To determine the clinical value of the Siewert classification for gastic-cancer patients in Korea, we evaluated and compared the clinicopathologic factors of type II and type III cancer.

The medical records of 89 consecutive patients who had undergone surgery for an adenocarcinoma of the gastroesophageal junction (GEJ) at the Department of Surgery, Hanyang University Hospital, between Jun. 1992 and Dec. 2003 were reviewed retrospectively.

There were one patient with type I, 12 pateints with type II and 77 patients with type III. During the same period, 1,341 patients underwent surgery for a gastric carcinoma, so proportion of GEJ cancer being 6.6%. The median follow- up duration was 31 months (range: 2~135 months), and the follow-up rate was 100%. Between type II and type III cancers, there were no significant differences in the clinicopathologic variables including age, sex, gross appearance, histologic type, depth of invasion, and pathologic stage. The longest diameter of the tumor was larger in type III (6.1±2.1 cm) than in type II (3.9±1.1 cm)(P=0.001). A total gastrectomy with Roux-en-Y esophagojejunostomy was done most frequently, while jejunal interposition was done in 3 cases of type II and 2 cases of type III. More than a D2 lymphadenectomy was done all cases. The numbers of dissected lymph nodes and metastatic lymph nodes in type II were 43.8 and 5.8 respectively, while they were 49.8 and 8.1 in type III, but the difference between the two groups were not statistically significant. The mean length of the proximal resection margin was 15±5 mm in type II and 21±13 mm in type III, but this difference was not statistically significanct. The time to recurrence after operation was 19.3 months in type II and 16.9 months in type III. The five-year survival rates of type II and III were 68.8% and 52.7% respectively, but difference was not significant.

There were no significant differences in the clinicopathologic variables, including survival rate, between type II and type III cancers in Korean patients. According to these findings, it appears to be reasonable to classify type III cancer as a cardia cancer in a broad sense.