Options to improve the quality of kidney grafts from expanded criteria donors − experimental study

Overview

Introduction:
There is a worldwide discrepancy between the number of donors and the number of organs needed for transplantation, leading to certain expansion of criteria, resulting in acceptance of donor organs. This means that organs with worse parenchymal qualities may be harvested for transplantation. One possibility, how to ameliorate the quality of these organs, is to change the concept of their preservation during cold ischemia or even after sudden circulatory arrest, prior to the collection of the organ itself. The goal of our experimental study was to try to improve organ quality from these marginal donors.

Methods:
In the first part of our experimental study, we tested retrograde oxygen persufflation (ROP) in preservation of ischemically damaged kidney grafts. In ten animals (domestic pig), we compared standard intra-arterial perfusion of the grafts using cooled perfusion solution (N=5), with a retrograde oxygen persufflation method (N=5). The main criterion for evaluation was the histopathological analysis of renal parenchyma after transplantation of the kidney graft. In the second part of our experimental study, again using an animal model of an ischemic kidney (laboratory rabbit), we compared machine perfusion of the kidney graft with routinely used perfusion by hydrostatic pressure. For evaluation, we added another two criteria – the volume of perfusate that flowed through the graft and the temperature drop during perfusion.

Results:
In the first part of the study, we proved the ability of ROP to preserve and even ameliorate the quality of ischemically damaged kidney grafts. Results of histopathological analysis of samples taken during ROP were without statistically significant difference in comparison with those taken during standard intra-arterial perfusion. In the second part of the study, we observed a significant difference in maximal flow rate measured during perfusion of the kidney grafts, favoring mechanical perfusion over perfusion using hydrostatic pressure (p=0.004). The same situation was seen with the drop of temperature measured in the parenchyma of the grafts (p<0.001). Finally, histopathological evaluation of the renal parenchyma found better washing out of blood particles from the capillaries during mechanical perfusion (p=0.005).

Conclusion:
The presented results of our experimental studies establish that alternative methods of preservation during cold ischemia and before removal of kidney grafts from the donor may be beneficial for its function after transplantation. We believe that these methods may be suitable especially for so-called marginal grafts from extended criteria donors.

Key words:
transplantation − donors after circulatory death − controlled organ perfusion − in situ perfusion − retrograde oxygen persufflation

INTRODUCTION

Optimal treatment for patients with irreversible kidney failure is kidney transplantation. This procedure returns the patient, who in the majority of cases is already dependent on elimination methods, back to a normal quality of life. There is a worldwide discrepancy between the number of patients awaiting a kidney and the number of available organs. Even in the Czech Republic, the number of patients on the waiting list is increasing, while the number of kidney transplantations remains the same [1].  This discrepancy logically leads to longer waiting times. One way to avoid this problem is the concept of expanded criteria donors – ECD. This concept is based on accepting donors, who do not satisfy ideal criteria. These grafts, sometimes labeled marginal grafts, are naturally of poorer quality. The aim is to perform a high quality harvest of these grafts and recondition them so that the function of the graft after transplantation will be the best possible.  Marginal grafts include, for example, kidneys from donors with non-beating hearts, or donors after circulatory death – DCD, where the poorer quality may be due to logistically more complicated harvesting. The most frequent cause is polytrauma, extensive myocardial infarction or devastating craniocerebral injury not compatible with life. The individual DCD categories are divided according to the Maastricht classification [2]. Even in donors after brain death – DBD, where for some reason there was a longer interval of warm ischemia, the graft may still benefit from some form of reconditioning. The aim of experimental research in this field is not only to immediately best perfuse the graft, but also to store it under optimal conditions during cold ischemia.  A number of experimental studies worldwide focused on optimizing graft perfusion, either in-situ (in the organism of the donor), or ex-situ (outside of the organism). Other experiments studied various forms of graft restitution. Our transplantation center also performs experimental research in this field [3].

The aim of our experimental study is to try to improve the quality of kidney grafts from DCD donors and from ECD donors. According to the literature, these grafts have a higher frequency of delayed graft failure (DGF) as well as primary non-function (PNF) [4]. Our experimentally tested changes in the algorithm of graft harvesting and perfusion technique could lead to a higher graft quality and thus to a better onset of graft function. The long-term goal of our experimental study was to create a new usable model for graft harvesting, perfusion and even restitution of kidney grafts from the abovementioned marginal donors and therefore increase the number of possible transplantations and accommodate more patients awaiting kidney transplantation. All experiments were approved by the Ethics Committee of the Faculty of Medicine of Charles University in Pilsen.  

