A woman's risk of undergoing a mastectomy for breast cancer ranges from zero to 80 per cent depending on the surgeon to whom she is referred. Jeremy Laurance, Health Editor, says the finding adds to fears about low standards of treatment in the UK.

A world-wide review of breast cancer treatment has established that for women with early-stage tumours removing the lump is as safe as removing the whole breast, provided it is accompanied by radiotherapy and drug treatment.

The NHS breast screening programme was set up in 1988 to detect cancers at a stage when the small size of the tumours allowed for less radical surgery.

Yet the treatment women actually receive varies widely, depending on the individual preference of the surgeon responsible for them, according to a study published tomorrow.

The review of 600 post-menopausal women with breast cancer who were treated by 35 surgeons has shown that the mastectomy rate varied from nil to 80 per cent. The variation was not linked to size of tumour or severity of disease.

The unnamed surgeon with the 80 per cent mastectomy rate had performed the operation on eight out of the ten breast cancer patients he had treated during the period of the study in 1991-92.

Five surgeons had treated only one patient and another five surgeons had managed five or fewer patients.

Those surgeons with higher caseloads of more than 20 patients had a lower mastectomy rate - at 15 per cent, suggesting a more modern approach to treatment - than those with smaller caseloads.

International comparisons show that 45 per cent of British women diagnosed with breast cancer die within five years compared with 21 per cent in the United States, 35 per cent in France and 40 per cent in Germany. The Cancer Research Campaign has blamed the discrepancy on poor standards of treatment in the UK.

The current study, published in the British Medical Journal, was conducted in the former south-east Thames region of the NHS, covering Kent, east Sussex and south-east London. Figures for 1995-96 show that that region had the highest number of screen detected cancers operated on by surgeons with a caseload of less than 10 breast cancer patients a year.

One in five patients in the study did not receive radiotherapy, including 51 of 317 patients with unfavourable tumours. Radiotherapy has been shown to reduce local recurrence of the cancer in the breast, regardless of whether the tumour is classified as favourable or unfavourable, although it does not affect overall survival.

The authors of the study, from the Breast Screening Quality Assurance Centre at King's College Hospital, London, say: "Women should not be put at increased risk of local recurrence by not having radiotherapy because the quality ... of life must suffer."

In addition, although most surgeons gave the hormonal drug tamoxifen, 26 of the 600 patients did not receive it, despite the fact that is has been shown to have a "clear survival benefit" for post-menopausal women, and the benefits outweigh the risk of side effects.

Chemotherapy, which may delay local recurrence of the disease in some patients, was given to less than 3 per cent of the women, but the authors say the evidence may have been too new to influence treatment decisions at the time of the study in 1991-92.

They warn that more consistency in treatment is required that does not depend on individual "surgical preference" and that the current guidelines issued to surgeons are deficient.

The NHS Breast Screening programme has a target of cutting deaths by a quarter by the year 2000. The authors say: "Sub-optimal treatment ... could compromise [the target's] success."