Prognostic Factors of Ampulla of Vater Carcinoma after Radical Surgery ====================================================================== * Dongbing Zhao * Yongkai Wu * Yi Shan * Chengfeng Wang * Ping Zhao ## Abstract **OBJECTIVE** Ampullary carcinoma is a rare disease with better prognosis than other periampullary neoplasms. This study investigated the association between clinicopathologic factors and prognosis after radical resection of ampulla of Vater carcinoma. **METHODS** Clinical data from 105 patients who underwent radical pancreaticoduodenectomy from January 1990 to December 2005 were retrospectively analyzed by the Kaplan-Meier method, log-rank test, and the Cox proportional hazard model. **RESULTS** The in-hospital mortality rate was 8.6%, the lymph node metastasis rate was 37.1%, and the five-year survival rate was 42.8%. Pancreatic involvement (*P* = 0.027), tumor diameter (*P* = 0.008), T stage (*P* = 0.003), TNM stage (*P* < 0.001), and number of metastatic lymph nodes (*P* < 0.001) were associated with prognosis when the univariate analysis was used. Multivariate analysis showed that the number of lymph node metastases (*P* < 0.001; OR: 1.923; CI: 1.367-2.705) and tumor diameter (*P* = 0.03; OR: 1.432; CI: 1.035-1.981) were the independent prognostic factors. **CONCLUSION** The number of metastatic lymph nodes and tumor diameter are important pathologic factors predicting prognosis of ampulla of Vater carcinoma after radical resection, and lymph node dissection during the radical surgery effectively improves the survival rate. KEY WORDS: * ampulla of Vater * lymph node * metastasis * carcinoma * prognosis ## Introduction Usually the ampulla of Vater is formed by the union of the common bile and pancreatic ducts in the middle/posterior inner wall of the descending duodenum. It opens at the duodenal papilla, which is also known as the Vater nipple. Cancer can occur in the ampulla or on the nipple. Histologically, the tumor can originate from the pancreatic duct epithelium, the epithelium of the common bile duct, or the duodenal mucosa. The incidence rate of ampulla of Vater carcinoma among all the periampullary cancers is lower than that of pancreatic head carcinoma. The standard surgical treatment procedure is to perform pancreaticoduodenectomy. However, pancreaticoduodenectomy has certain postoperative complications and mortality, and one study has reported that the effects of partial tumor resection are similar to those of pancreaticoduodenectomy while cause fewer postoperative complications and resulting in a good quality of life[1]. On the other hand, partial tumor resection cannot thoroughly dissect regional lymph nodes and may leave residual sites of metastasis. To provide additional guidance for selection of surgical methods, we have studied characteristics of lymph node metastasis and factors affecting prognosis of ampulla of Vater carcinoma after radical surgery. ## Materials and Methods ### Clinical data From January 1990 to December 2005, 105 patients with a pathologic diagnosis of ampulla of Vater carcinoma received radical resection. Fifty-eight patients were men, and 47 were women (gender ration of 1.23: 1). The ages at treatment ranged from 30-78 years with a median age of 52 years. The presenting symptom of jaundice accounted for 79.1% (83/105). All patients received routine ultrasound examination; 91 patients had computed tomography (CT) scan; and 45 had magnetic resonance imaging (MRI) or magnetic resonance cholangiopancreatography (MRCP) examination. Low-tension duodenal visualization or endoscopic retrograde cholangiopancreatography (ERCP) examination was performed in 21 patients, and duodenoscopic or endoscopic ultrasound examination was performed in 26 patients. Forty-nine patients were tested for the tumor