Pediatric Animal Model of Extracorporeal Cardiopulmonary Resuscitation After Prolonged Circulatory Arrest

Summary

This protocol describes a neonatal porcine model of cardiopulmonary bypass (CPB), with circulatory and cardiac arrest as a tool for studying severe brain damage and other complications secondary to CPB.

Abstract

Congenital heart disease (CHD) is the most prevalent congenital malformation, with about one million births impacted worldwide per year. Comprehensive investigation of this disease requires appropriate and validated animal models. Piglets are commonly used for translational research due to their analogous anatomy and physiology. This work aimed to describe and validate a neonatal piglet model of cardiopulmonary bypass (CPB) with circulatory and cardiac arrest (CA) as a tool for studying severe brain damage and other complications of cardiac surgery. In addition to including a list of materials, this work provides a roadmap for other investigators to plan and execute this protocol. After experienced practitioners performed several trials, the representative results of the model demonstrated a 92% success rate, with failures attributed to small piglet size and variant vessel anatomy. Furthermore, the model allowed practitioners to select from a wide variety of experimental conditions, including varying times in CA, temperature alterations, and pharmacologic interventions. In summary, this method uses materials readily available in most hospital settings, is reliable and reproducible, and can be widely employed to enhance translational research in children undergoing heart surgery.

Introduction

Congenital heart disease (CHD) is the most prevalent congenital malformation, with about one million births impacted worldwide per year¹. Though modern advances in cardiothoracic surgery (CTS) and intensive care treatment have improved mortality rates, comorbidities remain extremely common^2,3,4,5. Neurodevelopmental abnormalities, including cognitive and motor impairments as well as learning disabilities, are reported in around 25%-50% of these patients^6,7,8. Surgery during the first days of life, especially those that require circulatory and cardiac arrest (CA), has been demonstrated to increase morbidity⁹. Hemodynamic alterations during surgery may have an important effect on the vulnerable developing newborn brain. Experimental models are essential to better understand the origin of these abnormalities and investigate neuroprotective strategies to improve the prognoses of these patients.

The use of animal models to study this population has been widely documented^{5,10,11,12,13,14}. Notably, piglets offer an excellent option, given close approximations in cardiac anatomy (Figure 1), genome, and physiology, as well as their relatively larger size in comparison to other animal models¹⁵ (Figure 2). The use of piglet models to study the effects of both cardiopulmonary bypass (CPB) and CA has been previously described. These experimental animal models are useful for studying hemodynamic changes and associated end-tissue organ complications^{14,16,17,18,19,20}. These models were developed to allow researchers to study human conditions in a controlled setting, with flexibility for a variety of experimental conditions. Most studies report the use of central cannulation, a technique that demands advanced surgical skills, requires higher resource utilization, and makes it difficult to ensure long-term survival. Though previous studies have documented the use of piglets in studying CPB^12,15, few have proposed the peripheral cannulation technique.

This new peripheral cannulation technique is easier, less aggressive, and more feasible when compared to other published studies¹⁹. Moreover, validating this technique in newborns and small animals is novel and should be considered for use by all researchers interested in using an animal model to study CHD and its associated comorbidities. It is particularly appropriate for individuals with access to a laboratory equipped with supplies, resources, and personnel experienced in conducting animal model experiments.

In summary, the main aim of this study is to describe and validate a neonatal piglet model of CPB with CA. The protocol aims to study severe brain damage and other possible complications of CPB surgery in a controlled setting with varying experimental conditions. This method provides a model that is generalizable, reliable, and of high quality, which can be used for a wide variety of experimental protocols.

Protocol

The present procedure was approved by the Animal Experimentation Ethics Committee (CEEA) of the Comparative Medicine and Bioimage Centre of Catalonia (CEEA-CMCiB). The Government of Catalonia also authorized the experimental protocol (no. 11652), file identification number FUE-2022-02381434 and ID QBXQ3RY3J. Experienced practitioners, including certified veterinarians providing supervision and assistance, performed all experimentation. Piglets (Sus scrofa domestica), 4-6 days old, weighing 2.5-3.5 kg, were used for the present study. An attempt was made to balance gender distribution to avoid related biases.

