Refining murine heterotopic heart transplantation: A model to study ischemia and reperfusion injury in donation after circulatory death hearts

Abstract Heart transplantation is a lifesaving procedure, which is limited by the availability of donor hearts. Using hearts from donors after circulatory death, which have sustained global ischemia, requires thorough studies on reliable and reproducible models that developing researchers may not have mastered. By combining the most recent literature and our recommendations based on observations and trials and errors, the methods here detail a sound in vivo heterotopic heart transplantation model for rats in which protective interventions on the ischemic heart can be studied, and thus allowing the scientific community to advance organ preservation research. Knowledge gathered from reproducible animal models allow for successful translation to clinical studies.


| INTRODUC TI ON
Donation after brain death (DBD) donors are the primary source of donor hearts for transplantation. 1 Donation after circulatory death (DCD) donors represent an important but underutilized source of additional hearts for transplantation. 1 The main distinctions between DBD and DCD donors are (a) DCD donors do not have a complete loss of brain function, (b) due to the untreatable and grave nature of illness withdrawal of life support is agreed by the patient or a legal representative to pronounce the person dead, (c) warm ischemia of varying duration based on the cardiorespiratory reserve of donor to all organs including heart is inevitable, and (d) the amount of permanent damage to donor organs and their potential for recovery of function is unknown. Principle reasons for underutilization of DCD hearts are ischemic damage and uncertain viability of heart. 2 Developing a successful animal model to study the modulation of ischemia and reperfusion (I/R) injury is critical in gathering valuable information to promote DCD heart transplantation. A sound knowledge gathered from animal models would allow for successful designing and testing interventions that would pave the way for clinical studies. Survival rates after heterotopic heart transplantation in rats were extremely variable and poor prior to the seminal publication of murine heterotopic heart transplantation technique in 1969 by Oto and Lindsey. 3,4 They first described the proper method of heart procurement from donor rats and performed endto-side vascular anastomosis in the recipients. Subsequent to their publication, improvements were made in the choice and delivery of anesthesia (inhalation of isoflurane as anesthetic agent), 5 limiting dissection of lumbar vessels (to prevent spinal cord ischemia), 6 and the use of curved vascular clamps to obtain control of infrarenal aorta and vena cava. 6 These improvements resulted in shorter operative times, decreased complication rates (bleeding, paraplegia), and higher survival rates. 5,6 We successfully incorporated the improvements mentioned above for murine DBD heterotopic heart transplantation and observed reproducible survival outcomes. However, when we attempted murine DCD heterotopic heart transplantation, we encountered setbacks that challenged us to examine and modify our surgical techniques. For example, unlike DBD hearts, DCD hearts remain stiff, noncompliant, and require more space in the recipient abdomen during vascular anastomosis. In addition, since antemortem cardioplegia cannot be administered per simulated clinical DCD protocol a novel approach was necessary to deliver cardioplegia post-mortem. Incorporating several modifications as we gathered experience, we were able to perform successful DCD heterotopic heart transplantations. Many of the details we gathered during the learning process were not described in the published literature on murine heterotopic heart transplantation. Our objectives were to describe in detail the challenges encountered in performing DCD heterotopic heart transplantation, possible reasons for failures, and describe the modifications we adopted to overcome the challenges.
Here we describe the characteristics of the DCD process related to donor heart and hemodynamic changes associated with the surgery in the recipient rat. We consider these observations pivotal for successful DCD heterotopic transplantation experimental outcomes. It is our sincere desire that the reader of this manuscript will benefit from our experiences and be able to establish a successful DCD heart transplantation setup with reproducible survival outcomes.

