Thermal Preconditioning During Ex-vivo Lung Perfusion for the Rehabilitation of Damaged Lung Grafts before Transplantation

Ex-vivo lung perfusion (EVLP) has represented a major advance in the field of lung transplantation. Its foremost impact lies in expanding the pool of available donor lung grafts by enabling the functional assessment of marginal lungs to allow the use of organs that would otherwise be discarded for transplantation^1,2. Marginal (or extended criteria) donor lungs refer to one or more of the following criteria: donor age more than 55, PaO₂/FiO₂ of less than 300, ischemia time of more than 6 h, positive sputum microbiology, abnormal radiography³. In addition to physiological evaluation, EVLP may serve as a platform for the application of various therapeutic interventions aimed at improving graft quality prior to transplantation (concept of therapeutic reconditioning)⁴. As such, EVLP provides the unique opportunity to rehabilitate damaged (marginal) lungs and reduce the risk of primary graft dysfunction (PGD), a major complication impairing both early and long-term outcomes after lung transplantation^5,6. PGD is the clinical expression of the process of ischemia-reperfusion injury (IRI), whose pathophysiology relies on oxidative stress, inflammation, the activation of the complement and coagulation cascades, and the induction of cell death processes^7,8,9. Ultimately, these mechanisms contribute to disrupting normal functions of the lung endothelium and epithelium, leading to non-cardiogenic pulmonary edema with reduced oxygenation capacity, the two key hallmarks of PGD^6,10,11,12.

A growing number of experimental studies, both preclinical and clinical, have demonstrated the efficacy of various ex-vivo therapeutic strategies to alleviate IRI during lung transplantation. Examples include pharmacological agents targeting oxidants and free radicals, cell death pathways, inflammatory mediators, as well as cellular-based and gene therapies, to name only a few^{13,14,15,16,17,18,19,20}. Beyond such therapeutic approaches, EVLP may also serve as a platform for the application of physiological stress directly to the lung graft, to elicit intrinsic protective mechanisms providing an adaptive tolerance to subsequent pathophysiological insults. One such strategy involves the application of transient, non-lethal heat stress to the lung graft during EVLP. Indeed, it has been known for a long time that conditions of heat stress promote an adaptive heat shock response, resulting from the expression of a series of heat shock proteins (HSPs) acting as molecular chaperones and conferring resistance to further exposure to various stressful stimuli²¹. Besides HSP induction, some additional cellular adaptations have been identified following heat exposure, including (a) the NF-E2 related factor 2 (NRF2) antioxidant and cytoprotective pathway, whose activation upon heat stress notably upregulates the cytoprotective gene heme oxygenase^22,23; (b) the unfolded protein response (UPR), dealing with proteotoxic stress in the endoplasmic reticulum (ER stress), and whose activation by heat may be either protective or detrimental, depending on the intensity of the thermal stress^24,25,26; (c) autophagy, a cytoprotective process promoting lysosomal degradation of damaged organelles and macromolecules, which can be activated or inhibited by heat in a temperature and time-dependent manner^27,28.

We have demonstrated that such "thermal preconditioning" provides major benefits in terms of reduced oxidative stress, inflammation, cell death, and endothelial dysfunction, leading to significant protection against lung IRI. Therefore, we proposed that thermal preconditioning may represent a novel, non-pharmacological method for ex-vivo therapeutic rehabilitation of damaged lung grafts prior to transplantation^29,30,31. Thermal preconditioning may offer several advantages over existing pharmacological and non-pharmacological strategies. By eliciting an integrated cellular endogenous response, it may simultaneously provide protection against distinct pathophysiological pathways pertaining to IRI, whereas other strategies generally target a single pathway. Further, it avoids the risk of potential side effects associated with the use of pharmacological compounds. The aim of the present article is to present a detailed step-by-step protocol of thermal preconditioning during ex-vivo lung perfusion of damaged rat lungs, which could be applied across diverse settings, notably to larger animal models or human lungs for research purposes.

We utilize the rat isolated perfused lung system IPL-2 for EVLP. The setup is described in Figure 1 and includes the following: Perfusate reservoir (Figure 1-1); two roller pumps, for perfusate circulation (Figure 1-2A) and perfusate sampling (Figure 1-2B); a thermostatic water bath (Figure 1-3) for the maintenance of perfusate temperature; a membrane gas exchanger (Figure 1-4) connected to a gas tank (not shown) containing a mixture of 6% O₂, 10% CO₂, and 84% N₂ to deoxygenate and carboxylate the inflow perfusate; two pressure sensors (Figure 1-5) to monitor left atrial (LA) and pulmonary artery (PA) pressures (the PA pressure sensor is not visible on the figure); a heated and sealed chamber (Figure 1-6) housing the heart-lung block with dedicated connections for PA, LA, and tracheal cannulas; a ventilator (Figure 1-7) connected to the tracheal cannula for lung ventilation and for the measurement of airway pressure and static lung compliance; an electronic thermometer (Figure 1-8) to display the perfusate temperature at the level of the PA inflow. A connection platform (Figure 1-9) whose elements are displayed in detail in Figure 2: Connection for the PA (Figure 2-1), the LA (Figure 2-2) and tracheal (Figure 2-3) cannulas; connection to a weight module (Figure 2-4) for the continuous monitoring of the weight of the heart-lung block; A temperature sensor (Figure 2-5) for the monitoring of the perfusate temperature at the PA inflow. A detailed schematic description of the circuit can be found elsewhere¹⁸.

