detection rates, and biopsy rates.Clearly, if one waits for all breast can-cers to reach a larger size threshold anda higher probability of malignancy be-fore call back, biopsy, and diagnosis,then these statistical measures wouldappear to improve at the expense of thepatient’s clinical outcome.

Instead, one hopes that future stud-ies will not only seek to measure the po-tential benefit of CAD in more meaning-ful terms (ie, specific type, size, andstage of cancers detected) as Pai andcolleagues have done, but will also be-gin to examine results for the most cur-rent technology available for that task-FFDM with integrated CAD.

T. E. Cupples, MD

References1. Zheng B, Hardesty LA, Poller WR, et

al: Mammography with computer-aided detection: Reproducibility assessment-Initial experience.Radiology 228:58-62, 2003.

2. Fenton JJ, Taplin SH, Carney PA, etal: Influence of computer-aided detec-tion on performance of screeningmammography. N Engl J Med356:1399-1409, 2007.

3. Freer TW, Ulissey MJ: Screeningmammography with computer-aideddetection: Prospective study of 12,860patients in a community breast center.Radiology 220:781-786, 2001.

4. Birdwell RL, Bandodkar P, Ikeda DM:Computer-aided detection withscreening mammography in a univer-

sity hospital setting. Radiology236:451-457, 2005.

5. Morton MJ, Whaley DH, Brandt KR,et al: Screening mammograms:Interpretation with computer-aided detection-prospective evaluation.Radiology 239:375-383, 2006.

6. Dean JC, Ilvento C: Improved cancerdetection using computer-aided detec-tion with diagnostic and screeningmammography: Prospective study of104 cancers. AJR Am J Roentgenol187:20-28, 2006.

7. Cupples TE, Cunningham JE,Reynolds JC: Impact of computer-aided detection in a regional screeningmammography program. AJR Am JRoentgenol 185:944-950, 2005.

250 Breast Diseases: A Year Book®

Quarterly250 Vol 18 No 3 2007

3–11Full-Field DigitalMammography on LCD VersusCRT Monitors Zuley ML, Willison KM, Bonaccio E, et al(Elizabeth Wende Breast Clinic, Rochester,NY; Roswell Park Cancer Inst, Buffalo, NY;Ovation Research Group, Highland Park, Ill;et al)

AJR Am J Roentgenol 187:1492-1498, 2006

Objective.—Our purpose was to de-termine if the display of full-field digitalmammograms on a 5-megapixel liquidcrystal display (LCD) monitor is at leastequivalent to the display of the same on a5-megapixel cathode ray tube (CRT)monitor.

Materials and Methods.—Five radi-ologists evaluated normal anatomy andfeatures of 61 abnormalities in 48 full-field digital mammograms. A 9-pointLikert scale was used to compare imageson two identical soft-copy review work-stations, one equipped with two 5-

megapixel CRTs and the other with two5-megapixel LCDs. Outcomes were eval-uated using a random-effects analysis ofvariance model. Means and SEs were re-ported. Ninety-five percent confidenceintervals and p values were calculated.

Results.—The two systems wereequivalent for most features. The LCDswere rated better for the sharpness ofmass margins (p = 0.011) and mass con-spicuity (p = 0.050). For calcium fea-tures, the LCDs were rated better than theCRTs for lesion conspicuity (p = 0.010)and number of calcifications (p = 0.043).For architectural distortions, there was nostatistically significant difference be-tween the monitors in any of the featuresevaluated. For display characteristics, theLCDs were better for luminance (p =0.021). The CRTs were significantly bet-ter for image noise (p = 0.001). In theoverall ratings, there was no statisticallysignificant difference between the twodisplays.

Conclusion.—The 5-megapixelmonochrome active-matrix LCD is

equivalent to and in some respects betterthan the 5-megapixel CRT display forfull-field digital mammograms over arange of normal and abnormal findings.

In this study, Zuley and colleaguescollected and reviewed outcomes frommultiple radiologists who subjectivelyevaluated image quality in side-by-sidecomparisons of two 5-megapixelmonochrome active-matrix LCD moni-tors and two 5-megapixel CRT moni-tors. The authors concluded that LCDand CRT monitors were mostly equiva-lent, although lesion conspicuity andnumber of calcifications was betterwith LCD monitors, and noise was bet-ter with CRT monitors.