| ME THODS
All experimental animals were cared in accordance with institutional guidelines and the Guide for the Care and Use of Laboratory

| Induction of DCD process
The donation after circulatory death was induced in rats sedated in a 3% isoflurane chamber then anesthetized with ketamine/ xylazine (100/10 mg/kg intramuscular). The rats received intravenous injection of heparin (1000 units/kg), were intubated and connected to a ventilator (1 mL/kg at 90 rpm) and received an intravenous injection of vecuronium bromide (4 mg/kg). After one minute, the ventilator was disconnected and the DCD process initiated.
2.1.2 | Effects of duration of DCD-associated ischemia on ex vivo heart function recovery Duration of ischemia is a critical component of the DCD damage, and myocardial injury is proportionate to the duration of ischemia.
To determine the maximum duration of ischemia for DCD murine hearts with reversible damage, we examined the loss of myocardial function in rat hearts with 20, 25, and 30 minutes of warm ischemia by reanimating the hearts on a Langendorff system in preparation for standardizing procedures for heterotopic transplantation ( Figure 1). We noticed that with a shorter DCD ischemic duration (15 minutes), the rat hearts did not show a significant decrease in myocardial contractile function or oxidative phosphorylation in isolated mitochondria. The loss of myocardial function at 25 minutes of ischemia was between 40% and 45% of the control rat hearts, which improved with time on the Langendorff perfusion. With an ischemia time of 30 minutes, the loss of myocardial function was over 50%, and further deteriorated on the Langendorff perfusion system with time. We, therefore, identified 25 minutes of ischemia in rats as the maximum length of time that results in significant but reversible injury to the hearts to model DCD transplantation conditions.
To further determine maximum warm ischemic time, cardiac activity was followed using echocardiography and continuous electrocardiography, with two electrodes placed in the upper limbs and one in the hind limb after general anesthesia. A cannula was placed within the carotid artery for continuous blood pressure monitoring and serial blood gas values. Within 2 minutes of withdrawal from the ventilator, systolic blood pressure would fall below 50 mmHg with a blood oxygen saturation less than 70%. Within 10 minutes of withdrawal from the ventilator, electric asystole was observed which marks the beginning of the 5-minute standoff period required after asystole prior to initiating procurement.
For heterotopic heart transplantation, it takes approximately 5-7 minutes to procure the heart from the donor rat and administer cold cardioplegia. To achieve the target of 25 minutes of ischemia, we propose to start the procurement of the rat heart 18 minutes following the termination of ventilation, to allow a maximum time of 25 minutes of warm ischemia which includes standoff time. We suggest that the start of procurement can be changed based on the experience of the person performing the procurement.

| Effects of donor age and weight on ex vivo reanimation
Young rats tolerate ischemia to myocardium better than older rat hearts. It is our observation that rats older than 16 weeks are likely to sustain irreversible damage with 25 minutes of DCD ischemia. To maintain a homogeneity based on our limited observations with different ages and weights of rats, we suggest conducting DCD studies in rats of ages between 8 and 16 weeks weighing less than 400 g.

| Effect of gender
Differences based on gender were not studied in our experiments.
We only used male rats for our studies, and our results need to be interpreted with this limitation. We recommend defining the maximum tolerability and the appropriate age/weight groups to use in experiments designed to use female rats as DCD heart donors.