METHODS

1) Retrograde oxygen persufflation

One possibility for reconditioning explanted grafts is so-called retrograde oxygen persufflation (ROP). This method is based on retrograde perfusion by gaseous oxygen via the renal vein. ROP was first described in the 1970´s [5]. A number of experimental studies were performed in the past [6,7]. ROP was probably first used on kidneys in clinical practice by Rolles [8] in 1989. Repeated studies compared the quality of grafts preserved by ROP and standard preservation under cold temperatures or by intra-arterial perfusion. It was repeatedly confirmed that ROP may actually in many ways be more beneficial than standard preservation of both kidneys and liver [9,10]. The experiments were most often performed on small animals (rats, dogs). Other experiments using ROP usually studied the bioenergetics of the graft and their function after transplantation. Our experimental study aimed to verify the benefit of ROP on a large experimental animal (domestic pig).

Method

In group A (N=5 animals), the experimental animal (a pig weighing approx. 20 kg) was introduced into general anesthesia (Thiopental, Calypsol, Fentanyl i.v.) following premedication (Atropine, Stresnil i.m.), and via a middle laparotomy we accessed the retroperitoneum, where we dissected the hilum of the right kidney and the renal vascular pedicle was clamped for 20 min. to simulate warm ischemia. Subsequently, kidney explantation was performed. At this time, the first kidney biopsy sample was taken (a wedge excision of the kidney parenchyma including the cortex and medulla, biopsy - IA). Following explantation, the kidney graft was connected to an endosufflator, through which the kidney was retrogradely perfused by humidified gaseous oxygen with a perfusion pressure up to 18 mm mercury. During retrograde perfusion, the cannula was introduced into the renal vein. The graft was placed in a cooled perfusion solution (Custodiol) at a temperature of 4 ^oC, the temperature of the dialysate was maintained by a laboratory thermostat with cooling (Fig. 1). Using a needle, several perforations into the kidney parenchyme were made to release the gaseous oxygen. This perfusion lasted 60 minutes. At the end of this interval, the second biopsy sample from the kidney was collected (biopsy-IIA). The kidney graft was then again transplanted to an identical animal and a nephrectomy was performed on the opposite side (left-side). During transplantation, the renal artery was anastomosed to the aorta („end to side“anastomosis) and the renal vein to the inferior vena cava („end to side“ anastomosis). After 120 minutes, the graft was again explanted, a third kidney biopsy sample was collected (biopsy-IIIA) and the experimental animal was euthanized using a cardioplegic solution. This all took place under continued general anesthesia.

In group B (N=5 animals), renal ischemia was again simulated by clamping the renal vascular pedicle for 20 minutes, then following the nephrectomy the kidney graft was standardly intra-arterially perfused by identical perfusion solution using hydrostatic pressure.  The cannula was standardly introduced into the renal artery (Fig. 2). The dialysate itself was again cooled to a temperature of 4 °C. The washout was performed for 60 minutes similarly to Group A. The kidney was then autotransplanted in the same manner (Fig. 3) and a nephrectomy on the opposite side was performed. Biopsy samples from the kidney were collected in an identical manner (biopsy IB-IIIB). After 120 minutes, the animal was again euthanized by a cardioplegic solution.

Histopathological analysis of the biopsy samples from the graft was performed by pathologist, who evaluated the general degree of injury to the nephron (using the scale presented below). The rate of glomerular perfusion, the presence of microthrombi and the possible presence of acute tubular necrosis were evaluated (Figs. 4, 5). The study was blinded for the pathologist. 

Results

No technical difficulties with graft perfusion or with the transplantation itself were observed in any of the presented experimental animals. In two cases (one case in each group), worse segmental perfusion of the organ after occlusion of the hilar artery was observed.

Histopathological analysis of the biopsy samples:

Tab. 1 shows the quality of nephron perfusion as well as an overall evaluation of the quality of the parenchyma (presence of microthrombi, acute tubular necrosis)

Scale 1 – best to 4 − worst

Fisher´s exact test was used to express the quality of kidney perfusion at time intervals I-III.

The significance value according to Fisher´s exact test for time intervals I, II and even III was p=1.00, therefore greater than 0.05. Thus there is no statistically significant difference in the degree of kidney perfusion between groups A and B in any of the analyzed time intervals.

