Objectives: Hepatic ischemia-reperfusion injury and transfusion of red blood cells in liver surgery are well-known risk factors to induce acute tubular injury. Transfusion of stored red blood cells may affect hepatic ischemia-reperfusion injury-induced acute tubular injury. Here, we hypothesized whether preischemic (due to increased severity of hepatic injury) and postischemic (due to renal uptake of free heme and iron) transfusion of stored red blood cells may potentiate acute tubular injury in rats subjected to hepatic ischemia-reperfusion injury.

Materials and Methods: Sprague Dawley rats (n = 24) were divided into 4 groups: sham operation (sham group), hepatic ischemia-reperfusion injury only (injury-only group), red blood cell transfusion before hepatic ischemia-reperfusion injury (preinjury transfusion group), and red blood cell transfusion after hepatic ischemia-reperfusion injury (postinjury transfusion group). Partial hepatic ischemia was induced for 90 minutes, with reperfusion allowed for 12 hours. Hepatic and renal tubular injury markers, renal mRNA levels of oxidant stress markers, and inflammatory markers were assessed. Renal cortex samples were examined under hematoxylin and eosin staining for tubular histopathologic score and immunohistochemical staining for inflammatory cells.

Results: With regard to hepatic and renal tubular injury markers, serum alanine aminotransferase, serum urea nitrogen, and histopathologic scores were increased in the preinjury and postinjury transfusion groups versus injury-only group, with moderate to strong correlation between alanine aminotransferase and tubular injury markers. Renal oxidative stress markers (heme oxygenase-1 and neutrophil gelatinase-associated lipocalin) were correlated with increased alanine aminotransferase, with upregulation of oxidant stress markers in the preinjury transfusion group versus sham group (all markers), as well as in the injury-only and postinjury transfusion groups (heme oxygenase-1 only). We observed no changes in renal inflammatory responses among the groups.

Conclusions: Preischemic transfusion potentiated acute tubular injury without triggering renal inflam-matory responses. Exacerbation of hepatic injury may induce acute tubular injury via renal oxidant stress.

Key words : Acute kidney injury, Hepatic injury, Renal oxidant stress

Introduction

Hepatic ischemia-reperfusion injury (HIRI) is known to produce renal injury in animal models^1,2 and in humans.^3,4 Oxidative mediators and proinflam-matory cytokines released from the liver and the cells of the innate immune system, such as Kupffer cells and neutrophils, are believed to result in renal injury.³ In a murine model, serum alanine aminotransferase (ALT) levels were shown to correlate with serum creatinine 24 hours after HIRI1; in addition, in a prospective cohort human study, peak serum aspartate aminotransferase (AST) levels were the only independent risk factor for acute kidney injury after liver transplant.³ These reports suggest that the severity of hepatic injury after HIRI may affect the degree of the renal injury.

Transfusion of red blood cells (RBCs) has been reported to be an independent risk factor for clinical acute kidney injury in liver resection surgery⁵ and transplant^3,6 as a result of hemolysis and an increase in circulating free heme and iron, which causes oxidative stress or impaired oxygen delivery.^7,8 Hepatic surgery is usually considered to entail ischemia-reperfusion injury.

These studies raise a question of whether RBCs may potentiate renal injury when they are given to those exposed to HIRI, by increasing the severity of hepatic injury (with resultant increased oxidative mediators and cytokines) or the renal uptake of free heme and iron. We have recently reported that transfusion of stored RBCs (SRBCs) before HIRI (preischemic transfusion), but not after HIRI (postischemic transfusion), exacerbated hepatic injury in rats.⁹ We hypothesized whether not only preischemic (due to increased severity of hepatic injury) but also postischemic (due to renal uptake of free heme and iron) transfusion of SRBCs may potentiate acute tubular injury (ATI) in rats subjected to HIRI. To our knowledge, the effects of peri-ischemic blood transfusion on HIRI-induced acute renal injury have not been studied.

