BREAST MR IMAGING S147

Recognizing and Inter-preting Artifacts andPitfalls in MR Imagingof the Breast1

ONLINE-ONLYCME

See www.rsna.org/education/rg_cme.html.

LEARNINGOBJECTIVESAfter reading thisarticle and takingthe test, the reader

will be able to:

� List the indicationsand technical re-quirements for per-forming high-qualitybreast MR imaging.

� Describe the arti-facts and pitfallscommonly encoun-tered in breast MRimaging.

� Discuss techniquesand strategies for re-ducing or eliminatingthose artifacts andpitfalls.

Haydee Ojeda-Fournier, MD ● K. Ann Choe, MD ● Mary C. Mahoney,MD

Magnetic resonance (MR) imaging of the breast has evolved into animportant adjunctive tool in breast imaging with multiple and ever-increasing indications for its use. As with other types of MR imaging,there are a number of technical artifacts and pitfalls that can potentiallylimit interpretation of the images by masking or simulating disease.Because of the coils and computer-aided detection software specific tobreast MR imaging, there are additional technical considerations thatare unique to this type of MR imaging. Motion and misregistrationartifacts, wraparound artifact, susceptibility artifact, poor fat satura-tion, lack of contrast material, and poor timing of the contrast materialbolus are some of the artifacts and pitfalls that can make interpretationof breast MR images challenging and lead to misdiagnosis. Other im-portant considerations in proper interpretation of breast MR imagesinclude acquisition of a sufficient medical history, knowledge of benignand abnormal lesion enhancement, morphologic versus kinetic assess-ment, evaluation of areas outside the breast, and positioning. By usingthe recommended strategies, one can reduce or eliminate common ar-tifacts and pitfalls in breast MR imaging that prevent proper interpreta-tion of the results of this important diagnostic tool.©RSNA, 2007

Abbreviations: MIP � maximum intensity projection, SE � spin echo, TRAM � transverse rectus abdominis myocutaneous

RadioGraphics 2007; 27:S147–S164 ● Published online 10.1148/rg.27si075516 ● Content Codes:

1From the Department of Radiology, University of Cincinnati College of Medicine, Cincinnati, Ohio. Presented as an education exhibit at the 2006RSNA Annual Meeting. Received March 16, 2007; revision requested April 25 and received May 22; accepted May 30. K.A.C. consults for Johnson &Johnson (Peninsula Pharmaceuticals, Mountain View, Calif, and Ethicon Endo-Surgery, Cincinnati, Ohio); M.C.M. is a member of the speakers bu-reau for Johnson & Johnson (Ethicon Endo-Surgery); the other author has no financial relationships to disclose. Address correspondence to H.O.F.,Department of Radiology, Moores Cancer Center, University of California San Diego, 3855 Health Sciences Dr, #0846, La Jolla, CA 92093-0846(e-mail: hojeda@ucsd.edu).

©RSNA, 2007

Note: This copy is for your personal, non-commercial use only. To order presentation-ready copies for distribution to your colleagues or clients, use the RadioGraphics Reprints form at the end of this article.

See last page

TEACHING POINTS

IntroductionBreast cancer is the most common nonskin malig-nancy and the second most common cause ofcancer death in American women. Mammogra-phy has long been established as the only screen-ing study that can reduce breast cancer mortality.Despite this, mammography has significant limi-tations. Thus, it is imperative that other imagingmodalities be studied and developed to comple-ment mammography.

Breast magnetic resonance (MR) imaging hasemerged as an important adjunctive tool withmultiple and ever-increasing indications. As withall imaging modalities, a thorough understandingof the underlying technology, basic physics, andpotential limitations associated with breast MRimaging is imperative to maximizing its utility.Previous publications on general MR imagingphysics have described clinically significant arti-facts associated with this imaging modality (1–3).Subsequently, several authors have described arti-facts and pitfalls encountered in specific anatomicareas such as the abdomen, orbit, and spine (4–6). Despite this, little published literature is avail-able reviewing the artifacts and pitfalls specific tobreast MR imaging. In fact, a recent PubMedsearch yielded only a single reference to a pictorialreview of pitfalls in breast MR imaging (7).

The objective of this article is to present a pic-torial review of the common artifacts and pitfallsin MR imaging of the breast and recommendstrategies that will reduce or eliminate those is-sues that endanger proper interpretation of theresults of this important diagnostic tool. In addi-tion, we review the indications for and generaltechniques of breast MR imaging, as they are im-portant to the process of recognizing commonartifacts, pitfalls, and limitations of MR imagingof the breast.

Indications for Breast MR ImagingInitially, breast MR imaging was indicated for theevaluation of implant rupture in patients withbreast augmentation. At present, MR imagingremains the study of choice for evaluation of im-plant rupture. Since the introduction of gadolin-ium and the development of high-resolution MRimaging, multiple studies have shown MR imag-ing to be of value in evaluating the breast paren-chyma for breast cancer. The American Collegeof Radiology has developed a list of current indi-cations for breast MR imaging (8).

In addition to implant evaluation, these cur-rent indications include the following: evaluation

of newly diagnosed lobular or infiltrating ductalbreast cancer to assess extent of disease and toevaluate for possible contralateral disease; prob-lem solving for better lesion characterization; asan adjunct to screening in patients at high risk,including those with a personal history of breastcancer, a previous biopsy with proved high-riskresults at pathologic analysis, high-risk geneticmarkers, or a strong family history; evaluation ofresidual disease after lumpectomy with positivemargins; evaluation of chest wall invasion; evalua-tion of the breast parenchyma in metastatic dis-ease to the axilla from an unknown primary; as-sessment of response to neoadjuvant chemo-therapy; evaluation of breast cancer recurrence;and evaluation for recurrence in patients whohave undergone tissue transfer, such as transverserectus abdominis myocutaneous (TRAM) or la-tissimus dorsi flaps.

Breast MR imaging does not replace mam-mography. MR imaging may be helpful in se-lected cases where there are unanswered ques-tions after a complete clinical, mammographic,and sonographic evaluation.

Technique of Breast MR ImagingContraindications to MR imaging include thepresence of indwelling cardiac pacemakers, co-chlear implants, certain types of aneurysm clips,and a variety of metals that are susceptible to ahigh-strength magnetic field. Caution is recom-mended with the use of gadolinium in patientswith moderate to end-stage renal disease. Therehave been over 90 reported cases of nephrogenicsystemic fibrosis developing after the administra-tion of gadolinium contrast material (9). This hasled to the Food and Drug Administration and theAmerican College of Radiology recommendationsfor precautionary measures to be observed inhigh-risk patients undergoing infusion of gadolin-ium contrast material.

