Pancreas Transplantation From Living Donors: A Single Center Experience of 20 Cases

American Journal of Transplantation 2016; 16: 2413 2420 Wiley Periodicals Inc. Copyright 2016 The American Society of Transplantation and the American Society of Transplant Surgeons doi: 10.1111/ajt.13738 Pancreas Transplantation From Living Donors: A Single Center Experience of 20 Cases J. Y. Choi, J. H. Jung, H. Kwon, S. Shin, Y. H. Kim and D. J. Han* Department of Surgery, University of Ulsan College of Medicine and Asan Medical Center, Seoul, Korea *Corresponding author: Duck Jong Han, djhan@amc.seoul.kr Living donor pancreas transplantation (LDPT) has several advantages over deceased donor pancreas transplantation (DDPT), including better HLA matching, shorter ischemic time, and shorter waiting time. It remains an attractive option for diabetes mellitus (DM) patients with end stage renal disease. We reviewed 20 cases of LDPT performed in Asan Medical Center between October 1992 and March 2015. Six cases (30%) were pancreas transplantation alone (PTA), and the rest (70%) were simultaneous pancreas and kidney transplantation (SPK). Relations of donor and recipient were parents in 7 (35%), siblings in 6 (30%), spouse in 6 (30%), and cousin in 1 (5%). Graft survival in SPK at 1, 3, 5, and 10 years was 91.7%, 83.3%, 83.3%, and 83.3%, respectively, and that in PTA recipients was 50%, 33.3%, 16.7%, and 16.7%, respectively (p = 0.005). Causes of graft failure in SPK were thrombosis (one case), and rejection (one case), whereas those in PTA were noncompliance (two cases), thrombosis (one case), reflux pancreatitis (one case), and chronic rejection (one case). In terms of pancreas exocrine drainage, two grafts (25%) maintained their function in bladder drainage, while all grafts maintained in enteric drainage p < 0.05). Seven (35%) donors experienced minor pancreatic juice leakage and one underwent reoperation due to postoperative hematoma. Most donors maintained normoglycemia and normal renal function. However, two donors developed DM (at 1 and 90 months postdonation), and were treated with oral hypoglycemic agents. Graft survival in PTA recipients was poorer than in SPK due to poor compliance and bladder drainage related problems. The surgical and metabolic complication rates of donors can be minimized by applying strict donor criteria. Therefore, LDPT with enteric drainage is an acceptable treatment for SPK. Abbreviations: DD, deceased donor; DDPT, deceased donor pancreas transplantation; DM, diabetes mellitus; LD, living donor LDPT, living donor pancreas transplantation; PTA, pancreas transplantation alone; SPK, simultaneous pancreas and kidney transplantation Received 03 November 2015, revised 01 January 2016 and accepted for publication 19 January 2016 Introduction Diabetes mellitus (DM) is associated with various complications such as retinopathy, neuropathy, and nephropathy, and its prevalence has increased steadily worldwide (1 3). These complications are the leading causes of increased mortality and morbidity in DM patients (1,2). Although exogenous insulin therapy can be useful for maintaining normoglycemia, it does not prevent long-term complications (2,3). Pancreas transplantation is considered to be the most efficient treatment modality for restoring normoglycemia by suppling sufficient b cells (1,2,4,5). Since the first pancreas transplantation was performed in 1966 at the University of Minnesota, many centers have performed this operation and the outcomes have been improved due to the use of better surgical techniques and immunosuppressants (1,3). Based on this improvement, various types of pancreas transplantation from deceased donors (DD) have been performed (1). However, living donor pancreas transplantation (LDPT) was first reported in the late 1970s, although the procedure is not performed widely (1,6). LDPT has several advantages over deceased donor pancreas transplantation (DDPT), including better HLA matching, shorter ischemic and waiting times, less need for immunosuppression, and a lower risk of infection (7). Furthermore, the shortage of DDs and improved graft outcomes for LDPT can be an attractive treatment for DM patients with or without end-stage renal disease (ESRD) (2,7). In kidney transplantation, it has been already proven that the use of an organ from a living donor (LD) not only increases the number of transplants, but also shows excellent graft survival rates compared to a DD (2,8). However, because LDPT is technically more difficult and may be associated with increased donor morbidity (including the development of DM and surgical complications), its performance has been limited (2,7). In this study, we reviewed LDPT (performed at Asan Medical Center, as a treatment for DM) and analyzed the clinical characteristics that affect graft survival and donor safety. 2413

