Should women who have a familial history or genetic risk of breast cancer have a mastectomy just in case? No, according to a recent review.

According to Ian S Fentiman, professor of surgical oncology at Guy's Hospital in London, a doctor's powers of prediction are still very poor. He notes one retrospective study where, using a standard model of prediction, the authors concluded that 76 cases of cancer would appear in the 17 years of follow up. In the event only seven cases arose.

Even as these operations are being carried out, the relative risk of a woman with a genetic anomaly developing breast cancer is being revised. For instance. almost half of the cases of familial breast cancer are due to BRCA1 mutations located on chromosome 17q21. Yet the lifetime risk of a woman with BRCA1 mutation developing breast cancer has recently been lowered from 90 per cent to 56 per cent.

Neither, says Fentiman does a radical mastectomy guarantee freedom from cancer, since cancer cells can lurk in places on the breast which are not removed by the surgeon. Data on the protective effect of prophylactic mastectomy is thin on the ground. He suggests an urgent need for a register of women who have had this surgery so that its efficacy can be checked against eventual cases of cancer (BMJ, 1998; 317: 1402-3).

Only a third of women under 40 complaining of a breast lump actually had one when examined by an experienced surgeon, in a new study. The authors also found that if the result of the initial examination was normal, further investigations were unlikely to turn up many cases in which a tumour was present. Media scaremongering was highlighted as one reason why so many younger women felt fearful of what are normal or cyclical changes in breast tissue (Surgery, 1998; 124: 634-41).

Another study has found that women with painful breasts rarely have anything to worry about, either. They study looked at the value of x-ray examinations in cases of painful breasts. It found that pain is not generally associated with pathology only 0.8 per cent of cases revealed breast cancer and several of these were found, by chance, in the asymptomatic breast. Many of the mammographies were used simply for reassurance though this might be considered a questionable use of such technology, particularly one that is inherently inaccurate (BMJ, 1998; 317: 1492-5; see also WDDTY vol 3 no 10).