1. Sedation, intubation, and access

1. Initiate sedation through a snout mask with ketamine (20 mg/kg), midazolam (0.3 mg/kg), dexmedetomidine (0.02 mg/kg), and buprenorphine (0.01 mg/kg), followed by propofol (0.5 mg/kg) (see Table of Materials).
2. Position the piglet supine. Perform orotracheal intubation using a 2.5 mm cuffed endotracheal tube (see Table of Materials), using direct visualization of the trachea.
   1. Confirm appropriate endotracheal tube placement via direct auscultation of the lung bases.
3. Set the mechanical ventilation to deliver a respiratory rate of 30 breaths per minute, a tidal volume of 8-12 mL/kg, and an end-expiratory pressure of 4 cmH₂O.
4. Continuously monitor the depth of anesthesia during the protocol via heart rate (HR), blood pressure (BP), and oxygen saturation (spO2). Adjust ventilatory and sedation parameters as necessary.
   NOTE: The ideal values for vitals are a HR of 130-160 bpm, BP of 75-95/60-70, and an spO2 > 85.
5. Maintain sedation with 1.5% sevoflurane and fentanyl (25-200 µg/kg/min) (see Table of Materials).
6. Use direct visualization to place catheters in the femoral artery (3 Fr) and vein (4 Fr) (see Table of Materials).
   ​NOTE: These catheters will be used for medication administration and sample acquisition. As such, it is important to maintain access.

2. CPB circuit setup and priming

1. Customize and set up the CPB circuit following the steps below (Figure 3):
   1. Shorten the tubing as much as possible, while still allowing enough distance to reach the animal from the machine.
   2. Create and attach a tubing bridge that connects the outflow from the membrane oxygenator (see Table of Materials) to the inflow into the pump.
      NOTE: The bridge is vital to allow blood to continue circulating through the machine while the CAs of the animal are being performed.
2. Once all the connection points are sealed, prime the circuit with 300 mL of a heparin-saline solution (1,000 UI of heparin mixed into 1 L of saline) and 300 mL of fresh donor pig blood, followed by 3.5 mEq sodium bicarbonate, 350 UI of heparin, and 450 mg of calcium gluconate (see Table of Materials).
   1. "Sweep" the circuit by running the blood, heparin-saline, sodium bicarbonate, and calcium gluconate mixture through the entire circuit for 2 min at a rate of 0.3 L/min.

3. Surgery and CPB initiation

NOTE: Supplementary Figure 1 depicts the surgical materials required for cannula placement.

1. Expose the left internal jugular vein and the right carotid artery to prepare for cannulation (Figure 4).
2. For cannulation, use the Seldinger or "over-the-wire" technique²¹.
   1. First, insert a needle catheter into the left internal jugular vein. Once a flash of blood is visualized, carefully insert a guide wire into the vessel and remove the needle, ensuring the wire stays in place.
   2. Thread a dilator over the wire and into the vessel, then remove the dilator.
   3. With the wire still in place, thread an 8 Fr venous cannula and slowly advance it ~4 cm into the vessel. Carefully remove the wire, ensuring the cannula remains in place.
   4. Repeat the Seldinger wire technique with dilation to place a 6 Fr pediatric arterial cannula (see Table of Materials) into the right carotid artery.
   5. At the time of arterial cannulation, administer a bolus of intravenous heparin (50 IU/kg) via the newly placed arterial cannula.
3. Once access is achieved, securely fix both cannulas to the animal using 3-0 poly absorbable sutures and tape to prevent inadvertent removal (Figure 5).
4. Connect the cannulas to the CPB circuit, ensuring saline with heparin is added to connection points to prevent air in the circuit.
5. Set the initial flow to 80-85 mL/kg/min and slowly increase it to an ultimate flow rate of 150 mL/kg/min.
   ​NOTE: The animal can remain on the CPB for as long as the experiments require. A schema of steps 1-3 is depicted in Supplementary Figure 2.

4. Circulatory and cardiac arrest (CA)

1. To induce CA, administer 9 mEq of KCl. Use vitals to assess complete arrest and confirm with echocardiography. Administer additional KCl as necessary.
2. Once the heart is stopped, isolate the animal from the circuit to induce circulatory arrest.
3. Maintain CPB circuit flow using the previously described bridge (step 3.5) circulating at 1,500 rpm.

5. Extracorporeal cardiopulmonary resuscitation (eCPR)

1. Once the appropriate CA condition (0 min, 30 min, or 60 min) has been achieved, begin eCPR resuscitation.
2. Reconnect the piglet to the CPB circuit.
3. Administer 3 mL of calcium gluconate (2.25 mmol/10 mL, diluted 1:2) and 6 mL of sodium bicarbonate (1 M, diluted 1:2) via peripheral arterial access, adding doses as necessary.
   ​NOTE: Cardioversion or inotropic drugs (adrenaline or dopamine) may be used if necessary.

