- Email: [email protected]
Renal Perfusion Pump Vs Cold Storage for Donation After Cardiac Death Kidneys: A Systematic Review
Transplantation/Vascular Surgery
Renal Perfusion Pump Vs Cold Storage for Donation After Cardiac Death Kidneys: A Systematic Review Varunkumar Bathini, Thomas McGregor, Vivian C. McAlister, Patrick P. W. Luke and Alp Sener* From the Departments of Surgery (VB, TM, VCM, PPWL, AS) and Microbiology and Immunology (AS), University of Western Ontario and Multi-Organ Transplant Program (VCM, PPWL, AS) and Matthew Mailing Center for Translational Transplant Studies (PPWL, AS), University Hospital, London Health Sciences Centre, London (VB, TM, VCM, PPWL, AS), Ontario, Canada
Abbreviations and Acronyms DCD ⫽ donation after cardiac death DGF ⫽ delayed graft function ECD ⫽ expanded criteria donor NDD ⫽ neurologically deceased donor Accepted for publication November 29, 2012. * Correspondence: Department of Surgery, Western University, University Hospital, Fourth Floor, 339 Windermere Rd., London, Ontario, Canada N6A 5A5 (telephone: 519-685-8500, extension 33352; FAX: 519-663-3858; e-mail: Alp.Sener@ lhsc.on.ca).
Purpose: Static cold storage is generally used to preserve kidney allografts from deceased donors. Hypothermic machine perfusion may improve the outcome after transplantation but few studies with limited power have addressed this issue. We reviewed evidence of the effectiveness of storing kidneys from deceased donors after cardiac death before transplantation using cold static storage solution or pulsatile hypothermic machine perfusion. Materials and Methods: We searched electronic databases in September 2011 for systematic reviews and/or meta-analyses, randomized, controlled trials and studies of other designs that compared delayed graft function and graft survival. Sources included The Cochrane Library, PubMed® and EMBASE®. Studies excluded from review included those that did not discriminate between donation after cardiac death and donation from a neurologically deceased donor. Primary outcomes were delayed graft function and 1-year graft survival. Statistical analysis was done using RevMan (http://ims.cochrane.org/revman). Results: Nine studies qualified for review. Pulsatile perfusion pumped kidneys from donation after cardiac death donors had decreased delayed graft function compared to kidneys placed in cold storage (OR 0.64, 95% CI 0.43– 0.95, p ⫽ 0.03). There was a trend toward improved 1-year graft survival in the pulsatile perfusion group but statistical significance was not attained (OR 0.74, 95% CI 0.48 – 1.13, p ⫽ 0.17). Conclusions: Pulsatile machine perfusion of donation after cardiac death kidneys appears to decrease the delayed graft function rate. We noted no benefit in 1-year graft survival. Due to the great heterogeneity among the trials as well as several confounding factors, the overall impact on allograft function and survival requires more study. Key Words: kidney, kidney transplantion, cold ischemia, pulsatile flow, delayed graft function
2214
www.jurology.com
KIDNEY transplantation is the most effective treatment for end stage renal disease.1 Unfortunately, the gap between the number of patients on the transplant waiting list and the number of kidney transplants is widening. Most kidneys for transplantation are obtained from NDDs but this is not
sufficient to meet mounting demand. In the last 15 years there has been growing interest in North America in expanding the existing pool of organs using kidneys obtained from donors after cardiac death.2 Unlike NDD kidneys, DCD kidneys are subject to significant warm ischemia injury and
0022-5347/13/1896-2214/0 THE JOURNAL OF UROLOGY® © 2013 by AMERICAN UROLOGICAL ASSOCIATION EDUCATION
http://dx.doi.org/10.1016/j.juro.2012.11.173 Vol. 189, 2214-2220, June 2013 RESEARCH, INC. Printed in U.S.A.
AND
RENAL PERFUSION VS COLD STORAGE FOR DONATION AFTER CARDIAC DEATH KIDNEYS
prone to higher rates of DGF, primary nonfunction and decreased long-term graft survival.3,4 Of the various methods aimed at improving the early and late outcomes of DCD transplants, there has been renewed interest in using mechanical pulsatile perfusion of kidneys to potentially improve outcomes.5,6 Traditionally, kidneys have been preserved in static cold storage but a recent multicenter study suggested that DCD kidneys perfused via mechanical perfusion may perform better in regard to DGF and graft survival than those placed in cold storage.7 In contrast, a separate multicenter, randomized, controlled trial of cold mechanical perfusion vs static cold storage concluded that mechanical perfusion offered no advantage.8 These studies and a number of others were often underpowered and tended to show trends without achieving statistical significance. Clearly, there is debate on how best to preserve DCD kidneys. To our knowledge there has been no systematic review comparing clinical outcomes after mechanical pulsatile perfusion vs static cold storage for DCD allografts. We performed a systematic review to compare DCD kidneys on pulsatile perfusion and those in cold storage. We evaluated the rates of DGF and 1-year graft survival.
METHODS To identify eligible studies, we systematically searched the literature using electronic databases, including PubMed from 1966 to September 2011, EMBASE from 1980 to September 2011 and The Cochrane Library from 1995 to September 2011 (see table). Initially, our search strategy focused on all articles pertaining to pulsatile perfusion vs cold storage. We then narrowed the articles to only those with data on DCD kidneys since this was our area of interest. We thought that by focusing on the DCD population only we would be more likely to find a definitive answer to the ongoing debate of using the pump for these kidneys. The supplementary table (http://jurology.com/) shows our specific search strategy. Figure 1 shows how we excluded articles. No date or study publication type restrictions were applied. Our exclusion criteria were animal studies, noncomparative studies and reports in languages other than English. Studies that did not clearly discriminate DCD from NDD transplants were also excluded. Nine articles were analyzed and compared.5,8 –15 The main outcome measure used in most studies was DGF, most commonly defined as the requirement for hemodialysis in week 1 after transplantation. Since serum creatinine, graft biopsy results and long-term graft survival were not always reported or clearly stated, they could not be analyzed in our meta-analysis. The only 2 consistent end points reported in the reviewed literature were DGF and 1-year graft survival, which served as our 2 primary end points for this meta-analysis.
