Plasma sRAGE enables prediction of acute lung injury after cardiac surgery in children

Postoperative lung injury may occur in 12% to 50% of patients undergoing cardiac surgery necessitating cardiopulmonary bypass (CPB) [1‐4], and up to 20% of the patients need ventilation for more than 48 hours [5]. Acute respiratory distress syndrome (ARDS) occurs in 2% of these cases, resulting a mortality of 15% to 50% [3, 4]. Children are more prone to acute lung injury (ALI) under the detrimental stimulation during cardiac surgery with CPB [6‐8]. ALI/ARDS after CPB often leads to a prolonged length of hospital stay and increased therapeutic costs. Therefore, developing certain predictive biomarker for CPB-related ALI/ARDS in infants and young children undergoing cardiac surgery would be of great help for early diagnosis and efficient therapeutic decision making.

The receptor for advanced glycation end products (RAGE) is a member of the immunoglobulin superfamily that acts as a multiligand receptor and progression factor in propagating the inflammatory response [9, 10]. RAGE is expressed at a high level in the basal surface of alveolar type I cells, and its expression is upregulated under inflammatory conditions [9, 11, 12]. Soluble RAGE (sRAGE) is the extracellular form of RAGE and is produced by either cleaving the membrane by proteolysis or removing the transmembrane region via alternative splicing [12]. Increasing evidence suggests that local and circulating levels of sRAGE are promising as biomarkers of pulmonary tissue damage. The sRAGE concentration has been reported to be dramatically higher in pulmonary edema fluid and plasma from adult patients with ALI/ARDS, which were caused by sepsis, trauma, or primary pneumonia [13, 14]. Furthermore, increased plasma sRAGE levels are significantly correlated with the severity and clinical outcome of ALI/ARDS [15, 16].

S100A12 is a member of the S100 family of calcium-binding proteins, which is a newly identified extracellular prototypic ligand of RAGE [17, 18]. S100A12 expression may reflect neutrophil activation and contribute to pulmonary inflammation and endothelial activation via binding to RAGE [14, 18]. Blocking the interaction of S100A12 and RAGE results in improvement of inflammation in multiple experimental models [18, 19]. Wittkowski et al. [14] found that compared with healthy controls, patients with ARDS caused by pneumonia or peritonitis had significantly enhanced S100A12 expression in pulmonary tissue and higher S100A12 concentrations in bronchoalveolar lavage fluid.

Plasma sRAGE was recently reported to be a sensitive and rapid marker of lung distress after elective coronary artery bypass grafting in a study of 20 adult patients [20]. Additionally, Kikkawa and colleagues [21] found that sepsis patients who developed postoperative ALI had higher S100A12 levels compared with controls. Initial studies are promising. However, it is not clear whether these findings could extend to younger children undergoing cardiac surgery with CPB.

This pilot study aimed to observe the kinetics of plasma sRAGE and S100A12 in infants and young children undergoing cardiac surgery with CPB and to investigate whether plasma sRAGE and S100A12 levels are associated with the occurrence and severity of ALI after cardiac surgery.

In this prospective pilot study of the kinetics of plasma sRAGE and S100A12 levels in infants and young children undergoing cardiac surgery with CPB, we found that plasma concentrations of sRAGE and S100A12 remarkably increased immediately after CPB. The elevated levels were correlated with the severity of ALI and clinical outcomes of cardiac surgery with CPB and plasma sRAGE was a reliable early predictor for the occurrence of ALI after cardiac surgery.

Uniquely, RAGE is expressed at remarkably high basal levels in lung tissues [9, 11, 12], suggesting that it holds important functions in both pulmonary physiology and numerous pathologic states [9, 12]. The normal values of sRAGE were reported to range from 500 to 1,250 pg/ml in healthy volunteers [20, 21, 24, 25]. In addition, no significant difference was found in normal plasma sRAGE levels among various ages and in baseline levels with respect to the cause of lung injury, such as aspiration pneumonia, and to coexisting conditions, such as diabetes, end-stage renal disease, essential hypertension, or other vascular diseases [13]. Previous studies provided supportive evidence that the primary source of plasma sRAGE in ALI/ARDS patients was alveolar type I cells [11‐13, 24], and sRAGE served as a marker of alveolar epithelial injury [15, 16, 24]. As expected, in the current study, we found that the plasma levels of sRAGE were significantly higher in the ALI group than in the non-ALI group immediately after CPB. After adjusting the ALI-related variants, sRAGE remained as an independent risk factor for ALI development. Furthermore, we found the sRAGE levels were related with the impaired lung function, such as lower PaO₂/FiO₂ ratio and higher radiographic LIS. These findings suggested that plasma sRAGE levels may serve as a reliable quantitative predictor for the development of post-CPB ALI in infants and young children. Because plasma was diluted during CPB, the normalized plasma sRAGE and S100A12 concentrations were used to define the cut-off values (sRAGE, 54 pg/mg; S100A12, 4,326.5 pg/mg) in our study. Nevertheless, the cut-off value should be used with caution to diagnose ALI in a clinical setting, because the defined abnormal sRAGE or S100A12 levels were unequal among different studies in different patients [20, 21, 24, 25]. Unfortunately, to date, only a few studies exist on sRAGE and S100A12, which were always based on small number of patients. Hence, a larger-cohort validation study is needed to verify these results in future.

