Advertisement

Expert consensus on endoscopic papillectomy using a Delphi process

Open AccessPublished:April 19, 2021DOI:https://doi.org/10.1016/j.gie.2021.04.009

      Background and Aims

      Consensus regarding an optimal algorithm for endoscopic treatment of papillary adenomas has not been established. We aimed to assess the existing degree of consensus among international experts and develop further concordance by means of a Delphi process.

      Methods

      Fifty-two international experts in the field of endoscopic papillectomy were invited to participate. Data were collected between August and December 2019 using an online survey platform. Three rounds were conducted. Consensus was defined as ≥70% agreement.

      Results

      Sixteen experts (31%) completed the full process, and consensus was achieved on 47 of the final 79 statements (59%). Diagnostic workup should include at least an upper endoscopy using a duodenoscope (100%) and biopsy sampling (94%). There should be selected use of additional abdominal imaging (75%-81%). Patients with (suspected) papillary malignancy or over 1 cm intraductal extension should be referred for surgical resection (76%). To prevent pancreatitis, rectal nonsteroidal anti-inflammatory drugs should be administered before resection (82%) and a pancreatic stent should be placed (100%). A biliary stent is indicated in case of ongoing bleeding from the papillary region (76%) or concerns for a (micro)perforation after resection (88%). Follow-up should be started 3 to 6 months after initial papillectomy and repeated every 6 to 12 months for at least 5 years (75%).

      Conclusions

      This is the first step in developing an international consensus–based algorithm for endoscopic management of papillary adenomas. Surprisingly, in many areas consensus could not be achieved. These aspects should be the focus of future studies.

      Abbreviations:

      CBD (common bile duct), CT (computed tomography), EP (endoscopic papillectomy), EMR (endoscopic mucosal resection), ERCP (endoscopic retrograde cholangiopancreatograpy), EUS (endoscopic ultrasound), FCSEMS (fully covered self-expanding metal stent), IQR (interquartile range), MRCP (magnetic resonance cholangiopancreatography), MRI (magnetic resonance imaging), PD (pancreatic duct), PPP (postpapillectomy pancreatitis), PPI (proton pump inhibitor), RFA (radiofrequency ablation)
      The first endoscopic papillectomy (EP) for papillary adenoma (n = 2) was described in 1983, and the first substantial cohort (n = 25) was published 10 years later.
      • Suzuki K.
      • Kantou U.
      • Murakami Y.
      Two cases with ampullary cancer who underwent endoscopic resection.
      ,
      • Binmoeller K.F.
      • Boaventura S.
      • Ramsperger K.
      • et al.
      Endoscopic snare excision of benign adenomas of the papilla of Vater.
      A number of predominantly retrospective case and cohort studies have been published subsequently.
      • van der Wiel S.E.
      • Poley J.W.
      • Koch A.D.
      • et al.
      Endoscopic resection of advanced ampullary adenomas: a single-center 14-year retrospective cohort study.
      • Klein A.
      • Qi Z.
      • Bahin F.F.
      • et al.
      Outcomes after endoscopic resection of large laterally spreading lesions of the papilla and conventional ampullary adenomas are equivalent.
      • Ridtitid W.
      • Tan D.
      • Schmidt S.E.
      • et al.
      Endoscopic papillectomy: risk factors for incomplete resection and recurrence during long-term follow-up.
      • Napoleon B.
      • Gincul R.
      • Ponchon T.
      • et al.
      Endoscopic papillectomy for early ampullary tumors: long-term results from a large multicenter prospective study.
      • Sahar N.
      • Krishnamoorthi R.
      • Kozarek R.A.
      • et al.
      Long-term outcomes of endoscopic papillectomy for ampullary adenomas.
      Despite the lack of randomized controlled trials and prospective series, EP for papillary adenoma is considered a relatively safe, minimally invasive treatment for lesions without significant intraductal extension or invasive disease.
      • Ridtitid W.
      • Tan D.
      • Schmidt S.E.
      • et al.
      Endoscopic papillectomy: risk factors for incomplete resection and recurrence during long-term follow-up.
      It has proven difficult to generate high-level scientific knowledge on the best treatment algorithm primarily because of low incidence and therefore difficult to conduct large prospective or randomized controlled studies. Subsequently, a consensus for EP practices has not been established. In 2015 the Standards of Practice Committee of the American Society for Gastrointestinal Endoscopy attempted to develop an evidence-based guideline on the role of endoscopy in papillary and duodenal adenomas.
      • Chathadi K.V.
      • Khashab M.A.
      • Acosta R.D.
      • et al.
      ASGE Standards of Practice Committee
      The role of endoscopy in ampullary and duodenal adenomas.
      However, based on the current literature only limited recommendations regarding the optimal diagnostic workup, treatment, and follow-up protocol could be made. Unsurprisingly, at present there remains a wide variety in daily practice, mostly based on individual preferences.
      • El Hajj I.
      • Cote G.A.
      Endoscopic diagnosis and management of ampullary lesions.
      ,
      • Bassan M.
      • Bourke M.
      Endoscopic ampullectomy: a practical guide.
      The aim of this study was to assess the level of consensus among international experts and to obtain further consensus by using a Delphi process. The Delphi method was developed by the Rand Corporation in the 1950s and was originally used in forecasting.
      • Dalkey N.C.
      The Rand Corporation
      The Delphi method: an experimental study of group opinion.
      Since then, the Delphi method is considered to be a reliable instrument to develop clinical consensus guidelines. This iterative structured process is widely used to achieve expert consensus for subjects where no definitive evidence is available and where expert opinion is important.
      • Bennett C.
      • Vakil N.
      • Bergman J.
      • et al.
      Consensus statements for management of Barrett's dysplasia and early-stage esophageal adenocarcinoma, based on a Delphi process.
      • Powell C.
      The Delphi technique: myths and realities.
      • Goggins M.
      • Overbeek K.A.
      • Brand R.
      • et al.
      Management of patients with increased risk for familial pancreatic cancer: updated recommendations from the International Cancer of the Pancreas Screening (CAPS) Consortium.
      The process consists of a series of questionnaires (usually 3), and after each round the responses are summarized and anonymously redistributed for discussion in the next round(s).
      • Hasson F.
      • Keeney S.
      • McKenna H.
      Research guidelines for the Delphi survey technique.
      Accordingly, the ultimate goal is to conduct a proposal for a more standardized treatment protocol on the performance of EP.

      Methods

      Systematic literature search

      A systematic search of literature was performed in PubMed and EMBASE databases on December 18, 2020. The search strategy can be found in Supplementary Figure 1 (available online at www.giejournal.org).

      Expert panel selection

      Researchers were identified through authorship of relevant articles. Senior authors of original articles published in the last 15 years with a cohort of at least 20 patients were selected. Thus, 38 authors were invited. Furthermore, 14 longstanding and internationally recognized experts in the field of EP who did not appear in the search were selected to include broad clinical experience that might not be found in published literature. After the first round, only respondents who performed at least 30 EPs in their career were asked to join the consecutive rounds.

      Conduct of surveys

      Three rounds were conducted. The first survey was based on systematic literature and personal experience in daily practice of the senior authors (M.J. Bourke and R. P. Voermans). The survey consisted of 54 multiple-choice questions divided over 6 different sections: background of the respondent, diagnostic workup, lesion assessment and staging, technical aspects, adverse events and their management, and follow-up. After each question there was an option to add other answer options or to share general comments on the question. The questions of which the answers reached consensus were transformed into statements and presented again to respondents in the next round. Questions without consensus were extracted, modified based on the comments of the respondents, and proposed again in the consecutive round(s). After each round responses were summarized and anonymously redistributed for discussion in the next round(s) to provide respondents the opportunity to review and possibly change their answers based on group consensus.

      Consensus development process

      Data were collected between August and December 2019 using an online survey platform. The invited experts were given at least 2 weeks to complete the survey(s), with reminder e-mails sent twice during each completion period. Participants’ names and contact details were recorded to acknowledge their participation in the eventual article and to be able to ask them to participate in possible follow-up studies. However, participants were unaware of the names of other participants, and results were anonymized.

      Statistical analysis and grading of statements

      Respondents were able to show their level of agreement with the proposed statements by scoring on a 5-point Likert-scale (1 = completely disagree, 2 = disagree, 3 = neutral, 4 = agree, 5 = completely agree) and could comment on their reason to agree or disagree.
      • Likert R.
      Technique for the measurement of attitudes.
      Consensus was defined as at least 70% agreement between the respondents (either (dis)agree or completely (dis)agree). Questions in which participants were asked to rank the answer possibilities were analyzed using median and interquartile range (IQR) in which an IQR ≤1 was considered consensus.
      • Sumsion T.
      The Delphi technique: an adaptive research tool.
      ,
      • von der Gracht H.A.
      Consensus measurement in Delphi studies. Review and implications for future quality assurance.
      Strength of the consensus statements was based on the level of evidence of the supporting literature according to the definitions of the Oxford Centre for Evidence-Based Medicine.
      Centre for Evidence Based Medicine
      Oxford Centre for Evidence-based Medicine—levels of evidence.

      Results

      Systematic review

      After removing duplicates, 827 records were identified. Based on title and abstract screening, 627 records were excluded. Only an abstract was available for 92 records, resulting in 108 full-text articles that were screened for eligibility. The inclusion and exclusion process is summarized in Supplementary Figure 1. The relevant results are described and discussed together with the results of the Delphi process below.

      Participants

      Fifty-two experts were invited. Twenty-eight (53.8% response rate) completed the first round, 17 (32.7% response rate) completed the second round, and 16 joined round 3 (30.7% response rate). Three participants of round 1 were not asked to join consecutive rounds because they had performed fewer than 30 EPs in their career. The 16 final participants included gastroenterologists from 10 different countries and 3 continents (Asia, North America, and Europe). Most participants (15 [94%]) primarily worked in a university hospital setting. All participants performed at least 30 EPs, and 11 (69%) had at least 20 years of experience. A summary of the study process is depicted in Figure 1.
      Figure thumbnail gr1
      Figure 1Flowchart study process. EP; Endoscopic papillectomy.