A variety of imaging protocols can be used toevaluate the breast. The following sequences areperformed at our institution in the standardbreast protocol: axial T1-weighted gradient echo,axial T2-weighted fat-saturated fast spin echo(SE), and sagittal pre- and postcontrast T1-weighted spoiled gradient echo after administra-tion of a 0.1 mmol/kg dose of gadopentetatedimeglumine. We use both fat suppression andsubtraction in the evaluation of dynamic postcon-trast images. Although there is no standard rec-ommendation, we advocate bilateral breast imag-ing for several reasons, including the usefulness ofassessing symmetry and evaluation of the con-tralateral breast in patients with newly diagnosedbreast carcinoma (10,11).

S148 October 2007 RG f Volume 27 ● Special Issue

TeachingPoint

Imaging protocols may vary depending on theequipment being used and the radiologist’s pref-erences. At a minimum, a field strength of 1.5 T,a dedicated breast coil, high resolution and thinsection thickness, and gadolinium enhancementare required for MR imaging of the breast. Com-puter-aided detection systems, which simplifycreation of subtraction and maximum intensityprojection (MIP) images, kinetic assessment, andthree-dimensional reformatting, assist in the in-terpretation of breast MR images. A recent articleby Rausch and Hendrick (12) reviews techniquesto optimize breast MR imaging. Finally, the avail-ability of MR imaging biopsy technology is neces-sary in the development of a breast MR imagingprogram.

Artifacts

Motion ArtifactA significant artifact in MR imaging is motion(Fig 1). Even with an optimal prescribed proto-col, any amount of motion can degrade imagequality or even render a study completely nondi-agnostic. The resultant reduced signal intensity ofa moving structure as well as blurring can obscurelesions. Both physiologic and nonphysiologicmovement can cause artifact in the phase-encod-ing direction. Since there is movement of a struc-ture between the sampling of different lines of

k-space (phase encoding), some of the signal fromthe tissue is displaced in the phase-encoding di-rection.

Physiologic motion can be caused by fluid, in-cluding pleural fluid, bowel fluid, or blood in ves-sels (Fig 2). Periodic motion from vessel pulsa-tion is specifically referred to as ghosting. Theclassic appearance is that of duplicated high signalintensity of a normal structure in the phase-en-coding direction. Other physiologic motions areattributed to respiration and gastrointestinal peri-stalsis and can also mimic or obscure lesions.Physiologic motion can be difficult to correct.With a standard sequence, a saturation band canbe used to decrease or eliminate ghosting whenplaced over the moving structure.

Nonphysiologic motion due to patient motionalso results in unsatisfactory images. Claustro-phobia and patient anxiety contribute to this arti-fact. Patient motion can be reduced by optimizingpatient comfort. This includes providing seda-tives, using physical restraints (straps), optimizingexamination time, and providing earplugs andblankets or a fan. An explanation of the studyand frequent communication with the patientthroughout the study also help reduce patientmotion.

Figure 1. Motion artifact. (a) Sagittal postcontrast T1-weighted fat-suppressed subtraction image(repetition time msec/echo time msec � 8.9/1.89) shows motion artifact. The image is blurred, and con-sequently a benign intramammary lymph node (arrow) centrally located within the breast is obscured.(b) On a sagittal T1-weighted MIP image, the lymph node (arrow) is clearly visualized because the im-age is not compromised by motion.

RG f Volume 27 ● Special Issue Ojeda-Fournier et al S149

TeachingPoint

TeachingPoint

Figure 2. Artifacts due to physiologic motion. (a, b) Axial (4716/98) (a) and sagittal (4250/98) (b)T2-weighted fat-saturated fast SE images show pulsation artifacts caused by a blood vessel (arrow). Thisghosting artifact causes degradation of portions of the images. (c) Axial T2-weighted fat-saturated fastSE image (3150/98) shows an artifact in the phase-encoding direction caused by a small amount of pleu-ral fluid (arrow). This artifact partly obscures visualization of the axilla. (d) Axial T2-weighted fat-satu-rated fast SE image (5500/98) shows an artifact caused by peristalsis and fluid in the stomach (arrow).This artifact is also seen in the phase-encoding direction and partially obscures portions of the image.

Figure 3. Misregistration artifact. Sagittal post-contrast T1-weighted MIP image (8.9/1.89) showsrepeating breast structures, an example of misregis-tration. This type of motion artifact occurs whenthere has been motion between pulse sequences,images from which are later subtracted.

S150 October 2007 RG f Volume 27 ● Special Issue

Misregistration ArtifactA type of artifact specific to subtraction imagingused in interpretation of breast MR images is mis-registration (Fig 3). This type of motion artifact isencountered in subtraction images when there ismovement between the images to be subtracted(ie, postcontrast T1-weighted image � precon-trast T1-weighted image � subtraction image).The term edge artifact is used to describe the colormapping artifact caused by subtle misregistrationthat occurs in computer-aided detection (Fig 4).This artifact is identified as a mass appearance ina single section of one plane. Reformatted multi-planar images demonstrate no mass. The edge of

the fat-parenchyma interface is color mapped in aplanar fashion.

Wraparound ArtifactAliasing or wraparound artifact results in the ap-pearance of portions of anatomic structures wherethey do not belong (Fig 5).

Aliasing occurs when there is excited tissuelocated outside the prescribed imaging field ofview. Tissues that are outside the imaging field ofview are also excited as part of the image acquisi-tion process, resulting in positional misregistra-tion. Simply stated, owing to the cyclic nature offrequency functions (ie, 10° is viewed as the sameas 370°) in the Fourier transform process, thetissues outside the prescribed imaging area aremisregistered as being located within the recon-structed image.

Although aliasing occurs in both the phase-and frequency-encoding directions, aliasing in thefrequency-encoding direction is commonly sup-pressed by using a frequency filter or by oversam-pling (3,13). Therefore, aliasing (or wraparound)is of practical importance in the phase-encodingdirection. In three-dimensional acquisitions, thisis important in the section-selection direction,which is also phase encoded. Aliasing in thephase-encoding direction can be minimized byincreasing the field of view, which compromisesresolution (for a given matrix size), or by over-sampling in the phase-encoding direction (at thecost of increased imaging time).

