Rejectors and nonrejectors were separately analyzed. Demographic, clinical, and laboratory data are summarized in Table 2. A total of 17 patients were identified with impaired pancreatic graft function and elevated pancreas enzymes and deterioration of renal function such as pain and fever in some cases. Eleven of them demonstrated signs of acute graft rejection in the duodenal biopsy (Table 1) and were treated accordingly. Statistical analysis revealed rejectors to have significantly elevated amylase and lipase compared with nonrejectors (ρ=0.616; P<0.001 and ρ=0.623; P<0.001, respectively, Fig. 3). Furthermore, univariate analysis revealed C-peptide, need for insulin and fever as discriminating factors.
Nevertheless, most centers do not perform percutaneous allograft biopsies as the procedure can be associated with severe complications. CT- or ultrasound-guided biopsies have been reported to be successful in 80% of cases but are followed by a complication rate of 2.8% to 11% (10 – 12). This includes intraabdominal hemorrhage, gross hematuria (in bladder-drained grafts), allograft pancreatitis, exocrine leak, or inadvertent biopsies of other organs. In our series, enteroscopy was successful in 75% of cases, which remained without complications. On the basis of our experience, we would suggest enteroscopic biopsy of the duodenal segment as an alternative to biopsies of the pancreas graft parenchyma. Enteroscopy was performed under general anesthesia because of the fear of dehiscence of the fascial layer but in the majority of cases, this diagnostic procedure can also be done under analgetic sedation. We believe that short general anesthesia is less risky than a percutaneous pancreas allograft biopsy. When combined with the implantation technique described earlier (duodenal anastomosis to the proximal jejunum) enteroscopy seems to be helpful for centers that do not perform percutaneous pancreas allograft biopsy. Rejection diagnosis in those centers is based mainly on pancreatic enzymes, clinical symptoms and-if available-renal allograft function. This is the largest reported series on a novel clinical tool for monitoring enterically drained pancreatic allografts. Our study demonstrates that this technique is safe and permits the graft duodenal segment to be accessed and inspected in the majority of cases. A further advantage of this technique is that in addition to histologic monitoring of the graft, the upper gastrointestinal tract can be inspected at the same time. The results of our ROC analysis are certainly biased because rejection in our patients was suspected from clinical parameters. Apart from protocol biopsies, which were scheduled before discharge from our transplant unit, enteroscopy was indicated in 17 cases for suspected rejection, which was confirmed in 11 cases and treated accordingly. Elevated serum pancreas enzymes, fever, need for insulin, and C-peptide were the parameters significantly correlating with acute allograft rejection. In those cases in which re-enteroscopy was performed, the macroscopic aspect and histology showed complete recovery in all but two cases, in which loss of graft function could not be prevented despite appropriate antirejection therapy. The endoscopic appearance of duodenal rejection ranges from edema and hyperemia in mild cases to granularity and loss of the mucosal vascular pattern or mucosal ulceration in severe cases (Fig. 1) (20).