2215
Statistical Analysis Analysis was performed using RevMan, version 5.1. This software is designed to prepare and maintain Cochrane Reviews and it is especially designed for meta-analysis. The 2 primary end points that we analyzed were the rates of delayed graft function and 1-year graft survival with p ⫽ 0.05 considered statistically significant for comparisons. We used random effects analysis because we expected some heterogeneity by including trials that were not randomized.
RESULTS We analyzed 9 studies using RevMan (see table). Years of publication were 1975 to 2011. Four of the 9 studies were randomized, controlled trials, 1 was a prospective trial, 3 were retrospective studies and 1 was a cohort study. Six of the 9 studies provided DGF and 1-year graft survival data sets. Perfusion pumped kidneys from DCD donors had a decreased DGF rate compared to kidneys in static cold storage (OR 0.64, 95% CI 0.43– 0.95, p ⫽ 0.03, fig. 2, A). This finding held true on subgroup analysis using only randomized, controlled trials (OR 0.59, 95% CI 0.38 – 0.90) and only nonrandomized trials (OR 0.84, 95% CI 0.72– 0.97, each p ⫽ 0.02, fig. 2, B and C). Although 1-year graft survival showed a trend toward machine perfusion, the overall effect was not statistically significant (OR 0.74, 95% CI 0.48–1.13, p ⫽ 0.17, fig. 3, A). On subgroup analysis randomized trials, which unfortunately had small numbers and low power, showed no effect of pulsatile perfusion for 1-year graft survival (OR 0.64, 95% CI 0.28 –1.46, p ⫽ 0.29, fig. 3, B). Nonrandomized studies with a greater population weight average showed a significant beneficial effect of pulsatile perfusion compared to static storage alone (OR 0.40, 95% CI 0.25– 0.64, p ⫽ 0.0001, fig. 3, C).
DISCUSSION The renal transplant waiting list continues to grow each year, as does the demand for available organs. To decrease the disparity in supply and demand, new and innovative means of using suitable kidneys for transplantation have been established, including recruiting living donors as well as ECDs and DCD grafts. However, there is ongoing controversy on the quality of some ECD and DCD grafts compared to standard criteria organs. Thus, any method to optimize the quality of these kidneys would improve graft outcomes and gain greater support for ECD and DCD kidney transplantation. Several modifiable variables can be altered to try to optimize the quality and outcome of renal
2216
Mean Cold Ischemia (hrs) References
Center (site)
No. Kidneys
Cold Storage
Machine Perfusion
Study Design
Cold Storage Solution
6.08
11.9
Randomized
University of WI or Euro-Collins
Nikkiso APS-02
13
13
Machine Perfusion Pump
Cold Storage
Machine Perfusion
Matsuno et al11
Single (Japan)
Moers et al5
Multiple
15.9
15.0
Randomized
University of WI or histidinetryptophan-ketoglutarate
LifePort
82
82
van der Vliet et al15
Multiple
23.0
25.0
Randomized
University of WI
Gambro with Belzer solution
36
35
Watson et al8
Multiple
14.4
13.9
Randomized
University of WI
LifePort
43
41
Marshall et al10 Locke et al9
Single (Melbourne) United Network for Organ Sharing database Single (Oxford Transplant Unit) Single (Sydney)
13
Prospective Retrospective
Hypertonic citrate Not reported
Gambro PF-3A Not reported
68 1,440
62 1,122
15.2
Retrospective
University of WI
LifePort
18
18
18
Retrospective
Ice slush
Waters Mox 100
83
88
18.6
Cohort
Marshall
LifePort
30
30
Moustafellos et al12 Sheil et al14 Plata-Munoz et al13
Single (Oxford Transplant Center)
16 Not reported
16.7 4.5 17.9
Outcome (followup) DGF, immediate graft function, primary nonfunction, mean postop hemodialysis period, mean serum creatinine, graft survival (1 mo) DGF, decreased creatinine, primary nonfunction, acute rejection, calcineurin inhibitor toxicity, hospital stay, pt ⫹ graft survival (1 yr) Immediate function, DGF, primary nonfunction, 3-mo mean serum creatinine, graft survival (1 yr) DGF, primary nonfunction, decreased creatinine, acute rejection incidence, pt survival, graft function ⫹ survival (1 yr) DGF, graft function ⫹ survival (1 yr) DGF (5 yrs) DGF, discharge creatinine, hospital stay (immediate postop period) Immediate function, graft survival ⫹ rejection, drug toxicity (1 yr) Primary nonfunction, DGF, acute rejection incidence, hospital stay, graft function ⫹ survival (1 yr)
RENAL PERFUSION VS COLD STORAGE FOR DONATION AFTER CARDIAC DEATH KIDNEYS
Studies of usefulness of pulsatile machine perfusion vs static storage alone
RENAL PERFUSION VS COLD STORAGE FOR DONATION AFTER CARDIAC DEATH KIDNEYS
Search strategy from PubMed, Embase and Cochrane Library
726 Papers
338 Duplicate Papers
388 Papers
267 Reviews, Discussions and Background Papers
121 Full Text Papers
112 Inappropriate comparator groups/insufficient data
9 Papers for analysis
Figure 1. Flow chart shows studies selected for analysis
allografts. These variables include the type of cold storage (static cold storage vs machine perfusion), the perfusate used, and the addition of calcium
2217
channel blockers and small endogenous mediators (carbon monoxide releasing molecules as hydrogen sulfide releasing substances) to the perfusate.16 –19 One of the most promising variables in improving graft outcomes involves modifying kidney storage techniques in the period between procurement and transplantation. For kidney transplantation organ preservation classically consists of static cold storage, which continues to be the preferred method in the current era. However, machine perfusion has reemerged as a topic of interest with the potential to have a beneficial impact on the mounting clinical challenges to improve kidney transplantation outcomes. Several pumps have been used during the decades. Some older generation pumps include the nonpulsatile Nikkiso® (Japan), and the pulsatile Gambro® (Sweden) and Waters® (United States) pumps. In the more recent era of pumping kidneys the LifePort® Kidney Transporter has been used at most institutions.
