Any discussion of theutility and reliabilityof mammography mustfocus on what is good forindividual patients-not onwhat the results of a Europeanpublic health researchproject have determined isgood for society as a whole.What is good for most patientsis annual screeningmammography beginningat age 40 years (and for patientsat very high risk forbreast cancer, mammogramsbeginning much earlier,and possibly performedmore frequentlythan once a year).1
Any discussion of theutility and reliabilityof mammography mustfocus on what is good forindividual patients--not onwhat the results of a Europeanpublic health researchproject have determined isgood for society as a whole.What is good for most patientsis annual screeningmammography beginningat age 40 years (and for patientsat very high risk forbreast cancer, mammogramsbeginning much earlier,and possibly performedmore frequentlythan once a year).1Bias in Danish criteriafor study elimination.Whether to get a screeningmammogram should notbe the subject of continuedcontroversy. The evidenceis clearly in favor of mammography.The Danishmeta-analysis of mammographyscreening programsfocused on one end point--breast cancer-specific survival.The Danish investigatorseliminated severalstudies from the metaanalysisthrough the use ofcriteria many others believewere too stringent.For example, one study2was eliminated becausethere were about 100 morepatients in the controlgroup than in the studygroup--out of a study totalof tens of thousands of patients.The decision to eliminatethis study seemed tobe based on a bias theDanish investigators hadagainst it rather than onobjective scientific criteria.A reanalysis of the datafrom the entire group ofstudies showed a survivalbenefit for mammographicscreening.3Benefits beyond survivingbreast cancer.Moreover, screening mammographyprovides benefitsbeyond disease-specificsurvival. These include thepossibility of undergoinglumpectomy rather thanmastectomy, and of beingable to obtain treatment forbreast cancer precursors,such as ductal carcinoma insitu, lobular carcinoma insitu, and other atypical proliferativelesions. These lesionsare virtually alwaysidentified because calcificationsare noted on a mammogramrather than becauseof physical symptoms--and mammographyis the only technology capableof diagnosing thesehigh-risk precursor conditionsbefore malignancy develops.Diagnosis of precursorlesions makes it possibleto take actions that candecrease the risk that theywill develop into cancer.Regulation has improvedmammogramquality. Because of the limitationsof all imaging technologies,no screening testis 100% effective at identifyingdisease in asymptomaticpersons. However, establishedmethods and governmentregulations haveimproved the accuracy ofscreening mammography.FDA-accredited mammographycenters that employboard-certified radiologistsare likely to have lowerfalse-negative rates, lowerrecall rates, and optimalfilm processing. In suchfacilities, adjustments aremade daily to eliminatedust, fog, and humidityand to maintain optimaldarkroom conditions sothat proper film density isachieved. Mammograms inthese centers must be readon a viewbox with properillumination for mammography(higher-poweredlighting than that of a standardviewbox) and withoutambient light. Radiologistsare monitored for the numberof mammograms theyread each year and mustnot exceed set maximumnumbers of false-positives,recalls, and false-negatives.While mammographyis not a perfect screeningtool, it is a useful method ofscreening for asymptomaticbreast cancer (Figure 2),and its reliability has beenestablished by large clinicalstudies. Although not allpatients who get screeningmammograms can be saidto have improved diseasespecificsurvival as a result,the same can be said formany therapeutic modalities.Because we do notknow in advance which patientsmay--and whichmay not--benefit from aparticular diagnostic ortherapeutic technique, weremain obligated to offerthat diagnostic method ortherapy to all patients.
Smith RA, Saslow D, Sawyers KA,et al. American Cancer Society guidelinesfor breast cancer screening: update2003. CA
Cancer J Clin
Olsen O, GÃ¸tzsche PC. Cochranereview on screening for breast cancerwith mammography.
Fletcher SW, Elmore JG. Mammographicscreening for breast cancer.
N Engl J Med