American Journal of Transplantation 2010; 10: 846–851 Wiley Periodicals Inc.
C 2010 The Authors C 2010 The American Society of Journal compilation Transplantation and the American Society of Transplant Surgeons
doi: 10.1111/j.1600-6143.2010.03060.x
Pancreatic Graft Survival Despite Partial Vascular Graft Thrombosis due to Splenocephalic Anastomoses C. Margreitera , W. Marka , D. Wiedemannb , a a ¨ ¨ ¨ , C. Bosm uller , R. Suchera , R. Ollinger c a d M. Freund , H. T. Maier , A. Greiner , H. Fritsche , J. Pratschkea , R. Margreitera and F. Aignera, * a
Department of Visceral, Transplant and Thoracic Surgery, University Clinic of Cardiac Surgery, c Department of Diagnostic Radiology, d Department of Vascular Surgery, e Department of Anatomy, Histology and Embryology, Division of Clinical and Functional Anatomy, Innsbruck Medical University, Innsbruck, Austria *Corresponding author: Felix Aigner,
[email protected] b
Thrombotic complications following pancreas transplantation are still the most common cause of nonimmunologic graft loss. The aim of this study was to analyze pancreatic graft function after partial arterial graft thrombosis and the investigation of the pancreatic arterial anatomy with regard to intraparenchymal anastomoses. We retrospectively analyzed the data for 175 consecutive pancreas transplants performed between January 2002 and October 2007. Selective Ygraft angiography was performed in 10 and rubber-milk injection in 5 fresh pancreas specimens. Thrombosis of one leg of the Y-graft was diagnosed in 18 (10.3%) patients. Only one of these patients with thrombosis of the splenic artery required exogenous insulin. Sufficient graft perfusion was demonstrated in all of the remaining grafts. One graft was lost due to acute rejection. In all specimens angiography showed an excellent perfusion of the pancreaticoduodenal arcade, even after selective cannulation of the splenic artery. Arterial collaterals between the gastroduodenal, splenic artery and the superior mesenteric artery were demonstrated. Our results demonstrate that global perfusion of the pancreatic graft and sufficient graft function is sustained after the thrombotic occlusion of one branch of the Y-graft by a complex system of intraparenchymal anastomoses. These anatomical findings may have consequences for resection strategies in pancreas surgery. Key words: Graft survival, pancreas transplantation, thrombosis, vascularization Received 25 November 2009, revised 08 January 2010 and accepted for publication 29 January 2010
846
Introduction Despite the fact that pancreas transplantation is a routine procedure for patients suffering from end-stage diabetic nephropathy (1), surgical complications including vascular thrombosis is still the major cause for nonimmunologic graft failures after transplantation (2,3). Apart from venous thrombosis arterial reconstruction using a donor Y-graft still represents a risk factor in pancreas transplantation. The aim of this study was to retrospectively analyze the graft function in recipients experiencing a thrombotic occlusion of one branch of the Y-graft (partial Y-graft thrombosis, PYGT). We further investigated the intraparenchymal anastomoses between the supply areas of the superior mesenteric artery (SMA) and the splenic artery (SA) by angiography and dissection of anatomical specimens. Previous work of our group focused on radiological findings after pancreas transplantation (4,5) and raised the question for a clinical anatomical study. Anatomical data on the arterial blood supply of the pancreatic head has existed for more than two centuries since Haller published his anatomical findings in Iconae anatomicarum in 1745 (6). Bertelli’s review (7–11) summarized the various studies in detail. In brief: the arterial blood supply to the duodenum and pancreas comes from three different sources: hepatic artery (HA), SA and SMA. The head of the pancreas receives blood mainly from the HA via the gastroduodenal artery (GDA) and from the SMA via the inferior pancraticoduodenal artery (IPD). The GDA supplies the head of the pancreas via the superior pancreaticoduodenal artery, posterior superior pancreaticoduodenal and/or anterior superior pancreaticoduodenal arteries (ASPD). The IPD divides into the posterior inferior (PIPD) and the anterior inferior (AIPD) pancreaticoduodenal arteries. These form anastomoses between the anterior and posterior pancreaticoduodenal arches with complex and frequent variations (11). Body and tail of the pancreas are supplied with blood mainly by the SA and SMA. The dorsal pancreatic artery (DPA) (12), small pancreatic branches and the pancreatica magna artery (PMA) receive their blood supply from the SA. The DPA divides into two branches running in opposite directions: first, as ‘Kirk’s anastomosis’ to the ASPD or prepancreatic arcade, and second, as a branch to the body and tail of the pancreas. The IPD arises from the SMA as a common trunk for the AIPD and PIPD and the inferior pancreatic or transverse pancreatic artery. After having
Pancreatic Splenocephalic Anastomoses
observed two patients with thrombosis of one branch of the arterial Y-graft and normal graft function, the question about the importance of arterial intraparenchymal anastomoses was raised. As clinical radiomorphology is limited, anatomical models were used for this study to clarify this problem.