Figure 4. Misregistrationartifact. (a) Sagittal postcon-trast T1-weighted spoiled gra-dient-echo image (8.9/1.89)with color overlays shows mis-registration artifacts (arrows),which appear as masslikestructures in the sagittal plane.(b) Axial reformatted imageshows misregistration artifactsas bands of color overlay atfat-parenchyma interfaces (ar-rows) in the axial plane. Thistype of artifact is specific tothe computer-aided detectionprograms commercially avail-able for interpretation ofbreast MR images.

Figure 5. Wraparound artifact at imaging of a pa-tient who underwent mastectomy for invasive ductalcarcinoma. Axial T2-weighted fat-saturated fast SEimage (6300/98) shows wraparound artifact. The axillais completely obscured by a phantom arm (arrow).

RG f Volume 27 ● Special Issue Ojeda-Fournier et al S151

Susceptibility ArtifactBright spots, signal dropout, and tissue distortionare the imaging characteristics of susceptibilityartifact (Fig 6).

In the presence of the main magnetic field, tis-sues and other objects are magnetized to varyingdegrees. The subsequent effect on the image ismost noticeable around metallic objects due tothe larger induced field. The metallic object doesnot have mobile protons and therefore does notemit an MR signal. However, the induced fixedheterogeneities of the magnetic field cause addi-tional artifact around the metallic object. In thepresence of metal, there are larger changes in thelocal magnetic field, which cause rapid dephasingof spins with resultant signal loss in the region(3). In addition, since spatial position in the im-age is created by the intentional addition of mag-netic field gradients, the unexpected alterations ofthe local field will change the expected preces-sional frequencies, thereby artificially displacingvoxels in the image.

These artifacts appear more prominent on gra-dient-echo images due to the absence of the 180°

refocusing pulse. In contrast, owing to the mul-tiple 180° pulses in fast (turbo) SE imaging, theartifact is minimized when this sequence is used.Similar but less prominent effects are seen due tothe varying magnetic susceptibility of differenttissues, such as bone and soft tissue. The size ofthe susceptibility artifact is dependent on the sizeand composition of the metallic object (clip), withpure titanium producing the smallest artifact (14).

Artifacts Due to Body HabitusMany different types of artifacts can be caused bya patient’s body habitus. Until recently, breastcoils were manufactured in only one size. In obesepatients, artifacts occurred from breast tissue out-side the coil and from the breast touching the coil.Figure 7 demonstrates artifacts that occur when avery large breast is placed in a coil. There areweight limits for MR imaging tables, which varyby manufacturer. In addition, the width of thepatient can be a limiting factor. The bore of themagnet may not accommodate patients withbroad shoulders or hips. Placement of the pa-tient’s arms down to her side or above her headcan help fit some larger women into the magnetbore.

Figure 6. Susceptibility artifact. (a) Axial T1-weighted gradient-echo image (400/4.19)shows focal signal voids (arrowheads) in the axilla. The signal voids are the result of sus-ceptibility artifact from surgical clips placed at the time of lymph node dissection for breastcancer. The blooming artifact caused by the surgical clips obscures portions of the axilla.(b) Axial T1-weighted gradient-echo image (100/4.17) shows a signal void (arrow) from asurgical clip placed at the time of lumpectomy. The degree of signal void varies dependingon the size and composition of the metallic object. Titanium induces less susceptibility arti-fact than stainless steel. (c) Sagittal three-plane localizing image (91.8/1.7) shows signalvoids (arrowheads) at the sites of sternotomy wires after placement of a coronary artery by-pass graft. The mediastinal region including the internal mammary lymph node chain is ob-scured by the signal voids.

S152 October 2007 RG f Volume 27 ● Special Issue

Inhomogeneous Fat SaturationThe breast is composed of variable amounts offat. Fat is hyperintense on T1-weighted images,as is gadolinium. By suppressing the signal fromfat, gadolinium enhancement is easier to detect.Several techniques are available for fat saturation,including frequency-selective (chemical) fat satu-ration. This requires a field strength greater than1 T.

Frequency-selective fat suppression exploitsthe precessional frequency difference between theprotons in fat molecules and those in water mol-ecules, which is 220 Hz at 1.5 T. To effectivelysuppress the protons in the fat molecules, the cor-rect range of frequencies must be selected. Sincethe frequency of precession is dependent on themagnetic field experienced by the proton, varia-tions in the magnetic field will alter the actualprecessional frequency from the expected. There-fore, in the presence of an unexpected variation inthe magnetic field, there will be protons in fat thatare precessing out of the range of frequencies in-cluded in the suppression pulse. These protonswill not be suppressed, and the fat containing

these protons will maintain its brighter signal.This results in inhomogeneous suppression of thefat signal within the breast.

Many factors can alter the magnetic field, in-cluding metallic objects. Human tissue, which isdiamagnetic, also causes alterations in the mag-netic field and can cause frequency-selective fatsuppression to be problematic when variable tis-sue types are being imaged together. In breastimaging, this is particularly true due to the largeamount of air in the thoracic cavity and to the airgap that can exist between the breasts.

Inhomogeneous fat saturation is easily identi-fied as areas of hyperintense fat on fat-suppressedimages (Fig 8). This process can involve portionsof one breast or the entire imaged field of view.Enhancing lesions could be obscured by poor fatsaturation and easily missed. Although somecauses of inhomogeneous fat suppression cannotbe corrected for, tuning the shim (optimizing fieldhomogeneity) in the imaging unit can correctsome of the artifact.

Pitfalls

Insufficient HistoryThe availability of a thorough clinical history fa-cilitates breast MR image interpretation. Perti-nent information includes the indication for thestudy, prior surgical history, family history, andmenstrual cycle and hormone use history. At ourinstitution, this information is obtained from thereferring health care provider at the time thebreast MR imaging study is scheduled. At thetime of imaging, the patient completes two ques-tionnaires, an MR imaging safety form and a fo-cused medical history form.

Errors in interpretation can occur if the appro-priate clinical and surgical history is not takeninto consideration. For example, Figure 9 dem-onstrates MR imaging changes after TRAM flap

Figure 7. Artifacts due to large breast size.Sagittal T2-weighted fat-saturated fast SE image(4250/98) shows peripheral bright areas (arrows)caused by the breast coil touching the skin. Untilrecently, breast coils were available in only onesize. Patients with large breasts could demon-strate flattening (arrowheads) and bulging of thebreast tissue in addition to the bright spot artifactfrom the coil. The possibility of a delay in deliv-ery of contrast material to the breast caused bypinching of blood flow is a possible effect of largebreasts in the coil. In women with very largebreasts, MR imaging examination might not bepossible if the breast tissue cannot be accommo-dated in the coil.

