Outcomes of Simultaneous Pancreas-Kidney ... - Semantic Scholar

Article

Outcomes of Simultaneous Pancreas-Kidney Transplantation in Type 2 Diabetic Recipients Marcelo Santos Sampaio,*† Hung-Tien Kuo,*‡ and Suphamai Bunnapradist*

Summary Background and objectives Type 2 diabetic patients with end-stage renal disease may receive a simultaneous pancreas-kidney (SPK) transplant. However, outcomes are not well described. Risks for death and graft failure were examined in SPK type 2 diabetic recipients. Design, setting, participants, & measurements Using the United Network for Organ Sharing database, outcomes of SPK transplants were compared between type 2 and type 1 diabetic recipients. All primary SPK adult recipients transplanted between 2000 and 2007 (n ⫽ 6756) were stratified according to end-stage pancreas disease diagnosis (type 1: n⫽6141, type 2: n⫽582). Posttransplant complications and risks for death and kidney/pancreas graft failure were compared. Results Of the 6756 SPK transplants, 8.6% were performed in recipients with a type 2 diabetes diagnosis. Rates of delayed kidney graft function and primary kidney nonfunction were higher in the type 2 diabetics. Five-year overall and death-censored kidney graft survival were inferior in type 2 diabetics. After adjustment for other risk factors, including recipient (age, race, body weight, dialysis time, and cardiovascular comorbidities), donor, and transplant immune characteristics, type 2 diabetes was not associated with increased risk for death or kidney or pancreas failure when compared with type 1 diabetic recipients. Conclusions After adjustment for other risk factors, SPK recipients with type 2 diabetes diagnosis were not at increased risk for death, kidney failure, or pancreas failure when compared with recipients with type 1 diabetes. Clin J Am Soc Nephrol 6: 1198 –1206, 2011. doi: 10.2215/CJN.06860810

Introduction Simultaneous pancreas-kidney transplantation (SPK) is one of the treatment options for type 1 diabetes mellitus (T1DM) patients with end-stage renal disease (ESRD) (1). Compared with kidney transplant alone, a successful SPK may improve quality of life (2,3), diminish the progression of diabetic complications (4,5), and possibly prolong patient and kidney allograft survival (6 –9). In T1DM, SPK transplant outcomes are excellent, with a reported 5-year patient, kidney, and pancreas graft survival of 88%, 77% (10), and 69% (11), respectively. The outcomes of SPK in type 2 diabetes mellitus (T2DM) are less well described and mostly represent single-center experiences. The largest published study included 38 SPK T2DM recipients, defined by a serum C-peptide level ⬎0.8 ng/ml (12–14). In T1DM and T2DM, 5-year patient survival was 85% and 73%, pancreas survival was 71% and 67%, and kidney survival was 77% and 72%, respectively. Another smaller study defined T2DM by a C-peptide level ⱖ2.0 ng/ml (n ⫽ 7) and compared SPK outcomes with T1DM recipients. In T1DM and T2DM, recipients’ 3-year patient survival was 94% and 71%, death-censored 1198

Copyright © 2011 by the American Society of Nephrology

pancreas survival was 87% and 100%, and deathcensored kidney survival was 95% and 100%, respectively (15). Two additional studies reported outcomes in T2DM patients with pancreas transplants. One study with 17 recipients (7 were SPK) had 94% pancreas survival in 1 year, and 11 of the 12 recipients were alive and euglycemic after 3 years (16). The second study included only four SPK transplants, and the outcomes were one death, one pancreas failure within 2 years, and two recipients with euglycemic after 7 years (17). In this study we analyzed data from the Organ Procurement and Transplant Network/United Network for Organ Sharing (OPTN/UNOS), described characteristics of ESRD T2DM recipients who underwent a SPK transplant in the United States, and compared their outcomes with T1DM recipients. Diabetes was defined based on the diagnosis of end-stage pancreas disease (ESPD) as declared by each pancreas transplant center to UNOS. The objectives of this study were to describe characteristics of T2DM ESRD recipients considered for a SPK and to identify risks for death and graft failure compared with T1DM recipients.

*Department of Medicine, David Geffen School of Medicine, University of California–Los Angeles, Los Angeles, California; † Department of Nephrology, Pedro Ernesto University Hospital, Rio de Janeiro State University, Rio de Janeiro, Brazil; and ‡ Department of Nephrology, Kaohsiung Medical University Hospital, Kaohsiung Medical University, Taiwan, Republic of China Correspondence: Suphamai Bunnapradist, 1033 Gayley Avenue, Suite 208, Los Angeles, CA, 90024. Phone: 310794-8516; Fax: 310794-8589; E-mail: bunnapradist@mednet. ucla.edu

