Missed Breast Carcino- RadioGraphics ma: Pitfalls and Pearlsnucleus.iaea.org/HHW/Radiology/Clinical... · exceed 1 cm. Emphasis on the upper outer quad-rant, which demonstrates the

EDUCATION EXHIBIT 881

Missed Breast Carcino-ma: Pitfalls and Pearls1

Aneesa S. Majid, MD ● Ellen Shaw de Paredes, MD ● Richard D.Doherty, MD ● Neil R. Sharma ● Xavier Salvador

Mammography is the standard of reference for the detection of breastcarcinoma, yet 10%–30% of breast cancers may be missed at mam-mography. Possible causes for missed breast cancers include dense pa-renchyma obscuring a lesion, poor positioning or technique, percep-tion error, incorrect interpretation of a suspect finding, subtle featuresof malignancy, and slow growth of a lesion. Recent studies have em-phasized the use of alternative imaging modalities to detect and diag-nose breast carcinoma, including ultrasonography (US), magneticresonance imaging, and nuclear medicine studies. However, the radi-ologist can take a number of steps that will significantly enhance theaccuracy of image interpretation at mammography and decrease thefalse-negative rate. These steps include performing diagnostic as wellas screening mammography, reviewing clinical data and using US tohelp assess a palpable or mammographically detected mass, strictlyadhering to positioning and technical requirements, being alert tosubtle features of breast cancers, comparing recent images with earliermammograms to look for subtle increases in lesion size, looking foradditional lesions when one abnormality is seen, and judging a lesionby its most malignant features.©RSNA, 2003

Abbreviation: CAD � computer-aided detection

Index terms: Breast neoplasms, 00.32 ● Breast neoplasms, radiography, 00.11 ● Diagnostic radiology, observer performance

RadioGraphics 2003; 23:881–895 ● Published online 10.1148/rg.234025083

1From the Department of Radiology, Medical College of Virginia, Virginia Commonwealth University, 401 N 12th St, Richmond, VA 23298. Recipi-ent of a Certificate of Merit award and an Excellence in Design award for an education exhibit at the 2001 RSNA scientific assembly. Received April22, 2002; revision requested May 23; final revision received April 25, 2003; accepted April 25. Address correspondence to E.S.d.P. (e-mail:[email protected]).

©RSNA, 2003

Radio

Gra

phic

s

IntroductionMammography is the standard of reference forthe early detection of breast cancer. Screeningmammography is performed to detect an abnor-mality, whereas diagnostic mammography is usedto further evaluate the abnormality or a clinicalproblem.

The purpose of screening mammography issimply to detect a potential cancer; therefore, theradiologist should not try to make a diagnosis onthe basis of screening findings alone. Additionalviews are important in further assessing an identi-fied abnormality and suggesting appropriate pa-tient treatment. According to data from theBreast Cancer Detection Demonstration Project,the false-negative rate of mammography is ap-proximately 8%–10% (1). After evaluating retro-spective versus blinded interpretations of mam-mograms, others have concluded that the rate ofmissed breast cancers is as high as 35% (2). In aseries of 150 mammograms (27 cancers) read by10 radiologists, immediate work-up of the truecancers was recommended in 74%–96% of cases(3). Recent studies have emphasized the use ofalternative imaging modalities to detect and diag-nose breast carcinoma, including ultrasonography(US), magnetic resonance (MR) imaging, and

nuclear medicine studies. However, high-qualitymammography performed with meticulous atten-tion to detail and positioning can significantlyenhance the accuracy of image interpretation.

Breast cancers may be missed because of denseparenchyma that obscures a lesion (4), poor posi-tioning or technique, lesion location outside thefield of view, lack of perception of an abnormalitythat is present, incorrect interpretation of a sus-pect finding, subtle features of malignancy, or aslowly changing malignancy. Breast cancers areeasily missed when they appear as focal areas ofasymmetry or distortion (eg, invasive lobular car-cinoma) or when their appearance suggests a be-nign cause (eg, medullary and mucinous [colloid]invasive ductal carcinomas, which usually mani-fest as mostly circumscribed masses) (5). Bird etal (6) found that 77 of 320 cancers (24%) in ascreening population were missed, primarily dueto dense breasts and a developing density that wasnot identified by the radiologist. Goergen et al (7)found that cancers missed at screening mammog-raphy were significantly lower in density and weremore often seen on only one of two views thanwere detected cancers. In a review of interval can-cers in the Malmo Screening Trial, Ikeda et al (8)found that 10 of 94 cases were missed due to ob-server error and 21 of 94 showed subtle signs ofmalignancy.

