Breast-preserving surgery for tumors of limited size or reduced by neoadjuvant chemotherapy has definitely entered into the practice. Distant results of controlled studies demonstrated that conservative methods, when correctly indicated and performed, can provide the same results as mutilating procedures, in terms of overall survival. There is general agreement on the fact that conservation bears a major risk of intrabreast recurrences, whose meaning and impact on prognosis are still open to debate. Inadequate surgery, i.e., too-limited excision, or the lack of radiotherapy, certainly causes a higher rate of local failures. However, analysis of the patient series reported in the literature permits the conclusion that local failures and distant metastasis are events partially independent of each other. In other words, there are factors that are predictive of local recurrence, and not of distant spread, and vice versa, and factors that affect both the risks. Uncertainty about the meaning of local recurrences influences therapeutic attitudes, not only with regard to the choice between total mastectomy and re-resection, when possible, but also with reference to the identification of those local recurrences that merit systemic treatment. As far as the treatment of local failures is concerned, it is too soon to indicate undisputable guidelines. It is necessary to wait for distant results of the many experiences in progress on this issue. On the other hand, since local intrabreast recurrences fortunately are not very frequent (about 10% at 10 years from first treatment), the accrual of patients eligible for clinical trials would take a long time, even for cooperative groups. This is one of the reasons why local failure actually is an open problem.