- Email: [email protected]
Impact of duration of diabetes on outcome following pancreas transplantation
International Journal of Surgery 18 (2015) 21e27
Contents lists available at ScienceDirect
International Journal of Surgery journal homepage: www.journal-surgery.net
Original research
Impact of duration of diabetes on outcome following pancreas transplantation Burcin Ekser a, Richard S. Mangus a, John A. Powelson a, Michele L. Goble a, Muhammad A. Mujtaba b, Tim E. Taber b, Jonathan A. Fridell a, * a b
The Department of Surgery, Indiana University School of Medicine, Indianapolis, IN, USA The Department of Medicine, Indiana University School of Medicine, Indianapolis, IN, USA
h i g h l i g h t s Longstanding T1DM does not seem to negatively impact recipient outcomes following all types of pancreas transplantation. Duration of diabetes exposure significantly correlates with the need for kidney transplantation. The majority of pancreas recipients with longer standing diabetes (21e30 and >30 years) were also undergoing kidney transplantation as SPK or PAK, likely because renal failure is a late complication of diabetes. PTA recipients, on the other hand, tended to make up a much larger portion of the group with the shortest duration of diabetes.
a r t i c l e i n f o
a b s t r a c t
Article history: Received 12 February 2015 Accepted 8 April 2015 Available online 11 April 2015
Introduction: The impact of duration of T1DM on outcomes following simultaneous pancreas and kidney transplantation (SPK), pancreas after kidney transplantation (PAK), and pancreas transplantation alone (PTA) is currently unknown. Materials and methods: A total of 451 pancreas transplants performed at a single institution between January 2003 and April 2013 (SPK n ¼ 238, PAK, n ¼ 97, and PTA, n ¼ 116) were divided into three groups based on cumulative years of T1DM (0e20 years, 21e30 years, and >30 years). Early (7-day) and late (90day) pancreas allograft loss, patient and pancreas allograft survivals were analyzed. Results: While, PAK was more common in recipients with >30 years of T1DM (29%, p < 0.0047), PTA was more common in recipients with 0e20 years of T1DM (41%, p < 0.0011). In all transplant types, recipients age was significantly higher the longer the duration of diabetes. Although longer duration of T1DM correlated with a higher rate of major amputations in PAK recipients (p < 0.0032), no difference was observed in SPK or PTA. While early pancreas graft loss was 2e4% in SPK and PAK with shorter or longer T1DM (p ¼ n.s.), it reached to 10% in PTA with T1DM > 30 years (p < 0.0097). Longer duration of T1DM affected late pancreas graft loss in PAK patients (8%, p < 0.0349). Patient and death-censored graft survival rates were similar in all types of pancreas transplantation extracted by accumulation of years of T1DM prior to transplant. Conclusions: Longstanding T1DM does not seem to negatively impact recipient outcomes following all types of pancreas transplantation. © 2015 Published by Elsevier Ltd on behalf of IJS Publishing Group Limited.
Keywords: Diabetes mellitus Duration of diabetes Pancreas transplantation Outcome
Abbreviations: CAD, coronary artery disease; ESRD, end-stage renal disease; MMF, mycophenolate mofetil; n.s., not significant; PAK, pancreas after kidney transplantation; PTA, pancreas transplantation alone; rATG, rabbit antithymocyte globulin; SPK, simultaneous pancreas and kidney transplantation; T1DM, type 1 diabetes mellitus; UNOS, United Network for Organ Sharing. * Corresponding author. The Department of Surgery, Indiana University School of Medicine, 550 N University BLVD, #4258, Indianapolis, IN 46202, USA. E-mail address: [email protected] (J.A. Fridell). http://dx.doi.org/10.1016/j.ijsu.2015.04.031 1743-9191/© 2015 Published by Elsevier Ltd on behalf of IJS Publishing Group Limited.
22
B. Ekser et al. / International Journal of Surgery 18 (2015) 21e27
1. Introduction Recent analyses of national U.S. data from 1980 to 2012 suggest a doubling of the incidence and prevalence of diabetes during 1990e2008, and a plateauing between 2008 and 2012 [1]. An estimated 10e15% of the US population is affected by diabetes and 8e10% of these have type 1 diabetes mellitus (T1DM) [2]. T1DM is associated with severe late complications, including end-stage renal disease (ESRD), requiring renal replacement therapy or kidney transplantation [3]. Diabetes is also an independent major risk factor for cardiovascular and cerebrovascular disease, which is dependent on the years of exposure. Studies have shown that during the first 20 years of T1DM most of the excess mortality is attributed to renal failure, however after this period, it is considered to result largely from cardiovascular events [3e5]. International studies have revealed different mortality rates depending on the duration of diabetes exposure. The Finnish study demonstrated a cumulative mortality of 6.8% at 20 years and 15% at 30 years after diabetes diagnosis with a cumulative incidence of ESRD of 2.2.and 7.8%, respectively [6]. The impact of duration of diabetes on mortality has been reported by several groups, with rates varying between 6.8 and 13% after 20 years, and 15e29% after 30 years [3,6e9]. Pancreas transplantation is most commonly performed in the context of renal transplantation for end stage diabetic nephropathy (simultaneous pancreas and kidney transplantation (SPK) or pancreas after kidney (PAK) transplantation), or as an isolated transplant for poor glycemic control, particularly hypoglycemia unawareness (pancreas transplantation alone (PTA)), and is the only definitive long-term therapy which restores the euglycemic state and prevents and reverses secondary complications of diabetes [2,10]. The impact of recipient age, which can correlate with the duration of T1DM exposure, has been investigated in pancreas transplantation [11e13]. Schenker et al. and Ablorsu et al. have demonstrated that it is possible to achieve similar graft and patient survival rates in pancreas transplant recipients >50 years old compared to younger recipients [11,13]. Our group has recently reported on the impact of age on outcomes after pancreas transplantation. In that study, we also demonstrated that older recipients (50e59 years old, n ¼ 85 and >60 years old, n ¼ 18) have similar patient and graft survival rates compared to younger recipients [14]. Interestingly, in that study, the worst pancreas allograft survival was observed in the youngest recipients (<30 years of age). Although some studies have reported single center experiences of pancreas transplantation in the diabetic population in large cohorts [15,16] no studies to date have systemically examined the impact of duration of T1DM on outcomes following pancreas transplantation. The current study was designed to determine the impact of the exact duration of diabetes exposure on different types of pancreas transplant recipients.
included all patients undergoing pancreas transplantation (SPK, PAK, or PTA). Pancreas retransplantations, even if performed early after the first transplant, were included in the data analysis. Measurements of recipient pre-transplant HbA1c and cardiac ejection fraction, history of hypertension, smoking, and major amputation were included. Although it was limited, non-heart beating donors were also used and included in data analysis together with recipient demographics and risk factors to stratify the different risk levels in different types of pancreas transplantation. Table 1 and supplementary digital content (SDC) Tables 1s, 2s, and 3s show demographic data for all pancreas transplantation, SPK, PAK, and PTA by accumulated years of T1DM prior to transplant, respectively. Recipient listing for the transplant was according to standard procedures and protocols as established by the United Network for Organ Sharing (UNOS). All recipients were confirmed to be c-peptide negative prior to transplant. Regardless of type of pancreas transplantation (SPK, PAK, or PTA), the work-up for pancreas transplantation was similar in all recipients. All patients were required by the listing committee to have a negative cardiac stress test prior to pancreas transplantation. Criteria for cardiac catheterization included any patient with known history of coronary artery disease (CAD), multiple risk factors, or a positive finding on a cardiac stress test, as explained before [17]. 2.2. Pancreas procurement, preparation and transplantation Local pancreas allografts were typically procured using an en bloc technique following aortic flush with preservation solution and topical cooling with saline flush and ice packing, as previously described [18]. All pancreas allografts were prepared, as previously described [19]. We have previously shown that there was no difference in outcomes for local and import pancreas allografts in our experience [20]. Therefore, the data analyzed included those recipients of imported pancreas allografts as well. The recipient operation was performed through a midline incision, as previously described [21,22]. Briefly, the pancreas allograft was positioned with the tail toward the pelvis and the head and duodenum oriented superiorly to facilitate enteric exocrine drainage. All pancreas allografts, regardless of SPK, PAK or PTA, were drained enterically using a stapled technique [23]. Systemic venous drainage was performed to the vena cava or to the right common iliac vein. Arterial perfusion of the allograft was routinely established from the right common iliac artery, although on rare occasions where this vessel was found to be diseased or had been the site for arterial anastomosis for a prior transplant, the inflow would be established either from the aorta or the left common iliac artery. All SPK transplants were performed with ipsilateral placement of both the kidney and the pancreas to the right iliac vessels, as previously described [24]. Pulsatile perfusion was used routinely for the renal allograft portion of the SPK, as described [25]. 2.3. Immunosuppressive therapy
2. Materials and methods 2.1. Data collection and inclusion criteria The medical records for all deceased donor pancreas transplants performed at Indiana University between January 2003 and April 2013 were reviewed (n ¼ 451). Retrospective review of data from the transplant center database was approved by the Institutional Review Board of Indiana University School of Medicine. To obtain the data presented in the present manuscript, the comprehensive transplant recipient registry at our center, individual written and electronic medical records, and the original donor medical history were carefully reviewed. Inclusion criteria for the data analysis
The induction immunosuppressive regimen consisted of five doses of rabbit antithymocyte globulin (rATG) (1 mg/kg/dose). A single dose of rituximab (150 mg/m2) induction was also included in cases of PTA. The maintenance immunosuppressive regimen consisted of tacrolimus (through level of 8e10 ng/mL), and sirolimus (through level of 3e6 ng/mL) [26]. Steroids were only used as a premedication for rATG induction and were discontinued in all recipients, including PAK recipients receiving long-term steroids for a remote renal transplant [27]. Mycophenolate mofetil (MMF) (500 mg po b.i.d) was used as a part of the maintenance immunosuppressive regimen together with tacrolimus and sirolimus in cases of PTA or as a substitute for sirolimus in cases of drug
B. Ekser et al. / International Journal of Surgery 18 (2015) 21e27
23
Table 1 Demographic data for all pancreas transplantations by accumulated years of Type I diabetes prior to transplant.
