BACKGROUND: T1 gastric cancer can be diagnosed only by endoscopy and is almost curable by local treatment. It has been unclear how a multidetector-row computed tomography (CT) evaluation is valuable for clinical T1 patients. METHODS: Patients with clinical T1 disease, as diagnosed by endoscopy and treated with endoscopic submucosal dissection (ESD) or surgery between October 2000 and October 2007, were examined. The efficacy of CT was evaluated by the reversal rate of endoscopic T1 by CT, the incidence of clinical M1 disease, and the accuracy of diagnosing pathological N+ disease in patients who received surgery. To confirm metachronous distant and nodal metastases, the disease-free survival (DFS) also was evaluated. RESULTS: A total of 761 patients, 236 treated by ESD and 525 treated with surgery, were examined. None of the patients had an endoscopic diagnosis of clinical T1 reversed by CT. No clinical M1 disease was found. Among the 525 patients who underwent surgery, 8 showed clinical N+ disease (1.5 %), while 47 demonstrated pathological N+ disease (8.9 %). The accuracy, sensitivity, specificity, positive predictive value, and negative predictive values were 90.3, 4.3, 98.7, 25, and 91.3 %, respectively. The 5-year DFS rate was 93.6 % (95 % confidence interval 91.4-95.8 %). CONCLUSIONS: The present study suggests that diagnostic value of CT is limited for staging of clinical T1 gastric cancer patients, because the reversal rate of endoscopic T1 by CT was very low, clinical M1 disease was rare, the diagnosis of N+ status was unreliable, and metachronous M1 and N+ findings were rare.
BACKGROUND: T1 gastric cancer can be diagnosed only by endoscopy and is almost curable by local treatment. It has been unclear how a multidetector-row computed tomography (CT) evaluation is valuable for clinical T1 patients. METHODS:Patients with clinical T1 disease, as diagnosed by endoscopy and treated with endoscopic submucosal dissection (ESD) or surgery between October 2000 and October 2007, were examined. The efficacy of CT was evaluated by the reversal rate of endoscopic T1 by CT, the incidence of clinical M1 disease, and the accuracy of diagnosing pathological N+ disease in patients who received surgery. To confirm metachronous distant and nodal metastases, the disease-free survival (DFS) also was evaluated. RESULTS: A total of 761 patients, 236 treated by ESD and 525 treated with surgery, were examined. None of the patients had an endoscopic diagnosis of clinical T1 reversed by CT. No clinical M1 disease was found. Among the 525 patients who underwent surgery, 8 showed clinical N+ disease (1.5 %), while 47 demonstrated pathological N+ disease (8.9 %). The accuracy, sensitivity, specificity, positive predictive value, and negative predictive values were 90.3, 4.3, 98.7, 25, and 91.3 %, respectively. The 5-year DFS rate was 93.6 % (95 % confidence interval 91.4-95.8 %). CONCLUSIONS: The present study suggests that diagnostic value of CT is limited for staging of clinical T1 gastric cancerpatients, because the reversal rate of endoscopic T1 by CT was very low, clinical M1 disease was rare, the diagnosis of N+ status was unreliable, and metachronous M1 and N+ findings were rare.