During the last decades, several surgical and therapeutic advances in pancreas transplantation have improved both patient and graft survival [1, 2]. However, graft rejection is one of the main causes of graft failure (25% of the grafts) [3] and remains a diagnostic challenge. Clinical manifestations and laboratory markers are non-specific [4, 5]. Although imaging studies are key to rule out many other causes of graft dysfunction (graft pancreatitis, vascular events, relapse of type 2 diabetes mellitus …) [6–8], they do not provide specific findings for diagnosing graft rejection. Pancreatic graft biopsy is therefore the gold standard for diagnosing graft rejection. Histological evaluation provides additional information, distinguishing cellular from antibody-mediated rejection, grading its severity and excluding other causes of dysfunction.

There are different techniques for obtaining pancreatic graft samples. Laparoscopic access offers a good success rate for histological diagnosis (around 95%) [9] but it is more expensive, less available, and has been reported to have a 2.5% conversion to laparotomy [9, 10]. The endoscopic approach is less invasive but it is a complex technique only performed in some centers [11–13], with low accessibility rates [13] and a low success rate for histopathological diagnosis (50%) [11, 12]. Besides, it mainly uses sampling of the graft duodenal mucosa (instead of sampling the pancreatic graft), whose utility in diagnosing graft rejection is still under debate [14–16]. Percutaneous access is the most widely used technique because it has been demonstrated to be safe and effective for the classical intraperitoneal positioning of the graft, whether guided by ultrasound (US) [17, 18] or computed tomography (CT) [19]. In addition, it is a simple and cheap technique that does not require sedation or an operating room, thereby contributing to decreased costs and occurrence of comorbidities.

Classically, pancreatic exocrine secretions were derived to the jejunum through a duodeno-jejunostomy (DJ), placing the graft in an intraperitoneal position (Figure 1A). Recently, an alternative technique has emerged, in which the graft is placed retrocolically through a duodeno-duodenostomy (DD, Figure 1B), to mimic the physiology of the exocrine secretion in the native pancreas. In addition, it can improve the feasibility to reach the anastomotic sites [20, 21] and access to endoscopic procedures [22]. Despite the potential physiological and surgical benefits of this technique, some authors have pointed out that the retrocolic position of the pancreatic graft could limit the accessibility of percutaneous biopsy [20, 21]. Although accessibility for percutaneous graft biopsy after this surgical technique is an interesting topic, the recommendations of the first World Consensus Conference on pancreas transplantation published in 2021 [23], point out that percutaneous biopsy accessibility to retroperitoneally placed grafts still remains to be proven.

DJ intraperitoneal graft has been demonstrated to be accessible to percutaneous biopsy. To date, there are no reports on the accessibility to percutaneous biopsy on DD retrocolic grafts.

The aim of this study was to evaluate graft accessibility, the success rate for histological diagnosis and the safety of percutaneous biopsy of pancreatic grafts placed using the DD technique for enteric drainage. Furthermore, we compared these results with those obtained previously in intraperitoneal DJ grafts.

We received a total of 268 biopsy requests in 145 patients (83 in the DJ group, 60 in the DD group and two patients with a first DJ graft and a retransplantation with a DD graft). Patient characteristics are summarized in Table 1 and data related to the biopsies are summarized in Table 2.

This study demonstrates that retrocolic pancreatic grafts placed using the DD technique are accessible to percutaneous biopsy with an accessibility rate higher than 85%. The success rate for histological diagnosis was 87.9%, which is similar to that reported for percutaneous biopsies in grafts placed with the classical surgical techniques [17, 18, 27–29]. Therefore, accessibility for a subsequent biopsy should not be a limitation to implementing the novel DD technique.

Until 2016, the intestinal drainage in our center was performed with DJ, performed side-to-side to the jejunum, 70–80 cm from the ligament of Treitz, with good results: the incidence of intestinal complications of this DJ technique from 2000 to 2016 (337 pancreas transplants) published for our group was 6.8% [30]. From this date, the DD technique was adopted successfully and with a good level of acceptance by all members of the pancreatic transplant group, with a low rate of complications (initial data published in 2017), with no intestinal complications recorded in the first 10 pancreatic DD grafts [20]. The rationale of using retrocolic graft placement over the intraperitoneal position is the easy access and dissection of the vascular anastomosis site, and easy reconstruction of venous and arterial anastomosis. To be more specific describing the surgical postoperative complications, we published in 2022, the first retrospective single-center study comparing the effects of the four most commonly used preservation solutions in PTx, i.e., UW, CS, HTK, and IGL-1, on early pancreatic graft function as well as long-term patient and graft survival. A total of 43 out of 380 cases were performed using the duodenoduodenostomy, but this fact does not affect immediate reperfusion injury rates, as vascular anastomoses were performed with the same technique throughout the time period in question. When analyzing the surgical complications according Clavien-Dindo grade no statistical differences were found between the DJ and DD groups [31]. Recently, a descriptive review of 407 pancreas transplants performed at our center (1999–2019) by analyzing the type of arterial reconstruction technique and long-term survival were published. The DD was used in 57 patients with three of them presenting with acute arterial thrombosis [24]. Due to these good results, the DD technique is the one used in our center. Initially, it was feared that this technique would limit the percutaneous biopsy accessibility, but the results of the present study demonstrate that grafts placed with the DD technique are accessible for percutaneous biopsy. In fact, in our study, accessibility was even better in the retrocolic-DD group than in the intraperitoneal-DJ group. One of the factors favoring this higher accessibility is the more cranial and posterior position of the DD graft (Figure 1B). This position offers the possibility of performing a lateral or a posterior approach, avoiding the interposition of intestinal loops, which is the main cause of not accessing the graft, both in our study and in previously reported studies, including other surgical techniques [17, 27, 28]. The lower position of DJ grafts limits the posterior and lateral approaches because the iliac bone surrounds the posterior aspect of the graft (Figure 1A). Up to 36/132 patients (27.3%) in the DD group benefited from the lateral or posterior approach (in prone or lateral patient position), thereby increasing graft accessibility from 60.6% to 87.9% in this group.