Figure 1. Invasive ductal carcinoma in a 36-year-old woman with dense breasts and a pal-pable mass. (a) Left mediolateral oblique mammogram demonstrates no finding that corre-sponds to a palpable mass (arrow). (b) US image obtained in the area of the palpable abnor-mality reveals a heterogeneous, hypoechoic mass with irregular margins. Although there isno acoustic shadowing and the mass is wider than it is tall, the hypoechogenicity and irregu-lar margins are suspect for malignancy. Pathologic analysis demonstrated invasive ductal car-cinoma.

882 July-August 2003 RG f Volume 23 ● Number 4

Radio

Gra

phic

s

In this article, we discuss and illustrate theaforementioned pitfalls that can lead to missedbreast cancers and provide guidelines to help re-duce the false-negative rate of mammography.

Causes of Missed Breast Cancers

Dense ParenchymaBreast parenchyma that is inherently dense com-promises the ability to detect a mass, especially anoncalcified, nondistorting lesion. The radiologistmust be particularly attentive in searching for ar-eas of architectural distortion or faint microcalci-fications. Magnification views are used to evaluatethe morphologic features of suspect or faint mi-crocalcifications. Because architectural distortionmay be the only sign of malignancy in a densebreast, the tissue must be intensely evaluated forany areas of tethering or disruption of orientation

of normal parenchymal elements. Unless it isdocumented as a postsurgical scar, an area of ar-chitectural distortion must be further evaluatedwith additional views (eg, spot compression, mag-nification, off-angle). US may also be helpful indetermining the presence of a solid mass that cor-responds to an area of distortion.

Any patient with dense breast parenchyma, apalpable mass, and negative mammographic find-ings should undergo US for further evaluation ofthe mass (Fig 1). US is very important in theevaluation of mammographic abnormalities, be-ing useful in characterizing palpable masses indense tissue and circumscribed isodense masses(Fig 2). US can be especially helpful in the evalu-ation of asymmetric densities seen at mammogra-phy because it can help identify the density as ei-ther breast tissue or a true mass. Soo et al (9) andSkaane (10) found the negative predictive valueof US with mammography for a palpable lesion to

Figure 2. Invasive lobular carcinomain a 40-year-old woman with densebreasts. (a) Right mediolateral obliquescreening mammogram shows a small,oval obscured mass superiorly (arrow)that was not seen on the craniocaudalview. (b) US image of the mass demon-strates a simple cyst. (c) US image re-veals an incidentally detected irregularmass with acoustic shadowing in thelower outer quadrant. Pathologic analy-sis demonstrated invasive lobular carci-noma.

RG f Volume 23 ● Number 4 Majid et al 883

Radio

Gra

phic

s

be 99.8% and 100%, respectively. Moy et al (11)found the negative predictive value of US withmammography for a palpable mass to be 97.4%.However, a palpable mass that appears solid atUS warrants further evaluation with biopsy.

Poor PositioningProper positioning and image contrast are abso-lutely necessary in all aspects of radiology, butespecially in mammography. The technologistmust adhere to the positioning standards to maxi-mize the amount of tissue included on the image(12). Findings on the mediolateral oblique viewthat indicate proper positioning include visualiza-tion of the pectoralis muscle to the level of thenipple, a convex appearance of the pectoralis ma-jor muscle, complete visualization of posteriorbreast tissue, breast tissue that is well compressedand positioned in an up-and-out orientation, andan open inframammary fold (Fig 3). At cranio-caudal imaging, the technologist should verifythat the breast is pulled straight forward and notexaggerated laterally, and that the breast tissue iswell compressed. The difference between the pos-terior nipple line measurement on the mediolat-eral oblique and craniocaudal views should notexceed 1 cm. Emphasis on the upper outer quad-

rant, which demonstrates the greatest proportionof breast cancers, is necessary. However, the tech-nologist must use the craniocaudal view as acomplement to the mediolateral oblique view tovisualize the medial tissue as well.