Because biopsies were taken from the duodenal segment, the question arises whether duodenal histology accurately reflects episodes of ACR in the pancreatic portion of the graft. A few studies have shown that ACR of the pancreas can be reliably diagnosed from the duodenum. In a rat model, duodenal histology accurately reflected the diagnosis of rejection of the pancreas. The authors, however, found that morphologic changes readily respond to treatment in the duodenum but not in the pancreatic portion of the graft. Awareness for this phenomenon is important for the interpretation of duodenal biopsies after antirejection therapy (21). A pig model compared the degree of vascular and interstitial rejection of the duodenal and pancreatic part of the graft and revealed exact correlation between the pancreas and the duodenum in 47% of the cases. If the duodenum showed signs of rejection, the likelihood that the pancreas would also be rejected was 80%. Discordant cases have usually shown a higher degree of rejection in the pancreatic part of the graft. The interstitial architecture of the pancreas and the duodenum differs considerably, which may explain the poorer correlation among interstitial features of rejection (22 – 24). Still, the authors concluded that a positive duodenal biopsy is representative for the pancreas (25 – 27). The same group reported that in recipients of a combined bladder-drained pancreas-kidney transplant, the pancreas and duodenum can be rejected independently of each other and therefore both tissues should be biopsied (28). Most importantly, one group performed core-needle biopsies of the pancreas and duodenum in a series of dysfunctional pancreaticoduodenal allografts with exocrine drainage into the bladder under cystoscopic direction (29). The authors concluded that the histologic changes in the duodenum paralleled those in the pancreas in both rejection and nonrejection allografts. A histologic study of transplanted human duodenal tissue conducted by Nakhleh et al. (30) investigated 15 resected pancreaticoduodenal allografts and concluded that mucosal biopsy may be helpful in diagnosing rejection. Laftavi et al. (31) also emphasized the value of duodenal histology by demonstrating that in two thirds of cases, the duodenum was helpful in decision-making when no adequate pancreatic tissue was obtained. Kuhr et al. (32) even showed complete correlation in all cystoscopically guided allograft biopsies. In the case of a positive duodenal histology, the patient has to be treated accordingly.
The major drawback of this study is its descriptive nature and a validation of the samples is needed. It would certainly be helpful to have biopsies of the pancreatic parenchyma, which could be achieved with an enteroscope that allows transduodenal biopsy of the pancreas graft similar to the transcystoscopic biopsy in bladder-drained grafts.
In Europe, most pancreas transplants are performed as SPK. In the United States, a number of patients undergo pancreas transplant alone or pancreas after kidney transplant after living donor renal transplantation and in these cases renal function cannot be used for immunological monitoring of the pancreas. From our experience, we conclude that in cases of suspected rejection of an enterically drained pancreas allograft enteroscopic biopsy of the graft duodenum is a reliable and safe means of monitoring.
PTx performed at our Department between March 2005 and July 2009 were included in this study. Transplants were performed only after negative cross-matching. Pancreatic grafts were revascularized through a donor iliac Y-graft to the right common iliac artery in an end-to-side fashion and drained venously to the distal caval vein or the superior mesenteric vein. Exocrine drainage was completed as a stapled or hand-sewn side-to-side duodenojejunostomy 25 to 50 cm beyond the flexure of Treitz (Fig. 2). Table 2 shows demographic data of the study population. This is a retrospective analysis of all patients who underwent enteroscopy at our institution after PTx with enteric drainage. Enteroscopy with duodenal biopsy was indicated in cases in which rejection was suspected from clinical parameters or routinely performed before discharge from our transplant unit. All patients gave written informed consent, and the study was approved by the local ethics committee.
Standard induction therapy consisted of antithymocyte globulin, given as a single bolus at triple standard dose or at standard dose over three consecutive days. Mycophenolate mofetil was given at an oral dose of 1 g twice daily. A rapid steroid-tapering regimen was applied, starting with 500 mg methylprednisolone administered intraoperatively to reach a dose of 25 mg at the end of the first postoperative week and further reduction to a daily maintenance dose of 5 mg. Tacrolimus was given to achieve whole blood trough levels of 10 to 12 ng/mL for the first 3 months, 8 to 10 ng/mL for 6 to 12 months, and 6 to 8 ng/mL thereafter. A subset of patients received sirolimus with targeted trough levels of 6 to 10 ng/mL as initial immunosuppression in combination with tacrolimus. ACR was clinically determined from increased serum amylase and lipase, need for exogenous insulin, low C-peptide, impaired renal function, or abdominal pain and fever. Delayed endocrine graft function was defined as need for exogenous insulin to keep blood glucose levels below 150 mg/dL after postoperative day 10, because beyond this day patients were expected to be on regular care status and diet.
