The United Network for Organ Sharing scientific database shows an ever-increasing disparity between the number of patients listed for kidney transplantation and the total transplants performed (1). There are a number of reasons for this increasing disparity, including the increasing number of patients who are listed for retransplantation after the failure of one or more previous transplants (3937 in 1988 versus 7043 in 1995)(1). In addition, with the improvement in outcome over the last 10-15 years, there are more patients on the waiting list who previously might not have been considered suitable for transplantation, in particular, the older patient (403 patients older than 65 years in 1988 versus 2120 in 1995) (1). Candidacy for transplantation has been liberalized as experience is gained; the criteria for acceptable donors, particularly older donors, should be reevaluated as well.
We and other centers have recently expanded our acceptance criteria for kidney donors, so that we now use expanded criteria donors(ECD*) more commonly, particularly older donors with a history of hypertension (2, 3). In several instances, we have transplanted both kidneys from an older donor into a single older, size-matched adult recipient (dual transplant). Nationally, in 1995, of 8598 cadaveric renal transplants (CRTs) performed, there were 192 (2%) double transplants, the majority (158 of 192; 82%) being transplanted as pediatric en bloc kidneys; only 32 of 192 (17%) were transplanted as adult to adult dual transplants. In 1996, of 8561 total CRTs performed, there were 228 (3%) double transplants, but the number of these that were dual transplants“doubled” to 76 of 228 (33%) (1). Although there are several recent reports describing the use of dual kidneys(2, 4, 5), it is currently unclear when ECD kidneys should be used for dual versus single kidney transplants. We have been transplanting both kidneys into a single recipient when the ECD admission creatinine clearance (AdCCr) is <90 ml/min. This policy has evolved over the last 18 months and is based on our experience (and that of others) of worse outcome in recipients of kidney transplants when the creatinine clearance of the donor kidneys, transplanted singly or together, was <50 ml/min (3). As we continue to gain more experience, perhaps we can identify simple clinical characteristics that will predict outcome when considering ECD kidneys for single versus dual transplants. To better understand outcome based on clinical donor and recipient characteristics, we retrospectively reviewed our ECD program over the last 2 years. We found that donor admission clearance creatinine (D-AdCCr)<90 ml/min, donor age ≥59, and cold storage (Cld Stg) time >24 hr independently affected early outcome, as measured by the incidence of delayed graft function (DGF) and mean serum creatinine up to 6 months after transplant, in recipients of ECD single versus dual kidneys.
A retrospective review of all adult CRTs performed in the Division of Transplantation at Stanford University Medical Center between January 1, 1995 and November 15, 1996 was performed. Of the 263 renal transplants performed during this period, 33 were kidney/pancreas, 6 were kidney/liver, 43 were pediatric, 55 were live-donor adult, and 126 were adult CRT. In the CRT group, there were 52 recipients of ECD kidneys that all other local transplant centers declined [previously reported as “The Kidneys that Nobody Wanted” (2)]. Within this group of ECD, there were 15 recipients of dual kidney transplants and 37 recipients of single transplants.
Donor and recipient medical records were reviewed to obtain data for comparison. We compared 7 donor variables and 16 recipient variables between the recipients of single versus dual ECD kidneys. The control group was the remaining 74 recipients of CRT performed during the same period. In the donors, a comparison was made of age, diagnosed hypertension, nadir systolic blood pressure, duration of hypotension (defined as the number of minutes the systolic blood pressure was <90 mmHg), the lowest urine output over a 4-hr period, peak creatinine, and calculated AdCCr in ml/min using the Cockcroft-Gault formula. Recipient variables were age, cold ischemia time, degree of HLA match, length of stay in days, DGF (defined as requirement for dialysis during the first posttransplant week), serum creatinine on posttransplant day 7 and at 1, 3, 6, 12, and 18 months, graft loss, and 1-year actuarial patient and graft survival. Graft loss was defined as a permanent return to dialysis or patient death. Graft loss was not censored for patient death. In one instance, a recipient of a dual graft lost one kidney, but remains free of dialysis and is thus not considered a graft loss. The protocol for accepting ECD kidneys and for placement of dual kidneys has been previously published (2). Our definition of an ECD kidney is one which all other local centers have declined. The reasons these 52 ECD kidneys were declined by other local centers included 25 because of advanced age (13 of which also had a history of hypertension and 4 of which were also declined because of 10-20% glomerular sclerosis on biopsy), 18 because of donor quality alone, 4 for biopsy results alone (10-20% glomerular sclerosis), 2 for anatomic abnormalities (horseshoe kidney), 2 for hypertension alone, and 1 for multiple vessels. In general, we have not used the results of biopsies to make decisions regarding the use of kidneys for transplantation; we have relied instead on functional data, e.g., the AdCCr and the donor quality including the maintenance of blood pressure and urine output. We have biopsy results for only 2 of 15 ECDs for our dual recipients. In a previous report, we found no correlation in recipients of ECD kidneys between biopsy results and outcome in the few ECD kidneys for which we have biopsy data(2), but the numbers are too small to make any useful conclusions. Most donor biopsies are wedge biopsies, which are taken near the capsule. In more aged kidneys, there will be more sclerosed glomeruli near the capsule, and we believe this is not an accurate reflection of the functional potential.