Overall Duration of DM Recipient
(years) (mean ± SD) (range) Transplant type SPK PAK PTA Age (years) (mean ± SD) (range) Gender Male Female Race White African American Other Blood type A B AB O Body mass index (mean ± SD) (range) HbA1c (mean ± SD) Cardiac ejection fraction (mean ± SD) History of hypertension History of smoking History of major amputations
Total#
0 to 20 years
21 to 30 years
>30 years
p
451 (100%) 28.5 ± 10.1 (1e54)
111 (25%) 14.8 ± 5.0 (1e20)
157 (35%) 26.0 ± 4.0 (21e30)
183 (41%) 37.7 ± 5.4 (31e54)
<0.0001
53% 22% 26% 43.3 ± 9.9 (13e68)
47% 12% 41% 36.8 ± 10.8 (13e68)
58% 20% 22% 41.5 ± 8.4 (26e64)
52% 29% 19% 48.9 ± 7.3 (32e66)
n.s. <0.0047 <0.0011 <0.0001
60% 40%
50% 50%
62% 38%
65% 35%
<0.05 <0.05
93% 6% 1%
88% 10% 2%
93% 6% 1%
95% 5% 0%
n.s. n.s. n.s.
42% 8% 5% 45% 26 ± 4 (17e44) 8.0 ± 1.0 60 ± 10% 77% 31% 3%
36% 8% 6% 50% 25 ± 4 (17e38) 8.0 ± 0.9 60 ± 9% 72% 30% 1%
46% 7% 2% 45% 26 ± 7 (18e44) 8.1 ± 1.3 60 ± 11% 78% 36% 2%
42% 10% 7% 41% 26 ± 4 (18e39) 7.9 ± 1.2 61 ± 9% 79% 27% 6%
n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s.
26.8 ± 11.1 (4e54)
26.2 ± 11.3 (8e54)
26.9 ± 10.7 (6e51)
27.1 ± 11.4 (4e53)
n.s.
64% 36%
65% 35%
61% 39%
67% 33%
n.s. n.s.
79% 16% 5% 24 ± 5 (13e59)
77% 8% 7% 25 ± 7 (14e59)
87% 10% 3% 24 ± 4 (14e38)
74% 12% 14% 24 ± 5 (13e40)
<0.02 n.s. <0.01 n.s.
20% 59% 22% 62% 38% 1% 8.4 ± 2.5 (3.5e20.8)
23% 58% 19% 60% 40% 1% 8.7 ± 2.7 (4.5e20.8)
17% 60% 23% 64% 36% 0% 8.5 ± 2.5 (3.5e17.0)
19% 59% 22% 62% 38% 0% 8.2 ± 2.4 (4.4e18.0)
n.s. n.s. n.s. n.s. n.s. n.s. n.s.
Donor Age (years) (mean ± SD) (range) Gender Male Female Race White African American Other Body mass index (mean ± SD) (range) Cause of death Stroke Trauma Anoxia/other Procured by IU transplant team Shipped-in pancreas graft Non-heart beating donor Cold ischemia time (mean ± SD) (range) Statistically significant values are written bold.
nephrotoxicity or wound complications in SPK or PAK recipients. All patients received prophylaxis for cytomegalovirus with oral valganciclovir and Pneumocyctis jiroveci pneumonia with trimethoprim/sulfamethaxazole. 2.4. Outcomes and statistical analyses
PAK (22%), and 116 PTA (26%) (Fig. 1). For each type of pancreas transplantation, patients were divided into 3 subgroups by accumulated years of duration of diabetes exposure prior to transplant: (i) 0e20 years (n ¼ 111), (ii) 21e30 years (n ¼ 157), and (iii) > 30 years (range 31e56 years) (n ¼ 183). SPK represented approximately half of the cases regardless of duration of T1DM (Table 1).
Retrospective primary transplant outcomes included 7-day (early) and 90-day (late) pancreas transplant loss, and one-year pancreas allograft and patient survival. Secondary transplant outcomes included 5-year and 10-year pancreas allograft and patient survival. Demographic data and outcomes were compared between the three groups based on duration of diabetes, (i) 0e20 years, (ii) 21e30 years, and (iii) > 30 years, for each type of pancreas transplantation. Fisher's exact, standard chi-square, student t-test, and analysis of variance (ANOVA), log-rank (ManteleCox for survival curves) tests were used in appropriate calculations using GraphPad Prism 6 for Mac OS X (La Jolla, CA, USA). 3. Results Of 451 pancreas transplantations, there were 238 SPK (53%), 97
Fig. 1. Study design and exact number of patients in different types of pancreas transplantation by accumulated years of Type I diabetes prior to transplant.
24
B. Ekser et al. / International Journal of Surgery 18 (2015) 21e27
Although PTA was performed more than PAK and SPK (41%, p < 0.0011) in recipients with duration of T1DM 0e20 years, PAK was performed more (29%, p < 0.0047) in recipients with duration of T1DM > 30 years (Table 1). Data analyses were done comparing each subgroup for different types of pancreas transplantation. Fig. 1 shows the study design and exact numbers of recipients in each subgroup per type of pancreas transplantation. The mean duration of T1DM was not significant between different types of pancreas transplants for given subgroups; 15.9 ± 4.0 vs 15.8 ± 4.2 vs 12.0 ± 5.2 (p ¼ n.s.) in SPK, PAK, PTA, respectively for T1DM 0e20 years; 25.6 ± 3.1 vs 26.8 ± 2.7 vs 25.6 ± 3.0 (p ¼ n.s.) in SPK, PAK, PTA, respectively for T1DM 21e30 years, and 37.7 ± 5.4 vs 38.1 ± 5.8 vs 37.1 ± 4.6 (p ¼ n.s.) in SPK, PAK, PTA, respectively for T1DM > 31 years. In all types of transplant, mean age of recipients in subgroups was significantly higher and mirrored the duration of diabetes (SDC Tables 1s, 2s, and 3s). While >30 years of T1DM correlated with higher BMI in SPK patients (p < 0.01), no difference was observed in PAK, and PTA. In SPK, recipient and donor demographics were similar between different years of diabetes duration, except those mentioned above and the race of the donor, as shown in SDC Table 1s. In PAK, homogeneity between subgroups was not consistent, including race of recipients and donors, recipient blood type, and history of major amputations prior to transplant, donor cause of death, procurement by local team or shipped-in pancreas allografts (SDC Table 2s). The impact of duration of T1DM can be seen in recipients of PAK >30 years of T1DM with significantly higher major amputations rate (9%, p < 0.0032). In PTA, recipient and donor demographics were mainly comparable, except gender and race of the recipient, and history of smoking which was significantly lower in patients with >30 years of T1DM (p < 0.003) (SDC Table 3s). Median cold ischemia time (~8 h) and percentage of utilization of non-heart beating donor were similar in all 3 subgroups in all types of pancreas transplantation. Although the percentage of Indiana University (IU) Transplant Team procured pancreas grafts similar (52% vs 61%, p ¼ n.s.) in PAK and PTA, the percentage of IU Team procurement reached to 91% (p < 0.003) in SPK (SDC Tables 1s, 2s, and 3s).
While early (7-day) pancreas graft loss was similar among different duration of diabetes exposure in SPK and PAK transplant recipients (0e4%), it reached to 10% in PTA recipients with a history of >30 years of T1DM (Table 2). Late (90-day) pancreas graft loss was significantly higher only in PAK recipients with more than >20 years of diabetes (p < 0.0349). As expected, serum creatinine was lower in PTA recipients than SPK and PAK transplant recipients, whom had reflected higher glomerular filtration rate (GFR) 73 ± 30 vs 69 ± 33 vs 81 ± 43 in SPK, PAK, PTA, respectively (p < 0.01). Longer duration of diabetes exposure (>30 years) had a significant negative impact on GFR in all types of pancreas transplantation, regardless of the need for a kidney transplant or not (Table 2). There were a total of 86 (19%) pancreas graft losses in 451 patients over 10 years. Of 86 pancreas graft losses, 46 (54%) were due to patient death with a functioning allograft. Although longer duration of T1DM was correlated with higher rate of pancreas graft loss in recipients requiring kidney transplantation (19% in SPK and 23% in PAK in patients with >30 years of T1DM), it was contrary in PTA (35% pancreas graft loss in patients with 0e20 years of T1DM) (Table 3). Higher percentage of pancreas loss in PTA patients with 0e20 years of diabetes (p < 0.0009) was due to increased rate of chronic rejection (15%, p < 0.0003). Patient deaths were not significantly different among the causes in all types of pancreas transplantation, regardless of the duration of T1DM exposure, except the total percentage of death (19%) in PAK patients with >30 years of exposure (p < 0.037) (Table 3). Patient survival was found to be significantly reduced in recipients of pancreas transplantation (including all types) with T1DM > 30 years of duration (Fig. 2). However, longer duration of T1DM did not significantly affect pancreas allograft survival (with or without death-censorship) (Fig. 2). Despite differences in pancreas graft loss and patient death in different types of pancreas transplantation, KaplaneMeier survival curves did not show a significant difference in patient and death-censored pancreas graft survivals (SDC Fig. 1s). The only difference was observed in SPK patients with 21e30 years of T1DM in non-death-censored pancreas graft survival (p < 0.05). As expected, patient survival was higher in PTA recipients than SPK
Table 2 Post-transplant outcomes in different type of pancreas transplants by accumulated years of Type I diabetes prior to transplant.
