|Home | About | Journals | Submit | Contact Us | Français|
Rationale: The receptor for advanced glycation end products (RAGE) is an important marker of lung epithelial injury and may be associated with impaired alveolar fluid clearance. We hypothesized that patients with primary graft dysfunction (PGD) after lung transplantation would have higher RAGE levels in plasma than patients without PGD.
Objectives: To test the association of soluble RAGE (sRAGE) levels with PGD in a prospective, multicenter cohort study.
Methods: We measured plasma levels of sRAGE at 6 and 24 hours after allograft reperfusion in 317 lung transplant recipients at seven centers. The primary outcome was grade 3 PGD (PaO2/FiO2 < 200 with alveolar infiltrates) within the first 72 hours after transplantation.
Measurements and Main Results: Patients who developed PGD had higher levels of sRAGE than patients without PGD at both 6 hours (median 9.3 ng/ml vs. 7.5 ng/ml, respectively; P = 0.028) and at 24 hours post-transplantation (median 4.3 ng/ml vs. 1.9 ng/ml, respectively; P < 0.001). Multivariable logistic regression analyses indicated that the relationship between levels of sRAGE and PGD was attenuated by elevated right heart pressures and by the use of cardiopulmonary bypass. Median sRAGE levels were higher in subjects with cardiopulmonary bypass at both 6 hours (P = 0.003) and 24 hours (P < 0.001). sRAGE levels at 6 hours were significantly associated with intraoperative red cell transfusion (Spearman's ρ = 0.39, P = 0.002 in those with PGD), and in multivariable linear regression analyses this association was independent of confounding variables (P = 0.02).
Conclusions: Elevated plasma levels of sRAGE are associated with PGD after lung transplantation. Furthermore, plasma sRAGE levels are associated with blood product transfusion and use of cardiopulmonary bypass.
Soluble receptor for advanced glycation end products (sRAGE) is a novel marker of alveolar epithelial injury, and RAGE blockade attenuates experimental lung ischemia reperfusion injury. The role of sRAGE in clinical primary graft dysfunction is incompletely understood.
This multicenter cohort study demonstrates that sRAGE is elevated in primary graft dysfunction, suggesting that RAGE blockade may be a useful future therapy. In addition, sRAGE was elevated with use of cardiopulmonary bypass and red cell transfusion.
Primary graft dysfunction (PGD) is severe acute lung injury after lung transplantation (1–4). The pathophysiology of severe PGD is similar to other forms of acute lung injury (5) and represents the leading cause of early post-transplantation morbidity and mortality (1–3, 6–10). In humans, biomarkers of impaired coagulation and fibrinolysis (5), endothelial cell dysfunction (11), chemotaxis (7, 12), and cell adhesion (13) have been associated with PGD. However, the role of alveolar epithelial injury in PGD has not been fully evaluated.
The receptor for advanced glycation end products (RAGE) is a ubiquitously expressed receptor for nonenzymatically glycated adducts of endogenous proteins, lipids, and nucleic acids. These adducts form in the presence of hyperglycemia or oxidant stress and are termed advanced glycation end products (AGE). RAGE also binds to proinflammatory molecules, including S100 and high mobility group box 1 (14). Although RAGE is expressed in systemic vascular endothelium and nervous tissues, it is also highly localized to the lung where it is abundant in the basolateral membrane of alveolar epithelial type I cells (15). Recently, soluble RAGE (sRAGE) has been recognized as a potential plasma marker of type 1 cell injury. In both experimental and human acute lung injury, sRAGE is released into the alveolar and plasma compartments (16). A previous study in lung transplant recipients reported that plasma levels of sRAGE were associated with longer postoperative ICU stays; however, this investigation lacked sufficient power to detect an association between plasma sRAGE and clinical PGD (17).
We hypothesized that elevated plasma levels of RAGE within the first 24 hours post-transplant would be associated with the subsequent development of PGD. We also aimed to examine potential determinants of elevations in plasma sRAGE levels after lung transplantation.
We undertook a prospective cohort study of patients undergoing first lung transplantation at seven centers in the United States (5, 13). The sample for the current study consisted of 317 lung transplant recipients enrolled between June 2003 and February 2007 at these centers. Plasma samples were obtained in citrated tubes at 6 and 24 hours after reperfusion of the lung allograft(s). Samples were centrifuged within 30 minutes of collection and stored at −80°C. Clinical variables were categorized and defined using methods published previously (3). Use of blood and blood products was defined as only those products administered during the surgical procedure. Right heart catheterization was performed at induction of anesthesia. The study was approved by the Institutional Review Boards at each of the participating centers. Informed consent was obtained from all participants.