Statistical analysis. Data are expressed as mean ± SD unless indicated otherwise. Statistical significance was defined atP≤0.05. Statistical calculations were performed using the Statview 4.02 software program (Abacus Concepts, Berkeley, CA). Differences in donor and recipients variables were evaluated using analysis of variance(ANOVA) for continuous variables and the chi-square test for nominal variables. When nominal variables were less than 5, Fisher's exact test was used. One-year actuarial patient and graft survival rates were determined(SPSS Inc., Chicago, IL). Correlation coefficients to evaluate the degree of association between donor and recipient variables in the ECD single, dual, and control groups were tabulated in correlation matrices. Multiple regression was performed to identify variables related to recipient 1, 3, and 6 months after transplant serum creatinine.
Outcome in ECD used for single versus dual transplants. When evaluated independently, D-AdCCr <90 ml/min, D-age ≥59 years, and Cld Stg < or >24 hr, all had an impact on DGF and early function based upon mean serum creatinine up to 6 months after transplant(Table 1). Recipients of single kidneys from ECD with D-AdCCr <90 had a significantly higher incidence of DGF versus recipients of dual grafts (9 of 20 [45%] vs. 1 of 11 [9%]; P=0.04). Additionally, mean recipient serum creatinine at 1 and 4 weeks was significantly higher in recipients of single versus dual kidneys(5.3±3.3 and 2.8±2.0 vs. 1.7±0.6 and 1.4±0.5 mg/dl; P<0.05). By three months, these differences were no longer statistically significant; mean serum creatinines at 6 months and 1 year were also similar.
When donor age was ≥59 years, there was a twofold increase in DGF in the recipients of singles versus dual kidneys, although this did not reach statistical significance (5 of 11 [45%] vs. 3 of 14 [21%]; P=0.20). There was a statistically significant difference in mean recipient serum creatinine at 1 week, 1 month, and 3 months in the recipients of single versus dual kidneys, (5.1±3.3, 3.4±2.1, and 2.8±1.5 vs. 2.8±2.5, 1.5±0.6, and 1.6±0.5 mg/dl;P<0.05). Cld Stg also had a significant impact on outcome; the best outcome group was the recipients of dual transplants stored <24 hr, who had a significant improvement in DGF versus recipients of single kidneys stored >24 hr (1 of 10 [10%] vs. 7 of 11 [64%], respectively;P<0.05) and mean serum creatinine at 1 week, 1 month, and 3 months (1.9±0.8, 1.3±0.4, 1.5±0.2 vs. 6.6±3.4, 3.0±1.6, 2.9±1.9 mg/dl; P<0.05). We also compared the ratio of donor:recipient weight to identify any differences in outcome based upon size mismatch between the donors and recipients. In the ECD single kidneys, the ratio was 1.07±0.35 and in the ECD dual kidneys, it was 1.01±0.28. We achieved our goal of matching the donor to recipient weight in 6 of 15 (40%) of our dual recipients versus 21 of 37 (57%;P=NS) in our single recipients. Although we did not achieve a ratio of 1.0 in the remaining nine dual recipients, the ratio was at least 0.7 in all but one patient. Overall, 1-year actuarial patient and graft survival rates in the recipients of ECD single versus dual kidneys were 96% and 81% vs. 93% and 87% (P=NS), respectively. There tended to be a wide variation in the serum creatinine at the time points evaluated in the recipients of ECD single kidneys. The reasons for this are not clear, although there were not isolated “outliers.”
