- Email: [email protected]
Evolution of Renal Transplant Practice Over the Past Decade: A U.K. Center Experience
Evolution of Renal Transplant Practice Over the Past Decade: A U.K. Center Experience J. Natha,b, M. Fielda,*, S.R. Ebbsb, T. Smithb, D. McGrogana, J. Al-Shakarchia, J. Hodsonc, S. Mellora, and A. Readya a Department of Renal Surgery, University Hospitals Birmingham, Queen Elizabeth, UK; bSchool of Immunity and Infection, University of Birmingham, Birmingham, UK; and cWolfson Computer Laboratory, University Hospitals Birmingham, Queen Elizabeth, UK
ABSTRACT Objective. As renal transplantation continues to evolve, there appears to be a change in both donor and recipient populations. Traditional markers of high-risk donor (e.g. donation after cardiac death [DCD]/expanded criteria donor [ECD]) and recipient (e.g. obese, highly sensitized) operations appear to be more common without any noticeable worsening of patient outcome. The present study aimed to compare outcome and define the change in donor and recipient populations for cadaveric transplants over a 10-year period at a large U.K. center. Methods. Single-center analysis of all adult patients undergoing cadaveric renal transplantation between January 2004 and January 2014 (n ¼ 754). Transplants were divided into 3 groups (early, middle, and late) depending on the era, with donor, recipient and outcomes compared. Results. There were considerable changes in both donor and recipient factors between the 3 eras, with a greater proportion of high-risk operations performed, as reflected by significant increases in Donor Risk Index (median: 1.11e1.16, P ¼ .022), and the proportions of ECD (22.2%e33.9%, P ¼ .003) and DCD kidneys (10.8%e19.4% P ¼ .011). However, 1-year graft survival was comparable between the eras, with a decrease in the average 1-year serum creatinine between the early and late cohort (median: 161 mmol/L vs 132 mmol/L, P < .001). There was no significant increase in body mass index (BMI) in either the donor or recipient population across the eras. Conclusion. Improvement in transplant outcome continues despite a greater proportion of transplants previously considered as high risk being performed. This is likely to reflect a considerable improvement in pre- and postoperative management. BMI remains a major continuing block to transplantation.
T
HE EVOLUTION of renal transplantation over the past 60 years is a triumph of modern medicine, with patients undergoing organ transplant in the modern era expecting a 90% 1-year graft survival [1]. The decade-upondecade improvement in outcome is the consequence of many factors, including improved immunosuppressive, surgical, and perioperative standards. Such are the benefits of renal transplantation that greater numbers of patients are now offered this treatment, including those with complex medical problems, as there is a distinct survival advantage compared to dialysis even in high-risk patients [2e4].
Despite the ever-growing demand, transplantation is limited by the paucity of available organs, and great efforts have been made to expand the donor pool, reflected by increased transplant numbers both in the U.K. and worldwide [3,5]. Expansion of the living kidney program and widespread acceptance of both Extended Criteria Donor (ECD) and Donation after Circulatory Death (DCD) organs as suitable grafts has been central to this. *Address correspondence to Melanie Field, FRCS, Department of Renal Surgery, University Hospitals Birmingham, Queen Elizabeth, UK. E-mail: melfi[email protected]
0041-1345/15 http://dx.doi.org/10.1016/j.transproceed.2015.06.001
ª 2015 by Elsevier Inc. All rights reserved. 360 Park Avenue South, New York, NY 10010-1710
1700
Transplantation Proceedings, 47, 1700e1704 (2015)
EVOLUTION OF RENAL TRANSPLANT PRACTICE: UK
Various scoring systems have been derived to predict the likelihood and duration of post-transplant function. These have been based on multivariable analysis of donor and recipient factors on large datasets and have been prompted by the need for optimal organ allocation. The Kidney Donor Risk Index (KDRI) and the transplant calculator are the 2 most used in clinical practice, although the latter is only validated as a predictor of DGF [6,7]. An alternative to the KDRI was proposed based on a U.K. dataset and was based solely on donor characteristics [8]. Despite the encouraging outcomes for modern-day kidney transplants, there is a perception that both the donors and recipients of cadaveric kidneys have become more high risk. Some of the contributing factors (e.g. age, body mass index [BMI]) may simply be reflective of trends in the general population [9]. However, immunological complexity such as highly sensitized patients, often from multiple previous failed grafts, are transplant specific [10]. There is also an opinion that the expansion of the living donation kidney program has altered the overall group demographics of cadaveric recipients. As a group, recipients of living donor kidneys are younger, often with less comorbidity and often preemptive. Therefore, increased live donor numbers may have removed a cohort of straightforward transplants from the list, thus concentrating the more complex cases in the cadaveric pool. The aim of this study is to compare chronological cohorts of patients undergoing cadaveric renal transplant over the past decade at a single U.K. transplant center to determine whether perceived changed in donor and recipient risk are reflected in clinical practice and transplant outcome. METHODS Consecutive adult patients who underwent single-organ renal transplantation at Queen Elizabeth Hospital, Birmingham, between January 2004 and January 2014 were included. Donor and recipient details including age, gender, ethnicity, BMI, donor status, comorbidity, serum creatinine levels, HLA mismatch, peak PRA value, and dialysis status were obtained through the National Health Service Blood and Transplant data registry and a prospectively maintained in-house database. Renal transplantation was performed via a retroperitoneal approach as described elsewhere, with the right iliac fossa used preferentially [11]. Patients underwent standard transplantation workup, including chest radiograph, electrocardiogram, echocardiogram, myocardial perfusion imaging, coronary angiography, and pulmonary function tests, where appropriate.