Interpretation of results

Our results demonstrate an identical quality of grafts preserved by both classical intra-arterial perfusion and retrograde oxygen persufflation. We did not discover significant differences in the histopathological analysis of the harvested grafts for each of the time intervals I, II and III. We did not confirm the opinion of some authors, that ROP may actually be superior compared to classical perfusion or to „cold storage“[9]. However, we did undoubtedly confirm that basic ROP can protect, even restore, the parenchyma [10]. We see the benefit of this method in the possible combination with classical perfusion techniques. Despite certain logistic difficulties, the combination of intra-arterial perfusion and ROP in clinical practice could be considered, especially in donors with ECD. It is necessary to perform further experimental and clinical studies. 

2) Pulsatile intra-arterial perfusion

Immediately initiated pulsatile intra-arterial perfusion – experiment on a small animal.

This part of the experimental study followed numerous published experiments and tried to simulate in-situ perfusion in non-beating heart donors [11,12]. In this method, kidney perfusion is initiated while the kidney is still in the donor´s body by way of catheter, which is most often introduced into the femoral artery. The aim was to try to establish an improved graft quality if machine perfusion is immediately initiated in situ compared to standard hydrostatic perfusion. Not many published experimental studies in the literature address this problem on an animal model. In our experimental study, we selected the laboratory rabbit as the animal model to verify our hypothesis. 

Method

Again we worked with 10 experimental animals, which were divided into two groups of five animals. In both groups, the animal was premedicated (Midazolam i.m.), then introduced into general anesthesia (Ketamine i.v.) A middle laparotomy was performed, the retroperitoneum accessed and the abdominal aorta, inferior vena cava and the vascular pedicle of the left kidney were dissected.  The renal hilum was clamped for 30 minutes, which simulated warm ischemia of the left kidney (Fig. 6). The aorta was then ligated above the origin of the left-sided renal artery and above the bifurcation, and a thin perfusion catheter was introduced into the aorta (Fig. 7). A catheter was introduced in a similar manner into the inferior vena cava to drain the perfusate from the kidney.

In the first group (group A, N=5), the graft was perfused standardly by hydrostatic pressure using a bag which was hung at a standard height above the operating table. In the second group (group B, N=5), the kidney was perfused by a mechanical perfusion system of our own construction, using defined pressure ratios.  The perfusion time was 30 minutes in both groups. The flow rate through the kidney and the organ temperature were measured using a detector introduced into the renal parenchyma. The animal was then euthanized by cardioplegic solution and the kidney was removed for histological examination (Fig. 8). The extent of kidney perfusion and degree of parenchymal damage were objectively evaluated by pathologist, based on a standardized scale.

Parameters evaluating the quality of graft preservation included temperature of the kidney parenchyma (and the speed of its decrease), perfusion volume of the perfusion solution and the quality of graft perfusion assessed by the pathologist. Similarly, histopathological examination of the graft was performed blinded by the pathologist, who evaluated the degree of nephron damage in general (using the same scale that was used in the previous experiment).  

Results

The evaluated parameters in the individual animals are described in Tab. 2; Tab. 3 shows their statistical analysis. 

Even in such a small set of animals, we clearly found a significant difference in maximal perfusion flow, where in the group of mechanically perfused animals the flow rate was higher than in those where only hydrostatic pressure was used (Graph 1) (p=0.004).

We also observed a statistically significant difference in temperature decrease of the graft in favor of pulsatile perfusion (Graph 2) ( p<0.001).

The qualitative histopathological evaluation of graft perfusion was also statistically significant, again in favor of pulsatile perfusion in Group B (p=0.005).

Interpretation of results:

The results of our work established statistically significantly faster cooling of kidney grafts during mechanical perfusion, with a subsequent effect on metabolic processes, as well as a statistically significantly higher flow of perfusion solution per minute. The organ is therefore perfused by a greater volume of perfusion solution during the same time interval.  The currently used method uses cooled perfusate, which, however, is already heated by the surrounding room temperature when it enters the body of the deceased and is then heated in the organism by body temperature. By attaching the cooling system into our perfuser immediately before entering the donor’s body, we were able to achieve very low perfusate temperatures (4^oC). This makes subsequent organ cooling more effective. Objective evaluation by the pathologist regarding the quality of graft perfusion also unanimously favors mechanical pulsatile perfusion. 

Histological examination confirmed that the quality of kidney perfusion using the mechanical perfuser in the body of the animal is significantly better than when using hydrostatic pressure.  No blood elements were present in the glomeruli of the kidneys perfused by perfuser. It seems that guided mechanical perfusion is able to remove even small, already created, microthrombi and thus maintain a maximal vascular network for restoring blood flow.  One danger may be the use of excessively high perfusion pressure, which may cause edema of the organ and its subsequent damage. Therefore the perfusion pressure must be carefully monitored and guided, which our software is fully able to do.  