Figure 2. Effect of pulsatile renal perfusion vs static cold renal preservation on DGF in DCD kidneys using all available studies for comparison (A), only randomized, controlled trials (B) and only nonrandomized, controlled trials (C). Chi2, chi-square.
2218
RENAL PERFUSION VS COLD STORAGE FOR DONATION AFTER CARDIAC DEATH KIDNEYS
A Study or Subgroup Marshall et al10 Moers et al5 Plata-Munoz et al13 Sheil et al14 van der Vliet et al15 Watson et al8 Total (95% CI)
Pump Cold Storage Odds Ratio (Non-event) M-H, Fixed, 95% CI Events Total Events Total Weight 18 77 30 60 27 42
42 82 30 88 35 45 322
18.2% 7.7% 5.1% 57.7% 9.4% 1.9%
0.78 [0.29, 2.11] 1.27 [0.33, 4.89] 0.19 [0.01, 4.06] 0.50 [0.27, 0.93] 1.48 [0.46, 4.82] 3.14 [0.31, 31.42]
303 100.0%
0.74 [0.48, 1.13]
27 82 30 83 36 45
10 78 28 43 30 44
233 254 Total events Heterogeneity: Chi² = 5.73, df = 5 (P = 0.33); I² = 13% Test for overall effect: Z = 1.38 (P = 0.17)
Odds Ratio (Non-event) M-H, Fixed, 95% CI
0.1 0.2
0.5 1 2 5 10 Pump Cold Storage
B Moers et al5 van der Vliet et al15 Watson et al8
C Marshall et al10 Plata-Munoz et al13 Sheil et al14
Figure 3. Effect of pulsatile renal perfusion vs static cold renal preservation on 1-year graft survival in DCD kidneys using all available studies for comparison (A), only randomized, controlled trials (B) and only nonrandomized, controlled trials (C). Chi2, chi-square.
Recently, 2 randomized, controlled trials evaluated the effect of mechanical perfusion on renal transplantation outcomes.5,8 The first of these 2 studies was a large, multicenter, randomized, controlled trial of NDD and DCD grafts.5 Although the series showed an overall benefit of machine perfusion for decreasing DGF, it failed to demonstrate any overall improvement in 1-year graft survival. The more recent study evaluated the effects of pulsatile perfusion on DCD kidneys.8 This well designed, smaller study showed no clear benefit to pulsatile perfusion over static storage for DGF or 1-year graft function and survival. Given the evidence that the DGF rate is higher in recipients of DCD and ECD kidneys, the overall benefit of pulsatile perfusion would be expected to be greater in this population.20 In fact, looking at our own institutional data, for which we closely monitored the DGF rate, early and late graft function parameters, and pump parameters, while consistently using the same perfusate solution in all transplants, we found a trend toward a decreased DGF rate and improved 1-year graft survival in this patient cohort. In our retrospective
analysis 16 of 23 kidneys (69.6%) placed on pulsatile perfusion machines had DGF vs 22 of 28 (78.6%) placed in cold storage in the DCD population. Although there was a trend toward improved outcomes with pulsatile perfusion, our small data set showed no statistical significance (OR 0.62, 95% CI 0.18 –2.21). This difficulty with large, well performed studies adds further rationale to the need for our systematic review. While performing our systematic review, we noted that various mechanical pumps and solutions were used for cold storage depending on the era in which the study was published. This was an obvious confounder in our study with unknown significance. Not all series, especially older ones, provide details on pump technology, perfusates used and solutions used to flush the kidney before machine perfusion. Kidneys were flushed with citrate solution before being placed on a pump in the studies by Marshall10 and Plata-Munoz13 et al. In the other 7 studies the kidneys were not flushed or it was unspecified what was done before the kidneys were placed on the pump.5,8,9,11,12,14,15 Needless to say, this introduced
RENAL PERFUSION VS COLD STORAGE FOR DONATION AFTER CARDIAC DEATH KIDNEYS
significant variability when comparing study methodology. In addition, pump parameters, such as perfusion pressure, type of perfusate used and additives to the pump perfusate, greatly varied and were not always clearly reported. Examples of various additives to the perfusate included streptokinase, insulin, dexamethasone and mannitol, which all theoretically have the ability to affect graft function. Clearly, we have much to learn about different preservation solution additives. There does not appear to be a single preservation solution or gold standard that is considered universally ideal. However, the current systematic review highlights the fact that, regardless of the perfusate or device used, the usefulness of pulsatile perfusion is evident. As we further push the limits of organ donation, we need further well designed, randomized trials to elucidate the usefulness of the type of preservation solution as well as the importance of the minimum time that organs should be maintained on the pump. In contrast to static cold storage on ice, ex vivo perfusion of donor organs during the storage phase before transplantation is an old idea that is being revisited due to the prospect of providing better quality organs for transplantation, especially in the era of increased use of marginal grafts. Various reports reveal a beneficial effect of pulsatile perfusion but they show significant heterogeneity in storage methods and outcome measures, making objective assessment difficult. Most studies of the impact of pulsatile machine perfusion on DCD kidneys consist of retrospective reviews and nonrandomized series. Some studies are quite old and were done in the era before new machine perfusion technology and current perfusate solutions. Furthermore, some of these studies were not consistent with the type of pump, perfusate and procurement techniques used today. A presumed advantage of using pumps is that we can determine pump parameters. If a kidney does not meet the pump criteria, it would be discarded. We could not determine from the studies how many kidneys were discarded because they failed to meet pump parameter criteria. However, some kidneys that were discarded may have gone on to have function eventually, which would have skewed the results of DGF in favor of pulsatile perfusion. It is important to ascertain which population of donor kidneys would benefit the most from pulsatile machine perfusion. This distinction would be economically prudent with the cost associated with pulsatile perfusion and storage ranging from $1,500 per kidney and up. This does not include
2219
the cost of kidneys that are discarded because they do not meet pump parameter requirements. One would expect the costs to be much higher if this were factored in. By improving the DGF rate and graft survival other costs, such as postoperative dialysis and increased hospital stay, could be decreased. Therefore, we propose that pulsatile perfusion of DCD allografts may benefit the overall hospital budget in the long term. However, costbenefit analysis is required to support our findings. Setting costs aside, the benefit to transplant recipients should be the foremost concern. A decreased DGF rate and overall improved graft survival have obvious benefits to recipients mentally and physically, not to mention the time gained by earlier return to work. Our systematic review reveals that machine perfusion of DCD kidneys results in a statistically significant decrease in DGF. Although there was a trend toward improved 1-year graft survival, this result was not statistically significant. This finding at 1 year is not surprising, given that most grafts have good function 1 year after transplantation regardless of donor type. It was also difficult to ascertain from our literature review whether renal function was affected by pulsatile perfusion since studies did not provide the estimated glomerular filtration rate. We have been using mechanical pulsatile perfusion for all types of donor kidneys for several years. In our experience machine perfused DCD kidneys seem to benefit most and, thus, have a statistically lower glomerular filtration rate than their statically perfused counterparts (unpublished data). To our knowledge the more important question of whether machine perfusion will translate to superior outcomes after 5 or even 10 years remains to be studied.