| Measurement of hemodynamic parameters
As a first step to better understand the reasons for lack of recipient survival after DCD heart transplantation, we chose to monitor F I G U R E 1 Mitochondrial function (A) and myocardial function (B) of donation after circulatory death (DCD) hearts with 20, 25, and 30 min of warm ischemia compared to control beating donor (CBD) hearts. A significant decline is seen in both oxidative phosphorylation and calcium retention capacity (CRC) in subsarcolemmal (SSM) and interfibrillar mitochondria (IFM). While heart rate (HR) remain comparable between control and DCD hearts with variable periods of ischemia, a steady and significant decline was observed in left ventricle developed pressure (LVDP) and rate pressure product (RPP = HR × LVDP). MPTP = mitochondrial permeability transition pore, Sample size CBD n = 8, DCD 20 min n = 5, DCD 25 min n = 8, DCD 30 min n = 4 the hemodynamic changes occurring in the recipient rats (n = 4) at the time of surgery and immediately following reperfusion of the transplanted heart. To accomplish this, we cannulated the carotid artery with a polyethylene tube (0.8 mm external diameter) filled with normal saline (0.9% NaCl) and connected to a pressure probe and a power lab station (AD Instruments, Denver, CO) to transduce continuous blood pressure recording. We used the same carotid access to collect timed blood samples corresponding to the steps of heterotopic heart transplantation and define the impact of bleeding, lactic acidosis, hyperkalemia, hypocalcemia, and hypoxia. These parameters were measured using 0.2 mL of blood with the ABL-800 blood gas analyzer (Radiometer, Copenhagen, Denmark). Core temperature was carefully maintained between 37°C and 38°C using a water-heated pump with a therapy pad (Adroit Medical Systems, Loudon, TN). The following observations were made from the above-described monitoring protocol.
1. Systolic blood pressure increased by 20 mmHg from baseline after clamping the abdominal aorta and vena cava. With ongoing surgery, it returned to baseline in 5-7 minutes.
2. Systolic blood pressure decreased by 20-30 mmHg upon releasing the aortic and vena cava clamp following completion of graft anastomosis. In the absence of bleeding from anastomosis sites, blood pressure improved in less than 5 minutes and attained baseline status in ten minutes.
3. Hematocrit at the start of transplantation was between 33% and 36%. It remained steady during the surgery, but after releasing the clamp on the aorta and vena cava, the hematocrit dropped by 7-10 points. If hemostasis was not attained in the first five minutes, the hematocrit continued to drop below 20%. Which invariably led to the demise of the recipient rat. 4. Electrolytes, and in particular potassium levels, remained steady during the procedure and immediately after reperfusion of transplanted heart. Hyperkalemia was not detected in any of the four rats. 5. Lactic acid levels remained normal and steady during the procedure and immediately after the reperfusion of transplanted heart. 6. Oxygen content in the arterial blood samples remained over 150 mmHg with 3% isoflurane inhalation anesthesia for induction and 2% for maintenance. The gas anesthesia mix was 1.5%-3% isoflurane and the rest (98.5%-97%) oxygen. 7. Carbon dioxide levels continue to increase from 40 mmHg at the baseline to over 60 mmHg by the end of surgery. There was a corresponding respiratory acidosis but no metabolic acidosis.

| Examination for vascular anastomosis patency and thromboembolism
Twenty-four hours following transplantation, we euthanized the rats and then performed dissection of the aorta and inferior vena cava to look for patency of vascular anastomosis and recipient pulmonary artery to look for evidence of thromboembolism; the vascular anastomoses were patent and no clots were found in the recipient pulmonary artery. Following the anti-coagulation protocol, we recommend that we describe to prevent any thrombus formation in the anastomosis.
From the observations listed above and modifying the techniques that were described in the published literature, we describe here the factors that lead to a successful heterotopic DCD heart transplantation (Tables 1 and 2).

| Donor preparation
An isoflurane chamber was used for sedating the rats, and then ketamine/xylazine (100/10 mg/kg) was administered intramuscularly prior to shaving the fur of the neck and chest. The skin was cleansed with sterile alcohol wipes and povidone solution. Heparin (1000 U/ kg) was administered intravenously via tail vein with a 25G needle.

| Cardioplegia requirement and delivery in a DBD heart vs DCD heart
The literature on rat heart transplantation routinely mentions the need for cardioplegia. 5,6,8 Delivery of cardioplegia to a DBD heart protects the myocardium from ischemia and prevents the heart from stiffening during storage. It also washes off the remaining blood from the coronaries. Cardioplegia is most commonly administered through cannulation of the inferior vena cava (IVC) while the heart is still beating and occasionally delivered directly into the aorta with a 25G needle before the procurement of the heart. These delivery methods are non-physiologic or carry a serious risk of damage to the ascending aorta. Since there is no ongoing circulation in DCD donors, the above-described cardioplegia delivery methods are not practical. Direct access to the ascending aorta is possible in DCD hearts, but a significant risk of damage to the aortic valve or ascending aorta, which negatively affects the heart transplantation procedure, precludes its practice. In our laboratory, we developed a method of delivering cardioplegia to a donor's heart via right carotid artery cannulation, using a polyethylene tube with an external diameter of 0.8 mm, which has been effective in our practice. The following is a detailed description of our approach.

