Article. Simultaneous Pancreas Kidney Transplant versus Other Kidney Transplant Options in Patients with Type 2 Diabetes

Article Simultaneous Pancreas Kidney Transplant versus Other Kidney Transplant Options in Patients with Type 2 Diabetes Alexander C. Wiseman* and Jane Gralla Summary Background and objectives Current organ allocation policy prioritizes placement of kidneys (with pancreas) to patients listed for simultaneous pancreas kidney transplantation (SPK). Patients with type 2 diabetes mellitus (T2DM) may undergo SPK, but it is unknown whether these patients enjoy a survival advantage with SPK versus deceased-donor kidney transplantation alone (DDKA) or living-donor kidney transplantation alone (LDKA). Design, setting, participants, & measurements Using the Scientific Registry of Transplant Recipients database, patients with T2DM, age 18 59 years, body mass index 18 30 kg/m 2, who underwent SPK, DDKA, or LDKA from 2000 through 2008 were identified. Five-year patient and kidney graft survival rates were compared, and multivariable analysis was performed to determine donor, recipient, and transplant factors influencing these outcomes. Results Of 6416 patients identified, 4005, 1987, and 424 underwent DDKA, LDKA, and SPK, respectively. On unadjusted analysis, patient and kidney graft survival rates were superior for LDKA versus SPK, whereas patient but not graft survival was higher for SPK versus DDKA. On multivariable analysis, survival advantage for SPK versus DDKA was related not to pancreas transplantation but younger donor and recipient ages in the SPK cohort. *Division of Renal Diseases and Hypertension, Transplant Center, and Department of Pediatrics, University of Colorado Denver, Aurora, Colorado Correspondence: Dr. Alexander C. Wiseman, Transplant Center, University of Colorado Denver, Mail Stop F749, AOP 7089, 1635 North Aurora Court, Aurora, CO 80045. Email: Alexander.wiseman@ ucdenver.edu Conclusions Good outcomes can occur with SPK in selected patients with T2DM, but no patient or graft survival advantage is provided by added pancreas transplantation compared with DDKA; outcomes were superior with LDKA. These results support cautious use of SPK in T2DM when LDKA is not an option, with close oversight of the effect of kidney (with pancreas) allocation priority over other transplant candidates. Clin J Am Soc Nephrol 7: 656 664, 2012. doi: 10.2215/CJN.08310811 Introduction Outcomes of simultaneous pancreas kidney transplantation (SPK) have markedly improved over the past decade because of advances in immunosuppression and surgical techniques. With 1-year patient and pancreas graft survival rates of 96% and 85%, respectively (1) and evidence of long-term advantages of functional pancreas transplantation upon survival compared with living-donor and deceased-donor kidney transplantation alone (2, 3), an argument can be made that SPK is the transplantation option of choice for patients with type 1 diabetes mellitus (T1DM). In response to these improved outcomes and the comparatively poor outcomes of patients who remain on the waiting list for SPK, the United Network for Organ Sharing (UNOS) amended its allocation policy to expedite SPK transplantation. Patients who are waitlisted for SPK are now given allocation priority over other patients awaiting a kidney transplant when a deceased kidney pancreas organ donor is identified. Median waiting times for SPK are expected to decrease and to further separate from waiting times 656 Copyright 2012 by the American Society of Nephrology for deceased-donor kidneys. Before these policy changes, the most recent calculated median waiting times were 406 days for SPK and 1269 days for any kidney transplantation alone (1). The preceding outcomes primarily concerned patients with T1DM, but the initial UNOS policy amendment did not specifically segregate this policy by diabetes status. SPK is occasionally performed in patients with type 2 diabetes, and it is unknown whether these survival advantages apply to this subset of SPK recipients. A recent article published in this journal described similar pancreas transplant outcomes in T2DM and T1DM recipients (4), and an accompanying editorial encouraged the use of SPK in T2DM recipients (5). If such use were to expand, patients with T2DM could effectively shorten their waiting time by undergoing SPK rather than deceaseddonor kidney transplantation alone (DDKA), which may create a disadvantage for other DDKA wait-listed patients and T1DM SPK wait-listed patients. In response to this potential overrepresentation of T2DM candidates awaiting SPK, UNOS recently www.cjasn.org Vol 7 April, 2012