Figure 8. Inhomogeneous fat saturation. Axial T2-weighted fat-saturated fast SE image (6750/67.6)shows marked inhomogeneous fat saturation. The en-tire left breast (arrow) is not fat saturated, an appear-ance that can obscure lesions.

RG f Volume 27 ● Special Issue Ojeda-Fournier et al S153

reconstruction with a focal area of fat necrosis.Without the history, the postsurgical findingswould be perplexing and possibly misinterpreted.MR is an effective imaging modality in assess-ment of breast reconstruction with TRAM flapsand allows accurate differentiation of benign andmalignant conditions (15). Benign findings in the

reconstructed breast include skin thickening, fi-brosis, fat necrosis (as shown in our example),and seroma. Chest wall and axillary recurrencescan be identified with MR imaging even whenclinically and mammographically occult.

Knowledge of the patient’s last menstrual pe-riod in premenopausal women or hormone use inpostmenopausal women is also highly relevant toproper image interpretation. Hormone-induced

Figure 9. Importance of the surgical history in a 66-year-old patient with a history of ductal carcinoma insitu, mastectomy, and TRAM flap reconstruction.Axial T1-weighted gradient-echo (400/4.19) (a) andsagittal postcontrast T1-weighted MIP (89/1.891) (b)images show asymmetry in size and shape between theright and left breasts. The area of architectural distor-tion and enhancement (arrow) was known to repre-sent fat necrosis. If these images were to be interpretedwithout the surgical history, it could be easy to misin-terpret the area of fat necrosis as a suspicious mass.

Figure 10. MR imaging performed at two different times in the menstrual cycle of a 42-year-old woman with recently diagnosed invasive ductal carcinoma. (a) Sagittal postcontrastT1-weighted MIP image (8.9/1.89) shows the invasive ductal carcinoma (arrow). Ductalextension toward the nipple was suspected; however, this finding is obscured by extensiveproliferative changes. (b) Sagittal postcontrast T1-weighted MIP image (8.9/1.89) obtained12 days later clearly shows linear extension toward the nipple (arrows). The optimal time forperforming the examination is between days 3 and 21 of the menstrual cycle. A mass or amore subtle finding could be obscured by the presence of innumerable proliferative foci ofenhancement.

S154 October 2007 RG f Volume 27 ● Special Issue

proliferative changes can mimic or obscure truelesions (Fig 10). Although some investigatorshave reported that the optimal time to performbreast MR imaging is between days 3 and 14 ofthe menstrual cycle (16), we have found that im-aging up to day 21 yields accurate results.

Lack of Contrast EnhancementAbsence of contrast enhancement can be difficultto detect, particularly with inherent breast tissuecontrast present. Figure 11a–11c provides an

Figure 11. Lack of contrast enhancement. (a–c) Sagittal MR images of a patient who failed to receive intravenouscontrast material due to a technical error. There is lack of enhancement between pre- (a) and postcontrast (b) T1-weighted images (8.9/1.89) with normal intrinsic signal intensity in the breast parenchyma. Note the deceptively nor-mal appearance of the MIP image (c). Enhancing normal structures such as the nipple and blood vessels are also ab-sent, a strong indication of lack of contrast material. (d, e) MIP images (8.9/1.89) from two different studies of a 17-year-old girl with grade II infiltrating ductal carcinoma who underwent MR imaging to evaluate the extent of disease.The initial study was suspected to have suboptimal contrast enhancement owing to lack of prominent vessel enhance-ment throughout the entire study; the study was repeated. Image from the repeat study shows a prominent abnormalfocus of enhancement in the contralateral breast (arrow in e); the lesion is less apparent on the image from the initialstudy (arrow in d). At MR imaging–guided biopsy, the lesion proved to be ductal carcinoma in situ.

RG f Volume 27 ● Special Issue Ojeda-Fournier et al S155

TeachingPoint

example of complete lack of contrast enhance-ment. Even the best trained technologists occa-sionally overlook a missed contrast material bo-lus. It is up to the interpreting radiologist to assesseach study for contrast enhancement quality. Acompletely negative study, including negativesubtraction and MIP images, is a warning sign tofurther evaluate the study for adequate contrastenhancement. Lack of contrast enhancement inthe heart, absence of normal nipple enhancement,and absence of enhanced vessels within the breast(easily appreciated on MIP images) are excellentmarkers for evaluating the contrast material bo-lus.

In addition, inadequate contrast material ad-ministration may result from poor intravenousaccess, failure of power injectors, and intravenoustubing malfunction, as well as the patient’s hemo-dynamic status. A poor contrast material bolusmay be more difficult to perceive than completelack of enhancement. Figure 11d and 11e depictsa case in which the initial MR imaging examina-

tion demonstrated poor contrast enhancement.At follow-up examination, performed with a bet-ter contrast material bolus, an area of ductal car-cinoma in situ became apparent.

Complete lack of enhancement results in anondiagnostic examination and needs to be re-peated. In cases where the contrast material bolusis of questionable quality, either a repeat exami-nation or a short-interval follow-up can be con-sidered, taking into account the pretest probabil-ity of malignancy in each individual case.

Nipple EnhancementThe nipple enhances normally to varying intensi-ties at breast MR imaging (Fig 12a) (17). Thisenhancement is due to the rich blood supply ofthe nipple-areolar complex. Abnormal nipple en-hancement includes skin thickening and enhance-ment extending into the areola and periareolarbreast tissue (Fig 12b). The differential diagnosisof abnormal nipple enhancement and thickeningincludes Paget disease, lymphatic obstruction,inflammatory breast carcinoma, infection, andinflammation. Clinical evaluation of the area andpunch biopsy readily provide a diagnosis. MR

Figure 12. Nipple enhancement. (a) Sagittal MIP image shows normal nipple enhancement. (b) Sagittal MIP im-age shows abnormal nipple enhancement. The patient was a 19-year-old woman with nipple and periareolar skinthickening and inflammation at clinical examination. MR imaging was performed to rule out Paget disease and anunderlying mass. The final diagnosis from skin punch biopsy was eczema. (c) MR image (8.9/1.89) shows a nipplethat was pushed into the breast parenchyma (arrow) owing to the large size of the breast. This finding could be misin-terpreted as a subareolar enhancing mass lesion.

S156 October 2007 RG f Volume 27 ● Special Issue

imaging in these instances is indicated when thereis suspicion of underlying occult malignancy.