      Recommendation statements

      In the last round, a final 79 statements were proposed to the participants; the most important consensus statements are summarized in Table 1. Figure 2 depicts a consensus-based flowchart summarizing these statements. All statements that reached consensus are shown in Table 2. A selection of statements that did not reach consensus in the final round is shown in Table 3. The results of all 3 rounds are provided in Supplementary Table 1, Supplementary Table 2, Supplementary Table 3 (available online at www.giejournal.org).
      Table 1Important consensus statements
      StatementAgreementGrading
      Grading: A, level 1a-1b evidence; B, level 2a-3b evidence; C, level 4 evidence; D, level 5 evidence.
      Diagnostic workup
       1. Gastroduodenoscopy with side-viewing instrument should always be performed before resection.100%D
       2. Biopsy sampling should always be performed before resection.94%D
       3. Either MRI/MRCP or EUS should be performed in case of a lesion larger than 2 cm and/or in case of cholestatic laboratory features.75%D
       4. Either MRI/MRCP or CT should be performed in case of significant weight loss and/or in case of endoscopic signs of malignancy.81%D
       5. CT should be performed in case of jaundice.75%D
      Lesion assessment and staging
       6. When a lesion shows ulceration, this lesion should be defined as most likely malignant.94%D
       7. Patient should be referred for surgical management in the following cases, considering patient is suitable for surgery:
      a. Ingrowth in the PD >1 cm.76%D
      b. Ingrowth in the CBD >1 cm.81%D
       8. If there is ingrowth in the CBD >1 cm, endoscopic papillectomy with radiofrequency ablation can be considered in a patient who is not a surgical candidate because of age and/or comorbidity, considering the lesion seems favorable for endoscopic resection.75%C
      Technical aspects
       9. Submucosal injection should only be performed in case of a laterally spreading lesion.88%C
       10. PD stent should be routinely placed to prevent postintervention pancreatitis.100%B
       11. CBD stent should only be placed on indication, namely82%D
      a. If there are concerns for a perforation in the papillary region after resection, a fully covered self-expanding metal stent should be placed in the CBD.88%D
      b. In case of bleeding from the papillary region during the procedure.76%D
       12. Biliary sphincterotomy should be performed in case of concomitant bile duct stones and in case drainage is deemed suboptimal.81%D
      Adverse events and management
       13. Rectal nonsteroidal anti-inflammatory drugs should be given before resection.82%B
      Follow-up
       14. In case initial pathology shows low-grade dysplasia, first follow-up (after removal of possible placed stents) should be performed within 6 months.81%D
       15. In case initial pathology shows high-grade dysplasia, first follow-up (after removal of possible placed stents) should be performed within 3 months.94%D
       16. Follow-up should be performed for at least 5 years.75%D
      CBD, Common bile duct; CT, computed tomography; EUS, endoscopic ultrasound; MRCP, magnetic resonance cholangiopancreatography; MRI, magnetic resonance imaging; PD, pancreatic duct.
      Grading: A, level 1a-1b evidence; B, level 2a-3b evidence; C, level 4 evidence; D, level 5 evidence.
      Figure thumbnail gr2
      Figure 2Consensus-based flowchart. Percentages indicate degree of agreement. CBD, Common bile duct; CT, computed tomography; EUS, endoscopic ultrasound; FU, follow-up; HGD, high-grade dysplasia; LGD, low-grade dysplasia; LSL, laterally spreading lesion; MRCP, magnetic resonance cholangiopancreatography; MRI, magnetic resonance imaging, NSAID, nonsteroidal anti-inflammatory drug; PD, pancreatic duct.
      Table 2Final consensus statements
      StatementAgreementGrading
      Grading: A, level 1a-1b evidence; B, level 2a-3b evidence; C, level 4 evidence; D, level 5 evidence.
      Diagnostic workup
       1. Gastroduodenoscopy with side-viewing instrument should always be performed before resection.100%D
       2. Advanced imaging techniques (such as narrow-band imaging or chromoendoscopy) are not helpful to distinguish between benign and malignant lesions.71%D
       3. Biopsy sampling should always be performed before resection.94%D
       4. Either MRI/MRCP or EUS should be performed in case of cholestatic laboratory features with or without jaundice.81%D
       5. Either CT, MRI/MRCP, or EUS should be performed in case of cholestatic laboratory features with or without jaundice.75%D
       6. CT should be performed in case of jaundice.75%D
       7. Either MRI/MRCP or EUS should be performed in case of a lesion larger than 2 cm.75%D
       8. Either MRI/MRCP or CT should be performed in case of significant weight loss.81%D
       9. Either MRI/MRCP or CT should be performed in case of endoscopic signs of malignancy.81%D
      Lesion assessment and staging
       10. No predefined classification system to determine if a papillary adenoma is most likely benign or malignant exists.89%D
       11. When a lesion shows ulceration, this lesion should be defined as most likely malignant.94%D
       12. The following characteristics are not a sole reason to define the lesion as most likely malignant:
      a. Smooth surface96%D
      b. Spontaneous bleeding86%D
      c. Lesion size >4 cm86%D
       13. Patient should be referred for surgical management in case of ingrowth in the PD >1 cm, considering patient is suitable for surgery.76%D
       14. Patient should be referred for surgical management in case of ingrowth in the CBD >1 cm, considering patient is suitable for surgery.81%D
       15. The following situations are not a sole reason to refer for surgical management:
      a. Jaundice86%D
      b. Ingrowth in the PD ≤1 cm79%D
      c. Ingrowth in the CBD ≤1 cm86%D
      d. An umbilicated lesion82%D
       16. If biopsy sample shows LGD and ulceration is present, the lesion could still be resected endoscopically; there is no need to refer the patient for surgical management based on this sole characteristic, considering the lesion seems favorable for endoscopic resection.88%D
       17. If there is ingrowth in the CBD >1 cm, endoscopic resection can still be considered if the patient is not a surgical candidate because of age and/or comorbidity, considering the lesion seems favorable for endoscopic resection.81%D
       18. If there is ingrowth in the CBD >1 cm, EP with radiofrequency ablation can be considered in a patient that is not a surgical candidate because of age and/or comorbidity, considering the lesion seems favorable for endoscopic resection.75%C
       19. If biopsy sample shows adenocarcinoma in situ or well-differentiated adenocarcinoma, endoscopic resection can still be considered if the patient is not a surgical candidate because of age and/or comorbidity, considering the lesion seems favorable for endoscopic resection.75%D
      Technical aspects
       20. Submucosal injection should only be performed in case of a laterally spreading lesion.88%C
       21. Resection of the lesion should be performed at the plane of the duodenal wall.94%D
       22. EP should be performed with fractionated current.94%D
       23. If pancreatic sphincterotomy is indicated, then it should be performed after resection.88%D
       24. Biliary sphincterotomy should be performed in case of concomitant bile duct stones and in case drainage is deemed suboptimal.81%D
       25. If biliary sphincterotomy is indicated, then it should be performed after resection.100%D
       26. PD stent should be routinely placed to prevent postintervention pancreatitis.100%B
       27. PD should be cannulated after resection.100%D
       28. CBD stent should only be placed on indication, namely82%D
      a. If there are concerns for microperforations in the papillary region after resection.88%D
      b. In case of bleeding from the papillary region during the procedure.76%D
       29. In case there are concerns for microperforations in the papillary region a fully covered self-expanding metal stent should be placed in the CBD.88%D
      Adverse events and management
       30. Rectal nonsteroidal anti-inflammatory drugs should be given before resection.82%B
      Follow-up
       31. In case initial pathology shows LGD
      a. First follow-up (after removal of possible placed stents) should be performed within 6 months.81%D
      b. At first follow-up, biopsy specimens should only be taken when macroscopic abnormalities are present.94%D
      c. Follow-up interval should be 12 months or less.88%D
      d. At further follow-up, biopsy specimens should only be taken when macroscopic abnormalities are present.94%D
      e. Follow-up should be performed for at least 5 years.81%D
       32. In case initial pathology shows HGD
      a. First follow-up (after removal of possible placed stents) should be performed within 3 months.94%D
      b. At first follow-up, biopsy specimens should only be taken when macroscopic abnormalities are present.81%D
      c. Follow-up interval should be 6 months or less.94%D
      d. At further follow-up, biopsy specimens should only be taken when macroscopic abnormalities are present.81%D
      e. Follow-up should be performed for at least 5 years.75%D
      CBD, Common bile duct; CT, computed tomography; EUS, endoscopic ultrasound; EP, endoscopic papillectomy; HGD, high-grade dysplasia; LGD, low-grade dysplasia; MRCP, magnetic resonance cholangiopancreatography; MRI, magnetic resonance imaging; PD, pancreatic duct.
      Grading: A, level 1a-1b evidence; B, level 2a-3b evidence; C, level 4 evidence; D, level 5 evidence.
      Table 3Selection of final round statements that did not reach consensus
      StatementAgreement
      Diagnostic workup
       1. Either MRI/MRCP or EUS should be performed in every patient before resection.63%
       2. An endoscopic cholangiogram either before or during EP should only be performed if other performed tests are found inconclusive and there is still doubt about the presence of intraductal extension.44%
      Technical aspects
       3. STSC of the margins should not be performed after EP.56%
       4. STSC can be performed for the margins of the laterally spreading component but not the papillary margins.50%
       5. Pancreatic sphincterotomy after resection should only be performed in case of
      a. Extension in the pancreatic duct.38%
      b. Extension in the pancreatic duct or if drainage is deemed suboptimal.44%
       6. It can be helpful to inject the PD before resection to make it easier to find the PD after resection in case of extension in the pancreatic duct.44%
       7. In case there is bleeding during the procedure, an FCSEMS instead of a plastic stent should be placed in the CBD.63%
       8. In case there are concerns for residual adenomatous tissue in the distal part of the CBD, an FCSEMS should be placed in the CBD.31%
       9. Standard clip closure of the mucosal defect after resection should not be performed.38%
       10. Glucagon or scopulaminebutyl should be provided routinely before resection to reduce the risk of losing the specimen in the GI tract.56%
      Adverse events and management
       11. Vigorous hydration should be considered in patients without any cardiac comorbidity to further decrease the risk of postintervention pancreatitis.63%
       12. Every patient should be treated with PPI after performing an EP.69%
       13. Patients treated with PPI after resection should be treated for at least 2 weeks.69%
       14. If a bleeding occurs after EP and patient is hemodynamically stable after resuscitation with <1.2 mmol/L drop in hemoglobin
      a. Reintervention should be performed within 12 hours.38%
      b. Conservative treatment (continue or start PPI) is initially indicated.63%
      Follow-up
       15. Every patient should be admitted for observation after EP for
      a. At least 24 hours.69%
      b. At least 48 hours.44%
      CBD, Common bile duct; EP, endoscopic papillectomy; EUS, endoscopic ultrasound; FCSEMS, fully covered self-expanding metal stent; MRCP, magnetic resonance cholangiopancreatography; MRI, magnetic resonance imaging; PD, pancreatic duct; PPI, proton pump inhibitor; STSC, snare tip soft coagulation.

      Diagnostic workup

      The standard diagnostic workup of a patient with a papillary lesion should include a gastroduodenoscopy with a side-viewing instrument (100%) and biopsy sampling (94%) before resection. Additional abdominal imaging should only be performed for specific indications (75%-81%). Magnetic resonance imaging (MRI)/magnetic resonance cholangiopancreatography (MRCP) or endoscopic ultrasound (EUS) can be used to rule out intraductal extension and should at least be performed in case of cholestatic laboratory features (81%) and/or lesion size >2 cm (75%) (Table 1). Figure 3 shows intraductal extension on MRI/MRCP and EUS. No consensus was reached that intraductal extension should be ruled out standardly in every patient before resection (53%) (Supplementary Table 2). MRI/MRCP or computed tomography (CT) should be performed in case of significant weight loss and/or endoscopic signs of malignancy. In case of jaundice a CT should be performed as well (Table 1).
      Figure thumbnail gr3
      Figure 3A, MRCP showing ingrowth (∗) in the distal common bile duct (CBD) of approximately 15 mm. B, EUS showing ingrowth (∗) in the distal CBD of approximately 12.5 mm and dilatation of both CBD (14.4 mm) and pancreatic duct (PD) (5.5 mm).
      The only existing international guideline in EP from 2015 states that endoscopic retrograde cholangiopancreatography (ERCP) should be performed in every patient at the time of resection to assess for evidence of intraductal extension.
      • Chathadi K.V.
      • Khashab M.A.
      • Acosta R.D.
      • et al.
      ASGE Standards of Practice Committee
      The role of endoscopy in ampullary and duodenal adenomas.
      However, this was shown not to be common practice, with only 25% of respondents considering endoscopic cholangiography as part of the standard diagnostic workup (Supplementary Table 1). Our study does agree with the previous guideline statement that biopsy sampling should be performed in every patient before resection (94%). Nevertheless, the accuracy of biopsy sampling in papillary lesions is questionable; according to pathology studies preresection biopsy sampling accurately diagnosed 70% of papillary malignancies
      • Yamaguchi K.
      • Enjoji M.
      • Kitamura K.
      Endoscopic biopsy has limited accuracy in diagnosis of ampullary tumors.
      • Bellizzi A.M.
      • Kahaleh M.
      • Stelow E.B.
      The assessment of specimens procured by endoscopic ampullectomy.
      • Elek G.
      • Gyori S.
      • Toth B.
      • et al.
      Histological evaluation of preoperative biopsies from ampulla vateri.
      compared with at least 80% in the colon.
      • Petrelli N.J.
      • Létourneau R.
      • Weber T.
      • et al.
      Accuracy of biopsy and cytology for the preoperative diagnosis of colorectal adenocarcinoma.
      Biopsy sampling is also important when the diagnosis of adenoma is considered in a larger than normal papilla.