Creative positioning may be necessary to in-clude areas of palpable abnormalities on the im-ages. Radiopaque markers should be placed onpalpable areas, with repositioning of the markerbetween projections as needed to keep the markersuperimposed on the palpable finding. In addi-tion, a spot compression view obtained over apalpable mass with the skin in tangent can revealan underlying mass and demonstrate overlyingskin thickening or retraction. Creative positioningmay also be helpful in patients who are tense, whohave suffered a stroke, or who have shoulderproblems or other debilitating factors that limitvisualization of the posterior breast on standardmediolateral oblique views.

Off-angle or step oblique views are very helpfulin the evaluation of densities or abnormalitiesseen in only one projection (13). Densities seenon the craniocaudal view alone may be furthercharacterized and localized with use of spot com-pression and rolled craniocaudal views. If a lesionrolls medially when the top of the breast is rolledmedially, it is located superiorly; if it rolls later-ally, it is located inferiorly. The technologistshould label the image with the orientation in

Figure 3. Proper positioning. (a) Left mediolateral oblique (left) and craniocaudal (right) mam-mograms obtained with improper positioning demonstrate poor visualization of the posterior tis-sue. The margin of a mass is barely perceptible at the edge of the mediolateral oblique image (ar-row). (b) On a left mediolateral oblique mammogram obtained with improved positioning, a can-cer is seen near the chest wall. An exaggerated craniocaudal view may also help demonstrate such amass.


Radio

Gra

phic

s

which the top of the breast was rolled (eg, cranio-caudal RL � “craniocaudal rolled laterally”). If adensity is seen only on the mediolateral obliqueview, a mediolateral view is required to locate andfurther evaluate the lesion (Fig 4). In such a case,a medial lesion will move superiorly on the lateralview, whereas a lateral lesion will move inferiorly.This concept of triangulation is extremely impor-tant in identifying the actual position of a lesion.Off-angle or step oblique views, like standardviews, are most helpful when the lesion is super-imposed over fat and not dense tissue. Exagger-ated craniocaudal views may be helpful in demon-strating a posteriorly located lesion that is seen onthe mediolateral oblique view only. US may also

be helpful in verifying the location of a mass thatis clearly seen on only one view.

Poor TechniqueThe technologist must optimize image contrast toavoid obtaining over- or underpenetrated images.Proper positioning of the photocell is necessaryto achieve correct optical density on the image.Careful attention to daily processor quality con-trol is also necessary to optimize contrast. Thetechnologist should always review the images

Figure 4. Creative positioning for lesiondetection. (a) Bilateral mediolateral obliquemammograms show dense parenchyma withwell-defined masses (arrows) and a focal ir-regular density superoposteriorly on the rightside (arrowheads). The well-defined massesproved to be cysts at US. (b) On a right lat-eromedial mammogram, the irregular density(arrow) has moved upward, a finding that in-dicates a medial location. At lateromedialmammography, the medial aspect of thebreast is closer to the film and can thereforebe better evaluated. (c) Spot magnificationmammogram (right cleavage view) demon-strates a spiculated mass. Pathologic analysisrevealed invasive ductal carcinoma.


Radio

Gra

phic

s

under proper mammographic viewing conditionsto assess the adequacy of imaging technique (Fig5). Image blur is problematic, particularly in theassessment of microcalcifications. Rosen et al(14) found that in 62% of cancers that manifestedas microcalcifications and were incorrectly fol-lowed up with imaging rather than biopsy, imageblur on magnification views compromised imagequality.

Lack of PerceptionTwo major causes of missed breast cancers arerelated to radiologist error. The first of thesecauses is lack of perception. Perception error oc-

curs when the lesion is included in the field ofview and is evident but is not recognized by theradiologist. The lesion may or may not havesubtle features of malignancy that cause it to beless visible. Small nonspiculated masses, areas ofarchitectural distortion and asymmetry, and smallclusters of amorphous or faint microcalcificationsmay all be difficult to perceive.

