Cardiac surgery is associated with systemic inflammatory response syndrome (SIRS) that can feature cardiovascular, pulmonary, neurologic, splanchnic, hematologic, and immune system dysfunction, which directly influences the post-operative recovery (1, 2). The inflammatory response arises from many stimuli during the surgery, including tissue trauma, exposure to non-endothelialized cardiopulmonary bypass (CPB) circuit, ischemia-reperfusion, and hypothermia, yielding circulating pro-inflammatory mediators (2). This SIRS from cardiac surgery remains a clinical challenge as effective immunomodulatory therapies remain elusive, probably because of the diversity of immunologic pathways induced through the various stimuli (3, 4). A systematic understanding of the different immunologic reactions is paramount for future innovation and the development of efficacious treatments.

Two prominent immunologic stimuli during open-heart surgery include sternotomy and CPB. Sternotomy is performed at the beginning of open-heart surgery to access the patient's mediastinal and cardiac structures through skin and subcutaneous tissue incision, sternal division and pericardial entry (5). For neonates and young infants, a partial or complete thymectomy is also routinely performed to facilitate adequate visualization of the mediastinal structures (6). Acute immunologic reaction to tissue trauma includes hemostasis, which provides scaffolding for leukocytes, and inflammation, which are most relevant to the peri-operative setting (7, 8). Pro-inflammatory cytokines and chemokines – TNF, IL-1, IL-6, CCL2, CCL3, CXCL1, CXCL2, CXCL8, CXCL8, GM-CSF – are all upregulated in the acute inflammatory phase to stimulate leukocyte recruitment to the incision (7, 8).

CPB, initiated after sternotomy, is essential for open-heart surgery as the extracorporeal circuit provides circulation and oxygen delivery for the patient while providing a bloodless field for the surgeon (9). CPB activates the alternative complement pathway through the C3 roll-over mechanism, cleaving C3 into C3a and C3b, and subsequently releasing various cytokines and chemokines (10). C3b is a major player in the initiation, amplification, and effector functions of the whole complement pathway (10). Once CPB initiates the inflammatory cascade, this leads to an acute phase with pro-inflammatory reactions but also triggers a later resolution phase with anti-inflammatory interleukins 10 (IL-10) and receptor antagonist IL-1Ra (10). Importantly, the complement mediators (specifically C2, C3, C3a, C3b, C5, and C5a) have been shown to correlate with adverse outcomes after cardiac surgery, whereas the relevance of IL-6 and CXCL8 remains controversial (2, 11).

There has been limited research on the contribution of sternotomy to overall inflammatory sequelae in open cardiac surgery requiring CPB. Only a few human studies have investigated a limited number of inflammatory markers affected by cardiac surgery with a sternotomy in adults, with no consensus on the inflammatory effect of sternotomy (12, 13). Importantly, there have not been any pediatric studies investigating this topic, while adult studies have primarily focused on the inflammatory response for the entire cardiac surgery, either comparing mediators before sternotomy to the end of CPB or surgery and comparing surgeries with versus without sternotomy (5, 14). This study aims to describe inflammatory mediator profiles associated with sternotomy and to contextualize them within the inflammatory milieu at the end of CPB during children's heart surgery, to identify novel therapeutic targets.

Sternotomy has an active and distinct inflammatory profile during pediatric cardiac surgery. Our findings indicate that incisional trauma is associated with significant increases in circulating cytokine and chemokine concentrations, but not in any complement factors, because the patient's circulation has not yet been exposed to the non-endothelialized artificial surface of CPB. The sternotomy findings are contrasted against the significant CPB complement-mediated role in the inflammatory response, consistent with previous reports (11, 19). Several cytokines and chemokines – TNF, IL-1α, IL-1β, IL-2, TRAIL, CCL3, CCL4, CXCL1, CXCL2 and GM-CSF – were unique to sternotomy. IL-6, CXCL8, IL-10 and IL-1Ra were relevant to both sternotomy and CPB, exhibiting a multiplicative effect, as these mediators had a substantially higher fold increase during CPB compared to sternotomy. The continuous cytokine and chemokine production could relate to ongoing processes from sternotomy tissue injury, downstream effects from the complement reaction and also myocardial ischemia reperfusion during CPB (1, 2). C3a, C3b, IL-6 and CXCL8 are known to produce and stimulate direct and systemic neutrophil activation and release of pro-inflammatory cytokines, which directly facilitate endothelial leak syndrome, neutrophil recruitment, tissue edema, and subsequent injury that contributes to end-organ dysfunction in the post-operative period (1, 10, 11).

