The scientific literatureon screeningmammography can be confounding.This poses a continuingdilemma for bothpatients and clinicians. Nevertheless,objective analysisof the available data canprovide reasonable guidelinesfor the primary careclinician who must decidewhether screening mammographyhas benefit foran individual patient.
The scientific literatureon screeningmammography can be confounding.This poses a continuingdilemma for bothpatients and clinicians. Nevertheless,objective analysisof the available data canprovide reasonable guidelinesfor the primary careclinician who must decidewhether screening mammographyhas benefit foran individual patient.Evidence of benefit.In their meta-analysis,Gtzsche and Olsen1 criticized6 of the 8 trials theyreviewed for having inadequaciesin their randomizationprocedures. To teaseout their conclusion--thatmammographic screeningfor breast cancer is unjustified--they attempted toidentify and correct for theconfounding factors withinthe 8 trials. Unfortunately,despite Gtzsche andOlsen's protestations, thisprocess is highly susceptibleto investigator bias.A reanalysis by Nystrmand colleagues2 of 4randomized Swedish trialsthat were among those includedin Gtzsche andOlsen's meta-analysis revealeda statistically significant21% reduction inbreast cancer mortalitywith screening mammography.Their analysis of thedata confirmed thatscreening mammographyhas a real--if modest--effecton breast cancer mortalityand that this effectvaries with age.Limitations of mammography.Like any screeningtest, mammography hasinherent limitations. Therisk of a routine screeningmammogram yielding afalse-positive result may beas high as 10%.3 False-positivereadings are associatedwith psychological morbidityand financial burdens.The sensitivity ofmammography for occultbreast cancer ranges from83% to 95%.4 However,mammography fails to reveal17% to 30% of all intervalcancers.4 Moreover,normal findings on a mammogramcan produce afalse sense of security thatstops a woman from seekingappropriate medical attentioneven for a symptomaticlesion.Both false-negative andfalse-positive rates are especiallyhigh in women withdense breast tissue--whohave an increased risk ofbreast cancer. These highrates stem from the fact thatmammograms are particularlydifficult to interpretwhen breast tissue is dense.Finally, there is concernthat current screeningintervals may not identifysome aggressive tumors intime.Counseling individualwomen. Inform womenabout the benefits and potentialharms of mammographyfor their age group.Women younger than40 years. Less than 5% to7% of all breast cancersoccur in women youngerthan 40 years5; thus,women in their 30s shouldnot be screened unlessthey are at high risk forearly breast cancer (ie,they have first-degree relativesin whom breast cancerdeveloped before age45 years or they are knowncarriers of BCRA-1 orBCRA-2 mutations).Women aged 40 yearsand older. In the UnitedStates, nearly 25% of alldeaths and 33% of all yearsof life expectancy lost as aresult of breast canceroccur in women whosebreast cancers are found intheir 40s.6 Thus, it is reasonableto start screeningmammography at age 40years.Among women aged40 to 49 years, data on thebenefits of annual versusbiennial mammography arelimited. Some experts recommendannual mammographybecause of the lowersensitivity of the test inyounger women and theevidence that tumors growmore rapidly in this agegroup. Others are concernedabout the increasedrisk of false-positive resultsand diagnostic proceduresassociated with annualmammography in youngerwomen.3,7 Among womenaged 50 years and older,annual mammography haslittle additional benefit(compared with biennialscreening).Women older than 70years. The risk of breastcancer increases with agethroughout a woman's lifetime,and the absolute benefitof regular mammographyincreases along a continuumwith age. Theprecise age at which to discontinuescreening mammographyis uncertain becauseof the lack of data onwomen older than 70 years.Make the decision to continuescreening beyond age70 on an individual basis;consider such factors asshortened life expectancyand comorbid conditions.Follow-up of abnormalmammograms. Adoptan office system that ensurestimely and adequatefollow-up of abnormal mammograms.Timely follow-upcan optimize diagnosis andtreatment in women withabnormal results. TheAmerican College of Radiologydeveloped the BreastImaging Reporting andData System (BI-RADS) toincrease uniformity in theassessment of mammographyresults (Table 2).This system classifies resultsinto 5 categories andrecommends follow-up foreach category.8
GÃ¸tzsche PC, Olsen O. Is screeningfor breast cancer with mammographyjustifiable?
NystrÃ¶m L, Andersson I, BjurstamN, et al. Long-term effects of mammographyscreening: updatedoverview of the Swedish randomisedtrials.
Elmore JG, Barton MB, MoceriVM, et al. Ten-year risk of false positivescreening mammograms and clinicalbreast examinations.
N Engl J
Mushlin AI, Kouides RW, ShapiroDE. Estimating the accuracy of screeningmammography: a meta-analysis.
Am J Prev Med
Hankey BF, Miller B, Curtis R,Kosary C. Trends in breast cancer inyounger women in contrast to olderwomen.
Monogr Natl Cancer Inst
Ries L, Eisner M, Kosary C, et al.
SEER Cancer Statistics Review
, 1973-1997. Bethesda, Md: National CancerInstitute; 2000:1231.
Tabar L, Duffy SW, Vitak B, et al.The natural history of breast carcinoma:what have we learned fromscreening?
American College of Radiology.
Breast Imaging Reporting and Data System(BI-RADS
). 2nd ed. Reston, Va:American College of Radiology; 1995.