To avoid perception error, images should bereviewed as mirror images, with mediolateraloblique images placed together and craniocaudalimages placed together (Figs 6, 7). The radiolo-gist should compare like areas on the side-by-sideimages to identify any focal asymmetric density orlow-density mass. Identification of a focal densityshould prompt a search for this density on thecorresponding view in the same arc from the

Figure 5. Proper imaging technique. (a) Rightcraniocaudal screening mammogram obtained in a 65-year-old woman demonstrates underpenetration.(b) Right mediolateral oblique mammogram reveals anirregular density (arrow) that was obscured on thecraniocaudal view. (c) Right craniocaudal spot magni-fication mammogram demonstrates an irregular masswith microcalcifications. At pathologic analysis, themass proved to be invasive ductal carcinoma.


Radio

Gra

phic

s

Figure 6. Drawings illus-trate useful search patterns inmirror image interpretation.CC � craniocaudal, MLO �mediolateral oblique.

Figure 7. Mirror image in-terpretation. (a) Bilateral me-diolateral oblique mammo-grams reveal an irregular massposteriorly on the left side witha highly suspect appearance(arrow). In addition, a subtledistortion is noted more inferi-orly (arrowhead), a findingthat becomes more evidentwith mirror image interpreta-tion. (b, c) On left craniocau-dal spot compression mam-mograms, the posterior (b)and anterior (c) lesions dem-onstrate a spiculated appear-ance (arrowhead in c). Patho-logic analysis demonstratedmulticentric invasive ductalcarcinoma.


Radio

Gra

phic

s

Figure 8. Diagrams illustrate nipple-to-lesion arc measurements used to determinelesion depth. a � distance from nipple toanterior lesion, b � distance from nipple toposterior lesion, CC � craniocaudal,MLO � mediolateral oblique.

Figure 9. Multicentric breast can-cer in a 63-year-old woman. Rightmediolateral oblique (a) and rightexaggerated craniocaudal lateral (b)screening mammograms show aprominent area of architectural dis-tortion at the 10 o’clock position(solid arrow). Note also the twosmall, indistinct masses in the axil-lary tail (arrowheads) and the lin-early arranged microcalcifications atthe 7 o’clock position (open arrow).An indistinct high-density node isalso seen in the axilla and proved tobe malignant at surgery. Pathologicanalysis demonstrated multicentricinvasive ductal carcinoma and duc-tal carcinoma in situ.


Radio

Gra

phic

s

nipple (Fig 8). Additional views may be needed toverify the presence of a true lesion.

Failure to diagnose multifocal and multicentricbreast cancers can directly affect patient treat-ment. Multifocal breast cancer is defined as twoor more cancers in the same quadrant, whereasmulticentric breast cancer is defined as two ormore cancers in different quadrants (Fig 9). Inmulticentric disease, breast conservation therapyis contraindicated. These disease entities may notbe perceived owing to “satisfaction of search,” inwhich observation of an obvious finding misleadsthe radiologist into not looking carefully for otherlesions (Fig 10). Careful attention must also bepaid to the contralateral breast after observationof a suspect lesion because contralateral synchro-nous cancers have been reported in 0.19%–2.0%

of patients (15) and may actually be seen in 9%–10% of patients at MR imaging (16). Satisfactionof search can also occur in cases of an obviousbenign lesion with a subtle cancer. The radiolo-gist must not be satisfied with finding just onelesion, but must search carefully for others,whether benign or malignant.

Another special circumstance that can presenta perception problem involves a patient with apalpable node in the axilla that is evaluated withbiopsy and represents metastatic adenocarci-noma, likely of breast origin. The primary breastcancer may be occult and either not observed orvery subtle at mammography. Careful attention

Figure 10. Satisfaction of search. Right mediolateral oblique (a) and craniocaudal (b) mammo-grams demonstrate subtle architectural distortion (arrow) behind an obvious calcified fibroad-enoma. The first interpreting radiologist noted the fibroadenoma but missed the distortion, whichproved to be invasive ductal carcinoma.