www.cjasn.org Vol 6 May, 2011

Clin J Am Soc Nephrol 6: 1198 –1206, May, 2011

Materials and Methods Study Population Using the OPTN/UNOS database as of March 2009, we selected all recipients of a primary SPK transplant between 2000 and 2007. Recipients with a previous kidney- or pancreas-alone transplant were excluded. The cohort was stratified into two groups according to diabetes diagnosis: type 1 and 2 diabetes. Diabetes type definition was taken from the SPK transplant recipient registration form and was based on the diagnosis of ESPD (code 5001 and 5002, corresponding to T1DM and T2DM, respectively). Thirty-three recipients were excluded from the analysis because the cause of ESPD was retransplant/graft failure (n ⫽ 7), chronic pancreatitis (n ⫽ 1), cystic fibrosis (n ⫽ 1), or pancreatectomy (n ⫽ 2) and diabetes type other or unknown (n ⫽ 22). Only two recipients in the type 1 group were ⬍18 years old. Statistical Analyses Recipient, donor, and transplant baseline characteristics, including immunosuppressive regimens at discharge, were compared. Continuous variables were compared using the Kruskall–Wallis test and categorical variables using the ␹2 test. The results are shown as median values with 25th to 75th percentile ranges or percentages. Posttransplant events related to the kidney and pancreas allograft were identified and compared between study groups. The transplant date was used as the index date and patients were followed up to a maximum of 5 years after transplant. Kidney, pancreas, and patient survival were compared at the end of this period. Overall kidney graft survival was determined from the time of kidney transplantation until patient death, kidney retransplantation, return to dialysis, or end of follow-up. Overall pancreas graft survival was defined as pancreas failure and patient death. Death-censored graft survival analysis was different from overall survival and did not include patient death as a cause of graft loss. Patient survival was determined from time of kidney transplantation until death or end of follow-up. The Kaplan–Meier product limit method was used to generate survival curves, logistic regression was used to examine risks associated with posttransplant complications, and Cox regression was used to examine risks associated with graft loss and death. Baseline recipient, donor, and transplant characteristics were treated as potential confounders and were included in the regression model. The multivariate model was constructed using a stepwise method. First, covariates found to have P ⱕ 0.10 in the univariate analysis were included in the model. After initial covariate interactions, the model was optimized maintaining only the covariates with P ⱕ 0.05. The variable related with the diabetes type, used to test the association of T2DM versus T1DM with the outcomes, was forced into the model. The results were expressed as odds ratios or hazard ratios with their 95% confidence intervals and associated P values. Risks factors for death and kidney/pancreas failure were also examined separately for type 2 and type 1 diabetics. This analysis intended to define differences in the contribution of similar risk factors for outcomes in a type 1 and 2 diabetic population. Two models were analyzed. The first (model 1) was constructed as described above to the

SPK in Type 2 Diabetic Recipients, Sampaio et al.

1199

overall cohort, whereas the second (model 2) included a covariate to account for interaction between induction agents and steroid use. All P values were two-tailed and P ⬍ 0.05 was considered significant. STATA version 10 (Statacorp, College Station, TX) was used in all statistical analyses.

Results From 1992 to 2009, 24,084 candidates were waitlisted for a SPK transplant. Of these, 87.4% had T1DM, 7.8% had T2DM, and 4.8% had an alternate cause for diabetes. The percentage of T2DM patients on the SPK waitlist increased from 3% (n ⫽ 30) in 1994 to 8% (n ⫽ 149) in 2000. Between 2000 and 2007, these percentages varied from 7.7% to 10%. Of the 6756 SPK recipients transplanted between 2000 and 2007 (96.5% of all SPK), 582 (8.6%) were type 2 diabetics and 6141 (90.8%) were type 1 (Figure 1). SPK transplants in type 2 recipients were mainly performed in UNOS regions 2, 3, 5, and 11 (Figure 2). Baseline Characteristics Recipient, donor, and transplant baseline characteristics according to diabetes type are described in Table 1. Compared with T1DM recipients, T2DM recipients were older at diabetes onset and at the time of transplant; were more often obese, African American, and Hispanic; had a percent panel of reactive antibodies ⬎20%; and had a higher pretransplant dialysis time. Type 2 diabetic SPK recipients also had fewer years of diabetes disease, less frequently had private health insurance, and used fewer induction drugs when compared with T1DM patients. The only difference in pretransplant comorbidities was a higher proportion of hypertension in T1DM SPK recipients. Data on average insulin dose used before transplant were available in 2188 of the 6141 T1DM and in 184 of the 582 T2DM recipients. In type 1 and 2 diabetic recipients, the average pretransplant insulin dose was 34 (23.5 to 49) and 30 (17.5 to 45) IU/d, respectively. Posttransplant Complications Posttransplant complication rates in T1DM and T2DM recipients are shown in Table 2. Rates of delayed kidney graft function (11.7% versus 7.8%, P ⬍ 0.001) and kidney primary nonfunction (0.47% versus 1.03%, P ⫽ 0.03) were significantly more frequent in the T2DM recipients. Pancreas complications were similar between groups. The adjusted odds ratio (OR) for 1-year cumulative kidney and pancreas acute rejection was 1.19 (95% confidence interval [CI]: 0.99 to 1.43) and for delayed kidney graft function was 1.08 (95% CI: 0.79 to 1.48) for T2DM recipients, with T1DM recipients as the reference. Patient, Kidney, and Pancreas Survival Five-year unadjusted patient survival is shown in Figure 3. Overall kidney graft survival (77.8 versus 73.5%; P ⫽ 0.007) and death-censored kidney graft survival (85.3 versus 82.95%; P ⫽ 0.04) were inferior in T2DM. However, after adjustment for confounders, diabetes type was not identified as a risk factor for outcomes. Hazard ratios (HRs) were 1.10 (95% CI: 0.86 to 1.42) for patient death, 1.08 (95% CI: 0.91 to 1.28) for overall pancreas allograft failure, and 1.16 (95% CI 0.95 to 1.39) for kidney allograft failure in T2DM recipients, with

1200

Clinical Journal of the American Society of Nephrology

Figure 1. | Absolute number of candidates (black bars) and recipients (type 2, white bars and type 1, gray bars) of a SPK transplant. Type 2 diabetics represent a small percentage of those waitlisted and that received a SPK. Data from OPTN/UNOS as of March 2009.