An additional potential pitfall involving nippleenhancement is the possibility of misinterpretinga normal nipple for a mass when the nipple is flat-tened against the anterior surface of the coil ow-ing to the large size of the breast (Fig 12c). Ourpractice has received requests to biopsy “lesions”that were nothing more than a normal nipplepushed into the breast parenchyma. In thesecases, we find it helpful to view the MIP imagesand three-dimensional reformatted images to de-termine whether the enhancing lesion is indeedthe nipple.

Rim EnhancementRim enhancement can be seen with malignancy,fat necrosis, or cysts, particularly complicated

cysts (18). Use of T2-weighted sequences, non–fat-suppressed sequences, and second-look ultra-sonography (US) can help in evaluation (Fig 13).Postoperative seromas can also show peripheralenhancement due to inflammation (Fig 14a–14c).A thin rim of enhancement is likely benign whencorrelation with pathology results indicates nega-tive margins at resection. In contrast (Fig 14d,14e), necrotic invasive cancers can have a thickirregular enhancing rim as well as bright T2 sig-nal. However, in contrast to the T2 signal of cystsor seromas, the T2 signal of necrotic tumor isusually heterogeneous; manipulation of windowlevel settings is necessary to demonstrate thisfinding, since high brightness can obscure theunderlying heterogeneity.

Figure 13. Rim-enhancing cyst in a patientwith a history of invasive breast cancer who wasbeing evaluated for possible contralateral disease.(a) Sagittal postcontrast T1-weighted image(8.9/1.89) shows a lesion with peripheral highsignal intensity (arrow). The differential diagno-sis included fat necrosis, a proteinaceous cyst,and invasive cancer. (b) Axial T2-weighted im-age (5500/98) shows water signal intensity in thelesion (arrow). (c) Transverse second-look USimage shows that the lesion is a cyst (arrow),which was successfully aspirated.

RG f Volume 27 ● Special Issue Ojeda-Fournier et al S157

Morphologic ver-sus Kinetic AssessmentA combined model that uses both morphologiccharacteristics and kinetic assessment has beenshown to improve diagnostic accuracy (19). Theaccepted algorithm in breast MR image interpre-tation is to assess the morphologic features of anMR imaging–detected abnormality to determineif a biopsy is warranted. If there is an indetermi-nate lesion, then kinetic assessment is applied toincrease specificity. A postoperative scar shouldnot enhance (Fig 15).

Lesions Outside the BreastAs with all other imaging modalities, a systematicevaluation of other organ systems included in thestudy is necessary. Other organs typically in-cluded in the field of view for breast MR imaginginclude portions of the lung, heart, liver, gallblad-der, and stomach. Particular attention should bepaid to the axilla and other lymph node drainage

basins during interpretation of breast MR images.Findings in these areas may be clinically relevant.

Gastrointestinal Tract.—Common findings inthe liver include hemangiomas and cysts (Fig 16).Cysts are encountered in 2.5% of the populationand are more common in women (20). Cysts arecharacterized by low signal intensity on T1-weighted images and high signal intensity on T2-weighted images. Postcontrast images demon-strate a lack of enhancement. Hemangioma is

Figure 14. Rim enhancement in benign and malignant lesions.(a–c) Postoperative seroma in a patient who had undergone re-cent lumpectomy for breast cancer with negative margins. SagittalT2-weighted image (5150/98) (a), sagittal T1-weighted subtrac-tion fast SE image with color overlays (8.9/1.89) (b), and MIPimage (c) show a lesion (arrowhead in a) with rim enhance-ment (arrowheads in b and c). This appearance was interpretedas peripheral enhancement due to postoperative inflammation.(d, e) Grade III invasive ductal carcinoma in a 53-year-old woman.Sagittal T1-weighted subtraction fast SE (8.9/1.89) (d) and axialT2-weighted (5500/98) (e) images show a mass with irregular rimenhancement (arrow), in contrast to the smooth thin rim enhance-ment shown in a–c. The thick irregular rim enhancement is charac-teristic of malignancy. The mass has heterogeneous high signal in-tensity on the T2-weighted image, an appearance that should notbe misinterpreted as a cyst.

S158 October 2007 RG f Volume 27 ● Special Issue

another hyperintense T2 lesion in the liver that iscommonly encountered. This benign lesion dem-onstrates characteristic peripheral nodular en-hancement on postcontrast images. Metastases tothe liver are common in the setting of breast can-cer and appear as heterogeneous lesions with en-hancement.

The stomach and portions of the gastrointesti-nal tract are not well evaluated because of physi-ologic motion. In the gallbladder, cholelithiasiscan be detected as filling defects within the fluid-

filled gallbladder (Fig 17). This is well appreci-ated on T2-weighted images, paralleling the tech-nique of MR cholangiopancreatography.

Thorax.—In the thorax, pleural fluid is a com-mon finding. Pleural effusions can range fromtrace amounts of physiologic fluid (see Fig 2c) topleural effusions of varying sizes. Lung massesmay represent metastases (Fig 18).

Figure 15. Surgical scar in a patient whounderwent lumpectomy for breast cancerwith clear margins. Sagittal postcontrastT1-weighted image (8.9/1.89) shows themorphologic finding of suspicious architec-tural distortion (arrowheads) with lack ofenhancement. This appearance is concor-dant with the clinical history and repre-sents a surgical scar. However, architec-tural distortion without enhancement inany other setting would be considered asuspicious finding, with biopsy recom-mended.

Figures 16, 17. (16) Liver hemangioma. Axial (4716/98) (a) and sagittal (4550/98) (b)T2-weighted fat-saturated images, obtained at the level of the hepatic dome, show a hyperin-tense lesion (arrow) in the liver, a finding consistent with a hemangioma. Other liver lesionsthat are hyperintense on T2-weighted images, such as cysts and biliary hamartomas, are alsoconsidered benign and require no further diagnostic work-up. (17) Cholelithiasis. Axial T2-weighted fat-saturated image (6100/66) shows an incidentally detected gallstone (arrow).Also note the fluid-filled stomach.

RG f Volume 27 ● Special Issue Ojeda-Fournier et al S159

Lymph Node Evaluation.—Abnormal lymphnode findings (Fig 19) are of vital clinical signifi-cance. Knowledge of lymphatic drainage is a req-uisite for breast imaging. The majority of thelymph drainage is to the axilla, with a fractiondraining into the internal mammary chain. Thebreast MR imaging search pattern must include athorough evaluation of axillary, interpectoral(Rotter node) (Fig 20a), and internal mammarychain (Fig 20b) nodes (21). When coronal imag-ing is performed, the supraclavicular lymph nodebasins may also be visualized and need to be in-cluded in the search pattern.