CONCLUSIONS Our review shows that pulsatile perfusion appears to decrease the DGF rate in after DCD kidney transplantation. However, the impact of pulsatile perfusion on ECD and ECD-DCD donors requires further assessment. With the increasing age and comorbidity of organ donors, pulsatile perfusion may become an increasingly important tool for transplant programs to ensure improved shortterm and long-term outcomes. Also, with the advent of novel compounds added to preservation solutions that mitigate ischemia-reperfusion injury, at least in animal models, we must evaluate the influence on perfusion parameters and eventual graft function in future studies.
2220
RENAL PERFUSION VS COLD STORAGE FOR DONATION AFTER CARDIAC DEATH KIDNEYS
REFERENCES 1. Wolfe RA et al: Comparison of mortality in all patients on dialysis, patients on dialysis awaiting transplantation, and recipients of a first cadaveric transplant. N Engl J Med 1999; 341: 1725. 2. Stratta RJ, Rohr MS, Sundberg et al: Increased kidney transplantation utilizing expanded criteria deceased organ donors with results comparable to standard criteria donor transplant. Ann Surg 2004; 239: 688. 3. Yarlagadda SG, Klein CL and Jani A: Long-term renal outcomes after delayed graft function. Adv Chronic Kidney Dis 2008; 15: 248. 4. Perico N, Cattaneo D, Sayegh et al: Delayed graft function in kidney transplantation. Lancet 2004; 364: 1814. 5. Moers C, Smits JM, Maathuis MJ et al: Machine perfusion or cold storage in deceased-donor kidney transplantation. N Engl J Med 2009; 360: 7. 6. Kalbfleisch M, Lobb I, Mok A et al: Hydrogen sulphide improves real-time renal microvascular flow and attenuates renal injury following prolonged warm ischemia reperfusion injury. Can Urol Assoc J, suppl., 2011; 5: S165. 7. Jochmans I, Moers C, Smits JM et al: Machine perfusion versus cold storage for the preservation of kidneys donated after cardiac death: a multicenter, randomized, controlled trial. Ann Surg 2010; 252: 756.
8. Watson CJE, Wells AC, Roberts RJ et al: Cold machine perfusion versus static cold storage of kidneys donated after cardiac death: a UK multicenter randomized controlled trial. Am J Transplant 2010; 10: 1991. 9. Locke JE, Segev DL, Warren DS et al: Outcomes of kidneys from donors after cardiac death: Implications for allocation and preservation. Am J Transplant 2007; 7: 1797. 10. Marshall VC, Ross H, Scott DF et al: Preservation of cadaveric renal allografts-comparison of flushing and pumping techniques. Proc Eur Dial Transplant Assoc 1977; 14: 302. 11. Matsuno N, Sakurai E, Tamaki I et al: The effect of machine perfusion preservation versus cold storage on the function of kidneys from nonheart-beating donors. Transplantation 1994; 57: 293. 12. Moustafellos P, Hadjianastassiou V, Roy D et al: The influence of pulsatile preservation in kidney transplantation from non-heart-beating donors. Transplant Proc 2007; 39: 1323. 13. Plata-Munoz JJ, Muthusamy A, Quiroga I et al: Impact of pulsatile perfusion on postoperative outcome of kidneys from controlled donors after cardiac death. Transplant Int 2008; 21: 899. 14. Sheil AG, Drummond JM, Rogers JH et al: A controlled clinical trial of machine perfusion of
cadaveric donor renal allografts. Lancet 1975; 2: 287. 15. van der Vliet JA, Kievit JK, Héné RJ et al: Preservation of non-heart-beating donor kidneys: a clinical prospective randomised case-control study of machine perfusion versus cold storage. Transplant Proc 2001; 33: 847. 16. Nguan C, Sener A, Karnik V et al: Perfusion of renal allografts with verapamil improves graft function. Transplantation 2008; 86: 1463. 17. Alabbassi A, Tran KC, Bloch M et al: The effects of molecules CORM-3 and H2S on renal protection during pulsatile perfusion. Am J Transplant 2011; 2: A1614. 18. Lobb I, Mok A, Lan Z et al: Supplemental hydrogen sulphide protects transplant kidney function and prolongs recipient survival after prolonged cold ischaemia-reperfusion injury by mitigating renal graft apoptosis and inflammation. BJU Int 2012; 110: E1187. 19. Zhu JX, Kalbfleisch M, Yang YX et al: Detrimental effects of prolonged warm renal ischaemia-reperfusion injury are abrogated by supplemental hydrogen sulphide: an analysis using real-time intravital microscopy and polymerase chain reaction. BJU Int 2012; 110: E1218. 20. Wells AC, Rushworth L, Thiru S et al: Donor kidney disease and transplant outcome for kidneys donated after cardiac death. Br J Surg 2009; 96: 299.