Choi et al Materials and Methods Patients From October 1992 to July 2015, 300 cases of pancreas transplantation were performed at Asan Medical Center. Among these, 20 (6.67%) were LDPT. Here, we retrospectively review the clinical characteristics and outcomes of both recipients and donors. Before donation, all donors were assessed with respect to social and psychological status (to ensure that consent was voluntary and their reasons were altruistic). Donors underwent a general medical work-up, which included cross-matching of donor T-lymphocytes and recipient serum, and evaluation of pancreatic endocrine function (serum amylase and lipase, fasting plasma glucose, and fasting hemoglobin [Hb] A1C levels, an oral glucose tolerance test, an intravenous glucose test, and measurement of islet cell autoantibodies [anti-gad antibodies]). The donor recipient match was confirmed by the endocrinologist. In addition, three-dimensional angiography was performed by dynamic computed tomography (CT) of the abdominal cavity to evaluate the donors anatomy of the pancreas and kidney. Recipients underwent a general medical work-up. Surgical procedure LDPT was performed electively when both the recipient and donor were in optimal condition. The surgery was performed at a single center by the same surgical team. Donor surgery After an upper midline incision, either right or left nephrectomy was performed, followed by a distal pancreatectomy with splenectomy for simultaneous pancreas and kidney transplantation (SPK). However, pancreas transplant alone (PTA) patients underwent only distal pancreatectomy with splenectomy. After nephrectomy, the lesser sac was opened to visualize the body and tail of the pancreas. Mobilization and detachment of the inferior border of the distal pancreas was always performed meticulously with mobilization of the spleen. The upper border of the distal pancreas was mobilized to the level of the splenic artery. After visualizing the celiac trunk and the splenic, hepatic, and left gastric arteries, the pancreas neck was dissected from behind the portal vein. Transaction of the pancreatic neck over the left side of portal vein was then performed. Following identification of the pancreatic duct, bleeding from both sides of the transected pancreas neck was controlled by ligation with fine suture. The proximal pancreatic duct was ligated and the distal pancreatic duct was marked with a fine suture. The end of the pancreas was then oversewn. After systemic heparinization, the proximal splenic vein and artery at the junction of the superior mesenteric vein and common hepatic artery, respectively, were clamped and divided. Removal of the pancreas from a living donor must be done gently because it is critical to reduce the risk of graft pancreatitis and to preserve the splenic artery and vein. After procurement, the splenic artery was cannulated and the pancreas was flushed with low pressure (20 30 cm H 2 O) ice-cold histidine tryptophan ketoglutarate solution (200 300 ml) via the splenic artery to clear the blood. Splenectomy was performed ex vivo. Recipient surgery After a midline incision, the external iliac artery and vein were mobilized. The segmental pancreas was placed in the right iliac fossa. The donor splenic vein and artery were anastomosed to the recipient external iliac vein and artery, respectively, in an end-to-side fashion. For bladder drainage, the splenic arterial anastomosis was lateral and proximal to the splenic vein anastomosis due to the anatomical arrangement of the distal pancreas. For enteric drainage, the splenic arterial anastomosis was medial and distal to the splenic vein anastomosis due to cephalad position of the pancreas neck. After administration of intravenous heparin (70 U/kg), the pancreas graft was reperfused. In SPK, kidney transplantation was performed before pancreas transplantation in the left iliac fossa using standard techniques. For enteric drainage of the pancreas graft, a Roux-en-Y limb of the upper jejunum was anastomosed to the whole cut surface of the body of the pancreas using the double-layer invagination technique, and the pancreatic duct was cannulated with a stent. An end-to-side jejuno-jejunostomy was performed about 40 cm distal to the pancreaticojejunostomy. For bladder drainage, the anterior surface of the pancreas was anastomosed to the bladder by two-layer closure with pancreatic duct stent insertion. Immunosuppression Before 1999, OKT3 was used for induction and tacrolimus/cyclosporine, mycophenolate mofetil, and steroids for maintenance. From 1999 to 2004, basiliximab was used for induction, followed by maintenance with tacrolimus, mycophenolate