesthetics for anesthetic maintenance, which makes it difficult to understand the impact of volatile anesthetics. This led Bonanni et al. to conduct another meta-analysis. They performed a meta-analysis of randomized trials comparing volatile anesthetics versus propofol-based intravenous anesthetics for cardiac surgery requiring CPB.[97] The meta-analysis included 8,197 adult patients. Volatile anesthetics were not associated with a change in short-term mortality but were associated with lower 1-year mortality, myocardial infarction, cardiac troponin release, need for inotropic medications, shorter extubation time, and higher cardiac index/output.

Not all, but a number of volatile anesthetics showed favorable profiles in the setting of cardiac surgery. The mechanism of the cardioprotective effect of volatile anesthetics has been extensively studied and well established. For example, volatile anesthetics target the mitochondrial K channel, a channel important for cardioprotection.[98,99] In contrast, mechanistic analysis in the context of immunological aspects needs to be explored. Identifying the optimal anesthetic regimen, if any, would be important to enhance the outcome of cardiac surgery.

Although most of the studies focused on comparing volatile anesthetics vs intravenous anesthetics including propofol, benzodiazepines, and opioids, the use of dexmedetomidine is worth describing. The study of a total of 1,134 patients undergoing CABG surgeries and CABG plus valvular surgeries by Ji et al. showed that perioperative dexmedetomidine use was associated with a decrease in postoperative mortality up to one year and decreased incidence of postoperative complications.[100] Another study of 724 patients undergoing CABG surgeries by the same group also showed that perioperative dexmedetomidine use was associated with better in-hospital, 30-day, and 1-year survival rates.[101] A similar result was shown by Cheng et al.[102] In infant cardiac surgery, neurological complications after CPB are one of the serious comorbidities.[103] Dexmedetomidine has shown to be neuroprotective in animal models,[104,105] and the role of dexmedetomidine in infant cardiac surgery has been under evaluation currently.[106] Although dexmedetomidine did not affect TLRs and β2 integrin functions, and neutrophil functions in vitro,[86,107] Inada et al. found that neutrophil recruitment was significantly reduced by dexmedetomidine administration at least in part due to the reduction of neutrophil chemoattractants CXCL1 and CXCL2 production in vivo.[108]

CONCLUSION

We have reviewed the mechanism of the systemic inflammatory response associated with cardiac surgery, and the potential role of anesthetics from basic science to clinical study data. Cardiac surgery, especially when using CPB, is subjected to substantial systemic inflammatory response via several pathways. Extensive researches have revealed multiple molecules that act as DAMPs and their actions on various PRRs. This upstream interaction is theoretically critical for the proinflammatory response that contributes to postoperative organ dysfunction/failure. Novel technologies have been evolving to minimize these inflammatory processes. Immunomodulatory properties of anesthetic agents potentially influence the outcomes after cardiac surgery. Some of the volatile anesthetics may be favorable to attenuating DAMP-mediated activation of organ dysfunction.

Many of the current researches on DAMPs we have provided are performed in animal models. Future translation researches, as well as human studies, are needed to investigate the role of DAMPs mediated inflammation in cardiac surgery, including their mechanistic pathways to determine the optimal anesthetic regimen and the development of novel therapeutic strategies, if any, which would be essential to enhance the postoperative outcomes.

Financial support and sponsorship

This work is in part supported by NIH GM118277 and NICHD R21 HD109119.

Conflicts of interest

There are no conflicts of interest.