- Citado por SciELO
versión impresa ISSN 1130-0108
Rev. esp. enferm. dig. vol.96 no.12 dic. 2004
Transduodenal ampullectomy in the treatment of villous adenomas and adenocarcinomas of the Vater's ampulla
J. A. Fraguela Mariña
General Surgery Department "A". Complejo Hospitalario Universitario Juan Canalejo. A Coruña. Spain
Introduction: adenomas are the most frequent tumors of the Vater's ampulla. Their capacity for malignant transformation following the adenoma-carcinoma sequence is well known. It is because of this that resection after diagnosis is required. The identification of the appropriate technique according to tumor features would require that patients not be undertreated or overtreated, which would give rise to serious consequences derived from their location.
Patients and methods: villous adenomas and adenocarcinomas of the Vater's ampulla candidates for local resection were revised from January 1st, 1998 through June 30th, 2003. We describe the methods of diagnosis and ampulectomy techniques we performed.
Results: we performed an ampulectomy by first intention in all 8 patients included in this study. However, pancreatoduodenectomy was necessary in two patients because of the closeness of resection margins. We had no mortality in this series, and morbidity was limited to two episodes of digestive bleeding that were controlled by electrocoagulation and embolization. The mean follow-up was 28.5 months (range, 6-72 months).
Conclusions: the difficulty of precise preoperatory diagnosis in adenomas of the Vater's ampulla demands resection after identification. Ampulectomy is the treatment of choice for villous adenomas and T1 adenocarcinomas, with 1 cm of resection margin to avoid local recurrence.
Key words: Vater's ampulla. Papillary tumor. Ampullectomy.
Fraguela Mariña JA. Transduodenal ampullectomy in the treatment of villous adenomas and adenocarcinomas of the Vater's ampulla. Rev Esp Enferm Dig 2003; 96: 829-834.
Correspondencia: José Antonio Fraguela Mariña. C/ General Sanjurjo, 48, portal 3, 4º D. 15006 A Coruña. Tel.: 981 178 166. e-mail: firstname.lastname@example.org
In 1720, Abraham Vater described a new anatomic entity that today is called the Vater's ampulla (1). It is formed in most individuals by the union of the terminal segments of the pancreatic and common bile ducts. In 25% of individuals, the ampulla is formed by the ending of the common duct, having the Wirsung's duct its own entry point into the duodenum just adjacent to the ampulla. The ampulla opens into the duodenum through a small mucosal elevation known as the papilla of Vater. Tumors of the Vater's ampulla are called ampulomas and represent 2% of all the tumors of the gastrointestinal tract (2). The two most usual tumors are adenomas (about 70% villous adenomas) and adenocarcinomas (2,3).
The aim of this study is to attempt to identify patients that would benefit from transduodenal ampullectomy in the treatment of ampullary tumors, showing our experience in surgical ampullectomy in the treatment of villous adenomas and adenocarcinomas of the ampulla of Vater.
PATIENTS AND METHODS
Patients with villous adenomas and adenocarcinomas of the ampulla of Vater liable to local transduodenal resection and undergoing ampullectomy were retrospectively reviewed from January 1st, 1998 to June 30th, 2003 within the General Surgical Department "A" of the "Juan Canalejo Hospital", A Coruña, Spain. In that period of time, 28 patients underwent surgery with a diagnosis of adenocarcinoma of the Vater's ampulla and three villous adenomas of the papilla, one of them associated with an adenocarcinoma. Eight patients were treated by transduodenal ampullectomy, five males and three females. In the remaining patients we performed a pancreatoduodenectomy (PD). The mean age at presentation for patients undergoing transduodenal ampullectomy was 67 years (52-78 years). The two most common presenting symptoms were abdominal pain, in 5 cases, and jaundice, in 3. The presence of these symptoms led us to perform hepatic function tests and an abdominal ultrasonography. Endoscopic retrograde cholangiography (ERCP) showed a tumor of the ampulla in all cases, and biopsy confirmed the diagnosis. An abdominal computed tomography (CT) was further performed in all patients, and carcinoembryonic antigen (CEA) levels were measured. None of them was studied by endoscopic ultrasonography.
All of our patients were operated on by a right subcostal incision and, after cholecystectomy, the biliary duct was catheterized with a Fogarty catheter through the cystic duct in order to identify the papilla. After vertical duodenotomy a submucosal injection of 1/100.000 adrenaline was administered in order to avoid bleeding, and a resection of the papilla was made leaving 1-cm margins in all levels, including the pancreatic margin. After that, the bile duct and the pancreatic duct were reinserted with interrupted synthetic absorbable sutures (5-0), and then the duodenotomy was closed. Frozen sections of the resection margins were obtained during surgery. If the tumor was close to the resection margin, a PD was performed as a Whipple's procedure. Locoregional lymph node dissection was performed only after PD.
Hepatic function tests showed cholestasis in all cases. A dilatation of the bile duct was demonstrated in all patients by ultrasonography, but its origin was not identified. CEA was always normal, and the extension study by CT was normal as well. ERCP was diagnostic in 100% of cases, and the diagnostic accuracy of biopsy was correct in two patients with villous adenomas and in five with adenocarcinomas; a false negative diagnosis of villous adenoma was made in a patient with an adenocarcinoma.
