Safety and efficacy of endoscopic papillectomy for duodenal papillary adenomas: A single-center retrospective study

Abstract

BACKGROUND

Pancreaticoduodenectomy (PD) is currently the preferred surgical procedure to remove duodenal papillary adenoma. It offers a low recurrence rate, but a high rate of postoperative complications.

AIM

To determine the efficacy and safety of endoscopic papillectomy (EP) for duodenal papillary adenoma.

METHODS

This retrospective case-control study included 102 patients who received treatment at Nanjing Drum Tower Hospital between January 2010 and December 2024. Clinicopathological features, adverse events, and outcomes were compared between the two groups.

RESULTS

After propensity score matching for sex, year, and tumor size, 37 patients each were assigned to the PD and EP groups. All patients underwent successful removal of the diseased tissues. The pathological analysis revealed no significant differences between the two groups (P > 0.05). The PD group exhibited a significantly higher incidence of pancreatic fistula and intra-abdominal infection after surgery. Two patients in the PD group died from surgery-related complications, whereas no mortality was observed in the EP group (P = 0.152).

CONCLUSION

EP may be effective and safe for the removal of duodenal papillary adenoma in a minimally invasive manner, and can offer superior long-term survival compared with conventional PD.

Key Words: Duodenal papillary adenoma; Pancreaticoduodenectomy; Endoscopic resection; Recurrence; Complication

Core Tip: Endoscopic papillectomy (EP) is a minimally invasive, effective, and safe treatment for duodenal papillary neoplasms, offering comparable recurrence rates and superior long-term survival benefits to pancreaticoduodenectomy (PD). While PD has a low recurrence rate, its high postoperative complications limit its use. This study found similar recurrence rates between EP and PD, with fewer complications in the endoscopic group. EP is a viable alternative to PD for treating duodenal papillary neoplasms.

INTRODUCTION

Duodenal papillary adenoma is a rare condition, with an incidence of 0.3%–5%, and is usually discovered incidentally during upper gastrointestinal endoscopy[1,2]. While most ampullary lesions are non-neoplastic, duodenal papillary adenoma is prone to malignant transformation. Estimates show that up to 40% of villous adenomas of the duodenum may become malignant[3-5]. To prevent malignant transformation and improve prognosis, early recognition and surgical or endoscopic removal of duodenal papillary adenomas is critical.

Currently, there is no clear consensus regarding the choice of resection method for duodenal papillary adenomas; treatment is based on the experience of individual physicians. Considering its cancerous potential, surgical excision rather than endoscopic surveillance is recommended, such as pancreaticoduodenectomy (PD)[6] and transduodenal ampullectomy (TDA)[7], which are both invasive procedures. However, given the complex anatomy and poor blood supply of the ampulla, these surgical procedures remain challenging and are associated with high risks of perioperative and postoperative complications. Specifically, PD is associated with a low recurrence rate but a relatively high mortality[8], while TDA is commonly accompanied by recurrence[6]. Therefore, regular endoscopic surveillance is required after local excision of duodenal papillary adenomas after PD and TDA.

Recent advances have improved the safety and efficacy of endoscopic resection for gastrointestinal lesions[9]. Endoscopic resection is superior to conventional surgical procedures and is characterized by minimal invasion and shorter hospital stays[10,11]. Although this approach preserves the integrity of the digestive tract, the safety and efficacy of endoscopic resection for duodenal papillary adenomas remain to be validated in clinical settings.

Currently, long-term follow-up data from large cohorts of patients with duodenal papillary adenoma treated by endoscopic papillectomy (EP) are limited. This single-center retrospective study compared the safety and efficacy of EP and PD in treating papillary adenomas to provide a reference for clinical decision-making.

MATERIALS AND METHODS

Patients

From January 2010 to December 2024, 156 patients with duodenal papillary adenoma treated at Drum Tower Hospital were initially recruited. Patients with biopsy findings indicative of malignancy or those deemed to have advanced cancer based on endoscopic evaluation were excluded. Five patients were further excluded owing to postoperative pathological confirmation of neuroendocrine tumors or ectopic pancreas. An additional 21 patients were excluded owing to insufficient preoperative imaging data required for matching (Figure 1). The study was approved by the Ethics Committee of Drum Tower Hospital, Nanjing University School of Medicine (Approval No. 2021-401-02), and written informed consent was obtained from all the participants before enrolment.