Simultaneous pancreas-kidney transplantation (SPK), n (%) Pancreas graft loss within 7 days Pancreas graft loss within 90 days Hospital stay at the transplant (days) (mean ± SD) (median) Creatinine 1-year (mg/dL) (mean ± SD) GFR 1-year (mL/min/1.73 m2) (mean ± SD) HbA1c 1-year (%) (mean ± SD) C-peptide 1-year (ng/mL) (mean ± SD) Pancreas after kidney transplantation (PAK), n (%) Pancreas graft loss within 7 days Pancreas graft loss within 90 days Hospital stay at the transplant (days) (mean ± SD) (median) Creatinine 1-year (mg/dL) (mean ± SD) GFR 1-year (mL/min/1.73 m2) (mean ± SD) HbA1c 1-year (%) (mean ± SD) C-peptide 1-year (ng/mL) (mean ± SD) Pancreas transplantation alone (PTA), n (%) Pancreas graft loss within 7 days Pancreas graft loss within 90 days Hospital stay at the transplant (days) (mean ± SD) (median) Creatinine 1-year (mg/dL) (mean ± SD) GFR 1-year (mL/min/1.73 m2) (mean ± SD) HbA1c 1-year (%) (mean ± SD) C-peptide 1-year (ng/mL) (mean ± SD) Statistically significant values are written bold. a Median value is used for statistical calculation.
Total#
0 to 20 years
21 to 30 years
>30 years
238 (100%) 3% 6% 10.4 ± 12.1 (8) 1.4 ± 0.8 73 ± 30 5.5 ± 0.8 2.8 ± 2.1 97 (100%) 2% 6% 15.5 ± 46.9 (7) 1.4 ± 0.5 69 ± 33 5.5 ± 0.8 2.7 ± 1.3 116 (100%) 5% 7% 14.3 ± 42 (7) 1.2 ± 0.8 81 ± 43 5.6 ± 0.9 2.4 ± 1.7
52 (22%) 4% 6% 9.0 ± 5.7 (8) 1.5 ± 1.2 71 ± 31 5.5 ± 0.5 2.9 ± 1.7 13 (13%) 0% 0% 11.7 ± 13.7 (7) 1.4 ± 0.4 72 ± 22 5.7 ± 1.5 2.9 ± 1.7 46 (40%) 7% 9% 11.3 ± 13.0 (7) 1.3 ± 1.0 91 ± 50 5.7 ± 1.0 2.6 ± 1.9
91 (38%) 2% 3% 10.0 ± 8.3 (7.5) 1.3 ± 0.5 79 ± 29 5.5 ± 1.0 2.5 ± 1.6 31 (32%) 3% 7% 27.0 ± 85.1 (6.5) 1.4 ± 0.4 78 ± 36 5.6 ± 0.5 2.5 ± 1.3 35 (30%) 0% 3% 7.3 ± 2.1 (7) 1.3 ± 0.5 81 ± 37 5.4 ± 0.5 2.7 ± 2.0
95 (40%) 4% 8% 11.7 ± 17.0 (8) 1.4 ± 0.8 67 ± 29 5.5 ± 0.8 3.1 ± 2.8 53 (55%) 2% 8% 10.7 ± 10.1(7.5) 1.3 ± 0.6 63 ± 32 5.4 ± 0.6 2.7 ± 1.3 35 (30%) 10% 10% 25.9 ± 74.2 (7.5) 1.1 ± 0.5 68 ± 38 5.6 ± 1.0 1.7 ± 0.9
p n.s. n.s. n.s.a n.s. <0.0053 n.s. n.s. n.s. <0.0349 n.s.a n.s. <0.0314 n.s. n.s. <0.0097 n.s. n.s.a n.s. <0.0261 n.s. <0.0088
Table 3 Causes of pancreas graft loss and patient death in different type of pancreas transplants by accumulated years of Type I diabetes prior to transplant. Simultaneous pancreas-kidney (SPK) transplants
Pancreas transplants alone (PTA)
0 to 20 years 21 to 30 years >30 years p
Total#
0 to 20 years 21 to 30 years >30 years p
238 (100%) 36 (15%) 7 (3%) 6 (3%) 21 (9%) 1 (0.5%)
52 (22%) 10 (19%) 2 (4%) 1 (2%) 7 (14%) 0 (0%)
91 (38%) 8 (9%) 1 (1%) 2 (2%) 5 (5%) 0 (0%)
95 (40%) 18 (19%) 4 (4%) 3 (3%) 9 (10%) 1 (1%)
n.s. n.s. n.s. n.s. n.s.
97 (100%) 23 (24%) 3 (3%) 3 (3%) 13 (13%) 1 (1%)
13 (13%) 2 (16%) 0 (0%) 0 (0%) 1 (8%) 0 (0%)
31 (32%) 9 (29%) 2 (6%) 1 (3%) 4 (13%) 1 (3%)
53 (55%) 12 (23%) 1 (2%) 2 (4%) 8 (15%) 0 (0%)
116 (100%) 0.0414 27 (23%) n.s. 3 (3%) n.s. 8 (7%) n.s. 12 (10%) n.s. 4 (3%)
46 (40%) 16 (35%) 1 (2%) 7 (15%) 6 (13%) 2 (4%)
35 (30%) 6 (17%) 0 (0%) 0 (0%) 4 (11%) 2 (6%)
35 (30%) 5 (14%) 2 (6%) 1 (3%) 2 (6%) 0 (0%)
0.0009 n.s. 0.0003 n.s. n.s.
1 (0.5%) 24 (10%) 2 (1%) 5 (2%) 2 (1%) 1 (0.5%) 2 (1%) 1 (0.5%) 7 (3%) 4 (2%)
0 8 1 0 1 1 1 0 2 2
0 5 0 1 0 0 0 0 3 1
1 (1%) 11 (12%) 1 (1%) 4 (4%) 1 (1%) 0 (0%) 1 (1%) 1 (1%) 2 (2%) 1 (1%)
n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s.
3 (3%) 16 (17%) 0 (0%) 2 (2%) 1 (1%) 1 (1%) 2 (2%) 4 (4%) 0 (0%) 6 (6%)
1 1 0 0 0 0 0 0 0 1
1 5 0 1 1 1 0 0 0 2
1 (2%) 10 (19%) 0 (0%) 1 (2%) 0 (0%) 0 (0%) 2 (4%) 4 (8%) 0 (0%) 3 (6%)
0.0185 0.037 n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s.
0 2 0 1 0 0 0 0 0 1
0 1 0 0 0 0 0 0 1 0
0 3 1 0 0 0 1 0 0 1
n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s.
(0%) (15%) (2%) (0%) (2%) (2%) (2%) (0%) (4%) (4%)
(0%) (6%) (0%) (1%) (0%) (0%) (0%) (0%) (3%) (1%)
(8%) (8%) (0%) (0%) (0%) (0%) (0%) (0%) (0%) (8%)
(3%) (16%) (0%) (3%) (3%) (3%) (0%) (0%) (0%) (6%)
Total#
0 6 1 1 0 0 1 0 1 2
(0%) (5%) (1%) (1%) (0%) (0%) (1%) (0%) (1%) (2%)
0 to 20 years 21 to 30 years >30 years p
(0%) (4%) (0%) (2%) (0%) (0%) (0%) (0%) (0%) (2%)
(0%) (3%) (0%) (0%) (0%) (0%) (0%) (0%) (3%) (0%)
(0%) (9%) (3%) (0%) (0%) (0%) (3%) (0%) (0%) (3%)
Statistically significant values are written bold.
25
and PAK transplant recipients throughout the study (SDC Fig. 1s).
4. Discussion
In the USA, diabetes mellitus is the most common cause of ESRD, blindness, major limb amputations, and peripheral and coronary
Fig. 2. KaplaneMeier patient and graft (including death-censored graft) survival postpancreas transplant in all pancreas transplants by years of diabetes.
