Most surgeons performing mastectomy (MRM) use either the scalpel or electrocautery for developing flaps and performing node dissection. We report a retrospective analysis of all MRM performed over a five-year period by two surgical oncologists. One-hundred-and-ninety-six cases met criteria for inclusion into the study. There are 110 patients in the electrocautery (EC) group and 86 patients in the scalpel (Scpl) group. Mean patient age, mean specimen weight and mean number of lymph nodes removed were not significantly different for the two groups. The incidence of complications was similar for the two groups. Mean estimated blood loss was 134.1 cc for the EC group and 331.6 cc for the Scpl group (P < 0.001). Estimated blood loss ranged from 25 to 475 cc for EC group and from 88 to 1350 cc for the Scpl group. One patient in the Scpl group required transfusion. Mean operating time was less for the Scpl group, 111.0 vs 93.5 minutes (P < 0.009). Multiple regression analysis revealed that technique was the strongest predictor of estimated blood loss (standardized coefficient 0.61). Operating time, age, number of nodes removed and specimen weight were less predictive (standardized coefficient 0.21, 0.17, 0.11 and 0.09, respectively).