Clin J Am Soc Nephrol 7: 656 664, April, 2012 SPK or KTA in T2DM, Wiseman and Gralla 657 approved another amendment to pancreas transplant candidacy, defining an eligible T2DM patient as one who is receiving insulin and has a C-peptide level #2 ng/ml,or one who is receiving insulin and has a C-peptide level $2 ng/ml and a body mass index (BMI),28 kg/m 2 (this BMI cutoff may be adjusted according to the percentage of candidates in this category every 6 months, not to exceed 30 kg/m 2 ). These patients will be given allocation priority over other DDKA candidates when a kidney pancreas donor is identified. Given the recent changes in outcomes and allocation, an important yet unanswered question is whether pancreas transplantation actually provides a survival advantage over DDKA in the T2DM recipient, as has been shown repeatedly for the T1DM recipient. We tested the hypothesis that selected patients with T2DM who undergo SPK have superior graft and patient survival compared with those receiving a kidney transplant alone, a finding that would support organ allocation that favors SPK for patients with T2DM. Materials and Methods This study was reviewed and approved by the Colorado Institutional Review Board. The study population included all adult kidney (DDKA, living-donor kidney transplantation alone [LDKA], and SPK) recipients with a diagnosis of T2DM undergoing primary transplantation from 2000 through 2008 who were registered in the national Scientific Registry of Transplant Recipients (SRTR). The SRTR data system includes data on all donors, wait-listed candidates, and transplant recipients in the United States, submitted by the members of the Organ Procurement and Transplantation Network. The Health Resources and Services Administration, U.S. Department of Health and Human Services, provides oversight to the activities of the Organ Procurement and Transplantation Network and SRTR contractors. For this study, exclusion criteria were recipient age,18 or.59 years, prior transplantation, BMI,18 kg/m 2 or.30 kg/m 2, or any diagnosis of renal disease other than T2DM. These restrictions were used to more closely define potential candidates for SPK, as reflected in recent changes to UNOS pancreas allocation policy. The primary outcome was patient survival for DDKA versus SPK and for LDKA versus SPK at 5 years after transplantation. The secondary outcomes were kidney graft survival at 1 and 5 years after transplantation for DDKA versus SPK and LDKA versus SPK. For graft survival, both unadjusted graft survival and death-censored patient death were calculated. With the intent to define any potential biases from donor and recipient characteristics that may support the use of SPK versus DDKA, a multivariable Cox proportional hazards regression model was used to take into account prognostic factors considered to contribute to the risk for kidney graft loss and patient death: donor and recipient age and race, duration of pretransplant dialysis, time on the waiting list, recipient BMI, induction and maintenance immunosuppression, peak panel-reactive antibody 0% versus.0%, HLA mismatches, delayed graft function (need for dialysis within the first 7 days after transplantation), length of cold ischemia time, whether there was an extended-criteria donor, and donor cause of death (cerebrovascular accident, head trauma, other cause). A separate multivariable analysis was performed for T2DM recipients undergoing LDKA versus SPK, using the same statistical analysis as the preceding comparison. For the multivariable Cox proportional hazards regression model, prognostic factors accounted for were identical to those in the DDKA-versus-SPK analysis, with the following exceptions: (1) waiting time, because 29% of living donor recipients were never placed on the UNOS waiting