marker CA199, and the positive rate was 77.6%. Sixty-nine patients were tested for CEA, with a 15.9% positive rate. ### Surgical and pathological examinations All of the 105 patients underwent traditional pancreaticoduodenectomy (PD). The resection areas included the greater part of the stomach, part of the omentum, the head of the pancreas and the duodenum, about 15 cm of the upper jejunum, the lower part of the common bile duct, and regional lymph nodes. The gallbladder was retained for some patients. When enlarged lymph nodes were found around the superior mesenteric artery and abdominal aorta, selective dissection was conducted. The pathological diagnoses of the tissue specimens were made by two pathologists at our hospital. All patients had Radical resection, and the pathologic staging was determined using AJCC (2002) standards. Histologic types included adenocarcinoma in 101 patients, mucinous adenocarcinoma in 3 patients, and adenosquamous carcinoma in 1 patient. ### Complications Among the 105 patients with carcinoma of the ampulla of Vater who underwent the surgery, 9 (8.6%) hospital deaths. The causes of death were digestive tract or abdominal hemorrhage (4 patients), infection of the lung or abdominal cavity (2 patients), liver and kidney failure (1 patient), cerebral vascular accident (1 patient), and suicide (1 patient). Complications included digestive tract or abdominal hemorrhage in 10 patients, biliary fistula in 4, pancreatic fistula in 3, gastric emptying disorder in 4, incision infection in 3, intestinal obstruction in 2, gastrointestinal stoma stenosis in 1, and intestinal fistula in 1 patient. These patients recovered after conservative or surgical treatment. ### Statistical analysis Factors including patients’sex, age, medical history, the degree of jaundice, tumor diameter, histologic type, degree of differentiation, intestinal wall infiltration, T staging, and pancreatic involvement were entered into the computer database. Associations between the clinicopathologic parameters and lymph node metastasis were analyzed using χ2 test. The Kaplan-Meier method was used for calculation of the survival rate and the single-factor analysis of prognostic factors. The stepwise regression method was used to enter the factors that may affect prognosis into the Cox proportional hazard model for the multivariate analysis. SPSS 13.0 was used for the statistical analysis. ## Results ### Sites of lymph node metastasis and affecting factors Among the 105 patients who received surgery for carcinoma of the ampulla of Vater, lymph node metastasis was found in 39 (37.1%) in the postoperative pathological examination. A total of 903 lymph nodes were removed from the surgery, with an average of 8.6 per patient. The number of positive metastatic lymph nodes was 100. The cumulative frequency distribution of lymph node metastasis sites was: 61.5% (24/39) around the pancreas, 35.9% (14/39) adjacent to the duodenum, 10.3% (4/39) in the mesentery, 5.1% (2/39) in the pyloric area, 5.1% (2/39) adjacent to the abdominal aorta, 7.7% (3/39) beside the common bile duct, 2.6% (1/39) beside the celiac artery, and 2.6% (1/39) beside the arteria hepatica communis. Twenty, 4, 5, and 10 patients had 1, 2, 3, and more than 4 metastatic lymph nodes, respectively. The site distribution of single lymph node metastasis found in 20 patients was as follows: adjacent to the duodenum in 8 patients, around the head of the pancreas in 9, in the arcus minor ventriculi in 1, beside the common bile duct in 1, and in the mesentery in 1. Single lymph node metastasis around the pancreas and duodenum accounted for 85% (17/20). According to the chi-square test, the factors associated with lymph node metastasis were tumor diameter (*P* = 0.013), the degree of tumor differentiation (*P* = 