RESUMO
AIM: Our aim was to assess the effect on survival of circumferential resection margin (CRM) involvement in patients with resected oesophageal malignancy. METHODS: Patients undergoing potentially curative oesophageal resection between January 1994 and December 2003 were retrospectively analysed. CRM status was defined as either clear or involved (microscopic tumour within 1 mm of the inked resection margin). Univariate and multivariate survival analyses were performed using the Kaplan-Meier method and Cox proportional hazard model. Overall survival was used as the endpoint. RESULTS: The case records of 249 patients were analysed. CRM status was clear in 170 patients (T1-T3 tumours) and involved in 79 patients (all T3 tumours). Median survival in these groups was 37 months (range 28-47) and 18 months (range 13-23), respectively (p = 0.0001). When T3 tumours were analysed separately there was a trend for T3 CRM involved tumours to have a worse prognosis than T3 CRM clear tumours (p = 0.074). Substratification by percentage of lymph nodes involved by metastases (< or = or >25%) revealed that CRM status had a greater prognostic effect in T3 tumours with a low metastatic lymph node burden (p = 0.04). CONCLUSION: CRM involvement predicts poor prognosis in patients with resected oesophageal malignancy and was an independent prognostic factor in our study. There was only a trend for worse prognosis when T3 tumours were analysed separately. However, patients with T3 tumours and a low percentage of lymph node metastases had a better prognosis if the CRM was negative.