A reliable noninvasive approach for functional assessment of the donor heart during normothermic ex situ heart perfusion (NESP) is lacking. We herein describe a protocol for ex situ assessment of myocardial performance using the epicardial echocardiography and conductance catheter method.
Heart transplantation remains the gold standard treatment for advanced heart failure. However, the current critical organ shortage has resulted in the allocation of a growing number of donor hearts with extended criteria. These marginal grafts are associated with a high risk of primary graft failure and may benefit from ex situ perfusion before transplant. This technology allows for extended organ preservation using warm oxygenated blood perfusion with continuous metabolic monitoring. The only NESP device currently available for clinical practice perfuses the organ in an unloaded non-working state, which does not allow for functional assessment of the beating heart. We therefore developed an original platform of NESP in working mode conditions with adjustment of left ventricular preload and afterload. This protocol was applied in porcine hearts. Ex situ functional assessment of the heart was achieved with intracardiac conductance catheterization and surface echocardiography. Along with a description of the experimental protocol, we herein report the main results, as well as pearls and pitfalls associated with the acquisition of pressure-volume loops and myocardial power during NESP. Correlations between hemodynamic findings and ultrasound variables are of major interest, especially for further rehabilitation of donor hearts before transplantation. This protocol aims to improve the assessment of donor hearts to both increase the donor pool and reduce the incidence of primary graft failure.
Heart transplantation is the gold standard treatment for advanced heart failure, but is limited by current organ shortage1. A growing number of donor hearts with extended criteria (age >45 years, cardiovascular risk factors, prolonged low flow, acute left ventricular dysfunction secondary to catecholaminergic storm) are allocated with an increased risk of primary graft failure2. Moreover, hearts donated after controlled circulatory death (DCD) may be presented with myocardial injury secondary to prolonged warm ischemia3. Therefore, there is a need for a better assessment of these donor hearts before transplantation, especially to evaluate their eligibility for heart transplantation4,5.
Normothermic ex situ perfusion (NESP) preserves the beating heart using warm oxygenated blood. The only commercially available device for NESP preserves the heart in a non-working state (Langendorff mode). This approach was initially applied to expand the preservation of the graft beyond the critical 4 h period of cold ischemia6. Another major advantage of this technology is to provide continuous assessment of myocardial viability based on lactate concentration in the perfusate6. However, this biochemical assessment has never been correlated with post-transplant outcomes to date. Similarly, Langendorff mode for NESP does not allow for hemodynamic and functional evaluation of the heart prior to transplantation. Some authors have reported the potential benefit of intracardiac catheterization during NESP to predict myocardial recovery after transplantation7.
The present report aims to provide a reproducible methodology to evaluate donor heart performance during NESP. We modified the circuit to allow for working mode perfusion and, therefore, for the acquisition of noninvasive functional variables with epicardial echocardiography. Myocardial work index, a load-independent variable, was recorded using pressure-strain loops. We investigated the relationships between myocardial work and hemodynamic variables obtained from intracardiac conductance catheterization.
The present protocol was approved by the local ethics committee on animal experiments and by the Institutional Committee on Animal Welfare (APAFIS#30483-2021031811339219 v1, Animals Ethics Committee of the University of Paris Saclay, France). Animals were treated in accordance with the Guidelines for the Care and Use of Laboratory Animals developed by the National Institute of Health and with the Principles of Laboratory Animal Care developed by the National Society for Medical Research.
NOTE: Surgical procedures were performed under strict sterility using the same techniques used for a human being. Experimental procedures included large white piglets (45-60 kg) and were performed under general anesthesia.
1. Animal conditioning and anesthesia protocol
2. In situ hemodynamic and echocardiographic assessment of the heart
NOTE: Hemodynamic assessment is performed with a Swan Ganz Catheter, while baseline functional assessment of the heart is performed by transthoracic echocardiography.