T1DM as the reference. A trend toward increased risk for death-censored kidney graft failure (HR: 1.25; 95% CI: 0.97 to 1.61) was found in the T2DM group. A small but significant increased incidence of graft failure secondary to surgical complication (0.02% [1 of 6141] versus 0.2% [1 of 582], P ⫽ 0.04) was seen in T2DM. Risk factors independently associated with outcomes in the overall SPK cohort were shown in Table 3. The most significant recipient characteristics associated with death were age ⬎45 years old (versus ⱕ45 years), ⬎1 year of pretransplant dialysis (versus pre-emptive), comorbidities such as coronary and peripheral vascular diseases (versus no disease), and exocrine drainage to the bladder (versus enteric). Recipient characteristics associated with kidney and pancreas failure but not with risk of death were AA ethnicity (versus Caucasians) and a body mass index (BMI) ⬎30 (versus 18.5 to 25). Cold ischemia time ⬎24 hours (versus ⬍12 hours) was associated with increased kidney graft failure. Recipients with an HLA mismatch of five to six alleles (versus no mismatch) had a trend toward increased kidney graft failure. Figure 2. | Number of SPK transplants performed in ESRD type 1 and type 2 recipients according to UNOS regions between the years 2000 and 2007. UNOS regions: 1 ⫽ Connecticut, Maine, Massachusetts, New Hampshire, Rhode Island; 2 ⫽ Washington, DC, Delaware, Maryland, New Jersey, Pennsylvania, West Virginia; 3 ⫽ Arkansas, Florida, Georgia, Louisiana, Mississippi, Puerto Rico; 4 ⫽ Oklahoma, Texas; 5 ⫽ Arizona, California, Nevada, New Mexico, Utah; 6 ⫽ Alaska, Hawaii, Idaho, Montana, Oregon, Washington; 7 ⫽ Illinois, Minnesota, North Dakota, South Dakota, Wisconsin; 8 ⫽ Colorado, Iowa, Kansas, Missouri, Nebraska, Wyoming; 9 ⫽ New York, Vermont; 10 ⫽ Indiana, Michigan, Ohio; 11 ⫽ Kentucky, North Carolina, South Carolina, Tennessee, Virginia.

Comparison of Risk Factors for Death, Kidney Failure, and Pancreas Failure between Type 2 and Type 1 Diabetic Recipients To separately evaluate the contribution of each risk factor to T2DM and T1DM recipients, the Cox regression model was applied to T2DM recipients and then repeated in T1DM recipients. Covariates found to be significant risk factors for the outcomes of T2DM recipients are shown in Table 4, and their HRs were compared with those found in a T1DM population. Two models (as



1201

Table 1. Baseline recipient, donor, and transplant characteristics

Characteristics Age at diabetes onset, years BMI, kg/m2 ⬍18.5 18.5 to 25 25 to 30 ⬎30 Duration of diabetes, years Age at transplant, years Age ⬎45 years Race Caucasian AA Hispanic others Male gender PRA 0% 1% to 20% ⬎20% Degree of HLA mismatch Pretransplant dialysis duration, days Pre-emptive transplant EPD into bladder Private heath insurance Induction ALG/OKT3 ATG IL2-RA alemtuzumab multiple none Calcineurin inhibitor none cyclosporine A tacrolimus AP drug MMF SRL none multiple AZA Steroid CAD CVD Hypertension PVD Donor age ⬎45 years Donor race AA Donor COD/stroke CIT kidney, hours

T1DM (n ⫽ 6141)

T2DM (n ⫽ 582)

12 (8 to 18)

25 (18 to 31)

2.8 54.2 32.3 10.7 26 (21 to 32) 40 (34 to 46) 27.5

2.8 43.9 36.2 17.1 21 (16 to 27) 47 (40 to 52) 54.3

77.5 13.1 7.6 1.8 61.0

62.2 22.2 11.5 4.1 68.9

67.0 22.4 10.6 5 (4 to 5) 603 (339 to 982) 20.9 14.3 46.2

60.1 24.4 15.5 5 (4 to 5) 714 (403 to 1128) 17.5 15.5 38.5

3.3 37.3 24.2 7.7 4.1 23.6

2.4 40.9 10.8 4.5 3.8 37.6

6.3 8.9 84.7

8.9 4.5 86.6

75.9 11.3 7.16 5.06 0.5 94.6 10.2 2.2 79.1 7.0 7.8 14.3 19.7 12 (8.5 to 15.9)

76.6 7.0 11.0 4.6 0.7 93.0 10.5 2.2 71.9 6.2 7.4 18.0 22.5 12 (8.8 to 15)

