1. ¹University Department of Surgery, Western Infirmary, Dumbarton Road, Glasgow G11 6NT, UK
2. ²West of Scotland Cancer Surveillance Unit, Department of Public Health, University of Glasgow, 1 Lilybank Gardens, Glasgow G12 8RZ, UK

Correspondence: Dr D Hole, E-mail: d.j.hole@clinmed.gla.ac.uk

Received 9 February 2004; Accepted 3 March 2004; Published online 27 April 2004.

Abstract

Evidence that the survival of women with breast cancer treated by specialist surgeons is better than that by nonspecialists is limited. Previous reports have not identified the cause of this survival advantage. Our aim was to determine if the survival difference was due to case-mix, adjuvant treatment or the treatment provided by specialist surgeons. The case-records and pathology reports of 2776 women were reviewed. This represented 95% of all those diagnosed with breast cancer between 1/1/1986 and 31/12/1991 in a defined geographical area. Case-mix, surgery, pathology and adjuvant therapies of the 2148 women treated with curative intent were analysed. A standard of adequate surgical management was defined and confirmed as a valid predictor by examining rates of local recurrence, independent of all other prognostic factors. Against this standard, we compared the adequacy of surgical management, local recurrence rates and the survival outcomes of specialists and nonspecialists over an 8-year follow-up period. The inter-relationship between adequacy of surgical management, locoregional recurrence and survival was examined. While the case-mix and prescription of adjuvant therapies were comparable between specialist and nonspecialist surgeons, the efficacy and outcome of local treatment differed widely. Breast cancer patients treated in specialist compared to nonspecialist units had half the risk of inadequate treatment of the breast (24 vs 47%, P<0.001), a five-fold lower risk of inadequate axillary staging (8 vs 40%, P<0.001) and nine times lower risk of inadequate definitive axillary treatment (4 vs 38%, P<0.001). Local recurrence rates were 57% lower (13 vs 23% at eight years, P<0.001) and the risk of death from breast cancer was 20% lower for women treated in specialist units, after allowing for case-mix and adjuvant therapies. Adequacy of surgical management correlated with locoregional recurrence, which in turn correlated with the risk of death. The surgical management in specialised breast units is more often adequate, local and regional recurrence rates are lower, and survival is correspondingly better. We conclude that adequate surgical management of breast cancer is fundamental to improving the outcome from breast cancer irrespective of where it is delivered.