3. Description and priming of the normothermic ex situ perfusion (NESP) machine
NOTE: A modified NESP module is used to alternatively perform Langendorff perfusion and working mode perfusion. Briefly, connect the aortic line of the circuit to a compliance chamber via a Y-connector. Add a pediatric oxygenator and a cardiotomy reservoir (70-80 cm height above the aortic connector of the module) to provide a left ventricle afterload of approximately 70 mmHg during the working mode. Connect another cardiotomy reservoir (7-10 cm height above the aortic connector of the module) to the main inflow line using a Y-connector to provide a left atrium preload of approximately 10 mmHg during the working mode (Figure 2). Coronary flow is assessed with a flow sensor connected to the pulmonary cannula. A centrifugal pump, a membrane oxygenator, and a heater-cooler machine are connected to the circuit (Figure 2). For solution descriptions, refer to Table 1.
4. Heart procurement and instrumentation for normothermic ex situ heart perfusion
5. Connection to the NESP machine and resuscitation of the heart
NOTE: Before instrumentation of the heart, ensure that the materials necessary for resuscitation are available next to the perfusion circuit, especially a defibrillator with internal probes and an external pacemaker with epicardial electrodes. Ensure that the pressure line is connected to the aortic line, and that output sensor is placed on the coronary flow line. The afterload line must be clamped, as well as the preload line of the working mode circuit.
6. Working mode procedure
NOTE: Efficient arteriovenous clearance of lactate is usually achieved within 30 min after initiation of Langendorff perfusion. Working mode can then be initiated by connecting the preload cannula to the preload reservoir (this line was previously clamped during Langendorff mode). Similarly, the afterload line is connected to the aortic line (Figure 2). Set the flow sensor on the afterload line to measure cardiac output.
7. Pressure-volume (PV) loop assessment with the conductance method
NOTE: All calibration steps must be performed in working mode.
8. Epicardial echocardiography assessment of the heart in a working state
We herein described a NESP protocol in a monoventricular working state, using a modified heart perfusion module usually employed in clinical practice for Langendorff perfusion of the donor heart before transplantation. This piglet model of NESP using the present custom module was developed in 2019. The modifications of the circuit were minor, as most of the perfusion circuit was re-used for experiments. The cap of the module provided a flexible and waterproof membrane to protect the heart during transportation. It also allowed surface echocardiography while it remained in a sterile environment. The recommended priming volume with mixed blood and priming solution is about 1200-1500 mL in clinical practice. In the present protocol, the priming volume was higher (2000 mL) because longer tubing and additional reservoirs were necessary for working mode perfusion. Therefore, such considerations required animals over 50 kg for a blood collection of >1500 mL.
Placement of the porcine heart in the perfusion module was different compared to previously reported models of NESP in working mode10,11. Indeed, most of them described hearts suspended by the aorta, above a blood collection chamber, in a vertical position. In this protocol, we used a commercially custom module and set the heart with the anterior side laid in the perfusion box in a slightly tilted position and the posterior side facing the operator. However, Hatami et al. suggested that the heart's position during NESP was an important factor for optimal myocardial perfusion12 and would be better than the hanging position.
The present protocol used six animals to perform experimental Langendorff mode (LM) for 30 min, followed by working mode (WM) perfusion for 2 h. Mean aortic pressure (MAP) and cardiac output (CO) were continuously monitored and recorded every 30 min. Cardiac power output (CPO) was calculated as follows: CO x MAP/451. Assessment of lactate concentration in the perfusate was performed every 30 min to ensure that myocardial extraction of lactate (MEL) was effective as evidence for myocardial viability during NESP. Hemodynamic assessment was performed as soon as possible at T0, T60, and T120 during WM perfusion. Metabolic and hemodynamic measurements during NESP are summarized in Table 2.
Considering hemodynamic assessment by cardiac catheterization, optimal PV loops were achieved with a conductance catheter placed through the left atrial roof, then crossing the mitral valve, with the pigtail placed in the apex of the left ventricle. The position of the conductance catheter was checked using epicardial echocardiography (Figure 5). The quality of PV loop signal may change depending on catheter position and interference with external pacing (Figure 7).
