VCA supercooling in a swine partial hindlimb model

Abstract

Vascularized composite allotransplantations are complex procedures with substantial functional impact on patients. Extended preservation of VCAs is of major importance in advancing this field. It would result in improved donor-recipient matching as well as the potential for ex vivo manipulation with gene and cell therapies. Moreover, it would make logistically feasible immune tolerance induction protocols through mixed chimerism. Supercooling techniques have shown promising results in multi-day liver preservation. It consists of reaching sub-zero temperatures while preventing ice formation within the graft by using various cryoprotective agents. By drastically decreasing the cell metabolism and need for oxygen and nutrients, supercooling allows extended preservation and recovery with lower ischemia-reperfusion injuries. This study is the first to demonstrate the supercooling of a large animal model of VCA. Porcine hindlimbs underwent 48 h of preservation at - 5 °C followed by recovery and normothermic machine perfusion assessment, with no issues in ice formation and favorable levels of injury markers. Our findings provide valuable preliminary results, suggesting a promising future for extended VCA preservation.

Figure 1

Surgical model of the porcine VCA: partial hindlimb. (a,b) The partial hindlimb is harvested with the femoral vessels, distal femur, knee joint, and proximal tibia, and fibula bones. The skin paddle consists of an axial saphenous flap, vascularized by the saphenous pedicle, directly derived from the femoral vessels. This allows robust vascularization of the skin paddle. A careful dissection of the hindlimb muscles allows for conserving only deep muscle groups, ensuring complete vascularization of the VCA. Intraoperative bleeding from the bone is confirmed before control using bone wax and division of the vessels. (c) Pre- and (d) Post-injection iohexol angiography showing adequate vascularization of the harvested muscle groups.

Figure 2

Experimental design. Four hindlimbs were included in each of the three groups undergoing 48 h preservation and recovery. The experimental group underwent supercooling at – 5 °C followed by (1) 2 h SubNormothermic Machine Perfusion (SNMP) recovery and (2) 2 h Normothermic Machine Perfusion (NMP). The control groups 1 and 2 underwent 48 h static cold storage (4 °C, in HTK). Cold Storage + SNMP group received 2 h SNMP before 2 h normothermic reperfusion, while Cold Storage group directly received NMP. All groups received the same monitoring of weight gain, perfusion parameters, biochemical measurements, and histology at each step. HTK: Histidine-Tryptophan-Ketoglutarate solution.

Figure 3

Perfusion system used for loading and recovery, and detailed supercooling protocol. (a) Perfusion system. (i) The VCA is placed on a rack inside a stainless-steel bowl containing the perfusate solution. (ii) A peristaltic pump generates a flow in the silicon tubing to reach (iv) the Hollow-fiber membrane oxygenator, which also traps potential air bubbles. (v) A pressure sensor is connected to the closed system and to (iii) a pressure monitor, allowing for continuous monitoring. (vi) The oxygenated perfusate reaches the VCA through the arterial cannula. (vii) A hot plate stirrer keeps the perfusate at the desired temperature. White arrow = thermometer. (b) Supercooling protocol. The VCA is first loaded with 3-OMG as an intracellular CPA using SNMP (21 °C), before incremental switching to the extracellular CPA cocktail while decreasing the temperature until reaching 4 °C. The limb is then placed for 48 h in a sterile bag full of the extracellular CPA cocktail at − 5 °C in a cooler. The recovery is performed using SNMP with normal Steen + solution with incremental unloading of the extracellular CPA during gradual rewarming. The SNMP is pursued 2 h before switching to whole autologous blood warmed at 37 °C to perform the 2 h NMP phase. 3-OMG 3-O-methyl-glucose, CPA cryoprotective agent, HTK histidine-tryptophan-ketoglutarate, NMP normothermic machine perfusion, PEG polyethylene glycol, SNMP subnormothermic machine perfusion.

Figure 4

Perfusion parameters of recovered porcine VCAs following 48 h preservation. At t = 2 h, the line marks the perfusate switch to whole blood at 37 °C. (a) Flow rate profile during SNMP and NMP recovery showing a significantly higher flow allowed by (b) significantly lower vascular resistance in the SC group during the SNMP phase. No significant difference was found between the three groups in the NMP phase. Note: A mixed-effects model with the Geisser-Greenhouse correction was performed to compare groups during each phase. (c) Weight gain following SNMP and NMP was significantly lower in the SC group compared to CS + SNMP GROUP but not with CS GROUP. (d) Aspect of the stored VCA following 48 h supercooling, showing the absence of ice nucleation.

Figure 5

Perfusate analysis during recovery. (a) Oxygen consumption was higher in the Supercooled group during recovery versus control, but the difference was only significant in the SNMP phase. (b) The glucose consumption was not measurable during the SNMP recovery due to interference with the 3OMG. The supercooled limbs seemed to consume more glucose in the NMP phase versus Cold Storage + SNMP group, but the difference was not significant. (c) Lactate release levels were comparable between groups during both phases. (d) Hemoglobin arteriovenous difference was comparable between groups during the NMP phase. (e) Potassium release was significantly higher in the Supercooled group during the SNMP phase. (f) The pH was lower in the Supercooled group during the NMP phase and tended to reach similar values as control groups during the NMP phase.

Figure 6

Macroscopic and Microscopic results on the porcine VCAs following 48 h preservation and recovery. (a,d,g) The macroscopic aspect of the grafts following 2 h of 37 °C NMP with whole blood was similar in the Cold Storage + SNMP, Cold Storage and Supercooling groups, respectively. The skin paddle presented an ecchymotic aspect, which seemed milder in the Supercooling group, but no significant differences were found histologically. (b,c) At the microscopic level, the Cold storage + SNMP group showed significant interstitial edema, ischemic myocytes (black asterisks) and myocyte fiber injuries (yellow asterisks). (e,f) The Cold storage group showed major edema and mild fiber architecture disruption. (h,i) The Supercooling group showed mild interstitial edema and minor muscle fiber injuries when compared with both control groups. (j) A blinded microscopic score (Kruit et al. 2021) was performed. The overall scores were compared at several time points for each group: T0 (initial), EOP (end of preservation), EOS (end of Steen + SNMP recovery) and EOB (end of NMP blood recovery). A significantly lower muscle injury score was found between the Supercooling group and the Cold storage + SNMP group at the end of the Steen + recovery. The difference between experimental and control groups was not statistically significant at the other time points. (b,e,h) light microscopy (LM), X100, Hematoxylin and Eosin (H&E) staining; (c,f,i) LM X200, H&E.