SEARCH

SEARCH BY CITATION

To the Editor:

Lodhi et al. [1] provide a timely retrospective analysis demonstrating superiority for machine perfusion versus cold storage strategies in attenuation of delayed graft function (DGF) for donation after cardiac or brain death (DCD and DBD respectively) kidneys. Their results regarding benefit for machine perfusion of DCD kidneys agree [2] or disagree [3] with previous randomized controlled trials as acknowledged by the authors. However close inspection of these original studies may shed light on effective utilization of machine perfusion.

Jochmans et al. [2], analyzing DCD donors only, demonstrated reduction in DGF with machine perfusion versus cold storage. In this trial, trained perfusion specialists were part of the organ recovery team, resulting in immediate placement of kidneys onto machine perfusion. By contrast in Watson et al. [3], also performed on DCD kidneys, cold perfusion with subsequent machine perfusion was performed. Of total mean 14 h cold ischemic time for kidneys that underwent machine perfusion, initial use of cold storage accounted for mean of 3.9 h (representing 31% of total cold ischemic period, range of 3–73%). Therefore lack of immediate pulsatile perfusion may have rendered subsequent potential benefits of machine perfusion obsolete.

In support of this theory, Hosgood et al. [4] published data from porcine DCD kidney models where retrieved kidneys were allocated to either 18 h of cold storage, 18 h of hypothermic machine perfusion or hybrid strategy of 4 h of cold storage with subsequent 14 h of machine perfusion. Hybrid-strategy kidneys had similar graft function/injury profiles to kidneys that underwent cold storage alone, but significantly more intrarenal resistance. The authors speculate prolonged cold storage preservation injury may have been exacerbated by mechanical damage sustained by subsequent pulsatile perfusion.

Any retrospective analysis comparing organ storage techniques cannot tease out whether putative machine perfusion benefits relate to intrinsic properties of machine perfusion or due to confounders such as discard of kidneys with poor perfusion parameters. In addition, despite contrasting incidence of DGF rates, both Jochmans's and Watson's publications demonstrated equivalent patient and graft survival at 1 year posttransplantation. Although DGF is an important outcome (with adverse clinical consequences), “hard” clinical endpoints such as survival are of greater importance to influence our practice. Finally both studies had similar cold ischemic times in both arms—there is no clinical evidence to suggest prolonged machine perfusion can attenuate detrimental effects to allow prolongation of ischemic times for DCD kidneys.

We would argue DCD kidneys are probably better served being implanted as soon as possible regardless of whether they are machine perfused or not (unless viability is a concern). Machine perfusion may allow selection of good DCD kidneys, but is unlikely to resurrect a bad one. That “Lazarus” strategy may be possible with alternative organ storage strategies such as normothermic perfusion, recently demonstrated as efficacious in a proof-of-concept pilot study of DCD kidney transplantation in man [5]. Further trials with important clinical endpoints are clearly required but normothermic perfusion represents an exciting innovation and is likely to be the future of organ preservation.

Disclosure

  1. Top of page
  2. Disclosure
  3. References

The authors of this manuscript have no conflicts of interest to disclose as described by the American Journal of Transplantation.

References

  1. Top of page
  2. Disclosure
  3. References