Radio

Gra

phic

s

to mirror image abnormalities or focal asymmet-ric densities is important in identifying the pri-mary lesion (Fig 11). MR imaging has been use-ful in identifying the primary carcinoma when ametastatic node is found in the axilla and mam-mographic findings are negative (17).

US may be helpful (like MR imaging and, oc-casionally, scintimammography) in the searchfor occult breast malignancy in special circum-stances, such as those involving patients withmulticentric cancer or with metastases to the ax-illa and no obvious breast lesion (18,19). Bergand Gilbreath (20) found preoperative wholebreast US to be complementary to mammogra-phy in patients with known breast cancer and inwhom breast conservation was planned. MR im-aging is becoming increasingly important in dem-onstrating the local extent of disease in patientswith breast cancer.

Incorrect InterpretationThe second major cause of missed breast cancersthat is related to radiologist error is incorrect in-terpretation of a lesion, which occurs when anabnormality with suspect features is observed butis misinterpreted as being definitely or at leastprobably benign. Several factors may lead to mis-interpretation, such as lack of experience, fatigue,or inattention. Misinterpretation may also occurif the radiologist fails to obtain all the viewsneeded to assess the characteristics of a lesion orif the lesion is slow growing and prior images arenot used for comparison. The radiologist mayerroneously judge the abnormality by its mostbenign features and miss important malignantfeatures that necessitate biopsy (Fig 12).

The margins of masses are best evaluated withspot compression imaging. A mass that appearsrelatively smooth may be indistinct or microlobu-lated on spot compression images. Therefore,margins should not be characterized on the basis

Figure 11. Occult cancer with metastases in a 36-year-old woman. (a) Right mediolateral oblique mammo-gram that was thought to be otherwise negative reveals an enlarged axillary node (arrow) that was palpable.(b) On a right mediolateral oblique mammogram obtained 3 months later while the patient was being evalu-ated for adenopathy, the previously occult cancer in the 11 o’clock position (arrowhead) became visible. Patho-logic analysis demonstrated invasive ductal carcinoma with metastasis to the axilla.


Radio

Gra

phic

s

Figure 12. Apparent lesion thinning at spot compression mammography. (a, b) Right craniocaudal (a) andmediolateral oblique (b) mammograms demonstrate focal architectural distortion (arrow in a) that may corre-spond to a superiorly located lesion (arrowhead in b). (c, d) Mediolateral oblique (c) and craniocaudal (d)spot compression mammograms show a persistent but less prominent area of distortion. At 6-month follow-upmammography, the area appeared more prominent, and biopsy was performed. Pathologic analysis demon-strated invasive ductal carcinoma. Rolled craniocaudal views were also obtained and helped confirm the persis-tence of the lesion.


Radio

Gra

phic

s

of a screening study alone. Any areas of microcal-cifications should be evaluated with magnificationviews to accurately define their morphologic fea-tures as well as their number and distribution.Characterization of a lesion that is identified atscreening mammography should be based on di-agnostic mammographic findings and not onscreening findings alone.

Subtle Signs of MalignancyThe cancers that are the most challenging to diag-nose and that most often lead to interpretationerrors are those with subtle or indistinct featuresof malignancy. These features include areas ofarchitectural distortion, small groups of amor-phous or punctate microcalcifications, focalasymmetric densities, dilated ducts, and relativelywell circumscribed masses. In a study of nonpal-pable cancers, Sickles (21) found that only 39%manifested with classic signs, including spiculatedmasses and linear microcalcifications.

Although well-circumscribed cancers are rela-tively uncommon, they do exist (22). Medullary,colloid (mucinous), and papillary carcinomacommonly manifest as well-circumscribed masses(Fig 13). Invasive ductal carcinoma not otherwisespecified is usually not circumscribed; however,because it occurs frequently, it accounts for themajority of circumscribed cancers. Spot compres-sion magnification of a seemingly circumscribedmass that proves to be a cancer will often demon-strate some area of indistinctness or microlobula-tion of the margin.