B. Ekser et al. / International Journal of Surgery 18 (2015) 21e27
Overall transplants (n, %) Cause of pancreas graft loss (n, %) Vascular thrombosis Chronic rejection Patient death with function graft Pseudoaneurysm or arterioenteric fistula Other Cause of death (n, %) Stroke/neurologic event Cardiac etiology Post-transplant lymphoma disease Graft versus host disease Infection/sepsis Cancer Medical/surgical complications Other cause/unknown
Pancreas after kidney (PAK) transplants
Total#
26
B. Ekser et al. / International Journal of Surgery 18 (2015) 21e27
vascular complications, although its incidence and prevalence has plateaued in the last decade [1,2]. A steady increase in the number and age of T1DM patients implies that the pancreas transplant professional is going to encounter an increasing number of potential pancreas transplant candidates with decades of exposure to this disease [1,2]. In fact, among the three study groups in the present study, patients with diabetes duration >30 years represented the largest cohort (n ¼ 183) (Fig. 1). Jorgensen et al. recently showed that the absolute and relative mortality rates in patients with T1DM have decreased over the last 10 years confirming that the high mortality rate in T1DM patients is largely related to longterm diabetes complications [28]. Despite the amelioration in diabetes care, longer duration of diabetes is still associated with considerable incidence of morbidity and mortality [3,6e9]. The duration of T1DM is also an important factor for long-term complications, such as vascular diseases, including CAD and cerebrovascular disease, diabetic nephropathy, and neuropathy. Among the risk factors associated with longer duration of diabetes, CAD is one of the most important since it mainly determines the patient's longevity. Recently, our group investigated the impact of CAD on the outcome following pancreas transplantation [17]. We found that patients with a history of pretransplant cardiac interventions had worse clinical outcomes, including an 11-fold increased risk for myocardial infarction and 3fold increased risk for stroke after pancreas transplantation. Not surprisingly, the patients that had undergone prior cardiac interventions were also significantly older than those that did not [17]. There was no significant difference in patient and deathcensored pancreas graft survival in all types of transplants between subgroups of accumulated diabetes exposure prior to transplant. Five-year and 10-year patient survival was higher in PTA patients, regardless of duration of T1DM (91e97%) in comparison to other types of transplants. The negative impact of longer duration of T1DM was observed in recipients of PAK transplants at 10-year follow-up (89%, 65%, and 76% in 0e20 years, 21e30 years, and >30 years T1DM, respectively). The lowest pancreas survival was observed in PTA in 0e20 years exposed T1DM at 10-year follow up (61%). Although recipients of PTA received additional immunosuppressive treatment with a single dose rituximab induction and sirolimus maintenance with tacrolimus, chronic rejection rate was still the highest in this group (7% vs 3%). In PTA, patients with longer duration of diabetes had lower rejection rates. It is possible that older patients may have an immunological advantage posttransplantation, although it remains unclear whether these recipients provide a more favorable immunological environment because of better compliance or a less active immune system. Richardson et al. recently investigated the levels of antibodies to the diabetes-associated islet autoantigens, such as GAD, IA-2, and zinc transporter-8 (ZnT8) in patients with long-duration T1DM, and the influence of age at diagnosis and HLA genotype on autoantibody persistence [29]. They demonstrated that all autoantibodies against GAD, IA-2 and ZnT8 were significantly decreased in patients with long-duration diabetes but there was also a remarkable persistence of anti-GAD antibodies in long-term survivors of T1DM. The decreased titers of autoantibodies were explained by the weak expression of diabetes-associated islet autoantigens due to decreased number of surviving or regenerating islets in the diabetic pancreas. This could potentially explain the favorable immunological state of older recipients together with their high compliance for organ transplantation. Management of pancreas transplant recipients has also improved significantly over the last two decades. Earlier studies cautioned that one-year pancreas graft survival in patients over 49
years of age was approximately 50% and older age (>49) was considered a relative contraindication to pancreas transplantation [12,30]. However, in the present study, the longest diabetes exposure groups (>30 years) have a mean age of recipients of 48e49 years in all types of pancreas transplants. Our group also previously showed that recipients of pancreas transplantation between 50 and 59 years of age (n ¼ 85) and >60 years of age (n ¼ 18) had 95% and 93% 1-year patient survival, respectively [14]. In the present study, we observed that duration of diabetes exposure significantly correlates with the need for kidney transplantation. The majority of pancreas recipients with longer standing diabetes (21e30 and >30 years) were also undergoing kidney transplantation as SPK or PAK, likely because renal failure is a late complication of diabetes. PTA recipients, on the other hand, tended to make up a much larger portion of the group with the shortest duration of diabetes (40% vs 13% and 22% for PTA vs PAK and SPK, respectively, p < 0.001). In conclusion, although the patients with longstanding T1DM requires increased rate of kidney transplantation, their outcomes following all types of pancreas transplantation are not negatively impacted. Ethical approval Local hospital IRB was obtained. Sources of funding None. Author contribution B.E and R.S.M participated in data collection and analysis, and performance of research. B.E and J.A.F participated in writing of the article. R.S.M and J.A.F participated in research design and performance of research. J.A.P and J.A.F performed most of the surgeries. J.A.P, J.A.F, M.L.G, M.A.M, T.E.T participated in follow up of patients and data collection. All authors participated in critical review of the manuscript. Conflicts of interest Authors declare no conflict of interests. Guarantor Burcin Ekser, MD. Research registration unique identifying number (UIN) researchregistry15. Acknowledgments Part of this paper was presented in the 14th World Congress of the International Pancreas and Islet Transplant Association (IPITA 2013). Appendix A. Supplementary data Supplementary data related to this article can be found at http:// dx.doi.org/10.1016/j.ijsu.2015.04.031. References [1] L.S. Geiss, J. Wang, Y.J. Cheng, et al., Prevalence and incidence trends for
B. Ekser et al. / International Journal of Surgery 18 (2015) 21e27
[2] [3]
[4]
[5]
[6] [7]
[8]
[9]
[10]
[11]
[12]
[13]
[14] [15]
diagnosed diabetes among adults aged 20 to 79 years, United States, 1980e2012, JAMA 312 (2014) 1218e1226. R.W.G. Gruessner, A.C. Gruessner, The current status of pancreas transplantation, Nat. Rev. Endocrinol. 9 (2013) 555e562. M. Stadler, M. Auinger, C. Anderwald, et al., Long-term mortality and incidence of renal dialysis and transplantation in type 1 diabetes mellitus, J. Clin. Endocrinol. Metab. 91 (2006) 3814e3820. K. Asao, C. Sarti, T. Forsen, et al., Long-term mortality in nationwide cohorts of childhood-onset type 1 diabetes in Japan and Finland, Diabetes. Care 26 (2003) 2037e2042. J.S. Dorman, R.E. Laporte, L.H. Kuller, et al., The Pittsburgh insulin-dependent diabetes mellitus (IDDM) morbidity and mortality study. Mortality results, Diabetes 33 (1984) 271e276. P. Finne, A. Reunanen, S. Stenman, et al., Incidence of end-stage renal disease in patients with type 1 diabetes, JAMA 294 (2005) 1782e1787. I. Muhlhauser, H. Overmann, R. Bender, et al., Predictors of mortality and endstage diabetic complications in patients with type 1 diabetes mellitus on intensified insulin therapy, Diabet. Med. 17 (2000) 727e734. G. Pambianco, T. Costacou, D. Ellis, et al., The 30-year natural history of type 1 diabetes complications: the Pittsburgh epidemiology of diabetes complications study experience, Diabetes 55 (2006) 1463e1469. P. Rossing, P. Hougaard, K. Borch-Johnsen, et al., Predictors of mortality in insulin dependent diabetes: 10 year observational follow up study, BMJ 313 (1996) 779e784. P. Fioretto, M.W. Steffes, D.E.R. Sutherland, et al., Reversal of lesions of diabetic nephropathy after pancreas transplantation, N. Engl. J. Med. 339 (1998) 69e75. E. Ablorsu, A. Ghazanfar, S. Mehra, et al., Outcome of pancreas transplantation in recipients older than 50 years: a single-centre experience, Transplantation 86 (2008) 1511e1514. C.E. Freise, P.G. Stock, J.S. Melzer, Increased morbidity and mortality of simultaneous pancreas-renal transplantation in patients over 49 years of age, Transpl. Proc. 30 (1998) 292. P. Schenker, O. Vonend, B. Kruger, et al., Long-term results of pancreas transplantation in patients older than 50 years, Transpl. Int. 24 (2011) 136e142. A.P. Shah, R.S. Mangus, J.A. Powelson, et al., Impact of recipient age on whole organ pancreas transplantation, Clin. Transpl. 27 (2013) E49eE55. R. Ollinger, C. Margreiter, C. Bosmuller, et al., Evolution of pancreas transplantation. Long term results and perspectives from a high-volume center, Ann. Surg. 256 (2012) 780e787.
27
[16] H.W. Sollinger, J.S. Odorico, S.J. Knechtle, et al., Experience with 500 simultaneous pancreas-kidney transplants, Ann. Surg. 228 (1998) 284e296. [17] R.S. Mangus, J. Powelson, S.B. Kinsella, et al., Pretransplant coronary artery disease associated with worse clinical outcomes in pancreas transplantation, Clin. Transpl. 27 (2013) E442eE447. [18] J.A. Fridell, R.S. Mangus, J.A. Powelson, Histidine-tryptophan-ketoglutarate for pancreas allograft preservation: the Indiana university experience, Am. J. Transpl. 10 (2010) 1284e1289. [19] J.A. Fridell, J.A. Powelson, C.E. Sanders, et al., Preparation of the pancreas allograft for transplantation, Clin. Transpl. 25 (2011) E103eE112. [20] J.A. Fridell, R.S. Mangus, E.F. Hollinger, et al., No difference in transplant outcomes for local and import pancreas allografts, Transplantation 88 (2009) 723e728. [21] J.A. Fridell, A. Agarwal, J.A. Powelson, Pancreas transplantation at Indiana university: a brief overview of recent progress, in: R. Corry, R. Shapiro (Eds.), Pancreatic Transplantation: Informa Healthcare, USA, 2007, pp. 421e427. [22] J.A. Fridell, R.S. Mangus, J.A. Powelson, et al., Pancreas transplantation in the new millennium: the Indiana university experience, Clin. Transpl. (2011) 145e156. [23] J.A. Fridell, M.L. Milgrom, S. Henson, et al., Use of the end-to-end anastomotic circular stapler for creation of the duodenoenterostomy for enteric drainage of the pancreas allograft, J. Am. Coll. Surg. 198 (2004) 495e497. [24] J.A. Fridell, A. Shah, M.L. Milgrom, et al., Ipsilateral placement of simultaneous pancreas and kidney allografts, Transplantation 78 (2004) 1074e1076. [25] A.P. Shah, D.P. Milgrom, R.S. Mangus, et al., Comparison of pulsatile perfusion and cold storage for paired kidney allografts, Transplantation 86 (2008) 1006e1009. [26] J.A. Fridell, R.S. Mangus, E.F. Hollinger, et al., The case for pancreas after kidney transplantation, Clin. Transplant 23 (2009) 447e453. [27] J.A. Fridell, A. Agarwal, J.A. Powelson, et al., Steroid withdrawal for pancreas after kidney transplantation in recipients on maintenance prednisone immunosuppression, Transplantation 82 (2006) 389e392. [28] M.E. Jorgensen, T.P. Almdal, B. Carstensen, Time trends in mortality rates in type 1 diabetes from 2002 to 2011, Diabetologia 56 (2013) 2401e2404. [29] C.C. Richardson, J.A. Dromey, K.A. McLaughlin, et al., High frequency of autoantibodies in patients with long duration type 1 diabetes, Diabetologia 56 (2013) 2538e2540. [30] C.L. Manske, Y. Wang, W. Thomas, Mortality of cadaveric kidney transplantation versus combined kidney-pancreas transplantation in diabetic patients, Lancet 346 (1995) 1658e1662.