P ⬍0.001 ⬍0.001

⬍0.001 ⬍0.001 ⬍0.001 ⬍0.001

⬍0.001 ⬍0.001

0.02 ⬍0.001 0.05 0.73 ⬍0.001 ⬍0.001

⬍0.001

⬍0.01

0.11 0.83 0.77 0.001 0.52 0.71 0.01 0.11 0.83

Results presented as percent or as median (25th to 75th percentile). Data for BMI missing in 1.4% (n ⫽ 95), age of diabetes onset missing in 11.4% (n ⫽ 766, 656 in T1DM and 110 in T2DM), PRA missing in 2.6% (n ⫽ 178), and HLA mismatch missing in 0.06% (n ⫽ 4) of the cohort recipients. PRA, percent of reactive antibodies; EPD, exocrine pancreas drainage; ALG, anti-lymphocyte globulin; OKT3, muronab-anti CD3 antibody; ATG, rabbit anti-thymocyte globulin; IL2-RA, interleukin 2 receptor antibody (basiliximab and daclizumab); AP, antiproliferative; MMF, mycophenolate mofetil (includes nonenteric and enteric formulations); SRL, sirolimus; AZA, azathioprine; CAD, coronary artery disease; CVD, cerebrovascular disease; PVD, peripheral vascular disease; COD, cause of death; CIT cold ischemia time.

described in methods) were used in T2DM. In the first model (model 1), recipient age ⬎45 years old (versus ⱕ45 years old) and a period of ⬎1 year on dialysis before transplant (versus pre-emptive transplantation) were associated with increased death. One year or more on dialysis (versus pre-emptive transplantation) and use of alemtuzumab (versus no induction) were associated with

increased kidney allograft failure, and a BMI of 25 to 30 and ⬎30 kg/m2 (versus BMI of 18.5 to 25 kg/m2) were associated with increased risk for pancreas graft failure. In model 2, a variable to account for the interaction of steroid with induction agent use was included. This further adjusted the risks for kidney failure associated with alemtuzumab use, becoming not associated with increased kidney failure in type

1202


Table 2. Post-SPK transplant complications in type 1 and 2 diabetic recipients

Initial hospitalization time in days (posttransplantation) One-year cumulative combined kidney and pancreas rejection Kidney complications delayed graft function primary nonfunction Pancreas complications abscess/local infection anastomosis leak pancreatitis primary nonfunction surgery to convert bladder to enteric drainage

T1DM

T2DM

P

9 (7 to 14)

9 (7 to 14)

–

15.8

14.7

0.89

7.8 0.47

11.7 1.03

⬍0.001 0.03

4.37 2.15 2.43 0.86 1.43

2.95 1.91 1.56 0.86 1.37

0.23 0.91 0.42 0.75 0.91

Results in median (25th to 75th percentile) or in percent.

Figure 3. | Kaplan–Meier curves for (A) patient survival, (B) overall pancreas graft survival, (C) overall kidney graft survival, and (D) death-censored kidney graft survival in type 1 diabetic E and type 2 diabetic 䡲 recipients. Adjusted HR with 95% CI for the association of T2DM (T1DM as reference) with patient death and pancreas/kidney graft failure were shown inside of the graphics.

2 recipients and associated with decreased failure in type 1 diabetic recipients. When we compared risks between recipients with diabetes diagnosis (Table 4), we found that in T2DM and T1DM, risk for patient death increased respec-

tively in 116% and 29% with an older recipient age and in 160% and 59% with a pretransplant dialysis time ⬎1 year. Pretransplant dialysis ⬎1 year also increased the risk for overall kidney graft loss in 148% and 25% in T2DM and



1203

Table 3. Estimated HR for independent factors associated with SPK transplant outcomes Covariate Patient death diabetes recipient age pretransplant dialysis CAD PVD EPD calcineurin inhibitors

Reference

Levels

Unadjusted HR (95% CI)

P

Adjusted HR (95% CI)

P

T1DM ⱕ45 years Pre-emptive No No Enteric No drug

T2DM ⬎45 years ⬎1 year Yes Yes Bladder Cyclosporine Tacrolimus MMFa CVD ⬎45 years

1.25 (0.97 to 1.59) 1.39 (1.19 to 1.62) 1.56 (1.26 to 1.94) 1.51 (1.23 to 1.86) 1.51 (1.17 to 1.93) 1.37 (1.13 to 1.64) 0.49 (0.32 to 0.58) 0.33 (0.27 to 0.42) 0.43 (0.34 to 0.54) 1.38 (1.16 to 1.63) 1.60 (1.27 to 2.02)

0.08 ⬍0.001 ⬍0.001 ⬍0.001 0.001 0.001 ⬍0.001 ⬍0.001 ⬍0.001 ⬍0.001 ⬍0.001

1.10 (0.85 to 1.42) 1.29 (1.10 to 1.52) 1.57 (1.26 to 1.94) 1.27 (1.02 to 1.57) 1.34 (1.03 to 1.73) 1.29 (1.02 to 1.57) 0.46 (0.32 to 0.65) 0.40 (0.30 to 0.52) 0.69 (0.53 to 0.90) 1.18 (0.98 to 1.42) 1.29 (1.00 to 1.66)