list, and (2) donor characteristics, because the differences between living donors and deceased donors are so qualitatively different that comparisons between such variables as age and cold ischemia time are uninterpretable. Graft survival, death-censored graft survival, and patient survival were calculated using Kaplan-Meier estimates and compared using a log-rank test. Hazard ratios were obtained from Cox proportional hazards regression model, with DDKA versus SPK and LDKA versus SPK as covariates in the model. Donor and recipient characteristics were compared between SPK versus DDKA and LDKA groups at time of transplantation using t tests for continuous variables and chi-squared tests for categorical outcomes. All analyses were conducted using SAS software, version 9.2 (SAS Institute, Cary, NC). Results We identified 6416 patients classified as having T2DM, age 18 59 years, BMI 18 30 kg/m 2 without previous transplant who underwent DDKA, LDKA, or SPK during 2000 2008. Of these, 4005 had DDKA, 1987 had LDKA, and 424 had SPK. Baseline characteristics of these cohorts are presented in Table 1. Notably, the SPK cohort was younger, had a shorter waiting time to transplantation, was more likely to receive an organ from a younger donor, and was less likely to be African American. Surprisingly, 40% of the SPK recipients were age 50 59 years, and a significant percentage of these were older than age 55 years. Cold ischemia time was longer in the DDKA cohort, as would be expected given the general practice of minimization of cold ischemia time for SPK. Figure 1 depicts kidney and patient survival comparing DDKA to SPK outcomes to 60 months. One-year patient and kidney graft survival rates were similar for patients with T2DM undergoing SPK or DDKA. After 1 year, patient (Figure 1A) and graft (Figure 1B) survival began to favor SPK (Kaplan-Meier unadjusted survival rate at 5 years for SPK versus DDKA, 82.0% versus 75.5%; P=0.04). The difference in graft survival was primarily explained by differences in survival because death-censored kidney graft survival (Figure 1C) did not differ between groups at 1 or 5 years. A key question is whether the survival differences depicted in Figure 1A were attributable to the added pancreas transplant or, rather, to another factor, given the differences in baseline characteristics. Table 2 shows results of a multivariable analysis of death-censored kidney graft survival and of patient survival, examining all variables identified as potential covariates from Table 1. Of note, SPK was not significantly associated with survival (hazard ratio for death among DDKA versus SPK recipients, 0.90; 95%

658 Clinical Journal of the American Society of Nephrology Table 1. Baseline characteristics Characteristic DDKA (n=4005) SPK (n=424) LDKA (n=1987) SPK versus DDKA SPK versus LDKA Male recipient 2725 (68.0) 292 (68.9) 1354 (68.1) 0.73 0.77 Recipient age,0.001,0.001 18 34 yr 65 (1.6) 26 (6.1) 58 (2.9) 35 49 yr 1044 (26.1) 229 (54.0) 604 (30.4) 50 59 yr 2896 (72.3) 169 (39.9) 1325 (66.7) Mean recipient 25.862.7 24.762.8 25.962.7,0.001,0.001 BMI 6 SD (kg/m 2 ) Recipient race,0.001,0.001 white 1330 (33.2) 278 (65.6) 993 (50.0) African American 1325 (33.1) 73 (17.2) 346 (17.4) Hispanic 919 (22.9) 50 (11.8) 458 (23.1) Asian 288 (7.2) 16 (3.8) 128 (6.4) other 143 (3.6) 7 (1.6) 62 (3.1) Pretransplant dialysis,0.001,0.001 none 86 (2.2) 30 (7.8) 161 (9.1) 0,6 mo 93 (2.4) 22 (5.7) 245 (13.8) 6,12 mo 237 (6.2) 46 (12.0) 364 (20.6) 12,24 mo 716 (18.6) 119 (31.0) 565 (31.9) $24 mo 2710 (70.6) 167 (43.5) 436 (24.6) Wait time,0.001 NA 0,6 mo 839 (21.0) 179 (42.2) NA a 6,12 mo 569 (14.2) 90 (21.2) 12,24 mo 996 (24.9) 106 (25.0) $24 mo 1601 (40.0) 49 (11.6) Induction therapy,0.001,0.001 antithymocyte globulin 1397 (34.9) 217 (51.2) 490 (24.7) interleukin-2 receptor 1220 (30.5) 57 (13.4) 648 (32.6) antagonist other 339 (8.4) 29 (6.9) 189 (9.5) none 1049 (26.2) 121 (28.5) 660 (33.2) Maintenance therapy,0.001,0.001 immunosuppression tacrolimus/ 2478 (61.9) 325 (76.7) 1103 (55.5) mycophenolate other 1527 (38.1) 99 (23.3) 884 (44.5) Peak panel-reactive 1957 (49.0) 171 (40.5) 716 (36.6),0.001 0.13 antibody. 