Immunosuppression Regime All patients received 0.5 g methylprednisolone and 20 mg basiliximab (Simulect) at induction of anesthesia. Basiliximab was again given on day 4 postoperatively. Patients were routinely treated with a standardized triple immunosuppressive maintenance regime: mycophenolate mofetil, tacrolimus (Prograf) and prednisolone. However, for some patients in the early cohort, azathioprine, cyclosporine and mycophenolate mofetil were used.
1701
Statistical Methods The data were initially split into 3 “eras,” based on tertiles, to give approximately equal numbers of transplants in each group. A range of factors were then compared across these groups. JonckheereTerpstra tests were used for continuous factors, and Kendall’s s for dichotomous ones, in order to consider the chronological ordering of the periods being compared. Categorical variables with more than 2 levels were compared using Fisher’s exact test. Continuous data were assessed for normality prior to analysis, and all were found to follow skewed distributions. Hence, the continuous variables are reported as medians and ranges throughout. All analyses were performed using IBM SPSS Statistics 22 (IBM Corp., Armonk, N.Y., United States), with P < .05 deemed to be indicative of statistical significance.
RESULTS
During the 10-year study period, 1276 adult kidney transplants were performed at our center, of which 754 were cadaveric and met the inclusion criteria. These were divided into three chronological eras (n ¼ 251 in the first 2 eras, and n ¼ 252 in the final era) with the comparisons between the 3 eras for donor and recipient factors reported in Tables 1 and 2. Both donor and recipient age were found to have increased significantly over the period, from medians of 47 to 51 (P ¼ .035) and 46 to 49 years (P ¼ .018), respectively. There was no significant increase in either donor (P ¼ .154) or recipient (P ¼ .964) BMI during the study period. For cadaveric donors, the rates of hypertension increased from 16.8% to 25.5% (P ¼ .017), but there was no significant increase in the rate of diabetes (P ¼ .098). The distribution of causes of death also changed (P < .001), with deaths from anoxia increasing (4.8%e15.5%), whereas CVA deaths remained consistent (66.5%, 65.3%, and 67.1% in the 3 eras). There was a significant increase in the KDRI over the time period, which increased from a median of 1.11 in the first era to 1.16 in the last (P ¼ .022). Although modest in absolute terms, this has prognostic implications, with expected median graft survival falling from 10.8 to 9.2 years [6]. Furthermore, the increase in donor KDRI was tampered by a significant decrease in the median cold ischemia time (CIT) between groups from 17.5 to 15.8 hours (P < .001). Indeed the KDRI would have risen further if such hastening of CIT had not occurred. Other markers of high-risk grafts also increased during the study period. The proportion of ECD kidneys increased from 22.2% to 33.9% (P ¼ .003) with a similar pattern for DCD kidneys (10.8%e19.4%, P ¼ .011). The proportion of highly sensitized patients (PRA >80%) increased from 8.4% to 13.5% (P < .001), whereas the proportion of patients with more than 2 HLA mismatches rose from 43% to 60% (P ¼ .032). Of the outcomes considered (Table 3), both patient (P ¼ .157) and graft (death censored) (P ¼ .330) survival were similar across the 3 periods, changing from 96.4% to 98.4% and 91.3% to 88.7%, respectively. However, renal function
1702
NATH, FIELD, EBBS ET AL Table 1. Comparison of Donor Factors Across the Three Periods Jan 1, 2004 to Jun 10, 2007
Age (y) Sex (male) Ethnicity Asian Black Other White BMI (kg/m2) Obese (BMI >30 kg/m2) Diabetes Hypertension DCD kidney Cause of death† CVA Anoxia Other Terminal creatinine (mmol/L) Total HLA mismatches 0 1 2 3 4 5 6 CIT (h) KDRI Extended criteria donor
Jun 11, 2007 to Dec 25, 2010
Dec 26, 2010 to Jan 1, 2014
47 (7e74) 142 (56.6%)
48 (12e72) 126 (50.2%)
51 (13e76) 131 (52.0%)
11 1 2 237 24.9 33 6 41 27
(4.4%) (0.4%) (0.8%) (94.4%) (12.6e47.3) (13.1%) (2.4%) (16.8%) (10.8%)
0 5 4 242 25.7 45 11 47 52
(0.0%) (2.0%) (1.6%) (96.4%) (15.6e50.0) (17.9%) (4.5%) (19.3%) (20.7%)
5 3 6 233 25.5 45 13 63 49
(2.0%) (1.2%) (2.4%) (94.3%) (16.0e52.1) (17.9%) (5.2%) (25.5%) (19.4%)
167 12 72 76
(66.5%) (4.8%) (28.7%) (41e408)
164 28 59 74
(65.3%) (11.2%) (23.5%) (18e198)
169 39 44 73
(67.1%) (15.5%) (17.5%) (25e226)
36 14 83 49 39 7 6 17.5 1.11 55
(15.4%) (6.0%) (35.5%) (20.9%) (16.7%) (3.0%) (2.6%) (8.0e38.3) (0.61e2.51) (22.2%)
37 14 63 92 45 0 0 17.0 1.07 53
(14.7%) (5.6%) (25.1%) (36.7%) (17.9%) (0.0%) (0.0%) (7.5e39.0) (0.51e2.62) (21.9%)
33 9 58 97 54 1 0 15.8 1.16 82
(13.1%) (3.6%) (23.0%) (38.5%) (21.4%) (0.4%) (0.0%) (5.5e30.0) (0.58e2.19) (33.9%)
P-Value
.035* .303 .004*†
.154 .163 .098 .017* .011* <.001*†
.014* .032*
<.001* .022* .003*
Data reported as: “Median (Range)”, with P values from JonckheereeTerpstra test, or “N (%)”, with P values from Kendall’s s, as applicable, unless stated otherwise. *Significant at P < .05. † P value from Fisher’s exact test.