      Lesion assessment

      Because the accuracy of biopsy sampling in papillary lesions is questionable, careful endoscopic assessment is considered most important. However, final pathology may differ from the initial endoscopic diagnosis. Limited data of patients suspected to have a benign adenoma based on endoscopic appearance indicate that final pathology shows another diagnosis (such as normal mucosa, gastric heterotopia, adenomyomatosis, hamartoma, neuroendocrine tumor, or adenocarcinoma) in 10% to 20%.
      • Napoleon B.
      • Gincul R.
      • Ponchon T.
      • et al.
      Endoscopic papillectomy for early ampullary tumors: long-term results from a large multicenter prospective study.
      ,
      • Ahn D.W.
      • Ryu J.K.
      • Kim J.
      • et al.
      Endoscopic papillectomy for benign ampullary neoplasms: how can treatment outcome be predicted?.
      ,
      • Cheng C.L.
      • Sherman S.
      • Fogel E.L.
      • et al.
      Endoscopic snare papillectomy for tumors of the duodenal papillae.
      Furthermore, a recent series of patients who underwent EP showed that in only half of the lesions in which the resection specimen showed adenocarcinoma, malignancy was already suspected based on the endoscopic appearance, further questioning the accuracy of endoscopic assessment.
      • Fritzsche J.A.
      • Klein A.
      • Beekman M.J.
      • et al.
      Endoscopic papillectomy; a retrospective international multicenter cohort study with long-term follow-up.
      Advanced imaging techniques such as narrow-band imaging and chromoendoscopy have proven to be a valuable addition to white-light imaging in the diagnosis of colorectal lesions.
      • Bisschops R.
      • East J.E.
      • Hassan C.
      • et al.
      Advanced imaging for detection and differentiation of colorectal neoplasia: European Society of Gastrointestinal Endoscopy (ESGE) guideline—update 2019.
      However, the possible benefit of narrow-band imaging in the assessment of papillary lesions is only described in a small case series.
      • Uchiyama Y.
      • Imazu H.
      • Kakutani H.
      • et al.
      New approach to diagnosing ampullary tumors by magnifying endoscopy combined with a narrow-band imaging system.
      Accordingly, there is consensus that these techniques are not helpful in distinguishing between benign and malignant papillary lesions at this moment (71%) (Table 2).
      Despite the current lack of a predefined classification system to determine whether a papillary lesion is most likely benign or malignant (89%) (Table 2), consensus exists that features such as ulceration (median, 4; IQR, 1) and immobility (median, 4; IQR, 1) should be considered features of a potential malignant lesion, regardless of biopsy sample results (Supplementary Table 2). When ulceration is present, this lesion can even be defined as most likely malignant based on this sole feature, regardless of biopsy sample results (94%) (Table 1). In case the biopsy sample shows high-grade dysplasia, firm consistency is considered an important characteristic as well (median, 4; IQR, 1). Features such as smooth or irregular surface, tumor size >4 cm, spontaneous bleeding, and excessive friability are, as sole criteria, of less importance when assessing a papillary lesion (Supplementary Table 2).

      Patient selection

      Although certain features would define a lesion as most likely malignant, none of the mentioned characteristics should be considered as an independent reason to refer for surgical management (Supplementary Table 2).
      Considering the risk of incomplete endoscopic resection, consensus exists that patients should be referred for surgical management when ingrowth in the pancreatic duct (PD) or common bile duct (CBD) of >1 cm is present (76%) (Table 1). Jaundice (86%), ingrowth in the PD (79%) or CBD ≤1 cm (86%), and classification as an umbilicated lesion (a sign of central retraction) (82%) are not considered independent reasons to refer the patient for surgical management (Table 2). Nonetheless, no consensus exists on which additional techniques should be used to endoscopically resect ingrowth in the CBD or PD ≤1 cm (Supplementary Table 2, Supplementary Table 3). Furthermore, no agreement exists that intraductal extension should be ruled out routinely before resection (53%) (Supplementary Table 2).
      Moreover, in case a patient is unfit for surgery, endoscopic resection can, if technically feasible, still be considered in case of adenocarcinoma (75%) or ingrowth in the CBD >1 cm (81%) (Table 2). Additionally, intraductal radiofrequency ablation (RFA) has been successfully described in small studies.
      • Camus M.
      • Napoleon B.
      • Vienne A.
      • et al.
      Efficacy and safety of endobiliary radiofrequency ablation for the eradication of residual neoplasia after endoscopic papillectomy: a multicenter prospective study.
      • Rustagi T.
      • Irani S.
      • Reddy N.D.
      • et al.
      Endoscopic radiofrequency ablation for intraductal extension of ampullary neoplasms [abstract].
      • Choi Y.H.
      • Yoon S.B.
      • Chang J.H.
      • et al.
      The safety of radiofrequency ablation using a novel temperature-controlled probe for the treatment of residual intraductal lesions after endoscopic papillectomy.
      Accordingly, experts agree that EP with the additional use of RFA can be considered in a patient who is not a surgical candidate when ingrowth in the CBD >1 cm is present (75%) (Table 1). However, no consensus exists whether to consider endoscopic resection in combination with RFA when surgery is still an option (44%) (Supplementary Table 3).

      Technical aspects

      Resection should be performed at the plane of the duodenal wall (94%) with fractionated current (short, regular pulses of cutting current integrated in background of coagulation current), regardless of the size of the lesion (94%) (Table 2). A systematic review comparing fractionated and cutting current showed no difference in adverse events. This result was confirmed by a small randomized controlled pilot study that showed no difference in terms of safety and efficacy. Although the use of fractionated current might prevent immediate bleeding in larger adenomas, it may cause crush artefacts.
      • Iwasaki E.
      • Minami K.
      • Itoi T.
      • et al.
      Impact of electrical pulse cut mode during endoscopic papillectomy: pilot randomized clinical trial.
      ,
      • Minami K.
      • Iwasaki E.
      • Fukuhara S.
      • et al.
      Electric endocut and autocut resection for endoscopic papillectomy: a systematic review.
      Pancreatic and biliary sphincterotomy should, if indicated, only be performed after resection (100%) (Table 2). Indications for biliary sphincterotomy are concomitant bile duct stones or suboptimal drainage (81%) (Table 1). No consensus exists on indications for pancreatic sphincterotomy (38%-44%) (Table 3).
      Submucosal injection should in general only be performed in case of a laterally spreading lesion (88%) (Table 1). This statement is confirmed by a small randomized controlled trial that could not show advantages in the use of submucosal injection and concluded that resection without lifting would be simpler and therefore primarily the recommended technique.
      • Hyun J.J.
      • Lee T.H.
      • Park J.S.
      • et al.
      A prospective multicenter study of submucosal injection to improve endoscopic snare papillectomy for ampullary adenoma.
      ,
      • Chung K.H.
      • Lee S.H.
      • Choi J.H.
      • et al.
      Effect of submucosal injection in endoscopic papillectomy of ampullary tumor: propensity-score matching analysis.
      The effect of preventive PD stent placement on postpapillectomy pancreatitis (PPP) has been claimed by 1 small randomized controlled trial. However, this difference was only significant in the per-protocol analysis but not in the intention-to-treat analysis.
      • Harewood G.C.
      • Pochron N.L.
      • Gostout C.J.
      Prospective, randomized, controlled trial of prophylactic pancreatic stent placement for endoscopic snare excision of the duodenal ampulla.
      Nevertheless, more recently 2 systematic reviews and meta-analyses of available literature supported the preventive effect of PD stent placement as well.
      • Wang Y.
      • Qi M.
      • Hao Y.
      • et al.
      The efficacy of prophylactic pancreatic stents against complications of post-endoscopic papillectomy or endoscopic ampullectomy: a systematic review and meta-analysis.
      ,
      • Spadaccini M.
      • Fugazza A.
      • Frazzoni L.
      • et al.
      Endoscopic papillectomy for neoplastic ampullary lesions: A systematic review with pooled analysis.
      Accordingly, experts agree on the routine use of a PD stent to prevent PPP (100%) (Table 1). Different methods of PD stent placement have been studied; for example cannulating the PD before resection and performing resection with the guidewire in situ has been suggested. Because this method could potentially hamper complete en-bloc resection, consensus was achieved that the PD should be cannulated after resection (100%) (Table 2).
      • Lee T.Y.
      • Cheon Y.K.
      • Shim C.S.
      • et al.
      Endoscopic wire-guided papillectomy versus conventional papillectomy for ampullary tumors: a prospective comparative pilot study.
      • Kim S.H.
      • Moon J.H.
      • Choi H.J.
      • et al.
      Usefulness of pancreatic duct wire-guided endoscopic papillectomy for ampullary adenoma for preventing post-procedure pancreatitis.
      • Kobayashi M.
      • Ryozawa S.
      • Iwano H.
      • et al.
      The usefulness of wire-guided endoscopic snare papillectomy for tumors of the major duodenal papilla.
      • Moon J.H.
      • Cha S.W.
      • Cho Y.D.
      • et al.
      Wire-guided endoscopic snare papillectomy for tumors of the major duodenal papilla.
      However, it remains unclear if injecting the PD before resection would be helpful in finding the PD after resection (44%) (Table 3).
      • Poincloux L.
      • Scanzi J.
      • Goutte M.
      • et al.
      Pancreatic intubation facilitated by methylene blue injection decreases the risk for postpapillectomy acute pancreatitis.
      ,
      • Nakahara K.
      • Okuse C.
      • Suetani K.
      • et al.
      A novel endoscopic papillectomy after a pancreatic stent placement above the pancreatic duct orifice: inside pancreatic stenting papillectomy.
      Moreover, there was no consensus regarding the use of a PD stent with or without an internal flap (46% vs 54%) (Supplementary Table 1, Supplementary Table 2).
      No consensus exists that pancreas divisum should be routinely excluded before resection (65%) (Supplementary Table 2). In the final round only 63% agreed that either EUS or MRI/MRCP should be performed in every patient before resection (Table 3).
      A CBD stent should only be placed on indication (82%) (Table 1). No consensus was reached on the standard placement of a stent inside the CBD (18%) (Supplementary Table 1) to, for example, prevent postintervention cholangitis, which has been described in 0% to 7% of cases.
      • van der Wiel S.E.
      • Poley J.W.
      • Koch A.D.
      • et al.
      Endoscopic resection of advanced ampullary adenomas: a single-center 14-year retrospective cohort study.
      ,
      • Ridtitid W.
      • Tan D.
      • Schmidt S.E.
      • et al.
      Endoscopic papillectomy: risk factors for incomplete resection and recurrence during long-term follow-up.
      ,
      • Fritzsche J.A.
      • Klein A.
      • Beekman M.J.
      • et al.
      Endoscopic papillectomy; a retrospective international multicenter cohort study with long-term follow-up.
      Possible indications to place a CBD stent are concerns for a (micro)perforation (88%) or ongoing bleeding from the papillary region during the procedure (76%). In case of concerns for a (micro)perforation, a fully covered self-expanding metal stent (FCSEMS) is preferred over a plastic stent (88%) (Table 1). FCSEMSs could also be useful in case of bleeding from the biliary region by tamponading the bleeding vessel.
      • Donatelli G.
      • Dumont J.L.
      • Derhy S.
      • et al.
      New indications for fully covered lumen-apposing metal stents: biliary stenting to treat post-sphincterotomy bleeding or ampullary stenosis.
      Remarkably, no consensus was reached that a FCSEMS was preferred in case of bleeding as well, probably because the statement did not incorporate bleeding from the papillary region (63%) (Table 3). In case of residual tissue, stent placement could also facilitate the direct inspection of the distal CBD in the first follow-up procedure to exclude and treat possible intraductal extension.
      • Klein A.
      • Qi Z.
      • Bahin F.F.
      • et al.
      Outcomes after endoscopic resection of large laterally spreading lesions of the papilla and conventional ampullary adenomas are equivalent.
      Nonetheless, no consensus was achieved on the treatment of residual adenomatous tissue and the use of FCSEMSs in this manner (31%) (Table 3).
      Snare tip soft coagulation of the resection margins after endoscopic mucosal resection (EMR) of adenomas in the colon results in a 4-fold reduction in recurrence at first follow-up.
      • Klein A.
      • Tate D.J.
      • Jayasekeran V.
      • et al.
      Thermal ablation of mucosal defect margins reduces adenoma recurrence after colonic endoscopic mucosal resection.
      Recent (preliminary) data show similar results after EMR of duodenal adenomas, suggesting that snare tip soft coagulation could also prevent recurrence after EP.

      Sidhu M, Fritzsche JA, Klein A, Shahidi N, et al. Outcomes of thermal ablation of the defect margin after duodenal endoscopic mucosal resection (with videos). Gastrointest Endosc 2020;93:1373-80.

      However, considering the potential risk of perforation and/or pancreatitis this technique is, reasonably, not part of current practice at the moment (56%), and further prospective study is required to evaluate its potential utility (Table 3).
      Prophylactic clip closure of the mucosal defect to prevent bleeding has been successfully described after endoscopic resection of large colon polyps.
      • Pohl H.
      • Grimm I.S.
      • Moyer M.T.
      • et al.
      Clip closure prevents bleeding after endoscopic resection of large colon polyps in a randomized trial.
      Also, a recent small prospective study examined the preventive closure of the frenulum after EP, which led to a decrease in delayed bleeding without a shown increase in pancreatitis or perforation rates and without lengthening procedure time.
      • Kagawa K.
      • Kubota K.
      • Kurita Y.
      • et al.
      Effect of preventive closure of the frenulum after endoscopic papillectomy: A prospective pilot study.
      However, data are limited, and there is no agreement whether standard clip closure of the mucosal defect after resection should be performed (38%) (Table 3).
      The use of antispasmodic drugs such as glucagon and scopolaminebutyl can be helpful in reducing the risk of losing the specimen distally into the small bowel; however, no consensus exists whether to administer these drugs routinely before resection (56%) (Table 3). Probably because, in case of an adequate complete prone position of the patient, the risk of losing a specimen is considered low because it will generally migrate proximally with gravity toward the duodenal bulb. Figures 4 and 5 show the removal of a conventional and an extensive laterally spreading papillary adenoma, respectively, including placement of a PD and CBD stent, en-bloc removal of the adenoma at the plane of the duodenal wall, and clipping of the frenulum.
      Figure thumbnail gr4
      Figure 4A-C, Conventional en-bloc papillectomy at the level of the duodenal wall for 15 mm papillary adenoma. D-F, Exposure of biliary and pancreatic orifices with a 5F pancreatic stent and fully covered metal biliary stent. Clip closure of the frenulum for the prophylaxis of post-papillectomy bleeding.
      Figure thumbnail gr5
      Figure 5A and B, Extensive laterally spreading papillary adenoma involving greater than two-thirds of the duodenal circumference. C and D, Piecemeal EMR of the laterally spreading components resulting in 90% circumferential mucosal defect. E and F, En-bloc papillectomy followed by a 5F pancreatic stent and fully covered metal biliary stent.