Baseline data were prospectively collected and supplemented with retrospectively collected data on complications, graft loss, and death. Collected data on enteroscopies reported macroscopic appearance of the upper gastrointestinal tract, anastomosis, and duodenal segment and histological data from biopsies. Statistical analysis was performed using MS Excel and SPSS (version 17.0, SPSS Inc., Chicago, IL) including a chi-square test and a nonparametric Mann-Whitney U test. The diagnostic accuracy of duodenal histology for detection of pancreas rejection was determined by ROC curve analysis. In a separate analysis of serum pancreas enzymes, rejectors and nonrejectors were compared using the Spearman rank correlation test. Survival rates were calculated using the Kaplan-Meier method. Data are reported as median with minimum/maximum range, mean±SD or 25%/75% quartile values. A P value less than 0.05 was considered statistically significant.
The authors thank Hugo Bonatti who contributed to write the manuscript and Hanno Ulmer for statistical support.
1. Heilman RL, Mazur MJ, Reddy KS. Immunosuppression in simultaneous pancreas-kidney transplantation: Progress to date. Drugs 2010; 70: 793.
2. White SA, Shaw JA, Sutherland DE. Pancreas transplantation
. Lancet 2009; 373: 1808.
3. Sollinger HW, Odorico JS, Becker YT, et al.. One thousand simultaneous pancreas-kidney transplants at a single center with 22-year follow-up. Ann Surg 2009; 250: 618.
4. Klima G, Margreiter R, Königsrainer A, et al.. Pancreatic juice cytology after pancreas allotransplantation. Transplant Proc 1988; 20: 167.
5. Kubota K, Reinholt FP, Tyden G, et al.. Pancreatic juice cytology for monitoring pancreatic grafts in the early postoperative period. Transpl Int 1992; 5: 133.
6. Arjona-Sanchez A, Munos-Cesares FC, Ruiz-Rabelo J, et al.. Consolidation of enteric drainage for exocrine secretions in simultaneous pancreas-kidney transplant. Transplant Proc 2010; 42: 1815.
7. West M, Gruessner RW. Laparoscopic biopsy
after pancreaticoduodenal transplantation. A first report. Transplantation 1996; 62: 1684.
8. Allen TG, Wilson TG, Grierson JM, et al.. Percutaneous pancreas transplant fine needle aspiration and needle core biopsies are useful and safe. Transplant Proc 1990; 22: 663.
9. Aideyan OA, Schmidt AJ, Trenkner SW, et al.. CT-guided percutaneous biopsy
of pancreas transplants. Radiology 1996; 201: 825.
10. Klassen DK, Weir MR, Cangro CB, et al.. Pancreas allograft biopsy
: Safety of percutaneous biopsy
-results of a large experience. Transplantation 2002; 73: 553.
11. Atwell TD, Gorman B, Larson TS, et al.. Pancreas transplants: Experience with 232 percutaneous US-guided biopsy
procedures in 88 patients. Radiology 2004; 231: 845.
12. Gaber LW. Pancreas allograft biopsies in the management of pancreas transplant recipients: Histopathologic review and clinical correlations. Arch Pathol Lab Med 2007; 131: 1192.
13. Boggi U, Vistoli F, Signori S, et al.. A technique for retroperitoneal pancreas transplantation
with portal-enteric drainage. Transplantation 2005; 79: 1137.
14. Boggi U, Amorese G, Marchetti P. Surgical techniques for pancreas transplantation
. Curr Opin Organ Transplant 2010; 15: 102.
15. Rudroff C, Scheele J. Pankreastransplantation: Einen neue chirurgische Methode mit endoskopisch zugänglicher enteralen Drainage. Transplantationsmedizin Suppl 1998: 28.
16. Genzini AC, Dib R, Takahashi W, et al.. Balloon enteroscopy
for diagnosis and treatment of cytomegalovirus-induced small bowel gastrointestinal bleeding after whole-organ pancreas transplantation
. Am J Gastroenterol 2010; 105: 235.
17. De Melo SW, Gupta S, Arenas J, et al.. Single-balloon enteroscopy
-guided hemostasis of an anastomotic ulcer in a patient with simultaneous enteric-drained pancreas-kidney transplant. Endoscopy 2008; 40(suppl 2): E164.
