|Home | About | Journals | Submit | Contact Us | Français|
Autoimmune pancreatitis (AIP) often presents with a swollen duodenal papilla, however, the clinical significance of the duodenal papilla in AIP has not been fully elucidated. Data have shown swollen duodenal papillae shaped like a pear and/or with a submucosal tumor having IgG4-bearing plasma cells. Immunohistopathology has potentially verified duodenal papillitis associated with AIP. FOXP3-positive lymphocytes are also recognized in AIP. AIP has shown spontaneous remission and relapse irrelevance to corticosteroid therapy. The results of a multivariate analysis revealed the absence of a swollen duodenal papilla as the only significant independent factor predictive of spontaneous remission in AIP cases. In addition, the results of another multivariate analysis revealed the presence of a swollen duodenal papilla and the presence of extrapancreatic lesions as the significant independent factors predictive of relapse in these cases. Results suggest that the lack of a swollen duodenal papilla is a predictive factor for spontaneous remission, and thus negates the need to administer corticosteroids in those AIP patients. In contrast, a swollen duodenal papilla and the presence of extrapancreatic lesions are risk factors for relapse, and those AIP patients are candidates for maintenance corticosteroid therapy to reduce relapse. Therefore, the therapeutic strategy such as the indication for corticosteroid administration is subject to the endoscopic features of the duodenal papilla.
Useful evidence for the diagnostic criteria and therapeutic strategy for autoimmune pancreatitis (AIP) has been accumulating [Otsuki et al. 2008; Chari et al. 2006]. A swollen duodenal papilla has often been recognized in AIP [Unno et al. 2002]. However, interpretation of duodenal papillitis associated with AIP has been challenging. Rarely, the diagnosis has proved difficult when IgG4-seronegative findings have been observed and/or atypical imaging findings are present such as focal pancreas swelling mimicking pancreatic cancer (PC) [Woo et al. 2008; Nakazawa et al. 2007; Hardacre et al. 2003]. In such situations, only histopathological evidence has a diagnostic value for AIP. It is essential to obtain histopathological materials from the pancreas. However, it is sometimes difficult to retrieve sufficient material to enable a diagnosis of AIP [Zamboni et al. 2004]. This is partly because taking pancreatic tissue is difficult and is associated with the risk of complications, and partly because these specimens does not always present typical AIP pathologic features due to the small sample size [Bang et al. 2008]. Duodenal papilla findings reflect pancreaticobiliary diseases [Dimango et al. 1982]. Compared with the pancreatic biopsy material, the specimens taken from the duodenal papilla can be retrieved in an easy, safe and reliable way [Kubota et al. 2007a]. Data have suggested that a swollen duodenal papilla with positivity for IgG4 immunostaining was useful in both the diagnosis [Kamisawa et al. 2006] and prognosis of AIP [Kubota et al. 2007b]. In this review, we present the results of our study, the current understanding of duodenal papillitis related to AIP and the problems to be solved in the future.
Many AIP patients have undergone pancreas resection following a misdiagnosis of PC and/or bile duct cancer [Abraham et al. 2000]. Awareness of AIP is now much more widespread. However, it is regarded as a systemic disease involved with multiorgan systems such as sclerosing cholangitis (SC), sclerosing sialadenitis and retroperitoneal fibrosis [Kamisawa et al. 2003]. Sarles et al. have described pancreatitis diagnosed on the basis of the immune mechanism [Sarles et al. 1961]. They reported cases with noncalcifying chronic pancreatitis showing onset with obstructive jaundice, emaciation, hypergammaglobulinemia and pathological findings with massive infiltration of inflammatory cells in the pancreas. They also identified the self-immunization phenomenon. Subsequently, with only a small number of case reports on AIP having been published, the concept of AIP was defined [Yoshida et al. 1995]. The Japanese Pancreas Society (JPS) put together the first diagnostic criteria in the world in 2002 [Members of the criteria committee for autoimmune pancreatitis of the Japan Pancreas Society, 2002] and revised them in 2006 [Members of the criteria committee for autoimmune pancreatitis of the Japan Pancreas Society, 2006]. The JPS criteria weight the imaging and serological criteria, whereas the histology, imaging, serology, other organ involvement and response to therapy (HISORt) criteria proposed in the US can diagnose AIP only by histopathological findings [Chari et al. 2006]. However, the fundamental mechanism has not yet been proven. Hamano and colleagues described how serum IgG4 is a specific and highly sensitive marker of AIP [Hamano et al. 2001]. Following this, two types of AIP were recognized based on the histopathological findings: lymphoplasmacytic sclerosing pancreatitis (LPSP) and idiopathic centric pancreatitis (IDCP). The former was first described as LPSP by Kawaguchi and colleagues in 1991 [Kawaguchi et al. 1991] and IDCP was first reported by Notohara et al. [Notohara et al. 2003]. Both types of AIP responded well to corticosteroids, and the JPS regarded the LPSP and IDCP subsets of AIP as another type of pancreatitis. Characteristic features of AIP include spontaneous remission, and also relapse even after corticosteroid therapy has been administered [Kubota et al. 2007b; Hirano et al. 2007; Wakabayashi et al. 2005]. Also, the JPS criteria emphasized the unique type of pancreatitis characterized by diffuse narrowing of the main pancreatic duct (MPD) on endoscopic retrograde cholangio-pancreatography (ERCP). As for the therapeutic strategy, Kamisawa and colleagues described a therapeutic strategy based on the data collected from major Japanese institutes [Kamisawa et al. 2009].