US is helpful in predicting the likelihood ofmalignancy in a circumscribed mass. Simple cystsseen at US constitute a benign finding. Solid le-sions that are smooth, elliptic, and wider thanthey are tall are probably benign. However,masses that have irregular or angulated margins,are markedly hypoechogenic, and are taller thanthey are wide are probably malignant (23). Anonpalpable circumscribed mass at mammogra-phy that demonstrates what are likely benign solidfeatures at US may be reevaluated at an early in-terval (24). If, however, the mass is seen at US asa solid lesion with worrisome features such as a“taller-than-wide” shape or irregular margins(23), biopsy is indicated. Any increase in thesize of a circumscribed, noncystic mass shouldprompt further evaluation with biopsy.

Asymmetric densities are frequently seen atmammography. These findings in isolation have alow positive predictive value for malignancy; how-ever, when they are associated with microcalcifi-cations or architectural distortion, the risk of ma-lignancy is increased (Fig 14). In a retrospective

review of interval cancers, Ikeda et al (8) foundthat 21 of 94 cases (22%) showed subtle signs ofmalignancy, mostly asymmetric densities. Otherworrisome features associated with focal asym-metric densities include interval enlargement, anew asymmetric density, a nonhormonal findingat mammography, and a palpable mass. Clinicalhistory is important in evaluating focal areas ofasymmetry. In the absence of tumor or infection,focal developing densities should prompt furtherassessment and, usually, biopsy. Rosen et al (14)found that 10 of 12 malignant areas of asymmetry(83%) were new, yet were incorrectly followed upby the radiologist. Hormonal changes are typi-cally diffuse and bilateral, although a focal de-veloping density can result from hormone re-placement therapy. A developing density that isthought to be hormonally related calls for discon-tinuation of therapy for 3–4 weeks, followed byrepeat mammography.

Invasive lobular carcinoma accounts for ap-proximately 8%–10% of breast cancers and is eas-ily missed because common manifestations in-

Figure 13. Circumscribed cancer in a63-year-old woman. Right exaggeratedcraniocaudal lateral mammogram dem-onstrates a nonpalpable mass in the ax-illary tail. The mass is lobulated andcircumscribed and has high density.Spot compression mammographywould help verify the characteristics ofthe margins. Pathologic analysis dem-onstrated mucinous carcinoma.


Radio

Gra

phic

s

clude a focal asymmetric density, an area of archi-tectural distortion, and negative mammographicfindings (25). US may demonstrate prominentfocal shadowing.

The work-up of a focal asymmetric densityincludes a clinical examination; additional mam-mography (spot compression and off-angle views)and US may also be helpful. However, negativeUS findings at the site of a suspect asymmetricdensity should not preclude biopsy. Dilated ductsare infrequently associated with malignancy. Pat-terns of ductal dilatation that suggest malignancyinclude a unilateral solitary dilated duct (21) anddilated ducts associated with microcalcificationsor in a nonsubareolar location (26).

Slow-growing CancersThe doubling time for breast cancers has beenreported to range from 44 to 1,869 days (27).However, malignant calcifications have been re-ported to be stable at mammography for as longas 63 months (28). Low-grade malignancies maynot undergo obvious change between annual in-terval screenings. Therefore, a slowly changingcancer may go undetected if the radiologist failsto compare findings with those on older images(Fig 15). A lesion with features that strongly

Figure 14. Asymmetric density. Bilateral mediolateral oblique (a) and craniocaudal (b) mammograms demon-strate a new focal asymmetric area in the left axillary tail (arrow), a finding that becomes more evident with mirrorimage interpretation. Biopsy revealed infiltrating lobular carcinoma.

Figure 15. Slow-growing cancer.(a) Right mediolateral collimatedmammogram shows focal architec-tural distortion superiorly (arrow).The area was not noted on sub-sequent images because it hadchanged imperceptibly. (b) Rightmediolateral collimated mammo-gram obtained 8 years later demon-strates interval growth of the lesion.Biopsy was performed, and patho-logic analysis demonstrated tubularcarcinoma.


