J Am Soc Nephrol 14: 208 213, 2003 Kidney Allograft and Patient Survival in Type I Diabetic Recipients of Cadaveric Kidney Alone Versus Simultaneous Pancreas/Kidney Transplants: A Multivariate Analysis of the UNOS Database SUPHAMAI BUNNAPRADIST,* YONG W. CHO, J. MICHAEL CECKA, ALAN WILKINSON, and GABRIEL M. DANOVITCH *Kidney-Pancreas Transplant Program, Cedars-Sinai Medical Center, Los Angeles, California; UCLA Immunogenetics Center, Los Angeles, California; Kidney-Pancreas Transplant Program, UCLA Center for Health Science Center, UCLA, Los Angles, California; and UCLA School of Medicine, Los Angeles, California. Abstract. Simultaneous pancreas-kidney transplant (SPK) is now a common treatment for insulin-dependent diabetic patients with end-stage renal disease. Renal graft survival rates after SPK have been less well studied. This study compared the kidney survival results for 3642 SPK and 2374 cadaveric renal transplants (CRT) in type I diabetic patients at 112 US transplant centers reported to UNOS during 1994 through 1997. The analysis included follow-up information through September 2000. The kidney graft survival rates were significantly lower among recipients of CRT compared with SPK recipients (P 0.001). Patients who received SPK were younger, less often sensitized, transplanted after shorter periods on dialysis, and less often black. The donors of SPK organs were younger, more often died from head trauma, were less often female, and more often black. SPK renal grafts were transplanted with a shorter cold ischemia time to more poorly HLA-matched recipients. After adjustment of these and other factors, whether a patient was recipient of CRT or SPK was not associated with increased risk of kidney graft failure or patient death. SPK recipients experienced half the rate of delayed kidney function (11% versus 23%) but nearly double the rate of rejections during the initial hospitalization (15% versus 9%) compared with CRT recipients. SPK was associated with better renal allograft survival compared with CRT, despite a higher rate of renal allograft rejection. This observation was explained by favorable donor and recipient factors in the SPK group. After controlling for these factors, SPK provided no protective or detrimental effect on short-term renal allograft or patient survival. bunnapradist@cshs.org For insulin-dependent diabetes mellitus (IDDM) patients with ESRD, the options for renal replacement therapy include dialysis, a kidney transplant (cadaveric or living donor), or a simultaneous pancreas-kidney transplant (SPK) (1,2). Since 1995, more than half of IDDM patients who receive a kidney transplant receive a pancreas at the same time (3). SPK transplantation before age 50 has been associated with a long-term improvement in patient survival compared with CRT (4). Graft survival rates for SPK transplants are generally superior to those for kidney transplants in diabetic patients (3,5) despite a higher incidence of treated rejection episodes (6). In this study, we performed a retrospective analysis to compare the renal graft outcome in IDDM patients with ESRD who underwent Received July 19, 2002. Accepted August 19, 2002. Correspondence to Suphamai Bunnapradist, Medical Director, Kidney-Pancreas Transplantation, Cedars-Sinai Medical Center, Assistant Professor of Medicine, UCLA, 8635 W. 3 rd St., Suite 590W, Los Angeles, CA 90048. Phone: 310-423-2641; Fax: 310-423-0234; E-mail: bunnapradist@cshs.org 1046-6673/1401-0208 Journal of the American Society of Nephrology Copyright 2002 by the American Society of Nephrology DOI: 10.1097/01.ASN.0000037678.54984.41 SPK or cadaveric kidney transplantation alone (CRT). We used the United Network for Organ Sharing (UNOS) renal registry database. We also performed a multivariate analysis to assess factors responsible for any differences in outcome. Materials and Methods Graft and patient survival