NOTE: Adult male Sprague-Dawley rats were used for this study (n = 60; age, 8-11 weeks old; weight, 300-450 g). Animals were housed (2 rats/cage) in our local animal facility, maintained under a 12 h light/12 h dark cycle with unlimited access to food and water. All animals were treated in accordance with the "Guidelines for the Care and Use of Laboratory Animals" (NIH Publication no. 96-23), and all experiments were performed in compliance with the experimental protocol approved by our local animal care committee (Direction Générale de l'Agriculture, la Viticulture et Affaires Vétérinaires de l'Etat de Vaud (authorizations 3456a and 3456x1). The size and weight of the animals were selected on the basis of our previous studies in the rat EVLP model and for anatomical convenience (size of the lungs and vessels). The Sprague-Dawley strain was chosen, given that other strains (notably Wistar rats) may exhibit hypersensitivity reactions to the dextran solution present in the perfusate solution^1,32.

1. General description of the EVLP circuit

NOTE: The setup is described in Figure 1, and the elements of the connection platform (Figure 1-9) are displayed in detail in Figure 2. A detailed schematic description of the circuit has been described earlier¹⁸.

1. Use an acellular buffered extracellular solution for the ex-vivo perfusion. Ensure that the perfusate flows from the reservoir through the gas exchanger and is directed through the PA cannula for lung perfusion and then exits through the LA cannula back to the reservoir. Continuously monitor the inflow (PA) and outflow (LA) pressures.
2. Connect all transducers (PA pressure, LA pressure, lung weight, pump speed) to the Analog/Digital converter box (Figure 3). Open the Basic Data Acquisition Software (BDAS). Calibrate pressure signals using two-point calibration (0; 10 cm H₂O) using a water column. Calibrate the weight module using two-point calibration (0 g and 1 g). Calibrate pump speed using a 1 min perfusate sampling at 10 mL/min. Control the quality of the signals by applying known pressures and pump speed and assessing the adequacy of results (detailed protocol and video for signals calibration can be found in supplementary materials).
   NOTE: Detailed protocol for signals calibration can be found in supplementary materials (Supplementary Files 1 and 2).
3. Record all key parameters (PA, LA pressures, lung weight, and flow) in real time by the system software and observe their visual display on a personal computer.
   NOTE: Physiological signals recorded by the BDAS are exported to a text file using the File export command in the menu bar. Data extracted at selected time points (every 60 min) are transferred to a spreadsheet and then copy/paste to the software for statistical analysis.
4. Calculate the pulmonary vascular resistance (PVR) using equation (1):
         (1)
5. Obtain samples (1-2 mL) of the perfusate at dedicated time points for the measurements of various biological markers.

2. Thermal preconditioning application

1. Connect a thermostatic water bath (Figure 4-1) to ensure a constant, precise, and adjustable water temperature for perfusate heating. Use an integrated motor pump inside the water bath to circulate the heated water.
2. Use a double-walled glass container as the reservoir (Figure 4-2) to maintain the perfusate temperature by circulating heated water between the walls of the container.
3. Use a heating coil with serpentine tubing inside a heated water container (Figure 4-3) to drive the perfusate, permitting heat exchange from the water to the perfusate immediately before it enters the ex-vivo lung preparation.
4. House the heart-lung block in the organ chamber (Figure 4-4), a double-walled container filled with water circulating from the thermostatic water bath to maintain a warm environment.
5. For the temperature sensor and thermometer (Figure 2-5 and Figure 4-5), securely and firmly apply a Type K bead air/surface sensor against the inflow cannula for continuous monitoring of the actual perfusate temperature entering the pulmonary artery (PA). The temperature is displayed on a digital thermometer.
6. To maintain the temperature of the inflow at 37 °C, set the water bath at 38-38.5 °C.
   NOTE: To compensate for heat losses from the perfusate along its course in the tubing, we determined in our conditions that the water bath temperature had to be set at 38-38.5 °C to maintain 37 °C at the inflow. Depending on the heat losses from the perfusate, these settings may have to be adapted to guarantee the desired temperature at the inflow.
7. Initiate the ex-vivo heating protocol after 60 min of EVLP by increasing the water bath temperature up to 43 °C to reach an inflow temperature of 41.5 °C, which is obtained over 5 min and maintain over 30 min.
   NOTE: In our previous work²⁹, we determined the optimal temperature for thermal preconditioning (TP), using a range of temperatures from 40 °C to 43.5 °C and an EVLP protocol of 3 h, as depicted in Figure 5. The heating temperature providing optimal protection without any harmful effects (as summarized in Figure 5B) was found to be 41.5 °C. This temperature was therefore used in all our subsequent experiments, with EVLP durations up to 6 h^29,30,31.
8. After this transient heating, add ice to the water bath to rapidly (in 5 min) reduce the water bath temperature back to 38 °C to get the perfusate temperature in the inflow cannula back to 37 °C and maintain this temperature until the end of the EVLP protocol.
9. Determine physiological variables (lung compliance), edema formation, and molecular endpoints (expression of heat shock proteins, antioxidant enzymes, inflammatory cytokines, and biomarkers of apoptosis, autophagy, endoplasmic reticulum stress, and endothelial cell damage).