Radio

Gra

phic

s

suggest malignancy but that has been stable for1–2 years still requires biopsy because it may rep-resent a slowly changing cancer. In particular,caution should be used in evaluating stablemasses or lesions with suspect morphologic fea-tures that decrease in size in patients who are re-ceiving tamoxifen. Tamoxifen is used to treatbreast cancers and to prevent the development ofbreast cancer in high-risk women, but it can alsobe used to check the growth of occult malignan-cies.

Role of Double ReadingDouble reading of mammograms has been shownto increase the detection rate for breast cancer byup to 15% (29,30). Computer-aided detection(CAD) represents a relatively new technologythat has been implemented in some mammogra-phy facilities for double reading. Clinical studieshave shown that CAD increases the sensitivity ofbreast cancer detection by radiologists by up to20% (31,32). The sensitivity of the CAD systemsis greater for detecting calcifications than for de-tecting masses (33). In a study of 115 cancersretrospectively judged to merit recall on thescreening mammogram prior to the mammogramon which they were diagnosed, 77% of lesionswere identified with CAD (34). In all, 86% of 35missed areas of calcifications and 73% of 80missed malignant masses were detected withCAD. We may continue to see increasing use ofboth CAD and a second radiologist for doublereading of screening mammograms.

ConclusionsAlthough mammography is the standard of refer-ence for the detection of early breast cancer, asmany as 30% of breast cancers may be missed.To reduce the possibility of missing a cancer, the

radiologist should take the following steps wheninterpreting mammographic findings:

1. Do not rely on screening views alone to di-agnose a detected abnormality; complete theevaluation with diagnostic mammography.

2. Review clinical data and use US to help as-sess a palpable or mammographically detectedmass.

3. Be strict about positioning and technicalrequirements to optimize image quality.

4. Be alert to subtle features of breast cancers.5. Compare current images with multiple

prior studies to look for subtle increases in lesionsize.

6. Look for other lesions when one abnormal-ity is seen.

7. Judge a lesion by its most malignant fea-tures.

Acknowledgment: The authors wish to thank LouiseLogan for manuscript preparation.

References1. Baker LH. Breast Cancer Detection Demonstra-

tion Project: five-year summary report. CA CancerJ Clin 1982; 32:194–225.

2. Harvey JA, Fajardo LL, Innis CA. Preview mam-mograms on patients with impalpable breast carci-nomas: retrospective vs blind interpretation. AJRAm J Roentgenol 1993; 161:1167–1172.

3. Elmore JG, Wells CK, Lee CH, et al. Variability inradiologists’ interpretations of mammograms.N Engl J Med 1994; 331:1493–1499.

4. Patel MR, Whitman GJ. Negative mammogramsin symptomatic patients with breast cancer. AcadRadiol 1998; 5:26–33.

5. Burrell HC, Sibbering DM, Wilson AR, et al.Screening interval breast cancers: mammographicfeatures and prognosis factors. Radiology 1996;199:811–817.

6. Bird RE, Wallace TW, Yankaskas BC. Analysis ofcancers missed at screening mammography. Radi-ology 1992; 184:613–617.

7. Goergen SK, Evans J, Colen GPB, Macmillan JH.Characteristics of breast carcinomas missed byradiologist. Radiology 1997; 204:131–135.


Radio

Gra

phic

s

8. Ikeda DM, Andersson I, Wattsgard C, et al. Inter-val carcinomas in the Malmo MammographicScreening Trial: radiologic appearance and prog-nostic considerations. AJR Am J Roentgenol 1992;159:287–294.

9. Soo MS, Rosen EL, Baker JA, et al. Negative pre-dictive value of sonography with mammography inpatients with palpable breast lesions. AJR Am JRoentgenol 2001; 177:1167–1170.

10. Skaane P. Ultrasonography as adjunct to mam-mography in the evaluation of breast tumors. ActaRadiol 1999; 420(suppl):1–47.

11. Moy L, Slanetz P, Moore R, et al. Specificity ofmammography and ultrasound in the evaluation ofa palpable abnormality: retrospective review. Radi-ology 2002; 225:176–181.

12. Hendrick RE, Bassett L, Botsco MA, et al. Mam-mography quality control manual. Reston, Va:American College of Radiology, 1999.