There were several limitations tothis study. This study was conducted in asingle institution by dedicated breast ra-diologists; therefore, the conclusionsmight not reflect the opinions of generalradiologists. The study compared 1manufacturer’s CRT monitor versus 1manufacturer’s LCD monitor, and the

images were acquired by 1 manufactur-er’s full-field digital mammography(FFDM) system and technology. Thenumber of cases reviewed was small anddeliberately weighted by the authors, asdescribed in the study design. The radi-ologists’ interpretations of the displayswere subjective, and the authors ac-knowledged that some reader bias mighthave been introduced by the side-by-sidecomparative design of the study. In addi-tion, the authors did not specify whetherthe radiologists were aware that a studywas being conducted or whether theywere blinded to avoid additional bias.

Despite these potential limitationsand biases, this pioneer study is a signif-icant contribution to the literature.According to the U.S. Food and DrugAdministration Web site, FFDM is gain-ing increasing momentum in the United

States.1 Radiologists more commonlyread with soft-copy workstations be-cause of their ability to interpret imagesat different windows and levels and toutilize post-processing algorithms toevaluate images. Initially, CRT monitorswere introduced as a practical methodof achieving display in the soft-copyworkstation. However, LCD monitorsare replacing CRT monitors in other ar-eas of radiology because of their higherluminescence, longer life expectancy,lower weight, smaller volume, and re-duced heat load. With the decreasingcost of LCD monitors, more and morebreast centers are contemplating the im-plementation of LCD monitors. Cautionand scientific merit are required to en-sure that image quality is not negativelyimpacted in FFDM by this change insoft-copy display. To my knowledge, this

is the first study in the radiology litera-ture to investigate the display quality ofFFDM images using CRT versus LCDmonitors. Hopefully, this study will en-courage other institutions to scientifical-ly evaluate display systems for FFDM toensure image quality and patient safetyduring the FFDM transition in theUnited States.

J. Parikh, MD

Reference1. U. S. Food and Drug Administration.

FDA approved full field digital mam-mography systems (FFDM). Avail-able at http://www.fda.gov/CDRH/MAMMOGRAPHY/digital.html.Accessed April 14, 2007.

Breast Diseases: A Year Book®

Quarterly 251Vol 18 No 3 2007 251

3–12The Role of FDG-PET/CT inSuspected Recurrence ofBreast Cancer Radan L, Ben-Haim S, Bar-Shalom R, et al(Rambam Health Care Campus, Haifa,Israel; Univ College London; Israel Inst ofTechnology, Haifa)

Cancer 107:2545-2551, 2006

Background.—Early diagnosis of re-current breast cancer is crucial to selec-tion of the most appropriate therapy. Thecurrent study evaluated the role of FDG-PET/CT in the assessment of suspectedrecurrent breast cancer in patients whopresented with elevated serum tumormarkers.

Methods.—Forty-seven consecutiveFDG-PET/CT studies of 46 women (aged32-79 years; mean, 59.9 years) with a his-tory of breast cancer presented with ele-vated serum tumor markers 1-21 years

(mean = 6.2 years) after their initial diag-nosis and were retrospectively evaluated.PET/CT results were confirmed bypathology (n = 11), further imaging, andfollow-up (mean = 17.2 months; n = 36).Changes in further management based onPET/CT were recorded.

Results.—Thirty (65%) patients hadtumor recurrence, and 16 (35%) patientsshowed no further evidence of disease.Thirty-one patients had 32 abnormalPET/CT studies, and 15 patients had nor-mal studies with an overall sensitivity,specificity, and accuracy of 90%, 71%,and 83%, respectively. In 37 patients,PET/CT was compared with contrast-enhanced CT and had a higher sensitivity(85% vs 70%), specificity (76% vs 47%),and accuracy (81% vs 59%). PET/CT hadan impact on the management of 24 (5l%)patients. Of these, chemotherapy or radio-therapy was started in 16 patients, treat-ment was modified in 2 patients, and 6 pa-

tients were referred to biopsy, followedby referral to surgery for 2 patients.

Conclusions.—In patients withbreast cancer and rising tumor markers,FDG-PET/CT had high performance in-dices and was superior to CT for diagno-sis of tumor recurrence, which led tochanges in the subsequent clinical man-agement of 51% of these patients.

Radan and colleagues from Haifa,Israel, examined the performance offluorodeoxyglucose positron-emissiontomography/computed tomography(FDG-PET/CT) in a retrospectiveanalysis of 46 patients with a historyof breast cancer and elevated serumtumor markers. They found that 31(67%) patients had 32 abnormalPET/CT scans, similar to the findingsof other published reports in the sameclinical setting.1 By pathologic analy-sis, further imaging, or clinical follow-