The diagnostic criteria and therapeutic strategy are now well established. Atypical AIP cases were sometimes recognized as having a focal mass together with IgG4-seronegative findings [Kubota et al. 2007a]. Instead of obtaining pancreatic biopsy specimens, the usefulness of the endoscopic features and checking the reactivity of biopsy specimens to IgG4 and/or FOXP3 have been cited [Kubota et al. 2009a]. The features of duodenal papilla associated with AIP have been described, and we believe that they provide a safe, useful and reliable method to obtain and investigate biopsy materials in the diagnosis of AIP.
The pathogenesis of AIP remains unknown, however, there are several clues. An association of the HLA haplotype DRB1 *0405-DQB1*0401 with AIP [Kawa et al. 2002], antilactoferrin antibodies (anti-LF) mainly located in the pancreatic acini [Uchida et al. 2002], anti-carbonic anhydrase II antibodies (anti-CA-II) chiefly located in the duct cells [Aparisi et al. 2005; Uchida et al. 2002], mediation by the Th1/Th2-type immune response [Okazaki et al. 2000] and upregulation both of the Th2 cytokines and regulatory cytokines (interleukin-10 [IL-10] and tumor necrosis factor β) [Zen et al. 2007] might play an important role in pathogenesis of AIP. Cytotoxic T lymphocytes-associated antigen 4 (CTLA-4) was also reported as a susceptibility factor for AIP [Chang et al. 2007]. Microbes to break down immune tolerance have been one of the plausible mechanisms. Recently, a novel antibody, plasminogen-binding protein (PBP), of Helicobacter pylori was detected in most patients with AIP. As the PBP peptide is homologous to the human protein ubiquitin-protein ligase E3 component n-recognin 2, an enzyme highly expressed in the acinar cells of the pancreas, this suggests that pancreatic acinar cells but not ductal cell may be the target of the autoimmune activity in patients with AIP [Frulloni et al. 2009].
Up to the present, there have been seven articles regarding the association of AIP with immune-mediated inflammation of the duodenal papilla. Unno and colleagues. were the first to study the descriptions of the appearance of the main duodenal papilla in SC [Unno et al. 2002]. They stated that a swollen duodenal papilla was a characteristic finding in patients with sclerosing pancreatitis, and T-lymphocyte infiltration using CD3, CD4, CD8 and L26 antigens was presented in the same paper. The authors concluded that these features might be useful in the diagnosis of sclerosing pancreatitis. This is the flagship article regarding the use of duodenal papilla examination in the diagnosis of AIP. A case report on the association of AIP with immune-mediated inflammation of the papilla of Vater has been published [Sahin et al. 2004]. The authors additionally reported a study of HLA-DR-antigen expression regarding the duodenal papilla. Hisa and colleagues presented an interesting case with lymphoplasmacytic granuloma localized only in the duodenal papilla which was positively immunostained for IgG4 and originated from the submucosa [Hisa et al. 2008]. This case indicated that duodenal papillitis might be one of the other organ involvement in patient of AIP. The usefulness of a biopsy of the major duodenal papilla to diagnose AIP was reported by utilizing IgG4 immunostaining and the authors concluded that IgG4 immunostaining of biopsy specimens taken from the major duodenal papilla might support the diagnosis of AIP [Kamisawa et al. 2006]. We confirmed the findings of the previous study using 17 AIP patients with chronic alcoholic pancreatitis (CAP) and reported that a swollen duodenal papilla was recognized in 64.7% of AIP patients [Kubota et al. 2007a]. We also studied clinicopathological factors predictive of spontaneous remission or relapse in a small number of cases of AIP (n=20) and revealed that negative IgG4 immunostaining of the duodenal papilla appeared to predict a high frequency of remission without steroid therapy [Kubota et al. 2007b]. This article is the first to present the significance of the duodenal papilla findings contributing to the prognosis of AIP. As a diagnostic endoscopic tool for AIP, IgG4 immunostaining of biopsy specimens obtained from the duodenal papilla might be useful to support a diagnosis of AIP with pancreatic head involvement [Kamisawa et al. 2008]. Differentiating primary SC (PSC) and SC-AIP is challenging. We clarified the endoscopic findings of the duodenal papilla in patients with SC-AIP and those with PSC. We found characteristic duodenal papilla features such as a swollen duodenal papilla that was positive for IgG4 immunostaining; however, PSC patients tended to have a small duodenal papilla [Kubota et al. 2008]. We presented the significance of FOXP3+ Treg and IgG4 expression in the duodenal papilla in patients with AIP [Kubota et al. 2009a]. We also indicated that the features of the duodenal papilla could not only influence the diagnosis of AIP but could have a prognostic value. We believe that the therapeutic strategy can be subject to the features of the duodenal papilla [Kubota et al. 2009b].