0.44 0.002 ⬍0.001 0.03 0.03 0.01 ⬍0.01 ⬍0.01 0.007 0.07 0.04

T2DM AA ⬎30% ⬎1 year Yes Cyclosporine Tacrolimus MMFa 5 to 6 ⬎24 hours CVD ⬎45 years

1.29 (1.07 to 1.55) 1.55 (1.34 to 1.80) 1.29 (1.09 to 1.54) 1.29 (1.11 to 1.51) 0.51 (0.41 to 0.63) 0.43 (0.34 to 0.56) 0.37 (0.31 to 0.44) 0.46 (0.39 to 0.55) 1.27 (0.87 to 1.86) 1.31 (1.13 to 1.52) 1.33 (1.16 to 1.52) 1.56 (1.30 to 1.87)

0.007 ⬍0.001 0.003 0.001 ⬍0.001 ⬍0.001 ⬍0.001 ⬍0.001 ⬍0.001 ⬍0.001 ⬍0.001 ⬍0.001

1.16 (0.95 to 1.39) 1.51 (1.29 to 1.76) 1.25 (1.05 to 1.49) 1.26 (1.07 to 1.47) 0.76 (0.60 to 0.97) 0.53 (0.41 to 0.70) 0.46 (0.34 to 0.46) 0.72 (0.58 to 0.89) 1.44 (0.98 to 2.11) 1.26 (1.09 to 1.46) 1.19 (1.04 to 1.38) 1.38 (1.14 to 1.67)

0.136 ⬍0.001 0.01 0.004 0.03 ⬍0.001 ⬍0.001 0.003 0.06 0.002 0.01 0.001

T2DM AA ⬎30% ⬎1 year Cyclosporine Tacrolimus MMFa SRL CVD ⬎45 years

1.15 (0.96 to 1.35) 1.28 (1.12 to 1.47) 1.34 (1.15 to 1.55) 1.20 (1.05 to 1.37) 0.52 (0.42 to 0.65) 0.43 (0.37 to 0.49) 0.51 (0.43 to 0.59) 0.49 (0.40 to 0.61) 1.36 (1.21 to 1.53) 1.61 (1.37 to 1.88)

0.109 ⬍0.01 ⬍0.001 0.005 ⬍0.001 ⬍0.001 ⬍0.001 ⬍0.001 ⬍0.001 ⬍0.001

1.08 (0.91 to 1.28) 1.25 (1.08 to 1.43) 1.31 (1.13 to 1.53) 1.20 (1.05 to 1.38) 0.62 (0.48 to 0.79) 0.51 (0.42 to 0.62) 0.69 (0.57 to 0.83) 0.66 (0.52 to 0.83) 1.22 (1.07 to 1.37) 1.40 (1.17 to 1.65)

0.39 0.002 ⬍0.001 0.006 ⬍0.001 ⬍0.001 ⬍0.001 ⬍0.001 0.002 ⬍0.001

AP drug donor COD donor age Overall kidney graft loss diabetes ethnicity BMI pretransplant dialysis time steroid calcineurin inhibitors

No drug No CVD ⱕ45 years

AP drug HLA mismatch CIT donor COD donor age Overall pancreas graft loss diabetes ethnicity BMI pretransplant dialysis time calcineurin inhibitors

No drug 0 ⬍12 hours Others ⱕ45 years

T1DM Caucasian 18.5% to 25% Pre-emptive No drug No drug

T1Dm Caucasian 18.5% to 25% Pre-emptive No drug

AP drug

No drug

donor COD donor age

Others ⱕ45 years

Results are presented as HR with 95% CI. a Includes nonenteric and enteric formulations.

T1DM, respectively, and obesity was associated with a 71% and 27% increased risk for pancreas failure in T2DM and T1DM recipients, respectively. Recipient age, time on dialysis pretransplant, and BMI may have been the covariates responsible for adjusting risks in T2DM and for determining that T2DM recipients had a risk for SPK outcomes similar to T1DM recipients.

Discussion We studied characteristics and outcomes of all SPK recipients between the years 2000 and 2007 in the United States diagnosed as T2DM according to the pancreas transplant center. The large size of our cohort (n ⫽ 582) when compared with previous single-center reports allowed us to examine adjusted risks for death and kidney/pancreas failure associated with diabetes diagnosis (T1DM or T2DM). In contrast to previous studies (13–15), we did not find a significant difference in patient survival between T1DM and T2DM recipients. Also, after adjustments, T2DM was not associated with increased death when compared with T1DM recipients. In T2DM recipients, older age and a longer pretransplant dialysis time were associated with

higher risks for death than in T1DM recipients. These differences have to be taken in account when comparing survival. Comorbidities developed during dialysis time may be more intense in T2DM than in T1DM recipients (18 –20). However, this is speculative because there are few studies comparing dialysis outcomes between T1DM and T2DM with ESRD. Overall, in insulin-dependent ESRD candidates for a SPK transplant, cardiovascular disease and time on waitlist were identified as potentially modifiable risks for death. There is a concern that the insulin resistance in T2DM recipients would lead to an increase in pancreas allograft insulin production with ␤ cell exhaustion and a decrease in the pancreas functional survival (21). However, neither our study or previous studies support increased pancreas graft failure in T2DM. We found that 5-year pancreas survival in T2DM was 69.8%, comparable to the 72.4% in T1DM recipients. Light et al. found a 5-year pancreas survival of 67% in T2DM and of 71% in T1DM (12). Nath et al. showed a 1-year pancreas survival of 94% in T2DM (16). In our study T2DM was not an independent risk factor for pancreas failure even after adjusting for other risk factors. Only the Light et al. study reported long-term pancreas

