Phantom breast syndrome after mastectomy has already been reported by us and other authors. The temporal course, character, and extent of these phenomena, however, have not yet been elucidated. OBJECTIVE To investigate in a prospective study the incidence, clinical picture, and temporal course of phantom breast syndrome during a 6-year period. METHODS One-hundred twenty women who embarked on consecutive postoperative control or treatment at our department during a 1-year period were interviewed by a standard questionnaire 3 weeks after the operation. Of these, 110 patients were interviewed 1 year later and 69 were interviewed again 6 years later. METHODS The median age at the first interview was 54 years (Quartile (Q)1 = 45 years; Q3 = 62 years) and at the third interview 6 years later, 60 years (Q1 = 51 years; Q3 = 68 years). RESULTS The incidence of phantom pain and nonpainful phantom sensations was 13.3% and 15.0%, respectively, 3 weeks after mastectomy, 12.7% and 11.8%, respectively, after a year, and 17.4% and 11.8%, respectively, after 6 years. We found significant relationships between preoperative pain and phantom breast syndrome, but no significant relationship between age and the occurrence of this syndrome. Neither postoperative sequelae nor cancer treatment including radiotherapy seemed to affect the occurrence of phantom breast syndrome. Pain in the scar, which was clearly distinguishable from phantom pain, was present in 35.0% of the patients 3 weeks postoperatively, in 22.7% after 1 year, and persisted in 30.9% 6 years later. CONCLUSIONS The present incidence of phantom-related phenomena is close to the incidence reported by others. However, persistent phantom pain after mastectomy may be more common than usually expected. Also, the persistence of pain in the scar seems to be more common than generally expected.