rates were calculated for 3642 type I diabetic patients who received SPK transplants and 2374 type I diabetic patients who received CRT at 112 US transplant centers between January 1994 and December 1997 based on data reported to UNOS. The analyses included follow-up information received through September 2000. Renal allograft survival rates were estimated by the product-limit method. For graft survival analyses, patient deaths were counted as graft losses regardless of the functional status of the graft at the time of death. For patient survival, patients were followed until death or graft loss. To compare continuous variables, the Wilcoxon rank-sum test was used for comparing two groups. The 2 test was used to compare categorical variables. The log-rank test was used to compare entire survival curves. Ten variables that influenced either graft failure or patient death in the univariate analyses were included in a multivariate Cox regression analysis (Table 4). Missing values were imputed with modal values for categorical variables and mean values for continuous variables. In addition, continuous variables such as age, cold ischemia time, duration of dialysis, and percent peak value of panel reactive antibodies were cate-
J Am Soc Nephrol 14: 208 213, 2003 Cadaveric Kidney Alone Versus SPK Transplants 209 Table 1. Characteristics of recipients, donors, and renal grafts CRT SPK n (mean SD) n (mean SD) P Recipient Factors age (yr) 2374 44.9 10.3 3642 38.1 7.2 0.001 height (cm) 2193 170.4 11.1 3459 170.2 11.1 0.45 weight (kg) 2352 76.6 18.2 3625 70.3 14.3 0.001 peak PRA (%) 2372 12.6 24.0 3641 7.6 16.8 0.001 duration of dialysis (mo) 2153 29.8 24.5 2806 19.2 20.3 0.001 female (%) 941 40 1542 42 0.04 regraft (%) 241 10 142 4 0.001 Black (%) 434 18 336 9 0.001 Primary source of payment (%) Medicare 1401 65 1543 45 private insurance 640 29 1696 49 other sources 130 6 220 6 no assistance required (%) 1709 78 2931 84 0.001 Donor factors age (yr) 2342 33.5 17.5 3610 26.9 12.1 0.001 height (cm) 2125 163.8 29.3 3447 166.0 28.3 0.001 weight (kg) 2146 71.3 22.8 3455 70.4 17.1 0.03 female (%) 934 39 1247 34 0.001 Black (%) 193 8 414 11 0.001 head trauma death (%) 1157 51 2364 67 0.001 Transplant factors cold ischemia time (h) 2243 21.6 8.7 3466 14.1 6.0 0.001 No. of HLA-A, B, DR mismatches 2324 3.1 1.7 3536 4.2 1.3 0.001 gorized, because their effects on the hazard function were nonlinear. All reported P values were two-tailed. Results The characteristics of the recipients and the donors are shown in Table 1. When compared with patients who received CRT, patients who received SPK transplants were significantly younger, had lower levels of preformed anti-hla antibodies, were less likely to have been previously transplanted, and were more likely to have private insurance. The donors of SPK organs were significantly younger, less likely to be female, more likely to be African American, and more likely to have died from trauma death than donors for recipients of a kidney alone. SPK grafts were transplanted with a significantly shorter cold ischemia time and had more HLA-A, -B, -DR antigens mismatched than kidneys without a pancreas. The early outcome of transplantation in the two recipient groups is shown in Table 2. Anuria occurred on the first day in 4% of SPK grafts compared with 11% of grafts of the CRT group (P 0.001). Eleven percent of the SPK recipients required dialysis during the first week after transplantation compared with 23% of the recipients who received CRT. The poorest kidney function among CRT transplant recipients was reflected by significantly higher serum creatinine concentrations at the time of discharge compared with that of SPK transplant recipients (P 0.001). However, SPK transplant recipients received more anti-rejection Table 2. Early function of kidney grafts of simultaneous kidney/pancreas transplants (SPK) and kidney alone transplants (KA) Variable CRT SPK P Anuria, % 11 4 0.001 Delayed graft function, % 23 11 0.001 Antirejection treatment, % 9 15 0.001 Hospital stay, d (mean SD) 13 13 18 15 0.001 Serum creatinine at discharge, mg/dl (mean SD) 2.8 2.3 1.8 1.4 0.001