Contents lists available at ScienceDirect
International Journal of Surgery journal homepage: www.journal-surgery.net
Original research
Impact of duration of diabetes on outcome following pancreas transplantation Burcin Ekser a, Richard S. Mangus a, John A. Powelson a, Michele L. Goble a, Muhammad A. Mujtaba b, Tim E. Taber b, Jonathan A. Fridell a, * a b
The Department of Surgery, Indiana University School of Medicine, Indianapolis, IN, USA The Department of Medicine, Indiana University School of Medicine, Indianapolis, IN, USA
h i g h l i g h t s Longstanding T1DM does not seem to negatively impact recipient outcomes following all types of pancreas transplantation. Duration of diabetes exposure significantly correlates with the need for kidney transplantation. The majority of pancreas recipients with longer standing diabetes (21e30 and >30 years) were also undergoing kidney transplantation as SPK or PAK, likely because renal failure is a late complication of diabetes. PTA recipients, on the other hand, tended to make up a much larger portion of the group with the shortest duration of diabetes.
a r t i c l e i n f o
a b s t r a c t
Article history: Received 12 February 2015 Accepted 8 April 2015 Available online 11 April 2015
Introduction: The impact of duration of T1DM on outcomes following simultaneous pancreas and kidney transplantation (SPK), pancreas after kidney transplantation (PAK), and pancreas transplantation alone (PTA) is currently unknown. Materials and methods: A total of 451 pancreas transplants performed at a single institution between January 2003 and April 2013 (SPK n ¼ 238, PAK, n ¼ 97, and PTA, n ¼ 116) were divided into three groups based on cumulative years of T1DM (0e20 years, 21e30 years, and >30 years). Early (7-day) and late (90day) pancreas allograft loss, patient and pancreas allograft survivals were analyzed. Results: While, PAK was more common in recipients with >30 years of T1DM (29%, p < 0.0047), PTA was more common in recipients with 0e20 years of T1DM (41%, p < 0.0011). In all transplant types, recipients age was significantly higher the longer the duration of diabetes. Although longer duration of T1DM correlated with a higher rate of major amputations in PAK recipients (p < 0.0032), no difference was observed in SPK or PTA. While early pancreas graft loss was 2e4% in SPK and PAK with shorter or longer T1DM (p ¼ n.s.), it reached to 10% in PTA with T1DM > 30 years (p < 0.0097). Longer duration of T1DM affected late pancreas graft loss in PAK patients (8%, p < 0.0349). Patient and death-censored graft survival rates were similar in all types of pancreas transplantation extracted by accumulation of years of T1DM prior to transplant. Conclusions: Longstanding T1DM does not seem to negatively impact recipient outcomes following all types of pancreas transplantation. © 2015 Published by Elsevier Ltd on behalf of IJS Publishing Group Limited.
Keywords: Diabetes mellitus Duration of diabetes Pancreas transplantation Outcome
Abbreviations: CAD, coronary artery disease; ESRD, end-stage renal disease; MMF, mycophenolate mofetil; n.s., not significant; PAK, pancreas after kidney transplantation; PTA, pancreas transplantation alone; rATG, rabbit antithymocyte globulin; SPK, simultaneous pancreas and kidney transplantation; T1DM, type 1 diabetes mellitus; UNOS, United Network for Organ Sharing. * Corresponding author. The Department of Surgery, Indiana University School of Medicine, 550 N University BLVD, #4258, Indianapolis, IN 46202, USA. E-mail address: [email protected] (J.A. Fridell). http://dx.doi.org/10.1016/j.ijsu.2015.04.031 1743-9191/© 2015 Published by Elsevier Ltd on behalf of IJS Publishing Group Limited.
22
B. Ekser et al. / International Journal of Surgery 18 (2015) 21e27
1. Introduction Recent analyses of national U.S. data from 1980 to 2012 suggest a doubling of the incidence and prevalence of diabetes during 1990e2008, and a plateauing between 2008 and 2012 [1]. An estimated 10e15% of the US population is affected by diabetes and 8e10% of these have type 1 diabetes mellitus (T1DM) [2]. T1DM is associated with severe late complications, including end-stage renal disease (ESRD), requiring renal replacement therapy or kidney transplantation [3]. Diabetes is also an independent major risk factor for cardiovascular and cerebrovascular disease, which is dependent on the years of exposure. Studies have shown that during the first 20 years of T1DM most of the excess mortality is attributed to renal failure, however after this period, it is considered to result largely from cardiovascular events [3e5]. International studies have revealed different mortality rates depending on the duration of diabetes exposure. The Finnish study demonstrated a cumulative mortality of 6.8% at 20 years and 15% at 30 years after diabetes diagnosis with a cumulative incidence of ESRD of 2.2.and 7.8%, respectively [6]. The impact of duration of diabetes on mortality has been reported by several groups, with rates varying between 6.8 and 13% after 20 years, and 15e29% after 30 years [3,6e9]. Pancreas transplantation is most commonly performed in the context of renal transplantation for end stage diabetic nephropathy (simultaneous pancreas and kidney transplantation (SPK) or pancreas after kidney (PAK) transplantation), or as an isolated transplant for poor glycemic control, particularly hypoglycemia unawareness (pancreas transplantation alone (PTA)), and is the only definitive long-term therapy which restores the euglycemic state and prevents and reverses secondary complications of diabetes [2,10]. The impact of recipient age, which can correlate with the duration of T1DM exposure, has been investigated in pancreas transplantation [11e13]. Schenker et al. and Ablorsu et al. have demonstrated that it is possible to achieve similar graft and patient survival rates in pancreas transplant recipients >50 years old compared to younger recipients [11,13]. Our group has recently reported on the impact of age on outcomes after pancreas transplantation. In that study, we also demonstrated that older recipients (50e59 years old, n ¼ 85 and >60 years old, n ¼ 18) have similar patient and graft survival rates compared to younger recipients [14]. Interestingly, in that study, the worst pancreas allograft survival was observed in the youngest recipients (<30 years of age). Although some studies have reported single center experiences of pancreas transplantation in the diabetic population in large cohorts [15,16] no studies to date have systemically examined the impact of duration of T1DM on outcomes following pancreas transplantation. The current study was designed to determine the impact of the exact duration of diabetes exposure on different types of pancreas transplant recipients.
included all patients undergoing pancreas transplantation (SPK, PAK, or PTA). Pancreas retransplantations, even if performed early after the first transplant, were included in the data analysis. Measurements of recipient pre-transplant HbA1c and cardiac ejection fraction, history of hypertension, smoking, and major amputation were included. Although it was limited, non-heart beating donors were also used and included in data analysis together with recipient demographics and risk factors to stratify the different risk levels in different types of pancreas transplantation. Table 1 and supplementary digital content (SDC) Tables 1s, 2s, and 3s show demographic data for all pancreas transplantation, SPK, PAK, and PTA by accumulated years of T1DM prior to transplant, respectively. Recipient listing for the transplant was according to standard procedures and protocols as established by the United Network for Organ Sharing (UNOS). All recipients were confirmed to be c-peptide negative prior to transplant. Regardless of type of pancreas transplantation (SPK, PAK, or PTA), the work-up for pancreas transplantation was similar in all recipients. All patients were required by the listing committee to have a negative cardiac stress test prior to pancreas transplantation. Criteria for cardiac catheterization included any patient with known history of coronary artery disease (CAD), multiple risk factors, or a positive finding on a cardiac stress test, as explained before [17]. 2.2. Pancreas procurement, preparation and transplantation Local pancreas allografts were typically procured using an en bloc technique following aortic flush with preservation solution and topical cooling with saline flush and ice packing, as previously described [18]. All pancreas allografts were prepared, as previously described [19]. We have previously shown that there was no difference in outcomes for local and import pancreas allografts in our experience [20]. Therefore, the data analyzed included those recipients of imported pancreas allografts as well. The recipient operation was performed through a midline incision, as previously described [21,22]. Briefly, the pancreas allograft was positioned with the tail toward the pelvis and the head and duodenum oriented superiorly to facilitate enteric exocrine drainage. All pancreas allografts, regardless of SPK, PAK or PTA, were drained enterically using a stapled technique [23]. Systemic venous drainage was performed to the vena cava or to the right common iliac vein. Arterial perfusion of the allograft was routinely established from the right common iliac artery, although on rare occasions where this vessel was found to be diseased or had been the site for arterial anastomosis for a prior transplant, the inflow would be established either from the aorta or the left common iliac artery. All SPK transplants were performed with ipsilateral placement of both the kidney and the pancreas to the right iliac vessels, as previously described [24]. Pulsatile perfusion was used routinely for the renal allograft portion of the SPK, as described [25]. 2.3. Immunosuppressive therapy
2. Materials and methods 2.1. Data collection and inclusion criteria The medical records for all deceased donor pancreas transplants performed at Indiana University between January 2003 and April 2013 were reviewed (n ¼ 451). Retrospective review of data from the transplant center database was approved by the Institutional Review Board of Indiana University School of Medicine. To obtain the data presented in the present manuscript, the comprehensive transplant recipient registry at our center, individual written and electronic medical records, and the original donor medical history were carefully reviewed. Inclusion criteria for the data analysis
The induction immunosuppressive regimen consisted of five doses of rabbit antithymocyte globulin (rATG) (1 mg/kg/dose). A single dose of rituximab (150 mg/m2) induction was also included in cases of PTA. The maintenance immunosuppressive regimen consisted of tacrolimus (through level of 8e10 ng/mL), and sirolimus (through level of 3e6 ng/mL) [26]. Steroids were only used as a premedication for rATG induction and were discontinued in all recipients, including PAK recipients receiving long-term steroids for a remote renal transplant [27]. Mycophenolate mofetil (MMF) (500 mg po b.i.d) was used as a part of the maintenance immunosuppressive regimen together with tacrolimus and sirolimus in cases of PTA or as a substitute for sirolimus in cases of drug
B. Ekser et al. / International Journal of Surgery 18 (2015) 21e27
23
Table 1 Demographic data for all pancreas transplantations by accumulated years of Type I diabetes prior to transplant.