Figure 1 Flowchart of patient enrollment and treatment allocation. A total of 102 eligible patients with a single lesion in the duodenal papilla were enrolled. After propensity score matching, 74 eligible patients were enrolled in the final analysis, with 37 undergoing endoscopic papillectomy and 37 undergoing pancreaticoduodenectomy. PSM: Propensity score matching.

EP procedure

EP was performed by five experienced endoscopists using an Olympus TJF-Q260 endoscope. Other equipment and accessories used were a Micro-Tech snare, titanium clips, and an ERBE Erbotom ICC 200 Electrosurgical Unit. The EP procedure was performed in a manner similar to endoscopic mucosal resection for colonic polyps. The procedure started with submucosal injection of a solution containing 100 mL of saline and 1 mL of methylene blue around the lesion to elevate the duodenal papillary neoplasm and create a safety cushion. After adequate lifting, the lesion was resected using a snare. Any bleeding was controlled using hot biopsy forceps and bipolar electrocoagulation. Mucosal defects were partially closed with titanium clips to prevent postoperative bleeding and perforation. Finally, a plastic stent was inserted into the pancreatic duct to prevent the development of pancreatitis (Figure 2). Some patients had plastic stents implanted in the bile duct to prevent biliary stricture. The resected specimens were fixed in formalin and sectioned at intervals of 2–3 mm (Figure 3).

Figure 2 Endoscopic papillectomy for resection of duodenal papillary adenoma. A: Endoscopic visualization of duodenal papillary adenoma; B: Submucosal injection (e.g., saline with epinephrine) around the lesion to facilitate resection; C and D: En bloc snare excision of the lesion; E and F: Deployment of a 5-Fr, 3-cm pancreatic duct stent to prevent post-procedural pancreatitis, followed by closure of the mucosal defect using hemostatic clips; G: Gross view of the resected specimen; H: X-ray fluoroscopy demonstrating the pancreatic duct stent and titanium clips.

Figure 3 Histopathological evaluation of resected duodenal papillary adenoma. A: Macroscopic appearance of the formalin-fixed duodenal papillary specimen; B: Representative tissue sections prepared for histopathological examination; C–E: Hematoxylin and eosin-stained sections demonstrating the microscopic architecture of the neoplasms.

PD procedure

PD was performed by three experienced pancreatic surgeons. The head of the pancreas, gallbladder, duodenum, a portion of the stomach, and surrounding lymph nodes were removed. The remaining pancreas, gastric body, and common bile duct were reconnected with the jejunum. Anatomical sampling was conducted on the postoperative specimens (Figure 4).

Figure 4 Pancreaticoduodenectomy for resection of duodenal papillary adenoma. A: Endoscopic visualization of a duodenal papillary adenoma; B: Endoscopic ultrasound features of duodenal papillary lesions; C–E: Gross morphology of the surgically resected specimen.

Definitions and follow-up

Perioperative perforation was a serious complication of EP. It was identified by the presence of excessive duodenal tissues, a kidney shadow on X-ray imaging, or radiological evidence of intraperitoneal free gas. Hemorrhage was defined as bleeding associated with melena, unstable blood pressure, or a decrease in hemoglobin of > 2 g/dL. Successful resection was endoscopically confirmed by the complete removal of adenomatous tissue. Surveillance endoscopy was performed at 3, 6, 12, and 24 months postoperatively to monitor the ampullary wound. Patients with malignant lesions also underwent abdominal computed tomography (CT) at 3, 6, 12, and 24 months after surgery.

Statistical analysis

Clinical data were analyzed using SPSS version 25.0 (IBM, Armonk, NY, USA). Continuous variables were expressed as mean ± SD. Differences between groups were assessed using the t-test or Fisher’s exact test, as appropriate. Categorical variables were compared using the χ² test. A two-sided P < 0.05 was considered statistically significant. To minimize confounding bias, propensity score matching (PSM) was used to analyze the data from the two groups. The patients were matched at a 1:1 ratio according to age, sex, and tumor size. All statistical analyses were performed using R software, version 4.1.2 (R Project for Statistical Computing).

RESULTS

Baseline characteristics

A total of 102 eligible patients with a single lesion in the duodenal papilla were enrolled. After PSM, 74 patients were divided into the PD and EP groups (Figure 1). The serum levels of total bilirubin and direct bilirubin in the PD group (26.59 ± 33.75 mmol/L and 16.88 ± 26.77 mmol/L) were significantly higher than those in the EP group (12.08 ± 5.75 mmol/L and 3.71 ± 1.58 mmol/L). No significant differences were observed in sex, age, preoperative blood routine examination, preoperative coagulation function, and renal functions between the PD and EP groups. No significant differences were observed between the two groups with respect to past medical history or prior use of anticoagulant medications (Table 1).