210 Journal of the American Society of Nephrology J Am Soc Nephrol 14: 208 213, 2003 treatment (P 0.001) and stayed longer in hospital (P 0.001) than CRT recipients. The graft and patient survival rates of CRT kidneys were significantly lower than those of the SPK kidneys (P 0.001; Figure 1). To determine whether this difference was due to differences in donor/recipient selection or was due to simultaneous transplantation of the pancreas, we performed a multivariate analysis. Table 3 shows the list of potential confounding variables in the study of graft survival analysis, which were utilized in the multivariate Cox regression analysis. Multivariate regression analysis of ten potential risk factors showed that, after adjustment of the remaining factors, whether a patient was recipient of a CRT or an SPK was not significantly associated with the risk of kidney graft failure or patient death (Table 4). This suggests that SPK transplants did not have any significant advantage on kidney graft or patient survival, compared with CRT transplants. To confirm the absence of an advantage of SPK over CRT transplants, kidney graft survival curves for low-risk donor/patient pairs (recipient age 40 yr; donor age 35 yr) are shown in Figure 2. There was no significant advantage for low-risk SPK compared with low-risk CRT (P 0.12) for graft or patient survival. Discussion Our study demonstrated superior renal allograft outcome in SPK recipients compared with diabetic CRT recipients. However, after adjusting for donor and recipient demographic differences in a multivariate analysis, we found that only the age of the recipient and donor remained as significant factors that affected graft and patient survival rates (7). The recipient race, prior failed transplants, the time spent waiting on dialysis, the cold ischemia time, the retransplant rate, and HLA mismatch, Table 3. Univariate analyses for graft and patient survival: significant risk factors and their P values according to the log rank tests Graft Loss P value Patient Death P value Recipient age (yr) 41 to 50 versus 40 0.03 0.001 51 to 60 versus 40 0.001 0.001 60 versus 40 0.001 0.001 Source of primary pay Medicare versus private insurance 0.02 0.04 others versus private insurance 0.02 0.19 Recipient race African American versus others 0.02 0.71 Previous transplant regraft versus primary 0.05 0.57 Peak PRA (%) 20 versus 0 to 20 0.004 0.05 Duration of dialysis (mo) 36 versus 0 to 36 0.003 0.002 Donor age (yr) 16 versus 16 to 35 0.58 0.64 36 to 50 versus 16 to 35 0.001 0.09 50 versus 16 to 35 0.001 0.001 Cause of donor death head trauma versus others 0.001 0.001 Cold ischemia time (h) 24 versus 0to24 0.001 0.004 Figure 1. Comparison of graft and patient survival of cadaveric renal transplants (CRT) and simultaneous pancreas/kidney (SPK) transplant.
J Am Soc Nephrol 14: 208 213, 2003 Cadaveric Kidney Alone Versus SPK Transplants 211 Table 4. Risk factors for graft failure and patient death according to the Cox regression analyses n Graft Loss Patient Death RR (95% CI) P RR (95% CI) P Recipient age (yr) 40 3230 1.0 1.0 41 to 50 1924 1.13 (0.99 to 1.28) 0.07 1.38 (1.16 to 1.63) 0.001 51 to 60 667 1.22 (1.01 to 1.46) 0.04 1.75 (1.39 to 2.20) 0.001 60 195 1.68 (1.28 to 2.20) 0.001 2.82 (2.06 to 3.86) 0.001 Source of primary pay private insurance 2336 1.0 1.0 Medicare 2944 1.09 (0.96 to 1.24) 0.20 1.11 (0.94 to 1.32) 0.22 others 736 1.11 (0.94 to 1.32) 0.23 1.04 (0.82 to 1.31) 0.76 Recipient race Black 936 1.07 (0.91 to 1.26) 0.52 0.78 (0.62 to 0.98) 0.04 others 5080 1.0 1.0 Previous transplant primary 5634 1.0 1.0 regraft 382 1.17 (0.94 to 1.47) 0.16 1.01 (0.74 to 1.37) 0.93 Peak