No drug No drug

steroid calcineurin inhibitor

1.26 (0.45 to 3.48) 1.49 (1.02 to 2.16) 1.81 (1.18 to 2.79) 0.41 (0.25 to 0.68) 0.71 (0.35 to 1.44) 0.37 (0.24 to 0.57)

1.11 (0.39 to 3.09) 1.58 (1.08 to 2.29) 1.67 (1.08 to 2.59) 0.53 (0.31 to 0.91) 0.81 (0.39 to 1.66) 0.42 (0.27 to 0.66)

2.48 (1.38 to 4.47) 0.38 (0.21 to 0.67) 0.52 (0.22 to 1.24) 0.43 (0.26 to 0.71) 2.10 (1.01 to 4.35) 0.88 (0.57 to 1.34) 0.60 (0.30 to 1.24) 1.39 (0.43 to 4.51) 1.18 (0.49 to 2.81)

2.13 (1.26 to 3.59) 2.58 (1.11 to 6.00) 0.66 (0.23 to 1.85) 0.37 (0.19 to 0.69) 0.41 (0.20 to 0.84)

Model 1

T1DM

1.08 (0.38 to 3.01) 1.69 (1.09 to 2.33) 1.71 (1.10 to 2.64) 0.59 (0.33 to 1.05) 0.81 (0.39 to 1.66) 0.42 (0.27 to 0.66)


2.16 (1.27 to 3.64) 2.60 (1.12 to 6.06) 0.65 (0.23 to 1.84) 0.36 (0.19 to 0.69) 0.44 (0.20 to 0.99)

0.94 (0.67 to 1.31) 0.96 (0.85 to 1.08) 1.27 (1.07 to 1.49) 0.92 (0.72 to 1.19) 0.54 (0.42 to 0.67) 0.44 (0.37 to 0.53)


1.29 (1.09 to 1.52) 1.59 (1.21 to 1.89) 0.47 (0.34 to 0.65) 0.36 (0.28 to 0.46) 0.84 (0.58 to 1.21)

Model 2a

Adjusted HR (95% CI)

Results were presented as HRs with 95% CI. CsA, cyclosporine A, Tac, tacrolimus; Alem, alemtuzumab; thym, thymoglobulin; Others, ALG/OKT3. a Model 2 included a covariate to account for interaction between use of induction and steroid.

18.5 to 25

No drug

⬍18.5 25 to 30 ⬎30 Yes CsA Tac


⬎1 year Yes CsA Tac Alem Thym IL2-RA Others Multiple

Pre-emptive No drug No drug

No drug

2.16 (1.28 to 3.62) 2.57 (1.11 to 5.97) 0.54 (0.19 to 1.51) 0.30 (0.17 to 0.53) 0.32 (0.16 to 0.62)

Unadjusted HR (95% CI)

⬎45 years ⬎1 year CsA Tac Yes

Levels

ⱕ45 years Pre-emptive No drug

Reference

Overall pancreas graft loss BMI

induction

steroid Overall kidney graft loss pretransplant dialysis steroid calcineurin inhibitor

Patient death recipient age pretransplant dialysis calcineurin inhibitor

Covariate

T2DM

Table 4. Estimated HR for independent factors associated with SPK transplant outcomes in type 2 diabetic recipients and comparison to HR of similar covariates in recipients with T1DM diagnosis