      Adverse events and management

      In addition to the placement of a PD stent after resection, rectal nonsteroidal anti-inflammatory drugs should be given before resection to further aid in the prevention of PPP (82%) (Table 1). This consensus statement is supported by studies regarding pancreatitis prevention in conventional ERCP practice that showed a significant reduction in incidence when using nonsteroidal anti-inflammatory drug suppositories.
      • Elmunzer B.J.
      • Scheiman J.M.
      • Lehman G.A.
      • et al.
      A randomized trial of rectal indomethacin to prevent post-ERCP pancreatitis.
      • Elmunzer B.J.
      • Waljee A.K.
      • Elta G.H.
      • et al.
      A meta-analysis of rectal NSAIDs in the prevention of post-ERCP pancreatitis.
      • Khoshbaten M.
      • Khorram H.
      • Madad L.
      • et al.
      Role of diclofenac in reducing post-endoscopic retrograde cholangiopancreatography pancreatitis.
      In addition, studies suggest that preventive vigorous pre- and perprocedural hydration reduces the pancreatitis rate even further.
      • Buxbaum J.
      • Yan A.
      • Yeh K.
      • et al.
      Aggressive hydration with lactated Ringer's solution reduces pancreatitis after endoscopic retrograde cholangiopancreatography.
      • Shaygan-nejad A.
      • Masjedizadeh A.
      • Ghavidel A.
      • et al.
      Aggressive hydration with lactated Ringer's solution as the prophylactic intervention for postendoscopic retrograde cholangiopancreatography pancreatitis: a randomized controlled double-blind clinical trial.
      • Choi J.H.
      • Kim H.J.
      • Lee B.U.
      • et al.
      Vigorous periprocedural hydration with lactated Ringer's solution reduces the risk of pancreatitis after retrograde cholangiopancreatography in hospitalized patients.
      This outcome still needs to be confirmed by an adequately powered randomized controlled trial of which the results have not yet been published.
      • Smeets X.J.N.M.
      • da Costa D.W.
      • Fockens P.
      Dutch Pancreatitis Study Group. Fluid hydration to prevent post-ERCP pancreatitis in average- to high-risk patients receiving prophylactic rectal NSAIDs (FLUYT trial): study protocol for a randomized controlled trial.
      Accordingly, no consensus could be reached that vigorous hydration should be considered in, for example, cases without any major cardiac comorbidity (63%) (Table 3).
      Consensus could not be achieved on whether patients should be treated with a proton pump inhibitor (PPI) to decrease the risk of delayed bleeding after performing an EP (69%) (Table 3). Although the effect of acid suppression by a PPI in the treatment of upper GI tract bleeding caused by (artificial) ulcers is evident, the benefit of PPI use in the prevention of delayed bleeding after EP or, for example, sphincterotomy is not established and questionable given the relatively high pH in the duodenum.
      • Sachar H.
      • Vaidya K.
      • Laine L.
      Intermittent vs continuous proton pump inhibitor therapy for high-risk bleeding ulcers: a systematic review and meta-analysis.
      ,
      • Leung W.K.
      • But D.Y.K.
      • Wong S.-Y.
      • et al.
      Prevention of post-sphincterotomy bleeding by proton pump inhibitor: a randomized controlled trial.
      Considering bleeding, different methods were proposed to stop the bleeding such as the use of FCSEMSs, Hemospray (Cook Medical, Bloomington, Ind, USA), epinephrine injection, clips, or coagulation. No consensus could be achieved on the best treatment method in case of bleeding during the procedure (Supplementary Table 2, Supplementary Table 3). Furthermore, no consensus seems to exist as to whether to perform reintervention (38%) or initially treat a patient conservatively (63%) in case of delayed bleeding when the patient is hemodynamically stable after resuscitation (Table 3).

      Follow-up

      Biopsy sampling during follow-up only needs to be performed when macroscopic abnormalities are present (81%). The first follow-up should be performed within 3 months (94%) with an interval to the second follow-up of 6 months or less (94%) in case of high-grade dysplasia compared with, respectively, 6 (81%) and 12 months (88%) in case of low-grade dysplasia. In both cases follow-up should be continued for at least 5 years (75%-81%) (Table 2); however, 31-38% of experts would perform lifelong follow-up, as long as the patient is fit (Supplementary Table 3). Recent data on long-term follow-up after EP show that recurrence has been found even 5 years after the index procedure, confirming the need for consideration of longer follow-up.
      • Sahar N.
      • Krishnamoorthi R.
      • Kozarek R.A.
      • et al.
      Long-term outcomes of endoscopic papillectomy for ampullary adenomas.
      ,
      • Fritzsche J.A.
      • Klein A.
      • Beekman M.J.
      • et al.
      Endoscopic papillectomy; a retrospective international multicenter cohort study with long-term follow-up.
      Finally, there was consensus that patients should be admitted for observation after the procedure (82%) (Supplementary Table 2). However, participants could not agree on the length of necessary observation, 69% would observe at least 24 hours, and 44% at least 48 hours (Table 3).

      Discussion

      EP is established as the preferred method to manage benign papillary adenomas.
      • Ridtitid W.
      • Tan D.
      • Schmidt S.E.
      • et al.
      Endoscopic papillectomy: risk factors for incomplete resection and recurrence during long-term follow-up.
      ,
      • Klein A.
      • Tutticci N.
      • Bourke M.J.
      Endoscopic resection of advanced and laterally spreading duodenal papillary tumors.
      EP is minimally invasive, and modeling using well-validated scoring systems has shown it is safer and less expensive than surgical management.
      • Klein A.
      • Ahlenstiel G.
      • Tate D.J.
      • et al.
      Endoscopic resection of large duodenal and papillary lateral spreading lesions is clinically and economically advantageous compared with surgery.
      Moreover, in case of unsuspected cancer in the papillectomy specimen, it does not preclude or compromise subsequent surgery in a surgically fit patient. Although generally considered safe, EP is not without risks, with serious adverse events occurring in 15-35% of patients and recurrence in up to 20% during surveillance.
      • van der Wiel S.E.
      • Poley J.W.
      • Koch A.D.
      • et al.
      Endoscopic resection of advanced ampullary adenomas: a single-center 14-year retrospective cohort study.
      • Klein A.
      • Qi Z.
      • Bahin F.F.
      • et al.
      Outcomes after endoscopic resection of large laterally spreading lesions of the papilla and conventional ampullary adenomas are equivalent.
      • Ridtitid W.
      • Tan D.
      • Schmidt S.E.
      • et al.
      Endoscopic papillectomy: risk factors for incomplete resection and recurrence during long-term follow-up.
      • Napoleon B.
      • Gincul R.
      • Ponchon T.
      • et al.
      Endoscopic papillectomy for early ampullary tumors: long-term results from a large multicenter prospective study.
      • Sahar N.
      • Krishnamoorthi R.
      • Kozarek R.A.
      • et al.
      Long-term outcomes of endoscopic papillectomy for ampullary adenomas.
      ,
      • Fritzsche J.A.
      • Klein A.
      • Beekman M.J.
      • et al.
      Endoscopic papillectomy; a retrospective international multicenter cohort study with long-term follow-up.
      It is therefore incumbent on those managing these patients to further refine and optimize the EP procedure to mitigate against these adverse events. However, high-level scientific knowledge to guide this process is largely absent because papillary adenomas are uncommon, and thus large prospective multicenter or randomized controlled studies have not been executed. Therefore, unsurprisingly, an evidence-based consensus for EP has not been established. We sought to address this deficiency by using a Delphi process among international experts.
      Sixteen international experts joined the final round, and consensus was reached on 47 of the final 79 proposed statements. It was necessary to propose a large number of statements to fully evaluate this complex multifaceted procedure. Based on the consensus achieved, insight was given in the main characteristics that should be taken into account when classifying a papillary lesion and reasons to refer a patient for surgical management. Furthermore, a consensus-based algorithm regarding diagnostic workup, technical aspects, and follow-up is proposed and is depicted in Figure 2.
      As stated, biopsy sampling is only accurate in 70% of the papillary malignancies, and therefore endoscopic assessment can be considered most important.
      • Yamaguchi K.
      • Enjoji M.
      • Kitamura K.
      Endoscopic biopsy has limited accuracy in diagnosis of ampullary tumors.
      • Bellizzi A.M.
      • Kahaleh M.
      • Stelow E.B.
      The assessment of specimens procured by endoscopic ampullectomy.
      • Elek G.
      • Gyori S.
      • Toth B.
      • et al.
      Histological evaluation of preoperative biopsies from ampulla vateri.
      To improve this classification, it would be helpful to have a predefined scoring system to better assess these lesions. This does not currently exist, and although we were able to identify important characteristics deemed to be associated with malignancy, further study is needed to propose a useful scoring system to improve the assessment of papillary lesions. Also, advanced imaging techniques such as narrow-band imaging or chromoendoscopy could be further assessed to determine their possible additional value to the classification of papillary lesions.
      Besides the need for studies regarding optimal diagnosis, a lack of consensus on different technical aspects of the procedure warrants future prospective studies. For example, the benefit of FCSEMS placement after resection in case of bleeding from the papillary region or when residual tissue is present, RFA in case of ingrowth in the CBD, and the use of snare tip soft coagulation of the margins of the laterally spreading segment need to be further evaluated to determine their utility in daily practice. Furthermore, preventive measures such as vigorous hydration to prevent PPP and PPI to prevent delayed bleeding after resection are not part of the proposed consensus algorithm because, due to the paucity of data, no consensus could be reached on the benefit of these methods, showing the need for future well-targeted studies.
      This study is not free of limitations. According to the low patient numbers in the literature and the absence of systematic prospective studies, high-level scientific enquiry to inform evidence-based consensus is lacking. Consequently and unsurprisingly, because of the lack of consensus on certain critical steps in the process, it is impossible to propose a complete algorithm for EP. In addition, note should be taken that the proposed protocol in this study mainly focuses on patients with sporadic papillary adenoma and is not directly translatable to patients with a genetic predisposition as in familial adenomatous polyposis syndrome. Although the resection technique and the prevention and treatment of adverse events can be considered the same, the diagnostic and follow-up protocol contain important differences. Final limitations are inherent to the study design; the expert panel, although reliant on published literature, was partly composed on personal expert knowledge, which makes the study eligible for personal bias. Because an adequate sample of possible participants was not available, test–retest reliability could not be performed, and, most importantly, when consensus was reached, this does not certainly mean that the correct answer has been found.
      • Hasson F.
      • Keeney S.
      • McKenna H.
      Research guidelines for the Delphi survey technique.
      However, given the lack of large prospective, randomized studies, this Delphi consensus provides the best available evidence regarding the management of these relatively uncommon lesions. The invited expert panel can be considered an adequate reflection of the experience in EP because researchers and renowned experts (therefore, wide clinical and scientific experience) were included. Furthermore, the size of the panel can be considered reliable for content validation because 5 to 10 experts are considered sufficient.
      • Lynn M.R.
      Determination and quantification of content validity.
      By using a Delphi process, consensus could be achieved in a group of geographically spread experts, obviating the need for direct confrontation, leaving room for individual thought.
      • Dalkey N.C.
      The Rand Corporation
      The Delphi method: an experimental study of group opinion.
      The use of anonymous group feedback gave participants the opportunity to change their opinion based on group consensus. Moreover, consensus statements were supported by available literature after systematic review of the literature. Accordingly, this study provides a unique agreement and the best available, evidence-based guideline for the endoscopic management of papillary adenomas.
      In conclusion, this Delphi study provides the current highest level of evidence regarding the different aspects of the performance of EP. Although there are surprisingly many areas in which no consensus exists and scientific data are lacking, this study led to the first consensus-based management algorithm for papillary adenomas. Therefore, this study can be considered a vital step around which future studies can be designed to ultimately generate a more robust evidence-based consensus guideline for EP. Furthermore, important insights were observed in areas in which a high variety in daily practice still exists, and, accordingly, future studies could be targeted.