As with other imaging modalities, MR imagingevaluation of lymph nodes is based on size andmorphology. Micrometastases cannot be detectedwith MR imaging. Enhancement characteristics

are not helpful since normal lymph nodes avidlyenhance and abnormal lymph nodes may demon-strate lack of enhancement. Recognition of nor-mal lymph nodes, particularly intramammarylymph nodes, is important so that they are notmisinterpreted as masses (Fig 21). As with otherimaging modalities, at MR imaging normal lymphnodes contain a fatty hilum, are adjacent to a ves-sel, and have a reniform shape. Internal mam-mary lymph nodes can be difficult to visualize.The average size of an internal mammary lymphnode involved with metastasis is 6 mm, but it canbe as small as 4 mm (22). For this reason, weconsidered any lymph node visualized in the in-ternal mammary chain to be abnormal. Compari-son with previous MR imaging studies, mammo-grams, and follow-up US can be helpful when avisualized breast mass is suspected to be a lymphnode.

Figure 18. Lung metastases in a patientwith newly diagnosed breast cancer. MR im-aging was performed to evaluate the extent ofdisease and to check for contralateral disease.Sagittal postcontrast T1-weighted image (6.2/3.0) shows multiple lung metastases (arrow),an incidental but significant finding. The me-tastases were unsuspected, and their presencechanged the management of this case.

Figure 19. Abnormal lymph node in a 53-year-old woman with grade III invasive ductal carcinoma.Axial T2-weighted (5500/98) (a) and sagittal postcontrast T1-weighted MIP (8.9/1.89) (b) images showan abnormal axillary lymph node (arrow). Arrowheads in b � invasive ductal carcinoma. Fine-needleaspiration of the lymph node was performed under US guidance. Cytologic analysis demonstrated malig-nant cells, a finding consistent with metastatic high-grade adenocarcinoma. Because of this finding, asentinel lymph node biopsy was avoided and a full axillary dissection was performed.

S160 October 2007 RG f Volume 27 ● Special Issue

PositioningBreast MR imaging can be compromised bybreast tissue located outside the coil. Breast tissueexcluded from the coil may become folded andcompressed, resulting in loss of symmetry. Main-tenance of proper symmetry is ensured by properpositioning. We have found it helpful to train agroup of female MR imaging technologists to po-sition the breasts evenly within the coil, just as themammographic technologists manipulate andposition the breast within the compression

paddles. It is not prudent to allow the patient toposition herself in the coil. Finally, poor visualiza-tion of axillary tissue and the retroglandular tissuecan result from inadequate positioning of thebreast in the coil. Figure 22 shows an example ofa poorly positioned breast within the coil. Bloodflow can potentially be obstructed or delayed bypinching or compression caused by improperlypositioned breast tissue.

Figure 20. Other lymph node basins to be evaluated with breast MR imaging. (a) Sagittalpostcontrast T1-weighted image (550/8) shows an interpectoral (Rotter) lymph node (ar-row). (b) Sagittal T1-weighted MIP image (6.2/3.0) shows an internal mammary lymphnode (arrow). Note the internal mammary artery adjacent to the abnormal lymph node.

Figure 21. Normal intramammary lymph node. Axial T1-weighted (100/4.196) (a), close-up axial reformattedT2-weighted (6800/3.04) (b), and sagittal postcontrast T1-weighted spoiled gradient-echo (8.9/1.89) (c) imagesshow an intramammary lymph node (arrow) that mimics a mass. The well-defined margins, high signal intensity onthe T2-weighted image, fatty hilum, and proximity to a vessel are all characteristics of a benign lymph node.

RG f Volume 27 ● Special Issue Ojeda-Fournier et al S161

Tumor Heteroge-neity and Kinetic AssessmentHeterogeneous enhancement is identified by vari-able signal intensity within a mass. This is a signof malignancy, likely reflecting a polymorphouscell population and tumor necrosis. When evalu-ating the enhancement of lesions and assessingkinetic characteristics, the worst-appearing curveshould be considered in the final assessment (Fig23). Choosing the most benign-appearing curve

and ignoring the worst curve can lead to misdiag-nosis.

Interpretation of Kinetic CurvesKuhl and colleagues (23) have described threetime–signal intensity curves that are important indifferentiating benign from malignant lesions.The type I curve (Fig 23b) is a slow steady en-hancement curve. The type II curve demonstratesplateau signal intensity (Fig 23c). The type IIIcurve is associated with washout of signal inten-sity (Fig 23d) and is a strong indicator of malig-

Figure 22. Poor positioning of the breasts in the breastcoil. Axial T2-weighted fat-suppressed image (6000/66.6)shows that the posterior one-third of each breast (arrow-heads) is outside the breast coil. The large invasive ductalcarcinoma (arrow) in the lateral right breast is com-pressed; a more subtle tumor could easily be missed orcould not enhance optimally secondary to compression. Inour practice, a group of female MR imaging technologistshave been trained to properly position the breast tissuewithin the coil, paying attention to symmetry as well as thepatient’s comfort.

Figure 23. Tumor heteroge-neity in a 69-year-old patientwith grade I and III invasiveductal carcinoma. (a) Sagittalpostcontrast T1-weighted sub-traction image (8.9/1.89)shows heterogeneous tumorenhancement (arrow).(b–d) Computer-generatedkinetic curves obtained withinthe lesion show the three clas-sic types of kinetic assessmentcurves: the slow persistentcurve (type I) (b), plateaucurve (type II) (c), and wash-out curve (type III) (d).

S162 October 2007 RG f Volume 27 ● Special Issue

nancy. The use of time–signal intensity curves hasa sensitivity of 91% and specificity of 83% (23).

Artifactual curves are evident as multiple peaksand valleys, an appearance not consistent withcontrast material perfusion through vessels (Fig24). Kinetic time–signal intensity curves aregraphic representations of physiologic processesand should follow expected vascular flow pat-terns.

Mammographic and US FindingsInterpretation of breast MR images requiressimultaneous evaluation with mammographyand other available imaging studies. Evaluationof breast density and benign lesions is straight-forward if the appropriate correlative studiesare performed (Fig 25). At our institution, we

Figure 24. Interpretation ofa kinetic curve. Sagittal post-contrast T1-weighted image(8.9/1.89) (a) and computer-generated kinetic curve (b)show an area of artifactual en-hancement (arrow in a). Thecurve is easily identified as ab-normal because of its multiplepeaks and valleys. The artifac-tual enhancement was likelyproduced by subtle misregis-tration. We caution againstinterpreting color overlay im-aging findings in isolationfrom morphologic findings.