| Donor preparation and induction of the DCD process
Aseptic approach was practiced with donor and recipient animals routinely. Key instruments used are further described in Figure 2.     This needs to be cleaned with micro-iris scissors and blunt tip forceps, taking care not to damage either of the main blood vessels or remove too much of the adventitia from the ascending aorta. The atrial appendage tissue tends to bunch into this suture. Using a cotton-tip swab to gently pull the heart down toward abdomen while tying the pulmonary veins. This will keep the atrial appendages away from the tie. Once the tie is secured, divide the pulmonary veins with micro-iris scissors and collect the heart from the donor rat.
9. Immediately place the heart in the cold saline solution at 4°C. A well-protected heart will be soft at touch, pale from having all the blood flushed from the coronaries, and cold. The DBD hearts are soft, while DCD hearts tend to be firmer.

F I G U R E 3
Cannulation of the right common carotid artery and endotracheal intubation: Under general anesthesia, the anterior neck was exposed, strap muscles of the neck were retracted to expose trachea (A) in the midline and common carotid artery (B) to its left (as seen in the picture). The trachea is exposed on a curved iris forceps (C) and opened anteriorly with fine scissors (D) between the tracheal rings (red line). A 14-gage angiocath attached to the ventilator is inserted into the trachea and secured with a 4-0 silk tie. is facing upward so the operator is visualizing the anterior surface, and the donor aorta is slightly lower than the pulmonary artery ( Figure 9). We prefer the oblique orientation. With a horizontal orientation, we notice excess tension on the pulmonary artery anastomosis. With oblique orientation, the apex of the donor heart will be pointing to the right of the operator, and pulmonary artery anastomosis is relatively higher to the aortic anastomosis, resulting in less tension on the pulmonary artery anastomosis.
11. Prepare to perform anastomoses with 8-0 monofilament suture on a tapered 4mm needle loaded on a Castroviejo needle holder with fine tips and the lock removed to prevent accidental trauma to tissues when locking and unlocking. The pointed tip needle driver and precision tweezers with high precision tips are preferred instruments for microvascular anastomosis.
12. We recommend first placing a stay suture at 6 o'clock position on aortic anastomosis to provide for a symmetrical and hemostatic suture line. Once a stay suture is placed then we start the anastomosis at 12 o'clock position with outside-to-inside on the donor aorta and inside-to-outside on the recipient aorta ( Figure 9). Since this suture carries the most tension, a double square knot, also referred to as a surgeon's knot, is preferred. 13. Once the aortic anastomosis is done, free the pulmonary artery from the aorta if needed and orient it correctly.
14. Before starting pulmonary artery anastomosis, inject 3mL of subcutaneous (nape of the neck) normal saline in anticipation of a significant drop in afterload once the vascular clamp is released. Also, decrease isoflurane concentration from 3% to 2.5%, then to 2% as the pulmonary artery anastomosis is near completion. This is to prevent respiratory depression and CO 2 retention. Monitor the respiratory pattern of the rat and the ad- 19. Unclamp the vessels, and add more Surgicel ® over the needle holes as needed. Cover with small surgical gauze. The transplanted heart will start beating with occasional fibrillations before resuming rhythmic breathing. As long as the heart is beating and there is no overt bleeding, leave the surgical gauze in place for 5 minutes.