Table 1 Baseline characteristics of patients, n (%).

	PD (n = 37)	EP (n = 37)	t/χ2	P value
Age (year)	61.35 ± 8.11	62.83 ± 11.32	-0.649	0.518
Gender
Male	17 (45.95)	25 (67.75)	3.52	0.06
Female	20 (54.05)	12 (32.43)
Tumor size (cm)	1.94 ± 0.77	1.94 ± 0.85	0.123	0.481
Previous abdominal surgery
No	25 (67.57)	28 (75.68)	0.598	0.439
Yes	12 (32.43)	9 (24.32)
Hypertension
No	21 (56.76)	24 (64.86)	0.51	0.475
Yes	16 (43.24)	13 (35.14)
Diabetes
No	34 (91.89)	32 (86.49)	0.561	0.454
Yes	3 (8.11)	5 (13.51)
Coronary heart disease
No	36 (97.30)	35 (94.59)	0.347	0.556
Yes	1 (2.70)	2 (5.41)
Anticoagulant
No	36 (97.30)	36 (97.30)	0.01	0.99
Yes	1 (2.70)	1 (2.70)
CBC
Hemameba	6.14 ± 2.34	5.90 ± 1.48	0.534	0.595
Hemoglobin	124.70 ± 26.34	134.14 ± 19.14	-1.762	0.082
Blood platelet	212.76 ± 65.93	187.14 ± 60.09	1.747	0.085
Blood coagulation
PT	11.93 ± 1.10	12.92 ± 2.44	-2.239	0.58
APTT	26.96 ± 3.64	27.68 ± 3.95	-0.811	0.42
Biochemical function
Total bilirubin	26.59 ± 33.75	12.08 ± 5.75	2.578	0.012
Direct bilirubin	16.88 ± 26.77	3.71 ± 1.58	2.987	0.004
Albumin	39.70 ± 4.289	40.00 ± 2.45	-0.37	0.713
Urea nitrogen	4.61 ± 1.66	5.20 ± 1.42	-1.659	0.102
Creatinine	59.00 ± 12.43	68.03 ± 18.48	-2.465	0.058
Tumor marker
CA19-9	21.32 ± 27.53	13.92 ± 10.55	1.526	0.131

P < 0.05 indicated statistical significance. CBC: Complete blood count; PD: Pancreaticoduodenectomy; EP: Endoscopic papillectomy; PT: Prothrombin time; APTT: Activated partial thromboplastin time; CA19-9: Carbohydrate antigen 19-9.

Surgical outcomes

The surgical outcomes are shown in Table 2. The en bloc resection rate was 100% in each group. None of the patients who underwent endoscopic resection required surgical intervention owing to complications. Postoperative pathological findings revealed that 28 cases exhibited adenoma combined with carcinoma in situ, including 17 in the PD group and 11 in the EP group. No significant difference was observed in the pathological findings between the two groups (P > 0.05). To prevent postoperative stenosis, biliary stents were placed in 7 patients in the EP group and only 1 patient in the PD group (P < 0.05). To prevent postoperative pancreatitis and pancreatic duct stenosis, 75% of patients who underwent PD and 81% of patients who underwent EP had pancreatic duct stents placed, with no significant difference between the groups (P > 0.05; Table 2).

Table 2 Clinical results and complications, n (%).

	PD (n = 37)	EP (n = 37)	χ2	P value
Pathology
Adenoma	20 (54.05)	26 (70.27)	2.157	0.141
Carcinoma in situ	17 (45.95)	11 (29.73)
Biliary stent placement
No	36 (97.30)	30 (81.08)	5.045	0.025
Yes	1 (2.70)	7 (18.92)
Pancreatic stent placement
No	9 (24.32)	7 (18.92)	0.319	0.572
Yes	28 (75.68)	30 (81.08)
Complications
Hemorrhage
No	31 (83.78)	33 (89.19)	0.463	0.496
Yes	6 (16.22)	4 (10.81)
Perforation
No	35 (94.59)	36 (97.30)	0.347	0.556
Yes	2 (5.41)	1 (2.70)
Biliary fistula
No	36 (97.30)	37 (100.00)	1.014	0.314
Yes	1 (2.70)	0 (0.00)
Pancreatic fistula
No	32 (86.49)	37 (100.00)	5.362	0.021
Yes	5 (13.51)	0 (0.00)
Cholangitis
No	35 (94.59)	34 (91.89)	0.214	0.643
Yes	2 (5.41)	3 (8.11)
Pancreatitis
No	34 (91.89)	34 (91.89)	0.001	0.999
Yes	3 (8.11)	3 (8.11)
Abdominal infection
No	29 (78.38)	36 (97.30)	6.198	0.013
Yes	8 (21.62)	1 (2.70)
Mortality
No	35 (94.59)	37 (100.00)	2.056	0.152
Yes	2 (5.41)	0 (0.00)