mofetil, and low-dose prednisolone. From 2004, rabbit antithymocyte globulin (thymoglobulin) was used for induction, and tacrolimus and mycophenolate mofetil, coupled with the early withdrawal (within 1 week) of steroids, for maintenance. Anticoagulation therapy Anticoagulation therapy was administered both during and after surgery. Continuous intravenous heparin (400 1000 U/h) was administered and the activated partial thromboplastin time (aptt) was monitored every 6 h, after which oral warfarin was administered for 3 months. The target level of aptt and prothrombin time (international normalized ratio) was 1.5 to 29 the upper reference range. If a thrombus was found on CT angiography, the aptt level was targeted to 29 the upper reference range, with weekly or biweekly monitoring of graft patency by CT angiography (9). Postoperative monitoring and follow-up During hospitalization and after discharge, the serum glucose, amylase, lipase, blood cell count, electrolyte, creatinine, and C-peptide level of each donor was monitored. In cases of bladder-drained exocrine secretion, urine amylase levels were monitored to evaluate graft function. In some cases, intravenous insulin or oral hypoglycemic agents were used to maintain the glucose level at <200 mg/dl during the early postoperative period. Graft failure was defined as the time at which the reuse of exogenous insulin was required. The endocrine function of the residual pancreas in each donor was monitored by measuring serum glucose, amylase, lipase HbA1C, and serum C-peptide levels. Statistical analysis Categorical variables were analyzed using the absolute and relative frequencies. Quantitative variables were analyzed using the mean and the standard deviation. Posttransplant patient and graft survival were analyzed using Kaplan Meier analysis. Results Demographics and characteristics of donors and recipients The first case of LDPT was performed at Asan Medical Center in 1992, but failed due to graft thrombosis. LDPT was restarted from 2006. Until now, 20 cases of LDPT have been performed, comprising 6.67% of all pancreas transplantations (Figure 1). Among these, 6 cases (6/20, 2414 American Journal of Transplantation 2016; 16: 2413 2420

Pancreas Transplantation From Living Donors Figure 1: Type and number of living donor pancreas transplantations in Asan Medical Center. PTA, pancreas transplant alone; SPK, simultaneous pancreas and kidney transplantation. 30%) were PTA and the rest (14/20, 70%) were SPK. Two procedures for exocrine drainage were performed (Figure 2). During the early period, bladder drainage was performed in 8 (8/20, 40%) LDPT (6 PTA and 2 SPK cases); however, during the later period, all LDPTs (12/20, 60%) were SPK with enteric drainage. The characteristics of the recipients and donors are shown in Table 1. The mean recipient age was 30.80 8.02 (17 49) years and 13 (65%) were female. Most recipients had type I DM (n = 18, 90%) and the age at the onset was 16.50 6.98 (10 39) years. The mean donor age was 41.95 9.86 (27 60) years and five (50%) were female. Seven donors were the parents of the recipients, six were siblings, six were spouses, and one was a cousin. ABO blood type was incompatible in one donor. Recipients After operation, seven recipients (35%) had intra-abdominal bleeding and hematoma, and four of them required laparotomy for hematoma evacuation. Six recipients (30%) experienced partial vascular thrombosis of the graft and two (10%) experienced graft pancreatitis. Figure 2: The number of living donor pancreas transplantations according to the drainage and operation type in Asan Medical Center. PTA, pancreas transplant alone; SPK, simultaneous pancreas and kidney transplantation. American Journal of Transplantation 2016; 16: 2413 2420 2415

Choi et al Table 1: Donor selection criteria for living donor pancreas transplantation Body mass index <27 kg/m 2 No family history of diabetes Normal endocrine function Hemoglobin A1C <6.0% Plasma glucose level <150 mg/ml during 75 g oral glucose tolerance test Basal fasting insulin level <20 lmol/l per ml Insulin response to glucose or arginine >300% of basal insulin Glucose disposal rate >1% during intravenous glucose tolerance test Absence of anti-insulin and anti-islet antibody Donor age recipient age <10 years The patient survival rate was 100%. The mean follow-up period was 41.40 7.77 (1 116) months. The survival rates for the pancreatic graft at 1, 3, and 5 years were 79.1%, 67.8%, and 61.0%, respectively, and were maintained up until 10 years (Figure 3). We performed one ABO-incompatible LDSPK. The recipient blood type was B and the donor blood type was A. As a preconditioning regimen, the recipient received a single dose of rituximab (200 mg) 1 week before