Materials and Methods

Animals, collection and storage of red blood cells, and surgical protocolAnimals, collection and storage of RBCs, and the partial HIRI method have been described in detail elsewhere.⁹ All animal procedures were approved by our Institutional Animal Care and Use Committee. Briefly, 24 male Sprague Dawley rats were divided into 4 groups: sham operation (sham group), HIRI only (IR group), and transfusion before HIRI (Pre-T injury group), and transfusion after HIRI (Post-T injury group). For transfusion, allogeneic RBCs stored for 2 weeks were transfused before or after hepatic ischemia. Partial hepatic ischemia was induced for 90 minutes, and reperfusion was allowed for 12 hours.

Collected blood samples were analyzed for hepatic injury markers (AST and ALT) and renal tubular injury markers (serum urea nitrogen and creatinine). Kidney tissues were excised, and the cortices were prepared for reverse transcription-polymerase chain reaction for quantification of mRNA expression of oxidative stress-induced compensatory proteins, including neutrophil gelatinase-associated lipocalin (NGAL), heme oxygenase-1 (HO-1), extracellular superoxide dismutase (EC-SOD), endothelial nitric oxide synthase (eNOS); inflammatory markers, including tumor necrosis factor-alpha (TNF-α), interleukin (IL)-1β, IL-6, IL-18, and, cyclooxygenase-2 (COX-2); and kidney injury molecule-1 (KIM-1, as one of tubular injury markers). Tubular histopath-ologic scoring and immunohistochemistry were also performed.

Rats were euthanized after 12 hours of reper-fusion. Because it was previously demonstrated that upregulation of the redox-sensitive stress-induced proteins, including NGAL, HO-1, EC-SOD, and eNOS, is the consequence of oxidant stress,2,10 we used these as indices of renal oxidant stress.

Measurement of serum levels of aspartate aminotransferase, alanine aminotransferase, urea nitrogen, and creatinine
Serum levels of AST, ALT, urea nitrogen, and creatinine were measured by using a colorimetric assay, according to the manufacturer’s protocols (Asan Pharmacy, Seoul, Korea).

Tissue RNA preparation and reverse transcription-polymerase chain reaction
The primer sequences for COX-2, TNF-α, IL-1β, IL-6, IL-18, KIM-1, NGAL, EC-SOD, HO-1, eNOS, and β-actin are listed in Table 1. Tissue RNA preparation and reverse transcription-polymerase chain reaction have been described in detail elsewhere.⁹

Histology and immunohistochemistry
Tissues from kidneys were harvested, fixed in 10% phosphate-buffered formalin, washed, dehydrated in gradual ethanol, and finally embedded in paraffin. Tissue blocks were sectioned at a thickness of 5 μm and then stained with hematoxylin and eosin. After degree of tissue injury was measured using light microscopy (original magnification, ×400), the severity of renal tubular injury was evaluated based on the histopathologic score (HPS) according to the scoring system proposed by Miura and colleagues,11 which is as follows: score of 0 for normal tubules, score of 1 for mild blebbing and loss of brush border, score of 2 for intensive blebbing and mild vacu-olization, score of 3 for shrunken nuclei and intensive vacuolization, score of 4 for necrotic/apoptotic cells and denudation/rupture of basement membrane, and score of 5 for total necrosis of renal tubule. Immunohistochemical staining for myeloperoxidase and CD68 was performed for assessing neutrophil, dendritic cell, and macrophage infiltration (magnification, ×200). Acute tubular injury was defined as HPS > 0.

Statistical analyses
Data analyses were performed using SPSS statistical software (version 20.0 for Windows; SPSS, Chicago, IL, USA). Data are expressed as mean ± standard error or median with interquartile range. Normality of distribution was assessed using Q-Q plots, Z-test, and the Shapiro-Wilks test. One-way analyses of variance (ANOVA) with post hoc Scheffé test was used for continuous, normally distributed variables. Homogeneity of variance was assessed by Levene’s test, and the data of normal distribution with unequal variances (AST and serum urea nitrogen) were log-transformed. Nonnormally distributed data were analyzed by Kruskal-Wallis test with post hoc testing using one-way ANOVA and Tukey honestly significant difference test for multiple comparisons on the ranks of the observations. To assess the relationship between the severity of hepatic injury and degree of renal tubular injury and between the severity of hepatic injury and renal oxidant stress, Spearman rank correlation analysis was used. A value of P < .05 was considered significant.

