To the Editor:
The recent article by Nicholson and Hosgood, describing the first clinical study into the use of their developed ex vivo normothermic perfusion (EVNP) machine, was read with great interest . The use of the EVNP in marginal/extended criteria donors (ECD) organs demonstrated favorable delayed graft function (DGF) compared to static cold storage with no difference in primary nonfunction (PNF) and graft or patient survival at 12 months.
A recent meta-analysis reported the outcomes of kidney transplantation from donation after cardiac death (DCD) donors comparing cold storage (CS) and hypothermic machine perfusion (HMP) , concluding that despite lower rates of DGF with HMP, the overall survival and PNF were similar. In this context, the DGF rate with the EVNP may be a genuine effect, but its impact on outcome is questionable. The low numbers, as acknowledged by the authors, obviously support the need for a randomized multicenter trial (RCT). A detailed subgroup analysis of the ECDs either as DCD or donation after brain death (DBD) would add scientific value to the paper.
The EVNP grafts were implanted with an excised carrel patch and shortened vein to accommodate the EVNP cannulae. The technique of vessel handling described thus compromises the length, increasing the thrombosis risk . In our experience with HMP serial vessel connectors, length and the carrel patch can be retained. It would be helpful to know how multiple vessels were handled and if serial connectors were considered.
Once EVNP was commenced the evidence of perfusion, intrarenal-resistance (IRR) and production of urine were noted. The authors suggest these parameters may be used to predict outcome. The cases where a high IRR was recorded interestingly showed no associated DGF. A recent abstract published by the same group showed no advantageous effect on outcome with the use of sildenafil, where the IRR was significantly reduced .The use of the prostacyclin in the perfusate may confound any meaningful conclusion from the IRR data. Whether such simple parameters should be interpreted without caution when predicting outcomes remains to be answered.
All kidneys were successfully transplanted but despite a standard immunosuppression algorithm, both the EVNP and CS groups had high rates of rejection (27.7% vs. 23.4%, respectively). The rate of DGF in the cold storage group could possibly account for the higher rejection rate. Is EVNP an associated factor for the higher rate of rejection?
A recent report from China has shown that shorter cold ischemia times (CIT) could lead to survival [95.7% (1 year)/92.4% (3 years), respectively] on par with living donation . This highlights a move toward improving standards and logistics in reducing CIT/total ischemic times to improve outcomes. In our opinion EVNP would not only impact theatre/total ischemic times, it would require additional expert resources and impact further on infrastructure and costs. Would the proposed benefit of reduced DGF counterbalance these costs?
The authors also state that a potential use of EVNP could be in conjunction with stem cells. This raises two issues: (1) Is EVNP potentially only marginally better than simple, non-technical and cheaper refrigeration techniques? (2) Will 1 h of EVNP, prior to implantation, be adequate to obtain any beneficial effect with stem cell therapy? Controversy still exists in the exact mechanism of stem cell therapy  and presumably these effects may be due to their paracrine effects rather than homing.
Nicholson and associates should be lauded in the development of their normothermic circuit but their results, although impressive, should be met with caution. It is our belief that reorganization of infrastructure to improve graft outcomes should carry priority. Results from an RCT are awaited with interest.
M. A. Hossain and A. Bagul*
Department of Renal Transplantation, St. Georges Hospital NHS Trust, London, UK
*Corresponding author: Atul Bagul, firstname.lastname@example.org
This article is published with the permission of the Controller of HMSO and the Queen's Printer for Scotland.
The authors of this manuscript have no conflicts of interest to disclose as described by the American Journal of Transplantation.