| Abstract: | Macroscopic diagnosis for lymph node metastases was compared with histopathological diagnosis in 444 patients with carcinoma of the esophagus, stomach, colon, thyroid and breast. The former indicated lymph node metastases in 181 patients. In all of them, none or less than five node metastases were proven by routine histopathological diagnosis. Detailed histological study revealed lymph node metastases in 25 out of 263 patients with macroscopically negative nodes, the rate of false negative being 9.5 per cent. The study also demonstrated no lymph node metastases in 51 of 181 patients with macroscopically positive nodes. Three additional specimens were obtained from originally examined 693 lymph nodes and reexamined microscopically in these 51 patients. Involvement by cancer cells was detected in 9 nodes (1.3 per cent) in 8 patients. Metastases were found from additional specimens in 7 of 9 nodes, indicating that metastatic carcinoma had been overlooked in the remaining two nodes. Additional specimens or embedding-techniques were recommended in such cases as macroscopic metastases were strongly suspected or lymph vessel invasions were remarkable. In 24 patients with esophageal cancer, one to one correspondence was available in the analysis of macroscopic diagnosis. Seventy-eight out of 108 involved nodes were macroscopically judged as involved (sensitivity; 72.2 per cent), and 1166 out of 1260 nodes without macroscopical metastases were judged as cancer-free (specificity; 92.5 per cent). Overestimation of macroscopic diagnosis was due to thickened capsule, fibrosis, inflammation and enlargement in size more than 10 mm in diameter of the nodes. Underestimation was observed in case of nodes with metastatic area less than one-third and with smaller size less than 5mm in diameter. |