The primary outcome was PGD, defined as the International Society of Heart and Lung Transplantation grade 3 PGD on postoperative days 1, 2, or 3 (T24, T48, or T72) (18). As per the International Society of Heart and Lung Transplantation definition, PGD cases had diffuse alveolar infiltrates involving the lung allograft(s) that spared the native lung in the case of single lung transplantation; PaO2/FiO2 less than 200 mm Hg; and no other secondary cause of graft dysfunction, as defined in detail in prior publications (2, 3, 5, 13, 18). This definition of “any grade 3 PGD” has been adopted in a number of recent investigations (19–23). In addition, we conducted sensitivity analyses using patients with grade 3 PGD present at 72 hours, as in prior studies (3, 5, 8, 13).
Levels of sRAGE were measured in thawed plasma samples in duplicate by sandwich ELISA (R&D, Minneapolis, MN) compared with a standard curve of recombinant protein provided by the manufacturer. The intraassay coefficient of variation was 5.9%. Personnel responsible for performing assays had no knowledge of clinical data.
Continuous variables were compared using rank sum tests, and correlations were assessed using Spearman rank correlation coefficients. Contingency tables, Chi-square and Fisher exact tests were used, as appropriate. Multivariable logistic regression models were used to assess the potential confounding effects of clinical variables on the association of sRAGE levels at both 6 and 24 hours and the occurrence of PGD. sRAGE was assessed for proportionality of association with odds of PGD by plotting quartiles of sRAGE against PGD. In logistic regression equations, sRAGE was expressed as OR per 10-ng/ml increment (as this approximated the standard deviation of sRAGE). Potential confounders were included in the model one at a time to assess the confounding effects of each variable on the relationship of sRAGE (24). Linear regression was used to assess the relationship of continuous variables with sRAGE (such as units of blood transfused). In all linear regression equations, sRAGE was log transformed. Summary multivariable models included multiple potential confounding variables simultaneously if the confounders changed the coefficient by 15% or greater with adjustment (24). In addition, we examined the interaction of sRAGE levels with transplant procedure type and the use of cardiopulmonary bypass, in a priori specified analyses using multiplicative interaction terms in the logistic models. All statistical comparisons were performed using STATA version 10.0 (STATA Corp., College Station, TX).
The study sample included 317 lung transplant recipients. The incidence of PGD at any time point during the first 3 days after transplant was 27% (95% CI, 22–32%). The prevalence of PGD at 24 hours after transplant was 20% (95% CI, 16–25%), at 48 hours was 14% (95% CI, 10–19%), and at 72 hours was 14% (95% CI, 10–18%). Characteristics of the donor and recipient study population by PGD status are presented in Table 1. Recipients with PGD were more likely to be African American than those without PGD. Patients with PGD more commonly had diffuse parenchymal lung disease (pulmonary fibrosis) or pulmonary arterial hypertension and were less likely to have chronic obstructive pulmonary disease. Subjects with PGD had higher pulmonary arterial pressures at the induction of anesthesia. Use of cardiopulmonary bypass and blood products was more common in the patients with PGD.
Overall, plasma sRAGE levels were elevated in the immediate postoperative period and then decreased at 24 hours: median 7.9 ng/ml (interquartile range [IQR]: 3.8, 32.4) at 6 hours versus 2.3 ng/ml (IQR: 1.4, 4.5) at 24 hours after reperfusion (P < 0.001). sRAGE levels were significantly higher in patients with PGD compared with others at both 6 and 24 hours postoperatively (Figure 1). The median 6-hour sRAGE level was 9.3 ng/ml (IQR: 4.7, 19.7) in those with PGD versus 7.5 ng/ml (IQR: 3.8, 13.5) in those without (P = 0.028). Median sRAGE level at 24 hours post-transplant was 4.3 ng/ml (IQR: 2.6, 9.3) in those with PGD versus 1.9 ng/ml (IQR: 1.0, 3.8) in those without (P < 0.001).