with 3-D reconstruction (Figure 1) was routinely performed before discharge in 135 patients (77%) with stable renal function. Mean duration of hospital stay was 25 days. All partial Y-graft thromboses were diagnosed before discharge.
Methods
When graft thrombosis was suspected, ultrasound and angio-CT scan were performed. Acute graft rejection and primary vascular thrombosis were difficult to differentiate. Diagnosis was based mainly on clinical parameters such as function, pain and fever as well as angio-CT scans. Only patients with complete vascular graft thrombosis were explored and thrombectomized whereas asymptomatic patients with partial thrombosis were only heparinized as described above. All removed grafts were histologically examined. This study was done in accordance with the ethical principles of the Declaration of Helsinki and reported to the local IRB.
Patient study The medical records of all consecutive patients undergoing pancreatic transplantation (n = 175; 93 male and 82 female) between January 2002 and October 2007 (141/175 [80.6%] simultaneous pancreas–kidney, SPK; 27/175 [15.4%] pancreas-after-kidney, PAK; and 8/175 [4.6%] pancreas transplantations alone, PTA) were retrospectively analyzed. Twenty-six, 14.9%, were retransplantions including six SPK (23.1%). All whole-organ pancreatic transplantations were performed with enteric drainage following a technique described previously (13). Arterial reconstruction was performed with a Y-graft using an iliac segment. The majority of cases had systemic (n = 153, 87.4%) and some portal venous drainage (n = 22, 12.6%). Immunosuppression consisted of antithymocyte globulin in 164 patients and alemtuzumab in the remaining 11 cases, rapid steroid tapering, tacrolimus with target trough levels between 12 and 15 ng/dL for the first 3 months and mycophenolic acid mofetil (2g/day). To prevent thrombotic complications patients were heparinized with subcutaneous low molecular weight heparin according to body weight until discharge. When thrombotic complications occurred during the hospital stay, patients were given heparin intravenously to maintain PTT of 45 s for 1 week and then switched to acetylsalicylic acid. Graft perfusion was assessed with color Doppler ultrasound (HDI 5000, Philips Medical Systems, Bothell, WA) daily during the first week and three times a week thereafter or when clinically indicated. A multislice angio-CT
Thrombotic occlusion of either the SA or the SMA was defined as PYGT. Partial vein thrombosis was described as partial or complete occlusion of either the superior mesenteric or splenic vein.
Anatomical study Ten nonembalmed fresh deceased donor pancreases obtained from the Department of Anatomy, Histology and Embryology were harvested in a standardized fashion (13). As preservation solution, University of Wisconsin (UW) solution was applied backtable. The organs appeared macroscopically normal. The GDA and the common bile duct were ligated and the specimens were prepared as for pancreas transplantation including a Carrel patch with the SMA, but without creation of the Y-graft. Cannulas were inserted into the SA and the SMA and angiographies were performed following injection of increasing volumes of contrast medium (5, 10 and 15 mL) (Jopamiro, Bracco, Vienna, Austria), initially only via the SA for visualization of the perfusion status of the pancreas. Thereafter, the same procedure was repeated with the SMA. Additionally, five phenol-formaldehyde-preserved pancreas specimens obtained from the same institution were injected with colored rubber milk
Figure 1: Three-dimensional reconstruction of angio-CT scans of patients after simultaneous pancreas–kidney transplantation. (A) Thirty-nine-year-old male recipient with patent Y-graft and excellent global perfusion of the pancreatic graft; (B) Thirty-five-year-old male recipient with partial Y-graft thrombosis (occluded SMA) and sustained global perfusion via the SA; Y-graft (arrow), borders of graft perfusion (arrowheads), K simultaneously transplanted kidney. Insert (magnification) shows patent Y-graft and SA (black line) and occluded SMA (white dotted line).
