Riscos e benefícios do rastreamento mamográfico no Brasil.

Abstract

Evidence on the reduction of overall mortality with mammographic screening is inconclusive and to date no published study has covered all known harms associated with screening to assess the balance between risks and benefits of such intervention. In addition, the benefits and risks of screening vary across Brazilian states, but these differences have never been studied. The aim of the study was to estimate the risks and benefits of mammography screening in Brazil, in terms of deaths avoided and caused by screening. 196 life tables were constructed for benefit estimates and 13,440 for estimates of screening harms in Brazil and for the 27 Federative Units, for 5 different mammographic screening protocols, based on nationwide data from 2012 to 2015. The cumulative risk of cancer death was calculated in each cohort as well the absolute reduction in mortality with screening. Harms associated with screening were also estimated, including number of false positives, excess biopsies, overdiagnosis, overtreatment, and deaths due to: radio-induced breast cancer; suicides and cardiovascular diseases associated with overdiagnosis; surgical mortality; Stewart-Treves syndrome and cardiovascular disease, lung cancer, esophageal cancer, pulmonary thromboembolism and radio-induced breast sarcoma by overtreatment with adjuvant radiotherapy. The sensitivity analysis included three different scenarios for each screening protocol, in addition to additional simulations. In the most probable scenario, the reduction of the absolute risk of death from breast cancer with the biennial screening of 50 to 69 years in Brazil was 0.23% and, considering the screening harms, the net benefit was 21.2 deaths prevented by 10,000 invited to screening. The absolute benefit of the screening and the net benefit of the screening in the Federation Unit with the worst result, are three and a half times smaller than the one with the best result. The annual screening from 40 to 49 years resulted in 1.2 deaths per 10,000 invited to the screening, and even in the most favorable scenario, the difference between potentially avoided and caused deaths was 1.48 per 10,000 invited considering the most probable harms. In this age group, the greatest cause of death was breast cancers induced by mammograms (70% of the total), followed by cardiovascular diseases associated with adjuvant radiotherapy (16%), which were the main cause in the 50-69 years group (70% of the total), followed by radio-induced lung cancer (13%). The benefits of mammographic screening are lower in Brazil - and especially in the North and Northeast regions of the country - than in European countries. To reduce the harms of mammographic screening in Brazil, and to ensure that the balance between risks and benefits is likely to be favorable, the main measures that should be adopted are incentives for adherence to the official guidelines of the age group and target population; the institution of informed and shared decision prior to the request for screening; implementation of national quality programs for mammography and radiotherapy, and cessation of smoking in patients undergoing adjuvant radiotherapy.