plasmapheresis; plasmapheresis was performed four times. After preconditioning, the isoagglutinin titer of anti-a antibodies fell from 1:256 to 1:2. After SPK, the function of the kidney and pancreas graft was maintained without rejection for 42 months. During the follow-up period, acute cellular rejection (ACR) was observed in four recipients (4/20, 20%). All recipients with ACR received steroid pulse therapy. Among these, one recipient completely recovered from pancreas graft rejection. Seven recipients (five PTA and two SPK cases) lost their graft function. The causes of graft failure were graft thrombosis (two), rejection (two), reflux pancreatitis (one), and poor compliance (two) (Table 3). For PTA, two recipients lost graft function due to poor compliance. One recipient who received a graft from her mother experienced severe DM gastropathy and had difficulty taking oral immunosuppressants. She became hyperglycemic and insulin dependent 17 months after PTA. The other received a graft from her husband. At 35 months after the operation, she was under personal stress and did not take immunosuppressants properly. Her urine amylase level fell and graft rejection could not be recovered. For SPK, one recipient with bladder drainage experienced recurrent reflux pancreatitis and pancreatic juice leakage, resulting in pseudocyst formation at 14 months after SPK. At laparotomy, we transected the pancreas graft tail and performed retrograde pancreaticoenterostomy followed by bladder wall closure. After surgery, the peripancreatic fluid collection and pancreatic duct dilatation issues were resolved. However, the graft lost function. Graft survival at 1 and 3 years in SPK patients was 91.7% and 83.3%, respectively, and was maintained up until 10 years. Graft survival in PTA recipients at 1, 3, 5, Figure 3: Long-term graft survival after living donor pancreas transplantation compared with that after deceased donor transplantation. 2416 American Journal of Transplantation 2016; 16: 2413 2420

Pancreas Transplantation From Living Donors Table 2: Baseline characteristics of donors and recipients for living donor pancreas transplantation N = 20 (%) 20.36 1.69 (16.77 23.44) 21.66 2.55 (16.1 27.1) Recipients Age, mean (SD), years 30.80 8.02 (17 49) Gender (female) 13 (65) Body mass index, mean (SD), kg/m 2 Type I DM 18 (90) Onset of DM, 16.50 6.98 (10 39) mean (SD), years Duration of DM, 14.15 6.11 (4 25) mean (SD), years Amount of insulin 31.25 13.98 (14 68) (unit/day) DM complication Nephropathy 15 (75) Neuropathy 8 (40) Retinopathy 16 (80) HLA mismatching ABDR 3.12 1.76 (0 6) DR 1.00 0.79 (0 2) ABO incompatibility 1 (5) PRA class I positivity 2/13 PRA class I positivity 1/13 Mean follow-up 39.55 37.01 (1 113) period (months) Donors Age, mean (SD), years 41.95 9.86 (27 60) Gender (female) 10 (50) Body mass index, mean (SD), kg/m 2 Relationship Parent 7 (35) Sibling 6 (30) Spouse 6 (30) Uncle 1 (5) Mean follow-up 40.15 18.64 (1 270) periods (months) PRA, pannal reactive antibody. and 10 years was 50%, 33.3%, 16.7%, and 16.7%, respectively, illustrating that graft survival in SPK recipients was significantly better than that in PTA recipients (p = 0.005) (Figure 4). One graft (1/8, 12.5%) maintained function under bladder drainage, while all grafts (12/12, 100%) maintained function under enteric drainage (p < 0.05) (Figure 5). Kidney graft survival in SPK was 100%. Acute kidney graft rejection occurred in three recipients, but was recovered with steroid pulse therapy in all cases. Donors No donor suffered mortality from distal pancreatectomy and/or unilateral nephrectomy. One donor underwent reoperation the day after the initial operation due to hematoma at the nephrectomy site. Seven donors (35%) experienced minor pancreatic juice leakage at the distal pancreatectomy site, which was controlled with conservative management. Hyperglycemia developed in two donors (at 1 and 90 months after donation) and was treated with oral hypoglycemic agents. The rest maintained normoglycemia and had normal renal function (HbA1C, 5.87 2.80; C-peptide, 1.76 0.81 ng/ml; serum glucose level, 110.95 12.57 mg/dl; and scr, 0.94 0.20 mg/dl after donation). Discussion According to the Diabetes Atlas published by the International Diabetes Federation in 2014, 8.3% of the world population (about 387 million people) has DM. The prevalence of DM is rapidly increasing, especially in African and Asian regions, including Korea (1 3). According to the recently released Diabetes Fact Sheet in Korea, the prevalence of diabetes at the age of 30 and older is about 11.9% of the population, accounting for 3.2 million cases of DM patients (about one in every eight adults). Pancreas transplantation has become a widely accepted treatment modality