Results

Preischemic transfusion of stored red blood cells exacerbated hepatic injury
Serum levels of AST and ALT, as hepatic injury markers, were increased significantly in all groups subjected to HIRI compared with the sham group (P < .01). In the Pre-T injury group, ALT was increased compared with that shown in the IR group (P = .022). Neither AST nor ALT levels were increased when we compared the Post-T injury group versus the IR group (Figure 1).

Preischemic, but not postischemic, transfusion of stored red blood cells potentiated acute tubular injury with moderate to strong association between hepatic injury and renal tubular injury
To assess whether transfusion of SRBCs affected HIRI-induced ATI, HPS, serum urea nitrogen, serum creatinine levels, and KIM-1 mRNA expression levels were measured (Figure 2). We found that HPS was increased in all groups subjected to HIRI versus that shown in the sham group (P < .01); in addition, HPS was increased in the pre-T injury group versus the IR group (P = .004) and versus the Post-T injury group (P = .009). Serum urea nitrogen was increased in the IR group (P = .03) and Pre-T injury group (P = .001) but not in the Post-T injury group compared with the sham group. This increase was significant in the Pre-T injury versus IR group (P = .022) and Post-T injury group (P = .002). Serum creatinine and KIM-1 mRNA levels were only increased when we compared the pre-T injury group versus sham group (P = .002 and .001, respectively) and post-T injury group (P = .047 and .001, respectively). These levels were not increased when we compared the Pre-T injury group versus the IR group. To associate the severity of hepatic injury marker with renal injury markers, Spearman correlation coefficients were analyzed, showing r = 0.73 (P = .001), 0.61 (P = .01), and 0.52 (P = .029), respectively, for serum ALT versus serum creatinine, serum urea nitrogen, and KIM-1 mRNA levels (1 outlier was removed for analysis of Spearman coefficient for serum ALT vs creatinine). Correlations between hepatic injury markers and HPS were not observed.

Hepatic ischemia-reperfusion injury did not induce inflammatory responses in the renal cortex, irrespective of transfusion of stored red blood cells
Infiltration of myeloperoxidase-positive neutrophils and CD68-positive macrophages/dendritic cells was not identified in the renal interstitium in any of the groups (Figure 3). Transfusion of SRBCs did not affect renal infiltration of the inflammatory cells, despite development of ATI following HIRI. All measured markers of inflammation, the mRNA levels of COX-2, TNF-α, IL-1β, IL-6, and IL-18, were not increased in the IR, Pre-T injury, and Post-T injury groups compared with the sham group (Figure 3). Among them, only IL-6 mRNA expression was increased in the Pre-T injury group versus the IR group (P = .016), although no difference was found between the sham group and the Pre-T injury group. To assess the relationship between the severity of hepatic injury and the expression of renal inflammatory markers, the Spearman correlation coefficient was analyzed; no significant association was observed between ALT and the inflammatory markers.

Oxidative stress was correlated with hepatic injury
To assess the expression of oxidative stress-induced cytoprotective compensatory proteins in the renal tissue, mRNA expression levels of HO-1, EC-SOD, eNOS, and NGAL were measured (Figure 4). The HO-1 mRNA level was significantly increased in both the Pre-T injury and Post-T injury groups compared with that shown in the sham group (P = .001) and compared with that shown in the IR group (P = .004 and P = .008, respectively). These levels were not increased in the IR group. The mRNA levels of EC-SOD, eNOS, and NGAL were not increased when we compared the Pre-T injury versus IR group. The mRNA level of NGAL was significantly increased in the IR (P = .029) and the Pre-T injury group (P = .001) compared with the sham group, with its expression also increased in the Pre-T injury versus Post-T injury groups (P = .013). We found that EC-SOD (P = .043) and eNOS (P = .037) were only increased when we compared the Pre-T injury versus the sham group. To assess the relationship between severity of hepatic injury and expression of renal oxidative stress markers, Spearman correlation coefficient was used, which showed r = 0.52 (P = .028) and r = 0.49 (P = .037) for HO-1 and NGAL mRNA versus ALT, respectively.

