Radical Resection for Carcinoma of the Ampulla of Vater John R. T. Monson, MD, FRCSI, FRCS; John H. Donohue, MD; Gerard P. McEntee, MD, FRCSI; Donald C. Mcllrath, Jon A. van Heerden, MB, FRCSC; Roy G. Shorter, MD, FRCP; David M. Nagorney, MD; Duane M. Ilstrup, MS
\s=b\ One hundred four consecutive patients who underwent radical resection for ampullary cancer between 1965 and 1989 were retrospectively reviewed. Frequent clinical findings included jaundice (67%), significant (>10%) weight loss (42%), and anemia (27%). Eighty-seven patients (84%) underwent a subtotal pancreatectomy, and 17 patients (16%) underwent a total pancreatectomy. The postoperative mortality was 5.7% (six patients), and reoperation for postoperative complications was required in six patients. The 5-and 10-year survival rates were 34% and 25%, respectively. Eight patients died of tumor recurrence more than 5 years after resection. Patient survival was significantly impaired by microscopic lymphatic invasion, regional nodal metastasis, tumor grade, and the epithelium of origin. In a multivariate analysis, only microscopic lymphatic invasion significantly reduced patient survival. Radical resection for ampullary cancer can be performed with a low morbidity and mortality and should remain the procedure of choice for ampullary carcinoma.
(Arch Surg. 1991;126:353-357)
periampullary Thesenting junction
is anatomically complex, repre¬ the of three different epithelia: the pan¬ creatic and bile ducts and the duodenal mucosa. Tumors of the ampulla of Vater may, therefore, arise from any one of these epithelia.13 Macroscopically, the epithelium of origin is often impossible to determine, and these tumors are all classified as periampullary carcinomas. Regardless of the epithelium of origin, most authorities would favor similar curative surgical management, that is, radical resection. Traditionally, carci¬ noma of the ampulla has been considered an entity distinct from tumors arising from the head of the pancreas, but some authors have grouped this condition with pancreatic ductal
consecutive patients who had undergone radical resection for
ampullary cancer.
PATIENTS AND METHODS
Retrospective Analysis A retrospective analysis of all patients who had undergone radical resection for carcinoma of the ampulla of Vater at the Mayo Clinic, Rochester, Minn, between 1965 and 1989 was performed. We ex¬ cluded (1) patients who had undergone local tumor excision or biliary bypass and (2) patients for whom a full histologie review was not possible. Patient histories were reviewed for demographic data, details of preoperative symptoms and investigations, surgical treat¬ ment, postoperative complications, and follow-up. Complete followup was available by combining chart review, postal questionnaires, and direct telephone contact. A detailed histologie review was per¬ formed in all cases by a single pathologist (R.G.S.) and included evaluation of lymphatic and blood vessel invasion, lymph node in¬ volvement, and an estimate of the epithelium of origin.13'1
Statistical Analysis
area
•
carcinoma.
4-6
Recently, much attention has focused on the improved outcome of patients who are undergoing radical resection for pancreatic carcinoma with both reduced perioperative mor¬ bidity and improved long-term survival.6"12 We wondered whether patients with the clinical diagnosis of ampullary cancer have had a similar improvement in outcome. To deter¬ mine if such
a
trend has occurred and which factors affected
patient survival, we retrospectively reviewed a series of 104 Accepted for publication November 17,1990. From the Departments of Surgery (Drs Monson, Donohue, McEntee, McIlrath, van Heerden, and Nagorney), Pathology (Dr Shorter), and Biostatistics (Mr Ilstrup), Mayo Clinic and Foundation, Rochester, Minn. Read before the 43rd Annual Cancer Symposium of the Society of Surgical Oncology, Washington, DC, May 21,1990. Reprint requests to Department of Surgery, Mayo Clinic, Rochester, MN 55905 (Dr Donohue).
MD;
Survival rates were calculated with the use of an actuarial method, and comparison between survival curves were made with the use of either the log-rank test or the Mantel-Haenszel test.15,16 Univariate and multivariate analysis with the use of the Cox proportional haz¬ ards model17 was performed.