13. Pearson K, Sickles E, Frakel S. Efficacy of step-oblique mammography for confirmation and local-ization of densities seen on only one standardmammographic view. AJR Am J Roentgenol 2000;174:745–752.

14. Rosen EL, Baker JA, Soo MS. Malignant lesionsinitially subjected to short-term mammographicfollow-up. Radiology 2002; 223:221–228.

15. Kinne DW. Management of the contralateralbreast. In: Harris JR, Hellman S, Henderson P, etal, eds. Breast diseases. Philadelphia, Pa: Lippin-cott, 1987; 620–621.

16. Liberman L, Morris E, Kim C, et al. MR imagingfindings in the contralateral breast in women withrecently diagnosed breast cancer. AJR Am JRoentgenol 2003; 180:333–341.

17. Orel S, Weinstein SP, Schnall MD, et al. BreastMR imaging in patients with axillary node metas-tases and unknown primary malignancy. Radiol-ogy 1999; 212:543–549.

18. Khalkhali I, Vargas HI. The role of nuclear medi-cine in breast cancer detection: functional breastimaging. Radiol Clin North Am 2001; 39:1053–1068.

19. Orel SG. MR imaging of the breast. Radiol ClinNorth Am 2000; 38:899–913.

20. Berg WA, Gilbreath PL. Multicentric and multifo-cal cancer: whole breast ultrasound in preoperativeevaluation. Radiology 2000; 214:59–66.

21. Sickles EA. Mammographic features of 300 con-secutive nonpalpable breast cancers. AJR Am JRoentgenol 1986; 146:661–663.

22. Swann CA, Kopans DB, Koerner FC, McCarthyKA, White G, Hall DA. The halo sign and malig-nant breast lesions. AJR Am J Roentgenol 1987;149:1145–1147.

23. Stavros AT, Thickman D, Rapp CL, et al. Solidbreast nodules: use of sonography to distinguishbetween benign and malignant lesions. Radiology1995; 196:123–124.

24. Sickles EA. Nonpalpable circumscribed noncalci-fied solid breast masses: likelihood of malignancybased on lesion size and age of patient. Radiology1994; 192:439–442.

25. Mendelson EB, Harris KM, Doshi N, Tobon H.Infiltrating lobular carcinoma: mammographicpatterns with pathologic correlation. AJR Am JRoentgenol 1989; 153:265–271.

26. Huynh PT, Parellada JA, Shaw de Paredes E, etal. Dilated duct pattern at mammography. Radiol-ogy 1997; 204:137–141.

27. Fournier DV, Weber E, Hoeffken W, et al.Growth rate of 147 mammary carcinomas. Cancer1980; 45:2198–2207.

28. Lev-Toaff AS, Feig SA, Saitas VL, et al. Stabilityof malignant breast microcalcifications. Radiology1994; 198:153–156.

29. Anderson ED, Muir BB, Walsh JS, et al. The effi-cacy of double reading mammograms in breastscreening. Clin Radiol 1994; 49:248–251.

30. Thurjell EL, Lernevall KA, Taube AAS. Benefit ofindependent double reading in a population-basedmammography screening program. Radiology1994; 191:241–244.

31. Burhenne LJ, Wood SA, D’Orsi CJ, et al. Poten-tial contribution of computer-aided detection tothe sensitivity of screening mammography. Radiol-ogy 2001; 56:150–154.

32. Brem RF, Schoonjans JM. Radiologists detectionof microcalcifications with and without computer-aided detection: a comparative study. Clin Radiol2001; 56:150–154.

33. Lechner M, Nelson M, Elvecrog E. Comparison oftwo commercially available computer-aided detec-tion (CAD) systems. Appl Radiol 2002; 31:31–35.

34. Birdwell RL, Ikeda DM, O’Shaugnessy KF, Sick-les EA. Mammographic characteristics of 115missed cancers later detected with screening mam-mography and the potential utility of computer-aided detection. Radiology 2001; 219:192–202.


Radio

Gra

phic

s

Documents

Missed Breast Carcino- RadioGraphics ma: Pitfalls and Pearlsnucleus.iaea.org/HHW/Radiology/Clinical... · exceed 1 cm. Emphasis on the upper outer quad-rant, which demonstrates the