Overall Duration of DM Recipient
(years) (mean ± SD) (range) Transplant type SPK PAK PTA Age (years) (mean ± SD) (range) Gender Male Female Race White African American Other Blood type A B AB O Body mass index (mean ± SD) (range) HbA1c (mean ± SD) Cardiac ejection fraction (mean ± SD) History of hypertension History of smoking History of major amputations
Total#
0 to 20 years
21 to 30 years
>30 years
p
451 (100%) 28.5 ± 10.1 (1e54)
111 (25%) 14.8 ± 5.0 (1e20)
157 (35%) 26.0 ± 4.0 (21e30)
183 (41%) 37.7 ± 5.4 (31e54)
<0.0001
53% 22% 26% 43.3 ± 9.9 (13e68)
47% 12% 41% 36.8 ± 10.8 (13e68)
58% 20% 22% 41.5 ± 8.4 (26e64)
52% 29% 19% 48.9 ± 7.3 (32e66)
n.s. <0.0047 <0.0011 <0.0001
60% 40%
50% 50%
62% 38%
65% 35%
<0.05 <0.05
93% 6% 1%
88% 10% 2%
93% 6% 1%
95% 5% 0%
n.s. n.s. n.s.
42% 8% 5% 45% 26 ± 4 (17e44) 8.0 ± 1.0 60 ± 10% 77% 31% 3%
36% 8% 6% 50% 25 ± 4 (17e38) 8.0 ± 0.9 60 ± 9% 72% 30% 1%
46% 7% 2% 45% 26 ± 7 (18e44) 8.1 ± 1.3 60 ± 11% 78% 36% 2%
42% 10% 7% 41% 26 ± 4 (18e39) 7.9 ± 1.2 61 ± 9% 79% 27% 6%
n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s.
26.8 ± 11.1 (4e54)
26.2 ± 11.3 (8e54)
26.9 ± 10.7 (6e51)
27.1 ± 11.4 (4e53)
n.s.
64% 36%
65% 35%
61% 39%
67% 33%
n.s. n.s.
79% 16% 5% 24 ± 5 (13e59)
77% 8% 7% 25 ± 7 (14e59)
87% 10% 3% 24 ± 4 (14e38)
74% 12% 14% 24 ± 5 (13e40)
<0.02 n.s. <0.01 n.s.
20% 59% 22% 62% 38% 1% 8.4 ± 2.5 (3.5e20.8)
23% 58% 19% 60% 40% 1% 8.7 ± 2.7 (4.5e20.8)
17% 60% 23% 64% 36% 0% 8.5 ± 2.5 (3.5e17.0)
19% 59% 22% 62% 38% 0% 8.2 ± 2.4 (4.4e18.0)
n.s. n.s. n.s. n.s. n.s. n.s. n.s.
Donor Age (years) (mean ± SD) (range) Gender Male Female Race White African American Other Body mass index (mean ± SD) (range) Cause of death Stroke Trauma Anoxia/other Procured by IU transplant team Shipped-in pancreas graft Non-heart beating donor Cold ischemia time (mean ± SD) (range) Statistically significant values are written bold.
nephrotoxicity or wound complications in SPK or PAK recipients. All patients received prophylaxis for cytomegalovirus with oral valganciclovir and Pneumocyctis jiroveci pneumonia with trimethoprim/sulfamethaxazole. 2.4. Outcomes and statistical analyses
PAK (22%), and 116 PTA (26%) (Fig. 1). For each type of pancreas transplantation, patients were divided into 3 subgroups by accumulated years of duration of diabetes exposure prior to transplant: (i) 0e20 years (n ¼ 111), (ii) 21e30 years (n ¼ 157), and (iii) > 30 years (range 31e56 years) (n ¼ 183). SPK represented approximately half of the cases regardless of duration of T1DM (Table 1).
Retrospective primary transplant outcomes included 7-day (early) and 90-day (late) pancreas transplant loss, and one-year pancreas allograft and patient survival. Secondary transplant outcomes included 5-year and 10-year pancreas allograft and patient survival. Demographic data and outcomes were compared between the three groups based on duration of diabetes, (i) 0e20 years, (ii) 21e30 years, and (iii) > 30 years, for each type of pancreas transplantation. Fisher's exact, standard chi-square, student t-test, and analysis of variance (ANOVA), log-rank (ManteleCox for survival curves) tests were used in appropriate calculations using GraphPad Prism 6 for Mac OS X (La Jolla, CA, USA). 3. Results Of 451 pancreas transplantations, there were 238 SPK (53%), 97
Fig. 1. Study design and exact number of patients in different types of pancreas transplantation by accumulated years of Type I diabetes prior to transplant.
24
B. Ekser et al. / International Journal of Surgery 18 (2015) 21e27
Although PTA was performed more than PAK and SPK (41%, p < 0.0011) in recipients with duration of T1DM 0e20 years, PAK was performed more (29%, p < 0.0047) in recipients with duration of T1DM > 30 years (Table 1). Data analyses were done comparing each subgroup for different types of pancreas transplantation. Fig. 1 shows the study design and exact numbers of recipients in each subgroup per type of pancreas transplantation. The mean duration of T1DM was not significant between different types of pancreas transplants for given subgroups; 15.9 ± 4.0 vs 15.8 ± 4.2 vs 12.0 ± 5.2 (p ¼ n.s.) in SPK, PAK, PTA, respectively for T1DM 0e20 years; 25.6 ± 3.1 vs 26.8 ± 2.7 vs 25.6 ± 3.0 (p ¼ n.s.) in SPK, PAK, PTA, respectively for T1DM 21e30 years, and 37.7 ± 5.4 vs 38.1 ± 5.8 vs 37.1 ± 4.6 (p ¼ n.s.) in SPK, PAK, PTA, respectively for T1DM > 31 years. In all types of transplant, mean age of recipients in subgroups was significantly higher and mirrored the duration of diabetes (SDC Tables 1s, 2s, and 3s). While >30 years of T1DM correlated with higher BMI in SPK patients (p < 0.01), no difference was observed in PAK, and PTA. In SPK, recipient and donor demographics were similar between different years of diabetes duration, except those mentioned above and the race of the donor, as shown in SDC Table 1s. In PAK, homogeneity between subgroups was not consistent, including race of recipients and donors, recipient blood type, and history of major amputations prior to transplant, donor cause of death, procurement by local team or shipped-in pancreas allografts (SDC Table 2s). The impact of duration of T1DM can be seen in recipients of PAK >30 years of T1DM with significantly higher major amputations rate (9%, p < 0.0032). In PTA, recipient and donor demographics were mainly comparable, except gender and race of the recipient, and history of smoking which was significantly lower in patients with >30 years of T1DM (p < 0.003) (SDC Table 3s). Median cold ischemia time (~8 h) and percentage of utilization of non-heart beating donor were similar in all 3 subgroups in all types of pancreas transplantation. Although the percentage of Indiana University (IU) Transplant Team procured pancreas grafts similar (52% vs 61%, p ¼ n.s.) in PAK and PTA, the percentage of IU Team procurement reached to 91% (p < 0.003) in SPK (SDC Tables 1s, 2s, and 3s).
While early (7-day) pancreas graft loss was similar among different duration of diabetes exposure in SPK and PAK transplant recipients (0e4%), it reached to 10% in PTA recipients with a history of >30 years of T1DM (Table 2). Late (90-day) pancreas graft loss was significantly higher only in PAK recipients with more than >20 years of diabetes (p < 0.0349). As expected, serum creatinine was lower in PTA recipients than SPK and PAK transplant recipients, whom had reflected higher glomerular filtration rate (GFR) 73 ± 30 vs 69 ± 33 vs 81 ± 43 in SPK, PAK, PTA, respectively (p < 0.01). Longer duration of diabetes exposure (>30 years) had a significant negative impact on GFR in all types of pancreas transplantation, regardless of the need for a kidney transplant or not (Table 2). There were a total of 86 (19%) pancreas graft losses in 451 patients over 10 years. Of 86 pancreas graft losses, 46 (54%) were due to patient death with a functioning allograft. Although longer duration of T1DM was correlated with higher rate of pancreas graft loss in recipients requiring kidney transplantation (19% in SPK and 23% in PAK in patients with >30 years of T1DM), it was contrary in PTA (35% pancreas graft loss in patients with 0e20 years of T1DM) (Table 3). Higher percentage of pancreas loss in PTA patients with 0e20 years of diabetes (p < 0.0009) was due to increased rate of chronic rejection (15%, p < 0.0003). Patient deaths were not significantly different among the causes in all types of pancreas transplantation, regardless of the duration of T1DM exposure, except the total percentage of death (19%) in PAK patients with >30 years of exposure (p < 0.037) (Table 3). Patient survival was found to be significantly reduced in recipients of pancreas transplantation (including all types) with T1DM > 30 years of duration (Fig. 2). However, longer duration of T1DM did not significantly affect pancreas allograft survival (with or without death-censorship) (Fig. 2). Despite differences in pancreas graft loss and patient death in different types of pancreas transplantation, KaplaneMeier survival curves did not show a significant difference in patient and death-censored pancreas graft survivals (SDC Fig. 1s). The only difference was observed in SPK patients with 21e30 years of T1DM in non-death-censored pancreas graft survival (p < 0.05). As expected, patient survival was higher in PTA recipients than SPK
Table 2 Post-transplant outcomes in different type of pancreas transplants by accumulated years of Type I diabetes prior to transplant.