      Acknowledgments

      All co-authors contributed to at least 2 rounds of the Delphi process, and we sincerely thank everyone who contributed to the first round only for their help and valuable contribution to this study: T. H. Baron, E. L. Fogel, S. Irani, M. Kobayashi, R. A. Kozarek, S. K. Lo, J. H. Moon, B. Napoleon, A. Repici, S. Seewald, G. Webster, and U. Will.

      Appendix

      Supplementary Table 1Survey round 1
      QuestionAgreement
      Expertise
       1. In which country do you work?
      a. United States of America29%
      b. Japan14%
      c. France11%
      d. Korea11%
      e. Germany7%
      f. Italy7%
      g. Netherlands7%
      h. Belgium4%
      i. Finland4%
      j. Switzerland4%
      k. United Kingdom4%
       2. In which type of practice do you primarily work?
      a. Academic practice86%
      b. Private practice11%
      c. Nonacademic practice7%
      d. Other4%
       3. How many years have you been in practice?
      a. <5 years0%
      b. 5-10 years7%
      c. 10-20 years25%
      d. >20 years68%
       4. Do you perform colonoscopies every week?
      a. Yes82%
      b. No18%
       5. How many colonoscopies do you perform every week? (n = 23)
      a. <5 per week30%
      b. 5-10 per week35%
      c. 10-20 per week22%
      d. >20 per week13%
       6. Do you perform basic polypectomy?
      a. Yes93%
      b. No7%
       7. Do you perform colonic EMR?
      a. Yes82%
      b. No18%
       8. How many ERCPs do you perform yearly?
      a. <100 per year4%
      b. 100-200 per year21%
      c. 200-400 per year36%
      d. >400 per year39%
       9. How many endoscopic papillectomies are performed yearly in your center?
      a. <5 per year4%
      b. 5-10 per year18%
      c. 10-20 per year36%
      d. >20 per year43%
       10. How many endoscopic papillectomies did you perform in your career?
      a. <5 in total0%
      b. 5-10 in total4%
      c. 10-30 in total7%
      d. >30 in total89%
       11. Do you perform EUS?
      a. Yes79%
      b. No21%
       12. How many EUS do you perform yearly? (n = 21)
      a. <50 per year5%
      b. 50-100 per year5%
      c. 100-200 per year9%
      d. >200 per year82%
      Diagnostic workup
       1. Do you get patients referred to perform an endoscopic papillectomy?
      a. Yes100%
      b. No0%
       2. Which diagnostic modalities are usually already performed by the referring physician?
      a. CT50%
      b. EUS29%
      c. MRCP29%
      d. Cholangiography7%
      e. None of the above25%
      f. Other
      i. Gastroduodenoscopy29%
       3. What are possible reasons to perform a CT before resection?
      a. Part of the standard diagnostic workup43%
      b. Jaundice64%
      c. Cholestatic laboratory features, without clinical signs of jaundice43%
      d. Significant weight loss57%
      e. Double duct sign on other imaging61%
      f. Age older than 5011%
      g. Large (>2 cm) adenoma36%
      h. Endoscopic features of possible malignancy79%
      i. Other7%
       4. What are possible reasons to perform an MRCP before resection?
      a. Part of the standard diagnostic workup46%
      b. Jaundice71%
      c. Cholestatic laboratory features, without clinical signs of jaundice68%
      d. Significant weight loss36%
      e. Double duct sign on other imaging57%
      f. Age older than 507%
      g. Large (>2 cm) adenoma43%
      h. Endoscopic features of possible malignancy54%
      i. Other11%
       5. What are possible reasons to perform an EUS before resection?
      a. Part of the standard diagnostic workup50%
      b. Jaundice61%
      c. Cholestatic laboratory features, without clinical signs of jaundice57%
      d. Significant weight loss39%
      e. Double duct sign on other imaging61%
      f. Age older than 5011%
      g. Large (>2 cm) adenoma54%
      h. Endoscopic features of possible malignancy75%
      i. Other14%
       6. What are possible reasons to perform a cholangiography before resection?
      a. Part of the standard diagnostic workup25%
      b. Jaundice54%
      c. Cholestatic laboratory features, without clinical signs of jaundice29%
      d. Significant weight loss14%
      e. Double duct sign on other imaging50%
      f. Age older than 504%
      g. Large (>2 cm) adenoma14%
      h. Endoscopic features of possible malignancy50%
      i. Other21%
       7. Do you routinely perform biopsy sampling before resection?
      a. Yes75%
      b. No25%
      Lesion assessment and staging
       1. Do you use any kind of predefined classification system to determine if a lesion of the ampulla of Vater is most likely benign or malignant?
      a. Yes11%
      b. No89%
       2. Which of the following characteristics on endoscopic imaging do you normally take into account to determine if the lesion is benign or malignant?
      a. Presence of ulceration100%
      b. Aspect of the surface (granular vs smooth)43%
      c. Friability61%
      d. Firmness/rigidity82%
      e. Mobility61%
      f. Spontaneous bleeding54%
      g. Tumor size57%
      h. Umbilicated lesion46%
      i. Other0%
       3. Is there a specific tumor size that would raise your suspicion of the lesion being malignant?
      a. Tumor size does not raise the suspicion of being malignant54%
      b. >1 cm0%
      c. >2 cm18%
      d. >3 cm21%
      e. >4 cm7%
       4. Do you believe any of the characteristics mentioned below is specific enough by itself to define a lesion as most likely malignant? If yes, which?
      a. Ulceration79%
      b. Smooth surface4%
      c. Excessive friability18%
      d. Firm consistency32%
      e. Immobility36%
      f. Spontaneous bleeding14%
      g. Tumor size >4 cm14%
      h. No, only a combination of the characteristics makes it possible to define a lesion as most likely malignant21%
      i. Other4%
       5. What are possible reasons to refer the patient for surgical management instead of performing endoscopic papillectomy?
      a. Jaundice (without evident signs of malignancy based on endoscopic imaging)14%
      b. Ingrowth CBD ≤1 cm14%
      c. Ingrowth PD ≤1 cm21%
      d. Ingrowth CBD >1 cm75%
      e. Ingrowth PD >1 cm86%
      f. Tumor size >4 cm18%
      g. Other18%
       6. Do you use any advanced imaging techniques to distinguish adenomatous from nonadenomatous tissue?
      a. Yes, always29%
      b. Yes, sometimes36%
      c. No, never36%
       7. Do you look for signs of central retraction, also called an umbilicated lesion?
      a. Yes, this raises the suspicion of malignancy and patient should be referred for surgical management18%
      b. Yes, but the lesion could still be removed endoscopically if there are no further signs of malignancy75%
      c. No7%
       8. Considering a fit, 70-year-old patient without any significant comorbidities, how would you treat the lesion if biopsy before resection shows only adenomatous tissue with low-grade dysplasia but endoscopic imaging shows possible signs of malignancy?
      a. Endoscopically resect the lesion based on the biopsy sample result, disregarding the endoscopic imaging. Only refer the patient for surgery if the resected specimen shows malignancy.82%
      b. Refer the patient for surgical management.18%
       9. When there is ingrowth in the CBD of more than 1 cm, would you consider endoscopic papillectomy in combination with RFA?
      a. Yes50%
      b. No, I would refer the patient for surgical treatment50%
      Technical aspects
       1. Where do you perform the resection of the lesion?
      a. At the plane of the duodenal wall100%
      b. Just above the plane of the duodenal wall0%
       2. What type of electrosurgical current do you use most often while performing endoscopic papillectomy?
      a. Pure coagulation current0%
      b. Pure cutting current14%
      c. Fractionated current (short regular pulses of cutting current integrated in background of coagulation current)79%
      d. Other7%
       3. Do you standard use submucosal injection before endoscopic papillectomy?
      a. Yes4%
      b. No, only in case of lateral spreading79%
      c. No18%
       4. Do you routinely perform pancreatic sphincterotomy?
      a. Yes, always14%
      b. Yes, sometimes36%
      c. No, never50%
       5. When you perform pancreatic sphincterotomy do you in general perform it before or after resection? (n = 14)
      a. Before resection0%
      b. After resection100%
       6. Do you routinely place a stent in the PD to prevent postintervention pancreatitis?
      a. Yes89%
      b. Only if PD drainage is deemed suboptimal or if the PD is difficult to cannulate after the procedure11%
      c. No0%
       7. Do you cannulate the PD before or after resection?
      a. Cannulate before resection and perform the resection with guidewire inside the PD.7%
      b. Inject the PD before resection and cannulate it after resection.43%
      c. Cannulate after resection.50%
       8. What kind of PD stent do you use in general?
      a. Stent with an internal flap46%
      b. Stent without an internal flap54%
       9. Do you routinely perform biliary sphincterotomy?
      a. Yes, always32%
      b. Only in case of concomitant bile duct stones57%
      c. No, never11%
       10. When you perform biliary sphincterotomy do you in general perform it before or after resection? (n = 25)
      a. Before resection4%
      b. After resection96%
       11. Do you place a stent in the common bile duct after resection as standard practice?
      a. Yes18%
      b. No, only on indication71%
      c. No, never11%
       12. What are the indications to place a stent in the CBD after resection? (n = 20)
      a. Bleeding during the procedure80%
      b. Residual adenomatous tissue70%
      c. Concerns for microperforations after resection80%
      d. Advanced resection25%
      e. Multiple submucosal injections5%
      f. Extensive piecemeal resection20%
      g. Other30%
       13. What type of biliary stent do you prefer if the bile duct is 8 mm in diameter? (n = 25)
      a. Plastic72%
      b. FCSEMS28%
       14. What adjunct modality do you use most commonly to remove small (<1 cm) residual tissue during index procedure after papillectomy?
      a. Cold forceps biopsy32%
      b. Argon plasma coagulation43%
      c. Monopolar/multipolar coagulation probe4%
      d. Snare tip soft coagulation36%
      e. Other18%
       15. Do you use snare tip soft coagulation of the margins after resection?
      a. Yes0%
      b. Yes, but only in case of lateral spread25%
      c. No75%
       16. Would you perform RFA when EUS shows ingrowth in the CBD of <1 cm?
      a. Yes50%
      b. No50%
      Complications and management
       1. Do you standard give rectal indomethacin/diclofenac suppository before resection to prevent postintervention pancreatitis?
      a. Yes75%
      b. No25%
       2. Do you use vigorous hydration to prevent postintervention pancreatitis?
      a. Yes54%
      b. No46%
       3. Do you routinely start patients on PPI after performing an endoscopic papillectomy?
      a. Yes50%
      b. No, only on indication32%
      c. No, never18%
       4. What are the indications to start patients on PPI after performing an endoscopic papillectomy?
      a. Excessive friability of tissue during the procedure11%
      b. Use of antithrombotic medication33%
      c. Advanced resection67%
      d. Multiple submucosal injections0%
      e. Extensive piecemeal resection67%
      f. Other11%
       5. For how long will you normally treat patients with PPI after performing an endoscopic papillectomy?
      a. 24 hours9%
      b. Between 24 hours and 2 weeks39%
      c. Between 2 weeks and 1 month52%
      d. More than 1 month0%
       6. If bleeding occurs after endoscopic papillectomy and the patient is hemodynamic stable after resuscitation with <2 g/dL drop in hemoglobin, how would you manage it initially?
      a. Continue PPI, but no endoscopic intervention because patient is stable.25%
      b. Endoscopic reintervention to treat bleeding.64%
      c. Other11%
      Follow-up
       1. After which period would you perform the first endoscopic follow-up?
      a. 3 months54%
      b. 3-6 months4%
      c. 6 months18%
      d. 12 months4%
      e. Other21%
       2. Which follow-up interval do you use if no residual adenomatous tissue is seen at first follow-up endoscopy?