American Journal of Transplantation 2010; 10: 846–851
847
Margreiter et al. and carefully dissected under the operating microscope OPMI (Carl Zeiss Meditec AG, Jena, Germany) to demonstrate vascular collaterals and anastomoses within the pancreatic head.
Statistical analysis Values are expressed as mean and standard deviation. GraphPad Prism version 5.00 for Windows (GraphPad Software, San Diego, CA, www.graphpad.com) was used for all analyses. Kaplan–Meier survival calculations and the log rank test between patient groups were performed.
Results Patient study Major indication for pancreas transplantation was endstage type 1 diabetic nephropathy (n = 160, 91.4%) and 15 patients (8.6%) suffered from type 2 insulin-dependent diabetes mellitus (C-peptide >2 lg/L). Total vascular complication rate among 175 SPK, PAK and PTA including all kinds of vascular occlusion (partial and complete venous as well as Y-graft thrombosis) was 21.1% (37 cases). Total incidence of graft loss was 15.4% (27/175). Reasons for graft loss are depicted in Table 1. Eighteen cases (10.3%) of PYGT were diagnosed within the first 3 postoperative months. Thrombosis of the SMA leg occurred in 10 cases (5.7%), thrombosis of the SA leg in 8 cases (4.6%) (Table 1). Comparison of the PYGT group and the group without partial arterial thrombosis showed no difference in graft survival (Figure 2). One- and 3-year graft survival after pancreas transplantation was 92% and 83%, respec-
tively, as compared to 83% for both one- and 3-yearsurvival in the PYGT group (Figure 2, p = 0.712). In total, seven patients died with a functioning graft (two patients in the thrombosis groups and five without any vascular complication). All but one patient with PYGT showed excellent graft function (C-peptide >2 lg/L) without needing exogenous insulin. One patient required insulin 20 IU per day after discharge and lost the pancreatic graft after two episodes of rejection 2 years posttransplant. Another graft was lost in the PYGT group (SMA) 2 months posttransplant due to rejection and subsequent vein thrombosis (Table 1). Among 14 (8%) patients who lost their graft due to acute rejection only two were associated with arterial and/or venous thrombosis. No recipients with PYGT developed complete arterial thrombosis, but partial or complete venous thrombosis has been observed in three and one case, respectively. Anticoagulation therapy led to resolution of PYGT in three out of 18 patients within 1 year. In three patients CYGT was detected upon loss of graft function, requiring immediate surgical intervention. Preoperative ultrasound was judged unreliable and multisliced CT revealed arterial thrombosis of the graft 72 h after transplantation. Venous thrombosis was observed in 19 (10.9%) patients. Partial venous thrombosis was detected in 11 (6.3%) cases and complete venous thrombosis in eight (4.6%) cases
Table 1: Patient characteristics with regard to thrombotic complications; SMA-PYGT with concomitant partial portal vein thrombosis (n = 2), SA-PYGT with concomitant partial portal vein thrombosis (n = 1), SMA-PYGT with concomitant complete portal vein thrombosis (n = 1), CYGT with concomitant complete portal vein thrombosis (n = 1) Variables (%)