0.046), T-stage (*P* < 0.001), pancreatic involvement (*P* = 0.014), and vascular tumor thrombus (*P* = 0.019) (Table 1). View this table: [Table 1.](http://www.cancerbiomed.org/content/6/2/85/T1) Table 1. Relationship between lymph node metastasis in ampulla of Vater carcinoma and clinicopathologic parameters (*n* = 105) ### Univariate and multivariate analysis All of the patients were followed-up through February 2007. The five-year cumulative survival rate was 42.8%. Single-factor analysis revealed that factors associated with prognosis were pancreatic involvement (*P* = 0.027), tumor diameter (*P* = 0.008), T stage (*P* = 0.003), TNM stage (*P* < 0.001), and number of metastatic lymph nodes (*P* < 0.001) as shown in Fig. 1-5. Multivariate analysis showed that factors associated with prognosis were the number of lymph node metastases (*P* < 0.001, OR: 1.923, CI: 1.367-2.705) and tumor diameter (*P* = 0.03, OR: 1.432, CI: 1.035-1.981). ![Fig. 1.](http://www.cancerbiomed.org/https://www.cancerbiomed.org/content/cbm/6/2/85/F1.medium.gif) [Fig. 1.](http://www.cancerbiomed.org/content/6/2/85/F1) Fig. 1. **Survival curves of patients with different pancreatic involvement**. Patients without pancreatic involvement (*n* = 42) had a higher survival rate than those with pancreatic involvement (*n* = 63, *P* = 0.027). ![Fig. 2.](http://www.cancerbiomed.org/https://www.cancerbiomed.org/content/cbm/6/2/85/F2.medium.gif) [Fig. 2.](http://www.cancerbiomed.org/content/6/2/85/F2) Fig. 2. **Survival curves of patients with different tumor sizes**. The Tumor > 2 cm in diameter (*n* = 61) was associated with a lower survival rate as compared with those ≤ 2 cm in diameter (*n* = 38, *P* = 0.008). ![Fig. 3.](http://www.cancerbiomed.org/https://www.cancerbiomed.org/content/cbm/6/2/85/F3.medium.gif) [Fig. 3.](http://www.cancerbiomed.org/content/6/2/85/F3) Fig. 3. **Survival curves of patients with different tumor T Stage**. The survival rates of the patients with T1 (*n* = 5), T2 (*n* = 38), T3 (*n* = 48) and T4 (*n* = 14) vary significantly (*P* = 0.003). ![Fig. 4.](http://www.cancerbiomed.org/https://www.cancerbiomed.org/content/cbm/6/2/85/F4.medium.gif) [Fig. 4.](http://www.cancerbiomed.org/content/6/2/85/F4) Fig. 4. **Survival curves of patients with different tumor TNM Stage**. The survival rates of the patients with IA (n = 4), IB (*n* = 29), IIA (*n* = 32), IIB (*n* = 26) and III (*n* = 14) vary significantly (*P* < 0.001). ![Fig. 5.](http://www.cancerbiomed.org/https://www.cancerbiomed.org/content/cbm/6/2/85/F5.medium.gif) [Fig. 5.](http://www.cancerbiomed.org/content/6/2/85/F5) Fig. 5. **Survival curves of patients with number of metastatic lymph nodes**. The survival rates of the patients with number of metastatic lymph nodes 0 (*n* = 66), 1 (*n* = 20), ≥ 2 (*n* = 19), vary significantly (*P* < 0.001). ## Discussion Ampulla of Vater carcinoma has a relatively low incidence rate accounting for approximately 0.2% of the tumors of the gastrointestinal tract[2]. Although the early appearance of jaundice results in a better outcome from surgical treatment than that seen for carcinoma of the pancreatic head, the reported five-year survival rate ranged from 38% to 52% in the literature because of the limited number of patients studied[3-6]. The 42.8% of 5-year survival rate in this study is within this range. The 8.6% of mortality rate in this study is higher than that reported in several large medical centers[3-6], probably because the operations in this cohort were performed by doctors in multiple therapeutic groups. Richter et al.