The primary objective of this study is to compare the incidence of cardiac events \[(defined as asymptomatic systolic dysfunction (LVEF \< 50%, asymptomatic NYHA I) or symptomatic heart failure NYHA class II-IV either by Echo and/or by clinical exam (LVEF \< 50% and heart failure symptoms)\] between anthracycline and non-anthracycline chemotherapy treated node-positive breast cancer patients in a long-term follow-up. We will define a binary status for each patient: no cardiac event / one or more cardiac events. Analysis of the primary endpoint will involve a comparison of the CMF-treated patients versus the pooled anthracycline treated patients (EC and HEC). If a significant difference is detected, all paired comparisons will also be performed for the primary endpoint (CMF versus EC, CMF versus HEC, EC versus HEC).