1204 Clinical Journal of the American Society of Nephrology


survival (10 years), with a pancreas survival of 49% and 56% in their T1DM and T2DM recipients, respectively. Additional long-term follow-up trials are needed to study pancreas allograft survival related to diabetes type. In our study, the two main covariates associated with increased pancreas failure were older donor age and higher BMI of the recipient. Risk for pancreas failure associated with BMI was particularly high in T2DM. We cannot examine why increased BMI was a stronger predictor of pancreas failure in T2DM patients compared with T1DM patients. However, we can hypothesize that in T2DM recipients increased BMI may be a marker of plurimetabolic syndrome, with insulin resistance, hyperinsulinemia, dyslipidemia, vascular disease, and increased prothrombotic activity that may contribute to pancreas loss (22–26). Regarding kidney survival, we found an inferior 5-year overall (77.8% versus 73.5%; P ⫽ 0.007) and deathcensored survival (85.3% versus 82.9%, P ⫽ 0.04) in T2DM when compared with T1DM. However, after adjusting for other potential risk factors, diabetes type was not confirmed as an independent risk factor for kidney graft failure. Risks for kidney and pancreas failure in overall insulin-dependent recipients were most importantly associated with AA race (versus Caucasian) and obesity (versus normal BMI). This was not a surprise because AA race was associated with increased risk of nonimmune and immune posttransplant complications (27), and obesity has been shown to be a potential risk for posttransplant complications in T1DM (28). In nontransplant surgery, obese patients were more likely to have increased surgical time and blood loss during the procedure (29,30). The former can lead to prolonged cold ischemia time and the latter to hypotensive episodes during or after the surgery. Both complications may predispose to delayed kidney graft failure or allograft thrombosis. In line with this, we found a higher rate of delayed kidney graft function and primary kidney nonfunction in the T2DM group, which has a higher percentage of overweight and obese recipients when compared with T1DM recipients. When comparing hazards for graft failure between T2DM and T1DM recipients, time on dialysis was found to be the main confounder responsible for adjusted risk for kidney allograft failure in T2DM. The main limitation of our study is the lack of specific information on the criteria used to define diabetes type. Diabetes diagnosis reported in the UNOS prepancreas transplant forms reflects current practice in U.S. pancreas transplant centers. It may include laboratorial parameters in association with clinical judgment; however, we could not validate the reported diagnosis. Diabetes type diagnosis in ESRD patients relies mainly on the clinical history (31), and therefore occasional misdiagnoses can be expected. The antibody anti-glutamic acid decarboxylase has been studied as a marker to differentiate T2DM from latent autoimmune diabetes in adults, and it has been shown that 2% to 12% of patients initially diagnosed as T2DM in fact had latent autoimmune diabetes (32). On the basis of this information, our diabetes classification may incur an incorrect diagnosis of T2DM not over 12%. Anti-glutamic acid decarboxylase may be used at diabetes onset; however, it has yet to be vali-


1205

dated as a reliable diagnosis tool several years after the diabetes onset, as is the case when ESRD diabetics are listed for a SPK transplant. C-peptide level as an isolated parameter is unreliable for diabetes diagnosis in ESRD patients (33,34). Others limitations to the study are that criteria for pancreas allograft failure are poorly described in UNOS. In general, pancreas graft failure is assumed when a recipient returns to exogenous insulin after a period of insulin independence. Finally, our study is limited by missing records and a potential selection bias of recipients that can be found in retrospective analyses. In conclusion, 9% of all SPK transplants between 2000 and 2007 in the United States were done in recipients diagnosed with T2DM. Compared with recipients diagnosed with T1DM, type 2 diabetic recipients had a higher proportion of associated comorbidities. However, after adjusting for other risk factors, T2DM was associated with similar risks for death and kidney or pancreas failure after a SPK transplant when compared with T1DM recipients. Acknowledgments This work was supported in part by Health Resources and Services Administration contract 234-2005-370011C. Dr. Marcelo S. Sampaio’s participation in this work was made possible through an International Society of Nephrology funded fellowship. The content is the responsibility of the authors alone and does not necessarily reflect the views of the policies of the Department of Health and Human Services, nor does mention of trade names, commercial products, or organizations imply endorsement by the U.S. government. Part of this material was presented as an oral presentation at the American Transplant Congress; May 1 to 5, 2010; San Diego, CA. Disclosures None.

References 1. Lipshutz GS, Wilkinson AH: Pancreas-kidney and pancreas transplantation for the treatment of diabetes mellitus. Endocrinol Metab Clin North Am 36: 1015–1038; x, 2007 2. Milde FK, Hart LK, Zehr PS: Pancreatic transplantation. Impact on the quality of life of diabetic renal transplant recipients. Diabetes Care 18: 93–95, 1995 3. Sureshkumar KK, Patel BM, Markatos A, Nghiem DD, Marcus RJ: Quality of life after organ transplantation in type 1 diabetics with end-stage renal disease. Clin Transplant 20: 19 –25, 2006 4. Dean PG, Kudva YC, Stegall MD: Long-term benefits of pancreas transplantation. Curr Opin Organ Transplant 13: 85–90, 2008 5. 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 6. Fioretto P, Sutherland DE, Najafian B, Mauer M: Remodeling of renal interstitial and tubular lesions in pancreas transplant recipients. Kidney Int 69: 907–912, 2006 7. Mohan P, Safi K, Little DM, Donohoe J, Conlon P, Walshe JJ, O’Kelly P, Thompson CJ, Hickey DP: Improved patient survival in recipients of simultaneous pancreas-kidney transplant compared with kidney transplant alone in patients with type 1 diabetes mellitus and end-stage renal disease. Br J Surg 90: 1137–1141, 2003 8. Morath C, Zeier M, Dohler 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