[7] reported that surgical complications were associated with the experience of the surgeon and the establishment of a professional therapeutic group. Nevertheless, it is clear that the treatment of this disease is complex and difficult, which illustrates the need for better information when selecting treatment options. In our patients, the most important factor associated with prognosis was the number of lymph node metastases. Our results are consistent with those of other investigators who have reported many factors to be related to the 5-year survival rate of patients with ampulla of Vater carcinoma, including jaundice, intraoperative blood transfusion, lymph node metastasis, tumor differentiation, the diameter of the tumor, clinical stage, and positive margins; but who have identified lymph node metastasis as an important prognostic factor in all studies[3-5,8-10]. The reported incidence of lymph node metastasis was relatively high (29%-52%) in ampulla of Vater carcinoma[11,12], which was consistent with the incidence of 37.1% in the 105 patients in this study. Therefore, evaluating the location of lymph node metastasis will have a significant impact on the selection of the surgical methods. The lymphatic drainage is complex in the region of the head of the pancreas. The lymph nodes in front of the head of the pancreas, called anterior pancreaticoduodenal lymph nodes, distribute along the anterior pancreaticoduodenal blood vessels. The lymph either moves upwards along the gastroduodenal artery and returns to the adjacent area of the arteria hepatica propria or it enters the lymph nodes around the celiac artery via the pyloric lymph nodes. It may also interchange with the lymphatic return of the hepatic hilar region along the hepatoduodenal ligament. In addition, the lymph can move downward to enter the lymph nodes at the root of the mesentery and around the abdominal aorta via the inferior pancreatic lymph nodes. The large number of lymph nodes behind the head of the pancreas, called the posterior pancreaticoduodenal lymph nodes, can converge to the left leading into the superior mesenteric artery lymph nodes and enter the surrounding area of the celiac artery. They can also directly converge into the lymph nodes around the aorta, the upper back of which is also connected with lymphatic vessels at the hepatic hilar region. In clinical practice, the head of the pancreas is divided into the superior and the inferior parts by the ampulla. Lymph node metastasis of ampullary carcinoma can occur at any site within the above drainage pathways, as was demonstrated by the distribution of metastatic lymph nodes in this study. Because metastasis to posterior pancreaticoduodenal lymph nodes is most common and accounts for more than 50% of the metastasis in ampulla of Vater carcinoma[13], the surgical dissection of the lymph nodes at this drainage area is an important step to ensure thorough resection. In addition to the presence of metastasis, the number of lymph node metastases is also an important prognostic factor for the ampulla of Vater carcinoma[14]. This was confirmed in our study that the five-year survival rates were 46.6%, 28.4%, and 0% when metastasis occurred in 0, 1, and 2 lymph nodes, respectively. Sentinel lymph nodes (SLN) are the first stop in the lymphatic drainage of tumors[15]. Thus, a single metastatic lymph node can be considered to have the characteristics of the sentinel lymph node[16]. In this study, 84.6% (22/26) of the sites of single lymph node metastasis were in the region of the head of pancreas and duodenum, which should be counted as the first-stop lymph nodes according to the standard established by the Japan Pancreas Society[17]. Jumping transfer only occurred in 4 patients, suggesting that lymph node metastasis in ampullary carcinoma generally travels from the nearer to the farther nodes. These results provided a useful basis for the dissection of metastatic lymph nodes and examination of the SLN no matter whether partial tumor excision or PD was performed. The benefits of surgery with the expanded lymph node dissection and local tumor resection are still a matter of controversy. Beger et al.