      a. 6 months46%
      b. 12 months39%
      c. >12 months4%
      d. Other11%
       3. Until when would you perform endoscopic follow-up?
      a. Up to 2 years after endoscopic papillectomy11%
      b. Up to 3 years after endoscopic papillectomy14%
      c. Up to 5 years after endoscopic papillectomy32%
      d. Up to 10 years after endoscopic papillectomy7%
      e. Lifelong, as long as patient is fit32%
      f. Other4%
       4. Do you take biopsy samples during follow-up as standard practice?
      a. Yes36%
      b. No, only when macroscopic abnormalities are present64%
      c. No, never0%
      CBD, Common bile duct; CT, computed tomography; EMR, endoscopic mucosal resection; ERCP, endoscopic retrograde cholangiopancreatography; EUS, endoscopic ultrasound; FCSEMS, fully covered self-expanding metal stent; MRCP, magnetic resonance cholangiopancreatography; MRI, magnetic resonance imaging; PD, pancreatic duct; PPI, proton pump inhibitor; RFA, radiofrequency ablation.
      Supplementary Table 2Survey round 2
      QuestionAgreement
      Diagnostic workup
      Please fill out the order in which you believe the different diagnostic modalities should be performed for different indications (1 = first choice, 2 = only if test(s) of first choice are inconclusive, 3 = only if test(s) of second choice are inconclusive, 4 = only if test(s) of third choice are inconclusive, 5 = only if all earlier performed test(s) are inconclusive, 6 = never)
       1. In every patient
      a. CT3 (IQR 4)
      b. MRCP2 (IQR 2)
      c. EUS2 (IQR 1)
      d. Endoscopic cholangiography4 (IQR 2)
      e. Biopsy sampling1 (IQR 0)
       2. In case of jaundice
      a. CT2 (IQR 2)
      b. MRCP2 (IQR 1)
      c. EUS2 (IQR 1)
      d. Endoscopic cholangiography3 (IQR 3)
      e. Biopsy sampling1 (IQR 0)
       3. In case of cholestatic laboratory features without clinical signs of jaundice
      a. CT2 (IQR 1)
      b. MRCP1 (IQR 1)
      c. EUS2 (IQR 1)
      d. Endoscopic cholangiography3 (IQR 2)
      e. Biopsy sampling1 (IQR 1)
       4. In case of significant weight loss
      a. CT1 (IQR 1)
      b. MRCP2 (IQR 2)
      c. EUS2 (IQR 2)
      d. Endoscopic cholangiography4 (IQR 2)
      e. Biopsy sampling1 (IQR 1)
       5. In case of double duct sign on other imaging
      a. CT2 (IQR 1)
      b. MRCP2 (IQR 1)
      c. EUS1 (IQR 1)
      d. Endoscopic cholangiography3 (IQR 2)
      e. Biopsy sampling1 (IQR 0)
       6. In case patient is older than 50 years
      a. CT2 (IQR 2)
      b. MRCP2 (IQR 2)
      c. EUS2 (IQR 1)
      d. Endoscopic cholangiography4 (IQR 2)
      e. Biopsy sampling1 (IQR 0)
       7. In case of large (>2 cm) adenoma
      a. CT2 (IQR 2)
      b. MRCP2 (IQR 1)
      c. EUS1 (IQR 1)
      d. Endoscopic cholangiography4 (IQR 2)
      e. Biopsy sampling1 (IQR 0)
       8. In case of endoscopic signs of malignancy
      a. CT1 (IQR 1)
      b. MRCP2 (IQR 1)
      c. EUS2 (IQR 1)
      d. Endoscopic cholangiography3 (IQR 2)
      e. Biopsy sampling1 (IQR 0)
       9. Do you believe pancreas divisum should be ruled out in every patient before resection?
      a. Yes65%
      b. No35%
       10. Could you please fill out in which order the following diagnostic modalities should be performed to rule out pancreas divisum before resection? (1 = first choice, 2 = only if first choice is inconclusive, 3 = only if earlier performed test(s) are found inconclusive, 4 = never)
      a. MRCP2 (IQR 0)
      b. EUS2 (IQR 0)
      c. Endoscopic cholangiography3 (IQR 1)
       11. Do you believe local extension should be ruled out in every patient before resection?
      a. Yes53%
      b. Yes, except for small (<2 cm) adenoma or in case of familial adenomatous polyposis29%
      c. No18%
       12. Could you please fill out in which order the following diagnostic modalities should be performed to rule out local extension before resection? (1 = first choice, 2 = only if first choice is inconclusive, 3 = only if earlier performed test(s) are found inconclusive, 4 = never)
      a. MRCP2 (IQR 1)
      b. EUS1 (IQR 0)
      c. Endoscopic cholangiography2 (IQR 1)
      Lesion assessment and staging
       1. When a lesion shows ulceration, this lesion should be defined as most likely malignant.94%
       2. The following characteristics are not a reason to define the lesion as most likely malignant
      a. Smooth surface71%
      b. Spontaneous bleeding53%
      c. Lesion size >4 cm35%
       3. You should refer a patient for surgical management in case of
      a. Ingrowth in the PD >1 cm76%
      b. Ingrowth in the CBD >1 cm47%
       4. The following situations are not a reason to refer the patient for surgical management
      a. Jaundice41%
      b. Ingrowth in the PD ≤1 cm41%
      c. Ingrowth in the CBD ≤1 cm65%
      d. An umbilicated lesion53%
       5. When a lesion shows ulceration, do you believe this patient should be referred for surgical management? (n = 16)
      a. Yes, regardless of possible biopsy sampling results38%
      b. Only when biopsy specimen shows HGD63%
       6. If a patient is not a surgical candidate because of age and/or comorbidity, in which of the following situations would you still resect the lesion endoscopically?
      a. There is ingrowth of >1 cm in the CBD59%
      b. There is ingrowth of >1 cm in the PD35%
      c. The lesion shows ulceration35%
      d. Biopsy sample shows adenocarcinoma47%
      e. In none of the above described situations29%
      f. Other0%
       7. When there is ingrowth in the CBD of >1 cm, would you consider endoscopic papillectomy with RFA?
      a. In a fit 60-year-old patient47%
      b. In a patient who is not a surgical candidate because of age and/or comorbidity76%
       8. Do you believe any advanced imaging techniques (such as narrow-band imaging or chromoendoscopy) would be helpful to distinguish between benign and malignant lesions?
      a. Yes29%
      b. No71%
       9. If the biopsy sample shows LGD, which of the following characteristics would make you to refer the patient for surgical management? (0 = would not take this into account, 5 = would refer patient for surgical treatment based on the sole characteristic)
      a. Ulceration4 (IQR 1)
      b. Smooth or irregular surface2 (IQR 2)
      c. Excessive friability1 (IQR 3)
      d. Firm consistency3 (IQR 2)
      e. Immobility4 (IQR 1)
      f. Spontaneous bleeding2 (IQR 3)
      g. Tumor size >4 cm2 (IQR 2)
       10. If the biopsy sample shows HGD, which of the following characteristics would make you to refer the patient for surgical management?
      a. Ulceration4 (IQR 1)
      b. Smooth or irregular surface3 (IQR 3)
      c. Excessive friability4 (IQR 3)
      d. Firm consistency4 (IQR 1)
      e. Immobility4 (IQR 0)
      f. Spontaneous bleeding4 (IQR 2)
      g. Tumor size >4 cm3 (IQR 2)
      Technical aspects
       1. Resection of the lesion should be performed at the plane of the duodenal wall.94%
       2. Endoscopic papillectomy should be performed with fractionated current.94%
       3. Submucosal injection should only be performed in case of a laterally spreading lesion.88%
       4. If pancreatic sphincterotomy is performed, then it should be performed after resection.88%
       5. PD stent should be placed routinely to prevent postintervention pancreatitis.100%
       6. PD should be cannulated after resection.100%
       7. If biliary sphincterotomy is performed, it should be performed after resection.100%
       8. A CBD stent should only be placed on indication.82%
       9. A CBD stent should be placed in the following cases:
      a. In case of bleeding during the procedure.76%
      b. If there are any concerns for microperforations.88%
      c. In case of residual adenomatous tissue.59%
       10. STSC of the margins after endoscopic papillectomy should not be performed.65%
       11. When do you believe a pancreatic sphincterotomy should be performed?
      a. Never29%
      b. In case of intrapancreatic duct extent59%
      c. In case of pancreas divisum instead of PD stent6%
      d. Other12%
       12. Do you believe it is better to inject the PD before resection to make it easier to find the PD after resection?
      a. Yes41%
      b. No59%
       13. Which kind of PD stent would you choose in the following situations?
      a. Always with an internal flap47%
      b. Always without an internal flap35%
      c. Without an internal flap, but in case of large (>2 cm) lesions0%
      d. Without an internal flap, but in case of perforation18%
      e. Without an internal flap, but after coagulation or RFA18%
      f. Other0%
       14. When would you perform biliary sphincterotomy?
      a. Always18%
      b. In case of bile duct stones76%
      c. In case of drainage is deemed suboptimal53%
      d. Other6%
       15. Which stent do you prefer in a bile duct with diameter ≤8 mm in the following cases
      a. Bleeding during the procedure
      i. Plastic41%
      ii. FCSEMS59%
      iii. No stent0%
      b. Concerns for microperforations after resection
      i. Plastic35%
      ii. FCSEMS65%
      iii. No stent0%
      c. Residual tissue
      i. Plastic29%
      ii. FCSEMS71%
      iii. No stent0%
       16. Which stent do you prefer in a bile duct with diameter >8 mm in the following cases
      a. Bleeding during the procedure
      i. Plastic24%
      ii. FCSEMS65%
      iii. No stent12%
      b. Concerns for microperforations after resection
      i. Plastic18%
      ii. FCSEMS71%
      iii. No stent12%
      c. Residual tissue
      i. Plastic41%
      ii. FCSEMS35%
      iii. No stent24%
       17. What adjunct modalities would you consider to remove small (<1 cm) residual tissue during index procedure after papillectomy?
      a. Cold forceps biopsy sampling65%
      b. APC53%
      c. Monopolar/multipolar coagulation probe12%
      d. STSC18%
      e. Hot snare41%
      f. Cold snare53%
      g. Other6%
       18. How would you resect the lesion when EUS shows ingrowth in the CBD of <1 cm?
      a. Snare resection in combination with RFA24%
      b. Snare resection in combination with APC35%
      c. Resect intraductal residue in second procedure47%
      d. Resect intraductal residue in second procedure after placement of FCSEMS during first procedure18%
      e. Other6%
       19. Do you standardly use clips to close the defect after resection?
      a. Yes41%
      b. No59%
       20. Do you standardly give buscopan or glucagon before resection to reduce the risk of loss of specimen further in the GI tract?
      a. Yes59%
      b. No41%
       21. Would you use a different current to resect the lesion based on the size of the lesion?
      a. Large (>2 cm) lesion
      i. Pure coagulation current0%
      ii. Pure cutting current6%
      iii. Fractionated current94%
      b. Small (<2 cm) lesion
      i. Pure coagulation current6%
      ii. Pure cutting current24%
      iii. Fractionated current71%
      Complications and management
       1. Rectal indomethacin or diclofenac should be given before resection.82%
       2. If bleeding occurs during intervention, which modalities to stop the bleeding would you consider?
      a. FCSEMS71%
      b. Coagulation grasper65%
      c. Goldprobe12%
      d. Hemospray41%
      e. STSC24%
      f. APC12%
      g. Other35%
       3. When would you start patients on PPI after performing an endoscopic papillectomy?
      a. Always65%
      b. Extensive friability of the tissue during the procedure6%
      c. Use of antithrombotic medication0%
      d. Advanced resection of large lesion24%
      e. Multiple submucosal injections6%
      f. Extensive piecemeal resection18%
      g. Never12%
       4. How long do you believe patients should be treated with PPI after resection?
      a. 24 hours6%
      b. 48 hours0%
      c. 1 week24%
      d. 2 weeks18%
      e. 3 weeks0%
      f. 1 month47%
       5. If a bleeding occurs after endoscopic papillectomy and the patient is hemodynamic stable after resuscitation with <2-g/dL drop in hemoglobin, how would you manage it initially?
      a. Expectative (continue or start PPI), no endoscopic intervention because patient is stable41%
      b. Endoscopic reintervention to treat bleeding within working hours47%
