PURPOSE: To evaluate whether soft tissue surrounding the celiac axis, as seen on abdominal CT imaging after gastrectomy for gastric carcinoma, should be considered as the recurrence of carcinoma or postoperative change. MATERIALS AND METHODS: One hundred and forty-one abdominal CT examinations of 71 patients who had undergone subtotal or total gastrectomy for gastric carcinoma were included in our study. Conventional CT scans were obtained with 1 cm thickness and interval from the diaphragm to the kidneys after contrast enhancement. It was considered that carcinoma had not recurred if findings were negative on UGI series, endoscopy with biopsy and a normal level of carcinoembryonic antigen except for soft tissue surrounding the celiac axis on abdominal CT. We then divided subjects into a recurrence group (N = 20) and normal group (N = 51) and on initial follow-up CT (FU-CT), analyzed the incidence, margin, shape, extent, degree and pattern of attenuation of the soft tissue surrounding the celiac axis in both groups. Since the second FU-CT examination, we observed changes in the soft tissue surrounding the celiac axis. RESULTS: On initial follow-up CT, at mean 308 days after surgery, fifty-five percent(39/71) of total patients (70% (14/20) of the recurrence group and 49% (25/51) of the normal group) showed soft tissue surrounding the celiac axis. The margin was distinct in 12 (86%) of the recurrence group and indistinct in 21 (84%) of the normal group (P < 0.001). Twelve (86%) of the recurrence group showed a nodular or confluent nodular shape and 21 (84%) of the normal group showed a permeative shape (P < 0.001). Extent was unilateral in eight (57%) of the recurrence group and bilateral in 16 (64%) of the normal group. Attenuation was similar to that of the spleen and muscle in seven (50%) of the recurrence group and was similar to that of muscle in 18 (72%) of the normal group. The pattern of attenuation was homogeneous in 13 (93%) of the recurrence group and 21 (84%) of the normal group. There was no significant difference in extent, degree and pattern of attenuation between the two groups. Since the second FU-CT examination, soft tissue surrounding the celiac axis was seen to have changed. In one patient in the recurrence group it had a distinct margin, was nodular in shape, unilateral in extent and showed attenuation similar to that of the spleen. In one patient in the normal group, it had changed and had an indistinct margin, three patients showed a decrease in the amount of soft tissue and eight showed decreased attenuation. CONCLUSION: Follow-up abdominal CT is useful in the differentiation of cancer recurrence and postoperative change, and for observing changes in soft tissue surrounding the celiac axis.