Randomized preclinical study of machine perfusion in vascularized composite allografts

Background: Attempts to improve limb preservation for transplantation using ex vivo perfusion have yielded promising results. However, metabolic acidosis, aberrant perfusate biochemistry and significant perfusion-induced oedema are reported universally. Optimizing perfusion protocols is therefore essential for maintaining tissue health. Methods: A randomized, two-stage open preclinical trial design was used to determine the optimal temperature and mean arterial pressure for machine perfusion. Conditions compared were: Normothermic machine perfusion at 70mmHg (NMP-70); subnormothermic perfusion (28°C) at 70mmHg; subnormothermic (28°C) perfusion at 50mmHg; and hypothermic perfusion (10°C) at 30mmHg. Following this, a head-to-head experiment was undertaken comparing the optimal machine perfusion with static cold storage. Paired bilateral limbs (10 in total) were randomized to either 8 h of static cold storage, or 2 h of static cold storage and 6 h of optimal machine perfusion. Both groups of limbs were then reperfused on a circuit primed with matched blood from unrelated donors for 4 h without immunosuppression. Results: NMP-70 resulted in less tissue injury and stable perfusion biochemistry. Assessing reperfusion outcomes, static cold storage resulted in acidosis with increased lactate and a worsening electrolyte profile, necessitating bolus infusions of bicarbonate to prevent graft loss. Conversely, NMP-70 was associated with haemodynamic and biochemical stability. Histologically, on reperfusion with allogeneic whole blood, limbs subjected to static cold storage exhibited multifocal ischaemic injury and increased inflammation, which was absent with NMP-70. Static cold storage also resulted in significant oedema compared with NMP-70. Conclusion: Normothermic perfusion resulted in superior graft preservation and less reperfusion injury compared with the current static cold storage protocol.