P < 0.05 indicates statistical significance. PD: Pancreaticoduodenectomy; EP: Endoscopic papillectomy.

Adverse events

Five cases of pancreatic fistula were observed in the PD group, whereas no occurrence was noted in the EP group (P < 0.05; Table 2). There were 8 cases of abdominal cavity infection in the PD group, but only 1 in the EP group, showing a significant difference (P < 0.05). Two patients in the PD group died from surgery-related complications, whereas no mortality was observed in the EP group (P = 0.152). No significant differences were observed in complications, including bleeding, perforation, cholangitis, biliary fistula, and pancreatitis, between the two groups (all P > 0.05).

Risk factors related to complications and recurrence

Elevated serum amylase levels within 24 h postoperatively, prolonged activated partial thromboplastin time before surgery, and PD were associated with an increased risk of postoperative complications (Table 3). Follow-up data were available for 35 patients in the EP group, with a median follow-up of 24 months and a maximum of 60 months. Seven patients in the EP group experienced local recurrence; 2 underwent surgical treatment, while the remaining cases were managed with endoscopic radiofrequency ablation. The median follow-up duration for patients in the PD group was 26 months. Four patients were lost to follow-up. Among the remaining patients, 2 died from surgical complications, 1 from ovarian cancer, and 1 from cerebral infarction; in addition, 2 experienced recurrence at the choledochojejunostomy site during follow-up, and 1 patient developed distant metastasis (Table 4).

Table 3 Analysis of relevant risk factors, n (%)/mean ± SD.

	Non-complication (n = 49)	Complication (n = 25)	t/χ2	P value
Age (year)	62.63 ± 10.10	61.04 ± 9.33	0.658	0.513
Gender
Male	27 (55.10)	15 (60.0)	0.162	0.687
Female	22 (44.90)	10 (40.0)
Tumor diameter (cm)	1.95 ± 0.80	1.93 ± 0.83	0.124	0.902
Previous abdominal surgery
No	34 (69.39)	19 (76.0)	0.356	0.551
Yes	15 (30.61)	6 (24.0)
Hypertension
No	31 (63.27)	14 (56.0)	0.367	0.545
Yes	18 (36.73)	11 (44.0)
Diabetes
No	43 (87.76)	23 (92.0)	0.309	0.578
Yes	6 (12.24)	2 (8.0)
Coronary heart disease
No	46 (93.88)	25 (100.0)	1.595	0.207
Yes	3 (6.12)	0
CBC (24 h postoperatively)
Hemameba	8.60 ± 3.44	10.81 ± 4.51	-2.354	0.021
Hemoglobin	127.96 ± 18.12	124.28 ± 15.65	0.863	0.391
Blood coagulation
PT	12.53 ± 2.12	12.21 ± 1.57	0.685	0.496
APTT	26.63 ± 2.99	28.66 ± 4.78	-2.244	0.028
Pathology
Adenoma	25 (51.02)	12 (48.00)	0.06	0.806
Carcinoma in situ	24 (48.98)	13 (52.00)
Operation methods
PD	20 (40.82)	17 (68.00)	4.893	0.027
EP	29 (59.18)	8 (32.00)

P < 0.05 indicates statistical significance. CBC: Complete blood count; PD: Pancreaticoduodenectomy; EP: Endoscopic papillectomy; PT: Prothrombin time; APTT: Activated partial thromboplastin time.

Table 4 Follow-up results of different treatments.

	PD (n = 32)	EP (n = 35)	χ2	P value
Patients without recurrence	29	28	2.773	0.096
Dead patients	4	0
Living patients	25	28
Patients with recurrence	3	7	2.628	1.000
Radiofrequency ablation	0	5
Surgical treatment	0	2
Observation	3	0

P < 0.05 indicates statistical significance. PD: Pancreaticoduodenectomy; EP: Endoscopic papillectomy.