Previous studies have shown that the elevated circulation levels of sRAGE after onset of ALI/ARDS soon decreased to lower levels [13, 14, 17, 18]. Interestingly, we found similar kinetics of plasma sRAGE during CPB in children, to those in adult patients receiving coronary artery bypass grafting [20]. One possible explanation for this decrement is the reoxygenation resulting from reperfusion of the lung after CPB, because Lizotte et al. [26] reported that hyperoxic exposures lead to a loss of sRAGE in the neonatal rat lung. In addition, a great number of activated neutrophils during extracorporeal circulation may explain the decreasing sRAGE levels [27]. sRAGE is known to have multiple protease-sensitive sites, so it might be degraded by neutrophil-derived proteolytic enzymes [28].

Infiltration of activated neutrophils is an important hallmark of ALI [29]; therefore more neutrophils in ALI may account for the high levels of S100A12 in the ALI group. Although neutrophils and their secretory products play important roles in the pulmonary inflammation, such as ALI, they are a rather unspecific surrogate for lung-tissue injury [29‐31], which may contribute to the disassociation of plasma levels of S100A12 with the occurrence and the severity of post-CPB ALI in the logistic analysis and the multiple linear regression analysis.

In addition, the present study found that both plasma levels of sRAGE and S100A12 immediately after CPB were positively correlated with the duration of MV, surgical ICU LOS, and hospital LOS, and that sRAGE showed an independent association with the PaO₂/FiO₂ ratio on the second day. Although speculative, the following reasons might explain these correlations. In ALI/ARDS patients, Mauri et al. [15] reported that the first-day plasma sRAGE levels were correlated with the release of CRP and pentraxin 3 (PTX3) in the circulation [15], which promoted the coagulation/fibrinolysis dysfunction and organ failures [15, 32]. RAGE interacts with leukocyte Mac-1 integrin, monocyte chemoattractant protein-1, and CD11c/CD18 to facilitate inflammatory cell recruitment [33, 34]. Attraction of leukocytes to an inflammation site is additionally augmented by interaction with RAGE ligands such as S100A12 [10, 14, 18, 32]. Furthermore, interaction of RAGE and S100A12 causes production of proinflammatory cytokines including TNF-α, IL-6, and IL-8 [12, 21], which induced increased lung cell damage and vascular permeability [34, 35]. Taken together, RAGE and S100A12 contribute to more severe and prolonged inflammation in the lung and damage of the alveolar capillary barrier with increased lung water content and impaired oxygenation, resulting in longer MV support and more critical care in the SICU after cardiac surgery. There was no doubt that the S100A12-RAGE axis played an important role in the development of ALI after cardiac surgery with CPB. Because several single nuclear polymorphisms of the RAGE gene were recently found to affect the expression of sRAGE and associate with lung function in pathologic states [36‐38], it should be emphasized that genetic variation of RAGE must be considered in further studies.

Limitations of the present study should be acknowledged. Although the kinetics of plasma sRAGE levels in this study is consistent with that in a published study [20], our findings were based on a small number of patients in a single center, which might limit the application of these findings to other institutions. Furthermore, we enrolled the patients undergoing surgery for CHD in the absence of relevant comorbidities and particularly of known lung disease. Therefore, we are not sure whether the sRAGE and S100A12 changes would be the same in the presence of previous lung diseases. Because children often have pneumonias before cardiac operations, future studies must confirm this.

Interestingly, Jabaudon et al. [13] reported that sRAGE levels in ALI/ARDS were not influenced by the presence of RAGE-related diseases such as sepsis, diabetes mellitus, end-stage renal disease, coronary artery disease, rheumatoid arthritis, Alzheimer disease, and essential hypertension.

Third, the traditional inflammatory cytokines were not measured in this study. However, none of them was a specific biomarker for post-CPB ALI, although they were found to change perioperatively. In addition, in vitro study showed that the induction of S100A12 was more prominent than other cytokines, such as IL-8 [14], suggesting that the S100A12-RAGE axis may be more important in the CPB-induced inflammatory response.

We found that plasma sRAGE acts as a reliable early biomarker of ALI after cardiac surgery with CPB in infants and young children. The plasma levels of sRAGE and S100A12 are associated with lung function and clinical outcomes after CPB. These data may indicate that RAGE is a mediator of lung injury in patients after cardiac surgery and not merely a maker of disease. Insight into the role of the ligand-RAGE axis in the post-cardiac surgery inflammatory response holds the potential for better understanding of ALI induced by CPB.