0% No HLA mismatches 553 (13.8) 8 (1.9) 149 (7.6),0.001,0.001 Delayed graft function 1140 (28.6) 43 (10.1) 107 (5.4),0.001,0.001 Male donor 2377 (59.3) 263 (62.0) 780 (39.3) 0.29 NA Donor age,0.001 NA 0 17 yr 505 (12.6) 70 (16.5) 1 (0.1) 18 34 yr 1056 (26.4) 236 (55.7) 738 (37.1) 35 49 yr 1241 (31.0) 106 (25.0) 792 (39.9) 50 59 yr 831 (20.8) 11 (2.6) 410 (20.6) $60 yr 372 (9.3) 1 (0.2) 45 (2.3) Donor race 0.60 NA white 2764 (69.0) 296 (69.8) 1086 (54.7) African American 481 (12.0) 59 (13.9) 312 (15.7) Hispanic 606 (15.1) 54 (12.7) 433 (21.8) Asian 110 (2.8) 11 (2.6) 96 (4.8) other 44 (1.1) 4 (1.0) 60 (3.0) Donor cause of death,0.001 NA cerebrovascular 1651 (41.2) 99 (23.4) NA accident head trauma 1669 (41.7) 285 (67.2) other 685 (17.1) 40 (9.4) Expanded-criteria donor 750 (18.7) 3 (0.7) NA,0.001 NA

Clin J Am Soc Nephrol 7: 656 664, April, 2012 SPK or KTA in T2DM, Wiseman and Gralla 659 Table 1. (Continued) Characteristic DDKA (n=4005) SPK (n=424) LDKA (n=1987) SPK versus DDKA SPK versus LDKA Cold ischemia time,0.001 NA 0,12 h 731 (20.3) 178 (52.4) NA 12,24 h 1930 (53.7) 147 (43.2) $24 h 936 (26.0) 15 (4.4) Patients had type 2 diabetes, were age 18 59 years, had a body mass index # 30 kg/m 2, and were undergoing deceased-donor kidney transplantation alone, simultaneous pancreas kidney transplantation, or living-donor kidney transplantation alone from 2000 through 2008. Unless otherwise noted, values are expressed as the number (percentage) of patients. DDKA, deceased-donor kidney transplantation alone; SPK, simultaneous pancreas kidney transplantation; LDKA, living-donor kidney transplantation alone; BMI, body mass index; NA, not applicable. a Initial waiting list date was not provided for 29% of the LDKA cohort. confidence interval, 0.66 1.23). Numerous factors other than the added pancreas transplant were associated with higher risk for death, in particular older donor age and recipient age, use of maintenance immunosuppression other than tacrolimus and mycophenolate, HLA mismatches, use of expanded-criteria donor, and delayed graft function. Figure 2 depicts kidney and patient survival comparing LDKAtoSPKoutcomesto60months.Therewasearly separation for patient, graft, and death-censored graft survival curves favoring LDKA, which increased until 5 years (Kaplan-Meier unadjusted survival rate at 5 years for SPK versus LDKA, 82.0% versus 87.3%; P=0.003). When accounting for differences in baseline characteristics noted in Table 1, multivariable analysis of death-censored kidney graft survival and of patient survival indicate a significant survival advantage for T2DM patients undergoing LDKA compared with SPK (hazard ratio for death-censored graft survival among LDKA versus SPK recipients, 0.66; 95% confidence interval, 0.44 0.98; hazard ratio for death among LDKA versus SPK recipients, 0.50; 95% confidence interval, 0.35 0.71) (Table 3). Finally, pancreas allograft function as reported to the SRTR was assessed. Pancreas allograft survival for the T2DM cohort was similar to that reported earlier for patients with T1DM during the same era (4). Unadjusted pancreas allograft survival rates were 83.7% and 71.0% at 1 and 5 years, respectively, whereas death-censored pancreas graft survival rates were 87.7% at 1 year and 83.6% at 5 years. These results reiterate the findings of prior studies that describe excellent outcomes with SPK, irrespective of the recipient s type of diabetes. Discussion To our knowledge, this is the first study to compare outcomes of SPK, DDKA, and LDKA in patients with T2DM who may be considered eligible for SPK under current policies. Using a selected cohort of T2DM recipients who have a BMI of 18 30 kg/m 2 and are younger than 60 years of age (consistent with current practice and allocation policy), we report several important findings: (1) approximately 10% of DDKA recipients with T2DM who fit the preceding criteria underwent SPK from 2000 through 2008; (2) kidney, pancreas, and patient survival were excellent in the SPK cohort, with higher patient survival on unadjusted analysis than that in patients who had DDKA; (3) the survival advantage in the SPK cohort compared with the DDKA group at 5 years was not associated with the additional pancreas transplant but, rather, with other factors specific to this cohort (including