Simultaneous pancreas-kidney transplantation (SPK), n (%) Pancreas graft loss within 7 days Pancreas graft loss within 90 days Hospital stay at the transplant (days) (mean ± SD) (median) Creatinine 1-year (mg/dL) (mean ± SD) GFR 1-year (mL/min/1.73 m2) (mean ± SD) HbA1c 1-year (%) (mean ± SD) C-peptide 1-year (ng/mL) (mean ± SD) Pancreas after kidney transplantation (PAK), n (%) Pancreas graft loss within 7 days Pancreas graft loss within 90 days Hospital stay at the transplant (days) (mean ± SD) (median) Creatinine 1-year (mg/dL) (mean ± SD) GFR 1-year (mL/min/1.73 m2) (mean ± SD) HbA1c 1-year (%) (mean ± SD) C-peptide 1-year (ng/mL) (mean ± SD) Pancreas transplantation alone (PTA), n (%) Pancreas graft loss within 7 days Pancreas graft loss within 90 days Hospital stay at the transplant (days) (mean ± SD) (median) Creatinine 1-year (mg/dL) (mean ± SD) GFR 1-year (mL/min/1.73 m2) (mean ± SD) HbA1c 1-year (%) (mean ± SD) C-peptide 1-year (ng/mL) (mean ± SD) Statistically significant values are written bold. a Median value is used for statistical calculation.
Total#
0 to 20 years
21 to 30 years
>30 years
238 (100%) 3% 6% 10.4 ± 12.1 (8) 1.4 ± 0.8 73 ± 30 5.5 ± 0.8 2.8 ± 2.1 97 (100%) 2% 6% 15.5 ± 46.9 (7) 1.4 ± 0.5 69 ± 33 5.5 ± 0.8 2.7 ± 1.3 116 (100%) 5% 7% 14.3 ± 42 (7) 1.2 ± 0.8 81 ± 43 5.6 ± 0.9 2.4 ± 1.7
52 (22%) 4% 6% 9.0 ± 5.7 (8) 1.5 ± 1.2 71 ± 31 5.5 ± 0.5 2.9 ± 1.7 13 (13%) 0% 0% 11.7 ± 13.7 (7) 1.4 ± 0.4 72 ± 22 5.7 ± 1.5 2.9 ± 1.7 46 (40%) 7% 9% 11.3 ± 13.0 (7) 1.3 ± 1.0 91 ± 50 5.7 ± 1.0 2.6 ± 1.9
91 (38%) 2% 3% 10.0 ± 8.3 (7.5) 1.3 ± 0.5 79 ± 29 5.5 ± 1.0 2.5 ± 1.6 31 (32%) 3% 7% 27.0 ± 85.1 (6.5) 1.4 ± 0.4 78 ± 36 5.6 ± 0.5 2.5 ± 1.3 35 (30%) 0% 3% 7.3 ± 2.1 (7) 1.3 ± 0.5 81 ± 37 5.4 ± 0.5 2.7 ± 2.0
95 (40%) 4% 8% 11.7 ± 17.0 (8) 1.4 ± 0.8 67 ± 29 5.5 ± 0.8 3.1 ± 2.8 53 (55%) 2% 8% 10.7 ± 10.1(7.5) 1.3 ± 0.6 63 ± 32 5.4 ± 0.6 2.7 ± 1.3 35 (30%) 10% 10% 25.9 ± 74.2 (7.5) 1.1 ± 0.5 68 ± 38 5.6 ± 1.0 1.7 ± 0.9
p n.s. n.s. n.s.a n.s. <0.0053 n.s. n.s. n.s. <0.0349 n.s.a n.s. <0.0314 n.s. n.s. <0.0097 n.s. n.s.a n.s. <0.0261 n.s. <0.0088
Table 3 Causes of pancreas graft loss and patient death in different type of pancreas transplants by accumulated years of Type I diabetes prior to transplant. Simultaneous pancreas-kidney (SPK) transplants
Pancreas transplants alone (PTA)
0 to 20 years 21 to 30 years >30 years p
Total#
0 to 20 years 21 to 30 years >30 years p
238 (100%) 36 (15%) 7 (3%) 6 (3%) 21 (9%) 1 (0.5%)
52 (22%) 10 (19%) 2 (4%) 1 (2%) 7 (14%) 0 (0%)
91 (38%) 8 (9%) 1 (1%) 2 (2%) 5 (5%) 0 (0%)
95 (40%) 18 (19%) 4 (4%) 3 (3%) 9 (10%) 1 (1%)
n.s. n.s. n.s. n.s. n.s.
97 (100%) 23 (24%) 3 (3%) 3 (3%) 13 (13%) 1 (1%)
13 (13%) 2 (16%) 0 (0%) 0 (0%) 1 (8%) 0 (0%)
31 (32%) 9 (29%) 2 (6%) 1 (3%) 4 (13%) 1 (3%)
53 (55%) 12 (23%) 1 (2%) 2 (4%) 8 (15%) 0 (0%)
116 (100%) 0.0414 27 (23%) n.s. 3 (3%) n.s. 8 (7%) n.s. 12 (10%) n.s. 4 (3%)
46 (40%) 16 (35%) 1 (2%) 7 (15%) 6 (13%) 2 (4%)
35 (30%) 6 (17%) 0 (0%) 0 (0%) 4 (11%) 2 (6%)
35 (30%) 5 (14%) 2 (6%) 1 (3%) 2 (6%) 0 (0%)
0.0009 n.s. 0.0003 n.s. n.s.
1 (0.5%) 24 (10%) 2 (1%) 5 (2%) 2 (1%) 1 (0.5%) 2 (1%) 1 (0.5%) 7 (3%) 4 (2%)
0 8 1 0 1 1 1 0 2 2
0 5 0 1 0 0 0 0 3 1
1 (1%) 11 (12%) 1 (1%) 4 (4%) 1 (1%) 0 (0%) 1 (1%) 1 (1%) 2 (2%) 1 (1%)
n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s.
3 (3%) 16 (17%) 0 (0%) 2 (2%) 1 (1%) 1 (1%) 2 (2%) 4 (4%) 0 (0%) 6 (6%)
1 1 0 0 0 0 0 0 0 1
1 5 0 1 1 1 0 0 0 2
1 (2%) 10 (19%) 0 (0%) 1 (2%) 0 (0%) 0 (0%) 2 (4%) 4 (8%) 0 (0%) 3 (6%)
0.0185 0.037 n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s.
0 2 0 1 0 0 0 0 0 1
0 1 0 0 0 0 0 0 1 0
0 3 1 0 0 0 1 0 0 1
n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s.
(0%) (15%) (2%) (0%) (2%) (2%) (2%) (0%) (4%) (4%)
(0%) (6%) (0%) (1%) (0%) (0%) (0%) (0%) (3%) (1%)
(8%) (8%) (0%) (0%) (0%) (0%) (0%) (0%) (0%) (8%)
(3%) (16%) (0%) (3%) (3%) (3%) (0%) (0%) (0%) (6%)
Total#
0 6 1 1 0 0 1 0 1 2
(0%) (5%) (1%) (1%) (0%) (0%) (1%) (0%) (1%) (2%)
0 to 20 years 21 to 30 years >30 years p
(0%) (4%) (0%) (2%) (0%) (0%) (0%) (0%) (0%) (2%)
(0%) (3%) (0%) (0%) (0%) (0%) (0%) (0%) (3%) (0%)
(0%) (9%) (3%) (0%) (0%) (0%) (3%) (0%) (0%) (3%)
Statistically significant values are written bold.
25
and PAK transplant recipients throughout the study (SDC Fig. 1s).
4. Discussion
In the USA, diabetes mellitus is the most common cause of ESRD, blindness, major limb amputations, and peripheral and coronary
Fig. 2. KaplaneMeier patient and graft (including death-censored graft) survival postpancreas transplant in all pancreas transplants by years of diabetes.