[5] found that expanded lymph node dissection increased the survival rate, however, Yeo et al.[18] reported that a randomized study could not confirm the effects of expanded lymph node dissection on the prognosis (standard lymphadenectomy and extended groups: 56% *vs*. 60%). Because the rates of diagnosis and preoperative detection of lymph node metastasis are still low, studies of the pathologic factors associated with lymph node metastasis can assist in the selection of surgery. The clinicopathologic factors associated with lymph node metastasis found in this study were diameter of the tumor, the degree of tumor differentiation, T-stage, and pancreatic involvement, all of which should be taken into consideration when evaluating whether or not to perform local excision and expanded lymph node dissection. These results are similar to those of previous researches[19], which suggest that the tumors at Stage Tis and T1 with high and medium levels of differentiation are appropriate for local excision because these types of tumor have a low rate of lymph node metastasis. Despite of the uncertain effect of tumor size on the outcome of local tumor excision, our results indicated that the rate of lymph node metastasis was increased significantly and the 5-year survival rate was decreased significantly when the tumor diameter was greater than 2 cm. Thus, local tumor excision should not be performed for tumors larger than 2 cm in diameter. This study did not investigate the effects of adjuvant chemotherapy and expanded lymph node dissection on the prognosis, which is also a controversial topic[18,20]. Further studies are needed to demonstrate the benefits of adjuvant chemotherapy and expanded lymph node dissection. In conclusion, the number of metastatic lymph nodes and tumor diameter are important pathologic factors predicting prognosis of ampulla of Vater carcinoma after radical resection, and lymph node dissection during the radical surgery effectively improves the survival rate. * Received December 8, 2008. * Accepted February 26, 2009. * Copyright © 2009 by Tianjin Medical University Cancer Institute & Hospital and Springer ## References 1. Asbun HJ, Rossi RL, Munson JL. Local resection for ampullary tumors: is there a place for it? Arch Surg 1993; 128: 515-520. [CrossRef](http://www.cancerbiomed.org/lookup/external-ref?access_num=10.1001/archsurg.1993.01420170045006&link_type=DOI) [PubMed](http://www.cancerbiomed.org/lookup/external-ref?access_num=8098205&link_type=MED&atom=%2Fcbm%2F6%2F2%2F85.atom) 2. Yamaguchi K, Enjoji M. Carcinoma of the ampulla of Vater: a clinicopathologic study and pathologic staging of 109 cases of carcinoma and 5 cases of adenoma. Cancer 1987; 59: 506-515. [CrossRef](http://www.cancerbiomed.org/lookup/external-ref?access_num=10.1002/1097-0142(19870201)59:3<506::AID-CNCR2820590326>3.0.CO;2-%23&link_type=DOI) [PubMed](http://www.cancerbiomed.org/lookup/external-ref?access_num=3791159&link_type=MED&atom=%2Fcbm%2F6%2F2%2F85.atom) [Web of Science](http://www.cancerbiomed.org/lookup/external-ref?access_num=A1987F822700020&link_type=ISI) 3. Talamini MA, Moesinger RC, Pitt HA, et al. Adenocarcinoma of the Ampulla of Vater: A 28-Year Experience. Ann Surg 1997; 225: 590-600. [CrossRef](http://www.cancerbiomed.org/lookup/external-ref?access_num=10.1097/00000658-199705000-00015&link_type=DOI) [PubMed](http://www.cancerbiomed.org/lookup/external-ref?access_num=9193186&link_type=MED&atom=%2Fcbm%2F6%2F2%2F85.atom) [Web of Science](http://www.cancerbiomed.org/lookup/external-ref?access_num=A1997XE06300073&link_type=ISI) 4. Howe JR, Klimstra DS, Moccia RD, et al. Factors predictive of survival in ampullary