Duodenal papillitis associated with AIP was recognized not only macroscopically in swollen duodenal papillae detected by endoscopy, but also microscopically in duodenal papillae which were not swollen, as mentioned previously [Kamisawa et al. 2006; Sahin et al. 2004]. An expanding duodenal papilla with redness can be recognized as ordinary papillitis due to inflammation associated with gallstones and/or pancreatitis (Figure 1). Regarding the endoscopic features of duodenal papilla in patients with PC, we also studied using IgG4 immunostaining in eight patients with PC (not published data) and that there were no special findings on it as Kamisawa and coworkers reported [Kamisawa et al. 2008]. Tumorous swollen papillae with or without erosion on the surface have been recognized in duodenal papillary neoplasms (Figure 2) [Park et al. 2002; Yamaguchi and Enjoji, 1987]. Characteristic endoscopic features of AIP, such as a swollen duodenal papilla, were often overlooked until Unno and colleagues. described their usefulness [Unno et al. 2002]. They defined distinctive duodenal findings associated with AIP when an indistinct border existed between the papilla and the orad protrusion (Figure 3). We recognized that the swollen duodenal papilla tended to have dilated capillary veins on their surface. Therefore, we can diagnose AIP based on a swollen duodenal papilla if the endoscopic features show a swollen duodenal papilla with an indistinct border between the papilla and the orad protrusion with dilated papillary veins sometimes present on the surface. Swollen duodenal papillae have been recognized in some AIP patients but not in others: however, the significance of a swollen duodenal papilla in the diagnosis and/or prognosis of AIP remains to be clarified. Duodenal papilla findings in AIP patients should also be regarded as a supplementary diagnosis of AIP because it was positively reported in 41–59% AIP cases as seen in Table 1 [Kim et al. 2009; Kubota et al. 2007a]. We also reported that the presence of a swollen duodenal papilla is an independent factor for relapse of AIP [Kubota et al. 2007b]. To investigate the characteristic findings of duodenal papillae, a part of the duodenal papilla can be visualized endoscopically by side-view endoscopy which is often used in ERCP. We usually check and photograph the frontal view and the images are obtained while looking down at the papilla from the upper part of the duodenum before cannulation and at least two photographs of each papilla are obtained [Kubota et al. 2007a]. Endoscopic papillary biopsy is an easy and safe method of obtaining histopathological specimens from the duodenal papilla [Kubota et al. 2007a]. We recommended taking specimens from just below the pancreatic duct orifice of the duodenal papilla. Up to now, we have photographed 44 AIP cases and performed an endoscopic papillary biopsy in 32 AIP cases. We have never experienced postsampling pancreatitis and/or bleeding of AIP patients who have undergone sampling from the duodenal papilla. Table 2 summarizes the differences in the endoscopic findings of duodenal papillae among AIP, ordinary papillitis (cases with PC included) and neoplasm. The important points for the differentiation of AIP, ordinary papillitis and a neoplasm are the shape, the surface of the duodenal papilla, the presence of erosion and/or ulcers on the surface, the presence of an indistinct border, the color of the duodenal papilla, the presence of dilated capillary veins and the presence of spontaneous bleeding. In short, the swollen duodenal papilla associated with AIP tend to appear like a submucosal tumor, have a pear-like shape without any erosion/ulcer on the surface, with an indistinct border, a normal color, with dilated capillary veins and without spontaneous bleeding.