Functional assessment during working mode perfusion
Echocardiographic assessment during WM perfusion was performed in the custom setup used in this study and provided left ventricular ejection fraction (LVEF) assessment, global longitudinal strain (GLS), and myocardial work index (MWI) with reproducibility over the experiments. All three left ventricular views were obtained at any time point in all experiments (Figure 6). Mean LVEF, GLS, and MWI were 40.8 (± 11)%, -8.00 (± 2)%, and 652 (± 158) mmHg%, respectively. Conductance catheter measurements were performed during WM perfusion. Mean SW, maximum dP/dt, min dP/dt, end-systolic pressure-volume relationship (ESPVR), tau, and pre recruitable stroke work (PRSW) were respectively 877 (± 246) mmHg·mL, 1463 (± 385) mmHg/s, -1152 (± 383) mmHg/s, 5.13 (± 3.16), 79.4 (± 23) ms, and 63.4 (± 17.5) mmHg·mL during WM perfusion. Hemodynamic parameters assessed either by conductance catheter or by surface echocardiography during WM perfusion are summarized in Table 3 and Table 4.
A significant decrease in MWI was observed during WM perfusion over time in all experiments (Figure 8A), as well as cardiac output (Figure 8B) and other parameters related to ESPVR (Figure 8C). Global MWI was correlated with cardiac output measured by conductance catheter (r = 0.85, p < 0.001) (Figure 9).
Figure 1: Parasternal transthoracic echocardiography view of the aortic valve. The aortic valve and the ascending aorta are checked to ensure that there is no ascending aortic aneurysm and no significant aortic regurgitation above grade 2. The functional left ventricular ejection fraction is also assessed. Please click here to view a larger version of this figure.
Figure 2: Modified organ care system circuit for monoventricular working mode. (A) A compliance chamber is set on the afterload line to reproduce the vascular elasticity. A Y-connector is set in the principal arterial line to fill a reservoir at the height of 10 cm above the heart graft to provide a preload for the left atrium at 13-15 mmHg. Another Y-connector is placed on the principal arterial line before the aortic connector. (B) One of the branches of the Y-connector is connected to a 3/8 inch tubbing, connecting a pediatric oxygenator and a reservoir at the height of 70 cm to provide an afterload of the left ventricle of 60 mmHg Please click here to view a larger version of this figure.
Figure 3: Stable conductance signal provided by the pressure-volume conductance catheter. A stable signal of the pressure-volume loops recorded in the software is provided by a central position of the catheter inserted in the left ventricle through an 8 Fr sheath set into the left atrium. Please click here to view a larger version of this figure.
Figure 4: Progressive cross-clamping of the preload reservoir. The procedure of progressive occlusion of the tubbing from the preload reservoir and the left atrium provides a decrease in the volume injected into the left atrium. The pressure-volume loops are then recorded with the acquisition software. Please click here to view a larger version of this figure.
Figure 5: Transesophagus echographic probe position during the surface echocardiographic assessment of the heart graft during WM. (A) The probe is placed on the left atrium wall while the posterior face of the heart is facing the operator during NESP. (B) Such placement provides an echocardiographic view of the left atrium, the left ventricle, and the mitral valve. Please click here to view a larger version of this figure.
Figure 6: Left ventricular views obtained with TEE probe during NESP. The epicardial echocardiography using a transesophagus probe set on the posterior wall of the left atrium provides a two-chamber view of the left atrium and the left ventricle. Please click here to view a larger version of this figure.
Figure 7: Examples of poor conductance signal acquisition. (A) No-central positioned conductance catheter with signal disturbed by the movements of the ventricular septum. (B) Conductance signal disturbed by external pacing. Please click here to view a larger version of this figure.
Figure 8: Linear regression over time during WM perfusion. (A) Myocardial work index (MWI, mmHg%), (B) cardiac output (CO, mL.min-1), and (C) end-systolic pressure-volume relationship (ESPVR). Please click here to view a larger version of this figure.
Figure 9: Relationship between MWI and cardiac output during working mode perfusion. Correlation curve between the myocardial work index (mmHg %) and the cardiac output (mL·min-1) during ex situ heart perfusion in working mode. Please click here to view a larger version of this figure.