RESULTS Between 1965 and 1989, 104 patients were identified who underwent radical resection for a primary tumor arising from the ampulla of Vater. There were 59 men and 45 women who varied in age from 35 to 85 years, with a mean age of 64 years. Before operation, the mean duration of symptoms reported by the patients was 7.8 weeks. At the time of presentation, the majority of patients were icteric, although one third (34 patients) were not jaundiced either clinically or biochemical¬ ly. In contrast, only 28 patients (27%) were anemic (hemoglo¬ bin level,
\s=b\ One hundred four consecutive patients who underwent radical resection for ampullary cancer between 1965 and 1989 were retrospectively reviewed. Frequent clinical findings included jaundice (67%), significant (>10%) weight loss (42%), and anemia (27%). Eighty-seven patients (84%) underwent a subtotal pancreatectomy, and 17 patients (16%) underwent a total pancreatectomy. The postoperative mortality was 5.7% (six patients), and reoperation for postoperative complications was required in six patients. The 5-and 10-year survival rates were 34% and 25%, respectively. Eight patients died of tumor recurrence more than 5 years after resection. Patient survival was significantly impaired by microscopic lymphatic invasion, regional nodal metastasis, tumor grade, and the epithelium of origin. In a multivariate analysis, only microscopic lymphatic invasion significantly reduced patient survival. Radical resection for ampullary cancer can be performed with a low morbidity and mortality and should remain the procedure of choice for ampullary carcinoma.
(Arch Surg. 1991;126:353-357)
periampullary Thesenting junction
is anatomically complex, repre¬ the of three different epithelia: the pan¬ creatic and bile ducts and the duodenal mucosa. Tumors of the ampulla of Vater may, therefore, arise from any one of these epithelia.13 Macroscopically, the epithelium of origin is often impossible to determine, and these tumors are all classified as periampullary carcinomas. Regardless of the epithelium of origin, most authorities would favor similar curative surgical management, that is, radical resection. Traditionally, carci¬ noma of the ampulla has been considered an entity distinct from tumors arising from the head of the pancreas, but some authors have grouped this condition with pancreatic ductal
consecutive patients who had undergone radical resection for
ampullary cancer.
PATIENTS AND METHODS
Retrospective Analysis A retrospective analysis of all patients who had undergone radical resection for carcinoma of the ampulla of Vater at the Mayo Clinic, Rochester, Minn, between 1965 and 1989 was performed. We ex¬ cluded (1) patients who had undergone local tumor excision or biliary bypass and (2) patients for whom a full histologie review was not possible. Patient histories were reviewed for demographic data, details of preoperative symptoms and investigations, surgical treat¬ ment, postoperative complications, and follow-up. Complete followup was available by combining chart review, postal questionnaires, and direct telephone contact. A detailed histologie review was per¬ formed in all cases by a single pathologist (R.G.S.) and included evaluation of lymphatic and blood vessel invasion, lymph node in¬ volvement, and an estimate of the epithelium of origin.13'1
Statistical Analysis
area
•
carcinoma.
4-6
Recently, much attention has focused on the improved outcome of patients who are undergoing radical resection for pancreatic carcinoma with both reduced perioperative mor¬ bidity and improved long-term survival.6"12 We wondered whether patients with the clinical diagnosis of ampullary cancer have had a similar improvement in outcome. To deter¬ mine if such
a
trend has occurred and which factors affected
patient survival, we retrospectively reviewed a series of 104 Accepted for publication November 17,1990. From the Departments of Surgery (Drs Monson, Donohue, McEntee, McIlrath, van Heerden, and Nagorney), Pathology (Dr Shorter), and Biostatistics (Mr Ilstrup), Mayo Clinic and Foundation, Rochester, Minn. Read before the 43rd Annual Cancer Symposium of the Society of Surgical Oncology, Washington, DC, May 21,1990. Reprint requests to Department of Surgery, Mayo Clinic, Rochester, MN 55905 (Dr Donohue).
MD;
Survival rates were calculated with the use of an actuarial method, and comparison between survival curves were made with the use of either the log-rank test or the Mantel-Haenszel test.15,16 Univariate and multivariate analysis with the use of the Cox proportional haz¬ ards model17 was performed.
RESULTS Between 1965 and 1989, 104 patients were identified who underwent radical resection for a primary tumor arising from the ampulla of Vater. There were 59 men and 45 women who varied in age from 35 to 85 years, with a mean age of 64 years. Before operation, the mean duration of symptoms reported by the patients was 7.8 weeks. At the time of presentation, the majority of patients were icteric, although one third (34 patients) were not jaundiced either clinically or biochemical¬ ly. In contrast, only 28 patients (27%) were anemic (hemoglo¬ bin level,