3. Detailed EVLP protocol for thermal preconditioning

1. Animal preparation
   1. Prepare the required materials, including cannulas and surgical instruments (Figure 6A,B and Table of Materials).
   2. Weigh the animal to calculate the appropriate dosages for all administered substances, perfusate flow, and ventilation parameters.
   3. Prepare the anesthetic mixture: 80 mg/kg ketamine and 8 mg/kg xylazine in the same syringe.
      NOTE: The anesthetic agents must be stored in a secure cabinet.
   4. Prepare heparin, 1.7 U/g of body weight in isotonic saline (1 mL syringe).
   5. Anesthetize the animal, using 5% isoflurane (vaporized in a dedicated induction chamber), followed by intraperitoneal injection of the prepared anesthetic mixture. Maintain the anesthesia with 1-2% isoflurane via a face mask.
      NOTE: It is important to maintain deep anesthesia to avoid gasping during exsanguination.
      Use a setup with an isoflurane extraction system to protect the operator.
      ​Discard all the needles and syringes in appropriate biohazard and sharp waste containers. Sealed sharp and biohazard containers are then disposed through designated waste streams.
   6. Shave the animal from the throat to the penile area and remove any remaining fur using damp, warm paper.
   7. Verify the depth of anesthesia by pinching the hind paw; increase isoflurane concentration if reflexes are present.
   8. Place the animal in a supine position and secure all four limbs using adhesive tape.
2. Extraction of the heart-lung block for EVLP
   NOTE: The procedure requires intensive training to perform the different steps in a timely, atraumatic, and highly reproducible manner. The most delicate step is pulmonary artery cannulation, which should be performed under magnification using a specular operative microscope for the best results. The complete procedure can be performed by a single well-trained investigator. To achieve such experimental independence, a minimum of 15 complete procedures under the supervision of an experienced investigator is required.
   1. Gently lift the skin at the midline of the throat and make a 5 cm longitudinal incision along the caudal to cephalic axis using small scissors.
   2. Separate the salivary gland lobes and carefully dissect the fascia over the sternohyoid muscles using curved forceps until the larynx and trachea are clearly exposed.
   3. Pass a 4-0 silk suture underneath the trachea and prepare a pretied knot, leaving the forceps in place under the trachea.
   4. Calculate the tidal volume (Vt) for in-vivo ventilation and lung procurement using the Stahl formula³³:
       × 
   5. Set the ventilator dedicated for in-vivo ventilation and lung procurement (see the Table of Materials) to the calculated tidal volume with a respiratory rate of 70 breaths/min, an inspired fraction of oxygen (FiO₂) of 0.5, and a positive end-expiratory pressure (PEEP) of 3 cm H₂O.
      NOTE: To set the PEEP at 3 cm H₂O, connect the expiratory arm of the ventilator to a platinum-cured silicone tube inserted 3 cm under the surface of a water-filled container.
   6. Create a transverse incision between two cartilaginous rings, insert a 2 mm outer diameter cannula (14 mm length, EVLP ventilation cannula, Figure 6B) for ventilation and secure it by tightening the suture. 
   7. Connect the cannula to the ventilator and confirm effective ventilation by the presence of bubbles in the water-filled container connected to the expiratory arm of the ventilator.
   8. Using blunt-tipped scissors, incise the skin along the abdomen and enter the peritoneal cavity, continue by incising the midline abdominal wall muscles up to the diaphragm.
   9. Gently move the intestines aside using cotton swabs to expose the inferior vena cava.
   10. Inject the prepared heparin solution into the vein and allow it to circulate for 5 min. Then, transect the vena cava and aorta for exsanguination.
   11. Once the animal is fully exsanguinated and confirmed dead, stop ventilation and disconnect the ventilator from the tracheal cannula to allow the lungs to deflate, marking the onset of warm ischemia. Keep the total warm ischemic time (WIT) to 1 h.
       NOTE: Although the reported median WIT in clinical DCD lung donation is 33 min, no detrimental effects of WIT durations of up to 60 min on 1 year survival have been reported. Therefore, most centers currently agree on 60-90 min as an acceptable upper limit of WIT³⁴. As detailed previously, the use of 1 h WIT in our rat model is therefore relevant to the clinical scenario³⁵.
   12. Fifteen minutes before the end of the total warm ischemic time of 1 h (i.e., after 45 min of warm ischemia), fill up a 60 mL syringe (fixed on a holder) with cold (4 °C, stored in fridge) preservation solution (see composition in Table 1).
       NOTE: The bottom of the syringe is set at 20 cm above the animal. Connect the syringe to the PA cannula, using silicone tubing immersed in iced water (Figure 6C).
   13. Perform a median sternotomy using blunt-ended Lister scissors to avoid damaging the tissue and carefully retract the chest wall using blunt retractors attached to elastic bands and a micro-mosquito hemostat attached to the diaphragm to expose the lungs. Avoid any direct contact with the lung parenchyma.
       NOTE: From this step, to prevent tissue drying, apply a small amount of physiological saline to the surface of the lung when necessary.
   14. Make an "X" incision at the apex of the left ventricle of the heart and slide the scissors inside the heart up to the left atrium to facilitate later insertion of the LA cannula.
   15. Pass a 5-0 silk suture beneath the main PA and prepare a pre-tied double knot and hold the suture on one side with a home-made holder (Figure 6D and Figure 7A).
   16. Prepare an additional suture around the heart for later securement of the LA cannula (Figure 7B).
   17. Make a small incision in the main PA and insert the PA cannula, ensuring no air bubble is introduced.
   18. Immediately perfuse the lungs with 25 mL of the cold preservation solution via the PA cannula, initiating cold ischemia time.
       NOTE: Use passive pressure from the height of the syringe on the holder described above for this perfusion.
   19. Secure the cannula in place with the pretied suture.
   20. Immediately after PA cannulation, reconnect the tracheal cannula to the ventilator and resume ventilation with the initial settings, reducing the respiratory rate to 25 breaths/min.
   21. Insert the LA cannula through the incision in the left ventricle into the left atrium.
   22. Secure the cannula with the second pre-tied suture (around the heart), being careful to prevent flow from being impeded from the left atrium and allow the entire 25 mL volume of cold preservation solution to pass through (takes ~15 min).
   23. At the end of the preservation solution perfusion, clamp the PA cannula with a bulldog clamp (Figure 7C, blue circle) and disconnect it from the tubing.
   24. Carefully dissect the trachea and surrounding tissues to mobilize the lungs.
   25. Clamp the trachea using an aneurysm clip (Figure 7C, red circle), while lungs are fully inflated (one tidal volume).
   26. Dissect the descending aorta and other supporting vessels to free the heart-lung block, lifting it by gently grasping the trachea.
   27. Place the heart-lung block in a dish containing preservation solution in a prone position and store at 4 °C for a total cold ischemia time of 1 h.
       NOTE: At the end of surgery, place biological remains in waste disposal bags for further elimination by the dedicated service.
3. Ex-vivo lung perfusion and thermal preconditioning
   NOTE: Refer to the flowchart outlining the major steps to guide the progression through the different stages of the protocol (Figure 8).
   1. Place an ice bucket in the thermostatic water bath to reduce the perfusate temperature to 15 °C.
   2. Fill the reservoir with 100 mL of perfusate and allow it to circulate through the system. Use the human albumin-based perfusion solution (see composition in Table 1) as the acellular buffered extracellular solution for the ex-vivo perfusion.
   3. Measure the pH of the perfusate and maintain it at a physiological level (7.35-7.45), add tris(hydroxymethyl)aminomethane (THAM) buffer accordingly.
   4. Weigh the heart-lung block and mount it onto the EVLP system, beginning with the ventilation cannula while keeping the aneurysm clip in place on the trachea.
   5. Connect the PA cannula, carefully avoiding the introduction of air bubbles into the lung vasculature (ensure the bubble trap is filled).
   6. Allow a few drops of perfusion solution to exit and connect the LA cannula.
   7. Set left atrial pressure at 3 mm Hg by adjusting the height of the outflow line connected to the drainage circuit.
   8. Zero the weight module and monitor change in weight of the heart lung block throughout EVLP.
   9. Calculate the theoretical cardiac output (CO) using the formula developed by Lindstedt and Schaeffer:
      