the receipt of a transplant from a younger organ donor, younger recipient age at transplantation, and less waiting time before transplantation); and finally, (4) we confirm that LDKA is associated with survival advantages to 5 years compared with SPK and provides the greatest opportunity for optimal patient and graft outcomes for this population. Although these findings support the judicious use of SPK in selected T2DM recipients who do not have a living donor available, caution must be used before a generalized expansion of this treatment option can be considered for T2DM recipients. Aside from the differences noted in Table 1, it is highly likely that the approximately 10% of patients with T2DM who underwent deceased-donor SPK have different patient characteristics than the remaining DDKA cohort and are difficult to capture by age and BMI constraints alone. Such factors as insulin sensitivity and resistance, underlying C-peptide production, other comorbid conditions (e.g., cardiovascular disease), or transplant center effect may contribute to the selection and outcomes of the SPK cohort (6 9). These measures are poorly captured in registry data, and, even if they were available, the interpretation of these data can be even more problematic (e.g., the insulin sensitivity or resistance of the diabetic patient whose condition is controlled by diet and who has a hemoglobin A1c value of 7.0 should not be compared with that in the patient who is receiving 10 units of insulin daily and has a hemoglobin A1c value of 6.0). As described in the introduction, the recent change in pancreas allocation policy is a step toward improving the definition of patients who may be considered eligible for SPK by applying C-peptide and BMI criteria. Medication use and hemoglobin A1c data both before and after transplantation would further improve understanding of pancreas transplantation outcomes in both T2DM and T1DM recipients, even if interpatient comparisons may be challenging.

660 Clinical Journal of the American Society of Nephrology Figure 1. Kaplan-Meier estimated kidney graft and patient survival among patients with type 2 diabetes mellitus (T2DM), age 18 59 years, body mass index, 30 kg/m 2 undergoing deceased-donor kidney transplantation alone (DDKA) or simultaneous pancreas kidney transplantation (SPK) from 2000 through 2008. (A) Patient survival, SPK versus DDKA: 82.0% versus 75.5% (log-rank P=0.04). (B) Kidney graft survival, SPK versus DDKA: 75.2% versus 65.1% (log-rank P=0.004). (C) Death-censored kidney graft survival, SPK versus DDKA: 86.2% versus 82.6% (log-rank P=0.21).

Clin J Am Soc Nephrol 7: 656 664, April, 2012 SPK or KTA in T2DM, Wiseman and Gralla 661 Table 2. Risk factors associated with death and death-censored kidney graft loss Variable Patient Survival Hazard Ratio (95% CI) Death-Censored Kidney Graft Survival Hazard Ratio (95% CI) DDKA versus SPK 0.90 (0.66 1.23) 0.50 0.81 (0.57 1.16) 0.24 Male recipient 1.03 (0.87 1.23) 0.70 0.76 (0.63 0.92) 0.005 Recipient age $50 yr 1.39 (1.15 1.67),0.001 0.77 (0.64 0.94) 0.01 Recipient BMI 25 30 versus 18 24.9 kg/m 2 0.94 (0.81 1.11) 0.48 1.11 (0.92 0.135) 0.28 African-American recipient 1.05 (0.88 1.24) 0.60 1.56 (1.29 1.90),0.001 Dialysis $24 mo 1.16 (0.95 1.41) 0.14 0.97 (0.78 1.22) 0.82 Wait time $24 mo 1.11 (0.93 1.32) 0.27 1.02 (0.82 0.126) 0.89 IL-2ra induction versus antithymocyte globulin 1.03 (0.85 1.25) 0.77 1.15 (0.91 1.46) 0.24 Other induction versus antithymocyte globulin 0.92 (0.66 1.29) 0.63 1.38 (0.96 1.98) 0.08 No induction versus antithymocyte globulin 1.19 (0.97 1.45) 0.09 1.26 (0.99 1.60) 0.06 Maintenance immunosuppression other than TAC/MPA 1.34 (1.15 1.57) 0.0003 1.68 (1.40 2.03),0.001 Peak panel-reactive antibody.0% 0.98 (0.84 1.15) 0.83 0.99 (0.82 1.19) 0.87 Male donor 0.95 (0.81 1.12) 0.55 0.84 (0.70 1.02) 0.08 Donor age $35 yr 1.24 (1.02 1.51) 0.03 1.26 (1.00 1.60) 0.06 African American donor 1.11 (0.88 1.41) 0.38 1.43 (1.12 1.83) 0.004 Donor death: CVA versus trauma 0.88 (0.72 1.09) 0.24 0.91 (0.71 1.17) 0.48 Donor death: other versus trauma 0.92 (0.72 1.17) 0.49 0.96 (0.72 1.28) 0.78 Expanded-criteria donor 1.39 (1.13 1.72) 0.002 1.50 (1.17 1.91) 0.001.0 HLA mismatches 1.35 (1.02 1.77) 