We next evaluated outcome in the recipients of single versus dual ECD kidneys when the donor was >59 years of age and had a D-AdCCr <90 ml/min. In the recipients of these ECD single kidneys (n=6), the incidence of DGF was 33% versus 10% in the recipients of dual kidneys (n=10;P=NS). The serum creatinine was significantly higher at 1 month in recipients of single versus dual kidneys (3.4±2.6 vs. 1.3±0.3 mg/dl; P=0.03). The differences at 3, 6, and 12 months(2.2±1.1, 1.6±0.4, and 2.3±0.4 vs. 2.9±2.1, 1.7±0.5, and 1.9±0.2 mg/dl for single versus dual kidneys, respectively) were not statistically significant. Finally, the incidence of graft loss was fivefold higher, although not statistically significant, in recipients of single versus dual kidneys (3 of 6 [50%] vs. 1 of 10 [10%];P=0.07). There was one death in each group.
Ideal donors as single kidney versus ECD as dual kidneys. We also compared donor variables and recipient outcome using control transplants, assuming they represent ideal donor kidneys, performed during the same time period versus ECD dual kidneys (Table 2). As expected, the donor age was significantly lower in the controls versus the dual kidneys(39±2.2 vs. 63±2.2 years; P<0.0001). There was also a significant difference in the D-AdCCr (100±4.7 vs. 79±6.7 mg/dl; P=0.048) and in the incidence of hypertension(21 of 74 [28%] vs. 10 of 15 [67%]; P=0.008). Additionally, due to our approach of placing dual ECD kidneys into older recipients, recipient age was also significantly lower (47±1.4 vs. 57±3.4 years;P=0.013). There were also significant differences in two recipient outcome variables. In recipients of control donor kidneys, mean serum creatinine at 1 year was significantly lower (1.6±0.5 vs. 2.1±0.4 mg/dl; P=0.037) and initial length of hospital stay for transplantation was shorter (6.9±0.5 vs. 9.6±1.6 days;P=0.033). The mean serum creatinine at 18 months was similar in the two groups (1.7±0.5 vs. 2.1±0.4; P=NS). The 1-year actuarial patient and graft survival in the ideal donors versus the ECD dual kidney recipients was 94% and 91% vs. 93% and 87% (P=NS).
Correlation coefficients and multiple regression analysis. In all three groups (ECD single kidney, ECD dual kidneys, and ideal donors), donor age showed a correlation to recipient serum creatinine up to 6 months after transplant. Regression analysis also demonstrated a relationship between donor age and recipient creatinine up to 6 months.
In the recipients of ECD single kidneys, there was a positive correlation between donor age and creatinine at 6 months (r=0.62). Donor Ad-CCr and Cld Stg time showed no correlation. Regression analysis demonstrated a relationship between donor age and serum creatinine at 1 month(R2=0.21, P=0.004) and 6 months (R2=0.38,P=0.005).
In the recipients of ECD dual kidneys there was a negative correlation between donor age and serum creatinine at 3 and 6 months (r=0.53, r=0.72, respectively) suggesting a linear relationship between increasing donor age and decreasing mean creatinine at 3 and 6 months. The importance of this is unclear, although it may suggest that using both kidneys overcomes the positive relationship between donor age and creatinine in single kidneys. There was no relationship between the D-AdCCr or Cld Stg time and mean recipient serum creatinine. As expected, the creatinine at 1 month correlated strongly with creatinine at 3 and 6 months (r=0.78, r=0.79, respectively). Regression analysis identified a relationship between donor age and serum creatinine at 6 months (R2=0.52, P=0.004).