RESULTS

Because there are several parts to the experiment, the results are always presented after the described method. 

DISCUSSION

The method which ensures the most physiological replacement of kidney function is its transplantation, which enables the patient to return to everyday life.  In our region, donations from living donors are infrequent. The primary source of organs for transplantation is from deceased donors with brain death. Longterm, the number of patients awaiting a kidney exceeds the number of available organs [13]. This has led to the establishment of a program of deceased donors after irreversible cardiac arrest. The two main subgroups of these DCDs are termed controlled and uncontrolled [14]. Criteria for this classification depend on whether cardiac arrest is sudden or expected. This is associated with the method of harvesting and subsequent kidney graft preservation. After irreversible cardiac arrest is declared and the non-touch interval has passed, organ harvesting for transplantation is initiated.  The  „in situ“  method of perfusion, using a catheter introduced into the abdominal aorta to perfuse the visceral segment with special solutions, which aim to prevent ion passage across the cell membrane and thus stabilize the internal milieu, is usually used in cases of uncontrolled donors. They also bring about hypothermia, which decreases the metabolic demands of the organ. Some centers use extracorporal membrane oxygenation (ECMO), either in the normothermic or hypothermic mode [15]. However, this method is expensive and requires special equipment. In controlled donors, it is possible to use the method of quick laparotomy with direct visceral perfusion via the abdominal aorta, as used during harvesting from DBDs. A second possibility is also initially perfusion using a three-way catheter, with subsequent laparotomy and kidney harvesting. Mechanical pulsatile perfusion is the golden standard for subsequent graft preservation after its removal from the donor in DCDs. „Cold storage“ may be used in controlled donors with a minimal duration of warm ischemia, where the grafts were harvested by method of quick laparotomy. Especially in uncontrolled donors, where there was a longer duration of warm ischemia, mechanical pulsatile perfusion is able to improve the parameters of the harvested graft and enables reconditioning [16]. Warm ischemia is specific for DCDs. It begins during ineffective blood circulation during the agonal phase in controlled donors, and in uncontrolled donors it begins once the systolic pressure drops below 70 Torrs during sudden cardiac arrest. During warm ischemia, pathophysiological changes occur, which are caused by cessation of the cellular metabolism in mitochondria due to insufficient oxygen and metabolic substrate supply necessary for oxidative processes ensuring the creation of ATP. This halts the activity of membrane ATPases. This results in the transfer of ions down their concentration gradient in both directions across the cell membrane. In a similar manner, intracellular calcium deposits are released, which have significant signalization consequences inside the cell.  This leads to edema and lysis of the cell.  The amount of strictly intracellular enzymes released may be measured during mechanical perfusion. Another evaluated factor is the level of renal vascular resistance. The lower the resistance exuded by the circulatory system of the graft, the greater the chance of a good functional result after transplantation.

Other possible approaches for preserving kidney grafts prior to transplantation are the use of perfusion with perfusate temperatures below the physiological norm, or temperatures equal to those in the human body. The aim is to enable a certain restitution of energy supplies to the organ prior to transplantation thus ensuring an earlier functional start. To date, these methods are mainly experimental, without greater use in clinical practice. A special method is retrograde oxygen persufflation [10]. This method consists of gaseous oxygen persufflated through a catheter via the renal vein into the harvested organ. Positive experimental results have led to the usage of this method in clinical practice, especially in the field of liver transplantation. Our own results also seem promising for possible use in the future. 

The DCD program is undoubtedly associated with a higher percentage of PNF occurrence, which is a reflection of the severity of warm ischemia with possible organ injury which took place during the agonal phase prior to harvesting.  The occurrence of PNF in DCDs ranges from 2–19 %, whilst in grafts harvested from DBDs the range is 1−9%. A more frequently discussed issue of DCD is DGF, which widely ranges from 39–77% [17].

Our aim in both experiments was to analyze the possibilities of perfusion of the kidney graft and try to find possible ways of improving the quality of the graft harvested from DCDs. 