All patients underwent ampullectomy; the tumor was near the resection margin in 3 of 6 adenocarcinomas, and surgery was completed by means of PD in two cases; the third patient was a poor candidate for this type of surgery; therefore, the operation was not completed, and the patient died 13 months later. All patients were informed on the possibility of a PD; six of them agreed, but a patient with a high surgical risk refused, as did another patient with an initial diagnosis of villous adenoma in whom, resection confirmed the presence of an adenocarcinoma.
None of the patients died, but two patients developed upper gastrointestinal bleeding (UGB), which were solved by endoscopic electrocoagulation 24 hours after surgery in one patient, and by embolization of the gastroduodenal artery in the other one.
Follow-up ranged from 6 to 72 months. One patient with PD because of a resection margin close to a moderately differentiated tumor showed 58 months after surgery a metastasis at the abdominal wall, which was resected; 12 months later a new metastasis in the right iliac chain with compression of the crural nerve was also removed. The other patients remained without symptoms (Table I).
Ampullary tumors include a wide variety of pathological entities. Adenoma is the most frequent tumor, and its progression to villous adenoma and adenocarcinoma is admitted, much in the same way as the adenoma-carcinoma sequence in colon polyps. Villous adenomas of the Vater's ampulla are considered premalignant lesions, their resection being indicated to prevent the development of adenocarcinomas, which takes place in 25 to 40% of patients. However, 40% of cases are "in situ" carcinomas (4,5), as it was the case in one of our three patients (33%). There is no single treatment desirable for all lesions; moreover, not all adenocarcinomas require PD, and not all benign lesions are suitable for local resection (5). An appropriate selection of those patients in whom local resection would be safe and effective would allow this suitable treatment, depending on the kind of tumor.
The pathological characteristics of the papilla and the difficulties of biopsy sampling make a histologically accurate diagnosis at the time of endoscopy not an easy task. Macrobiopsies and a brushing cytology help in diagnostic accuracy, but a suitable classification requires a complete resection of the tumor and its histological examination, since the presence of an "in situ" or infiltrating carcinoma within an adenoma cannot be excluded without complete resection (5-7). In our series, only one patient (case 7) was first diagnosed as having a benign tumor and then turned out to reveal a malignant lesion after total resection, which confers us a diagnostic accuracy of 87.5%. Endoscopy and cholangioresonance, performed before sphincterotomy and endoprosthesis insertion to minimize interpretation mistakes, allow lesion staging, overall and regarding depth of penetration into the duodenal wall and the pancreas. However, overstaging due to submucosal edema from associated pancreatitis has been reported in as much as one third of lesions. Both techniques are less helpful in accurately determining the presence or absence of lymph node metastases (8,9). However, they would be able to select patients who may be optimally managed by endoscopic resection. In our patients we performed an extension study by means of CT to identify parameters against surgery. Biopsy at the time of surgery with a total resection of the lesion allowed a final diagnosis and clearly revealed the depth of tumor penetration into the wall, with no problems regarding biopsy interpretation in patients with a previous papillotomy (6), perhaps because all patients underwent surgery 3 to 7 days after papillotomy, thus minimizing interpretation errors due to regenerative changes.
It is considered nowadays that all ampulla tumors should be resected because of their risk of malignant transformation (10). Controversy arises as to what kind of resection is suitable for each tumor and each patient.
Ampullectomy was first described by Halsted in 1899 (11). After a golden age at the beginning of the last century, it was progressively abandoned due to postoperative complications and mortality, a high rate of recurrences, and the introduction of PD. Nowadays it is associated with a lower rate of morbidity and mortality, and with less hospitalizations than PD (5,7,12); moreover, local recurrence rates are low if indication and surgical technique are suitable, thus allowing a safe resection margin of 1 cm (7) in adenocarcinomas and quite lower in villous adenomas. Endoscopic snare resection does not allow a wide resectional margin on the pancreatic side, and the possibilities of local recurrence are higher, which restricts its indication as a palliative treatment for patients unfit for surgery or rejecting it (13). A wide series showed that the most important prognostic factor for survival was complete tumor resection (14-16). In villous adenomas and adenocarcinomas, pT1 ampullectomy results in the same, or even better, overall outcomes than PD (7,17). We performed it in 6 patients, and a follow-up between 8 and 60 months showed no evidence of recurrence. However, an episode of upper gastrointestinal bleeding occurred in two patients. This has been reported as a complication quite often (10) probably because of the use of adrenaline, which makes the resection easier but may result in rebleeding once metabolized. When the resection margin was affected or close to the tumor, which usually occurs with poorly or moderately differentiated tumors, we performed a PD with a previous informed consent of the patients. Case 7 refused to go on with PD if there was an indication for that procedure; so, although the resection margin was lower than 1 cm, we performed only an ampullectomy. Eight months after being operated on he remains without symptoms, and had no signs of recurrence during an endoscopy performed at 6 months.