Figure 25. Correlation with mammographic findings. (a) Close-up mediolateral oblique mammogram shows awell-circumscribed mass (arrow) containing coarse calcifications in the inferior aspect of the breast. (b) Sagittal post-contrast T1-weighted image (8.9/1.8) shows that the lesion has well-defined margins (arrow) and nonenhancing cen-tral septa (arrowhead). The diagnosis of a hyalinizing fibroadenoma at breast MR imaging is straightforward, particu-larly when the MR imaging findings are interpreted in conjunction with the characteristically benign appearance ofthe lesion at mammography. All of our MR images are interpreted along with a correlating recent mammographicstudy and any other available imaging study.

RG f Volume 27 ● Special Issue Ojeda-Fournier et al S163

TeachingPoint

recommend that a mammographic study per-formed within 6 months be available at the timeof MR image interpretation. In addition, targetedsecond-look US is recommended in cases wherenonspecific lesions are encountered. The use ofUS allows more definitive interpretation and canguide biopsy if indicated. It cannot be overem-phasized that breast MR imaging does not replacemammography and that mammographically sus-picious findings are to be primarily addressed re-gardless of MR imaging findings.

ConclusionsBreast MR imaging has evolved into an acceptedand widely used adjunctive tool in evaluation ofthe breast. Numerous indications have emergedfor breast MR imaging, and its use continues togrow. However, mammography remains the pri-mary imaging modality for evaluation of thebreast. Recognizing the potential artifacts andpitfalls associated with breast MR imaging is nec-essary for appropriate and accurate interpretationof the results of this important imaging tool.

References1. Arena L, Morehouse HT, Safir J. MR imaging ar-

tifacts that simulate disease: how to recognize andeliminate them. RadioGraphics 1995;15(6):1373–1394.

2. Pusey E, Lufkin RB, Brown RK, et al. Magneticresonance imaging artifacts: mechanism and clini-cal significance. RadioGraphics 1986;6(5):891–911.

3. Zhuo J, Gullapalli RP. MR artifacts, safety, andquality control. RadioGraphics 2006;26(1):275–297.

4. Mirowitz SA. Diagnostic pitfalls and artifacts inabdominal MR imaging: a review. Radiology1998;208(3):577–589.

5. Herrick RC, Hayman LA, Taber KH, Diaz-Marchan PJ, Kuo MD. Artifacts and pitfalls inMR imaging of the orbit: a clinical review. Radio-Graphics 1997;17(3):707–724.

6. Taber KH, Herrick RC, Weathers SW, Kumar AJ,Schomer DF, Hayman LA. Pitfalls and artifactsencountered in clinical MR imaging of the spine.RadioGraphics 1998;18(6):1499–1521.

7. Coulthard A, Potterton AJ. Pitfalls of breast MRI.Br J Radiol 2000;73(870):665–671.

8. American College of Radiology. ACR practiceguidelines for the performance of magnetic reso-nance imaging (MRI) of the breast. Reston, Va:American College of Radiology, 2004; 419–424.

9. Radiological Society of North America. Searchintensifies for link between new disease and gado-linium-based agents. RSNA News [newsletter]2007;17(2):7–8.

10. Morris EA. Breast cancer imaging with MRI. Ra-diol Clin North Am 2002;40(3):443–466.

11. Friedman PD, Swaminathan SV, Herman K, Kal-isher L. Breast MRI: the importance of bilateralimaging. AJR Am J Roentgenol 2006;187(2):345–349.

12. Rausch DR, Hendrick RE. How to optimize clini-cal breast MR imaging practices and techniqueson your 1.5-T system. RadioGraphics 2006;26(5):1469–1484.

13. Smith R, Lange R. Understanding magnetic reso-nance imaging. Boca Raton, Fla: CRC Press,1998.

14. Shellock FG, Kanal E. Aneurysm clips: evaluationof MR imaging artifacts at 1.5 T. Radiology 1998;209(2):563–566.

15. Devon RK, Rosen MA, Mies C, Orel SG. Breastreconstruction with a transverse rectus abdominismyocutaneous flap: spectrum of normal and ab-normal MR imaging findings. RadioGraphics2004;24(5):1287–1299.

16. Delille JP, Slanetz PJ, Yeh ED, Kopans DB, Gar-rido L. Physiologic changes in breast magneticresonance imaging during the menstrual cycle:perfusion imaging, signal enhancement, and influ-ence of the T1 relaxation time of breast tissue.Breast J 2005;11(4):236–241.

17. Friedman EP, Hall-Craggs MA, Mumtaz H,Schneidau A. Breast MR and the appearance ofthe normal and abnormal nipple. Clin Radiol1997;52(11):854–861.

18. Erguvan-Dogan B, Whitman GJ, Kushwaha AC,Phelps MJ, Dempsey PJ. BI-RADS-MRI: aprimer. AJR Am J Roentgenol 2006;187(2):W152–W160.

19. Schnall MD, Rosten S, Englander S, Orel SG,Nunes LW. A combined architectural and kineticinterpretation model for breast MR images. AcadRadiol 2001;8(7):591–597.

20. Mortele KJ, Ros PR. Cystic focal liver lesions inthe adult: differential CT and MR imaging fea-tures. RadioGraphics 2001;21(4):895–910.

21. Sharma A, Fidias P, Hayman LA, Loomis SL,Taber KH, Aquino SL. Patterns of lymphad-enopathy in thoracic malignancies. RadioGraphics2004;24(2):419–434.

22. Kinoshita T, Odagiri K, Andoh K, et al. Evalua-tion of small internal mammary lymph node me-tastases in breast cancer by MRI. Radiat Med1999;17(3):189–193.

23. Kuhl CK, Mielcareck P, Klaschik S, et al. Dy-namic breast MR imaging: are signal intensitytime course data useful for differential diagnosisof enhancing lesions? Radiology 1999;211(1):101–110.

This article meets the criteria for 1.0 AMA PRA Category 1 Credit TM. To obtain credit, see www.rsna.org/education/rg_cme.html.

S164 October 2007 RG f Volume 27 ● Special Issue

RG Volume 27 • Special Issue • October 2007 Ojeda-Fournier et al

Recognizing and Interpreting Artifacts and Pitfalls in MR Imaging of the Breast Haydee Ojeda-Fournier, MD, et al

Page S148 Breast MR imaging does not replace mammography. MR imaging may be helpful in selected cases where there are unanswered questions after a complete clinical, mammographic, and sonographic evaluation. Page S149 At a minimum, a field strength of 1.5 T, a dedicated breast coil, high resolution and thin section thickness, and gadolinium enhancement are required for MR imaging of the breast. Page S149 A significant artifact in MR imaging is motion (Fig 1). Even with an optimal prescribed protocol, any amount of motion can degrade image quality or even render a study completely nondiagnostic. Page S155 Absence of contrast enhancement can be difficult to detect, particularly with inherent breast tissue contrast present. Page S163 Interpretation of breast MR images requires simultaneous evaluation with mammography and other available imaging studies.