      c. Endoscopic reintervention to treat bleeding as an emergency procedure12%
      Follow-up
       1. Do you standardly admit patients for observation after endoscopic papillectomy? If yes, how long, if no complications occur?
      a. 6 hours0%
      b. 24 hours29%
      c. 48 hours35%
      d. Other
      i. No18%
      ii. >48 hours18%
       2. After which period should first follow-up be performed after presumed complete removal (excluding the removal of possible placed stents within 1-3 months)?
      a. In case initial pathology shows LGD
      i. 3 months35%
      ii. 3-6 months29%
      iii. 6 months24%
      iv. 6-12 months12%
      v. >12 months0%
      b. In case initial pathology shows HGD
      i. 3 months76%
      ii. 3-6 months24%
      iii. 6 months0%
      iv. 6-12 months0%
      v. >12 months0%
       3. What should be the follow-up interval if no residual tissue is seen at first follow-up?
      a. In case initial pathology shows LGD
      i. 6 months35%
      ii. 12 months53%
      iii. >12 months12%
      b. In case initial pathology shows HGD
      i. 6 months65%
      ii. 12 months35%
      iii. >12 months0%
       4. Until when should endoscopic follow-up be performed?
      a. In case initial pathology shows LGD
      i. Up to 2 years12%
      ii. Up to 3 years12%
      iii. Up to 5 years41%
      iv. Up to 10 years12%
      v. Lifelong, as long as patient is fit24%
      b. In case initial pathology shows HGD
      i. Up to 2 years12%
      ii. Up to 3 years0%
      iii. Up to 5 years47%
      iv. Up to 10 years6%
      v. Lifelong, as long as patient is fit35%
       5. Should biopsy samples be taken routinely at first follow-up?
      a. In case initial pathology shows LGD
      i. Yes29%
      ii. No, only when macroscopic abnormalities are present71%
      b. In case initial pathology shows HGD
      i. Yes35%
      ii. No, only when macroscopic abnormalities are present65%
       6. Should biopsy samples be taken routinely at further follow-up?
      a. In case initial pathology shows LGD
      i. Yes18%
      ii. No, only when macroscopic abnormalities are present82%
      b. In case initial pathology shows HGD
      i. Yes35%
      ii. No, only when macroscopic abnormalities are present65%
      CBD, Common bile duct; CT, computed tomography; EUS, endoscopic ultrasound; FCSEMS, fully covered self-expanding metal stent; IQR, interquartile range; HGD, high-grade dysplasia; LGD, low-grade dysplasia; MRCP, magnetic resonance cholangiopancreatography; MRI, magnetic resonance imaging; STSC, snare tip soft coagulation; APC, argon plasma coagulation; PD, pancreatic duct; PPI, proton pump inhibitor; RFA, radiofrequency ablation.
      Supplementary Table 3Survey round 3
      QuestionAgreement
      Diagnostic workup statements
       1. Gastroduodenoscopy (with side-viewing instrument) should always be performed before resection.100%
       2. Biopsy sampling should always be performed before resection.94%
       3. Either MRI/MRCP or EUS should be performed in every patient before resection.63%
       4. Either MRI/MRCP or EUS should be performed in case of cholestatic laboratory features with or without jaundice.81%
       5. Either CT, MRI/MRCP or EUS should be performed in case of cholestatic laboratory features with or without jaundice.75%
       6. CT should be performed in case of jaundice.75%
       7. Either MRI/MRCP or EUS should be performed in case of a lesion larger than 2 cm.75%
       8. Either MRI/MRCP or CT should be performed in case of significant weight loss.81%
       9. Either EUS, MRI/MRCP, or CT should be performed in case of significant weight loss.63%
       10. Either MRI/MRCP or CT should be performed in case of endoscopic signs of malignancy.81%
       11. Either CT or EUS should be performed in case of endoscopic signs of malignancy.69%
       12. Either MRI/MRCP, EUS, or CT should be performed in case of endoscopic signs of malignancy.81%
       13. An endoscopic cholangiogram either before or during endoscopic papillectomy should only be performed if other performed tests are found inconclusive and there is still doubt about the presence of intraductal extension.44%
      Lesion assessment and staging
       1. Patient should be referred for surgical management in case of ingrowth in the CBD of more than 1 cm, considering patient is suitable for surgery.81%
       2. If biopsy sample shows LGD and ulceration is present, the lesion could still be resected endoscopically; there is no need to refer the patient for surgical management based on this sole characteristic, considering the lesion seems favorable for endoscopic resection.88%
       3. If biopsy sample shows HGD and ulceration is present, the lesion could still be resected endoscopically; there is no need to refer the patient for surgical management based on this sole characteristic, considering the lesion seems favorable for endoscopic resection.63%
       4. If there is ingrowth in the CBD of more than 1 cm, endoscopic resection can still be considered if the patient is not a surgical candidate because of age and/or comorbidity, considering the lesion seems favorable for endoscopic resection.81%
       5. If biopsy sample shows adenocarcinoma in situ or well-differentiated adenocarcinoma, endoscopic resection can still be considered if the patient is not a surgical candidate because of age and/or comorbidity, considering the lesion seems favorable for endoscopic resection.75%
       6. If there is ingrowth in the CBD of more than 1 cm, endoscopic papillectomy with RFA should not be performed in a fit 60-year-old patient.44%
       7. If there is ingrowth in the CBD of more than 1 cm, endoscopic papillectomy with RFA can be considered in a patient that is not a surgical candidate because of age and/or comorbidity, considering the lesion seems favorable for endoscopic resection.75%
      Technical aspects
       1. Snare tip soft coagulation of the margins should not be performed after endoscopic papillectomy.56%
       2. Snare tip soft coagulation can be performed for the margins of the laterally spreading component, but not the papillary margins.50%
       3. Pancreatic sphincterotomy after resection should only be performed in case of extension in the pancreatic duct.38%
       4. Pancreatic sphincterotomy after resection should be performed in case of extension in the pancreatic duct or if drainage is deemed suboptimal.44%
       5. It can be helpful to inject the PD before resection to make it easier to find the PD after resection in case of extension in the pancreatic duct.44%
       6. Biliary sphincterotomy should be performed only in case of concomitant bile duct stones.44%
       7. Biliary sphincterotomy should be performed in case of concomitant bile duct stones and in case drainage is deemed suboptimal.81%
       8. In case there is bleeding during the procedure, an FCSEMS instead of a plastic stent should be placed in the CBD.63%
       9. In case there are concerns for microperforations in the papillary region, an FCSEMS should be placed in the CBD.88%
       10. In case there are concerns for residual adenomatous tissue in the distal part of the CBD, an FCSEMS should be placed in the CBD.31%
       11. In case there are concerns for residual adenomatous tissue in the distal part of the CBD, either a plastic stent or FCSEMS should be placed in the CBD.44%
       12. Please rank the following items in which order you would consider them if there is ingrowth in the CBD of <1 cm, which seems accessible from the duodenum. (1 = first choice, 5 = never)
      a. Snare resection in combination with RFA2 (IQR 1.5)
      b. Snare resection in combination with APC3 (IQR 3.25)
      c. Resect intraductal residue in second procedure after placement of FCSEMS during first procedure3 (IQR 1.25)
      d. Resect with snare inside duct2.5 (IQR 2.25)
       13. Standard clip closure of the mucosal defect after resection should not be performed.38%
       14. Glucagon or buscopan should be provided routinely before resection to reduce the risk of losing the specimen into the GI tract.56%
      Adverse events and management
       1. Vigorous hydration should be considered in patients without any cardiologic comorbidity to further decrease the risk of postintervention pancreatitis.63%
      Please rank the following methods in which order you would consider them to use in case of intraprocedural bleeding in the following hypothetical situations (1 = first choice, 6 = never)
       2. In case of bleeding from the papillectomy site (not from lateral spreading component) considering the PD is already protected by a stent.
      a. FCSEMS3.5 (IQR 1.5)
      b. Hemospray5 (IQR 2)
      c. Epinephrine2 (IQR 1.25)
      d. Clips3 (IQR 1)
      e. Coagulation probe (or other instrument used to coagulate)2 (IQR 1)
       3. In case of bleeding from the papillectomy site (not from lateral spreading component) however the PD is not yet protected by a stent.
      a. FCSEMS4 (IQR 2.25)
      b. Hemospray4 (IQR 3)
      c. Epinephrine1 (IQR 1)
      d. Clips3 (IQR 3.25)
      e. Coagulation probe (or other instrument used to coagulate)2 (IQR 2)
       4. In case of bleeding from the lateral spreading component.
      a. FCSEMS5.5 (IQR 1)
      b. Hemospray4 (IQR 2)
      c. Epinephrine2 (IQR 2)
      d. Clips2.5 (IQR 2)
      e. Coagulation probe (or other instrument used to coagulate)2 (IQR 1)
       5. Every patient should be treated with PPI after performing an endoscopic papillectomy.69%
       6. Patients treated with PPI after resection should be treated for at least 2 weeks.69%
       7. Patients treated with PPI after resection should be treated for at least 1 month.56%
       8. If a bleeding occurs after endoscopic papillectomy and the patient is hemodynamic stable after resuscitation with <2-g/dL drop in hemoglobin, reintervention should be performed within 12 hours.38%
       9. If a bleeding occurs after endoscopic papillectomy and the patient is hemodynamic stable after resuscitation with <2-g/dL drop in hemoglobin, conservative treatment (continue or start PPI) is initially indicated.63%
      Follow-up
       1. Every patient should be admitted for at least 24 hours for observation after endoscopic papillectomy.69%
       2. Every patient should be admitted for at least 48 hours for observation after endoscopic papillectomy.44%
       3. In case initial pathology shows LGD
      a. First follow-up (after removal of possible placed stents) should be performed within 6 months.81%
      b. First follow-up (after removal of possible placed stents) should be performed within 3 months.56%
      c. At first follow-up biopsy samples should only be taken when macroscopic abnormalities are present.94%
      d. Follow-up interval should be 12 months or less.88%
      e. Follow-up interval should be 6 months or less.38%
      f. At further follow-up biopsy samples should only be taken when macroscopic abnormalities are present.94%
      g. Follow-up should be performed for at least 5 years.81%
      h. Follow-up should be performed lifelong, as long as patient is fit.31%
       4. In case initial pathology shows HGD
      a. First follow-up (after removal of possible placed stents) should be performed within 6 months.31%
      b. First follow-up (after removal of possible placed stents) should be performed within 3 months.94%
      c. At first follow-up biopsy samples should only be taken when macroscopic abnormalities are present.81%
      d. Follow-up interval should be 12 months or less.25%
      e. Follow-up interval should be 6 months or less.94%
      f. At further follow-up biopsy samples should only be taken when macroscopic abnormalities are present.81%
      g. Follow-up should be performed for at least 5 years.75%
      h. Follow-up should be performed lifelong, as long as patient is fit.38%
      CBD, Common bile duct; CT, computed tomography; EUS, endoscopic ultrasound; MRCP, magnetic resonance cholangiopancreatography; MRI, magnetic resonance imaging; FCSEMS, fully covered self-expanding metal stent; HGD, high-grade dysplasia; LGD, low-grade dysplasia; STSC, snare tip soft coagulation; APC, argon plasma coagulation; PD, pancreatic duct; PPI, proton pump inhibitor; RFA, radiofrequency ablation.