The first way of reaching this goal was to compare ROP with kidney graft perfusion by hydrostatic pressure. The persufflator was constructed specifically for this experimental project. Technically, this is a very simple and inexpensive procedure.  We used histopathological examination to evaluate the quality of the graft (first after preservation, and then after subsequent transplantation). We used our own classification to evaluate the quality of the graft (1−4); this classification has been used long-term at out center (along with the standard Remuzzi classification) to evaluate grafts from DCDs [18]. We are aware of the obviously simplified view; however, our aim was to determine whether ROP may be a certain alternative for reconditioning the kidney graft. Results show an identical quality of grafts preserved by conventional intra-arterial perfusion, as well as by retrograde oxygen persufflation [10]. We did not find significant differences in the histopathological analysis of the collected grafts. The presented results are in accordance with the literature. We did not confirm the opinion of some authors, that ROP may actually be superior compared to classical perfusion or possibly with „cold storage“. We did, however, prove, without a doubt, that even simple ROP can protect, or even restore, the kidney parenchyma. Although ROP showed certain potential in reconditioning the kidney graft after warm ischemia, it seems more appropriate introduced into the algorithm of mechanical perfusion after graft collection. We asked ourselves, whether it would be possible to save more glomeruli and achieve faster cooling of the organ already in the body of the donor. An already somewhat used possibility is ECMO, which is however a very complicated and expensive method [19]. Thus in our experiment, we proposed in-situ washout of the visceral segment by introducing a catheter into the abdominal aorta, using hypothermic pulsatile perfusion. Hypothermia decreases the cells´ need for oxygen by slowing down enzymatic reactions.  A 50% reduction in metabolic activity is reportedly achieved at a cooling to 10^oC [20]. Although hypothermia partially increases renal vascular resistance, according to recent literary data, it prolongs the lifespan of the graft and reduces DGF even during longer intervals of cold ischemia [21]. The experimental animal in this phase of the study was a domestic rabbit, primarily due to easier logistics and manipulation.

The results of our original work on a small animal established a statistically significant improvement in kidney graft cooling during mechanical perfusion compared to perfusion using hydrostatic pressure. In-situ perfusion by hydrostatic pressure decreased the temperature by 3.5^oC on average, mechanical perfusion by 17.5^oC on average. The maximum flow of the perfusion solution through the organ per minute was also statistically significantly higher. In mechanical perfusion, the organ is perfused by a higher volume of perfusion solution during the same time interval. The routinely used method works with a cooled perfusion solution flowing from a bag in which it is stored at a temperature of approximately 8^oC. However, once it is introduced into the body of the dead donor, it is warmed by the surrounding room temperature in the set, through which it is administered. By attaching the cooling system of our perfuser directly before entry into the body of the donor, we were able to attain very low perfusate temperatures around 4^oC. This makes subsequent organ cooling more effective.  The objective evaluation by pathologist regarding the quality of graft perfusion also unambiguously supports mechanical pulsatile perfusion. Histological examination established that kidney perfusion using the mechanical perfuser in the body of the animal is significantly better than when hydrostatic pressure is used.  The control group also had significantly worse kidney perfusion during macroscopic observation. It seems that guided mechanical perfusion is able remove even small microthrombi from the kidney while maintaining a maximum vascular network for restoration of blood flow. We are aware of the limitations of this experiment, especially in the small number of animals and the measuring of only three basic parameters. This was, however, a pilot project, where the aim was to establish functionality of the model of in situ perfusion as well as the usability of the experimental perfuser. 

From the above-mentioned results, we believe that immediately initiated pulsatile perfusion of the grafts in the body of the donor after irreversible cardiac arrest should ensure better perfusion of the graft thus preserving a greater number of available glomerules and nephrons in the parenchyma, more effective cooling should lead to earlier cessation of already ongoing ischemic changes and also lead to a decrease in the incidence of PNF and DNF following subsequent transplantation. These conclusions mostly correspond to literary data [22]. The financial demands for implementing our proposed method would not be too great, commercial perfusers are easily available on the market and most transplantation centers already use them. In the next phases of the experiment, we would like to again use the experimental model of the domestic pig, this time evaluating long-term survival after autotransplantation with mechanical organ perfusion in a time interval of several weeks. A number of experimental studies will still need to be performed prior to possible implementation into practice; however, every positive result of this experimental study brings us closer to this goal.

CONCLUSION

The presented results of the experimental study are, in our opinion, promising in possibly improving the algorithm of harvesting and reconditioning organs from expanded criteria donors. Additional experimental and clinical studies will provide definite confirmation of the benefit of the aforementioned methods.

Supported by Charles University Research Fund

(Progres Q39).

Conflict of Interests

The authors declare that they have not conflict of interest in connection with the emergence of and that the article was not published in any other journal.

doc. MUDr. Jiří Moláček. Ph.D.

Department of Vascular Surgery

University Hospital in Pilsen

Faculty of Medicine in Pilsen Charles University

alej Svobody 80

304 60 Pilsen

e-mail: molacek@fnplzen.cz