improved across the study period, with a significant decrease in 1-year creatinine levels between the early and later era noted (median: 161 mmol/L vs. 132 mmol/L, P < .001). Table 2. Comparison of Recipient Factors Across the 3 Periods Jan 1, 2004 to Jun 11, 2007 to Dec 26, 2010 to Jun 10, 2007 Dec 25, 2010 Jan 1, 2014
Age (y)† Sex (male) Ethnicity Asian Black Other White BMI (kg/m2)
46 (16e72) 47 (16e77) 148 (59.0%) 148 (59.0%)
49 (17e75) 153 (60.7%)
60 (23.9%) 56 (22.3%) 18 (7.2%) 20 (8.0%) 2 (0.8%) 4 (1.6%) 171 (68.1%) 171 (68.1%) 26.1 26.1 (15.2e40.6) (16.9e38.8) BMI >30 (kg/m2) 49 (24.0%) 54 (26.7%) Graft number 2þ 28 (11.2%) 33 (13.1%) PRA 0 193 (77.2%) 177 (70.5%) 1e80 36 (14.4%) 48 (19.1%) >80 21 (8.4%) 26 (10.4%)
76 (30.2%) 20 (7.9%) 2 (0.8%) 154 (61.1%) 26.3 (17.1e37.9) 62 (24.8%) 28 (11.1%)
P Value
.018* .699 .457†
.964 .902 .961 <.001*
133 (52.8%) 85 (33.7%) 34 (13.5%)
Data reported as: “Median (Range)”, with P values from Jonckheere-Terpstra test, or “N (%)”, with P values from Kendall’s s, as applicable, unless stated otherwise. *Significant at P < .05. † P value from Fisher’s exact test.
DISCUSSION
Our data support the long-held assertion that recipients and donors are becoming increasingly complex. We have demonstrated significant increases in age (in both donor and recipient), donor rates of hypertension, donor deaths to anoxia, increase in KDRI, increase in the number of ECD and DCD kidneys transplanted, increases in the sensitisation level of patients, and increases in the number of HLA mismatches. Despite such increases in marginality of both donors and recipients, outcomes have not only been maintained but have shown improvement, with superior graft function noted in the later cohort. This is reassuring against Table 3. Transplant Outcome Measures Across the 3 Periods Jun 11, 2007 to Dec 26, 2010 Jan 1, 2004 to Jun 10, 2007 Dec 25, 2010 to Jan 1, 2014 P Value
1-year patient survival 1-year graft survival† 1-year creatinine
242 (96.4%) 242 (96.4%) 247 (98.4%)
.157
221 (91.3%) 221 (91.3%) 216 (88.7%)
.330
161
127
132
< .001*
Data reported as: “Median” with P values from JonckheereeTerpstra test, or “N (%)”, with P values from Kendall’s Tau, as applicable. *Significant at P < .05. † Censored at death.
EVOLUTION OF RENAL TRANSPLANT PRACTICE: UK
the background of increased complexity and is likely to reflect improvements in perioperative management. There have been several changes in practice that have occurred during the study period that counter the changes in recipient and donor characteristics. Although no new major class of drug entered into routine use during the study period, the evolution of immunosuppression has continued and may be responsible for some of the improvement in outcome [12]. Over the study period, the proportion of donor deaths attributed to anoxia rose significantly from 4.8% to 15.5%. It is hard to determine exactly the reasons for this. It may be this rise represents an increase in the number of DCD donors in the latter era. Kidneys in the latter era were less well matched than in the earlier era. It may be this reflects the changes to the allocation system, with greater priority being weighted towards other factors other than the HLA mismatch. It may be this represents an inherent mismatch in our donor population (predominantly Caucasian) and our recipient population (with a high proportion of non-Caucasians). During the study period, hypothermic machine perfusion was adopted into routine clinical practice. Results from the machine perfusion trial have demonstrated that this can improve 1- and 3-year graft survival for cadaveric kidneys overall [13,14]. The benefit of machine perfusion appears most pronounced for expanded criteria kidneys and may be partly responsible for limiting the effects of these marginal organs within our population [15]. Not all donor/recipient factors worsened over the study period, with an improvement in the donor terminal creatinine and CIT noted. The decrease in terminal creatinine is perhaps surprising, particularly in light of the increasing age of donors and by association the likelihood of reduced glomerular filtration rate. A possible explanation for this finding could be due to improved donor management in the intensive care setting, with goal-directed therapy often begun by specialist teams within the U.K. prior to the organ retrieval process [16]. The reduction in CIT at our center, the only modifiable transplant risk factor, has been an active development and the reasons for this observation are likely to be multifactorial. However, routine implementation of the “virtual crossmatch” for patients with undetectable panel reactive antibodies has negated the need for a lengthy crossmatch period in this cohort of patients and has been a major contributing factor. Live donor transplants have been excluded from this analysis to avoid nonmeaningful comparisons between inherently different groups, especially given the expansion of the live donor program during the study period. Interestingly, live donors and recipients were found to be younger than for cadaveric kidneys (median: 46 years and 44 years) with comparable BMIs. However, waiting list times for live donor recipients are much lower (384 days vs 1344 days) in our population. The notion that the live donor population has filtered some of the “easier” predialysis
1703
patients from the transplant list therefore appears valid, perhaps increasing the complexity in the remaining cadaveric pool. However, to balance this it must also be considered that the live donor pool also absorbs some of the most complex transplants (e.g. ABO-I, HLA-I) and some potential recipients may have technical factors rendering them only suitable for a planned live donor implant. This study is limited by only looking at relatively shortterm outcomes from a single, albeit large, center. However, despite these limitations it does serve to illustrate the point that the characteristics of donor and recipient groups have changed and transplants that were once considered high risk are now performed routinely. Although overall, donors and recipients have undoubtedly become more complex, there has been no change in the proportion of obese patients transplanted. Reported outcomes for obese patients receiving kidney transplants are inconclusive [17e20], with no clear upper guidelines as to the upper limit of acceptable BMI for transplantation. In our population, outcomes for this group were worse, with risks exaggerated in kidneys with multiple arteries [21]. The static BMI of recipients in this study does not reflect the increase in BMI seen in the general population or the number of obese patients assessed for transplantation at our center and, as such, obesity appears to remain a significant barrier to transplantation. In the future, both the donor and recipient pool are likely to show further evolution and present further challenges for transplantation. Indeed it is estimated that almost a third of patients commencing dialysis are over 70 [22]. If the changes in the last decade can be maintained, provision for this evolution should be adequate. BMI remains a major continuing block to transplantation and careful consideration will be needed in order to best serve these patients. CONCLUSION
Although limited to a single center, our results show that both donors and recipients of cadaveric kidneys are increasingly more complex. Despite increased risk, outcomes have improved over the study period, suggesting a significant improvement in transplant care. It is likely that both donors and recipients will become increasingly more complex and that the challenge will be how to improve or maintain renal transplant outcomes in this era of evolution. ACKNOWLEDGEMENTS The authors are grateful to NHSBT for supplying some of the data presented. This work was supported through a grant from University Hospital Birmingham Charities.