   10. Start perfusion (2% of theoretical CO) and gradually increase perfusate flow (7.5% of theoretical CO) and temperature according to the specified parameters, as outlined in Table 2.
       NOTE: During EVLP, perfusate flow is set between 2% and 7.5% of the calculated theoretical CO. This low perfusate flow permits to keep the pulmonary arterial pressure within normal ranges and to reduce the risk of early pulmonary edema, especially in warm ischemic damaged lungs. This allows users to maintain the preparation for extended periods of time. Other investigators used comparable flows for rat EVLP, whereas some others used flows up to 20% of theoretical CO³⁸.
   11. Adjust the temperature of the perfusate by progressively increasing the temperature of the water bath.
       NOTE: Take care to keep hands clear of all heating equipment to avoid possible injuries.
   12. Once perfusate temperature reaches 35 °C, initiate mechanical ventilation by removing the aneurysm clip from the trachea and starting the ventilator dedicated for ex-vivo ventilation (see Table of Materials).
       NOTE: Pressure calibration is performed every 14 days using two-point calibration (0 and 30 cm H₂O). Tube calibration is performed at each utilization according to a programmed protocol, to correct for external factors related to the resistance of tubes and cannula.
       1. Use the referenced software for ventilator settings and data acquisition, as it provides continuous recordings of respiratory signals and lung compliance at dedicated time points (see flowchart, Figure 8), according to a programmed protocol of stepwise increase of lung volume with concomitant measurement of airway pressure.
       2. Compliance is then automatically calculated by an equation fitted to the deflation limb of the pressure volume loop (Salazar-Knowles equation).
       3. Display all ventilatory data, store it on a dedicated personal computer, and export it to a spreadsheet. Extract data at selected time points, transfer it to another spreadsheet, and then copy/paste it into the statistical analysis software.
   13. Set respiratory rate and tidal volume at 7 strokes/min and 3 mL/kg, respectively, for 10 min, and then increase to 15 strokes/min and 6 mL/kg for 10 min and finally, 30 strokes/min and 6 mL/kg until the end of EVLP. Set the positive end-expiratory pressure (PEEP) at 3 cm H₂O throughout EVLP (Table 2).
   14. Apply a recruitment maneuver every 30 min of EVLP to prevent atelectasis by setting airway pressure at 15 cm H₂O for 6 s.
   15. At 1 h, 2 h, 3 h, 4.5 h, and 6 h EVLP, apply a stepwise lung inflation up to a total lung volume of 10 mL/kg of body weight to determine lung compliance and peak inspiratory pressure (Pmax).
       NOTE: Compliance measurements are performed following a recruitment maneuver described above.
   16. Increase perfusate temperature to reach 37 °C  at the inflow PA cannula after 60 min EVLP.
   17. Initiate the ex-vivo heating protocol after 60 min of EVLP.
   18. Increase the water bath temperature up to 43 °C to reach 41.5 °C at the inflow PA cannula (this takes ~ 5 min), as displayed in Figure 2, Figure 3, and Figure 4-5.
   19. Maintain this temperature from 60 min to 90 min EVLP (total of 30 min).
       NOTE: In our previous work, we showed that 41.5 °C for 30 min did not promote thermal injury at the level of the whole lung²⁹. We did not perform specific in vitro cell viability assays to test for thermal injury in such conditions.
   20. After 30 min of heating, add ice to the water bath to rapidly (5 min) set the temperature at the inflow cannula back to 37 °C.
       NOTE: In the control group, the perfusate temperature is maintained at 37 °C throughout EVLP, without thermal preconditioning application (see Table 2). We do not incorporate a "sham heating" group (e.g., by water bath heat adjustment without temperature rise).
   21. Keep perfusate temperature at 37 °C at the inflow PA cannula until the end of the EVLP.
   22. Collect samples of perfusate solution at 1 h, 2 h, 3 h, 4.5 h (1-2 mL at each time point) and 6 h (10 mL) of EVLP, from one of the sampling ports connected to the circuit (upstream or downstream of the lung preparation), as shown in Figure 1-2B. Use samples immediately for gas analyses (measurement of pH, PCO₂ and PO₂, using the CG4+ cartridge). For further biochemical measurements, centrifuge the samples (500 × g, 10 min at 4 °C) and keep the clear supernatant at -80 °C until use.
   23. After 6 h EVLP, reapply the aneurysm clip to the trachea (once the lung is fully inflated) and clamp the PA cannula with a bulldog clamp.
   24. Weigh the heart-lung block to obtain its final weight.
       NOTE: increase in weight during EVLP can be assessed as the difference between final and initial weights or can be monitored continuously during EVLP with the weight module described above.
   25. Carefully dissect the lungs, separating the individual lobes.
   26. Preserve the tissue for downstream applications: snap-freeze in liquid nitrogen and store at -80 °C for biochemical analyses or fix in paraformaldehyde (PFA) for histological evaluation. The left lung can be kept and prepared for lung transplantation.