In the recipients of control donor kidneys, there was no correlation between donor age, Ad-CCr, or Cld Stg time and recipient serum creatinine at 1, 3, or 6 months. There was a strong correlation between recipient serum creatinine at 1 month and 3 months (r=0.73). Regression analysis did show a relationship between donor age and serum creatinine at 1 month (R2=0.12, P=0.005), 3 months (R2=0.14,P=0.002), and 6 months (R2=0.2, P=0.001). As with the other groups, there was no correlation or relationship between donor Ad-CCr or Cld Stg time and recipient serum creatinine at 1, 3, or 6 months.
The use of two kidneys as a dual transplant from a single older donor has been a novel practice used to help decrease the organ shortage; these kidneys otherwise would have been discarded. In this report, we describe our results with our dual kidney transplant program and compare the outcome to that in recipients of single kidneys from ECDs with similar donor characteristics and to recipients of kidneys from ideal donors. We have not compared outcome in all of the recipients of single ECD kidneys as a whole versus recipients of the dual kidneys. We believe the most appropriate comparison is between donors with similar characteristics when trying to determine when these ECD kidneys should be used as single versus dual transplants. We believe using dual transplants selectively from ECDs can yield outcome similar to the recipients of ideal donor kidneys, and we have compared outcome between these two groups.
In this analysis of our ECD program over the last 2 years, we have identified three characteristics that impact early graft function and outcome. Donor AdCCr <90 ml/min and D-age ≥59 years were characteristics that predicted less DGF and better mean creatinine up to 6 months after transplant when ECD kidneys were transplanted as dual rather than single kidneys. Additionally, Cld Stg time >24 hr was associated with more DGF and higher mean serum creatinine up to 6 months after transplant. Other reports have also identified Cld Stg time >24 hr as a predictor of DGF(6).
It is important to recognize that the majority of recipients of ECD dual grafts received kidneys from an older ECD who usually had a history of hypertension. We attempt to place these ECD dual kidneys into smaller, aged adults; overall our donor to recipient weight ratio is 1.01, which demonstrates adequate size matching. Other reports have demonstrated the importance of size matching and long-term outcome (7, 8). Therefore, having the flexibility to place these ECD kidneys into appropriately age- and size-matched adults is important.
The major improvement in using the ECD kidneys as dual versus single grafts was a decrease in the incidence of DGF. Most reports from recent series uniformly describe improvement in short-term (1 year) and long-term outcome in patients without DGF versus patients with DGF(9-12). In general, DGF predicts worse outcome independent of acute rejection. A decrease in 1-year graft survival from 10-30% has been reported in association with DGF, even in the cyclosporine era (12). DGF also has an impact on outcome in recipients of aged kidneys (13). In general, DGF increases hospital costs because of the need for dialysis and additional studies and biopsies (12). Occasional reports do not demonstrate a decrease in graft survival in patients with DGF except in association with acute rejection (6, 14). A multicenter study demonstrated a clear correlation between early graft function (within the first month after transplant) and long-term graft survival (15). We believe DGF and early outcome can be improved when using ECD kidneys by carefully evaluating donor and recipient characteristics and placing kidneys into age and size-matched adults, using dual transplants when appropriate.
It is unclear what the absolute cutoff should be when considering D-AdCC4 in the decision algorithm for using single versus dual grafts. Many patients over the age of 60 could have a CCr >90 ml/min, e.g., a 63-year old man who weighs 85 kg and has an admission creatinine of 0.9 mg/dl would have a calculated CCr of 101 ml/min. Providing other characteristics of this donor were suitable, these kidneys could be used as single grafts. In contrast, a 60-year-old woman who weighs 60 kg and has an CCr of 1.0 mg/dl would have a calculated AdCCr of 57 ml/min. Other donor characteristics, e.g., the rate of rise and peak creatinine after a brief period of hypotension or dehydration, and the characteristics of the urine output, are important considerations. We would be reluctant to use either of these ECD for a young, otherwise healthy, active patient. We agree, in principle, with the concept that “the supply of viable donor nephrons and the physiologic demands of the transplant recipient are important determinants of long-term kidney allograft failure”(8). We have found that using dual kidneys from an ECD who has been turned down locally as single grafts will provide excellent renal function for age and size-matched adults. We are opposed, however, to offering ECD kidneys, which have been turned down by all local centers as single grafts, nationally as single grafts before being used locally as dual grafts. We believe that the unnecessary Cld Stg time will result in worse outcome in the recipients of these single kidneys. We also believe that donor characteristics and recipient outcome should be carefully evaluated in centers willing to accept ECD as single and dual grafts, to establish clear guidelines for their use.