REFERENCES [1] NHSBT. Organ Donation and Transplantation: Acitvity Report 2013-2014. 2014. [2] Wolfe R, Ashby V, Milford E, Ojo A, Ettenger R, Agodoa L, et al. Comparison of mortality in all patients on dialysis, patients on
1704 dialysis awaiting transplantation, and recipients of a first cadaveric transplant. N Engl J Med 1999;341:1725e30. [3] Gilg J, Rao A, Fogarty D. UK renal replacement therapy Incidence in 2012: national and centre-specific analyses. Nephron Clin Pract 2013;125:1e27. [4] Oniscu G, Brown H, Forsythe J. How great is the survival advantage of transplantation over dialysis in elderly patients? Nephrol Dial Transplant 2004;19:945e51. [5] Orandi B, Garonzik-Wang J, Massie A, Zachary A, Montgomery RA, Van Arendonk K, et al. Quantifying the risk of incompatible kidney transplantation: a multicenter study. Am J Transplant 2014;14:1573e80. [6] Rao P, Schaubel D, Guidinger M, Andreoni K, Wolfe R, Merion R, et al. A comprehensive risk quantification score for deceased donor kidneys: the kidney donor risk index. Transplantation 2009;88:231e6. [7] Irish W, Ilsley J, Schnitzler M, Feng S, Brennan D. A risk prediction model for delayed graft function in the current era of deceased donor renal transplantation. Am J Transplant 2010;10: 2279e86. [8] Watson C, Johnson R, Birch R, Collett D, Bradley J. A simplified donor risk index for predictin outcome after deceased donor kidney transplantation. Transplantation 2012;93:314e8. [9] Health &Social Care Information Centre: Statistics on Obesity, Physical Activity and Diet - England, 2013 [updated Feb 20, 2013]. Available at: http://www.hscic.gov.uk/catalogue/PUB10364. Accessed April 2015. [10] Sharif A, Alachkar N, Kraus E. Incompatible kidney transplantation: a brief overview of the past, present and future. QJM 2012;105:1141e50. [11] Allen RDM. Vascular complications after kidney transplantation. In: Morris P, Knechtle SJ, editors. Kidney Transplantation: Principles and Practice. 6th ed. UK: Saunders; 2008. p. 439e61. [12] Ekberg H, Tedesco-Silva H, Demirbas A, Vitko S, Nashan B, Gurkan A, et al. Reduced exposure to calcineurin
NATH, FIELD, EBBS ET AL inhibitors in renal transplantation. N Engl J Med 2007;357: 2562e75. [13] Moers C, Smits J, Maathuis M, Treckmann J, van Gelder F, Napieraslski B, et al. Machine perfusion or cold storage in deceased-donor kidney transplantation. N Engl J Med 2009;360: 7e19. [14] Moers C, Pirenne J, Paul A, Ploeg R, Group MPTS. Machine perfusion or cold storage in deceased-donor kidney transplantation. N Engl J Med 2012;366:770e1. [15] O’Callaghan J, Morgan R, Knight S, Morris P. Systematic review and meta-analysis of hypothermic machine perfusion versus static cold storage of kidney allografts on transplant outcome. BJS 2012;94(suppl):277. [16] Kustogiannis D, Pagliarello G, Doig C, Ross H, Shemie S. Medical management to optimize donor organ potential : a review of the literature. Can J Anaesth 2006;53:820e30. [17] Chang S, Coates P, Mcdonald S. Effects of body mass index at transplant on outcomes of kidney transplatation. Transplantation 2007;84:981e7. [18] Bennett W, McEvoy K, Henell K, Valete J, Douzdjian V. Morbid obesity does not preculde successful renal transplantation. Clin Transplant 2004;8:89e93. [19] Grosso G, Corona D, Mistretta A, Zerbo D, Sinagra N, Giaquinta A, et al. The role of obesity in kidney transplantation outcome. Transplant Proc 2012;44:1864e8. [20] Meier-Kriesche H, Arndorfer J, Kaplan B. The impact of body mass index on renal transplant outcomes: a signficiant independent risk factor for graft failure and patient death. Transplantation 2002;73:70e4. [21] Nath J, Mastoridis S, van Dellen D, Guy A, McGrogan D, Krishnan H, et al. Complex kidneys for complex patients: the risk associated with transplantation of kineys with multiple arteries into obese patients. Transplant Proc 2015;47:373e8. [22] Knoll G. Kidney transplantation in the older adult. Am J Kidney Dis 2013;61(5):790e7.