4. Statistics

1. Allocate animals to specific treatment groups (control versus thermal preconditioning, n = 5/group for each experimental endpoint).
   NOTE: We do not randomly assign animals to one or the other groups, but groups are alternated each day. The protocol (heat vs control) cannot be performed in a blinded manner, owing to the necessity to strictly know the temperature of the perfusate during the experiments. In contrast, biological and histological analyses are done in a blinded fashion, with no identification of the allocated treatment group.
2. Analyze the data.
3. Express results as means ± SEM.
4. Verify normal distribution of the data using the Kolmogorov-Smirnov test.
5. For time point experiments, assess the effects of time and group allocation using two-way ANOVA, followed by Sidak's post-hoc test for group effects and Bonferroni correction for time effects, using 1 h as the reference. Consider a p value < 0.05 as statistically significant. Consider log transformation of the data in case of non-normal distribution before applying ANOVA.
6. For histological analyses, determine (in a blinded fashion) the lung injury score of each sample, by quantifying perivascular edema in 10 fields/section.
   1. Use the relative number of vessels with perivascular edema (A): 0: absent; 1: mild (<25%); 2: moderate (25-50%); 3: severe (>50%).
   2. Use the thickness of perivascular edema (B), calculated as a percentage of the inner vessel diameters (0: no edema; 1: <25%; 2: 25-50%; 3: >50%).
   3. Use the sum of A + B in the 10 sections examined to calculate the score³¹.
   4. Carry out statistical comparisons using the non-parametric Mann-Whitney test (due to not normal distribution of the data), with significance assigned to p < 0.05.