The reasons why duodenal papillae become swollen in duodenal papillitis associated with AIP remain unknown. We speculate that duodenal papillitis represents the inflammation of the pancreaticobiliary condition of AIP. We studied duodenal papillitis from the points of the subsets of lymphocytes such as AIP, compared with ordinary papillitis due to alcoholic pancreatitis and/or gallstones [Kubota et al. 2007a]. We identified histopathologically that mild to moderate lymphocytic infiltration was present in the stroma of the duodenal papilla in AIP cases (Figure 4). No dense fibrosis was detected in the duodenal papilla. There were numerous stromal inflammatory cells infiltrating (SICI) the region just beneath the duodenal papilla in both patients with AIP and those with CAP. Marked infiltration was particularly observed in swollen duodenal papillae in patients with AIP. We assessed the SICI count in AIP patients and found that the average number of cells in SICI per high-power field (HPF) in AIP patients was significantly higher than that in the patients with CAP (p<0.05). Unno and colleagues reported that the cellular infiltrate in the duodenal papilla consisted predominantly of CD3-positive T lymphocytes and IgG4-positive plasma cells [Unno et al. 2002]. It was found that T-cell infiltration tended to be pronounced in the duodenal papillae of AIP patients compared with CAP patients. We used CD3 as the T-cell marker (Figure 5) and 79a as the B-cell marker (Figure 6) to compare cell subtypes in AIP patients. With regards to the cases with a swollen duodenal papilla, more pronounced T-cell infiltration tended to be recognized in patients with a swollen duodenal papilla than in those without. Furthermore, more pronounced T-cell infiltration of the duodenal papilla was also recognized in the patients with immunopositivity for IgG4 than in those without. Eosinophils tended to be recognized in AIP patients. In summary, T-cell dominant lymphoplasmacytic infiltration of the duodenal papilla tended to be recognized in AIP patients. Table 3 summarizes the histopathological differences in biopsy specimens obtained from duodenal papillae in AIP, ordinary papillitis and a neoplasm.
Sahin and colleagues and Kamisawa and colleagues have reported the usefulness of an endoscopic biopsy of the duodenal papilla with immunostaining in AIP patients [Kamisawa et al. 2006; Sahin et al. 2004]. Sahin and coworkers demonstrated that immune-mediated inflammation of the main duodenal papilla might be associated with AIP because HLA-DR-antigen expression was recognized in duodenal papillary epithelial cells [Sahin et al. 2004]. Positivity for serum IgG4 is significantly high in AIP [Hamano et al. 2001]; however, IL-10 induced IgG4 did not include any proliferative aspect induced by TGF-β. In addition to IgG4, the forkhead/winged helix family of transcription factor P3 regulatory T cells (FOXP3+ Treg), a master gene of regulatory T cells (Treg), is expressed in patients with AIP [Zen et al. 2007]. However, its local expression and value in biopsy specimens remains unknown. Even if the macroscopic findings of a swollen duodenal papilla were not present, we could demonstrate it microscopically with biopsy specimens taken from the duodenal papilla. We reported that IgG4 and FOXP3 immunostaining was beneficial to the diagnosis of AIP [Kubota et al. 2009a]. The results of immunostaining are shown in Table 4. Kamisawa and colleagues revealed that the results of endoscopic biopsy using IgG4 immunostaining was useful in the differential diagnosis between patients with AIP and PC [Kamisawa et al. 2008]. We hypothesized that FOXP3 showed more comprehensive positivity in the diagnosis of AIP than that of IgG4. Therefore, we analyzed the endoscopic features and immunohistopathological findings using FOXP3 and IgG4 in biopsy specimens obtained from duodenal papillae in patients with AIP to determine their clinical significance as useful adjuncts for the differential diagnosis in potential AIP patients with a swollen duodenal papilla [Kubota et al. 2009a]. As for FOXP3+ Treg expression in the duodenal papilla (Figure 7), significantly higher levels were detected in AIP than in ordinary papillitis, similarly to IgG4 expression (Figure 8) (p<0.05 for both). Those patients in whom the FOXP3+ Treg and IgG4 expression rates normalized showed a favorable clinical course more than 2 years after corticosteroid therapy. These expression levels might also have therapeutic value for patients with AIP. In summary, the result of FOXP3+ and IgG4 upregulation in the main duodenal papilla improves the sensitivity and specificity in the diagnosis of AIP.