0.03 1.41 (1.00 1.99) 0.05 Cold ischemia time $24 h 1.05 (0.87 1.27) 0.61 1.07 (0.85 1.34) 0.57 Delayed graft function 1.40 (1.18 1.66),0.001 2.02 (1.66 2.45),0.001 Patients had type 2 diabetes, were age 18 59 years, had a body mass index # 30 kg/m 2, and were undergoing deceased-donor kidney transplantation alone or simultaneous pancreas kidney transplantation from 2000 through 2008. Hazard ratios were obtained from Cox proportional hazards regression model. CI, confidence interval; DDKA, deceased-donor kidney transplantation alone; SPK, simultaneous pancreas kidney transplantation; BMI, body mass index; IL-2ra, interleukin-2 receptor antagonist; TAC, tacrolimus; MPA, mycophenolate mofetil; CVA, cerebrovascular accident. Although the findings on multivariable analysis do not suggest a beneficial effect of the added pancreas transplant on kidney graft or patient survival per se, thisshouldnot dissuade consideration of SPK for the selected T2DM candidate who does not have a living donor available. Our findings do not consider the benefits of euglycemia on quality of life (10, 11) or the possible effect on secondary complications of diabetes, as has been described in T1DM recipients (12). In addition, the effect of euglycemia on graft or patient survival may not be apparent until beyond 5 years (12). The survival benefit of SPK over DDKA has been consistently reported for T1DM recipients, with difference in survival noted as early as 12 months after transplantation (13). This benefit appears to be related to both kidney donor quality and a functional pancreas transplant for the first era (approximately 5 years) after transplantation (2, 14). However, in a large retrospective analysis of T1DM recipients, after 5 years SPK was associated with a reduction in mortality risk, and after 10 years, with a reduction in risk for kidney graft loss, suggesting that over time the benefits of SPK become more evident (3, 15). In contrast to the relatively similar survival outcomes noted with LDKA versus SPK in T1DM recipients in previous studies (2,3,13,15), for the T2DM recipients LDKA provides both graft and patient survival advantages versus SPK, at least for the follow-up duration of this study. Whether the euglycemic effects of the added pancreas transplant ultimately may lead to a survival advantage compared with LDKA cannot be ruled out, but it is not evident on the basis of 5-year estimated survival rates. This is different from the findings in T1DM recipients, who show similar unadjusted survival before the 5-year point with SPK versus LDKA (13). The results from these studies in total suggest that a priority for SPK allocation is more strongly supported for the T1DM candidate than the T2DM candidate. From the perspective of deceased-donor organ allocation, the results of this study support the use of SPK for T2DM recipients under current clinical practice, and expansion of this population without continued consideration of longer-term outcomes may have untoward effects on other patient populations. As noted earlier, T2DM SPK recipients who underwent transplantation from 2000 through 2008 are likely to have unaccounted-for differences from other T2DM candidates that contributed to their eligibility for SPK. Expansion of this cohort would reduce the number of donor pancreases available and reduce the number of SPK procedures for T1DM recipients, a population in whom survival benefits are much better defined. Further, the priority given to SPK versus DDKA in the current allocation algorithm favors the T2DM patient undergoing SPK at a pancreas transplant center over all

662 Clinical Journal of the American Society of Nephrology Figure 2. Kaplan-Meier estimated kidney graft and patient survival among patients with type 2 diabetes mellitus (T2DM), age 18 59 years, body mass index, 30 kg/m 2 undergoing living-donor kidney transplantation alone (LDKA) or simultaneous pancreas kidney transplantation (SPK) from 2000 through 2008. (A) Patient survival, SPK versus LDKA: 82.0% versus 87.3% (log-rank P=0.003). (B) Kidney graft survival, SPK versus LDKA: 75.2% versus 81.2% (log-rank P=0.002). (C) Death-censored kidney graft survival, SPK versus LDKA: 86.2% versus 91.1% (log-rank P=0.003).