Lung physiological variables are continuously monitored throughout the procedure, including pulmonary artery pressure (PAP), calculated pulmonary vascular resistance, and ventilatory parameters, including static pulmonary compliance (SPC) and peak airway pressure (Pmax). Additionally, the weight module enables the evaluation of edema formation by providing an indirect measure of fluid accumulation in the lung. Benefits of thermal preconditioning (TP) with respect to control lungs (Ctrl) include an improvement of SPC (Figure 9A, expressed as the ratio of SPC at each time point to the initial SPC value, n = 5/group) and a reduction of lung weight gain (Figure 9B, n = 5/group) during EVLP. The data are presented as means ± SEM.

Perfusate fluid can be assayed for various biomarkers. For example, thermal preconditioning is associated with a significantly reduced release of von Willebrand (vWF), an endothelial cell biomarker at the end of EVLP (Figure 9C, n = 5/group, means ± SEM).

Morphological analyses of the lung tissue obtained at the end of EVLP include histology and immunohistochemistry. Examples of hematoxylin and eosin (H & E)-stained sections are shown in Figure 9D, showing reduced histological damage in lungs exposed to thermal preconditioning, expressed as a lung injury score (Figure 9E, n = 5/group). The score was established by quantifying perivascular edema in 10 fields/section (see protocol steps 4.6.1 and 4.6.2). The sum of A + B in the 10 sections examined was used to calculate the score31.

Molecular analyses in the lung tissue at the end of EVLP include RNA or protein expression profiling. As an example, the protein expression levels of the inducible heat shock protein HSP70 can be measured by ELISA, to show the development of the heat shock response in lungs exposed to thermal preconditioning (Figure 9F, showing a time-course experiment with different durations of EVLP, n = 5/group at each time-point). The expression of HSP70 in the TP group peaked after 3h EVLP, that is, 90 min after the end of TP, and remained stable thereafter until the end of EVLP. The data are presented as means ± SEM.

Figure 1: Description of the EVLP platform. (1) Double-walled reservoir for perfusate. (2A) Roller pump for perfusate circulation. (2B) Roller pump for perfusate sampling (biochemical measurements and gas analyses). (3) Thermostatic water bath. (4) Membrane gas exchanger connected to a gas tank (not shown). (5) Left Atrial Pressure sensor. (6) Heated and sealed chamber. (7) EVLP ventilator. (8) Temperature probe and thermometer. (9) Connection platform for the heart-lung block, whose elements are detailed in Figure 2. (Note: the PA pressure sensor is not visible). Please click here to view a larger version of this figure.

Figure 2: Connection platform for the heart-lung block. (1) Inflow cannula to the pulmonary artery. (2) Outflow cannula from the left atrium. (3) Connection for the tracheal cannula. (4) Weight module. (5) In-house designed temperature probe firmly attached against the inflow cannula. Please click here to view a larger version of this figure.

Figure 3: Signal transducers/amplifiers and Analog/Digital converter box. (A). Pulmonary artery pressure transducer/amplifier. (B). Pump speed flow controller. (C). Left atrial pressure transducer/amplifier. (D). Lung weight module. (E). Analog/Digital converter. (F). Personal computer for data display and recording. Please click here to view a larger version of this figure.

Figure 4: Temperature maintenance of the perfusate during EVLP. (1) Thermostatic water bath. (2) Perfusate reservoir with double-walled glass container where water from the thermostatic water bath circulates. (3) Heating coil tubing inside a heated water container connected to the thermostatic water bath. (4) Organ chamber with double-walled glass container for warm water circulation from the water bath. (5) Electronic thermometer and in-house designed temperature probe firmly attached against the inflow cannula. Please click here to view a larger version of this figure.

Figure 5: Identification of the optimal temperature for thermal preconditioning. (A) EVLP Protocol for the evaluation of different heating temperatures. From 60 min to 90 min of EVLP, the perfusate temperature was increased at the indicated temperatures. At 90 min of EVLP, temperature was returned to 37 °C for 90 min until the end of EVLP, where various physiological and molecular measurements were performed. (B) Result summary of the effects of the different heating temperatures on physiological and molecular parameters (green indicates favorable effects; red indicates unfavorable effects; and orange indicates mixed effects). The heating temperature of 41.5 °C provided the most protective effects. Abbreviations: EVLP = ex-vivo lung perfusion; TP = thermal preconditioning. This figure was modified from Ojanguren et al.21. Please click here to view a larger version of this figure.