      References

        • Suzuki K.
        • Kantou U.
        • Murakami Y.
        Two cases with ampullary cancer who underwent endoscopic resection.
        Prog Dig Endosc. 1983; : 236-239
        • Binmoeller K.F.
        • Boaventura S.
        • Ramsperger K.
        • et al.
        Endoscopic snare excision of benign adenomas of the papilla of Vater.
        Gastrointest Endosc. 1993; 39: 127-131
        • van der Wiel S.E.
        • Poley J.W.
        • Koch A.D.
        • et al.
        Endoscopic resection of advanced ampullary adenomas: a single-center 14-year retrospective cohort study.
        Surg Endosc. 2019; 33: 1180-1188
        • Klein A.
        • Qi Z.
        • Bahin F.F.
        • et al.
        Outcomes after endoscopic resection of large laterally spreading lesions of the papilla and conventional ampullary adenomas are equivalent.
        Endoscopy. 2018; 50: 972-983
        • Ridtitid W.
        • Tan D.
        • Schmidt S.E.
        • et al.
        Endoscopic papillectomy: risk factors for incomplete resection and recurrence during long-term follow-up.
        Gastrointest Endosc. 2014; 79: 289-296
        • Napoleon B.
        • Gincul R.
        • Ponchon T.
        • et al.
        Endoscopic papillectomy for early ampullary tumors: long-term results from a large multicenter prospective study.
        Endoscopy. 2014; 46: 127-134
        • Sahar N.
        • Krishnamoorthi R.
        • Kozarek R.A.
        • et al.
        Long-term outcomes of endoscopic papillectomy for ampullary adenomas.
        Dig Dis Sci. 2020; 65: 260-268
        • Chathadi K.V.
        • Khashab M.A.
        • Acosta R.D.
        • et al.
        • ASGE Standards of Practice Committee
        The role of endoscopy in ampullary and duodenal adenomas.
        Gastrointest Endosc. 2015; 82: 773-781
        • El Hajj I.
        • Cote G.A.
        Endoscopic diagnosis and management of ampullary lesions.
        Gastrointest Endosc Clin North Am. 2013; 23: 95-109
        • Bassan M.
        • Bourke M.
        Endoscopic ampullectomy: a practical guide.
        J Interv Gastroenterol. 2012; 2: 23-30
        • Dalkey N.C.
        • The Rand Corporation
        The Delphi method: an experimental study of group opinion.
        Rand Corporation, Santa Monica, CA1969
        • Bennett C.
        • Vakil N.
        • Bergman J.
        • et al.
        Consensus statements for management of Barrett's dysplasia and early-stage esophageal adenocarcinoma, based on a Delphi process.
        Gastroenterology. 2012; 143: 336-346
        • Powell C.
        The Delphi technique: myths and realities.
        J Adv Nurs. 2003; 41: 376-382
        • Goggins M.
        • Overbeek K.A.
        • Brand R.
        • et al.
        Management of patients with increased risk for familial pancreatic cancer: updated recommendations from the International Cancer of the Pancreas Screening (CAPS) Consortium.
        Gut. 2020; 69: 7-17
        • Hasson F.
        • Keeney S.
        • McKenna H.
        Research guidelines for the Delphi survey technique.
        J Adv Nurs. 2000; 32: 1008-1015
        • Likert R.
        Technique for the measurement of attitudes.
        Arch Psychol. 1932; : 1932-1953
        • Sumsion T.
        The Delphi technique: an adaptive research tool.
        Br J Occup Ther. 1998; 61: 153-156
        • von der Gracht H.A.
        Consensus measurement in Delphi studies. Review and implications for future quality assurance.
        Technol Forecast Soc Change. 2012; 79: 1525-1536
        • Centre for Evidence Based Medicine
        Oxford Centre for Evidence-based Medicine—levels of evidence.
        Centre for Evidence-Based Medicine, Oxford, UK2009
        • Yamaguchi K.
        • Enjoji M.
        • Kitamura K.
        Endoscopic biopsy has limited accuracy in diagnosis of ampullary tumors.
        Gastrointest Endosc. 1990; 36: 588-592
        • Bellizzi A.M.
        • Kahaleh M.
        • Stelow E.B.
        The assessment of specimens procured by endoscopic ampullectomy.
        Am J Clin Pathol. 2009; 132: 506-513
        • Elek G.
        • Gyori S.
        • Toth B.
        • et al.
        Histological evaluation of preoperative biopsies from ampulla vateri.
        Pathol Oncol Res. 2003; 9: 32-41
        • Petrelli N.J.
        • Létourneau R.
        • Weber T.
        • et al.
        Accuracy of biopsy and cytology for the preoperative diagnosis of colorectal adenocarcinoma.
        Surg Oncol Clin North Am. 1999; 71: 46-49
        • Ahn D.W.
        • Ryu J.K.
        • Kim J.
        • et al.
        Endoscopic papillectomy for benign ampullary neoplasms: how can treatment outcome be predicted?.
        Gut Liver. 2013; 7: 239-245
        • Cheng C.L.
        • Sherman S.
        • Fogel E.L.
        • et al.
        Endoscopic snare papillectomy for tumors of the duodenal papillae.
        Gastrointest Endosc. 2004; 60: 757-764
        • Fritzsche J.A.
        • Klein A.
        • Beekman M.J.
        • et al.
        Endoscopic papillectomy; a retrospective international multicenter cohort study with long-term follow-up.
        Surg Endosc. Epub 2020 Nov 6;
        • Bisschops R.
        • East J.E.
        • Hassan C.
        • et al.
        Advanced imaging for detection and differentiation of colorectal neoplasia: European Society of Gastrointestinal Endoscopy (ESGE) guideline—update 2019.
        Endoscopy. 2019; 51: 1155-1179
        • Uchiyama Y.
        • Imazu H.
        • Kakutani H.
        • et al.
        New approach to diagnosing ampullary tumors by magnifying endoscopy combined with a narrow-band imaging system.
        J Gastroenterol. 2006; 41: 483-490
        • Camus M.
        • Napoleon B.
        • Vienne A.
        • et al.
        Efficacy and safety of endobiliary radiofrequency ablation for the eradication of residual neoplasia after endoscopic papillectomy: a multicenter prospective study.
        Gastrointest Endosc. 2018; 88: 511-518
        • Rustagi T.
        • Irani S.
        • Reddy N.D.
        • et al.
        Endoscopic radiofrequency ablation for intraductal extension of ampullary neoplasms [abstract].
        Gastrointest Endosc. 2016; 83: AB298
        • Choi Y.H.
        • Yoon S.B.
        • Chang J.H.
        • et al.
        The safety of radiofrequency ablation using a novel temperature-controlled probe for the treatment of residual intraductal lesions after endoscopic papillectomy.
        Gut Liver. 2021; 15: 307-314
        • Iwasaki E.
        • Minami K.
        • Itoi T.
        • et al.
        Impact of electrical pulse cut mode during endoscopic papillectomy: pilot randomized clinical trial.
        Dig Endosc. 2020; 32: 127-135
        • Minami K.
        • Iwasaki E.
        • Fukuhara S.
        • et al.
        Electric endocut and autocut resection for endoscopic papillectomy: a systematic review.
        Intern Med. 2019; 58: 2767-2772
        • Hyun J.J.
        • Lee T.H.
        • Park J.S.
        • et al.
        A prospective multicenter study of submucosal injection to improve endoscopic snare papillectomy for ampullary adenoma.
        Gastrointest Endosc. 2017; 85: 746-755
        • Chung K.H.
        • Lee S.H.
        • Choi J.H.
        • et al.
        Effect of submucosal injection in endoscopic papillectomy of ampullary tumor: propensity-score matching analysis.
        United Eur Gastroenterol J. 2018; 6: 576-585
        • Harewood G.C.
        • Pochron N.L.
        • Gostout C.J.
        Prospective, randomized, controlled trial of prophylactic pancreatic stent placement for endoscopic snare excision of the duodenal ampulla.
        Gastrointest Endosc. 2005; 62: 367-370
        • Wang Y.
        • Qi M.
        • Hao Y.
        • et al.
        The efficacy of prophylactic pancreatic stents against complications of post-endoscopic papillectomy or endoscopic ampullectomy: a systematic review and meta-analysis.
        Therap Adv Gastroenterol. 2019; 12 (1756284819855342)
        • Spadaccini M.
        • Fugazza A.
        • Frazzoni L.
        • et al.
        Endoscopic papillectomy for neoplastic ampullary lesions: A systematic review with pooled analysis.
        United Eur Gastroenterol J. 2020; 8: 44-51
        • Lee T.Y.
        • Cheon Y.K.
        • Shim C.S.
        • et al.
        Endoscopic wire-guided papillectomy versus conventional papillectomy for ampullary tumors: a prospective comparative pilot study.
        J Gastroenterol Hepatol. 2016; 31: 897-902
        • Kim S.H.
        • Moon J.H.
        • Choi H.J.
        • et al.
        Usefulness of pancreatic duct wire-guided endoscopic papillectomy for ampullary adenoma for preventing post-procedure pancreatitis.
        Endoscopy. 2013; 45: 838-841
        • Kobayashi M.
        • Ryozawa S.
        • Iwano H.
        • et al.
        The usefulness of wire-guided endoscopic snare papillectomy for tumors of the major duodenal papilla.
        PLoS One. 2019; 14e0211019
        • Moon J.H.
        • Cha S.W.
        • Cho Y.D.
        • et al.
        Wire-guided endoscopic snare papillectomy for tumors of the major duodenal papilla.
        Gastrointest Endosc. 2005; 61: 461-466
        • Poincloux L.
        • Scanzi J.
        • Goutte M.
        • et al.
        Pancreatic intubation facilitated by methylene blue injection decreases the risk for postpapillectomy acute pancreatitis.
        Eur J Gastroenterol Hepatol. 2014; 26: 990-995
        • Nakahara K.
        • Okuse C.
        • Suetani K.
        • et al.
        A novel endoscopic papillectomy after a pancreatic stent placement above the pancreatic duct orifice: inside pancreatic stenting papillectomy.
        J Clin Gastroenterol. 2014; 48: 796-800
        • Donatelli G.
        • Dumont J.L.
        • Derhy S.
        • et al.
        New indications for fully covered lumen-apposing metal stents: biliary stenting to treat post-sphincterotomy bleeding or ampullary stenosis.
        Endoscopy. 2018; 50: E36-E37
        • Klein A.
        • Tate D.J.
        • Jayasekeran V.
        • et al.
        Thermal ablation of mucosal defect margins reduces adenoma recurrence after colonic endoscopic mucosal resection.
        Gastroenterology. 2019; 156: 604-613
      1. Sidhu M, Fritzsche JA, Klein A, Shahidi N, et al. Outcomes of thermal ablation of the defect margin after duodenal endoscopic mucosal resection (with videos). Gastrointest Endosc 2020;93:1373-80.

        • Pohl H.
        • Grimm I.S.
        • Moyer M.T.
        • et al.
        Clip closure prevents bleeding after endoscopic resection of large colon polyps in a randomized trial.
        Gastroenterology. 2019; 157: 977-984
        • Kagawa K.
        • Kubota K.
        • Kurita Y.
        • et al.
        Effect of preventive closure of the frenulum after endoscopic papillectomy: A prospective pilot study.
        J Gastroenterol Hepatol. 2020; 35: 374-379
        • Elmunzer B.J.
        • Scheiman J.M.
        • Lehman G.A.
        • et al.
        A randomized trial of rectal indomethacin to prevent post-ERCP pancreatitis.
        N Engl J Med. 2012; 366: 1414-1422
        • Elmunzer B.J.
        • Waljee A.K.
        • Elta G.H.
        • et al.
        A meta-analysis of rectal NSAIDs in the prevention of post-ERCP pancreatitis.
        Gut. 2008; 57: 1262-1267
        • Khoshbaten M.
        • Khorram H.
        • Madad L.
        • et al.
        Role of diclofenac in reducing post-endoscopic retrograde cholangiopancreatography pancreatitis.
        J Gastroenterol Hepatol. 2008; 23: e11-e16
        • Buxbaum J.
        • Yan A.
        • Yeh K.
        • et al.
        Aggressive hydration with lactated Ringer's solution reduces pancreatitis after endoscopic retrograde cholangiopancreatography.
        Clin Gastroenterol Hepatol. 2014; 12: 303-307
        • Shaygan-nejad A.
        • Masjedizadeh A.
        • Ghavidel A.
        • et al.
        Aggressive hydration with lactated Ringer's solution as the prophylactic intervention for postendoscopic retrograde cholangiopancreatography pancreatitis: a randomized controlled double-blind clinical trial.
        J Res Med Sci. 2015; 20: 838-842
        • Choi J.H.
        • Kim H.J.
        • Lee B.U.
        • et al.
        Vigorous periprocedural hydration with lactated Ringer's solution reduces the risk of pancreatitis after retrograde cholangiopancreatography in hospitalized patients.
        Clin Gastroenterol Hepatol. 2017; 15: 86-92
        • Smeets X.J.N.M.
        • da Costa D.W.
        • Fockens P.
        Dutch Pancreatitis Study Group. Fluid hydration to prevent post-ERCP pancreatitis in average- to high-risk patients receiving prophylactic rectal NSAIDs (FLUYT trial): study protocol for a randomized controlled trial.
        Trials. 2018; 19: 207
        • Sachar H.
        • Vaidya K.
        • Laine L.
        Intermittent vs continuous proton pump inhibitor therapy for high-risk bleeding ulcers: a systematic review and meta-analysis.
        JAMA Intern Med. 2014; 174: 1755-1762
        • Leung W.K.
        • But D.Y.K.
        • Wong S.-Y.
        • et al.
        Prevention of post-sphincterotomy bleeding by proton pump inhibitor: a randomized controlled trial.
        J Dig Dis. 2018; 19: 369-376
        • Klein A.
        • Tutticci N.
        • Bourke M.J.
        Endoscopic resection of advanced and laterally spreading duodenal papillary tumors.
        Dig Endosc. 2016; 28: 121-130
        • Klein A.
        • Ahlenstiel G.
        • Tate D.J.
        • et al.
        Endoscopic resection of large duodenal and papillary lateral spreading lesions is clinically and economically advantageous compared with surgery.
        Endoscopy. 2017; 49: 659-667
        • Lynn M.R.
        Determination and quantification of content validity.
        Nurs Res. 1986; 35: 382-385

      Linked Article

      • Endoscopic ampullectomy: Can expert input shape endoscopic practice?
        Gastrointestinal EndoscopyVol. 94Issue 4
        • Preview
          Endoscopic ampullectomy has been performed since 1983 and is considered a standard endoscopic approach for benign papillary adenomas without evidence of significant intraductal extension or malignancy.1,2 Ampullectomy has a reported adverse event rate of up to 15% and an adenoma recurrence rate of up to 20%; yet, it has been found to be safer and more cost effective when compared with surgical approaches.3 Importantly, endoscopic ampullectomy does not compromise subsequent surgical intervention if the resected specimen is found to harbor malignancy on pathologic evaluation.
        • Full-Text
        • PDF