We next assessed the association of sRAGE levels with PGD after adjustment for potential confounding variables in multivariable analyses. The unadjusted odds ratio (OR) for PGD associated with a 10-ng/ml increase in sRAGE level at 6 hours after reperfusion was 1.28 (95% CI, 1.10–1.49; P = 0.002). Adjustment for pulmonary artery systolic pressure in the model attenuated the relationship between the 6-hour sRAGE level and subsequent PGD (Table 2). Inclusion of additional covariates to the model did not change this result. There was no interaction (effect modification) between use of cardiopulmonary bypass (P = 0.64) or transplant procedure type (P = 0.99) and the relationship of sRAGE levels at 6 hours with PGD. Results were similar in examining 24-hour sRAGE level and PGD risk. The unadjusted OR for each 10-ng/ml increase in sRAGE at 24 hours was 1.46 (95% CI, 1.01–2.12; P = 0.043). There was no confounding by clinical variables except for use of cardiopulmonary bypass (adjusted OR, 1.29; 95% CI, 0.93–1.80; P = 0.14), and pulmonary artery systolic pressure (adjusted OR, 1.27; 95% CI, 0.93–1.73; P = 0.13).
Results were unchanged when the definition of PGD was restricted to grade 3 PGD present at 72 hours. Using this definition in the total population (n = 317), the unadjusted OR for PGD associated with a 10-ng/ml increase in 6-hour sRAGE level was 1.23 (95% CI, 1.05–1.45; P = 0.01) and 24-hour level was 1.35 (95% CI, 1.01–181; P = 0.044). Adjustment for pulmonary artery systolic pressure or the use of cardiopulmonary bypass attenuated the relationship between both the 6-hour and 24-hour sRAGE levels and PGD.
Patients undergoing cardiopulmonary bypass had higher levels of sRAGE than those who did not (Figure 2). The median 6-hour sRAGE level was 10.5 ng/ml (IQR: 6.0, 20.6) in the 109 subjects with use of cardiopulmonary bypass versus 7.4 ng/ml (IQR: 3.6, 13.2) in the 208 subjects without (P = 0.003), and at 24 hours was 3.9 ng/ml (IQR: 2.1, 7.3) versus 1.9 ng/ml (IQR: 1.0, 3.8) in those without (P < 0.001). The volume of transfused red blood cells was associated with sRAGE at both 6 hours (Spearman ρ = 0.21, P < 0.001), and 24 hours (ρ = 0.22, P < 0.001). In multivariable linear regression models, the association of 6-hour sRAGE levels remained significant when simultaneously adjusting for transplant type, preoperative diagnosis, and use of cardiopulmonary bypass (P = 0.020); however, 24-hour sRAGE levels lost statistical significance on adjustment for the same variables (P = 0.19). The association of 6-hour sRAGE levels was strongest among subjects with PGD: at 6 hours both in unadjusted analysis (ρ = 0.36, P = 0.002), and adjusted analyses (P < 0.001 in a multivariable linear regression model simultaneously adjusting for transplant type, preoperative diagnosis, and use of cardiopulmonary bypass). In contrast, there was no correlation between pulmonary artery systolic pressure at the induction of anesthesia and sRAGE levels (Spearman's ρ = 0.0005; P = 0.99 at 6 h).
Higher sRAGE levels at 6 and 24 hours were associated with greater length of mechanical ventilation (Spearman ρ = 0.34, P < 0.001 at 6 h; ρ = 0.38, P < 0.001 at 24 h). However, sRAGE levels were not associated with total hospital length of stay at either time point (Spearman ρ = 0.03, P = 0.69 at 6 h; ρ = 0.05, P = 0.50 at 24 h). A total of 17 of 317 subjects died at 30 days, of whom 13 subjects (76%) had PGD. sRAGE levels were significantly higher in those who died: median sRAGE measured at 6 hours was 25.9 ng/ml (IQR: 5.1, 44.2) in those who died versus 7.7 ng/ml (IQR: 3.8, 14.1) in 30-day survivors (P = 0.05); and at 24 hours was 13.0 ng/ml (IQR: 4.9, 36.6) versus 2.3 ng/ml (IQR 1.3, 4.3) in 30-day survivors (P < 0.001).
PGD increases morbidity and mortality and affects longer-term outcomes after lung transplantation (1, 2, 8, 19). In this study, we demonstrated that elevated plasma levels of sRAGE in the post-transplant period are associated with an increased risk of PGD. In addition, sRAGE levels were higher in those subjects who underwent cardiopulmonary bypass and who received blood product transfusions.