Renal Perfusion Pump Vs Cold Storage for Donation After Cardiac Death Kidneys: A Systematic Review Varunkumar Bathini, Thomas McGregor, Vivian C. McAlister, Patrick P. W. Luke and Alp Sener* From the Departments of Surgery (VB, TM, VCM, PPWL, AS) and Microbiology and Immunology (AS), University of Western Ontario and Multi-Organ Transplant Program (VCM, PPWL, AS) and Matthew Mailing Center for Translational Transplant Studies (PPWL, AS), University Hospital, London Health Sciences Centre, London (VB, TM, VCM, PPWL, AS), Ontario, Canada
Abbreviations and Acronyms DCD ⫽ donation after cardiac death DGF ⫽ delayed graft function ECD ⫽ expanded criteria donor NDD ⫽ neurologically deceased donor Accepted for publication November 29, 2012. * Correspondence: Department of Surgery, Western University, University Hospital, Fourth Floor, 339 Windermere Rd., London, Ontario, Canada N6A 5A5 (telephone: 519-685-8500, extension 33352; FAX: 519-663-3858; e-mail: Alp.Sener@ lhsc.on.ca).
Purpose: Static cold storage is generally used to preserve kidney allografts from deceased donors. Hypothermic machine perfusion may improve the outcome after transplantation but few studies with limited power have addressed this issue. We reviewed evidence of the effectiveness of storing kidneys from deceased donors after cardiac death before transplantation using cold static storage solution or pulsatile hypothermic machine perfusion. Materials and Methods: We searched electronic databases in September 2011 for systematic reviews and/or meta-analyses, randomized, controlled trials and studies of other designs that compared delayed graft function and graft survival. Sources included The Cochrane Library, PubMed® and EMBASE®. Studies excluded from review included those that did not discriminate between donation after cardiac death and donation from a neurologically deceased donor. Primary outcomes were delayed graft function and 1-year graft survival. Statistical analysis was done using RevMan (http://ims.cochrane.org/revman). Results: Nine studies qualified for review. Pulsatile perfusion pumped kidneys from donation after cardiac death donors had decreased delayed graft function compared to kidneys placed in cold storage (OR 0.64, 95% CI 0.43– 0.95, p ⫽ 0.03). There was a trend toward improved 1-year graft survival in the pulsatile perfusion group but statistical significance was not attained (OR 0.74, 95% CI 0.48 – 1.13, p ⫽ 0.17). Conclusions: Pulsatile machine perfusion of donation after cardiac death kidneys appears to decrease the delayed graft function rate. We noted no benefit in 1-year graft survival. Due to the great heterogeneity among the trials as well as several confounding factors, the overall impact on allograft function and survival requires more study. Key Words: kidney, kidney transplantion, cold ischemia, pulsatile flow, delayed graft function
2214
www.jurology.com
KIDNEY transplantation is the most effective treatment for end stage renal disease.1 Unfortunately, the gap between the number of patients on the transplant waiting list and the number of kidney transplants is widening. Most kidneys for transplantation are obtained from NDDs but this is not
sufficient to meet mounting demand. In the last 15 years there has been growing interest in North America in expanding the existing pool of organs using kidneys obtained from donors after cardiac death.2 Unlike NDD kidneys, DCD kidneys are subject to significant warm ischemia injury and
0022-5347/13/1896-2214/0 THE JOURNAL OF UROLOGY® © 2013 by AMERICAN UROLOGICAL ASSOCIATION EDUCATION
http://dx.doi.org/10.1016/j.juro.2012.11.173 Vol. 189, 2214-2220, June 2013 RESEARCH, INC. Printed in U.S.A.
AND
RENAL PERFUSION VS COLD STORAGE FOR DONATION AFTER CARDIAC DEATH KIDNEYS
prone to higher rates of DGF, primary nonfunction and decreased long-term graft survival.3,4 Of the various methods aimed at improving the early and late outcomes of DCD transplants, there has been renewed interest in using mechanical pulsatile perfusion of kidneys to potentially improve outcomes.5,6 Traditionally, kidneys have been preserved in static cold storage but a recent multicenter study suggested that DCD kidneys perfused via mechanical perfusion may perform better in regard to DGF and graft survival than those placed in cold storage.7 In contrast, a separate multicenter, randomized, controlled trial of cold mechanical perfusion vs static cold storage concluded that mechanical perfusion offered no advantage.8 These studies and a number of others were often underpowered and tended to show trends without achieving statistical significance. Clearly, there is debate on how best to preserve DCD kidneys. To our knowledge there has been no systematic review comparing clinical outcomes after mechanical pulsatile perfusion vs static cold storage for DCD allografts. We performed a systematic review to compare DCD kidneys on pulsatile perfusion and those in cold storage. We evaluated the rates of DGF and 1-year graft survival.
METHODS To identify eligible studies, we systematically searched the literature using electronic databases, including PubMed from 1966 to September 2011, EMBASE from 1980 to September 2011 and The Cochrane Library from 1995 to September 2011 (see table). Initially, our search strategy focused on all articles pertaining to pulsatile perfusion vs cold storage. We then narrowed the articles to only those with data on DCD kidneys since this was our area of interest. We thought that by focusing on the DCD population only we would be more likely to find a definitive answer to the ongoing debate of using the pump for these kidneys. The supplementary table (http://jurology.com/) shows our specific search strategy. Figure 1 shows how we excluded articles. No date or study publication type restrictions were applied. Our exclusion criteria were animal studies, noncomparative studies and reports in languages other than English. Studies that did not clearly discriminate DCD from NDD transplants were also excluded. Nine articles were analyzed and compared.5,8 –15 The main outcome measure used in most studies was DGF, most commonly defined as the requirement for hemodialysis in week 1 after transplantation. Since serum creatinine, graft biopsy results and long-term graft survival were not always reported or clearly stated, they could not be analyzed in our meta-analysis. The only 2 consistent end points reported in the reviewed literature were DGF and 1-year graft survival, which served as our 2 primary end points for this meta-analysis.