ABSTRACT Objective. As renal transplantation continues to evolve, there appears to be a change in both donor and recipient populations. Traditional markers of high-risk donor (e.g. donation after cardiac death [DCD]/expanded criteria donor [ECD]) and recipient (e.g. obese, highly sensitized) operations appear to be more common without any noticeable worsening of patient outcome. The present study aimed to compare outcome and define the change in donor and recipient populations for cadaveric transplants over a 10-year period at a large U.K. center. Methods. Single-center analysis of all adult patients undergoing cadaveric renal transplantation between January 2004 and January 2014 (n ¼ 754). Transplants were divided into 3 groups (early, middle, and late) depending on the era, with donor, recipient and outcomes compared. Results. There were considerable changes in both donor and recipient factors between the 3 eras, with a greater proportion of high-risk operations performed, as reflected by significant increases in Donor Risk Index (median: 1.11e1.16, P ¼ .022), and the proportions of ECD (22.2%e33.9%, P ¼ .003) and DCD kidneys (10.8%e19.4% P ¼ .011). However, 1-year graft survival was comparable between the eras, with a decrease in the average 1-year serum creatinine between the early and late cohort (median: 161 mmol/L vs 132 mmol/L, P < .001). There was no significant increase in body mass index (BMI) in either the donor or recipient population across the eras. Conclusion. Improvement in transplant outcome continues despite a greater proportion of transplants previously considered as high risk being performed. This is likely to reflect a considerable improvement in pre- and postoperative management. BMI remains a major continuing block to transplantation.
T
HE EVOLUTION of renal transplantation over the past 60 years is a triumph of modern medicine, with patients undergoing organ transplant in the modern era expecting a 90% 1-year graft survival [1]. The decade-upondecade improvement in outcome is the consequence of many factors, including improved immunosuppressive, surgical, and perioperative standards. Such are the benefits of renal transplantation that greater numbers of patients are now offered this treatment, including those with complex medical problems, as there is a distinct survival advantage compared to dialysis even in high-risk patients [2e4].
Despite the ever-growing demand, transplantation is limited by the paucity of available organs, and great efforts have been made to expand the donor pool, reflected by increased transplant numbers both in the U.K. and worldwide [3,5]. Expansion of the living kidney program and widespread acceptance of both Extended Criteria Donor (ECD) and Donation after Circulatory Death (DCD) organs as suitable grafts has been central to this. *Address correspondence to Melanie Field, FRCS, Department of Renal Surgery, University Hospitals Birmingham, Queen Elizabeth, UK. E-mail: melfi[email protected]
0041-1345/15 http://dx.doi.org/10.1016/j.transproceed.2015.06.001
ª 2015 by Elsevier Inc. All rights reserved. 360 Park Avenue South, New York, NY 10010-1710
1700
Transplantation Proceedings, 47, 1700e1704 (2015)
EVOLUTION OF RENAL TRANSPLANT PRACTICE: UK
Various scoring systems have been derived to predict the likelihood and duration of post-transplant function. These have been based on multivariable analysis of donor and recipient factors on large datasets and have been prompted by the need for optimal organ allocation. The Kidney Donor Risk Index (KDRI) and the transplant calculator are the 2 most used in clinical practice, although the latter is only validated as a predictor of DGF [6,7]. An alternative to the KDRI was proposed based on a U.K. dataset and was based solely on donor characteristics [8]. Despite the encouraging outcomes for modern-day kidney transplants, there is a perception that both the donors and recipients of cadaveric kidneys have become more high risk. Some of the contributing factors (e.g. age, body mass index [BMI]) may simply be reflective of trends in the general population [9]. However, immunological complexity such as highly sensitized patients, often from multiple previous failed grafts, are transplant specific [10]. There is also an opinion that the expansion of the living donation kidney program has altered the overall group demographics of cadaveric recipients. As a group, recipients of living donor kidneys are younger, often with less comorbidity and often preemptive. Therefore, increased live donor numbers may have removed a cohort of straightforward transplants from the list, thus concentrating the more complex cases in the cadaveric pool. The aim of this study is to compare chronological cohorts of patients undergoing cadaveric renal transplant over the past decade at a single U.K. transplant center to determine whether perceived changed in donor and recipient risk are reflected in clinical practice and transplant outcome. METHODS Consecutive adult patients who underwent single-organ renal transplantation at Queen Elizabeth Hospital, Birmingham, between January 2004 and January 2014 were included. Donor and recipient details including age, gender, ethnicity, BMI, donor status, comorbidity, serum creatinine levels, HLA mismatch, peak PRA value, and dialysis status were obtained through the National Health Service Blood and Transplant data registry and a prospectively maintained in-house database. Renal transplantation was performed via a retroperitoneal approach as described elsewhere, with the right iliac fossa used preferentially [11]. Patients underwent standard transplantation workup, including chest radiograph, electrocardiogram, echocardiogram, myocardial perfusion imaging, coronary angiography, and pulmonary function tests, where appropriate.
Immunosuppression Regime All patients received 0.5 g methylprednisolone and 20 mg basiliximab (Simulect) at induction of anesthesia. Basiliximab was again given on day 4 postoperatively. Patients were routinely treated with a standardized triple immunosuppressive maintenance regime: mycophenolate mofetil, tacrolimus (Prograf) and prednisolone. However, for some patients in the early cohort, azathioprine, cyclosporine and mycophenolate mofetil were used.