carcinoma. Ann Surg 1998; 228: 87-94. [CrossRef](http://www.cancerbiomed.org/lookup/external-ref?access_num=10.1097/00000658-199807000-00013&link_type=DOI) [PubMed](http://www.cancerbiomed.org/lookup/external-ref?access_num=9671071&link_type=MED&atom=%2Fcbm%2F6%2F2%2F85.atom) [Web of Science](http://www.cancerbiomed.org/lookup/external-ref?access_num=000074711200014&link_type=ISI) 5. Beger HG, Treitschke F, Gansauge F, et al. Tumors of the ampulla of Vater: experience with local or radical resection in 171 consecutively treated patients. Arch Surg 1999; 134: 526-532. [CrossRef](http://www.cancerbiomed.org/lookup/external-ref?access_num=10.1001/archsurg.134.5.526&link_type=DOI) [PubMed](http://www.cancerbiomed.org/lookup/external-ref?access_num=10323425&link_type=MED&atom=%2Fcbm%2F6%2F2%2F85.atom) [Web of Science](http://www.cancerbiomed.org/lookup/external-ref?access_num=000080173400012&link_type=ISI) 6. Qiao QL, Zhao YG, Ye ML, et al. Carcinoma of the Ampulla of Vater: factors influencing long-term survival of 127 patients with resection. World J Surg 2007; 31: 137-143. [CrossRef](http://www.cancerbiomed.org/lookup/external-ref?access_num=10.1007/s00268-006-0213-3&link_type=DOI) [PubMed](http://www.cancerbiomed.org/lookup/external-ref?access_num=17171495&link_type=MED&atom=%2Fcbm%2F6%2F2%2F85.atom) [Web of Science](http://www.cancerbiomed.org/lookup/external-ref?access_num=000243365100023&link_type=ISI) 7. Richter A, Niedergethmann M, Sturm JW, et al. Long-term results of partial pancreaticoduodenectomy for ductal adenocarcinoma of the pancreatic head: 25-year experience. World J Surg 2003; 27: 324-329. [CrossRef](http://www.cancerbiomed.org/lookup/external-ref?access_num=10.1007/s00268-002-6659-z&link_type=DOI) [PubMed](http://www.cancerbiomed.org/lookup/external-ref?access_num=12607060&link_type=MED&atom=%2Fcbm%2F6%2F2%2F85.atom) [Web of Science](http://www.cancerbiomed.org/lookup/external-ref?access_num=000181517800016&link_type=ISI) 8. Yokoyama N, Shirai Y, Wakai T, et al. Jaundice at presentation heralds advanced disease and poor prognosis in patients with ampullary carcinoma. World J Surg 2005; 29: 519-523. [CrossRef](http://www.cancerbiomed.org/lookup/external-ref?access_num=10.1007/s00268-004-7709-5&link_type=DOI) [PubMed](http://www.cancerbiomed.org/lookup/external-ref?access_num=15770375&link_type=MED&atom=%2Fcbm%2F6%2F2%2F85.atom) 9. Roder JD, Schneider PM, Stein HJ, et al. Number of lymph node metastases is significantly associated with survival in patients with radically resected carcinoma of the ampulla of Vater. Br J Surg 1995; 82: 1693-1696. [CrossRef](http://www.cancerbiomed.org/lookup/external-ref?access_num=10.1002/bjs.1800821235&link_type=DOI) [PubMed](http://www.cancerbiomed.org/lookup/external-ref?access_num=8548244&link_type=MED&atom=%2Fcbm%2F6%2F2%2F85.atom) [Web of Science](http://www.cancerbiomed.org/lookup/external-ref?access_num=A1995TG97000029&link_type=ISI) 10. Giorgio AD, Alfieri S, Rotondi F, et al. Pancrratoduodenectomy for Tumors of Vater’s Ampulla: Report on 94 consecutive Patients. World J Surg 2005; 29: 513-518. [CrossRef](http://www.cancerbiomed.org/lookup/external-ref?access_num=10.1007/s00268-004-7498-x&link_type=DOI) [PubMed](http://www.cancerbiomed.org/lookup/external-ref?access_num=15776300&link_type=MED&atom=%2Fcbm%2F6%2F2%2F85.atom) [Web of Science](http://www.cancerbiomed.org/lookup/external-ref?access_num=000228835800023&link_type=ISI) 11. Allema JH, Reinders ME, van Gulik TM, et al. Results of pancreaticoduodenectomy for ampullary carcinoma and analysis of prognostic factors for survival. Surgery 1995; 117: 247-253. [CrossRef](http://www.cancerbiomed.org/lookup/external-ref?access_num=10.1016/S0039-6060(05)80197-7&link_type=DOI) [PubMed](http://www.cancerbiomed.org/lookup/external-ref?access_num=7878528&link_type=MED&atom=%2Fcbm%2F6%2F2%2F85.atom) [Web