2215
Statistical Analysis Analysis was performed using RevMan, version 5.1. This software is designed to prepare and maintain Cochrane Reviews and it is especially designed for meta-analysis. The 2 primary end points that we analyzed were the rates of delayed graft function and 1-year graft survival with p ⫽ 0.05 considered statistically significant for comparisons. We used random effects analysis because we expected some heterogeneity by including trials that were not randomized.
RESULTS We analyzed 9 studies using RevMan (see table). Years of publication were 1975 to 2011. Four of the 9 studies were randomized, controlled trials, 1 was a prospective trial, 3 were retrospective studies and 1 was a cohort study. Six of the 9 studies provided DGF and 1-year graft survival data sets. Perfusion pumped kidneys from DCD donors had a decreased DGF rate compared to kidneys in static cold storage (OR 0.64, 95% CI 0.43– 0.95, p ⫽ 0.03, fig. 2, A). This finding held true on subgroup analysis using only randomized, controlled trials (OR 0.59, 95% CI 0.38 – 0.90) and only nonrandomized trials (OR 0.84, 95% CI 0.72– 0.97, each p ⫽ 0.02, fig. 2, B and C). Although 1-year graft survival showed a trend toward machine perfusion, the overall effect was not statistically significant (OR 0.74, 95% CI 0.48–1.13, p ⫽ 0.17, fig. 3, A). On subgroup analysis randomized trials, which unfortunately had small numbers and low power, showed no effect of pulsatile perfusion for 1-year graft survival (OR 0.64, 95% CI 0.28 –1.46, p ⫽ 0.29, fig. 3, B). Nonrandomized studies with a greater population weight average showed a significant beneficial effect of pulsatile perfusion compared to static storage alone (OR 0.40, 95% CI 0.25– 0.64, p ⫽ 0.0001, fig. 3, C).
DISCUSSION The renal transplant waiting list continues to grow each year, as does the demand for available organs. To decrease the disparity in supply and demand, new and innovative means of using suitable kidneys for transplantation have been established, including recruiting living donors as well as ECDs and DCD grafts. However, there is ongoing controversy on the quality of some ECD and DCD grafts compared to standard criteria organs. Thus, any method to optimize the quality of these kidneys would improve graft outcomes and gain greater support for ECD and DCD kidney transplantation. Several modifiable variables can be altered to try to optimize the quality and outcome of renal
2216
Mean Cold Ischemia (hrs) References
Center (site)
No. Kidneys
Cold Storage
Machine Perfusion
Study Design
Cold Storage Solution
6.08
11.9
Randomized
University of WI or Euro-Collins
Nikkiso APS-02
13
13
Machine Perfusion Pump
Cold Storage
Machine Perfusion
Matsuno et al11
Single (Japan)
Moers et al5
Multiple
15.9
15.0
Randomized
University of WI or histidinetryptophan-ketoglutarate
LifePort
82
82
van der Vliet et al15
Multiple
23.0
25.0
Randomized
University of WI
Gambro with Belzer solution
36
35
Watson et al8
Multiple
14.4
13.9
Randomized
University of WI
LifePort
43
41
Marshall et al10 Locke et al9
Single (Melbourne) United Network for Organ Sharing database Single (Oxford Transplant Unit) Single (Sydney)
13
Prospective Retrospective
Hypertonic citrate Not reported
Gambro PF-3A Not reported
68 1,440
62 1,122
15.2
Retrospective
University of WI
LifePort
18
18
18
Retrospective
Ice slush
Waters Mox 100
83
88
18.6
Cohort
Marshall
LifePort
30
30
Moustafellos et al12 Sheil et al14 Plata-Munoz et al13
Single (Oxford Transplant Center)
16 Not reported
16.7 4.5 17.9
Outcome (followup) DGF, immediate graft function, primary nonfunction, mean postop hemodialysis period, mean serum creatinine, graft survival (1 mo) DGF, decreased creatinine, primary nonfunction, acute rejection, calcineurin inhibitor toxicity, hospital stay, pt ⫹ graft survival (1 yr) Immediate function, DGF, primary nonfunction, 3-mo mean serum creatinine, graft survival (1 yr) DGF, primary nonfunction, decreased creatinine, acute rejection incidence, pt survival, graft function ⫹ survival (1 yr) DGF, graft function ⫹ survival (1 yr) DGF (5 yrs) DGF, discharge creatinine, hospital stay (immediate postop period) Immediate function, graft survival ⫹ rejection, drug toxicity (1 yr) Primary nonfunction, DGF, acute rejection incidence, hospital stay, graft function ⫹ survival (1 yr)
RENAL PERFUSION VS COLD STORAGE FOR DONATION AFTER CARDIAC DEATH KIDNEYS
Studies of usefulness of pulsatile machine perfusion vs static storage alone
RENAL PERFUSION VS COLD STORAGE FOR DONATION AFTER CARDIAC DEATH KIDNEYS
Search strategy from PubMed, Embase and Cochrane Library
726 Papers
338 Duplicate Papers
388 Papers
267 Reviews, Discussions and Background Papers
121 Full Text Papers
112 Inappropriate comparator groups/insufficient data
9 Papers for analysis
Figure 1. Flow chart shows studies selected for analysis
allografts. These variables include the type of cold storage (static cold storage vs machine perfusion), the perfusate used, and the addition of calcium
2217
channel blockers and small endogenous mediators (carbon monoxide releasing molecules as hydrogen sulfide releasing substances) to the perfusate.16 –19 One of the most promising variables in improving graft outcomes involves modifying kidney storage techniques in the period between procurement and transplantation. For kidney transplantation organ preservation classically consists of static cold storage, which continues to be the preferred method in the current era. However, machine perfusion has reemerged as a topic of interest with the potential to have a beneficial impact on the mounting clinical challenges to improve kidney transplantation outcomes. Several pumps have been used during the decades. Some older generation pumps include the nonpulsatile Nikkiso® (Japan), and the pulsatile Gambro® (Sweden) and Waters® (United States) pumps. In the more recent era of pumping kidneys the LifePort® Kidney Transporter has been used at most institutions.