DISCUSSION

Duodenal papillary neoplasms are a heterogeneous group of lesions that originate from the biliary, duodenal, and pancreatic epithelial cells of the ampulla of Vater and are usually detected during an endoscopic examination of the upper gastrointestinal tract. Among all its subtypes, duodenal papillary adenoma is most commonly observed. Ampullary adenoma and adenocarcinoma are the benign and malignant neoplasms that arise from the glandular epithelium of the ampulla of Vater. Prophylactic excision is therefore anticipated to eliminate the risk of malignant transformation in the ampulla of Vater. However, surgical removal of ampullary adenomas remains challenging because of the complex regional anatomy.

PD involves resection of the head of the pancreas, duodenum, and part of the stomach. However, as a technically demanding procedure, PD is associated with a high risk of postoperative complications, such as intraperitoneal hemorrhage, infection, biliary fistula, pancreatic fistula, and gastroparesis[12,13]. Consequently, it is contraindicated in individuals with advanced age or those with poor physical condition. In our study, 2 patients with papillary adenoma in the PD group died of postoperative complications, while others experienced varying degrees of postoperative abdominal infections, pancreatitis, and bleeding. Therefore, there is an urgent need for a minimally invasive treatment approach that offers greater efficacy with fewer complications for the management of duodenal papillary adenoma.

TDA is a less invasive alternative for the excision of papillary neoplasms that does not require the removal of surrounding organs of the lesion[14], thus offering a faster postoperative recovery, shorter hospital length of stay, and higher long-term success rate (83.6%)[7]. However, owing to the high recurrence rate following TDA and the persistence of complications such as bile fistula and intra-abdominal infection, this approach has not gained widespread acceptance in clinical practice. Consequently, most surgeons continue to prefer PD for the management of lesions located in the ampulla of Vater.

Endoscopic interventions are emerging modalities to avoid the need for surgery[15-17]. EP is an endoscopic intervention to treat tumors of the duodenal papilla[18]. Compared with PD and TDA, EP is associated with significantly lower morbidity and mortality[19-21]. Previous studies mainly focus on the short-term outcomes of EP, and long-term follow-up data are limited. In the present study, patients with duodenal papillary adenoma were followed up for a median of 24 months, with the longest follow-up of 60 months. Seven patients experienced recurrence after EP during the follow-up period. No serious postoperative complications were reported. The integrity of the digestive tract was intact, and patients typically maintained a high quality of life. In contrast, 2 patients died of postoperative complications in the PD group. Postoperative abdominal infection, pancreatic fistula, pancreatitis, and other serious complications were also more frequent in the PD than the EP group. Notably, EP also provides a valuable therapeutic option for recurrent cases requiring repeated intervention. Some patients achieved complete remission through repeat endoscopic resection or its combination with radiofrequency ablation[22,23].

In patients with appropriate indications, EP can minimize the risks of local recurrence and metastasis. Papillary adenoma is a benign disease, and thus is the best indication for endoscopic resection. However, papillary lesions with intraductal extension usually recur. We therefore recommend endoscopic ultrasound to exclude submucosal invasion, along with computed tomography or magnetic resonance imaging to assess regional lymph nodes in patients with suspected malignant lesions[24,25]. A previous study demonstrated that endoscopic ultrasound has a specificity of 88% and a negative predictive value of 90% for evaluating the infiltration depth of duodenal adenomas[26]. Although preoperative imaging plays a critical role in determining the indications for EP, postoperative pathological analysis remains necessary. We suggest processing EP specimens using endoscopic submucosal dissection to ensure accurate histopathological assessment. If intravascular tumor thrombus or submucosal invasion is identified, surgical intervention is warranted[27].

Several limitations of this study should be acknowledged. First, this was a single-center retrospective study, which may be subject to selection bias. Second, duodenal papillary adenoma is a rare condition, and the relatively small sample size may have influenced the reliability of our findings. Further large-scale multicenter randomized controlled trials are required to validate the findings of this research.

CONCLUSION

EP is safe and effective for treating duodenal papillary adenoma, offering a lower recurrence rate and better long-term prognosis than standard PD. In cases of post-EP recurrence, repeat endoscopic treatment or surgical intervention may be considered as appropriate therapeutic options.

ACKNOWLEDGEMENTS

The authors thank the members of the Department of Gastroenterology, Drum Tower Hospital, Nanjing University School of Medicine, for their valuable contributions to the discussion and critical review of the manuscript.