PRA (%) 0 to 20 5262 1.0 1.0 21 to 100 754 1.15 (0.98 to 1.36) 0.08 1.15 (0.93 to 1.42) 0.20 Duration of dialysis (mo) 0 to 36 5113 1.0 1.0 36 903 1.10 (0.94 to 1.29) 0.24 1.19 (0.97 to 1.46) 0.10 Donor age (yr) 0 to 15 854 1.04 (0.87 to 1.24) 0.68 1.02 (0.81 to 1.29) 0.84 16 to 35 3105 1.0 1.0 36 to 50 1406 1.28 (1.10 to 1.49) 0.001 1.09 (0.90 to 1.34) 0.38 50 593 1.71 (1.42 to 2.07) 0.001 1.59 (1.25 to 2.03) 0.001 Cause of donor death head trauma 3509 1.0 1.0 others 4117 1.04 (0.91 to 1.18) 0.57 1.07 (0.91 to 1.27) 0.41 Cold Ischemia Time (h) 0 to 24 5090 1.0 1.0 24 926 1.13 (0.97 to 1.31) 0.13 1.03 (0.84 to 1.25) 0.80 Transplant SPK 3642 1.0 1.0 kidney only 2374 0.98 (0.85 to 1.12) 0.73 1.06 (0.88 to 1.28) 0.53 all of which are known to affect the outcome of cadaver kidney transplants in general (8-11) had no significant effect on graft and patient survival rates among diabetic patients included in this study. Although the incidence of rejection treatment before discharge was higher among SPK (15% versus 9%) than CRT recipients, the difference may be due in part to the longer hospitalization noted for SPK recipients. Despite the higher reported incidence of early acute rejection, graft survival was higher among SPK recipients. Even among low-risk patients (under age 41 yr) who were transplanted from donors under age 36 yr, there was no significant difference in graft of patient and graft survival associated with the type of transplant. Thus, young diabetic recipients of optimal donor kidneys generally have good posttransplant graft survival regardless of whether or not they also receive a pancreas transplant. Our results should be interpreted with some caution. We considered only patients transplanted after 1994 to reflect outcomes using current practices (12,13). Nevertheless, fewer than half of US kidney transplant centers perform SPK for type 1 diabetic patients with ESRD. To reduce the influence of center effects, we evaluated data only from centers that offer both procedures. Among the centers providing both options, 60% of type I diabetic patients received SPK versus 40% for CRT. This observation suggests that there is likely to be some referral bias because some patients are obviously referred just for consideration of SPK. Most transplant centers have more strict criteria for SPK compared with CRT, including comorbid
212 Journal of the American Society of Nephrology J Am Soc Nephrol 14: 208 213, 2003 Figure 2. Comparison of graft and patient survival of CRT and SPK transplant for low-risk group. conditions such as cardiovascular and other vascular diseases (14). Different referral patterns and selection bias may contribute to the poorest outcome in the CRT group. We did not compare live donor kidney transplant outcome, a viable treatment for diabetic patients with ESRD. Our study did not examine the effects of different immunosuppressive protocols on graft survival (15,16). We controlled for center effects; therefore, we felt that variations in transplant management were unlikely to be relevant to our analysis. Finally, other factors that were not considered in this study are also important in assessing solid organ transplant outcome, patient survival, cost, quality of life, and morbidity from cardiovascular and peripheral vascular diseases (4,17,18). In summary, SPK transplantation was associated with better short-term renal allograft and patient survival compared with CRT, despite an apparently higher rate of renal allograft rejection. The superior survival results were due to more favorable donor and recipient factors, in particular the younger age of patients and donors in the SPK group. After controlling for the effects of these confounding factors, SPK provided no protective or detrimental effect on renal allograft outcome. In a low-risk group, comprising younger recipients of younger donor allografts, we observed no differences