B. Ekser et al. / International Journal of Surgery 18 (2015) 21e27
Overall transplants (n, %) Cause of pancreas graft loss (n, %) Vascular thrombosis Chronic rejection Patient death with function graft Pseudoaneurysm or arterioenteric fistula Other Cause of death (n, %) Stroke/neurologic event Cardiac etiology Post-transplant lymphoma disease Graft versus host disease Infection/sepsis Cancer Medical/surgical complications Other cause/unknown
Pancreas after kidney (PAK) transplants
Total#
26
B. Ekser et al. / International Journal of Surgery 18 (2015) 21e27
vascular complications, although its incidence and prevalence has plateaued in the last decade [1,2]. A steady increase in the number and age of T1DM patients implies that the pancreas transplant professional is going to encounter an increasing number of potential pancreas transplant candidates with decades of exposure to this disease [1,2]. In fact, among the three study groups in the present study, patients with diabetes duration >30 years represented the largest cohort (n ¼ 183) (Fig. 1). Jorgensen et al. recently showed that the absolute and relative mortality rates in patients with T1DM have decreased over the last 10 years confirming that the high mortality rate in T1DM patients is largely related to longterm diabetes complications [28]. Despite the amelioration in diabetes care, longer duration of diabetes is still associated with considerable incidence of morbidity and mortality [3,6e9]. The duration of T1DM is also an important factor for long-term complications, such as vascular diseases, including CAD and cerebrovascular disease, diabetic nephropathy, and neuropathy. Among the risk factors associated with longer duration of diabetes, CAD is one of the most important since it mainly determines the patient's longevity. Recently, our group investigated the impact of CAD on the outcome following pancreas transplantation [17]. We found that patients with a history of pretransplant cardiac interventions had worse clinical outcomes, including an 11-fold increased risk for myocardial infarction and 3fold increased risk for stroke after pancreas transplantation. Not surprisingly, the patients that had undergone prior cardiac interventions were also significantly older than those that did not [17]. There was no significant difference in patient and deathcensored pancreas graft survival in all types of transplants between subgroups of accumulated diabetes exposure prior to transplant. Five-year and 10-year patient survival was higher in PTA patients, regardless of duration of T1DM (91e97%) in comparison to other types of transplants. The negative impact of longer duration of T1DM was observed in recipients of PAK transplants at 10-year follow-up (89%, 65%, and 76% in 0e20 years, 21e30 years, and >30 years T1DM, respectively). The lowest pancreas survival was observed in PTA in 0e20 years exposed T1DM at 10-year follow up (61%). Although recipients of PTA received additional immunosuppressive treatment with a single dose rituximab induction and sirolimus maintenance with tacrolimus, chronic rejection rate was still the highest in this group (7% vs 3%). In PTA, patients with longer duration of diabetes had lower rejection rates. It is possible that older patients may have an immunological advantage posttransplantation, although it remains unclear whether these recipients provide a more favorable immunological environment because of better compliance or a less active immune system. Richardson et al. recently investigated the levels of antibodies to the diabetes-associated islet autoantigens, such as GAD, IA-2, and zinc transporter-8 (ZnT8) in patients with long-duration T1DM, and the influence of age at diagnosis and HLA genotype on autoantibody persistence [29]. They demonstrated that all autoantibodies against GAD, IA-2 and ZnT8 were significantly decreased in patients with long-duration diabetes but there was also a remarkable persistence of anti-GAD antibodies in long-term survivors of T1DM. The decreased titers of autoantibodies were explained by the weak expression of diabetes-associated islet autoantigens due to decreased number of surviving or regenerating islets in the diabetic pancreas. This could potentially explain the favorable immunological state of older recipients together with their high compliance for organ transplantation. Management of pancreas transplant recipients has also improved significantly over the last two decades. Earlier studies cautioned that one-year pancreas graft survival in patients over 49
years of age was approximately 50% and older age (>49) was considered a relative contraindication to pancreas transplantation [12,30]. However, in the present study, the longest diabetes exposure groups (>30 years) have a mean age of recipients of 48e49 years in all types of pancreas transplants. Our group also previously showed that recipients of pancreas transplantation between 50 and 59 years of age (n ¼ 85) and >60 years of age (n ¼ 18) had 95% and 93% 1-year patient survival, respectively [14]. In the present study, we observed that duration of diabetes exposure significantly correlates with the need for kidney transplantation. The majority of pancreas recipients with longer standing diabetes (21e30 and >30 years) were also undergoing kidney transplantation as SPK or PAK, likely because renal failure is a late complication of diabetes. PTA recipients, on the other hand, tended to make up a much larger portion of the group with the shortest duration of diabetes (40% vs 13% and 22% for PTA vs PAK and SPK, respectively, p < 0.001). In conclusion, although the patients with longstanding T1DM requires increased rate of kidney transplantation, their outcomes following all types of pancreas transplantation are not negatively impacted. Ethical approval Local hospital IRB was obtained. Sources of funding None. Author contribution B.E and R.S.M participated in data collection and analysis, and performance of research. B.E and J.A.F participated in writing of the article. R.S.M and J.A.F participated in research design and performance of research. J.A.P and J.A.F performed most of the surgeries. J.A.P, J.A.F, M.L.G, M.A.M, T.E.T participated in follow up of patients and data collection. All authors participated in critical review of the manuscript. Conflicts of interest Authors declare no conflict of interests. Guarantor Burcin Ekser, MD. Research registration unique identifying number (UIN) researchregistry15. Acknowledgments Part of this paper was presented in the 14th World Congress of the International Pancreas and Islet Transplant Association (IPITA 2013). Appendix A. Supplementary data Supplementary data related to this article can be found at http:// dx.doi.org/10.1016/j.ijsu.2015.04.031. References [1] L.S. Geiss, J. Wang, Y.J. Cheng, et al., Prevalence and incidence trends for
B. Ekser et al. / International Journal of Surgery 18 (2015) 21e27
[2] [3]
[4]
[5]
[6] [7]
[8]
[9]
[10]
[11]
[12]
[13]
[14] [15]
diagnosed diabetes among adults aged 20 to 79 years, United States, 1980e2012, JAMA 312 (2014) 1218e1226. R.W.G. Gruessner, A.C. Gruessner, The current status of pancreas transplantation, Nat. Rev. Endocrinol. 9 (2013) 555e562. M. Stadler, M. Auinger, C. Anderwald, et al., Long-term mortality and incidence of renal dialysis and transplantation in type 1 diabetes mellitus, J. Clin. Endocrinol. Metab. 91 (2006) 3814e3820. K. Asao, C. Sarti, T. Forsen, et al., Long-term mortality in nationwide cohorts of childhood-onset type 1 diabetes in Japan and Finland, Diabetes. Care 26 (2003) 2037e2042. J.S. Dorman, R.E. Laporte, L.H. Kuller, et al., The Pittsburgh insulin-dependent diabetes mellitus (IDDM) morbidity and mortality study. Mortality results, Diabetes 33 (1984) 271e276. P. Finne, A. Reunanen, S. Stenman, et al., Incidence of end-stage renal disease in patients with type 1 diabetes, JAMA 294 (2005) 1782e1787. I. Muhlhauser, H. Overmann, R. Bender, et al., Predictors of mortality and endstage diabetic complications in patients with type 1 diabetes mellitus on intensified insulin therapy, Diabet. Med. 17 (2000) 727e734. G. Pambianco, T. Costacou, D. Ellis, et al., The 30-year natural history of type 1 diabetes complications: the Pittsburgh epidemiology of diabetes complications study experience, Diabetes 55 (2006) 1463e1469. P. Rossing, P. Hougaard, K. Borch-Johnsen, et al., Predictors of mortality in insulin dependent diabetes: 10 year observational follow up study, BMJ 313 (1996) 779e784. P. Fioretto, M.W. Steffes, D.E.R. Sutherland, et al., Reversal of lesions of diabetic nephropathy after pancreas transplantation, N. Engl. J. Med. 339 (1998) 69e75. E. Ablorsu, A. Ghazanfar, S. Mehra, et al., Outcome of pancreas transplantation in recipients older than 50 years: a single-centre experience, Transplantation 86 (2008) 1511e1514. C.E. Freise, P.G. Stock, J.S. Melzer, Increased morbidity and mortality of simultaneous pancreas-renal transplantation in patients over 49 years of age, Transpl. Proc. 30 (1998) 292. P. Schenker, O. Vonend, B. Kruger, et al., Long-term results of pancreas transplantation in patients older than 50 years, Transpl. Int. 24 (2011) 136e142. A.P. Shah, R.S. Mangus, J.A. Powelson, et al., Impact of recipient age on whole organ pancreas transplantation, Clin. Transpl. 27 (2013) E49eE55. R. Ollinger, C. Margreiter, C. Bosmuller, et al., Evolution of pancreas transplantation. Long term results and perspectives from a high-volume center, Ann. Surg. 256 (2012) 780e787.
27
[16] H.W. Sollinger, J.S. Odorico, S.J. Knechtle, et al., Experience with 500 simultaneous pancreas-kidney transplants, Ann. Surg. 228 (1998) 284e296. [17] R.S. Mangus, J. Powelson, S.B. Kinsella, et al., Pretransplant coronary artery disease associated with worse clinical outcomes in pancreas transplantation, Clin. Transpl. 27 (2013) E442eE447. [18] J.A. Fridell, R.S. Mangus, J.A. Powelson, Histidine-tryptophan-ketoglutarate for pancreas allograft preservation: the Indiana university experience, Am. J. Transpl. 10 (2010) 1284e1289. [19] J.A. Fridell, J.A. Powelson, C.E. Sanders, et al., Preparation of the pancreas allograft for transplantation, Clin. Transpl. 25 (2011) E103eE112. [20] J.A. Fridell, R.S. Mangus, E.F. Hollinger, et al., No difference in transplant outcomes for local and import pancreas allografts, Transplantation 88 (2009) 723e728. [21] J.A. Fridell, A. Agarwal, J.A. Powelson, Pancreas transplantation at Indiana university: a brief overview of recent progress, in: R. Corry, R. Shapiro (Eds.), Pancreatic Transplantation: Informa Healthcare, USA, 2007, pp. 421e427. [22] J.A. Fridell, R.S. Mangus, J.A. Powelson, et al., Pancreas transplantation in the new millennium: the Indiana university experience, Clin. Transpl. (2011) 145e156. [23] J.A. Fridell, M.L. Milgrom, S. Henson, et al., Use of the end-to-end anastomotic circular stapler for creation of the duodenoenterostomy for enteric drainage of the pancreas allograft, J. Am. Coll. Surg. 198 (2004) 495e497. [24] J.A. Fridell, A. Shah, M.L. Milgrom, et al., Ipsilateral placement of simultaneous pancreas and kidney allografts, Transplantation 78 (2004) 1074e1076. [25] A.P. Shah, D.P. Milgrom, R.S. Mangus, et al., Comparison of pulsatile perfusion and cold storage for paired kidney allografts, Transplantation 86 (2008) 1006e1009. [26] J.A. Fridell, R.S. Mangus, E.F. Hollinger, et al., The case for pancreas after kidney transplantation, Clin. Transplant 23 (2009) 447e453. [27] J.A. Fridell, A. Agarwal, J.A. Powelson, et al., Steroid withdrawal for pancreas after kidney transplantation in recipients on maintenance prednisone immunosuppression, Transplantation 82 (2006) 389e392. [28] M.E. Jorgensen, T.P. Almdal, B. Carstensen, Time trends in mortality rates in type 1 diabetes from 2002 to 2011, Diabetologia 56 (2013) 2401e2404. [29] C.C. Richardson, J.A. Dromey, K.A. McLaughlin, et al., High frequency of autoantibodies in patients with long duration type 1 diabetes, Diabetologia 56 (2013) 2538e2540. [30] C.L. Manske, Y. Wang, W. Thomas, Mortality of cadaveric kidney transplantation versus combined kidney-pancreas transplantation in diabetic patients, Lancet 346 (1995) 1658e1662.