American Journal of Kidney Diseases
Volume 54, Issue 3 , Pages 410-412, September 2009

The Value of Machine Perfusion in Deceased Donor Kidney Transplantation

  • Jeffrey T. Cooper, MD

      Affiliations

    • Corresponding Author InformationAddress correspondence to Jeffrey T. Cooper, MD, Department of Surgery, Division of Transplant Surgery, Tufts Medical Center, 750 Washington St, Boston, MA 02111
    • ,
  • Richard B. Freeman, MD

Tufts Medical Center, Boston, Massachusetts

published online 29 June 2009.

Article Outline

 

Commentary on Moers C, Smits J, Maathuis MH, et al: Machine Perfusion or Cold Storage in Deceased-Donor Kidney Transplantion. N Engl J Med 360:7-19, 2009.

Two methods of preservation exist for deceased donor kidney transplantation: static cold storage and hypothermic machine perfusion. In static cold storage, the kidneys are flushed with preservation solution, along with the other organs to be procured, and after removal from the cadaveric donor, are placed on ice. In hypothermic machine perfusion, the kidneys, when procured, are placed on a perfusion machine that maintains pulsatile flow of the preservative solution at a temperature between 1°C and 8°C. Since the initial description of pulsatile perfusion by Folkert Belzer in 1967, there has been controversy about the relative benefit of hypothermic pulsatile perfusion over cold storage.1 Recently, Moers et al2 reported results from a randomized trial in the New England Journal of Medicine.

Back to Article Outline

What Does This Important Study Show? 

Moers et al2 prospectively randomly assigned kidneys from 336 deceased kidney donors such that 1 kidney of the pair underwent hypothermic machine perfusion and the other was preserved by using cold storage. Deceased donors included donors after cardiac death, donors after neurological death, and standard- and expanded-criteria donors. There were no significant differences in such recipient characteristics as age, duration of pretransplantation dialysis therapy, panel-reactive antibody level, number of previous transplantation surgeries, and immunosuppressive regimen. There also were no significant differences in HLA mismatching, cold ischemic times, or number of renal arteries between the 2 groups. Subgroup analyses also were performed to evaluate any differences between specific donor types: donors after cardiac death versus donors after neurological death and standard-criteria donors versus expanded-criteria donors. The primary end point of the study was delayed graft function, with secondary end points consisting of length of delayed graft function, primary nonfunction, area under the curve of the serum creatinine level, creatinine clearance at postoperative day 14, acute rejection, length of stay, and 1-year patient/graft survival.

The investigators found that machine perfusion was associated with a significant decrease in delayed graft function compared with cold storage (20.8% and 26.5%, respectively; P = 0.01). Importantly, they saw no difference in the magnitude of the treatment effect in their subgroup analyses: in both the expanded-criteria-donor group and donors-after-cardiac-death group, delayed graft function was decreased, but did not reach clinical significance. In addition, use of machine perfusion reduced the duration of delayed graft function by 3 days (P = 0.04). In terms of secondary end points, they found a significant 1-year graft survival benefit with the use of machine perfusion over cold storage (94% versus 90%; P = 0.04). No benefit was seen in rate of acute rejection, length of stay, or creatinine clearance at 14 days. There was a trend in favor of machine perfusion when examining the incidence of primary nonfunction (2.1% versus 4.8%), but this did not reach statistical significance (P = 0.08).

Back to Article Outline

How Does This Study Compare with Prior Studies? 

The study by Moers et al2 follows a long series of retrospective and nonrandomized studies. The few prospective trials were small scale and with insufficient power to reach definitive conclusions regarding important outcome measures.

Early results were mixed. In 1990, in a prospective controlled study of 60 cadaveric renal donors, Merion et al3 could find no benefit in either delayed graft function (the need for hemodialysis within 1 week after transplantation) or early graft loss. This finding was in keeping with other prospective trials performed during the cyclosporine era.4 Given the lack of demonstrable clinical benefit and in view of the additional costs associated with machine perfusion, its value was questioned.

However, as the 1990s proceeded, a growing body of evidence pointed to the relative benefit of pulsatile perfusion over cold storage. In 1997, Burdick et al,5 using the United Network for Organ Sharing (UNOS) database, reviewed results of 60,827 cadaveric kidney transplants performed between 1988 and 1995. They found that cold storage was associated with a 2.13-fold increase in the need for first-week dialysis compared with hypothermic machine perfusion. Of note, this risk increased to 2.33 in donors younger than 55 years. It was estimated that the additional costs inherent in machine perfusion were balanced by the decreased need for postoperative dialysis. When looking at 1-year graft survival, Burdick et al5 could find an effect only when preservation method was combined with donor age. In particular, they found a 1.29-fold greater odds of 1-year graft failure in cold-stored kidneys from older donors in comparison to those that were machine perfused. Similar results were found in a meta-analysis that also included data from outside the United States.6

In 2000, a multicenter retrospective review conducted by Polyak et al7 showed that machine perfusion led to significant reductions in delayed graft function and length of stay, as well as a significant improvement in 1-year graft survival. Of particular significance, these benefits extended to kidneys procured from both standard- and expanded-criteria donors.7 In particular, graft survival at 1 year was 7% greater with machine perfusion than cold storage in standard-criteria kidneys and 10% greater in expanded-criteria kidneys (96% versus 89% and 88% versus 79%, respectively; P = 0.02). Subsequent studies have confirmed that the use of machine perfusion is associated with improved early graft function.