6. Postoperative care

1. Once resuscitated, monitor vitals for 15 min to ensure stability.
   NOTE: Ideal parameters in the intensive care unit period are: HR of 100-150 bpm, BP of 75-95/60-70, and spO2 > 85.
2. Transfer the animals for magnetic resonance imaging (MRI).
   NOTE: In this study, after imaging, the animals were euthanized for histological study of the brain lesions¹⁵. A timeline of the experimental portion of the procedure, including the sample collection strategy, can be viewed in Supplementary Figure 3.

Representative Results

During a 6 month period, the complete protocol was performed 12 times by an interdisciplinary team of pediatric critical care physicians, pediatric cardiologists, veterinarians, and technicians (Supplementary Figure 2 and Supplementary Figure 3).

Figure 1 and Figure 2 demonstrate the expected anatomy of the animals used in this protocol. The included piglets were an average of 4.8 days old (4-6 days) and on average weighed 2,814 g (2,400-3,140 g). A total of 42% were female. All of them were successfully sedated and intubated. The mean time from sedation to cannulation was 3.2 h (1.75-3.72 h). Vessel cannulation and CPB with CA were successfully achieved in 92% of the experiments. The one failure occurred during cannulation, likely secondary to the small piglet size and variant vessel anatomy (Table 1). The mean time from KCl administration to arrest was 2 min (1-6 min), and the mean time from eCPR initiation to heartbeat recovery was 12 min (3-28 min). These results provide guidelines for the ideal characteristics of animals to be used in this protocol and offer practitioners an estimate of experimental success rates. They also provide expectations for the time needed for both arrest and eCPR medications to take effect. Figure 3, Figure 4, and Figure 5 show key steps in the protocol that must be followed to achieve these results.

MRIs taken after the experiments demonstrated clear signs of early brain damage in the CPB with CA groups, not visualized in the control animals. The acute diffusion tensor imaging (DTI) modality uses magnetic field gradients to detect diffusivity parameters of water molecules as they undergo diffusion in biological tissues. These diffusivities can be impacted by acute stroke. As such, signs of infarction are better identified in DTI than other conventional MRI sequences. Representative MRI results from an animal after 30 min of CA with CPB can be viewed in Figure 6.

Figure 1: Human versus piglet cardiac anatomy. A schematic comparison of the internal features of the human heart and the piglet heart, exemplifying the close similarities of the features of each organ. Please click here to view a larger version of this figure.

Figure 2: Piglet anatomy. Visualization of key piglet anatomical structures, demonstrating the close approximation to human anatomy and supporting the selection of the species in the investigation of congenital heart defects. The right carotid artery and left internal jugular vein are bolded to indicate sites of cannulation. Please click here to view a larger version of this figure.

Figure 3: Cardiopulmonary bypass (CPB) setup. Labeled image of a complete CPB setup, including blood from an adult pig donor that is added to the circuit, the pump to drive the circuit, the membrane oxygenator, tubing secured at each point, and the bridge, which is highlighted in bright red. When the circuit is disconnected from the animal during CA, this bridge allows blood to continuously pass from the membrane oxygenator directly back to the pump to prevent clotting of the circuit. Please click here to view a larger version of this figure.

Figure 4: Exposure of the vessels. Bilateral exposure of both vessels to be cannulated. The right carotid artery and the left internal jugular vein are clearly visualized. Both vessels are marked with sutures prior to cannulation with the Seldinger or "over-the-wire" technique.²¹ Please click here to view a larger version of this figure.

Figure 5: Cannula fixation. An example of the fixation required to ensure cannulas remain securely in place throughout the experiment. Bilateral cannulas are fixed with both sutures and tape to prevent inadvertent removal. Please click here to view a larger version of this figure.

Figure 6: Preliminary post-surgical brain MRI results. A moderate acute infarct, identified by yellow arrows, is well-visualized in the right brain hemisphere and best depicted inDTI. Please click here to view a larger version of this figure.

Experiment	Weight (g)	Sex	Success?
1	2746	Female	+
2	2400	Male	+
3	2970	Male	+
4	2710	Male	+
5	2600	Male	+
6	2600	Male	-
7	2990	Male	+
8	2950	Female	+
9	2762	Female	+
10	3140	Female	+
11	2900	Male	+
12	2960	Female	+
SUMMARY	2814 (SD 213)	58% M, 42% F	92%

Table 1: Summary of experimental results. Table outlining all CPB attempts and results, including both failures and successes.