1701
Statistical Methods The data were initially split into 3 “eras,” based on tertiles, to give approximately equal numbers of transplants in each group. A range of factors were then compared across these groups. JonckheereTerpstra tests were used for continuous factors, and Kendall’s s for dichotomous ones, in order to consider the chronological ordering of the periods being compared. Categorical variables with more than 2 levels were compared using Fisher’s exact test. Continuous data were assessed for normality prior to analysis, and all were found to follow skewed distributions. Hence, the continuous variables are reported as medians and ranges throughout. All analyses were performed using IBM SPSS Statistics 22 (IBM Corp., Armonk, N.Y., United States), with P < .05 deemed to be indicative of statistical significance.
RESULTS
During the 10-year study period, 1276 adult kidney transplants were performed at our center, of which 754 were cadaveric and met the inclusion criteria. These were divided into three chronological eras (n ¼ 251 in the first 2 eras, and n ¼ 252 in the final era) with the comparisons between the 3 eras for donor and recipient factors reported in Tables 1 and 2. Both donor and recipient age were found to have increased significantly over the period, from medians of 47 to 51 (P ¼ .035) and 46 to 49 years (P ¼ .018), respectively. There was no significant increase in either donor (P ¼ .154) or recipient (P ¼ .964) BMI during the study period. For cadaveric donors, the rates of hypertension increased from 16.8% to 25.5% (P ¼ .017), but there was no significant increase in the rate of diabetes (P ¼ .098). The distribution of causes of death also changed (P < .001), with deaths from anoxia increasing (4.8%e15.5%), whereas CVA deaths remained consistent (66.5%, 65.3%, and 67.1% in the 3 eras). There was a significant increase in the KDRI over the time period, which increased from a median of 1.11 in the first era to 1.16 in the last (P ¼ .022). Although modest in absolute terms, this has prognostic implications, with expected median graft survival falling from 10.8 to 9.2 years [6]. Furthermore, the increase in donor KDRI was tampered by a significant decrease in the median cold ischemia time (CIT) between groups from 17.5 to 15.8 hours (P < .001). Indeed the KDRI would have risen further if such hastening of CIT had not occurred. Other markers of high-risk grafts also increased during the study period. The proportion of ECD kidneys increased from 22.2% to 33.9% (P ¼ .003) with a similar pattern for DCD kidneys (10.8%e19.4%, P ¼ .011). The proportion of highly sensitized patients (PRA >80%) increased from 8.4% to 13.5% (P < .001), whereas the proportion of patients with more than 2 HLA mismatches rose from 43% to 60% (P ¼ .032). Of the outcomes considered (Table 3), both patient (P ¼ .157) and graft (death censored) (P ¼ .330) survival were similar across the 3 periods, changing from 96.4% to 98.4% and 91.3% to 88.7%, respectively. However, renal function
1702
NATH, FIELD, EBBS ET AL Table 1. Comparison of Donor Factors Across the Three Periods Jan 1, 2004 to Jun 10, 2007
Age (y) Sex (male) Ethnicity Asian Black Other White BMI (kg/m2) Obese (BMI >30 kg/m2) Diabetes Hypertension DCD kidney Cause of death† CVA Anoxia Other Terminal creatinine (mmol/L) Total HLA mismatches 0 1 2 3 4 5 6 CIT (h) KDRI Extended criteria donor
Jun 11, 2007 to Dec 25, 2010
Dec 26, 2010 to Jan 1, 2014
47 (7e74) 142 (56.6%)
48 (12e72) 126 (50.2%)
51 (13e76) 131 (52.0%)
11 1 2 237 24.9 33 6 41 27
(4.4%) (0.4%) (0.8%) (94.4%) (12.6e47.3) (13.1%) (2.4%) (16.8%) (10.8%)
0 5 4 242 25.7 45 11 47 52
(0.0%) (2.0%) (1.6%) (96.4%) (15.6e50.0) (17.9%) (4.5%) (19.3%) (20.7%)
5 3 6 233 25.5 45 13 63 49
(2.0%) (1.2%) (2.4%) (94.3%) (16.0e52.1) (17.9%) (5.2%) (25.5%) (19.4%)
167 12 72 76
(66.5%) (4.8%) (28.7%) (41e408)
164 28 59 74
(65.3%) (11.2%) (23.5%) (18e198)
169 39 44 73
(67.1%) (15.5%) (17.5%) (25e226)
36 14 83 49 39 7 6 17.5 1.11 55
(15.4%) (6.0%) (35.5%) (20.9%) (16.7%) (3.0%) (2.6%) (8.0e38.3) (0.61e2.51) (22.2%)
37 14 63 92 45 0 0 17.0 1.07 53
(14.7%) (5.6%) (25.1%) (36.7%) (17.9%) (0.0%) (0.0%) (7.5e39.0) (0.51e2.62) (21.9%)
33 9 58 97 54 1 0 15.8 1.16 82
(13.1%) (3.6%) (23.0%) (38.5%) (21.4%) (0.4%) (0.0%) (5.5e30.0) (0.58e2.19) (33.9%)
P-Value
.035* .303 .004*†
.154 .163 .098 .017* .011* <.001*†
.014* .032*
<.001* .022* .003*
Data reported as: “Median (Range)”, with P values from JonckheereeTerpstra test, or “N (%)”, with P values from Kendall’s s, as applicable, unless stated otherwise. *Significant at P < .05. † P value from Fisher’s exact test.