N Age (SD) Hours cold ischemia (SD) Min warm ischemia (SD) Donor age (SD) Sex m w Diabetes type 1 Type 2 Death with functioning graft Need for insulin Graft loss Cause of graft loss Rejection Venous thrombosis Arterial thrombosis Sepsis Hemorrhage Noncompliance Retransplantation PTA SPK PAK
848
Total
No vascular compl
PVT
CVT
CYGT
SA-PYGT
SMA-PYGT
175 (100) 44.2 (9) 13.4 (3.2) 32.5 (8.3) 31.6 (11.2)
138 (78.9) 44.6 (9.2) 13.5 (3.2) 32 (7.7) 31.1 (11.3)
11 (6.3) 46 (35–52) 12 (9–18) 36.5 (23–43) 34 (15–49)
8 (4.6) 41.5 (25–56) 12.5 (10–18) 38.5 (26–59) 34.5 (13–46)
3 (1.7) 46 (40–54) 15 (10–17) 41 (18–59) 23 (19–27)
8 (4.6) 46.5 (38–56) 11.5 (6–15) 27 (20–33) 37 (15–44)
10 (5.7) 41 (36–52) 13 (7–19) 35 (22–54) 40 (21–50)
106 (60.6) 69 (39.4) 160 (91.4) 15 (8.6) 7 (4) 22 (12.6) 27 (15.4)
87 (63) 51 (37) 126 (91.3) 12 (8.7) 5 (3.6) 10 (7.2) 13 (9.4)
3 (27.3) 8 (72.7) 9 (81.8) 2 (18.2) − 2 (18.2) 2 (18.2)
4 (50) 4 (50) 8 (100) − − 8 (100) 8 (100)
2 (66.7) 1 (33.3) 2 (66.7) 1 (33.3) − 2 (66.7) 3 (100)
5 (62.5) 3 (37.5) 8 (100) − 2 (25) 1 (12.5) 1 (12.5)
7 (70) 3 (30) 9 (90) 1 (10) − − 1 (10)
14 (8) 7 (4) 3 (1.7) 1 (0.6) 1 (0.6) 1 (0.6) 26 (14.8) 8 (4.6) 141 (80.6) 27 (15.4)
12 (8.7) − − − 1 (0.7) − 21 (15.2) 4 (2.9) 113 (81.9) 21 (15.2)
1 (9.1) − − 1 (9.1) − − 1 (9.1) 0 (0) 9 (81.8) 2 (18.2)
− 7 (87.5) 1 (12.5) − − − − 3 (37.5) 5 (62.5) −
− 1 (33.3) 2 (66.7) − − − 1 (33.3) − 3 (100) −
1 (12.5) − − − − − 2 (25) − 4 (50) 4 (50)
1 (10) − − − − − 1 (10) − 9 (90) 1 (10)
American Journal of Transplantation 2010; 10: 846–851
Pancreatic Splenocephalic Anastomoses
medium into the SA (Figure 3). The occurrence of anastomoses between the SA and SMA, but particularly to the pancreaticoduodenal arcade, was observed in all specimens. Intraparenchymal anastomoses between the pancreatic head, body and tail were illustrated in detail in the rubber milk-injected anatomical specimens (Figure 4A). Figure 4B demonstrates a collateral branch between the SA and SMA. In the case of PYGT of the SMA imitated by clamping the vessel and vice versa in the case of partial thrombosis of the SA sufficient perfusion of the pancreatic head was provided by intraparenchymal anastomoses as demonstrated in the 3-D CT angio reconstructions. Figure 1B shows a 35-year-old male recipient with partial Y-graft thrombosis (occluded SMA) and sustained global perfusion via the SA.
Figure 2: Cumulative graft survival after pancreas transplantation with patent (constant line) and partially occluded (dotted line) Y-graft.
requiring pancreatectomy. A majority of transplants were SPK (81.8% in the partial and 62.5% in the complete venous thrombosis group, Table 1). Complete venous thrombosis occurred within 48 h after transplantation in all but two cases where thrombosis was seen 1 month following transplantation. All of the grafts with complete Y-graft or complete venous thrombosis were lost. Two of the patients with partial venous thrombosis required exogenous insulin, but subsequently lost their graft due to rejection and septic complications (Table 1).
