Clinical significance of the discrepancy between preoperative and postoperative diagnoses in gastric cancer patients.

BACKGROUND: Discrepancies between pre- and postoperative diagnoses can lead to dilemma for operative management adequacy. PATIENTS AND METHODS: A total of 2,910 patients with gastric adenocarcinoma underwent curative surgery at the Samsung Medical Center between 2001 and 2003. Patients were divided into four groups: early gastric cancer (EGC)-EGC group that consisted of subjects who were diagnosed as having EGC pre- and postoperatively, advanced gastric cancer (AGC)-EGC group, EGC-AGC group, and AGC-AGC group. Clinicopathologic features and survival rates of groups were analyzed retrospectively. RESULTS: Of the 2,910 patients, 1,491 (51.2%) patients were included in the EGC-EGC group, 132 (4.5%) in the AGC-EGC group, 120 (4.1%) in the EGC-AGC group, and 1,167 (40.1%) in the AGC-AGC group. The EGC-AGC group showed higher proportions of the followings than the EGC-EGC group: upper-third and middle-third tumor localizations, a tumor size from 2 to 5 cm, undifferentiated adenocarcinoma, Lauren's diffuse type, endolymphatic invasion, vascular invasion, and perineural invasion. Five-year survival rates were dependent on the final pathologic stages, not on the preoperative stages. Multivariate analysis revealed that age and American Joint Committee of Cancer stage were independent prognostic factors of patient survival. CONCLUSION: A decision regarding minimally invasive treatment for EGC must be made having considered tumor location, size, and cellular differentiation, because of the possibility of an incorrect preoperative diagnosis.

Clinical significance of splenic hilar lymph node metastasis in proximal gastric cancer.

BACKGROUND: Prophylactic splenectomy for splenic hilar node removal is generally not advised because of the high morbidity and mortality rates and the uncertain impact on patient survival. The aim of this study was to compare the clinicopathologic characteristics and effect on survival of the following two groups: the splenic hilar lymph node metastasis group and the non-metastasis group. METHODS: Three hundred and nineteen patients with proximal gastric adenocarcinoma who underwent curative total gastrectomy with simultaneous splenectomy and D2 lymph node dissection at the Samsung Medical Center between 1995 and 2004 were analyzed retrospectively. RESULTS: Forty one patients (12.9%) had splenic hilar node metastasis. The splenic hilar metastasis group was shown to have a higher proportion of females (48.8%), Borrmann type IV (34.1%), tumor size >5 cm (82.9%), poorly differentiated adenocarcinoma (51.2%), signet ring cell carcinoma (31.7%), Lauren diffuse-type (80.5%), endolymphatic invasion (65.5%), and nerve invasion (46.4%; p < 0.05). There was no splenic hilar node metastasis in early gastric cancer. The 5-year survival rate was 11.04% for the hilar node metastasis group (p < 0.001), which was significantly lower than in the non-metastasis group, in which it was 51.57%. Multivariate analysis revealed that hilar node metastasis was an independent prognostic factor [hazard ratio 1.671; 95% confidence interval (CI) 1.075-2.595; p = 0.022]. CONCLUSION: Splenic hilar node metastasis was not apparent in early gastric cancer and had a very poor prognosis, even though curative resection was done, so the effectiveness of prophylactic splenectomy is uncertain.