Clin J Am Soc Nephrol 7: 656 664, April, 2012 SPK or KTA in T2DM, Wiseman and Gralla 663 Table 3. Risk factors associated with death and death-censored kidney graft loss Variable Patient Survival Hazard Ratio (95% CI) Death-Censored Kidney Graft Survival Hazard Ratio (95% CI) LDKA versus SPK 0.50 (0.35 0.71),0.001 0.66 (0.44 0.98) 0.04 Male recipient 0.83 (0.62 1.11) 0.21 0.75 (0.54 1.05) 0.09 Recipient age $50 (yr) 1.74 (1.28 2.36),0.001 0.88 (0.64 1.23) 0.46 Recipient BMI 25 30 versus 18 24.9 kg/m 2 0.80 (0.60 1.06) 0.12 0.81 (0.58 1.12) 0.20 African-American recipient 1.29 (0.91 1.82) 0.15 1.70 (1.17 2.46) 0.005 Dialysis $24 mo 1.18 (0.86 1.60) 0.31 0.94 (0.64 1.38) 0.74 IL-2ra induction versus antithymocyte globulin 1.13 (0.77 1.66) 0.53 0.71 (0.45 1.13) 0.15 Other induction versus antithymocyte globulin 1.24 (0.69 2.22) 0.47 1.29 (0.70 2.37) 0.41 No induction versus antithymocyte globulin 1.28 (0.88 1.84) 0.19 1.06 (0.71 1.58) 0.77 Maintenance immunosuppression other than 1.58 (1.19 2.10) 0.002 1.46 (1.05 2.02) 0.02 TAC/MPA Peak panel-reactive antibody.0% 1.10 (0.84 1.46) 0.49 1.05 (0.76 1.46) 0.76.0 HLA mismatches 0.93 (0.53 1.61) 0.78 1.29 (0.62 2.65) 0.50 Delayed graft function 2.26 (1.47 3.46),0.001 4.55 (3.01 6.87),0.001 Patients had type 2 diabetes, were age 18 59 years, had a body mass index # 30 kg/m 2, and were undergoing simultaneous pancreas kidney transplantation or living-donor kidney transplantation alone from 2000 through 2008. Hazard ratios were obtained from Cox proportional hazards regression model. CI, confidence interval; LDKA, living-donor kidney transplantation alone; SPK, simultaneous pancreas kidney transplantation; BMI, body mass index; IL-2ra, interleukin-2 receptor antagonist; TAC, tacrolimus; MPA, mycophenolate mofetil. other kidney transplant recipients at transplant centers that do not perform pancreas transplantations for a given region. The waiting time for transplantation is expected to be significantly shorter for the SPK candidate. UNOS policy has therefore placed a 6-month review process in place in order to reduce the BMI eligibility criteria by 2 kg/m 2 if more than 10% of the SPK waiting list is composed of patients with T2DM. Aside from the limitations of this comparison noted above, other limitations of this study deserve mention. This is a retrospective database analysis; the definition of T2DM is left to the discretion of the individual reporting center and does not take into account variations in diabetes phenotype, such as the subset of T2DM recipients with mature-onset diabetes of the young (16). Until recently, the definition of diabetes as reported to the SRTR does not require data regarding medication use, C-peptide values, or any other feature that may confirm that the patient does not have T1DM. However, the diagnosis of T2DM is less likely to be in error than is the diagnosis of T1DM (e.g., it is more common to mislabel a patient with T2DM who requires insulin as having T1DM than the converse). In addition, when a cohort with BMI #30 kg/m 2 is being selected, it should be acknowledged that BMI can change over time and that the BMI analyzed represents the BMI at listing rather than at time of transplantation. Finally, because of the study s retrospective design, we could analyze and report only the variables available through the Standard Analysis Files from SRTR, in which certain variables are universally collected (e.g., death, graft loss), whereas others are less complete (acute rejection, immunosuppression). Given that our primary endpoints were graft and patient survival, our data form the best interpretation of available data. In summary, carefully selected patients with T2DM enjoy excellent patient and kidney and pancreas graft survival with SPK, but graft and patient survival at 5 years are not associated with the added pancreas transplant compared with DDKA. For a similar cohort of T2DM candidates, LDKA provides better patient and graft