The diagnosis of SC, which can be caused by AIP and/or PSC, can be difficult. AIP is often complicated with SC (SC-AIP), which Nakazawa and coworkers classified using cholangiography to correctly diagnose AIP [Nakazawa et al. 2006]. According to their classification, type 1 occurs only in the lower parts of the common bile duct and can mimic PC. Type 2 can be misdiagnosed as PSC, and types 3 and 4 are easily mistaken for cholangiocarcinoma. Awareness of this classification could diminish the misdiagnosis of AIP; nevertheless, some cases remain difficult to diagnose [Nakazawa et al. 2007], and unnecessary surgery sometimes occurs. The endoscopic differentiation of these two conditions has not yet been fully clarified. We clarified the endoscopic findings of the duodenal papilla in patients with SC-AIP and those with PSC and determined criteria for the differentiation of these conditions. We reviewed these records to determine whether the duodenal papillary findings (swollen papilla/normal papilla/small papilla) might be potentially useful for differentiating SC-AIP from PSC. Immunohistopathological findings for the duodenal papilla were also examined using IgG4 among the infiltrating lymphocytes. The main outcome measurements were as follows: the presence of a swollen duodenal papilla with IgG4-positive plasma cells was useful for discriminating between SC-AIP and PSC. A swollen duodenal papilla was observed in 70.4% of the SC-AIP patients, whereas no swelling of the duodenal papilla was seen in the PSC patients. A small papilla was recognized in 41.7% of the PSC patients. IgG4-positive plasma cells in the duodenal papilla were significantly detected in the SC-AIP patients but not in the PSC patients. Characteristic duodenal endoscopic papillary features in SC-AIP patients, such as a swollen duodenal papilla and positive immunostaining for IgG4, might be helpful for discriminating between this condition and PSC.
The correct timing for terminating corticosteroid therapy remains unknown because inappropriate or a lack of corticosteroid therapy can cause relapse in AIP [Hirano et al. 2007]. The effect of corticosteroid therapy on the immunopositivity using FOXP3 and IgG4 of the duodenal papilla were reviewed. A tendency towards a marked decrease in the stainability of FOXP3-positive lymphocytes and IgG4-positive plasma cells in the duodenal papilla after corticosteroid administration as compared with that before the therapy was observed in each case [Kubota et al. 2009a]. AIP cases showed a relapse rate of around 20–45% [Kamisawa et al. 2009; Hirano et al. 2007; Kubota et al. 2007b]. The need to maintain steroid therapy to reduce relapse for AIP was indicated. Predictive factors including duodenal papillary features have been cited [Kubota et al. 2007b]. Regarding the potential therapeutic value for AIP, the comparison before and after steroid therapy has not been reviewed. We showed that the FOXP3-positive lymphocyte count diminished with corticosteroid therapy. We believe that FOXP3 in the duodenal papilla also has therapeutic value for AIP.
AIP is considered to have a favorable prognosis, as AIP patients usually respond well to corticosteroid therapy [Erkelens et al. 1999]. First-line standard therapy for patients with AIP is corticosteroids. Based on the analysis of 563 AIP patients nationwide, first induction in AIP patients can be achieved with prednisolone (PSL) at a dose of 30mg/day for 2–4 weeks. After that, the dose of PSL is tapered gradually depending on the improvement of serum data and/or imaging findings. Once remission took place, maintenance therapy with PSL at a dose of 5–7.5mg/day follows for up to 3 years to reduce relapse [Kamisawa et al. 2009]. However, some patients with AIP show spontaneous remission [Kubota et al. 2007b; Wakabayashi et al. 2005], while others relapse even after corticosteroid therapy [Takayama et al. 2004]. As for relapse, the relapse rate of AIP was reported to be 30–50 % based on the long-term follow-up studies [Kamisawa et al. 2009; Kubota et al. 2007b], in particular within the first 3 years after the initial presentation of AIP. A high serum IgG level, a high serum IgG4 level, the presence of jaundice, diffuse pancreatic swelling, duodenal papilla swelling and the presence of extrapancreatic lesions have been reported to increase the risk of relapse. Hirano and colleagues showed that adequate maintenance therapy is required to reduce the relapse rate [Hirano et al. 2007]. According to the JPS-published standard steroid therapy for AIP, they stressed that the remission rate of steroid-treated AIP was 98%, which was significantly higher than that of patients without steroid therapy including cases of spontaneous remission (74%). Moreover, the relapse rate of AIP patients was significantly lower in those who received steroid therapy (24%) than those not given steroid therapy (42%). In other words, the relapse rate of AIP could be reduced by steroid therapy and AIP cases showing spontaneous remission without relapse may exist [Kamisawa et al. 2009].