Anatomical study Angiography showed excellent perfusion of the pancreaticoduodenal arcade via anastomoses in all cases after isolated injection of even low volumes of contrast
Discussion Since Kelly and Lillehei (14) performed the first SPK in 1966, vascular complications have remained the most common cause of nonimmunologic graft loss (3). Graft loss to technical failure is reported to be 6.8–11% depending on the graft type (15). Graft thrombosis occurs predominantly within the first postoperative days (1). At our center the thrombosis rate resulting in graft loss due to portal vein and/or Y-graft occlusion following pancreas transplantation performed was 11/175 (6.3%) (Table 1). Most of the complications were observed in the PTA group (37.5%; 3/8). In contrast, Gruessner et al. (15) reported for PAK a thrombosis rate twice as high as for SPK or PTA recipients. Preexisting chronic immunosuppression may unfavorably alter the coagulatory system of the recipient, making sequential transplants more susceptible to thrombotic events. Among predisposing factors donor age and cardiocerebrovascular cause of donor death, type of preservation
Figure 3: Angiography of a fresh pancreatic specimen after injection of contrast medium into the splenic artery (arrow). Global perfusion of the pancreatic graft is provided by intraparenchymal anastomoses.
American Journal of Transplantation 2010; 10: 846–851
849
Margreiter et al.
one patient with thrombosis of the SA branch required exogenous insulin. In contrast, 2 of 11 patients with partial venous thrombosis had impaired graft function (18.2%) (Table 1).
Figure 4: Intraparenchymal anastomoses (‘splenocephalic anastomoses’) between the pancreatic head, body and tail. (A) Schematic illustration demonstrating possible collateral branches (asterisks); GDA = gastroduodenal artery; SPD = superior pancreaticoduodenal artery; IPD = inferior pancreaticoduodenal artery; SMA = superior mesenteric artery; SA = splenic artery. (B) Rubber milk-injected anatomical specimen with one collateral branch (arrows) from the SA to the supply area of the IPD.
solution (16), cold ischemia time, type of pancreatic transplantation, drainage technique, immunosuppression and postoperative anticoagulation have been described. With regard to arterial reconstruction, the iliac Y-graft became the method of choice (17). Risk factors specific for arterial thrombosis include arteriosclerotic lesions or iatrogenic intimal lesions as well as inappropriate anastomotic techniques. Furthermore, impaired microcirculation due to acute rejection (alloimmune vasculitis) may contribute to graft thrombosis. In our series a total of 18 isolated PYGT (10.3%) in either the SA branch (n = 8) or the SMA branch (n = 10) were diagnosed because CT with contrast medium was performed according to our protocol before patient discharge if their renal function was stable. Among them, 850
The possibility of pancreatic allograft survival following PYGT due to dual blood supply (SMA, SA) was first mentioned by Troppmann et al. (2) and radiologically documented with MR angiography by Hagspiel et al. (18) PYGT may remain undetected unless specific diagnostic steps are undertaken. Four of eighteen patients with PYGT (22.2%) developed concomitant venous thrombosis and were heparinized and later switched to acetylsalicylic acid. In three of them thrombosis was completely reversed within 1 year as diagnosed by CT scan. Several authors share our strategy of using low molecular weight heparin in combination with long-term administration of acetylsalicylic acid (19). Others recommend dextran followed by low-dose heparin or a combination of acetylsalicylic acid and dipyrimidole (20). Multislice and contrastenhanced CT is the most sensitive method for detecting thrombotic lesions as well as perfusion deficits via parenchymal enhancement. This method, however, is limited by the nephrotoxic side-effect of intravenous contrast agents with regard to constricted renal function after SPK (4,5). Treatment options for thrombotic complications are limited (e.g. thrombectomy or thrombolysis) with poor outcome (2,21), often leading to graft pancreatectomy (22,23). In selected cases successful immediate retransplantation was described (24). In our series all grafts with complete venous or arterial thrombosis were lost. The role of rejection in graft thrombosis has been thoroughly discussed (1). Our findings, however, seem to demonstrate that rejection may play only a minor role, since in only two out