survival than does SPK. Current allocation policy that encourages consideration of SPK for T2DM candidates deserves close monitoring to better define the outcomes of the T2DM SPK recipient and the potential effect on organ availability for other populations on the waiting list. Disclosures None. References 1. Axelrod DA, McCullough KP, Brewer ED, Becker BN, Segev DL, Rao PS: Kidney and pancreas transplantation in the United States, 1999-2008: The changing face of living donation. Am J Transplant 10: 987 1002, 2010 2. Weiss AS, Smits G, Wiseman AC: Twelve-month pancreas graft function significantly influences survival following simultaneous pancreas-kidney transplantation. Clin J Am Soc Nephrol 4: 988 995, 2009 3. Morath C, Zeier M, Döhler B, Schmidt J, Nawroth PP, Opelz G: Metabolic control improves long-term renal allograft and patient survival in type 1 diabetes. J Am Soc Nephrol 19: 1557 1563, 2008 4. Sampaio MS, Kuo HT, Bunnapradist S: Outcomes of simultaneous pancreas-kidney transplantation in type 2 diabetic recipients. Clin J Am Soc Nephrol 6: 1198 1206, 2011 5. Kaufman DB, Sutherland DE: Simultaneous pancreas-kidney transplants are appropriate in insulin-treated candidates with uremia regardless of diabetes type. Clin J Am Soc Nephrol 6: 957 959, 2011 6. Sener A, Cooper M, Bartlett ST: Is there a role for pancreas transplantation in type 2 diabetes mellitus? Transplantation 90: 121 123, 2010

664 Clinical Journal of the American Society of Nephrology 7. Light JA, Barhyte DY: Simultaneous pancreas-kidney transplants in type I and type II diabetic patients with end-stage renal disease: Similar 10-year outcomes. Transplant Proc 37: 1283 1284, 2005 8. Singh RP, Rogers J, Farney AC, Hartmann EL, Reeves-Daniel A, Doares W, Ashcraft E, Adams PL, Stratta RJ: Do pretransplant C-peptide levels influence outcomes in simultaneous kidneypancreas transplantation? Transplant Proc 40: 510 512, 2008 9. Nath DS, Gruessner AC, Kandaswamy R, Gruessner RW, Sutherland DE, Humar A: Outcomes of pancreas transplants for patients with type 2 diabetes mellitus. Clin Transplant 19: 792 797, 2005 10. Smith GC, Trauer T, Kerr PG, Chadban SJ: Prospective quality-oflife monitoring of simultaneous pancreas and kidney transplant recipients using the 36-item short form health survey. Am J Kidney Dis 55: 698 707, 2010 11. Isla Pera P, Moncho Vasallo J, Torras Rabasa A, Oppenheimer Salinas F, Fernández Cruz Pérez L, Ricart Brulles MJ: Quality of life in simultaneous pancreas-kidney transplant recipients. Clin Transplant 23: 600 605, 2009 12. Gremizzi C, Vergani A, Paloschi V, Secchi A: Impact of pancreas transplantation on type 1 diabetes-related complications. Curr Opin Organ Transplant 15: 119 123, 2010 13. Young BY, Gill J, Huang E, Takemoto SK, Anastasi B, Shah T, Bunnapradist S: Living donor kidney versus simultaneous pancreas-kidney transplant in type I diabetics: An analysis of the OPTN/UNOS database. Clin J Am Soc Nephrol 4: 845 852, 2009 14. Weiss AS, Smits G, Wiseman AC: Simultaneous pancreas-kidney versus deceased donor kidney transplant: can a fair comparison be made? Transplantation 87: 1402 1410, 2009 15. Morath C, Zeier M, Döhler B, Schmidt J, Nawroth PP, Schwenger V, Opelz G: Transplantation of the type 1 diabetic patient: The long-term benefit of a functioning pancreas allograft. Clin J Am Soc Nephrol 5: 549 552, 2010 16. Raile K, Klopocki E, Wessel T, Deiss D, Horn D, Müller D, Ullmann R, Grüters A: HNF1B abnormality (mature-onset diabetes of the young 5) in children and adolescents: High prevalence in autoantibody-negative type 1 diabetes with kidney defects. Diabetes Care 31: e83, 2008 Received: August 15, 2011 Accepted: January 25, 2012 Published online ahead of print. Publication date available at www. cjasn.org. See related editorial, Type 2 Diabetes: The Best Transplant Option Is Still Uncertain, on pages 530 532. Access to UpToDate on-line is available for additional clinical information at www.cjasn.org.