The standard steroid therapy for AIP has been proposed, however, the factors related to spontaneous remission and relapse have not been fully evaluated from the viewpoint of the features of endoscopic duodenal papilla. We retrospectively analyzed the clinicopathological features of AIP patients who exhibited spontaneous remission and/or relapse in an attempt to identify the relapse-related risk factors in AIP [Kubota et al. 2009b]. Regarding spontaneous remission, patients who entered remission without corticosteroid therapy had a better outcome without relapse in the study. The results of a univariate analysis revealed significant association of spontaneous remission with serum negativity for IgG (<1800mg/dl), absence of obstructive jaundice, absence of underlying diabetes mellitus, absence of swelling of the duodenal papilla and localized as opposed to diffuse pancreatic swelling (p<0.05). The results of the multivariate analysis revealed the absence of swollen duodenal papilla as the only significant independent factor predictive of spontaneous remission in these cases. On the other hand, according to the results of the univariate analysis, relapse was associated with diffuse pancreatic swelling, the presence of duodenal papilla, the presence of extrapancreatic lesions and a history of corticosteroid therapy (p<0.05), while a multivariate analysis revealed the presence of swollen duodenal papillae and the presence of extra pancreatic lesions as the significant independent factors predictive of relapse in these cases. Our results suggest that the lack of a swollen duodenal papilla in AIP patients is a predictive factor for spontaneous remission, and the administration of corticosteroid in those patients with AIP is not necessary. In contrast, a swollen duodenal papilla and the presence of extrapancreatic lesions are risk factors for relapse, and those AIP patients are candidates for maintenance corticosteroid therapy to reduce relapse. Therefore, the therapeutic strategy such as the administration of a corticosteroid is subject to the features of endoscopic duodenal papilla. First, the indications for corticosteroid therapy should be examined. If the patient does not have jaundice, exhibits only focal pancreatic swelling and has a normal duodenal papilla, corticosteroid therapy may not be necessary. As relapse typically occurs within 2 years following initial treatment and/or diagnosis, patients with an increased risk of relapse, such as those with a high serum IgG level, diffuse pancreas swelling, a swollen duodenal papilla and extrapancreatic lesions, should receive maintenance therapy for more than 6 months and for as long as 3 years. On the other hand, the long-term natural history of AIP patients with spontaneous remission remains unknown, however, spontaneous remission is likely to be recognized in AIP patients without duodenal papilla swelling and/or diffuse pancreas swelling. We therefore do not always administer steroids in such patients; however, those patients are at risk for potential relapse in the future.
Recognition of endoscopic duodenal papilla findings may be useful in AIP diagnosis. However, clinically AIP is recognized as being one of two types, one pathologically demonstrating LPSP, the other IDCP [Park et al. 2009]. A swollen duodenal papilla might not be seen in the case of the IDCP type. A future study is warranted to assess whether our results using the endoscopic features of duodenal papilla as for AIP could be useful for type II AIP. The assessment of duodenal papillary features observed endoscopically and using biopsy specimens is a simple, safe, easy and reliable method to make a supplemental diagnosis of AIP. A swollen duodenal papilla is not always recognized in AIP, however, if it can be recognized, it is the safest way to obtain histopathological materials. In addition to the endoscopic features, immunostaining using FOXP3 and/or IgG4 of duodenal papilla specimens enhanced the supplemental diagnosis of patients with suspected AIP based on the clinical findings, and enabled deciding on the therapeutic regimen. Cytokine identification using specimens from the duodenal papilla is promising, therefore these data have provided with information on which to base the diagnosis of AIP and an indication whether corticosteroid therapy would be appropriate. A further prospective, large-number and multi-center study is needed to verify the usefulness of the duodenal papilla in the diagnosis of AIP.
None of the authors have any disclosure to make.