In the lung, sRAGE appears primarily to be a marker of alveolar epithelial cell injury (16). Uchida and colleagues showed that sRAGE was present in high concentrations in the alveolar compartment in experimental and human lung injury (16), and sRAGE levels were associated with clinical outcomes in a large clinical trial of two mechanical ventilation strategies in patients with acute lung injury (25). In isolated perfused human lungs, airspace sRAGE levels were inversely correlated with the rate of alveolar fluid clearance, supporting the theory that sRAGE is a relevant marker of alveolar epithelial injury (26). Likewise, a clinical study of alveolar epithelial fluid transport function in patients with PGD suggests that the alveolar epithelium plays a major role in both the pathogenesis and resolution of PGD (27). Therefore, elevated plasma sRAGE levels in our PGD subjects may indicate alveolar epithelial injury.
RAGE may also produce vascular inflammation by perpetuating endothelial cell dysfunction, as RAGE and its ligands have been implicated in vascular atherogenesis through a variety of mechanisms (28–30). Indeed, endothelial RAGE serves as a counter-receptor for leukocyte integrins such as Mac-1 (31), and the importance of leukocytes in the pathogenesis of acute lung injury had been well established. Because we did not perform RAGE expression analyses in cells from our subjects, we cannot specify the origin of the elevated plasma sRAGE levels in PGD (12).
Regardless of the cellular etiology, RAGE activation may be important in PGD. There is emerging evidence that RAGE blockade attenuates ischemia reperfusion injury in a murine pulmonary PGD model (32). Similar findings of the effect of RAGE blockade have been observed in hepatic ischemia reperfusion models (33, 34). Furthermore, administration of sRAGE attenuated lung injury in a murine model of LPS-induced lung injury suggesting a role for RAGE ligands in the augmentation of lung injury that is blocked by sRAGE (35). These data suggest a role for RAGE ligands in the augmentation of lung injury and should provide an impetus for further research into the relationship between RAGE, RAGE ligands, and the initiation and perpetuation of transplant-associated lung injury.
An important finding of our study was the association of plasma sRAGE levels with blood product transfusion. Accumulating evidence suggests an association between blood transfusion and the development of acute lung injury in the critically ill (36–38). The association of elevated sRAGE levels with red blood cell (RBC) transfusion may reflect increased alveolar or endothelial cell injury in patients receiving RBC transfusion, a finding that is supported by the stronger association between plasma sRAGE and RBC transfusion in patients with PGD in our study. Alternatively, RBC transfusion may promote cellular injury through activation of pathways initiated by the RAGE receptor. Recent evidence suggests that AGEs form in stored RBC units with increasing duration of storage (40). It is plausible that administration of RAGE ligands, such as RBC units containing AGEs, may increase detectable sRAGE levels in recipients, because RAGE expression is reported to be elevated in the presence of RAGE ligands (41). Further studies are necessary to determine the role of RBC transfusion and RAGE in the pathogenesis of PGD.
The association of sRAGE with PGD was in part attenuated by the use of cardiopulmonary bypass, and higher sRAGE levels were found in subjects who underwent cardiopulmonary bypass. The association of sRAGE levels with use of cardiopulmonary bypass could reflect alveolar epithelial injury that is caused by cardiopulmonary bypass itself. Alternately, the elevation of sRAGE levels associated with use of cardiopulmonary bypass could be due to systemic endothelial production of sRAGE (39). Similarly, the attenuation of the sRAGE association with PGD by elevated pulmonary artery pressures may indicate either shared pathophysiology, systemic inflammation, or simply may be an epiphenomenon of lung injury. In this cohort study, it was difficult to assess the individual contributions of elevated pulmonary artery pressures perhaps related to pretransplant diagnosis, use of bypass, and blood product transfusion, due to colinearity of these factors.
Our study is the first to show that plasma levels of sRAGE are elevated in clinical PGD. A prior study by Calfee and colleagues studied sRAGE at a single 4-hour time point in 20 subjects after lung transplantation, 7 of whom had grade 3 PGD (17). Our data are consistent with those of Calfee and colleagues in that RAGE levels were associated with length of mechanical ventilation. However, in contrast to our investigation, the prior study did not have sufficient power to test the association of RAGE with PGD.
Our study has several potential limitations. First, we did not measure preoperative RAGE levels in this study. Therefore, our ability to comment on the predictive value of RAGE is limited. Further, we measured RAGE levels in plasma, not bronchoalveolar lavage fluid, and thus our ability to localize RAGE to alveolar epithelial cells may be limited (14). Likewise, we did not measure other concurrent markers to discern if the source of sRAGE was predominantly epithelial or endothelial. In addition, the clinical definition of PGD may not represent pathological acute lung injury in all cases, particularly in the early postoperative period. Other mimicking conditions may have caused some misclassification of this outcome.
