Article révisé par les pairs
Résumé : The value of systematic mammographic screening has been studied in a total of 11 randomized trials - some of doubtful methodological quality - on various populations of women; the earliest dates from the 1960s. However, older women have been under-represented in these trials, as only six of them registered patients over 60 years of age. The proportion of women aged over 70 years was low and there are no data for patients older than 74. Several meta-analyses have been conducted and a plausible estimate of the impact of screening is a 16% reduction in breast cancer mortality (95% confidence interval 9-23%). Some meta-analyses provide "post hoc subgroup analyses" by age and these have found a greater benefit of screening for women of older age, with estimates of risk reductions of 17%, 27% and 22% for cohorts of women aged 55-64, 60-69 and 65-74; these figures are significantly different from 0. In comparison, for women in the age groups 40-49, 45-54 and 50-59 the estimated risk reductions were 9%, 7% and 12%. Nevertheless, there has been no formal demonstration of an interaction with age. When these estimates are translated into the number of women who need to be screened in order to avoid one death from breast cancer during 15 years of follow-up, the figure lies between 1370 and 4120. The probability of not dying from breast cancer within 15 years of screening for a 65-year-old woman is estimated to be in the range 98.85-99.11%, compared with 98.73% for a woman who does not undergo screening. These benefits should be balanced against the possible disadvantages of screening, including false positive results, overdiagnosis and overtreatment. One of the meta-analyses concluded that the rate of total mastectomies was significantly increased after screening. Although most scientific recommendations agree that screening should be proposed to all women aged 50-69 years, the benefits and disadvantages of mammography should be fully explained, in order to allow them to make a truly informed decision.