Discussion

In this study, we assessed whether peri-ischemic transfusion of SRBCs potentiates ATI in rats subjected to HIRI. Our main finding was that preischemic, but not postischemic, transfusion of SRBCs increased serum ALT and potentiated ATI, with moderate to strong association between hepatic injury and tubular injury. Inflammatory responses, indicated by inflammatory markers and cell infiltration, were not observed in the renal cortex. However, the degree of renal oxidative stress was correlated with the severity of hepatic injury that induced ATI.

The present study showed that preischemic transfusion of SRBCs potentiated renal tubular injury. The 2 underlying causes of renal tubular injury, severity of hepatic injury1 and renal uptake of free hemoglobin and iron,^7,8 may induce tubular injury by producing oxidative mediators and proinflammatory cytokines in our model. Uptake of free hemoglobin and iron released from the transfused SRBCs did not appear to play a significant role in inducing tubular injury because RBC transfusion after HIRI revealed limited tubular injury. In contrast, preischemic RBC transfusion increased serum ALT, a marker of hepatic injury, in addition to tubular injury markers correlated with ALT. Therefore, the main mechanism for tubular injury seems to be exacerbation of hepatic injury, rather than renal uptake of free heme and iron from the transfused SRBCs. However, the results of our present study were not able to show distinct differences in oxidative stress markers among groups. Instead, correlation of hepatic injury with renal oxidative stress (HO-1 and NGAL) in the absence of inflammatory responses suggested that hepatic injury induced tubular injury via renal oxidative stress. The increasing severity of hepatic injury seemed to release more proinflammatory cytokines and oxidative mediators, including the hepatic heme-based protein (cytochrome c) to induce renal oxidative stress^2,12 because the severity of hepatic injury was increased in the Pre-T injury group and was correlated with HO-1 mRNA expression levels.

Inflammatory cell infiltration was not observed in the renal cortex; in addition, dendritic cells did not seem to be involved in ATI in our study, as the cell surface marker CD68 did not distinguish between dendritic cells and resident or infiltrating macrophages.¹³ These results are in line with a previous study in which hepatic ischemia did not trigger an inflammatory response and neutrophil accumulation in the rat kidney.¹⁴ It has been previously suggested that acutely increased iron load to the monocyte-macrophage system due to blood transfusion may contribute to organ dysfunction after major surgery.^15,16 As a result, we expected that priming of renal innate immunity by preischemic transfusion of SRBCs⁹ might induce hyperres-ponsiveness and infiltration of neutrophils and monocyte/macrophage infiltration in the kidney, like in the liver, which might lead to potentiation of renal tubular injury, based on our previous study. However, the infiltration of these inflammatory cells was not observed. Thus, priming of renal innate immunity appears not to be involved in potentiation of ATI by preischemic transfusion of SRBCs. We speculate that the relatively smaller number of renal macrophages than hepatic or splenic tissue macrophages may partly explain the reason. Furthermore, because systemic proinflammatory cytokines, oxidants, and other nephrotoxic mediators from injured liver are generally considered to induce renal proinflammatory cytokine transcription and induction of the cytokines evokes renal inflammatory responses, it is surprising that expression of inflammatory markers was also not observed. Previous studies found that renal mRNA expression levels of inflammatory markers (proinflammatory cytokines and COX-2) in response to HIRI are variable.^1,2 Collectively, the renal inflammatory responses were not triggered in our remote renal injury model.

Notably, only HO-1 mRNA, among the measured stress-induced compensatory proteins, increased in both the Pre-T and Post-T injury groups versus the sham and IR groups. Thus, HO-1 mRNA levels seem to be upregulated by renal uptake of free heme and iron from transfused SRBCs. Our results showed that HO-1 mRNA upregulation did not correspond with tubular injury. Previous studies on HO-1 induction have revealed mixed results in acute kidney injury models,¹⁷ which could explain the reason why HO-1 was not upregulated in the IR group. On the other hand, transfusion-induced systemic iron overload can cause increased iron uptake in renal tubular cells,⁷ and NGAL is upregulated to tolerate high iron concentrations¹⁸ and to ameliorate oxidative stress-mediated toxicity.^18-20 Thus, the decreased NGAL mRNA expression levels shown in the Post-T versus Pre-T injury group contradicts the finding with HO-1 mRNA upregulation. It is difficult to understand the reason for this in the present study.

