Normothermic Machine Perfusion Reconstitutes Porcine Kidney Tissue Metabolism But Induces an Inflammatory Response, Which Is Reduced by Complement C5 Inhibition

Abstract

Normothermic machine perfusion (NMP) is a clinical strategy to reduce renal ischemia-reperfusion injury (IRI). Optimal NMP should restore metabolism and minimize IRI induced inflammatory responses. Microdialysis was used to evaluate renal metabolism. This study aimed to assess the effect of complement inhibition on NMP induced inflammatory responses. Twenty-two pig kidneys underwent 18 h of static cold storage (SCS) followed by 4 h of NMP using a closed-circuit system. Kidneys were randomized to receive a C5-inhibitor or placebo during SCS and NMP. Perfusion resulted in rapidly stabilized renal flow, low renal resistance, and urine production. During SCS, tissue microdialysate levels of glucose and pyruvate decreased significantly, whereas glycerol increased (p < 0.001). In the first hour of NMP, glucose and pyruvate increased while glycerol decreased (p < 0.001). After 4 h, all metabolites had returned to baseline. Inflammatory markers C3a, soluble C5b-9, TNF, IL-6, IL-1β, IL-8, and IL-10 increased significantly during NMP in perfusate and kidney tissue. C5-inhibition significantly decreased perfusate and urine soluble C5b-9 (p < 0.001; p = 0.002, respectively), and tissue IL-1β (p = 0.049), but did not alter other inflammatory markers. Microdialysis can accurately monitor the effect of NMP on renal metabolism. Closed-circuit NMP induces inflammation, which appeared partly complement-mediated. Targeting additional immune inhibitors should be the next step.

Keywords: inflammation; ischemia-reperfusion injury; microdialysis; normothermic machine perfusion; renal metabolism.

FIGURE 1

Closed-circuit normothermic machine perfusion. Graphic illustration of the different components of the closed-circuit normothermic machine perfusion model. Blood was driven in a sinusoidal manner at a fixed rate of 60 oscillations per minute by a centrifugal pump. Pump speed was adjusted by setting a mean arterial pressure target in the software. Blood was warmed to 39°C, oxygenated, and cleared for CO2 in an oxygenator prior to entering the kidney through an arterial cannula. At the venous side, treatment, whole blood, Ringer’s solution and urine were infused and pushed into a bubble trap to prevent perfusion of air bubbles from entering the circuit. Perfusate sampling ports were placed before and after the kidney chamber, and the microdialysis syringe pump was placed beside the organ chamber, allowing sampling throughout the perfusion period.

FIGURE 2

Perfusion characteristics. Arterial renal blood flow (A) and renal resistance (B) over a 240 min period of normothermic machine perfusion. Data are presented as median ± IQR. General mixed model analyses.

FIGURE 3

Renal function and injury. The renal function and injury markers in the control and the C5 inhibited group were compared over a 240 min period of normothermic machine perfusion. NGAL levels in perfusate and NGAL levels in urine (A, B), excretion rates of protein in urine (C) and oxygen consumption creatinine clearance and urine production (D–F). Data are presented as median ± IQR. Generalized mixed model analyses. NGAL, neutrophil gelatinase-associated lipocalin; NMP, normothermic machine perfusion.

FIGURE 4

Renal tissue metabolism. Glucose, pyruvate and glycerol (A–C) were measured in the renal microdialysate during SCS and NMP. Data are presented as median ± IQR. General mixed model analyses. SCS, static cold storage; NMP, normothermic machine perfusion; BL, in vivo baseline measurements (mean ± 2x sd).

FIGURE 5

Effect of C5 complement inhibition on complement activation. The complement markers in the control and the C5 inhibited group were compared during and after a 240 min period of normothermic machine perfusion. sC5b-9 levels and C3a levels in the perfusate (A, B), sC5b-9 and C3a levels in medulla and cortex tissue (C, D), sC5b-9 and C3a levels in the urine (E, F) and urine sC5b-9-to-proteinuria ratio (G). Data are presented as median ± IQR. General mixed model analyses, Wilcoxon signed rank test and Mann-Whitney-U test. * = p < 0.05, ** = p < 0.01, *** = p < 0.001. BL, in vivo baseline measurements; CAU, complement arbitrary units; NMP, normothermic machine perfusion.

FIGURE 6

Effect of C5 complement inhibition on cytokine levels in renal tissue. The complement markers in the control and the C5 inhibited group were compared after a 240 min period of normothermic machine perfusion. IL-1β, IL-6, IL-8, TNF and IL-10 cytokine concentrations measured in medulla and cortex tissue (A–E). Data are presented as median ± IQR. Wilcoxon signed rank test and Mann-Whitney-U test. * = p < 0.05. IL, interleukin; TNF, tumor necrosis factor.

FIGURE 7

Histology. Loss of glomerular basement membrane integrity depicted by the arrow (A) and protein casts in the lumen of tubules depicted by the arrows (B) assessed after a 240 min period of normothermic machine perfusion in PAS-stained biopsies. No differences were observed between the control and the C5 inhibited group.