Early gastric cancer: ten years of experience

Gastric cancer is a disease in which the main treatment is surgical extirpation. The modifications introduced in the surgical treatment over the last decades were accompanied by a clear increase of survival, which reaches global values of 61% at 5 years in Japan. One of the reasons that contribute to this improvement is early diagnosis of the lesions. In the period between January 1, 1990 and December 31, 1999 662 patients with gastric adenocarcinoma were treated in the Service of Surgery 1 of our hospital; 110 were refused surgical treatment. Of the resected patients, 91 (21.4%) were classified as early gastric cancer according to the definition of the Japanese Society of Digestive Endoscopy. There were 30 women and 61 men, with a median age of 60.2 +/- 15 years; 3 patients had a preoperative diagnosis of gastric ulcer; 2 others were operated without recent histology; and 1 patient was urgently resected for a bleeding ulcer. In all the remaining patients biopsy confirmed the presence of cancer (89%) or serious dysplasia (4.6%). The lesions had been distributed essentially in the medium 1/3 (48.3%) and distal 1/3 of the stomach. Subtotal gastrectomy was accomplished in 48 patients, total gastrectomy in 40, total desgastrogastrectomy in 3, and in 9 patients the surgery involved the spleen (8 patients) and the spleen and tail of the pancreas in 1 patient. Lymphadenectomy was not performed in 5 patients, lymph nodes by the first lymph node barrier were removed in 25 patients and by the second barrier in 61 patients (67%). Median tumor size was 26 +/- 1.8 mm. The lesion reached the mucosa in 46 patients and the mucosa and submucosa in 45. In 6 patients the removed lymph nodes were microscopically invaded (6.7%). Five patients died (5.7%). The median follow-up of the patients is 41 +/- 26 months; 7 patients died (8.1%) during this period; 4 died unequivocally of disease progression. The median survival of patients was 85% at 5 years and 80% at 10 years. In our series, survival was affected by the presence of invaded lymph nodes, not by the penetration in depth of the lesion or the size of the tumor.