improved across the study period, with a significant decrease in 1-year creatinine levels between the early and later era noted (median: 161 mmol/L vs. 132 mmol/L, P < .001). Table 2. Comparison of Recipient Factors Across the 3 Periods Jan 1, 2004 to Jun 11, 2007 to Dec 26, 2010 to Jun 10, 2007 Dec 25, 2010 Jan 1, 2014
Age (y)† Sex (male) Ethnicity Asian Black Other White BMI (kg/m2)
46 (16e72) 47 (16e77) 148 (59.0%) 148 (59.0%)
49 (17e75) 153 (60.7%)
60 (23.9%) 56 (22.3%) 18 (7.2%) 20 (8.0%) 2 (0.8%) 4 (1.6%) 171 (68.1%) 171 (68.1%) 26.1 26.1 (15.2e40.6) (16.9e38.8) BMI >30 (kg/m2) 49 (24.0%) 54 (26.7%) Graft number 2þ 28 (11.2%) 33 (13.1%) PRA 0 193 (77.2%) 177 (70.5%) 1e80 36 (14.4%) 48 (19.1%) >80 21 (8.4%) 26 (10.4%)
76 (30.2%) 20 (7.9%) 2 (0.8%) 154 (61.1%) 26.3 (17.1e37.9) 62 (24.8%) 28 (11.1%)
P Value
.018* .699 .457†
.964 .902 .961 <.001*
133 (52.8%) 85 (33.7%) 34 (13.5%)
Data reported as: “Median (Range)”, with P values from Jonckheere-Terpstra test, or “N (%)”, with P values from Kendall’s s, as applicable, unless stated otherwise. *Significant at P < .05. † P value from Fisher’s exact test.
DISCUSSION
Our data support the long-held assertion that recipients and donors are becoming increasingly complex. We have demonstrated significant increases in age (in both donor and recipient), donor rates of hypertension, donor deaths to anoxia, increase in KDRI, increase in the number of ECD and DCD kidneys transplanted, increases in the sensitisation level of patients, and increases in the number of HLA mismatches. Despite such increases in marginality of both donors and recipients, outcomes have not only been maintained but have shown improvement, with superior graft function noted in the later cohort. This is reassuring against Table 3. Transplant Outcome Measures Across the 3 Periods Jun 11, 2007 to Dec 26, 2010 Jan 1, 2004 to Jun 10, 2007 Dec 25, 2010 to Jan 1, 2014 P Value
1-year patient survival 1-year graft survival† 1-year creatinine
242 (96.4%) 242 (96.4%) 247 (98.4%)
.157
221 (91.3%) 221 (91.3%) 216 (88.7%)
.330
161
127
132
< .001*
Data reported as: “Median” with P values from JonckheereeTerpstra test, or “N (%)”, with P values from Kendall’s Tau, as applicable. *Significant at P < .05. † Censored at death.
EVOLUTION OF RENAL TRANSPLANT PRACTICE: UK
the background of increased complexity and is likely to reflect improvements in perioperative management. There have been several changes in practice that have occurred during the study period that counter the changes in recipient and donor characteristics. Although no new major class of drug entered into routine use during the study period, the evolution of immunosuppression has continued and may be responsible for some of the improvement in outcome [12]. Over the study period, the proportion of donor deaths attributed to anoxia rose significantly from 4.8% to 15.5%. It is hard to determine exactly the reasons for this. It may be this rise represents an increase in the number of DCD donors in the latter era. Kidneys in the latter era were less well matched than in the earlier era. It may be this reflects the changes to the allocation system, with greater priority being weighted towards other factors other than the HLA mismatch. It may be this represents an inherent mismatch in our donor population (predominantly Caucasian) and our recipient population (with a high proportion of non-Caucasians). During the study period, hypothermic machine perfusion was adopted into routine clinical practice. Results from the machine perfusion trial have demonstrated that this can improve 1- and 3-year graft survival for cadaveric kidneys overall [13,14]. The benefit of machine perfusion appears most pronounced for expanded criteria kidneys and may be partly responsible for limiting the effects of these marginal organs within our population [15]. Not all donor/recipient factors worsened over the study period, with an improvement in the donor terminal creatinine and CIT noted. The decrease in terminal creatinine is perhaps surprising, particularly in light of the increasing age of donors and by association the likelihood of reduced glomerular filtration rate. A possible explanation for this finding could be due to improved donor management in the intensive care setting, with goal-directed therapy often begun by specialist teams within the U.K. prior to the organ retrieval process [16]. The reduction in CIT at our center, the only modifiable transplant risk factor, has been an active development and the reasons for this observation are likely to be multifactorial. However, routine implementation of the “virtual crossmatch” for patients with undetectable panel reactive antibodies has negated the need for a lengthy crossmatch period in this cohort of patients and has been a major contributing factor. Live donor transplants have been excluded from this analysis to avoid nonmeaningful comparisons between inherently different groups, especially given the expansion of the live donor program during the study period. Interestingly, live donors and recipients were found to be younger than for cadaveric kidneys (median: 46 years and 44 years) with comparable BMIs. However, waiting list times for live donor recipients are much lower (384 days vs 1344 days) in our population. The notion that the live donor population has filtered some of the “easier” predialysis
1703
patients from the transplant list therefore appears valid, perhaps increasing the complexity in the remaining cadaveric pool. However, to balance this it must also be considered that the live donor pool also absorbs some of the most complex transplants (e.g. ABO-I, HLA-I) and some potential recipients may have technical factors rendering them only suitable for a planned live donor implant. This study is limited by only looking at relatively shortterm outcomes from a single, albeit large, center. However, despite these limitations it does serve to illustrate the point that the characteristics of donor and recipient groups have changed and transplants that were once considered high risk are now performed routinely. Although overall, donors and recipients have undoubtedly become more complex, there has been no change in the proportion of obese patients transplanted. Reported outcomes for obese patients receiving kidney transplants are inconclusive [17e20], with no clear upper guidelines as to the upper limit of acceptable BMI for transplantation. In our population, outcomes for this group were worse, with risks exaggerated in kidneys with multiple arteries [21]. The static BMI of recipients in this study does not reflect the increase in BMI seen in the general population or the number of obese patients assessed for transplantation at our center and, as such, obesity appears to remain a significant barrier to transplantation. In the future, both the donor and recipient pool are likely to show further evolution and present further challenges for transplantation. Indeed it is estimated that almost a third of patients commencing dialysis are over 70 [22]. If the changes in the last decade can be maintained, provision for this evolution should be adequate. BMI remains a major continuing block to transplantation and careful consideration will be needed in order to best serve these patients. CONCLUSION
Although limited to a single center, our results show that both donors and recipients of cadaveric kidneys are increasingly more complex. Despite increased risk, outcomes have improved over the study period, suggesting a significant improvement in transplant care. It is likely that both donors and recipients will become increasingly more complex and that the challenge will be how to improve or maintain renal transplant outcomes in this era of evolution. ACKNOWLEDGEMENTS The authors are grateful to NHSBT for supplying some of the data presented. This work was supported through a grant from University Hospital Birmingham Charities.