Figure 2. Effect of pulsatile renal perfusion vs static cold renal preservation on DGF in DCD kidneys using all available studies for comparison (A), only randomized, controlled trials (B) and only nonrandomized, controlled trials (C). Chi2, chi-square.
2218
RENAL PERFUSION VS COLD STORAGE FOR DONATION AFTER CARDIAC DEATH KIDNEYS
A Study or Subgroup Marshall et al10 Moers et al5 Plata-Munoz et al13 Sheil et al14 van der Vliet et al15 Watson et al8 Total (95% CI)
Pump Cold Storage Odds Ratio (Non-event) M-H, Fixed, 95% CI Events Total Events Total Weight 18 77 30 60 27 42
42 82 30 88 35 45 322
18.2% 7.7% 5.1% 57.7% 9.4% 1.9%
0.78 [0.29, 2.11] 1.27 [0.33, 4.89] 0.19 [0.01, 4.06] 0.50 [0.27, 0.93] 1.48 [0.46, 4.82] 3.14 [0.31, 31.42]
303 100.0%
0.74 [0.48, 1.13]
27 82 30 83 36 45
10 78 28 43 30 44
233 254 Total events Heterogeneity: Chi² = 5.73, df = 5 (P = 0.33); I² = 13% Test for overall effect: Z = 1.38 (P = 0.17)
Odds Ratio (Non-event) M-H, Fixed, 95% CI
0.1 0.2
0.5 1 2 5 10 Pump Cold Storage
B Moers et al5 van der Vliet et al15 Watson et al8
C Marshall et al10 Plata-Munoz et al13 Sheil et al14
Figure 3. Effect of pulsatile renal perfusion vs static cold renal preservation on 1-year graft survival in DCD kidneys using all available studies for comparison (A), only randomized, controlled trials (B) and only nonrandomized, controlled trials (C). Chi2, chi-square.
Recently, 2 randomized, controlled trials evaluated the effect of mechanical perfusion on renal transplantation outcomes.5,8 The first of these 2 studies was a large, multicenter, randomized, controlled trial of NDD and DCD grafts.5 Although the series showed an overall benefit of machine perfusion for decreasing DGF, it failed to demonstrate any overall improvement in 1-year graft survival. The more recent study evaluated the effects of pulsatile perfusion on DCD kidneys.8 This well designed, smaller study showed no clear benefit to pulsatile perfusion over static storage for DGF or 1-year graft function and survival. Given the evidence that the DGF rate is higher in recipients of DCD and ECD kidneys, the overall benefit of pulsatile perfusion would be expected to be greater in this population.20 In fact, looking at our own institutional data, for which we closely monitored the DGF rate, early and late graft function parameters, and pump parameters, while consistently using the same perfusate solution in all transplants, we found a trend toward a decreased DGF rate and improved 1-year graft survival in this patient cohort. In our retrospective
analysis 16 of 23 kidneys (69.6%) placed on pulsatile perfusion machines had DGF vs 22 of 28 (78.6%) placed in cold storage in the DCD population. Although there was a trend toward improved outcomes with pulsatile perfusion, our small data set showed no statistical significance (OR 0.62, 95% CI 0.18 –2.21). This difficulty with large, well performed studies adds further rationale to the need for our systematic review. While performing our systematic review, we noted that various mechanical pumps and solutions were used for cold storage depending on the era in which the study was published. This was an obvious confounder in our study with unknown significance. Not all series, especially older ones, provide details on pump technology, perfusates used and solutions used to flush the kidney before machine perfusion. Kidneys were flushed with citrate solution before being placed on a pump in the studies by Marshall10 and Plata-Munoz13 et al. In the other 7 studies the kidneys were not flushed or it was unspecified what was done before the kidneys were placed on the pump.5,8,9,11,12,14,15 Needless to say, this introduced
RENAL PERFUSION VS COLD STORAGE FOR DONATION AFTER CARDIAC DEATH KIDNEYS
significant variability when comparing study methodology. In addition, pump parameters, such as perfusion pressure, type of perfusate used and additives to the pump perfusate, greatly varied and were not always clearly reported. Examples of various additives to the perfusate included streptokinase, insulin, dexamethasone and mannitol, which all theoretically have the ability to affect graft function. Clearly, we have much to learn about different preservation solution additives. There does not appear to be a single preservation solution or gold standard that is considered universally ideal. However, the current systematic review highlights the fact that, regardless of the perfusate or device used, the usefulness of pulsatile perfusion is evident. As we further push the limits of organ donation, we need further well designed, randomized trials to elucidate the usefulness of the type of preservation solution as well as the importance of the minimum time that organs should be maintained on the pump. In contrast to static cold storage on ice, ex vivo perfusion of donor organs during the storage phase before transplantation is an old idea that is being revisited due to the prospect of providing better quality organs for transplantation, especially in the era of increased use of marginal grafts. Various reports reveal a beneficial effect of pulsatile perfusion but they show significant heterogeneity in storage methods and outcome measures, making objective assessment difficult. Most studies of the impact of pulsatile machine perfusion on DCD kidneys consist of retrospective reviews and nonrandomized series. Some studies are quite old and were done in the era before new machine perfusion technology and current perfusate solutions. Furthermore, some of these studies were not consistent with the type of pump, perfusate and procurement techniques used today. A presumed advantage of using pumps is that we can determine pump parameters. If a kidney does not meet the pump criteria, it would be discarded. We could not determine from the studies how many kidneys were discarded because they failed to meet pump parameter criteria. However, some kidneys that were discarded may have gone on to have function eventually, which would have skewed the results of DGF in favor of pulsatile perfusion. It is important to ascertain which population of donor kidneys would benefit the most from pulsatile machine perfusion. This distinction would be economically prudent with the cost associated with pulsatile perfusion and storage ranging from $1,500 per kidney and up. This does not include
2219
the cost of kidneys that are discarded because they do not meet pump parameter requirements. One would expect the costs to be much higher if this were factored in. By improving the DGF rate and graft survival other costs, such as postoperative dialysis and increased hospital stay, could be decreased. Therefore, we propose that pulsatile perfusion of DCD allografts may benefit the overall hospital budget in the long term. However, costbenefit analysis is required to support our findings. Setting costs aside, the benefit to transplant recipients should be the foremost concern. A decreased DGF rate and overall improved graft survival have obvious benefits to recipients mentally and physically, not to mention the time gained by earlier return to work. Our systematic review reveals that machine perfusion of DCD kidneys results in a statistically significant decrease in DGF. Although there was a trend toward improved 1-year graft survival, this result was not statistically significant. This finding at 1 year is not surprising, given that most grafts have good function 1 year after transplantation regardless of donor type. It was also difficult to ascertain from our literature review whether renal function was affected by pulsatile perfusion since studies did not provide the estimated glomerular filtration rate. We have been using mechanical pulsatile perfusion for all types of donor kidneys for several years. In our experience machine perfused DCD kidneys seem to benefit most and, thus, have a statistically lower glomerular filtration rate than their statically perfused counterparts (unpublished data). To our knowledge the more important question of whether machine perfusion will translate to superior outcomes after 5 or even 10 years remains to be studied.