Priming solution | Maintenance solution | Adrenaline solution | Del Nido cardioplegia |
500 mL NaCl solution | 60 mg of Adenosine | 0.25 mg of adrenaline | 500 mg of Ringer solution |
150 mg of Magnesium | 40 mL of NaCl solution | 500 mL of Glucose 5% | 10 mL of KCl 10% |
250 mg of Methylprednisolone | (concentration: 1.5 mg/mL) | 3 mL of Xylocaïne 2% | |
1 g of Cefotaxime | 6 mL of Mannitol 20% | ||
6 mL of Sodium Bicarbonate 8.4% | |||
7 mL of Magnesium Sulfate 15% |
Table 1: Solution descriptions. The table provides the volumes and concentrations of constituents used to prepare the priming, maintenance, adrenaline, and Del Nido cardioplegia solutions used in this protocol. The Del Nido cardioplegia solution is used to achieve cardiac arrest along with myocardial protection during cold ischemic time. The priming solution is infused in the perfusion machine, along with the blood collected during the experimental protocol. The maintenance solution and the adrenaline solution are infused during the ex situ heart perfusion to maintain stable perfusion parameters.
T0 | T120 | |
Lactate concentration (mmol/L) | 2.4 (0.97–2.83) | 1.27 (0.36–2.48) |
Myocardial Extraction of Lactate (mmol/L) | 0.15 (0.14–0.19) | 0.08 (0.04–0.09) |
pH | 7.37 ( 7.31–7.45) | 7.41 (7.31–7.47) |
Potassium (mmol/L) | 4.6 ( 4.4–5.1) | 4.9 (4.3–5.5) |
Systolic aortic pressure (mmHg) | 132.5 (101.0–142.3) | 101.0 (96.2–109.3) |
Mean aortic pressure (mmHg) | 97.5 (73.0–106.8) | 77.0 (69.0–85.5) |
Coronary Flow (mL/min) | 925 (550–1050) | 700 (550–875) |
Cardiac Power Output | 326.5 (116.5–485.5) | 228.0 (185.5–361.0) |
Table 2: Hemodynamic and metabolic parameters assessed during WM perfusion. Data are provided with the median and interquartile range.
SW (mmHg·mL) | maximum dP/dt (mmHg/s) | min dP/dt (mmHg/s) | ESPVR | Tau (ms) | PRSW | |
Mean | 877 | 1463 | -1152 | 5.13 | 79.4 | 63.4 |
Median | 816 | 1423 | -1025 | 4.01 | 73.9 | 62.8 |
Standard deviation | 246 | 385 | 383 | 3.16 | 23.0 | 17.5 |
Minimum | 528 | 778 | -1856 | 2.19 | 52.0 | 40.0 |
Maximum | 1244 | 2119 | -755 | 13.8 | 134 | 101 |
Table 3: Mean and median values obtained by the conductance catheter method during WM perfusion. Abbreviations: ESPVR: end-systolic pressure-volume ratio; PRSW: pre-recruitable stroke work; SW: stroke work.
GLS (%) | LVEF (SB) | MWI | GCW | |
Mean | -8.04 | 40.8 | 652 | 936 |
Median | -8.00 | 37 | 642 | 919 |
Standard deviation | 2.03 | 11.0 | 158 | 208 |
Minimum | -11.5 | 27 | 389 | 579 |
Maximum | -5.00 | 59 | 898 | 1268 |
Table 4: Mean and median values obtained by surface echocardiography during WM perfusion. Abbreviations: GLS: global longitudinal strain; LVEF: left ventricular ejection fraction; MWI: myocardial work index; MWE: myocardial work efficiency; GCW: global constructive work.