Supplementary Figure 1: Surgical materials required for cannula placement. Sample surgical tray containing tools required for the protocol, including (clockwise starting from the top left) gauze, sutures, guide wire, needle catheter, normal saline, 20 mL syringes, basin for disposal of liquids, needle forceps, Adson forceps, surgical scissors, mosquito forceps, and an 8 Fr pediatric venous cannula. Please click here to download this File.

Supplementary Figure 2: Schematic of sections 1-3 of the protocol. Diagram of key surgical steps with accompanying images. Please click here to download this File.

Supplementary Figure 3: Experimental timeline. Timeline of the experimental protocol, with temperature in blue and mean arterial pressure (MAP) in black on the y-axis, and time on the x-axis. The timing of neuromonitoring is denoted in purple, and sample collection timing suggestions are marked in orange. While the time spent in CA will vary depending on the experimental conditions, this schematic serves as a general guideline for the protocol. Please click here to download this File.

Discussion

Cardiopulmonary bypass is commonly used during cardiac surgery for adults, children, and neonates. It relies on a motorized extracorporeal circuit and membrane oxygenator that work together to oxygenate blood and provide pulmonary and cardiac stabilization. Previous studies have demonstrated that CPB may adversely impact many organ systems (renal, cerebral, pulmonary, cardiac, gastrointestinal) both in ill and formerly healthy patients^22,23,24. This is likely driven by a variety of factors, including the loss of arterial pulse pressure, blood cell exposure to foreign material, and/or regional hypoperfusion secondary to cannulation²⁵.

Several animal models of CPB have been proposed, including rats, mice, rabbits, goats, dogs, and pigs^26,27,28,29. Pig models have been validated as models for studying cardiovascular disease, cardiac physiology, transplantation, and the clinical use of CPB, given their anatomic and physiological similarities to humans. However, these models present several problems, including the friability of certain organs, anesthesia difficulties, ventricular fibrillation, and edema^30,31. These challenges are exacerbated if the animal is small, driven by technical problems associated with accessing and working with small structures. Of course, pediatric and neonatological patients differ greatly from their adult counterparts, so it is vital to have age-specific models. For this reason, neonatal models of CPB have been used to study a variety of congenital pathologies^16,17. Though the model presented here is currently being used to study neurological damage and different neuroprotection strategies, it may be adapted to study other organ systems.

Sedation and anesthesia were accomplished using medications appropriate to the animal size and species by a team with experience working with porcine models. Endotracheal intubation was performed by direct visualization of the trachea. Though the animals were small, this step did not pose great difficulty, as, particularly compared to other animal models, piglet airways are large enough to place a 2.5 mm endotracheal tube. Sedation and intubation were successful in all trials, demonstrating the efficacy of this strategy.

The animals used in this model all weighed approximately 3 kg, meriting the use of a neonatal extracorporeal circuit with the Quadrox oxygenator. According to the protocol, the circuit is customized to include a bridge that connects the oxygenator to the pump, and its length is reduced to minimize the volume of distribution while still providing enough tubing to reach the subject. The circuit was primed with adult pig donor blood, and the animals were premedicated with dexamethasone to prevent transfusion reactions. Although crystalloid and other colloid solutions have been used with this model³², adult pig blood was selected to maintain stable hemoglobin levels, particularly given the small size of the animals.

Smooth, efficient vascular access is of paramount importance for the success of the model, particularly in experiments aiming to assess functional postoperative outcomes. Most procedures described in other articles are based on central cannulation³³. Given the friability of cardiac tissue and access complications using the central cannulation method, this protocol instead described a peripheral cannulation technique. Specifically, the left internal jugular vein and the right carotid artery were cannulated. Though other alternatives were considered, the Seldinger or "over-the-wire" technique²¹ was selected due to superior results. The cannulas with optimal flow were 8 Fr and 6 Fr, for venous and arterial access, respectively. Larger cannulas did not improve the flow significantly and presented increased challenges during insertion. In some animals, as an alternative, an 8 Fr hemodialysis catheter was used for venous cannulation. In this protocol, the venous side was cannulated first, followed by the arterial side. Echocardiography was used to confirm the location of the cannulas post-placement.