of 14 patients, who had lost their pancreas graft for rejection, concomitant thrombosis was diagnosed. We hypothesize that survival of the pancreatic graft after PYGT is related to the presence of intraparenchymal arterial anastomoses providing global perfusion of the graft, which could be demonstrated in the anatomical specimens (Figure 4). This is confirmed by long-term graft survival rates without any significant difference between the group with patent and the group with partially occluded Y-graft (p = 0.712; Figure 2). Differences in 1-year graft survival (92% vs. 83%) may be due to the graft loss within 1 year in one case or concomitant portal vein thrombosis in four cases in the PYGT group. Kirk’s arcade plays a key role in global blood supply to the graft following ligation of the GDA (12). It consists of collaterals between the small pancreatic branches and the PMA connecting the pre- and postpancreatic arcade. Even after occlusion of the SMA, sufficient global perfusion of the pancreatic graft was provided by the SA through retrograde perfusion of the pancreaticoduodenal arcade, thus bringing into question the need for a Y-graft at all (Figures 1B, 4). These intraparenchymal anastomoses have never been described in detail on the basis of a clinical–radiological study. However, an anatomical study by Fiedor et al. described the variability of the arteries of the pancreas American Journal of Transplantation 2010; 10: 846–851
Pancreatic Splenocephalic Anastomoses
especially the dorsal pancreatic artery. The authors conclude that arteriography through the celiac trunk and the SMA in the donor should be performed prior to procurement (25). Our anatomical–radiological data confirm the constant appearance of intraparenchymal anastomoses between the PMA connecting the pre- and postpancreatic arcade. In light of their functional importance, we therefore suggest the term ‘splenocephalic anastomoses’ (Figure 4A, B). This knowledge can be useful for example, during multiorgan procurement when dealing with variations in the arterial blood supply for example common trunk or a right hepatic artery arising from the SMA. These anatomical characteristics may be of general interest in pancreatic surgery—as already discussed for pancreatic head anastomoses (26)—and may be a route of collateral circulation in addition to the two pancreatic arcades. According to our data, the ‘point of no return’ after ligation of the GDA during pancreas resection may therefore be reconsidered. Preoperative determination of the arterial vascularisation status (by CT scan or MRI) is therefore mandatory before resection of tumors of the pancreatic head whose resection is questionable.
10.
11.
12. 13.
14.
15.
16.
References 17. 1. Delis S, Dervenis C, Bramis J, Burke GW, Miller J, Ciancio G. Vascular complications of pancreas transplantation. Pancreas 2004; 28: 413–420. 2. Troppmann C, Gruessner AC, Benedetti E et al. Vascular graft thrombosis after pancreatic transplantation: Univariate and multivariate operative and nonoperative risk factor analysis. J Am Coll Surg 1996; 182: 285–316. ´ 3. Gilabert R, Fernandez-Cruz L, Real MI, Ricart MJ, Astudillo E, ˜ X. Treatment and outcome of pancreatic venous graft Montana thrombosis after kidney-pancreas transplantation. Br J Surg 2002; 89: 355–360. 4. Freund MC, Steurer W, Gassner EM et al. Spectrum of imaging findings after pancreas transplantation with enteric exocrine drainage: Part 1, posttransplantation anatomy. Am J Roentgenol 2004; 182: 911–917. 5. Freund MC, Steurer W, Gassner EM et al. Spectrum of imaging findings after pancreas transplantation with enteric exocrine drainage: Part 2, posttransplantation complications. Am J Roentgenol 2004; 182: 919–925. 6. Haller Von A. Iconae Anatomicarum. Goettingen: Vandenhoeck; 1754; 2: 1–50. 7. Bertelli E, Di Gregorio F, Bertelli L, Mosca S. The arterial blood supply of the pancreas: A review. I. The superior pancreaticoduodenal and the anterior superior pancreaticoduodenal arteries. An anatomical and radiological study. Surg Radiol Anat 1995; 17: 97– 106. 8. Bertelli E, Di Gregorio F, Bertelli L, Civeli L, Mosca S. The arterial blood supply of the pancreas: A review. II. The posterior superior pancreaticoduodenal artery. An anatomical and radiological study. Surg Radiol Anat 1996; 18: 1–9. 9. Bertelli E, Di Gregorio F, Bertelli L, Civeli L, Mosca S. The arterial blood supply of the pancreas: A review. III. The inferior pancreati-
American Journal of Transplantation 2010; 10: 846–851
18.
