Abstract
rile worst prognosis of the periampullary carcinomas. This retrospective study was to determine prognostic actors for survival after pancreaticoduodenectomy in patients had pancreatic carcinoma. Methods: We retospectively studied all patients who underwent PD for pancreatic adenocarcinoma originating from the head, neck or uncinate process from January 1996 to January 2011 in our center. Preoperative variables, intraoperative variables and postoperative variables were collected. Results: The study included 480 patients (282 males and 198 females with a median age of 53 years. At the time of analysis, 180 (37.5%) patients were still alive. The median survival was 19 months. This corresponded to a -, 3-, an -year actuaria surviva of 44, %, 20%, and 15% respectively. Mass size less than 2 c (P=0.0001), lymph node ratio (P=0.0001), safety mari gin (P=0.0001), perineural, perivascular infiltration, age; above 60 years (P=0.03), gender, preoperative bilirubinl f SGPT, liver status, pre and postoperative CEA, CA199 (P=0.0001) were significant predictors of survival Conclusion: Mass size less than 2 cm, lymph node ratio, safety margin, perineural, perivascular infiltration, age" above 60 years, gender, liver status, pre and postoper. tive CEA, CA19-9 are important predictors of survival i patients undergoing PD for pancreatic cancer.