There are some critical steps to consider in the NESP protocol. In situ preliminary assessment of the heart remained important, especially considering the aortic valve that should not present with significant aortic regurgitation (grade 2 or more); otherwise, the resuscitation of the heart will be compromised during the Langendorff period because of impaired coronary perfusion and myocardial ischemia. The initiation of the WM after Langendorff perfusion was a challenging maneuver, requiring at least two persons to regulate the filling of the preload reservoir, the pump flow, the pressure of the left atrium, and the aortic outflow line. This transition period was performed once metabolic effective myocardial extraction for lactate was achieved. During this period, the perfusion circuit could stop because of pump defusing related to major air embolism. The optimal placement of the ultrasound probe on the left atrial wall to obtain steady two- and three-chamber views was partially disturbed by the cumbersome cannulas and materials set around the heart. Echocardiographic data had to be recorded with a very steady ultrasound signal, with at least three contraction cycles.
Resuscitation of the heart during NESP is not explicitly reported in the literature. Only a few studies describe in detail the resuscitation procedure to initiate NESP13. Preliminary resuscitation approaches were developed in this protocol to achieve an optimal technique for resuscitation, including progressive reperfusion, by slowly increasing coronary flow and blood temperature (from room temperature to 37 °C). The main issue for ultrasound imaging was finding the optimal location for the probe on the left atrial roof. The position of the perfused heart, with its posterior wall facing the operator, allowed performing surface echocardiography without moving the heart and without the risk of aortic valve regurgitation. The presence of bubbles in the circuit altered imaging quality, and this problem must be avoided as much as possible. Optimization of the circuit was performed to reduce blood turbulences, especially considering blood drainage from the afterload reservoir to the main reservoir. A non-steady position of the conductance catheter into the left ventricle provided poor quality PV loop curves. The PV loop signal could however be significantly improved by introducing the catheter in the center of the left atrial posterior wall, through the center of the mitral valve, and positioned in the middle part of the left ventricle.
Loading left heart cavities is essential for ex situ echocardiographic evaluation. Even if the decline of the cardiac output has been previously described in other studies while the lactate trend remained stable, only a few articles described such consideration using a real monoventricular working mode perfusion11. Biventricular working mode perfusion was not performed in this model for technical reasons, because such a system is even more complex and cumbersome. However, the lack of working mode for the RV is questionable because of LV and RV interdependence, a confounding factor in LV assessment. The lack of right ventricular assessment may also be questionable since RV failure is a common complication after transplant, associated with high mortality. Potassium concentration constantly increased in the perfusate without the possibility of clearing it because no blood filtration membrane was included in our custom circuit. The main issue concerning this perfusion mode is the fact that the organ itself is isolated from the other organs that could regulate its metabolism and clear all the metabolites produced by the myocardial metabolism. Some authors have described a perfusion model that included a hemofiltration system to provide prolonged NESP in working mode14, with a significant decrease of myocardial edema at the end of perfusion, which certainly participates in the decline of the myocardial performances over time.
Myocardial hemodynamic and echocardiographic performances decreased in NESP during working mode in our experience, as well as cardiac hemodynamics recorded by conductance catheterization. This suggests that perfusion should not be considered as a preservative method for donor hearts before transplant. During WM, trends of biochemical were different compared to Langendorff mode. Myocardial extraction of lactate during WM was continuously effective, while hemodynamic performance decreased progressively. This finding suggests that the lactate trend may not be a relevant parameter for assessing the myocardial performance in WM, as previously observed in other studies15.
Functional assessment of the heart during NESP would be of great interest to clinicians. Invasive assessment methods (PV loop technique) present several limitations. Indeed, the conductance technique should be considered to carefully draw reliable results, because of the isolation of the heart graft without a physiological biological environment that usually conduces the electric signal along with the myocardium itself16. The decision to transplant marginal grafts preserved with NESP technology is currently based only on lactate trends17. We trust this approach could be easily applied to resolve this major issue before transplantation. It can provide both anatomical (valvular disease, myocardial thickness) and functional assessments of the donor heart. Echocardiographic assessment of the left ventricle was achieved in the preclinical model and allowed to obtain MWI, a load-independent parameter that was significantly correlated to cardiac output assessed by a conductance catheter. These preliminary results highlight the role of surface echocardiographic evaluation during NESP in a working mode.
The authors have nothing to disclose.
