BACKGROUND: The effectiveness and extent of regional lymph node dissection in primary duodenal cancer (DC) remains unclear. This study aimed to analyze the prognostic factors and lymph node metastasis (LNM) patterns in DC. METHODS: Fifty-three patients who underwent surgical resection for DC between January 1998 and December 2018 at two institutions were retrospectively analyzed. Univariate and multivariate analyses were performed on the prognostic factors of resected DC. Moreover, the relationships between depth of tumor invasion and incidence of LNM and between tumor location and LNM stations were analyzed. RESULTS: The five-year survival rate of the study population was 68.9%. Multivariate survival analysis demonstrated that histologic grade G2-G4, presence of LNM, pT3-4, and elevated preoperative CA19-9 were the independent poor prognostic factors. No patient with pTis-T2 had LNM. On the other hand, LNM was found in 70% of patients with pT3-4. Among 36 patients who underwent pancreaticoduodenectomy (PD), LNM around the pancreatic head was observed, regardless of the duodenal cancer site, including the duodenal bulb and the third to the fourth portion. CONCLUSIONS: Histologic grade G2-G4, presence of LNM, pT3-T4, and elevated preoperative CA19-9 were the independent poor prognostic factors in patients with resected DC. Our results suggested that lymph node dissection could be omitted for DC Tis-T1a. Moreover, based on the high frequency of LNM in T3-4 cases, PD with lymph node dissection in the pancreatic head region was considered necessary for T3-4 DC at any site.
BACKGROUND: The effectiveness and extent of regional lymph node dissection in primary duodenal cancer (DC) remains unclear. This study aimed to analyze the prognostic factors and lymph node metastasis (LNM) patterns in DC. METHODS: Fifty-three patients who underwent surgical resection for DC between January 1998 and December 2018 at two institutions were retrospectively analyzed. Univariate and multivariate analyses were performed on the prognostic factors of resected DC. Moreover, the relationships between depth of tumor invasion and incidence of LNM and between tumor location and LNM stations were analyzed. RESULTS: The five-year survival rate of the study population was 68.9%. Multivariate survival analysis demonstrated that histologic grade G2-G4, presence of LNM, pT3-4, and elevated preoperative CA19-9 were the independent poor prognostic factors. No patient with pTis-T2 had LNM. On the other hand, LNM was found in 70% of patients with pT3-4. Among 36 patients who underwent pancreaticoduodenectomy (PD), LNM around the pancreatic head was observed, regardless of the duodenal cancer site, including the duodenal bulb and the third to the fourth portion. CONCLUSIONS: Histologic grade G2-G4, presence of LNM, pT3-T4, and elevated preoperative CA19-9 were the independent poor prognostic factors in patients with resected DC. Our results suggested that lymph node dissection could be omitted for DC Tis-T1a. Moreover, based on the high frequency of LNM in T3-4 cases, PD with lymph node dissection in the pancreatic head region was considered necessary for T3-4 DC at any site.