In this multicenter study, we have shown that elevated sRAGE levels in the plasma are associated with PGD after lung transplantation. We also have demonstrated an association between sRAGE levels and use of cardiopulmonary bypass and administration of blood products. These results suggest the need for future basic and clinical studies into the role of RAGE in PGD and in the pathogenesis of acute lung injury. Further, our results suggest that investigation of RAGE blockade may lead to a novel therapeutic direction in PGD research.
The authors thank Karen Murphy for her administrative help in preparing this manuscript.
Participants in the Lung Transplant Outcomes Group by site: Columbia University: David Lederer, M.D., M.S. (P.I.), Selim Arcasoy, M.D., Joshua Sonett, M.D., Jessie Wilt, M.D., Frank D'Ovidio, M.D., Nilani Ravichandran, N.P., Matthew Bacchetta, M.D., Nadine Al-Naamani, M.D., Debbie Rybak, B.A., Michael Koeckert, B.A., Robert Sorabella, B.A. University of Pennsylvania (Coordinating site): Jason Christie, M.D., M.S. (P.I.), Steven M. Kawut, M.D., M.S., Alberto Pocchetino, M.D., Y. Joseph Woo, M.D., Ejigayehu Demissie, M.S.N., Robert M. Kotloff, M.D., Vivek N. Ayha, M.D., James Lee, M.D., M.S., Denis Hadjiliadis, M.D., M.H.S., Melanie Doran, B.S., Richard Aplenc, M.D., Clifford Deutschman, M.D., M.S., Benjamin Kohl, M.D. Vanderbilt University: Lorraine Ware, M.D. (P.I.), Stacy Kelley-Blackburn, R.N. Stanford University: Ann Weinacker, M.D., (P.I.), Ramona Doyle, M.D., Susan Spencer Jacobs, M.S.N., Val Scott, M.S.N. University of Alabama, Birmingham: Keith Wille, M.D. (P.I.), Joao deAndrade, M.D., Tonja Meadows, R.N. Johns Hopkins University: Jonathan Orens, M.D. (P.I.), Pali Shah, M.D., Ashish Shah, M.D., John McDyer, M.D. University of Michigan: Vibha Lama, M.D., M.S. (P.I.), Fernando Martinez, M.D., M.S., Emily Galopin, B.S. Duke University: Scott M. Palmer, M.D., M.H.S. (P.I.), David Zaas, M.D., M.B.A., R. Duane Davis, M.D., Ashley Finlen-Copeland, M.S.W.
Supported by NIH HL081619, NIH HL087115, NIH HL081332, NIH HL088263, and the Craig and Elaine Dobbin Pulmonary Research Fund.
Originally Published in Press as DOI: 10.1164/rccm.200901-0118OC on August 6, 2009
Conflict of Interest Statement: J.D.C. has served on an advisory board for Hospira, GlaxoSmithKline, Discovery Labs ($1,001–$5,000), served as an expert witness for Rasmussen Willis Dickey and Moore, LLP, Dehay and Elliston, LLP, and Rawle Henderson, LLP ($10,001–$50,000), and has received industry-sponsored grants from Eli Lilly and Company ($100,001 or more). C.V.S. has received a consultancy for the American College of Physicians ($1,001–$5,000). S.M.K. has received a consultancy from Gilead ($1,001–$5,000), commercial entity from Gilead ($10,001–$50,000), lecture fees paid by Gilead ($1,001–$5,000). N.M. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript. D.J.L. has received a consultancy from CanAccord Adams ($1,001–$5,000) and industry-sponsored grants from Gilead ($50,001–$100,000) and Broncus Technologies ($10,001–$50,000). J.R.S. has received a consultancy from Covidian ($5,001–$10,000) and lecture fees paid by Covidian ($5,001–$10,000). V.N.A. has received lecture fees from Astellas Inc. (>$1,000), industry-sponsored grants from Astellas, Enzon Pharmaceuticals ($100,001 or more), and Alnylam ($10,001–$50,000). S.M.P. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript. K.W. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript. V.L. received lecture fees from Pfizer ($1,001–$5,000). P.D.S. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript. A.S. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript. A.W. has received an industry-sponsored grant from Lilly ($10,001–$50,000). C.S.D. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript. B.A.K. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript. E.D. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript. S.B. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript. L.B.W. has received an industry-sponsored grant from Luminex ($1,001–$5,000) and Sirius Genomics ($10,001–$50,000).