P-selectin was one of the adhesion molecules that increased significantly with sternotomy. This mediator plays a key role in wound hemostasis, in addition to leukocyte recruitment, explaining the immediate relevance to sternotomy (20). Other adhesion molecules are elicited in a transcription-dependent way by cytokines and chemokines and thus require several hours to achieve peak expression (21, 22). Therefore, this might explain why the other leukocyte adhesion molecules did not change significantly, as these are downstream mediators in the inflammatory cascade that would likely be elevated later in the post-operative period.

This was the first investigation to comprehensively characterize the dynamics of inflammatory mediators of sternotomy in children. Prior studies in adults primarily focused on the inflammatory response throughout the cardiac procedure, comparing mediators measured before sternotomy with those at the end of CPB (5, 14). A few adult studies have data on a limited number of inflammatory mediators affected by sternotomy, most of which contradict our results. Three studies reported no change in IL-6 concentrations during sternotomy (12, 23, 24), whereas two others reported a decrease (25, 26), in contrast to the robust 2.6x median fold increase observed here. Our results demonstrate that IL-6 concentration increases over time, a finding that may differ across studies. Additionally, the latter studies observed no changes in TNF and CXCL8, whereas our study demonstrated statistically significant increases in these mediators (27). C3a and C5b-9 were found to be active after anterolateral thoracotomy for CABG, which again conflicts with our observations (13). ET1 was the only mediator that showed a similar profile between pediatric and adults, as a single study reported dynamic increases over sternotomy in a CABG population (27). Allowing for differences in patient populations and study designs across these prior reports, one might conclude that the inflammatory response to sternotomy may differ between children and adults. However, this contrast should be interpreted with caution for several reasons. First, the adult studies were relatively small, with the largest being 30 patients divided into subgroups, and were not specifically designed to understand the inflammatory profile of sternotomy, as this report does. Secondly, the concept of immunosenescence is relevant to the elderly and comorbid patient populations (28). Third, pediatric congenital heart disease and adult acquired cardiac disease are distinct patient populations in terms of age and pathophysiology. Finally, the modern Luminex technology used in this study is more sensitive than historical enzyme-linked immunosorbent assay (ELISA) approaches.

IL-1β was the most active mediator during sternotomy, although the related mediators TNF and IL-1α also showed significant increases in circulation. IL-1α is released during cellular injury; TNF expression is primarily induced within activated macrophages, while IL-1β requires transcription and cleavage by caspase for biological activity (22, 29). These findings support current immunological understanding that, together, these are early pro-inflammatory signals produced mainly by macrophages, monocytes, and neutrophils in response to tissue trauma and damage-associated molecular patterns (30). Specifically, they induce the production of other innate cytokines and chemokines that activate macrophage and neutrophil machinery required for healing (29, 30). Interestingly, IL-1α, IL-1β, and TNF were predominantly active during sternotomy and inactive during CPB with continuous ultrafiltration, during which their concentrations decreased.

CXCL2 showed the second-largest median fold increase during sternotomy in our study. CXCL1 and CXCL8 also showed significant increases in concentration over the course of the sternotomy. These chemokines are all involved in the primary response to the inflammatory phase of wound healing and angiogenesis (31). In early phases of wound healing after tissue injury, this group of chemokines is released to recruit inflammatory cells, including neutrophils predominantly and macrophages, and later on, it mediates the proliferation phase of the wound healing response (31). Consistent with previous adult sternotomy reporting, we also observed GM-CSF concentration increase with sternotomy, as it is known to be upregulated in injured skin and helps orchestrate the proliferation of endothelial cells for restorative angiogenesis (7, 32).