In 2005, Schold et al,8 extending the initial analysis of Burdick et al5 of the UNOS database, found significant reductions in delayed graft function that were magnified in kidneys at greatest risk, including those procured from donors after cardiac death, older donors, and those associated with prolonged cold ischemic times. However, in a subanalysis of their main cohort, consisting of 907 paired transplanted kidneys, of which 1 underwent machine perfusion and the other underwent cold storage, no difference in either 1- or 6-year graft survival was noted. Additional studies specifically of kidneys procured from expanded-criteria donors also found that machine perfusion led to improved early graft function and less need for early dialysis, but also could not show a benefit in graft survival.9, 10

More recently, Shah et al,11 in a prospective randomized trial consisting mostly of kidneys from standard-criteria donors, also found a significant decrease in delayed graft function, but this study did not have enough power to assess differences in graft survival.

Back to Article Outline

What Should Clinicians and Researchers Do? 

The study of Moers et al2 is unique in that it is one of the very few large-scale prospective randomized trials to assess the value of machine perfusion in deceased donor kidney transplantation. As such, its findings are uniquely valid. Early kidney graft function clearly is improved by machine perfusion. This study also shows that this improvement occurs in all deceased donor subtypes: the benefit was not restricted to a high-risk group, such as expanded-criteria donors. As such, it argues against the selective use of machine perfusion dependent on deceased donor type. In addition, this study points to an improvement in graft survival with machine perfusion. Unfortunately, a subset analysis was not performed when looking at graft survival; therefore, we cannot tell whether the graft survival benefit occurred in all subgroups, which had not been seen in other studies.9, 10

Interestingly, the transplantation teams were blinded to perfusion characteristics of the kidneys before implantation. It would be interesting to know whether perfusion characteristics correlated with either delayed graft function or, ultimately, graft survival. In this regard, the study of Moers et al2 is unique because a common criticism of previous studies was that machine perfusion introduced a selection bias, allowing for surgeons to discard kidneys with poor perfusion characteristics and thus accounting for improved graft function. Because the teams were blinded, no selection bias was introduced, arguing in favor of a beneficial physiological effect from machine perfusion.

In the editorial that accompanies the article by Moers et al,2 Drs Tullius and Garcia-Cardena12 relate the importance of such flow-dependent transcription factors as Kruppel-like factor 2 in kidney preservation. Despite the long-standing use of pulsatile perfusion, the means by which it works remains the subject of active investigation. In addition, more experience has been gained with normothermic perfusion, the end results of which remain to be seen. Results of the study of Moers et al2 provide additional evidence that larger prospective trials are necessary, particularly in regard to the benefits of machine perfusion in kidneys at greatest risk of delayed graft function, especially those from donors after cardiac death and extended-criteria donors. Until that time, the study of Moers et al2 remains the best evidence that we have of the benefits of machine perfusion in early and late graft function in deceased donor kidney transplantation.

Back to Article Outline

Acknowledgements 

Financial Disclosure: None.

Back to Article Outline

References 

  1. Belzer F, Ashby B, Dunphy J. 24-Hour and 72-hour preservation of canine kidneys. Lancet. 1967;2(7515):536–538
  2. Moers C, Smits J, Maathuis MH, et al. Machine perfusion or cold storage in deceased-donor kidney transplantation. N Engl J Med. 2009;360:7–19
  3. Merion R, Oh H, Port F, Toledopereyra L, Turcotte J. A prospective controlled trial of cold-storage versus machine-perfusion preservation in cadaveric renal transplantation. Transplantation. 1990;50:230–233
  4. Halloran P, Aprile M. A randomized prospective trial of cold storage versus pulsatile machine perfusion for cadaver kidney preservation. Transplantation. 1987;43:827–832
  5. Burdick J, Rosendale J, McBride M, Kauffman H, Bennett L. National impact of pulsatile perfusion on cadaveric renal transplantation. Transplantation. 1997;64:1730–1733
  6. Wight J, Chilcott J, Holmes M, Brewer N. Pulsatile machine perfusion vs. cold storage of kidneys for transplantation: A rapid and systematic review. Clin Transplant. 2003;17:293–307
  7. Polyak M, Arrington B, Stubenbord W, et al. The influence of pulsatile preservation on renal transplantation in the 1990's. Transplantation. 2000;69:249–258
  8. Schold J, Kaplan B, Howard R, Reed A, Foley D, Meier-Kriesche HU. Are we frozen in time? (Analysis of the utilization and efficacy of pulsatile perfusion in renal transplantation). Am J Transplant. 2005;5:1681–1688
  9. Stratta R, Moore P, Farney A, et al. Influence of pulsatile perfusion on outcomes in kidney transplantation from expanded criteria donors. J Am Coll Surg. 2007;204:873–882
  10. Matsuoka L, Shah T, Aswad S, et al. Pulsatile perfusion reduces the incidence of delayed graft function in expanded criteria donor kidney transplantation. Am J Transplant. 2006;6:1473–1478
  11. Shah A, Milgrom D, Mangus R, et al. Comparison of pulsatile perfusion and cold storage for paired kidney allografts. Transplantation. 2008;86:1006–1009
  12. Tullius S, Garcia-Cardena G. Organ procurement and perfusion before transplantation. N Engl J Med. 2009;360:78–79

 Originally published online as doi: 10.1053/j.ajkd.2009.05.004 on June 29, 2009.

PII: S0272-6386(09)00758-6

doi:10.1053/j.ajkd.2009.05.004

American Journal of Kidney Diseases
Volume 54, Issue 3 , Pages 410-412, September 2009

Article Tools