for DM that consistently restores normoglycemia and returns HbA1C levels to normal (1,2). Indeed, 300 consecutive patients underwent pancreas transplantation at Asan Medical Center; 20 of these were LDPT (6.6%). The outcomes of the 20 LDPT cases are comparable with those at other centers. Patient survival was 100% and graft survival at 1, 3, and 5 years was 79.1%, 67.8%, Table 3: Characteristics of recipients who experienced pancreas graft failure Operation type Exocrine drainage Rejection therapy Graft function maintenance period (months) Cause of graft failure Graftectomy 1 PTA Bladder 1 Graft thrombosis No 2 PTA Bladder 17 Noncompliance No 3 SPK Bladder Steroid pulse therapy 14 Reflux pancreatitis No 4 PTA Bladder Steroid pulse therapy 37 Noncompliance Yes 5 PTA Bladder 1 Graft thrombosis Yes 6 SPK Bladder Steroid pulse therapy, 6 Rejection No Antithymocyte therapy 7 PTA Bladder 8 Rejection No PTA, pancreas transplantation alone; SPK, simultaneous pancreas and kidney transplantation. American Journal of Transplantation 2016; 16: 2413 2420 2417

Choi et al Figure 4: Long-term graft survival after living donor pancreas transplantation according to operation type. PTA, pancreas transplant alone; SPK, simultaneous pancreas and kidney transplantation. Figure 5: Long-term graft survival after living donor pancreas transplantation according to exocrine drainage type. and 61.0%, respectively, and was maintained for 10 years. Plasma glucose levels started to decrease at the time of the operation in almost all recipients. We experienced seven cases of graft failure, five of which developed during the early period (1992 2006). After 2007, there were no graft failures. At Minnesota (5), 125 cases LDPT were performed between 1978 and 2010 and all graft survival at 1, 5, and 10 years was 62%, 50%, and 34%, 2418 American Journal of Transplantation 2016; 16: 2413 2420

Pancreas Transplantation From Living Donors respectively, and pancreas graft survival rates for technically successful cases were 79%, 64%, and 44%, respectively. When analyzed according to time (Era 1, 1978 1986; Era 2, 1987 1997; and Era 3, 1988 2010), graft survival was significantly better in Era 3 (1 year, 44% vs. 58% vs. 100%; p < 0.001; and 10 year, 33% vs. 58% vs. 74%; p < 0.001). This reflects the high technical difficulty of LDPT due to the relatively small size of the splenic artery and vein. In early LDPT cases, technical failure was a problem; however, this was overcome by technical improvements and declined gradually over time (5,7). In patients with minimal risk of surgical complications, LDPT should not be avoided as a DM treatment for technical reasons. LDPT shows better graft survival than DDPT due to the immunologic advantages over technical difficulty (2,4,7). However, after the introduction of tacrolimus, the graft survival rate of DDPT improved remarkably and showed results comparable with those for LDPT (4,7). In our center, the 1- and 5-year graft survival rates for LDPT and DDPT were not significantly different (79.1% vs. 89.5% and 61.0% vs. 80.9%, respectively; p = 0.073). The 1- and 5-year graft survival rates for LDSPK and DDSPK were comparable (92.3% vs. 91.0% and 83.9% vs. 86.8%, respectively; p = 0.917) (Figure 6). Improvements in surgical technique, immunosuppressants, and the use of anticoagulants mean that the indications for LDPT have changed (2,5,7). Interest in LDPT in solitary categories (pancreas transplantation after kidney transplantation and PTA) has declined whereas that in LDSPK has increased (7). Here, we found that the graft survival rates differed according to operation type. During the early period, PTA was most often performed; however, we experienced high graft failure in PTA recipients (5/6, 83.3%). The Minnesota data suggest that technical failure due to graft thrombus and rejection were more common in PTA than SPK (5). Therefore, we only performed SPK in LDPT after 2008. In SPK, only two graft failures developed (significantly fewer than for PTA). When considering the exocrine pancreas drainage procedure, LDPT with enteric drainage showed a better graft outcome than those with bladder drainage. Therefore, LDSPK with enteric drainage seems to be a recommendable treatment for DM patients with ESRD. LDPT has several advantages (3 7). In SPK from DD, the waiting time for transplantation can be extended from 2 to 3 years in our country, which is similar to that reported by others (4,10). While DM patients wait for a transplant, their physical condition deteriorates rapidly on dialysis and they can develop severe complications. Shortage of DD kidneys is the main factor that limits our ability to transplant potential SPK recipients in a timely manner. However, if the candidates have a LD, the operation time can be adjusted, especially for SPK recipients (4,5,10). However, in case of PTA from a DD, the waiting time is shorter (about 3 6 months) than that for SPK. Thus, the Figure 6: Long-term graft survival after simultaneous pancreas and kidney transplantation according to donor type. SPK, simultaneous pancreas and kidney transplantation. American Journal of Transplantation 2016; 16: 2413 2420 2419