REFERENCES [1] NHSBT. Organ Donation and Transplantation: Acitvity Report 2013-2014. 2014. [2] Wolfe R, Ashby V, Milford E, Ojo A, Ettenger R, Agodoa L, et al. Comparison of mortality in all patients on dialysis, patients on
1704 dialysis awaiting transplantation, and recipients of a first cadaveric transplant. N Engl J Med 1999;341:1725e30. [3] Gilg J, Rao A, Fogarty D. UK renal replacement therapy Incidence in 2012: national and centre-specific analyses. Nephron Clin Pract 2013;125:1e27. [4] Oniscu G, Brown H, Forsythe J. How great is the survival advantage of transplantation over dialysis in elderly patients? Nephrol Dial Transplant 2004;19:945e51. [5] Orandi B, Garonzik-Wang J, Massie A, Zachary A, Montgomery RA, Van Arendonk K, et al. Quantifying the risk of incompatible kidney transplantation: a multicenter study. Am J Transplant 2014;14:1573e80. [6] Rao P, Schaubel D, Guidinger M, Andreoni K, Wolfe R, Merion R, et al. A comprehensive risk quantification score for deceased donor kidneys: the kidney donor risk index. Transplantation 2009;88:231e6. [7] Irish W, Ilsley J, Schnitzler M, Feng S, Brennan D. A risk prediction model for delayed graft function in the current era of deceased donor renal transplantation. Am J Transplant 2010;10: 2279e86. [8] Watson C, Johnson R, Birch R, Collett D, Bradley J. A simplified donor risk index for predictin outcome after deceased donor kidney transplantation. Transplantation 2012;93:314e8. [9] Health &Social Care Information Centre: Statistics on Obesity, Physical Activity and Diet - England, 2013 [updated Feb 20, 2013]. Available at: http://www.hscic.gov.uk/catalogue/PUB10364. Accessed April 2015. [10] Sharif A, Alachkar N, Kraus E. Incompatible kidney transplantation: a brief overview of the past, present and future. QJM 2012;105:1141e50. [11] Allen RDM. Vascular complications after kidney transplantation. In: Morris P, Knechtle SJ, editors. Kidney Transplantation: Principles and Practice. 6th ed. UK: Saunders; 2008. p. 439e61. [12] Ekberg H, Tedesco-Silva H, Demirbas A, Vitko S, Nashan B, Gurkan A, et al. Reduced exposure to calcineurin
NATH, FIELD, EBBS ET AL inhibitors in renal transplantation. N Engl J Med 2007;357: 2562e75. [13] Moers C, Smits J, Maathuis M, Treckmann J, van Gelder F, Napieraslski B, et al. Machine perfusion or cold storage in deceased-donor kidney transplantation. N Engl J Med 2009;360: 7e19. [14] Moers C, Pirenne J, Paul A, Ploeg R, Group MPTS. Machine perfusion or cold storage in deceased-donor kidney transplantation. N Engl J Med 2012;366:770e1. [15] O’Callaghan J, Morgan R, Knight S, Morris P. Systematic review and meta-analysis of hypothermic machine perfusion versus static cold storage of kidney allografts on transplant outcome. BJS 2012;94(suppl):277. [16] Kustogiannis D, Pagliarello G, Doig C, Ross H, Shemie S. Medical management to optimize donor organ potential : a review of the literature. Can J Anaesth 2006;53:820e30. [17] Chang S, Coates P, Mcdonald S. Effects of body mass index at transplant on outcomes of kidney transplatation. Transplantation 2007;84:981e7. [18] Bennett W, McEvoy K, Henell K, Valete J, Douzdjian V. Morbid obesity does not preculde successful renal transplantation. Clin Transplant 2004;8:89e93. [19] Grosso G, Corona D, Mistretta A, Zerbo D, Sinagra N, Giaquinta A, et al. The role of obesity in kidney transplantation outcome. Transplant Proc 2012;44:1864e8. [20] Meier-Kriesche H, Arndorfer J, Kaplan B. The impact of body mass index on renal transplant outcomes: a signficiant independent risk factor for graft failure and patient death. Transplantation 2002;73:70e4. [21] Nath J, Mastoridis S, van Dellen D, Guy A, McGrogan D, Krishnan H, et al. Complex kidneys for complex patients: the risk associated with transplantation of kineys with multiple arteries into obese patients. Transplant Proc 2015;47:373e8. [22] Knoll G. Kidney transplantation in the older adult. Am J Kidney Dis 2013;61(5):790e7.