19.
20.
21.
22.
23.
24.
25.
26.
coduodenal artery. An anatomical review and a radiological study. Surg Radiol Anat 1996; 18: 67–74. Bertelli E, Di Gregorio F, Bertelli L, Orazioli D, Bastianini A. The arterial blood supply of the pancreas: A review. IV. The anterior inferior and posterior pancreaticoduodenal aa., and minor sources of blood supply for the head of the pancreas. An anatomical review and radiologic study. Surg Radiol Anat 1997; 19: 203–212. Bertelli E, Di Gregorio F, Mosca S, Bastianini A. The arterial blood supply of the pancreas: A review. V. The dorsal pancreatic artery. An anatomic review and a radiologic study. Surg Radiol Anat 1998; 20: 445–452. Nghiem DD. Revascularization of the gastroepiploic artery in pancreas transplant. Transpl Int 2008; 21: 774–777. ¨ Margreiter R, Konigsrainer A, Schmid T, Takahashi N, Pernthaler H, Ofner D. Multiple organ procurement-a simple and safe procedure. Transplant Proc 1991; 23: 2307–2308. Kelly WD, Lillehei RC, Merkel FK, Idezuki Y, Goetz FC. Allotransplantation of the pancreas and duodenum along with the kidney in diabetic nephropathy. Surgery 1967; 61: 827–837. Gruessner AC, Sutherland DE. Pancreas transplant outcomes for United States (US) and non-US cases as reported to the United Network for Organ Sharing (UNOS) and the International Pancreas Transplant Registry (IPTR) as of June 2004. Clin Transplant 2005; 19: 433–455. Schneeberger S, Biebl M, Steurer W et al. A prospective randomized multicenter trial comparing histidine-tryptophaneketoglutarate versus University of Wisconsin perfusion solution in clinical pancreas transplantation. Transpl Int 2009; 22: 217–224. Ciancio G, Lo Monte A, Julian JF, Romano M, Miller J, Burke GW. Vascular complications following bladder drained, simultaneous pancreas-kidney transplantation: The University of Miami experience. Transpl Int 2000;13(Suppl 1): S187–S190. Hagspiel KD, Nandalur K, Burkholder B et al. Contrast-enhanced MR angiography after pancreas transplantation: Normal appearance and vascular complications. Am J Roentgenol 2005; 184: 465–473. ¨ Hopt UT, Busing M, Schareck W et al. Prevention of early postoperative graft thrombosis in pancreatic transplantation. Transplant Proc 1993; 25: 2607–2608. Bynon JS, Stratta RJ, Taylor RJ, Lowell JA, Cattral M. Vascular reconstruction in 105 consecutive pancreas transplants. Transplant Proc 1993; 25: 3288–3289. MacMillan N, Fernandez-Cruz L, Ricart MJ et al. Venous graft thrombosis in clinical pancreas transplantation: Options for a rescue treatment. Transplant Proc 1998; 30: 425–426. Ozaki CF, Stratta RJ, Taylor RJ, Langnas AN, Bynon JS, Shaw BW Jr. Surgical complications in solitary pancreas and combined pancreas-kidney transplantations. Am J Surg 1992; 164: 546–551. Douzdjian V, Abecassis MM, Cooper JL, Smith JL, Corry RJ. Incidence, management and significance of surgical complications after pancreatic transplantation. Surg Gynecol Obstet 1993; 177: 451–456. Fernandez-Cruz L, Gilabert R, Sabater L. Pancreas graft thrombosis: Prompt diagnosis and immediate thrombectomy or retransplantation. Clin Transplant 1993; 7: 230–234. Fiedor P, Rezulski J, Wierzbicki Z et al. Study of the arterial blood supply of the pancreas harvested for transplantation. Folia Morphol (Warsz.) 1989; 48: 123–128. Hollinger EF, Powelson JA, Mangus RS et al. Immediate retransplantation for pancreas allograft thrombosis. Am J Transplant 2009; 9: 740–745.
851