US has proven to be an excellent technique to guide percutaneous biopsy for DJ grafts [17, 18, 27, 28, 32], and it has some advantages over CT [19], as it is a faster procedure without radiation exposure. Our results demonstrated that US is also an excellent tool for guiding biopsies of retrocolic-DD graft. In our series, 89.7% of the performed biopsies in the DD group were guided by US (Table 2). This differs from the native pancreas, which is also retroperitoneal but is located in the midline position and is not accessible using the posterior approach.

In addition, percutaneous biopsy is a safe technique. Only one minor complication was recorded in the DD group, with a total complication rate of 0.5%, lower than that reported in the literature (2%–3.6%) [17, 18, 27, 28, 32]. One factor that could contribute is the low number of passes performed related to other studies [19, 27, 32]. Aideyan et al. [19], point out that CT-guided biopsy is associated with a higher risk of severe hemorrhage. This could be explained by the static imaging provided by CT, which could lead to the possibility of traversing both sides of the graft with the needle [19]. US-guided biopsy allows continuous control over the needle trajectory, which might favor a lower complication rate. The experience level of the operator could also play a role in reducing complication rates, as all our biopsies were performed by senior radiologists with vast experience in percutaneous biopsies. Another contributing factor could be the needle gauge. In our study, an 18G automatic needle was used in all patients, but no clear relationship between needle gauge and bleeding in pancreatic biopsies has been demonstrated. Lee et al. [32] compared the performances between 18G and 20G needles, reporting only a minor complication (bleeding) in the 20G group. The safety of this technique, as shown in our study, could favor the recommendation of a standardized surveillance biopsy program to detect subclinical rejection for an early treatment.

This study has some limitations. First, the DD technique was implemented in May 2016, thus all biopsies performed in the first part of the inclusion period (November 2009–May 2016) belonged to the DJ group. This means that the learning curve limitations affect only DJ patients. However, their impact on the results was reduced by excluding the biopsies performed during the first year (n = 16), which still showed significant differences in accessibility rates between groups.

Second, some grafts experience atrophy of the gland over time [7, 26, 27], potentially due to multiple episodes of undiagnosed acute rejection that may lead to chronic rejection [8, 33–36], limiting access to percutaneous biopsy. This fact is also supported by our study in which atrophy was a relevant cause for not accessing grafts older than 5 years (Table 3). Due to the implementation of the DD technique in May 2016, the DD group included patients with grafts younger than 5 years (the study ended in July 2021). Although this fact could have contributed to a decrease in the accessibility of the DJ grafts, statistically significant differences remained between groups when analyzing the accessibility rates only for young grafts (<5 years).

Third, the retrospective and monocentric nature of the study, may also be considered a limitation. But, in the scenario of pancreas transplantation with a wide variety of surgical techniques used throughout the world and the fact that it is a minority type of solid organ transplantation, it makes it difficult to carry out a multicenter study. This fact becomes more important if we take into account that the application of pancreatic biopsy in the different centers is in its infancy, both due to the worry of complications and the obvious learning curve that is needed in the context of a minority transplant. To our knowledge, this is the largest series analyzed using two different positions of the pancreatic graft, including a significant number of biopsies performed, not only in clinically indicated cases but also in those performed per protocol; without losing sight of the fact that analysis has also been carried out from an intention to treat point of view. In the absence of a reliable and proven method for the diagnosis of rejection, beyond surrogate blood analytical data, the results of the present study are of vital importance for the scientific community since it offers the possibility of making a very precise histopathological diagnosis to treat subclinical rejection with impact on long-term graft survival.

In conclusion, our results demonstrate that US-guided percutaneous biopsy of retrocolic pancreatic grafts placed by DD is a safe and effective method for the histologic diagnosis of rejection, with an accessibility rate even better than that obtained for intraperitoneal pancreatic grafts. We firmly believe that this is the first step to eliminate fears of associated morbidity to the detriment of the benefits provided, and move towards the worldwide implementation of pancreas graft percutaneous biopsy in real life to improve the outcomes of such a challenging type of transplant.