of Science](http://www.cancerbiomed.org/lookup/external-ref?access_num=A1995QK68900002&link_type=ISI) 12. Warren KW, Choe DS, Plaza J, et al. Results of radical resection for periampullary cancer. Ann Surg 1975; 181: 534-540. [PubMed](http://www.cancerbiomed.org/lookup/external-ref?access_num=1130871&link_type=MED&atom=%2Fcbm%2F6%2F2%2F85.atom) 13. Shirai Y, Tsukada K, Ohtani T, et al. Carcinoma of the Ampulla of Vater: is radical lympadenecyomy beneficial to patients with nodal disease? J Surg Oncol 1996; 61: 190-194. [CrossRef](http://www.cancerbiomed.org/lookup/external-ref?access_num=10.1002/(SICI)1096-9098(199603)61:3<190::AID-JSO5>3.0.CO;2-6&link_type=DOI) [PubMed](http://www.cancerbiomed.org/lookup/external-ref?access_num=8637205&link_type=MED&atom=%2Fcbm%2F6%2F2%2F85.atom) 14. de Castro SMM, van Heek NT, Kuhlmann KFD, et al. Surgical management of neoplasms of the ampulla of Vater: local resection or pancreatodenectomy and prognostic factors for survival. Surgery 2004; 136: 994-1002. [CrossRef](http://www.cancerbiomed.org/lookup/external-ref?access_num=10.1016/j.surg.2004.03.010&link_type=DOI) [PubMed](http://www.cancerbiomed.org/lookup/external-ref?access_num=15523392&link_type=MED&atom=%2Fcbm%2F6%2F2%2F85.atom) 15. Cabanas RM. An approach for the treatment of penile carcinoma. Cancer 1977; 39: 456-466. [CrossRef](http://www.cancerbiomed.org/lookup/external-ref?access_num=10.1002/1097-0142(197702)39:2<456::AID-CNCR2820390214>3.0.CO;2-I&link_type=DOI) [PubMed](http://www.cancerbiomed.org/lookup/external-ref?access_num=837331&link_type=MED&atom=%2Fcbm%2F6%2F2%2F85.atom) [Web of Science](http://www.cancerbiomed.org/lookup/external-ref?access_num=A1977CX83800013&link_type=ISI) 16. Kosaka T, Ueshige N, Sugaya J, et al. Lymphatic routes of the stomach demonstrated by gastric carcinomas with solitary lymph node metastasis. Surg Today 1999; 29: 695-700. [CrossRef](http://www.cancerbiomed.org/lookup/external-ref?access_num=10.1007/BF02482311&link_type=DOI) [PubMed](http://www.cancerbiomed.org/lookup/external-ref?access_num=10483741&link_type=MED&atom=%2Fcbm%2F6%2F2%2F85.atom) 17. Japan Pancreas Society. Classification of pancreatic carcinoma. 2nd ed. Tokyo: Kanehara. 2003. 18. Yeo CJ, Cameron JL, Lillemoe KD, et al. Pancreaticoduodenectomy with or without distal gastrectomy and extended retroperitoneal lymphadenectomy for periampullary adenocarcinoma, part 2: randomized controlled trial evaluating survival, morbidity, and mortality. Ann Surg 2002; 236: 355-368. [CrossRef](http://www.cancerbiomed.org/lookup/external-ref?access_num=10.1097/00000658-200209000-00012&link_type=DOI) [PubMed](http://www.cancerbiomed.org/lookup/external-ref?access_num=12192322&link_type=MED&atom=%2Fcbm%2F6%2F2%2F85.atom) [Web of Science](http://www.cancerbiomed.org/lookup/external-ref?access_num=000177741500012&link_type=ISI) 19. Rattner DW, Fernandez-del Castillo C, Brugge WR, et al. Defining the criteria for local resection for ampullar neoplasm. Arch Surg 1996; 131: 366-371. [CrossRef](http://www.cancerbiomed.org/lookup/external-ref?access_num=10.1001/archsurg.1996.01430160024003&link_type=DOI) [PubMed](http://www.cancerbiomed.org/lookup/external-ref?access_num=8615720&link_type=MED&atom=%2Fcbm%2F6%2F2%2F85.atom) 20. Todoroki T, Koike N, Morishita Y, et al. Patterns and predictors of failure after curative resections of carcinoma of the Ampulla of Vater. Ann Surg Oncol 2003; 10: 1176-1183. [CrossRef](http://www.cancerbiomed.org/lookup/external-ref?access_num=10.1245/ASO.2003.07.512&link_type=DOI) [PubMed](http://www.cancerbiomed.org/lookup/external-ref?access_num=14654474&link_type=MED&atom=%2Fcbm%2F6%2F2%2F85.atom) [Web of Science](http://www.cancerbiomed.org/lookup/external-ref?access_num=000187066100012&link_type=ISI)