Liver resection or transplant may require transfusion of RBCs. The use of stored RBCs cannot be avoided, and standard-issue RBCs are known to vary widely in age.^21,22 Moreover, some patients may need transfusion before hepatic ischemia, in cases of anemia, preoperative trauma, and hemorrhage. Our results indicate that transfusion before ischemia of the primary organ (liver) exacerbates ischemia-reperfusion injury of the primary organ and subsequently potentiates injury of the remote organ (the kidney), confirming our previous results that transfusion should be delayed until after an ischemic event during liver surgery.⁹

There are several key limitations in the present study. First, we did not use sham operation with transfusion of SRBCs. Transfusion of SRBCs may induce oxidative stress in the kidney.8 Thus sham operation plus transfusion of SRBCs may be helpful in determining the effects of SRBCs as a transfusion control. Second, renal mRNA expression levels of the stress-induced cytoprotective proteins may be influenced by circulatory levels of proinflammatory cytokines from the injured liver. Thus, it may be necessary to confirm the role of systemic proin-flammatory cytokines. However, a question of how many systemic proinflammatory cytokines should be measured remains. In addition, plasma cytokines may be also variable or may not be increased in a HIRI model.² Furthermore, many oxidant mediators also affect the renal mRNA expression levels of stress proteins and proinflammatory cytokines. Thus, it may be impractical to measure plasma concentrations of cytokines. Third, because tubular injury and expression of renal compensatory proteins are ongoing and serum creatinine levels rise slowly, the monitoring period may need to be extended to longer than 12 hours. For example, NGAL mRNA expression is increased tremendously at 18 hours after ischemia versus at 4 hours after ischemia with the same ischemic time.² Continuing exposure of the kidney to circulatory inflammatory cytokines and oxidants may increase tubular injury markers if sufficient time for reperfusion is allowed. Fourth, measurement of renal oxidative stress, such as the malondialdehyde level, can provide direct assessment of renal oxidative injury. However, renal malondialdehyde levels may be less sensitive.

Conclusions

In the present study, we showed that transfusion of SRBCs before HIRI potentiates ATI without trig-gering renal inflammatory responses. Exacerbation of hepatic injury may induce ATI via renal oxidant stress.

References:

1. Lee HT, Park SW, Kim M, D'Agati VD. Acute kidney injury after hepatic ischemia and reperfusion injury in mice. Lab Invest. 2009;89(2):196-208.
   CrossRef - PubMed
   
2. Zager RA, Johnson AC, Frostad KB. Acute hepatic ischemic-reperfusion injury induces a renal cortical “stress response,” renal “cytoresistance,” and an endotoxin hyperresponsive state. Am J Physiol Renal Physiol. 2014;307(7):F856-F868.
   CrossRef - PubMed
   
3. Jochmans I, Meurisse N, Neyrinck A, Verhaegen M, Monbaliu D, Pirenne J. Hepatic ischemia/reperfusion injury associates with acute kidney injury in liver transplantation: Prospective cohort study. Liver Transpl. 2017;23(5):634-644.
   CrossRef - PubMed
   
4. Slankamenac K, Breitenstein S, Held U, Beck-Schimmer B, Puhan MA, Clavien PA. Development and validation of a prediction score for postoperative acute renal failure following liver resection. Ann Surg. 2009;250(5):720-728.
   CrossRef - PubMed
   
5. Tomozawa A, Ishikawa S, Shiota N, Cholvisudhi P, Makita K. Perioperative risk factors for acute kidney injury after liver resection surgery: an historical cohort study. Can J Anaesth. 2015;62(7):753-761.
   CrossRef - PubMed
   
6. Wang Y, Li Q, Ma T, et al. Transfusion of older red blood cells increases the risk of acute kidney injury after orthotopic liver transplantation: a propensity score analysis. Anesth Analg. 2018;127(1):202-209.
   CrossRef - PubMed
   
