Although not essential for the intended endocrine functions, the duodenal segment has prompted significant considerations in the development of surgical techniques for whole-pancreas transplantation.1
Early experience with gross and microscopic examination of rejecting organs showed that acute rejection affected both the pancreas and its duodenal cuff.1 In the era when bladder-drainage was the rule, cystoscopic biopsies helped define the features of acute duodenal cuff rejection and determine the degree of concordance with rejection in the pancreas itself. Indeed, although acute rejection could be isolated to the duodenal cuff or the pancreas, more often rejection coincided in both components.2,3 More specifically, the largest study concluded that in 72% of the cases, treatment decisions would have been the same if only one of the components had been biopsied.3
Based on the experience with cystoscopic samples as well as with experimental models, histology of the duodenal cuff was considered a very useful tool, because it could generally reveal the rejection status of the pancreas parenchyma. With the introduction of the percutaneous pancreas biopsy technique and abandonment of bladder drainage, evaluation of the duodenal cuff fell into disfavor and was forgotten.4
The report of Margreiter et al5 published in 2012, brought the duodenal cuff back to the spotlight, now in the context of enteric drainage (proximal duodenojejunostomy or duodenoduodenal anastomosis). In this setting, the duodenal cuff becomes accessible through routine upper endoscopy and can be easily biopsied.5,6 This approach was also adopted by others who use it successfully for surveillance and diagnosis after pancreas transplantation.7,8
Because the utilization of the duodenal cuff as a surrogate relies on the potential concordance or rejection with the pancreas, a retrospective analysis by the Banff pancreas working group was done by Holanda et al.9 Based on 72 concurrent samples, it was concluded that acute T cell–mediated rejection (TCMR) to a large extent occurred concurrently in the duodenal cuff and pancreas, particularly in the higher rejection grades (Figure 1). Interestingly, features of acute rejection in the duodenum were highly predictive of pancreas rejection, but the opposite was not true.
In this issue, Brockmann et al,10 report their experience with 167 endoscopic duodenal biopsies. Adequate tissue was obtained in all instances, and acute rejection was diagnosed in 18% of samples. Subsequent duodenal follow-up biopsies were useful to identify resolution or persistence of the acute rejection process. In the hands of this group, endoscopic duodenal cuff biopsies served for early diagnosis of acute rejection and were also useful to identify other processes, such as CMV infection. The positive conclusions of this study are in stark contradiction with the recent study by Nordheim et al6 who in a side-by-side comparison with concurrent pancreas biopsies did not find significant value in using the duodenal cuff as a surrogate for the diagnosis of pancreas rejection.6
In addition to center-to-center variations that are difficult to assess, 2 issues can be considered in the comparison of these 2 studies.
- (1) Brockmann et al10 routinely used biopsies of the native intestine as negative controls, which would provide significant help in the interpretation of ambiguous inflammatory changes in the grafted intestinal segment. Inherent to the histological organization of the intestine, which normally contains a prominent population of lymphoid cells between the crypts, the diagnosis of intestinal rejection con be problematic and is liable to subjective interpretations. In other words, the 2 studies may have used different thresholds for the diagnosis of acute duodenal rejection. Furthermore, this is a comparison between studies that used 2 different tissues/settings for the diagnosis of rejection.
- (2) Per study design, Nordheim et al6 performed half the number of biopsies per patient in the first year posttransplantation (3, 6, and 52 weeks vs 14, 30, 60, 90, 180, and 365 days), and was limited by sampling issues to 113 biopsies sets (duodenum and pancreas) in 67 patients (1.7 biopsies per patient). In contrast, Brockmann et al did not have the advantage of evaluating concurrent pancreas tissue, but performed 167 biopsies in 27 patients (6.1 biopsies per patient). The intensive surveillance protocol of Brockmann et al, also included biopsies at 2, 3, and 6 months and may have resulted in increased identification of acute rejection episodes favoring the conclusion that that duodenal biopsies are indeed useful for diagnosis.
The study published in this issue provides a strong incentive for the utilization of duodenal biopsies for monitoring pancreas allografts, and help prevent the damages associated with untreated or ongoing acute rejection in the pancreas parenchyma.
Going forward, several diagnostic and logistic issues should be better defined for an ideal utilization of the duodenal biopsy pool. This include, better histological definitions of the features of duodenal rejection, in the specific context of pancreas transplantation. Whereas features of TCMR appear to be relatively well understood, it is not clear what the features of antibody-mediated rejection are, or even if antibody-mediated rejection can be diagnosed in the duodenal cuff. Cases with vascular involvement were problematic for Brockmann et al and are overall poorly defined in intestinal samples. Further attempts toward clarifying the correlation between the duodenum and the pancreas with evaluation of concurrent samples would be helpful.
Lastly, it is necessary to emphasize that there are pathological processes that can only be diagnosed with evaluation of the pancreas proper, including islet disorders, such as type 1 DM recurrence, type 2 diabetes/islet exhaustion, and drug toxicity. Duodenal cuff biopsies can facilitate the follow-up of pancreas transplant recipients, but cannot replace the utilization of pancreas biopsies entirely.
1. Gruessner R, Sutherland D. Transplantation of the Pancreas: History of Pancreas Transplantation. 2004 1st ed. New York, NY:Springer-Verlag.
2. Laftavi MR, Gruessner AC, Bland BJ, et al. Significance of pancreas graft biopsy in detection of rejection. Transplant Proc. 1998;30:642–644.
3. Nakhleh RE, Benedetti E, Gruessner A, et al. Cystoscopic biopsies in pancreaticoduodenal transplantation. Are duodenal biopsies indicative of pancreas dysfunction? Transplantation. 1995;60:541–546.
4. Laftavi MR, Gruessner AC, Bland BJ, et al. Diagnosis of pancreas rejection: cystoscopic transduodenal versus percutaneous computed tomography scan-guided biopsy. Transplantation. 1998;65:528–532.
5. Margreiter C, Aigner F, Resch T, et al. Enteroscopic biopsies in the management of pancreas transplants: a proof of concept study for a novel monitoring tool. Transplantation. 2012;93:207–213.
6. Nordheim E, Horneland R, Aandahl EM, et al. Pancreas transplant rejection episodes are not revealed by biopsies of the donor duodenum in a prospective study with paired biopsies. Am J Transplant. 2018;18:1256–1261.
7. Perosa M, Noujaim H, Ianhez LE, et al. Experience with 53 portal-duodenal drained solitary pancreas transplants. Clin Transplant. 2014;28:198–204.
8. Biglarnia A, Wallgren A, Larsson E, et al. Systematic protocol biopsy of the duodenal graft for surveillance after pancreas transplantation. Transplantation. 2012;9410S85.
9. Holanda D, Drachenberg CB, Minervini MI, et al. Allograft duodenal cuff biopsy as surrogate in evaluation of pancreatic transplant rejection – a multicenter data effort. Transplantation. 2018;102:S447.
10. Brockmann JG, Butt A, Alhussaini H, et al. Protocol duodenal graft biopsies aid pancreas graft surveillance. Transplantation. 2018;103:622–629.