The factors affecting long-term survival following oesophagectomy for oesophageal cancer are poorly understood. We examined the significance of microscopic tumour involvement at the circumferential resection margin (CRM) on postoperative survival following oesophagectomy. The case notes of 329 patients who underwent a potentially curative oesophagectomy for squamous or adenocarcinoma were reviewed retrospectively. As part of the procedure, all patients underwent an en-bloc resection of their perioesophageal tissue. The presence of tumour either at, or within, 1 mm of the CRM was recorded and correlated with their TNM and survival data. A total of 67 patients (20%) were noted to have a positive CRM, of which 40 cases (12%) had tumour at the resection margin and the remainder had tumour within 1 mm of the margin. Univariate analysis showed no statistically significant association between survival and either category of CRM involvement. Multivariate analysis showed that only T-stage, nodal status and tumour grade were prognostic markers. In conclusion, the presence of microscopic tumour at the CRM following an en-bloc oesophagectomy is not a significant prognostic marker. Despite advances in the management of oesophageal cancer, survival following oesophagectomy remains poor with a 5-year survival of approximately 25% (Alexiou et al, 1998; Ellis, 1999). A number of studies (Edwards et al, 1989;Robey-Cafferty et al, 1991;Patti and Owen, 1997) have attempted to identify histological characteristics that correlate with long-term postoperative survival. The presence of microscopic tumour at the circumferential margin of excision is one such histological factor that has recently been investigated as a possible prognostic marker. A retrospective study carried out in 1991 showed that tumour involvement at the circumferential resection margin (CRM) was associated with a higher incidence of local recurrence (Sagar et al, 1993). Further work by the same group (Dexter et al, 2001) showed that the presence of microscopic tumour at the CRM reduces median postoperative survival; and that the prognostic effect of this factor was most pronounced in those patients with a few metastatic lymph nodes. However, neither of these series examined the longterm prognostic effect of CRM involvement. In addition, both studies were limited by their relatively small sizes. The purpose of our study was to analyse our experience of the significance of microscopic tumour involvement at the CRM on long-term postoperative survival following oesophagectomy in a large cohort of patients.
METHODSThe case notes of 431 patients who underwent an oesophagectomy for squamous or adenocarcinoma between January 1987 and July 1996 at Nottingham City Hospital were reviewed retrospectively. We then excluded all cases of surgical mortality (defined as death occurring within 30 days of operation), incomplete excision (defined as cases with the presence of microscopic tumour within 1 mm of the proximal or distal margins of excision), primary gastric carcinomas, sync...