Figure 6: Surgical material. (A) Instruments for the extraction of the heart-lung block: Aneurysm clip holder and aneurysm microclip (Yasargil system), bulldog clamp, scissors, microscissors, retractors attached to elastic bands. (B) EVLP cannulas for the pulmonary artery (PA), the left atrium (LA) and the trachea. (C) Home-made perfusion system for cold perfusate flush of the lungs. Distance between the level of the PA in the animal and the bottom of the syringe is set at 20 cm. (C) Home-made holder. (Not shown: surgical microscope, ventilator, and anesthesia setup). Please click here to view a larger version of this figure.

Figure 7: Representative pictures of the surgical preparation (A) Pre-tied knot around the pulmonary artery. (B) Pre-tied knot around the tip of the heart. (C) Bulldog clamp (blue circle) on the PA cannula and aneurysm clip (red circle) on the trachea. Abbreviation: PA = pulmonary artery. Please click here to view a larger version of this figure.

Figure 8: Flowchart. Major steps to guide the progression through the different stages of the protocol. The period between 1 h and 1.5 h EVLP is magnified to indicate the treatment protocol according to the group assignment (control vs thermal preconditioning). Abbreviations: CI = cold ischemia; EVLP = ex-vivo lung perfusion; PA = pulmonary artery; SPC = static pulmonary compliance; TP = thermal preconditioning; WI = warm ischemia. Please click here to view a larger version of this figure.

Figure 9: Representative results of the effects of thermal preconditioning in damaged rat lungs. Lungs damaged by 1 h of warm ischemia were perfused in an EVLP system for 6 h. From 60 to 90 min of EVLP, a group of lungs was subjected to thermal preconditioning at 41.5 °C (TP group). A control group of lungs was maintained at 37 °C throughout the EVLP. (A) Static pulmonary compliance (SPC), expressed as the ratio of the initial value, in control (Ctrl, n = 5) and TP lungs (n = 5). SPC was better preserved in TP lungs. (B) Weight gain of the lungs during EVLP in control (n = 5) and TP lungs (n = 5). There was a significant reduction in edema formation in the TP group. (C) The release in the perfusate of the endothelial cell biomarker von Willebrand Factor was suppressed in the TP group (n = 5) compared to controls (n = 5). (D) Lung macroscopic and microscopic (hematoxylin-eosin staining) appearance at the end of EVLP in control (n = 5) and TP lungs (n = 5). (E) Lung injury score (for the quantification of the histological damage) was lower in the TP group (n = 5) compared to the control group (n = 5). (F) Protein levels of the inducible heat shock protein HSP70 in lung tissue extracts measured after 1, 2, 3, 4.5, and 6 h EVLP in control and TP groups (n = 5/group at each time point), showing marked heat shock response in the TP group. The expression of HSP70 in the TP group peaked after 3 h EVLP (90 min after the end of TP) and remained stable thereafter until the end of EVLP.
All graphs show means ± SEM. For time course experiments (A-C,F), statistical comparisons were performed using two-way ANOVA followed by Sidak's test for the group effect and Bonferroni's adjustments for the effect of time (with time 1 h as the reference). For lung injury score (E), statistical comparisons were done using Mann-Whitney test. * p < 0.05 (intergroup differences), † p < 0.05 vs 1 h EVLP. Abbreviations: Ctrl = control; EVLP = ex-vivo lung perfusion; HSP70 = heat shock protein 70; SPC = static pulmonary compliance; TP = thermal preconditioning. Modified from Parapanov et al.23. Please click here to view a larger version of this figure.

Table 1: Composition of the solutions used in the protocol. Please click here to download this Table.

Table 2: EVLP settings for perfusion flow, inflow temperature (at the level of the pulmonary artery cannula) and ventilation over time. Please click here to download this Table.

Table 3: Troubleshooting. Please click here to download this Table.

Supplementary File 1: Protocol detailing the major steps for signal calibration in the ex vivo lung perfusion system, including calibration of pulmonary artery pressure, weight, and flow signals. Please click here to download this File.

Supplementary File 2: Demonstration of the signal calibration procedure for pulmonary artery pressure, weight, and flow during ex vivo lung perfusion. Please click here to download this File.

The development of a reproducible and physiologically relevant rat EVLP model represents a significant step forward in lung transplantation research. Despite significant species differences between the rat and human lung (recently reviewed in¹), notably the greater vulnerability of the rat lung, the rat EVLP model offers pharmacological and non-pharmacological interventions with high reproducibility and low cost. EVLP offers a unique platform to evaluate donor lung function, simulate clinical injury scenarios such as ischemia-reperfusion injury, and implement therapeutic interventions prior to transplantation^2,39,40. In this context, the rat EVLP model enables high-throughput mechanistic investigations under tightly controlled conditions³³. In recent years, several groups have contributed to the refinement of rat EVLP techniques. Nelson et al. provided foundational procedural guidance³⁹. Wang et al. proposed a modular EVLP setup tailored to mimic donation after circulatory death (DCD) simulation⁴¹. Gouchoe et al. explored novel perfusate formulations using artificial oxygen carriers⁴² and Oliveira et al. focused on inflammation after long cold ischemia preservation³⁸. Building on this existing knowledge, our current protocol introduces a novel, non-pharmacological strategy, namely thermal preconditioning, designed to activate endogenous protective mechanisms for the reconditioning of rat lung grafts damaged by warm ischemia. Our damaged lung model, which simulates DCD conditions, was performed in male Sprague-Dawley rats³³.