18. Durlik M, Kosmala W, Milewski J, et al.. Feasibility of visualization and biopsy
of donor duodenum by double-balloon enteroscopy
technique in a recipient of simultaneous enteric-drained pancreas-kidney transplant: Case report. Transplantation 2006; 82: 578.
19. Margreiter C, Mark W, Wiedemann D, et al.. Pancreatic graft survival despite partial vascular graft thrombosis due to splenocephalic anastomoses. Am J Transplant 2010; 10: 846.
20. Wu T, Abu-Elmagd K, Bond G, Nalesnik MA, Randhawa P, Demetris AJ. A schema for histologic grading of small intestine allograft acute rejection
. Transplantation 2003; 75: 1241.
21. Mark W, Hechenleitner P, Dietze O, et al.. Duodenal histology for monitoring treatment of acute rejection
in pancreaticoduodenal allografts in rats. Transplantation 2002; 73: 198.
22. Oberhuber G, Schmid T, Thaler W, et al.. The pattern of rejection
after combined stomach, small bowel, and pancreas transplantation
in the rat. Transpl Int 1993; 6: 296.
23. Steiniger B, Klempnauer J, Wonigeit K. Altered distribution of class I and class II MHC antigens during acute pancreas allograft rejection
in the rat. Transplantation 1985; 40: 234.
24. Yamamoto S, Ito T, Nakata S, et al.. The rejection
mechanism of rat pancreaticoduodenal allografts with a class I MHC disparity. Transplantation 1994; 57: 1217.
25. Nakhleh RE, Sutherland DE, Tzardis P, et al.. Correlation of rejection
of the duodenum with rejection
of the pancreas in a pig model of pancreaticoduodenal transplantation. Transplantation 1993; 56: 1353.
26. Nakhleh RE, Sutherland DE, Benedetti E, et al.. Diagnostic utility and correlation of duodenal and pancreas biopsy
tissue in pancreaticoduodenal transplants with emphasis on therapeutic use. Transplant Proc 1995; 27: 1327.
27. Gruessner RW, Nakhleh R, Tzardis PJ, et al.. Correlation between duodenal and kidney rejection
: A histologic comparative study in a pig model of pancreaticoduodenal-kidney transplantation. Transplant Proc 1994; 26: 541.
28. Nakhleh RE, Benedetti E, Gruessner A, et al.. Cystoscopic biopsies in pancreaticoduodenal transplantation. Are duodenal biopsies indicative of pancreas dysfunction? Transplantation 1995; 60: 541.
29. Carpenter HA, Engen D, Munn S, et al.. Histologic diagnosis of rejection
by using cystoscopically directed needle biopsy
specimens from dysfunctional pancreatoduodenal allografts with exocrine drainage into the bladder. Am J Surg Pathol 1990; 14: 837.
30. Nakhleh RE, Gruessner RW, Tzardis PJ, et al.. Pathology of transplanted human duodenal tissue: A histologic study, with comparison to pancreatic pathology in resected pancreaticoduodenal transplants. Clin Transplant 1991; 5: 241.
31. Laftavi MR, Gruessner AC, Bland BJ, et al.. Significance of pancreas graft biopsy
in detection of rejection
. Transplant Proc 1998; 30: 642.
32. Kuhr CS, Davis CL, Barr D, et al.. Use of ultrasound and cystoscopically guided pancreatic allograft biopsies and transabdominal renal allograft biopsies: Safety and efficacy in kidney-pancreas transplant recipients. J Urol 1995; 153: 316.
33. Hawthorne WJ, Allen RD, Greenberg ML, et al.. Simultaneous pancreas and kidney transplant rejection
: Separate or synchronous events? Transplantation 1997; 63: 352.
34. Solez K, Colvin RB, Racusen LC, et al.. Banff '07 classification of renal allograft pathology: Updates and future directions. Am J Transplant 2008; 8: 753.