Adenomas of the ampulla of Vater: a comparison of outcomes of operative and endoscopic resections.

BACKGROUND: Data comparing operative and endoscopic resection of adenomas of the ampulla of Vater are limited. Our aims were to evaluate and compare the long-term results and outcomes of endoscopic and operative resections of benign tumors of the ampulla of Vater as well as to determine which features of benign periampullary neoplasms would predict recurrence or failure of endoscopic therapy and therefore need for operative treatment.
METHODS: Retrospective review of all patients treated for adenomas of ampulla of Vater at our institution from 1994 to 2009.
RESULTS: Over a 15-year span, 180 patients (mean age 59 years) were treated for benign adenomas of the ampulla of Vater with a mean follow-up of 4.4 years. Obstructive jaundice was more common in the operative resection group (p = 0.006). The adenomas were tubular in 83 patients (44%), tubulovillous in 77 (45%) and villous in 20 (11%). Endoscopic resection alone was performed in 130 patients (78%). Operative resection was performed in 50 patients (28%), with pancreatoduodenectomy in 40, transduodenal local resection in 9, and pancreas-sparing total duodenectomy in 1. Nine patients who underwent endoscopic resection initially were endoscopic treatment failures. Fifty-eight percent of endoscopically treated patients required one endoscopic resection, while 58 (42%) required two or more endoscopic resections (range 2-8). Patients who underwent operative resection had larger tumors with a mean size of 3.7 ± 2.8 versus 1.8 ± 1.5 cm in those treated by endoscopic resection (p < 0.001) or intraductal extension (p = 0.02). Intraductal extension and ulceration had no effect on recurrence if completely resected endoscopically (p = 0.41 and p = 0.98, respectively). Postoperative complications occurred in 58% of patients, and post-endoscopic complications in 29% (p < 0.001). Endoscopic resection was associated with a greater than fivefold risk of recurrence than operative resection (p = 0.006); 4% of recurrences had invasive carcinomas. When comparing patients who underwent local resections only (endoscopic and operative), there was no difference in the recurrence rate between endoscopic resection and operative transduodenal resection (32 versus 33%; p = 0.49). The need for two or more endoscopic resections for complete tumor removal was associated with 13-fold greater risk of recurrence (p < 0.001).
CONCLUSION: There is no significant difference between endoscopic and local operative resections of benign adenomas of ampulla of Vater; recurrences are more common when two or more endoscopic resections are required for complete tumor removal. Appropriate adenomas for endoscopic resection included tumors <3.6 cm that do not extend far enough intraductally (on EUS) to preclude an endoscopic snare ampullectomy.