The Vater's ampulla has a distinct pattern of lymphatic drainage, and in contrast to pancreatic tumors, ampullary tumors spread into the nearby retroduodenal nodes, even in non-advanced cases. Because of this, some authors advocate for both local lymphadenectomy and ampullectomy in order to improve prognosis (18,19). Only when local nodes are involved in T1 poorly differentiated tumors with suitable resectional margins, PD would be justified as a complementary treatment (7). For T2 well differentiated tumors with suitable resection margins and no nodal involvement, ampullectomy would be probably chosen. Pancreatic involvement will lead to PD with lymphadenectomy, because of lymphatic spread after pancreatic dissemination, as it happened in two of our patients.
Endoscopic follow-up after ampullectomy is important, and we performed endoscopy every six months during the first three years, and yearly thereafter in order to detect recurrences that might lead to PD.
To conclude, we may say that transduodenal ampullectomy is an accurate treatment for villous adenomas of the papilla if resection is complete, and it may be a safe treatment for well-differentiated adenocarcinomas whenever we succeed in having a safe resection margin of at least 1 cm at all levels.
1. Vater A. Disertatio anatomica qua novum bilis diverticulum circa orificium ductus choledochi, at et valvulosum colli vesical fellae constructionem ad discepterdum proponit. Wittenberg, Lit Gerdesianis, 1720. p. 2-32. [ Links ]
2. Seifert W, Schulte F, Stolte M. Adenoma and carcinoma of the duodenum and papilla of Vater: a clinicopathologic study. Am J Gastroenterol 1992; 87: 37-41. [ Links ]
3. Sharp KW, Brandes JL. Local resection of tumors of the ampulla of Vater. Am Surg 1990; 56: 214-7. [ Links ]
4. Seifert E, Schulte F, Stolte M. Adenoma and carcinoma of the duodenum and papilla of Vater: a clinicopathologic study. Am J Gastroenterol 1992; 87: 37-42. [ Links ]
5. Rattner DX, Fernández C, Brugger WR, Warshaw AL. Defining the criteria for local resection of ampullary neoplasms. Arch Surg 1996; 131: 366-71. [ Links ]
6. Rodríguez C, Borda F, Elizalde I, Jiménez FJ, Carral D. ¿Hasta qué punto es seguro el diagnóstico preoperatorio de la biopsia endoscópica en el tumor ampular? Rev Esp Enferm Dig 2002; 94 (10): 585-8. [ Links ]
7. Beger H, Treitschke F, Gansauge F, Harada H, Hiki N, Mattfeldt T. Tumor of the ampulla of Vater. Experience with local or radical resection in 171 consecutively treated patients. Arch Surg 1999; 134: 526-32. [ Links ]
8. Souquet JC, Napoleon B, Pujol B, Ponchon T, Keriven O, Lambert R. Ecoendoscopy prior to endoscopy tumor therapy: more safety? Endoscopy 1993; 25: 475-8. [ Links ]
9. Géier A, Nguyen HN, Gartung C, Matern S. MRCP and ERCP to detect small ampullary carcinoma. Lancet 2000; 356 (9241): 1607-8. [ Links ]
10. Chareton B, Coiffic J, Landen S, Bardaxoglou E, Campion JP, Launois B. Diagnosis and therapy for ampullary tumors: 63 cases. World J Surg 1996; 20: 707-12. [ Links ]
11. Halsted W. Contributions to the surgery of the bile passages especially of the common bile duct. Boston Med Surg 1889; 141: 645-54. [ Links ]
12. Farouk M, Niotis M, Branum G, Cotton PB, Meyers WC. Indications for and the technique of local resection of tumors of the papila of Vater. Arch Surg 1991; 126: 650-2. [ Links ]
13. Tokunaga Y, Hosogi H, Hoppou T, Nagakami M, Tokuka A, Ohsumi K. A case of ampullary carcinoma successfully managed with endoscopic snare resection. Surg Laparosc Endosc Percutan Tech 2002; 12 (4): 273-8. [ Links ]
14. Monson JRT, Donohue JH, McEntee GP. Radical resection for carcinoma of the ampulla of Vater. Arch Surg 1991; 126: 353-7. [ Links ]
15. Matory YL, Gaynor J, Brennan M. Carcinoma of the ampulla of Vater. Surg Gynecol Obstet 1993; 177: 366-70. [ Links ]
16. Allema JH, Reinders ME, van Gulik TM. Results of pancreaticoduodenectomy for ampullary carcinoma and analysis of prognostic factors for survival. Surgery 1995; 117: 247-53. [ Links ]
17. Asbun HJ, Rossi RL, Munson Jl. Local resection for ampullary tumors: is there a place for it? Arch Surg 1993; 128: 515-20. [ Links ]
18. Hanyu F, Imaizumi T, Nakamura M, Yoshikawa T. Proposed definition of early cancer of the papilla of Vater. J Biliary Tract Pancreas 1984; 5: 847-52. [ Links ]
19. Willet CG, Warshaw AL, Convery K, Compton CC. Patterns of faillure after pancreaticoduodenectomy for ampullary carcinoma. Surg Gynecol Obstet 1993; 176: 33-8. [ Links ]