in kidney graft survival between SPK and CRT. Thus, donor and recipient factors, rather than pancreas transplant itself appear to account for the reported improved kidney survival in SPK recipients. Acknowledgments This study was presented as an abstract and oral presentation at the 2000 American Society of Transplantation meeting in Chicago, Illinois. References 1. Consensus conference on standardized listing criteria for renal transplant candidates: Transplantation 66: 962 967, 1998 2. Pancreas transplantation for patients with type 1 diabetes: American Diabetes Association: Diabetes Care 23: 117, 2000 3. Cecka JM: The UNOS Scientific Renal Transplant Registry: Clin Transplants 1 21, 1999 4. Ojo AO, Meier-Kriesche HU, Hanson JA, Tyden G, Bolinder J, Solders G, Brattstrom C, Tibell A, Groth CG: The impact of simultaneous pancreas-kidney transplantation on long-term patient survival. Transplantation 71: 82 90, 2001 5. Tyden G, Bolinder J, Solders G, Brattstrom C, Tibell A, Groth CG: Improved survival in patients with insulin-dependent diabetes mellitus and end-stage diabetic nephropathy 10 years after combined pancreas and kidney transplantation. Transplantation 67: 645 648, 1999 6. Douzdjian V, Abecassis MM, Corry RJ, Hunsicker LG: Simultaneous pancreas-kidney versus kidney-alone transplants in diabetics: increased risk of early cardiac death and acute rejection following pancreas transplants. Clin Transplant 8: 246 251, 1994 7. Terasaki PI, Gjertson DW, Cecka JM, Takemoto S, Cho YW: Significance of the donor age effect on kidney transplants. Clin Transplant 11: 366 372, 1997 8. Takemoto SK, Cho YW, Gjertson DW: Transplant risks. Clin Transplant 325 334, 1999 9. Terasaki PI, Cecka JM, Gjertson DW: Impact analysis: A method for evaluating the impact of factors in clinical renal transplantation. Clin Transplant 437 441, 1998 10. Meier-Kriesche HU, Port FK, Ojo AO, Rudich SM, Hanson JA, Cibrik DM, Leichtman AB, Kaplan B: Effect of waiting time on renal transplant outcome. Kidney Int 58: 1311 1317, 2000 11. Wolfe RA, Ashby VB, Milford EL, Ojo AO, Ettenger RE, Agodoa LY, Held PJ, Port FK: Comparison of mortality in all
J Am Soc Nephrol 14: 208 213, 2003 Cadaveric Kidney Alone Versus SPK Transplants 213 patients on dialysis, patients on dialysis awaiting transplantation, and recipients of a first cadaveric transplant [see comments]. N Engl J Med 341: 1725 1730, 1999 12. Sutherland DE, Gruessner RW, Dunn DL, Matas AJ, Humar A, Kandaswamy R, Mauer SM, Kennedy WR, Goetz FC, Robertson RP, Gruessner AC, Najarian JS: Lessons learned from more than 1,000 pancreas transplants at a single institution. Ann Surg 233: 463 501, 2001 13. Hariharan S, Johnson CP, Bresnahan BA, Taranto SE, McIntosh MJ, Stablein D: Improved graft survival after renal transplantation in the United States, 1988 to 1996 [see comments]. N Eng J Med 342: 605 612, 2000 14. Steinman TI, Becker BN, Frost AE, Olthoff KM, Smart FW, Suki WN, Wilkinson AH: The Clinical Practice Committee, American Society of Transplantation. Guidelines for the referral and management of patients eligible for solid organ transplantation. Transplantation 71: 1189 1204, 2001 15. Takemoto SK: Maintenance immunosuppression. Clin Transplant 481 495, 2000 16. Gruessner AC, Sutherland DE: Report for the international pancreas transplant registry-2000. Transplantat Proc 33: 1643 1646, 2001 17. Douzdjian V, Escobar F, Kupin WL, Venkat KK, Abouljoud MS: Cost-utility analysis of living-donor kidney transplantation followed by pancreas transplantation versus simultaneous pancreas-kidney transplantation. Clin Transplant 13: 51 58, 1999 18. Gross CR, Limwattananon C, Matthees B, Zehrer JL, Savik K: Impact of transplantation on quality of life in patients with diabetes and renal dysfunction. Transplantation 70: 1736 1746, 2000 Access to UpToDate on-line is available for additional clinical information at http://www.jasn.org/