| D ISCUSS I ON
The first successful heterotopic heart transplantation (HTx) in a rat  Table 1).
The anesthetic used by the original description of rodent HTx was pentobarbital sodium. Pentobarbital is a potent anesthetic, but its duration of action is variable, requiring repeat administration and potential for cardiotoxicity to the recipient. Lately, its availability also is limited. An alternate anesthetic agent that was commonly used in combination is ketamine plus dexmedetomidine. Ketamine is a potent analgesic, but it has less sedative action requiring dexmedetomidine to supplement as a sedative agent. Experience with this combination resulted in inconsistent duration and depth of anesthesia with poor surgical results. 5 Isoflurane is a volatile anesthetic with rapid onset, deep anesthesia, and fast recovery. It is easy to administer and has been the choice of anesthetic in rodent surgeries.
It is administered at a concentration of 3%-5% for induction and between 1.5% and 2.5% for maintenance. In our pilot study of rat HTx, we noticed CO 2 retention with 3% maintenance for the duration of surgery. However, when we decreased the concentration to 1.5%-2% for the last few minutes of the procedure the CO 2 retention has not been an issue, and the rats spontaneously resumed unassisted breathing at the end of the procedure. We recommend for DCD rat heart transplantation an induction with 3% isoflurane followed by a maintenance dose of 2.0%-2.5% until the last 5 minutes, where the isoflurane concentration can be decreased to between 1.5% and 2% depending on the respiratory pattern of the recipient rat.
A DCD donor heart, in general, is stiff and takes up more space in the abdomen of the recipient during anastomosis, thereby adding tension on suture lines, leading to a higher bleeding risk. Due to poor recovery of DCD heart function upon reperfusion, the blood from the coronary sinus pools into the right ventricle and if not propelled forward, will clot and prevent the forward flow of blood in coronaries leading to graft failure. Thus, special attention is required toward the preparation of both the donor heart and recipient in addition to the technique for a successful heterotopic DCD heart transplantation. We described in detail the steps to overcome the limitations set by the DCD heart.
We recommend the duration of global ischemia for DCD hearts to be 25 minutes or less, which correlates with other published reports 9 and clinical observations. 2 DCD-related ischemia is better tolerated in younger rats ages 2-3 months, 5 and as such, we recommend limiting the donor rat age to less than 16 weeks, with better results achieved with rats younger than 12 weeks. Without any cardioplegia, a rat DCD heart remains stiff and in asystole upon reperfusion. Delivery of cardioplegia with the cannulation of the common carotid artery and occlusion of the transverse aorta allows for accurate delivery of cardioplegia and better protection of the heart.  6 This clamp provides immediate isolation of infra-renal abdominal aorta and IVC without the need for dissecting them or separating the lumbar vessels. We were able to complete the vascular anastomosis consistently in 30 minutes or less by utilizing this clamp.
Microvascular instruments with pointed ends are essential for rat heterotopic heart transplantation. We adopted the instrument set as described by MacDonald et al, 10 which included pointed forceps, pointed end needle driver and angled microvascular scissors.
We used 8-0 monofilament suture for both aortic and pulmonary artery anastomosis. A short tapered needle (4 mm, 3/8 of a circle) is preferred over a longer needle.
Hemostasis is critical to the survival of the recipient, and we ad- During the pulmonary artery anastomosis, we recommend to decrease the isoflurane concentration to 2% or less in preparation for waking up from anesthesia. In addition, we routinely administer 3 mL of normal saline (0.9% NaCl) subcutaneously over the nape of the neck to account for insensible losses and potential blood loss from the suture line. We recommend administering long-acting analgesic (eg, buprenorphine) to allow adequate time for release and desired effect.
Before releasing the vascular occlusion clamp, we advocate applying small amounts of Surgicel ® as a hemostatic agent to prevent needle hole bleeding. We recommend placing the rat in a pre-heated cage on a warm pad at 37-38°C to prevent post-operative hypothermia.
We incorporated most of the advancements that are described in the literature for successful heterotopic rat heart transplantation and added several steps that we learned from our own experience to achieve over 90% survival in recipients following the above detailed techniques. The result is improved success with DCD heterotopic heart transplantation in rats.

ACK N OWLED G M ENTS
Gratitude is expressed for the American Heart Association SDG (award number 16SDG31080002) and Veterans Administration Merit Review grants awarded to Mohammed Quader (Grant ID CARA-015-17S, Award No. I01 BX003859), and for funds from the Pauley Heart Center to Mohammed Quader and Stefano Toldo.

CO N FLI C T O F I NTE R E S T
The authors have no conflicts of interest to declare that compromise the quality of this article.