Arrest, via KCl, did not occur immediately in this method, likely because the medication was not distributed directly into the coronary vessels, as occurs during typical neonatal cardiac surgery. Instead, the time from medication administration to arrest was 2 min on average, with some animals requiring an additional dose of KCl to reach complete arrest. A complete arrest must be confirmed with echocardiography to ensure appropriate experimental conditions. Similarly, eCPR was not immediate. The time from resuscitation medication administration to heartbeat recovery was 12 min, with some animals requiring additional doses, transfusions, and cardioversion. This portion of the protocol requires close attention from experienced practitioners to assess and respond to each animal's specific needs. One should ensure that additional doses of medications are readily available along with fresh adult pig donor blood, and vitals should be continued to be monitored for 15 min after heartbeat recovery.

One of the greatest challenges encountered during early trials was the clotting of the circuit. To combat this, intravenous heparin was used for anticoagulation. The first heparin dose was administered once the arterial cannula was placed. To prevent thrombosis of arterial and venous indwelling catheters during CA, a 1 mL/h infusion of heparin solution remained continuously infusing both cannulas. Similarly, to prevent clotting of the circuit during CA, the extracorporeal circuit remained circulating through the bridge at a set rate of 1,500 rpm. If these steps are followed, clotting is significantly less likely.

In conclusion, this model is generalizable, reliable, and uses materials readily available in most hospital settings. Cannulation and CPB induction, while requiring skill and precision, is possible, reproducible, and successful up to 92% of the time with experienced practitioners. While similar models have been described in the literature, few have described the peripheral cannulation technique, and none are accompanied by supplementary video instructions. Providing video instructions to accompany this complex protocol may 1) improve the success rates of those already using this strategy, 2) encourage previously hesitant investigators to attempt it, and 3) standardize the protocol to provide improved inter-study comparisons. The use of animal models to study children with CHD is vital to further identify mechanisms of damage and strategies for mediation and treatment. In the future, this model may be used not only to study neurological damage, but all comorbidities associated with cardiac surgery, to improve the lives and prognoses of children with CHD.

Materials

Name	Company	Catalog Number	Comments
1.5% sevofluorane	Zoetis	20070289
2.5 mm endotracheal tube	Henry Schein	988-1782
3 Fr catheter for peripheral arterial access	Prodimed	3872.1
4 Fr catheter for peripheral venous access	Prodimed	3872.13
6 French ECMO pediatric arterial cannula	Medtronic	77206
8 French ECMO pediatric venous cannula	Medtronic	68112
Adrenaline	B Braun	469801-1119
Adson forceps	Allgaier instruments	08-030-130	Any brand may be substituted
BP cuff	Mindray
Buprenorfine (0.01 mg/kg)	Richter Pharma	#9004114000537
Calcium gluconate (2.25 mmol/10 mL)	B Braun	570-12606194-1119
Dexmedetomidine (0.5-2.0 µg/kg/min)	Orion farma	GTN 064321000017253
Dolethol	vetoquinol	#3605870004904
Dopamine	hikma	A044098010
Fentanyl (25-200 µg/kg/min)	Kern Pharma	756650.2H
Fresh donor pig blood Type O	Any
Heat Exchanger	Maquet Gmbh & Co	MCP70107.2130
Heparin (1350 UI)	ROVI	641641.1
Irwin retractor	Aesculap	BV104R	Any brand may be substituted
Ketamine (20 mg/kg)	Richter Pharma	#9004114000452
Lubricant	Any orotracheal lubricant
Midazolam (0.3 mg/kg)	Serra Pamies	619627.4
Mosquito forceps	Aesculap	BH109R	Any brand may be substituted
Needle forceps	Aesculap	BM016R	Any brand may be substituted
Normal saline (0.9%)	B Braun Fisiovet	5/469827/0610	Any brand may be substituted
Plastic clamps for tubing	Achim Schulz-Lauterbach	DBGM	Any brand may be substituted
Potassium chloride (9 mEq)	B Braun	3545156
Propofol (0.5 mg/kg)	Zoetis	579742.7
Quadrox Membrane Oxygenator	Maquet Gmbh & Co	BE-HMOSD 300000
Rectal thermometer	Any
RotaFlow Console ECMO system	Maquet Gmbh & Co	MCP00703177	Neonatal ECMO System
Scalpel	Aesculap	BB074R	Any brand may be substituted
Sodium bicarbonate (1 M)	Fresenius Kabi	634477.4 OH
Surgical scissors	Talmed Inox	112	Any brand may be substituted
Suture (3/0 poly absorbable)	B Braun Novosyn (R)	0068030N1	Any brand may be substituted