Choi et al shorter waiting time for DD transplant, coupled with the poor graft survival of LDPTA transplants, means that LDPTA is unnecessary. Also, LDPT can be performed despite HLA antibody or major ABO incompatibilities (4,5,7). If recipients are highly sensitized or must avoid high-dose immunosuppression, LDPT can still be performed in case of well-matched HLA (5,6). For ABO and/ or cross-match incompatible recipients, LDPT can be performed by desensitization (5,6). Here, we performed one SPK with ABO incompatibility from her father. As a preconditioning regimen, the recipient received a single dose of rituximab (200 mg) at 1 week before plasmapheresis. After SPK, the function of the kidney and pancreas grafts has been maintained up to 42 months after operation. There is no doubt that the evaluation of the outcome of LDPT should focus not only on the recipient but also on the donor (2,3,7,10,11). The Minnesota group (5) experienced relatively low levels of surgical complications (<5%), which included pancreatitis, leakage, pseudocyst formation, or reoperation. However, HbA1C levels were elevated in 10/115 donors, 3 of whom required insulin treatment. Kenmochi et al (7) performed 12 LDPT operations and experienced one pancreatic pseudocyst at 6 months after surgery. During the 5-year follow-up period, no patient developed diabetes. However, HbA1C increased in two donors. We also experienced some surgical (hematoma and minor pancreas leakages) and metabolic (DM) complications, but, they were not critical and were comparable with those reported in other centers. However, it is clear that efforts to reduce complications and improve donor safety are required. Under these conditions, LDPT can be considered a secure treatment modality for DM. In conclusion, graft survival in LDPTA was poor when compared with that in LDSPK and DDPTA. However, graft survival in LDSPK was comparable with that after DDSPK. As the surgical and metabolic complication rates of donors and recipients become less, LDPT should be an acceptable and safe treatment, particularly SPK with enteric drainage, due to shorter waiting times and good graft survival. LDPT will reduce the mortality of DM patients on waiting list, and expand the donor pool to overcome organ shortages. Disclosure The authors of this manuscript have no conflicts of interest to disclose as described by the American Journal of Transplantation. References 1. Han DJ, Sutherland DE. Pancreas transplantation. Gut Liv 2010; 4: 450 465. 2. Zielinski A, Nazarewski S, Bogetti D, et al. Simultaneous pancreas-kidney transplant from living related donor: A singlecenter experience. Transplantation 2003; 76: 547 552. 3. Reynoso JF, Gruessner CE, Sutherland DE, Gruessner RW. Short- and long-term outcome for living pancreas donors. J Hepatobiliary Pancreat Sci 2010; 17: 92 96. 4. Boggi U, Amorese G, Marchetti P, Mosca F. Segmental live donor pancreas transplantation: Review and critique of rationale, outcomes, and current recommendations. Clin Transplant 2011; 25: 4 12. 5. Sutherland DE, Radosevich D, Gruessner R, Gruessner A, Kandaswamy R. Pushing the envelope: Living donor pancreas transplantation. Curr Opin Organ Transplant 2012; 17: 106 115. 6. Sutherland DE, Goetz FC, Najarian JS. Living-related donor segmental pancreatectomy for transplantation. Transplant Proc 1980; 12: 19 25. 7. Kenmochi T, Asano T, Maruyama M, et al. Living donor pancreas transplantation in Japan. J Hepatobiliary Pancreat Sci 2010; 17: 101 107. 8. Knight RJ, Burrows L, Bodian C. The influence of acute rejection on long-term renal allograft survival: A comparison of living and cadaveric donor transplantation. Transplantation 2001; 72: 69 76. 9. Kim YH, Park JB, Lee SS, Byun JH, Kim SC, Han DJ. How to avoid graft thrombosis requiring graftectomy: Immediate posttransplant CT angiography in pancreas transplantation. Transplantation 2012; 94: 925 930. 10. Sutherland DE. Extra-renal living donor transplants with special reference to segmental pancreas transplantation. Clin Transplant 2011; 25: 1 3. 11. Otsuki K, Yoshikawa K, Kenmochi T, et al. Evaluation of segmental pancreatic function using 11C-methionine positron emission tomography for safe living donor operation of pancreas transplantation. Transplant Proc 2011; 43: 3273 3276. 2420 American Journal of Transplantation 2016; 16: 2413 2420