In conclusion, excellent outcome can be achieved using ECD kidneys both as single and dual grafts. Donors with AdCCr <90 ml/min and age ≥59 are suitable for use as dual kidney transplants and can be expected to have a very low incidence of DGF and provide excellent early posttransplant serum creatinine when used as dual transplants placed into appropriately size- and age-matched adults. Careful consideration should be given before accepting these kidneys as single grafts, particularly when the donor has a long history of hypertension. The decision to use ECD kidneys as single versus dual grafts should include donor history, AdCCr, and age. Every effort should be made to keep the Cld Stg time <24 hr. When appropriately selected, kidneys from ECDs can be used as dual transplants with the expectation for excellent outcome, similar to recipients of single cadaveric kidney transplants.
Presented at the 16th annual meeting of the American Society of Transplant Physicians.
Abbreviations: AdCCr, admission creatinine clearance; ANOVA, analysis of variance; Cld Stg, cold storage; CRT, cadaveric renal transplant; D, donor; DGF, delayed graft function; ECD, expanded criteria donor; R, recipient.
1. UNOS 1996 Annual Report, and the UNOS Scientific Registry of Transplant Recipients. Richmond, VA: UNOS.
2. Lee CM, Scandling JD, Shen GK, Salvatierra O, Dafoe DC, Alfrey EJ. The kidneys that nobody wanted: support for the utilization of expanded criteria donors. Transplantation 1996; 62: 1832.
3. Kuo PC, Johnson LB, Schweitzer EJ, Alfrey EJ, Waskerwitz J,Bartlett ST. Utilization of the older donor for renal transplantation. Am J Surg 1996; 172: 551.
4. Johnson LB, Kuo PC, Dafoe DC, et al. The use of bilateral adult renal allografts: a method to optimize function from donor kidneys with suboptimal nephron mass. Transplantation 1996; 61: 1261.
5. Johnson LB, Kuo PC, Dafoe DC, et al. Double adult renal allografts: a technique for expansion of the cadaveric kidney donor pool. Surgery 1996; 120: 580.
6. Troppmann C, Gillingham KJ, Gruessner RW, et al. Delayed graft function in the absence of rejection has no long-term impact: a study of cadaver kidney recipients with good graft function at one year after transplantation. Transplantation 1996; 61: 1331.
7. Feldman HI, Fazio I, Roth D, et al. Recipient body size and cadaveric renal allograft survival. J Am Soc Nephrol 1996; 7: 151.
8. Chertow GM, Milford EL, Mackenzie HS, Brenner BM. Antigen-independent determinants of cadaveric kidney transplant failure. JAMA 1996; 276: 1732.
9. Nicholson ML, Wheatley TJ, Horsburg T, Edwards CM, Weitch PS, Bell PR. The relative influence of delayed graft function and acute rejection on renal transplant survival. Transplant Int 1996; 9: 415.
10. Feldman HI, Gayner R, Berlin JA, et al. Delayed function reduces renal allograft survival independent of acute rejection. Nephrol Dial Transplant 1996; 11: 1306.
11. Carmellini M, Stefano RD, Filipponi F, Rindi P, Rizzo G, Mosca F. Delayed graft function adversely affects one-year graft survival of cadaveric renal transplants. Transplant Proc 1996; 28: 359.
12. Shoskes DA, Halloran PF. Delayed graft function in renal transplantation: etiology, management and long-term significance. J Urol 1996; 155: 1831.
13. Wyner LM, Novick AC, Hodge EE, Flechner SM, Sankari BR, Streem SB. Long-term follow-up of kidneys transplanted from elderly cadaveric donors. World J Urol 1996; 14: 265.
14. Perez FM, Rodriquez-Carmona A, Bouza P, et al. Outcome of grafts with long-lasting delayed function after renal transplantation. Transplantation 1996; 62: 42.
15. Opelz G, Sasaki N, Terasaki PI. Prediction of long-term kidney transplant survival rates by monitoring early graft function and clinical grades. Transplantation 1978; 25: 212.