1206


9. Salvalaggio PR, Dzebisashvili N, Pinsky B, Schnitzler MA, Burroughs TE, Graff R, Axelrod DA, Brennan DC, Lentine KL: Incremental value of the pancreas allograft to the survival of simultaneous pancreas-kidney transplant recipients. Diabetes Care 32: 600 – 602, 2009 10. 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 11. Gruessner AC, Sutherland DE: Pancreas transplant outcomes for United States (US) and non-US cases as reported to the United Network for Organ Sharing (UNOS) and the International Pancreas Transplant Registry (IPTR) as of June 2004. Clin Transplant 19: 433– 455, 2005 12. 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 13. Light JA, Sasaki TM, Currier CB, Barhyte DY: Successful longterm kidney-pancreas transplants regardless of C-peptide status or race. Transplantation 71: 152–154, 2001 14. Sasaki TM, Gray RS, Ratner RE, Currier C, Aquino A, Barhyte DY, Light JA: Successful long-term kidney-pancreas transplants in diabetic patients with high C-peptide levels. Transplantation 65: 1510 –1512, 1998 15. 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 kidney-pancreas transplantation? Transplant Proc 40: 510 – 512, 2008 16. 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 17. Decker E, Coimbra C, Weekers L, Detry O, Honore P, Squifflet JP, Meurisse M, De Roover A: A retrospective monocenter review of simultaneous pancreas-kidney transplantation. Transplant Proc 41: 3389 –3392, 2009 18. Miguel A, Garcia-Ramon R, Perez-Contreras J, GomezRoldan C, Alvarino J, Escobedo J, Garcia H, Lanuza M, Lopez-Menchero R, Olivares J, Tornero F, Albero D: Comorbidity and mortality in peritoneal dialysis: A comparative study of type 1 and 2 diabetes versus nondiabetic patients. Peritoneal dialysis and diabetes. Nephron 90: 290 –296, 2002 19. Kervinen M, Lehto S, Ikaheimo R, Karhapaa P, GronhagenRiska C, Finne P: Improved survival of type 2 diabetic patients on renal replacement therapy in Finland. Nephrol Dial Transplant 25: 892– 896, 2010 20. Haapio M, Helve J, Groop PH, Gronhagen-Riska C, Finne P: Survival of patients with type 1 diabetes receiving renal replacement therapy in 1980 –2007. Diabetes Care 33: 1718 – 1723, 2010 21. Cavaghan MK, Ehrmann DA, Polonsky KS: Interactions between insulin resistance and insulin secretion in the development of glucose intolerance. J Clin Invest 106: 329 –333, 2000 22. Grant PJ: Diabetes mellitus as a prothrombotic condition. J Intern Med 262: 157–172, 2007 23. Meigs JB, Mittleman MA, Nathan DM, Tofler GH, Singer DE,

24.

25. 26.

27. 28.

29. 30.

31. 32. 33.

34.

Murphy-Sheehy PM, Lipinska I, D’Agostino RB, Wilson PW: Hyperinsulinemia, hyperglycemia, and impaired hemostasis: The Framingham Offspring Study. JAMA 283: 221–228, 2000 Rosito GA, D’Agostino RB, Massaro J, Lipinska I, Mittleman MA, Sutherland P, Wilson PW, Levy D, Muller JE, Tofler GH: Association between obesity and a prothrombotic state: The Framingham Offspring Study. Thromb Haemost 91: 683– 689, 2004 Fryirs M, Barter PJ, Rye KA: Cholesterol metabolism and pancreatic beta-cell function. Curr Opin Lipidol 20: 159 –164, 2009 Brunham LR, Kruit JK, Hayden MR, Verchere CB: Cholesterol in beta-cell dysfunction: The emerging connection between HDL cholesterol and type 2 diabetes. Curr Diab Rep 10: 55– 60, 2010 Young CJ, Gaston RS: Renal transplantation in black Americans. N Engl J Med 343: 1545–1552, 2000 Sampaio MS, Reddy PN, Kuo HT, Poommipanit N, Cho YW, Shah T, Bunnapradist S: Obesity was associated with inferior outcomes in simultaneous pancreas kidney transplant. Transplantation 89: 1117–1125, 2010 Lee CT, Dunn RL, Chen BT, Joshi DP, Sheffield J, Montie JE: Impact of body mass index on radical cystectomy. J Urol 172: 1281–1285, 2004 Williams TK, Rosato EL, Kennedy EP, Chojnacki KA, Andrel J, Hyslop T, Doria C, Sauter PK, Bloom J, Yeo CJ, Berger AC: Impact of obesity on perioperative morbidity and mortality after pancreaticoduodenectomy. J Am Coll Surg 208: 210 – 217, 2009 Sener A, Cooper M, Bartlett ST: Is there a role for pancreas transplantation in type 2 diabetes mellitus? Transplantation 90: 121–123, 2010 Falorni A, Brozzetti A: Diabetes-related antibodies in adult diabetic patients. Best Pract Res Clin Endocrinol Metab 19: 119 –133, 2005 Chakkera HA, Bodner JK, Heilman RL, Mulligan DC, Moss AA, Mekeel KL, Mazur MJ, Hamawi K, Ray RM, Beck GL, Reddy KS: Outcomes after simultaneous pancreas and kidney transplantation and the discriminative ability of the C-peptide measurement pretransplant among type 1 and type 2 diabetes mellitus. Transplant Proc 42: 2650 –2652, 2010 Covic AM, Schelling JR, Constantiner M, Iyengar SK, Sedor JR: Serum C-peptide concentrations poorly phenotype type 2 diabetic end-stage renal disease patients. Kidney Int 58: 1742–1750, 2000

Received: August 9, 2010 Accepted: January 1, 2011 M.S. and S.B. contributed to the design, performance of the research, data analysis, and writing of the paper. H.-T.K. contributed to design, performance of the research, and data analysis. Published online ahead of print. Publication date available at www.cjasn.org. See related editorial, “Simultaneous Pancreas-Kidney Transplants Are Appropriate in Insulin-Treated Candidates with Gremia Regardless of Diabetes Type,” on pages 957–959.