7. Martines AM, Masereeuw R, Tjalsma H, Hoenderop JG, Wetzels JF, Swinkels DW. Iron metabolism in the pathogenesis of iron-induced kidney injury. Nat Rev Nephrol. 2013;9(7):385-398.
   CrossRef - PubMed
   
8. Deuel JW, Schaer CA, Boretti FS, et al. Hemoglobinuria-related acute kidney injury is driven by intrarenal oxidative reactions triggering a heme toxicity response. Cell Death Dis. 2016;7:e2064.
   CrossRef - PubMed
   
9. Choi EK, Baek J, Park S, et al. Preischemic transfusion of old packed RBCs exacerbates early-phase warm hepatic ischemia reperfusion injury in rats. J Surg Res. 2018;222:26-33.
   CrossRef - PubMed
   
10. Zhen J, Lu H, Wang XQ, Vaziri ND, Zhou XJ. Upregulation of endothelial and inducible nitric oxide synthase expression by reactive oxygen species. Am J Hypertens. 2008;21(1):28-34.
    CrossRef - PubMed
    
11. Miura M, Fu X, Zhang QW, Remick DG, Fairchild RL. Neutralization of Gro alpha and macrophage inflammatory protein-2 attenuates renal ischemia/reperfusion injury. Am J Pathol. 2001;159(6):2137-2145.
    CrossRef - PubMed
    
12. Tanaka Y, Maher JM, Chen C, Klaassen CD. Hepatic ischemia-reperfusion induces renal heme oxygenase-1 via NF-E2-related factor 2 in rats and mice. Mol Pharmacol. 2007;71(3):817-825.
    CrossRef - PubMed
    
13. Ferenbach D, Hughes J. Macrophages and dendritic cells: what is the difference? Kidney Int. 2008;74(1):5-7.
    CrossRef - PubMed
    
14. Behrends M, Hirose R, Park YH, et al. Remote renal injury following partial hepatic ischemia/reperfusion injury in rats. J Gastrointest Surg. 2008;12(3):490-495.
    CrossRef - PubMed
    
15. Vincent JL, Lelubre C. Preoperative transfusions to limit the deleterious effects of blood transfusions. Anesthesiology. 2012;116(3):513-514.
    CrossRef - PubMed
    
16. Hod EA, Zhang N, Sokol SA, et al. Transfusion of red blood cells after prolonged storage produces harmful effects that are mediated by iron and inflammation. Blood. 2010;115(21):4284-4292.
    CrossRef - PubMed
    
17. Nath KA. Heme oxygenase-1: a provenance for cytoprotective pathways in the kidney and other tissues. Kidney Int. 2006;70(3):432-443.
    CrossRef - PubMed
    
18. Srinivasan G, Aitken JD, Zhang B, et al. Lipocalin 2 deficiency dysregulates iron homeostasis and exacerbates endotoxin-induced sepsis. J Immunol. 2012;189(4):1911-1919.
    CrossRef - PubMed
    
19. Roudkenar MH, Halabian R, Bahmani P, Roushandeh AM, Kuwahara Y, Fukumoto M. Neutrophil gelatinase-associated lipocalin: a new antioxidant that exerts its cytoprotective effect independent on Heme Oxygenase-1. Free Radic Res. 2011;45(7):810-819.
    CrossRef - PubMed
    
20. Pavlakou P, Liakopoulos V, Eleftheriadis T, Mitsis M, Dounousi E. Oxidative stress and acute kidney injury in critical illness: pathophysiologic mechanisms-biomarkers-interventions, and future perspectives. Oxid Med Cell Longev. 2017;2017:6193694.
    CrossRef - PubMed
    
21. Fergusson DA, Hebert P, Hogan DL, et al. Effect of fresh red blood cell transfusions on clinical outcomes in premature, very low-birth-weight infants: the ARIPI randomized trial. JAMA. 2012;308(14):1443-1451.
    CrossRef - PubMed
    
22. Lacroix J, Hebert PC, Fergusson DA, et al. Age of transfused blood in critically ill adults. N Engl J Med. 2015;372(15):1410-1418.
    CrossRef - PubMed