Endoscopic vs. Surgical Ampullectomy: an Algorithm to Treat Disease of the Ampulla of Vater

Eugene P Ceppa, MD, Rebecca A Burbridge, MD, Kristy L Rialon, MD, Philip A Omotosho, MD, Dawn Emick, MD MPH, Paul S Jowell, MD, M S Branch, MD, Theodore N Pappas, MD. Duke University Medical Center

Introduction: Proposed management of benign ampullary lesions includes local resection (endoscopic or surgical ampullectomy) and en-bloc resection (pancreaticoduodenectomy). Most agree that en-bloc resection entails a significant morbidity and mortality. The objective of this study was to compare the effectiveness, morbidity, and mortality of endoscopic ampullectomy (EA) and surgical ampullectomy (SA). No study has previously compared EA versus SA for the treatment of benign ampullary lesions.

Methods: Medical records of patients selected for ampullectomy at Duke University Medical Center from 1991-2010 were reviewed. Results: After review, 109 patients were confirmed to have undergone ampullectomy for a suspected benign ampullary lesion. Sixty-eight patients underwent EA, while 41 patients underwent SA. Patients in each group were identical in terms of age, sex, race, and comorbid conditions, except EA had a higher rate of morbid obesity (BMI > 35). EA was found to have a significantly reduced length of stay, lower morbidity and readmission rates, but had similar rates of mortality, margin-positive excisions, and re-interventions (Table 1).

Conclusion: In patients selected for ampullectomy for benign ampullary lesions, EA was found to be equally effective, have a lower morbidity and identical mortality when compared to SA. These findings suggest that patients would likely benefit from an aggressive endoscopic approach prior to consideration for surgery.

Table 1.
Patient Data EA(n = 68) SA(n = 41) P Value
Length of Stay
(days) 0.6 + 0.2 10.1 + 1.0 0.0000
Mortality 0% 0% n/a
Complications 18% 42% 0.006
Readmission 16% 34% 0.03
Margins Positive 20% 10% 0.19
Re-intervention 26% 15% 0.15


Session: Resident/Fellow
Program Number: S115

« Return to SAGES 2011 abstract archive