Georges Lopez Institute, Lissieu, 69380, France
Claudia Lacerda, General Electric Healthcare, Buc, France
3T Heater Cooler System | Liva Nova, Châtillon, France | IM-00727 A | Extracorporeal Heater Cooler device |
4-0 polypropylene suture | Peters, bobigny, France | 20S15B | sutures |
5-0 polypropylene suture | Peters, bobigny, France | 20S10B | sutures |
Adenosine | Efisciens BV, Rotterdam, Netherlands | 9088309 | Drugs for the ex-vivo perfusion |
Adrenaline | Aguettant, Lyon, France | 600040 | Drugs for the ex-vivo perfusion |
Atracurium | Pfizer Holding France, Paris, France | 582547 | Drugs for the induction of the anesthesia |
DeltaStream | Fresenius Medical Care, L’Arbresle, France | MEH2C4024 | Extracorporeal blood pump |
EKG epicardial electrodes | Cardinal Health LLC, Waukegan, Illinois, USA | 31050522 | EKG detection electrodes |
External pacemaker | Medtronic Inc. Minneapolis, Minneapolis, USA | 5392 | Pacemaker device |
Glucose 5% | B.Braun Melsungen AG, Melsungen, Germany | 3400891780017 | Drugs for the priming solution |
Heart Perfusion Set, Organ Care System | Transmedics, Andover, MA, USA | Ref#1200 | Normothermic ex-vivo heart perfusion device |
Intellivue MX550 | Philips Healthcare, Suresnes, France | NA | Permanent monitoring system |
Istat 1 | Abbott, Chicago, Ill, USA | 714336-03O | Blood Analyzer machine |
Labchart | AD Instruments Ltd, Paris, France | LabChart v8.1.21 | Pressure Volume loops aquisition software |
Magnesium | Aguettant, Lyon, France | 564 780-6 | Drugs for the cardioplegia |
Magnesium Sulfate | Aguettant, Lyon, France | 600111 | Drugs for the cardioplegia |
Mannitol 20% | Macopharma, Mouvoux, France | 3400891694567.00 | Drugs for the cardioplegia |
Methylprednisolone | Mylan S.A.S, Saint Priest, France | 400005623 | Drugs for the priming solution |
Millar Conductance Catheter | AD Instruments Ltd, Paris, France | Ventri-Cath 507 | Pressure Volume loops conductance catheter |
MWI software | General Electric Healthcare, Chicago, Ill, USA | NA | software used for the Ultrasound echocardiographic machine |
Orotracheal probe | Smiths medical ASD, Inc., Minneapolis, Minneapolis, USA | 100/199/070 | probe for the intubation during anesthesia |
Potassium chloride 10% | B.Braun Melsungen AG, Melsungen, Germany | 3400892691527.00 | Drugs for the cardioplegia |
Propofol | Zoetis France, Malakoff, France | 8083511 | Drugs for the induction of the anesthesia |
Quadrox-I small Adult Oxygenator | Getinge, Göteborg, Sweden | BE-HMO 50000 | Extracorporeal blood oxygenator |
Ringer solution | B.Braun Melsungen AG, Melsungen, Germany | DKE2323 | Drugs for the cardioplegia |
Sodium Bicarbonate | Laboratoire Renaudin, itxassou, France | 3701447 | Drugs for the cardioplegia |
Sodium chloride | Aguettant, Lyon, France | 606726 | Drugs for the priming solution |
Swan Ganz Catheter | Merit Medical, south jordan, utah, USA | 5041856 | Right pressure and cardiac output probe |
Tiletamine | Virbac France, Carros, France | 3597132126021.00 | Drugs for the induction of the anesthesia |
Transesophagus probe (3–8 MHz 6VT) | General Electric Healthcare, Chicago, Ill, USA | NA | Ultrasound echocardiographic transesophagus probe |
Vivid E95 ultraSound Machine | General Electric Healthcare, Chicago, Ill, USA | NA | Ultrasound echocardiographic machine |
Xylocaïne 2% | Aspen, Reuil-malmaison, France | 600550 | Drugs for the cardioplegia |
Zolazepam | Virbac France, Carros, France | 3597132126021.00 | Drugs for the induction of the anesthesia |