The profiles of IL-6, CXCL8, IL-1Ra and IL-10 were unique from other mediators assessed in this study, as they consistently increased through the sternotomy and CPB phase, perhaps a sequential stimulation by two different stimuli (1, 10). IL-6 is a pro-inflammatory mediator released in response to tissue injury or other immunologic stimuli. Elevated concentrations after surgery and trauma are necessary for the timely resolution of wound healing by regulating leukocyte infiltration, angiogenesis, and collagen deposition (33). Consistent with our results of IL-6 production through sternotomy, other studies have detected IL-6 within 90 minutes of skin incision, lasting for several hours, and concentrations were related to the extent of the surgically induced trauma during elective procedures (34, 35). The anti-inflammatory mediator IL-1Ra is tightly coupled to pro-inflammatory cytokine production, primarily the IL-1 family of mediators, to prevent a hyper-inflammatory state and ultimately return the system to homeostasis (36). IL-10 is another anti-inflammatory mediator with a major role in the tissue repair process (37), observed to be stimulated by the sternotomy. IL-10 also has a vital role in inflammatory resolution in a variety of models of tissue repair, including cutaneous wound healing and in re-establishing tissue integrity after injury in multiple models of skin and myocardial injury (37).

The collection of active pro-inflammatory cytokines and chemokines observed during the sternotomy phase is characterized by their downstream NF-κβ-mediated responses (7, 8, 38). Innate immune cells resident in the skin and connective tissue include macrophages, dendritic cells and mast cells, which monitor damage-associated molecular patterns (DAMPs) through pattern recognition receptors (PRRs) (38–40). Upon stimulation, these surveilling leukocytes initiate the NF-κβ response observed in this sternotomy response with cascading TNF, IL-1, IL-6, IL-1Ra and IL-10, along with a host of chemokines. Interestingly, there was virtually no complement activation during the sternotomy period, which was limited to CPB exposure. Altogether, the multiple immune stimuli of tissue trauma, CPB non-endothelialized circuit, and myocardial ischemia-reperfusion, among others, culminate in a pro-inflammatory systemic response (Figure 4) (1, 11).

IL-2 is traditionally described as an adaptive mediator with pleiotropic effects on T-cell function, which is largely unrelated to non-specific adaptive responses during cardiac surgery (41, 42). IL-2 has been sparsely studied during the proliferative and remodeling phases of wound healing, but has been measured in the local wound environment and in the systemic circulation in the days following tissue injury to orchestrate keratinocytes, fibroblasts and endothelial cells (43, 44). For the first time, our results indicate that circulating IL-2 concentrations increase substantially (1.7x median fold increase) immediately after the sternotomy tissue trauma. IL-2 is known to be produced by activated dendritic, mast cells and macrophages, which are stimulated during incisional damage and trauma and could explain these unique observations (42). IL-2 may have an unexplored innate immune function in the acute setting of tissue injury, particularly given that the majority of data in this study arises from immunologically immature neonates and infants. This postulated mechanism requires dedicated investigation and validation.

The authors recognize that this descriptive study is not without limitations. First, the study population was heterogeneous across several variables relevant to systemic inflammation, including patient age, congenital heart disease pathology, steroid administration, type of operative repair, CPB time, myocardial ischemia-reperfusion time, intraoperative transfusion, and hypothermia. The sample size of 40 is prohibitive for subgroup analyses by age group or for interrogating these specific perioperative variables relative to immunologic response. The impact of heterogeneity in patient baseline demographics was partially mitigated by including paired within-individual comparisons through the sternotomy and CPB phases. Second, the data are drawn from a single center, which limits generalizability. In particular, the results should be interpreted in the context of pediatric CPB with continuous ultrafiltration, where cytokines, chemokines, and complement anaphylatoxins C3a and C5a are known to be removed from the circulation (45, 46). Mediator changes throughout CPB exposure could theoretically exceed the values reported here if ultrafiltration were not used.

In conclusion, sternotomy during pediatric cardiac surgery is a major immunologic stimulus and primarily results in increased expression of circulating pro- and anti-inflammatory cytokines and chemokines – TNF, IL-1α, IL-1β, IL-2, TRAIL, CCL3, CCL4, CXCL1, CXCL2, and GM-CSF – through an NF-κB response to tissue trauma. In contrast, CPB primarily activates the complement system and further increases specific cytokines and chemokines, including IL-6, CXCL8, IL-1Ra, and IL-10. CPB and, perhaps, the myocardial ischemia-reperfusion reaction during CPB appear to make a dominant contribution to these mediators versus sternotomy. These results clearly delineate two temporally distinct, and perhaps synergistic, immunologic stimuli during pediatric cardiac surgery and lay the foundation for exploring innovative therapeutic options to dampen inflammation and enhance post-operative recovery in children with congenital heart disease.