CONCLUSIONS Our review shows that pulsatile perfusion appears to decrease the DGF rate in after DCD kidney transplantation. However, the impact of pulsatile perfusion on ECD and ECD-DCD donors requires further assessment. With the increasing age and comorbidity of organ donors, pulsatile perfusion may become an increasingly important tool for transplant programs to ensure improved shortterm and long-term outcomes. Also, with the advent of novel compounds added to preservation solutions that mitigate ischemia-reperfusion injury, at least in animal models, we must evaluate the influence on perfusion parameters and eventual graft function in future studies.
2220
RENAL PERFUSION VS COLD STORAGE FOR DONATION AFTER CARDIAC DEATH KIDNEYS
REFERENCES 1. Wolfe RA et al: Comparison of mortality in all patients on dialysis, patients on dialysis awaiting transplantation, and recipients of a first cadaveric transplant. N Engl J Med 1999; 341: 1725. 2. Stratta RJ, Rohr MS, Sundberg et al: Increased kidney transplantation utilizing expanded criteria deceased organ donors with results comparable to standard criteria donor transplant. Ann Surg 2004; 239: 688. 3. Yarlagadda SG, Klein CL and Jani A: Long-term renal outcomes after delayed graft function. Adv Chronic Kidney Dis 2008; 15: 248. 4. Perico N, Cattaneo D, Sayegh et al: Delayed graft function in kidney transplantation. Lancet 2004; 364: 1814. 5. Moers C, Smits JM, Maathuis MJ et al: Machine perfusion or cold storage in deceased-donor kidney transplantation. N Engl J Med 2009; 360: 7. 6. Kalbfleisch M, Lobb I, Mok A et al: Hydrogen sulphide improves real-time renal microvascular flow and attenuates renal injury following prolonged warm ischemia reperfusion injury. Can Urol Assoc J, suppl., 2011; 5: S165. 7. Jochmans I, Moers C, Smits JM et al: Machine perfusion versus cold storage for the preservation of kidneys donated after cardiac death: a multicenter, randomized, controlled trial. Ann Surg 2010; 252: 756.
8. Watson CJE, Wells AC, Roberts RJ et al: Cold machine perfusion versus static cold storage of kidneys donated after cardiac death: a UK multicenter randomized controlled trial. Am J Transplant 2010; 10: 1991. 9. Locke JE, Segev DL, Warren DS et al: Outcomes of kidneys from donors after cardiac death: Implications for allocation and preservation. Am J Transplant 2007; 7: 1797. 10. Marshall VC, Ross H, Scott DF et al: Preservation of cadaveric renal allografts-comparison of flushing and pumping techniques. Proc Eur Dial Transplant Assoc 1977; 14: 302. 11. Matsuno N, Sakurai E, Tamaki I et al: The effect of machine perfusion preservation versus cold storage on the function of kidneys from nonheart-beating donors. Transplantation 1994; 57: 293. 12. Moustafellos P, Hadjianastassiou V, Roy D et al: The influence of pulsatile preservation in kidney transplantation from non-heart-beating donors. Transplant Proc 2007; 39: 1323. 13. Plata-Munoz JJ, Muthusamy A, Quiroga I et al: Impact of pulsatile perfusion on postoperative outcome of kidneys from controlled donors after cardiac death. Transplant Int 2008; 21: 899. 14. Sheil AG, Drummond JM, Rogers JH et al: A controlled clinical trial of machine perfusion of
cadaveric donor renal allografts. Lancet 1975; 2: 287. 15. van der Vliet JA, Kievit JK, Héné RJ et al: Preservation of non-heart-beating donor kidneys: a clinical prospective randomised case-control study of machine perfusion versus cold storage. Transplant Proc 2001; 33: 847. 16. Nguan C, Sener A, Karnik V et al: Perfusion of renal allografts with verapamil improves graft function. Transplantation 2008; 86: 1463. 17. Alabbassi A, Tran KC, Bloch M et al: The effects of molecules CORM-3 and H2S on renal protection during pulsatile perfusion. Am J Transplant 2011; 2: A1614. 18. Lobb I, Mok A, Lan Z et al: Supplemental hydrogen sulphide protects transplant kidney function and prolongs recipient survival after prolonged cold ischaemia-reperfusion injury by mitigating renal graft apoptosis and inflammation. BJU Int 2012; 110: E1187. 19. Zhu JX, Kalbfleisch M, Yang YX et al: Detrimental effects of prolonged warm renal ischaemia-reperfusion injury are abrogated by supplemental hydrogen sulphide: an analysis using real-time intravital microscopy and polymerase chain reaction. BJU Int 2012; 110: E1218. 20. Wells AC, Rushworth L, Thiru S et al: Donor kidney disease and transplant outcome for kidneys donated after cardiac death. Br J Surg 2009; 96: 299.