The establishment of this protocol required careful attention to several technically demanding steps that critically influence lung viability and reproducibility. Lung retrieval must be performed with minimal manipulation to avoid parenchymal injury. The depth of anesthesia is paramount to prevent gasping-induced pulmonary congestion and poor exsanguination. Stringent avoidance of air bubbles during cannulation and circuit priming is essential, as even small emboli can precipitate edema⁴³. Specific troubleshooting advice for common protocol challenges is indicated in Table 3. While ventilation and perfusion settings can be adjusted within safe limits, lung-protective strategies must be maintained to prevent barotrauma or volutrauma⁴⁴. The perfusate flow can be increased up to 20% of the estimated cardiac output in rats, compared to 40% in larger models like pigs and humans, but must be carefully titrated⁴⁵. In our protocol, we use lower perfusate flow (7.5% of theoretical cardiac output), to prevent surges of pulmonary artery pressure and the development of early pulmonary edema in warm ischemic damaged lungs, thereby allowing extended preparation. Although some laboratories administer corticosteroids during perfusion to reduce inflammation^46,47, their known anti-inflammatory effects could confound some experimental outcomes, and we therefore do not use them in our model to preserve the ability to study lung injury and inflammatory processes.

Thermal preconditioning requires precise temperature control. Because heat is lost along the tubing from reservoir to lung, the perfusate temperature must be monitored at the inflow cannula, and the water-bath setting kept slightly above the desired inflow temperature to achieve the target within the preparation. Sensors are calibrated annually using a Multi-Channel PCE-T 1200 Digital Thermometer to ensure accuracy.

Our findings also address key limitations of conventional heat therapy. Systemic heat application is constrained by imprecise temperature regulation, systemic side effects, and lack of organ specificity. EVLP circumvents these barriers by providing a controlled, localized environment, allowing precise delivery of thermal stress directly to the lung tissue²⁹. This method enables rigorous modulation of both the magnitude and duration of the thermal stress, two critical parameters for optimizing the protective cellular response without triggering thermal injury. Moreover, our EVLP platform allows multiple downstream evaluations, including molecular, histological, and functional analyses. Importantly, lungs subjected to EVLP can also be transplanted, allowing for the direct evaluation of post-EVLP graft function in vivo²⁹. Thus, the model supports assessment at both the isolated-organ and post-transplantation levels. In this context, the integration of thermal preconditioning, a transient exposure to 41.5 °C during EVLP, emerges as a promising intervention. By stimulating a heat shock response and other stress-adaptive pathways, this method enhances graft resilience without pharmacological confounders. Our results show that thermal preconditioning improves lung graft physiology, reduces edema formation and the release of inflammatory mediators and biomarkers of cellular injury (as exemplified by the reduced levels of von Willebrand factor in the EVLP perfusate of heat-treated lungs). Morphological analyses further support the protective effect of this intervention, with reduced structural damage and preserved tissue integrity. Most experimental rat EVLP data have been generated in males, limiting sex-specific insight; two studies reported divergent outcomes in females-reduced reperfusion injury in one and increased inflammation in another^48,49. Although our study used males, we previously observed TP benefits in a porcine EVLP model using females²⁹, suggesting relevance across sexes.

Important constraints and drawbacks of our method need to be emphasized. First, to guarantee that the lungs are perfused with the actual desired temperature, the latter must be monitored at the inflow cannula by firmly applying the temperature sensor. Second, it is mandatory that the temperature changes be obtained within very short time periods, which requires the use of a rapid response thermal water bath for heating and the addition of ice to the water bath for rapid cooling. Third, the physiological response to heat stress may differ according to the species used and is highly dependent on the level and duration of the stress. Therefore, it is mandatory to perform pilot experiments using various heating temperatures for various durations in case this protocol is applied to different animal species or human lungs and evaluate specific endpoints pertaining both to protective and to potential detrimental effects of the applied heat stress. Fourth, owing to the kinetic expression of heat-responsive genes, the effects of thermal preconditioning depend on the recovery time following heat stress application. Experiments using different post-heat stress durations (that is, with different durations of EVLP) are required to assess such time-dependent effects of heat application, as shown in our recent publication on this topic³⁰.

Altogether, our current study establishes a detailed, reproducible rat EVLP protocol incorporating thermal preconditioning as a viable strategy to improve marginal donor lung quality. The protocol is scalable and adaptable to larger animal or human EVLP systems, reinforcing its translational relevance. By providing both methodological rigor and physiological fidelity, this model opens new avenues for exploring non-pharmacological interventions aimed at reducing ischemia-reperfusion injury and improving lung graft outcomes after transplantation.