RadioGraphics 2007; 27:S147–S164 ● Published online 10.1148/rg.27si075516 ● Content Codes:

RadioGraphics 2007 This is your reprint order form or pro forma invoice

(Please keep a copy of this document for your records.)

Author Name _______________________________________________________________________________________________ Title of Article _______________________________________________________________________________________________ Issue of Journal_______________________________ Reprint # _____________ Publication Date ________________ Number of Pages_______________________________ KB # _____________ Symbol RadioGraphics Color in Article? Yes / No (Please Circle) Please include the journal name and reprint number or manuscript number on your purchase order or other correspondence. Order and Shipping Information Reprint Costs (Please see page 2 of 2 for reprint costs/fees.) ________ Number of reprints ordered $_________

________ Number of color reprints ordered $_________

________ Number of covers ordered $_________

Subtotal $_________

Taxes $_________ (Add appropriate sales tax for Virginia, Maryland, Pennsylvania, and the District of Columbia or Canadian GST to the reprints if your order is to be shipped to these locations.)

First address included, add $32 for each additional shipping address $_________

TOTAL $_________

Shipping Address (cannot ship to a P.O. Box) Please Print Clearly Name ___________________________________________ Institution _________________________________________ Street ___________________________________________ City ____________________ State _____ Zip ___________ Country ___________________________________________ Quantity___________________ Fax ___________________ Phone: Day _________________ Evening _______________ E-mail Address _____________________________________ Additional Shipping Address* (cannot ship to a P.O. Box)

Name ___________________________________________ Institution _________________________________________ Street ___________________________________________ City ________________ State ______ Zip ___________

Country _________________________________________ Quantity __________________ Fax __________________ Phone: Day ________________ Evening ______________ E-mail Address ____________________________________ * Add $32 for each additional shipping address

Payment and Credit Card Details Enclosed: Personal Check ___________ Credit Card Payment Details _________ Checks must be paid in U.S. dollars and drawn on a U.S. Bank. Credit Card: __ VISA __ Am. Exp. __ MasterCard Card Number __________________________________ Expiration Date_________________________________ Signature: _____________________________________ Please send your order form and prepayment made payable to: Cadmus Reprints P.O. Box 751903 Charlotte, NC 28275-1903 Note: Do not send express packages to this location, PO Box.

FEIN #:541274108

Invoice or Credit Card Information Invoice Address Please Print Clearly Please complete Invoice address as it appears on credit card statement Name ____________________________________________ Institution ________________________________________ Department _______________________________________ Street ____________________________________________ City ________________________ State _____ Zip _______ Country ___________________________________________ Phone _____________________ Fax _________________ E-mail Address _____________________________________ Cadmus will process credit cards and Cadmus Journal

Services will appear on the credit card statement. If you don’t mail your order form, you may fax it to 410-820-9765 with

your credit card information. Signature __________________________________________ Date _______________________________________ Signature is required. By signing this form, the author agrees to accept the responsibility for the payment of reprints and/or all charges described in this document.

Reprint order forms and purchase orders or prepayments must be received 72 hours after receipt of form either by mail or by fax at 410-820-9765. It is the policy of Cadmus Reprints to issue one invoice per order.

Please print clearly.

Page 1 of 2 RB-9/22/06

RadioGraphics 2007 Black and White Reprint Prices

Domestic (USA only) # of

Pages 50 100 200 300 400 500

1-4 $213 $228 $260 $278 $295 $313 5-8 $338 $373 $420 $453 $495 $530

9-12 $450 $500 $575 $635 $693 $755 13-16 $555 $623 $728 $805 $888 $965 17-20 $673 $753 $883 $990 $1,085 $1,185 21-24 $785 $880 $1,040 $1,165 $1,285 $1,413 25-28 $895 $1,010 $1,208 $1,350 $1,498 $1,638 29-32 $1,008 $1,143 $1,363 $1,525 $1,698 $1,865

Covers $95 $118 $218 $320 $428 $530

International (includes Canada and Mexico) # of

Pages 50 100 200 300 400 500

1-4 $263 $275 $330 $385 $430 $485 5-8 $415 $443 $555 $650 $753 $850

9-12 $563 $608 $773 $930 $1,070 $1,228 13-16 $698 $760 $988 $1,185 $1,388 $1,585 17-20 $848 $925 $1,203 $1,463 $1,705 $1,950 21-24 $985 $1,080 $1,420 $1,725 $2,025 $2,325 25-28 $1,135 $1,248 $1,640 $1,990 $2,350 $2,698 29-32 $1,273 $1,403 $1,863 $2,265 $2,673 $3,075

Covers $148 $168 $308 $463 $615 $768 Minimum order is 50 copies. For orders larger than 500 copies, please consult Cadmus Reprints at 800-407-9190. Reprint Cover Cover prices are listed above. The cover will include the publication title, article title, and author name in black. Shipping Shipping costs are included in the reprint prices. Domestic orders are shipped via UPS Ground service. Foreign orders are shipped via a proof of delivery air service. Multiple Shipments Orders can be shipped to more than one location. Please be aware that it will cost $32 for each additional location. Delivery Your order will be shipped within 2 weeks of the journal print date. Allow extra time for delivery.

Color Reprint Prices

Domestic (USA only) # of

Pages 50 100 200 300 400 500

1-4 $218 $233 $343 $460 $579 $697 5-8 $343 $388 $584 $825 $1,069 $1,311 9-12 $471 $503 $828 $1,196 $1,563 $1,935 13-16 $601 $633 $1,073 $1,562 $2,058 $2,547 17-20 $738 $767 $1,319 $1,940 $2,550 $3,164 21-24 $872 $899 $1,564 $2,308 $3,045 $3,790 25-28 $1,004 $1,035 $1,820 $2,678 $3,545 $4,403 29-32 $1,140 $1,173 $2,063 $3,048 $4,040 $5,028

Covers $95 $118 $218 $320 $428 $530

International (includes Canada and Mexico)) # of

Pages 50 100 200 300 400 500

1-4 $268 $280 $412 $568 $715 $871 5-8 $419 $457 $720 $1,022 $1,328 $1,633 9-12 $583 $610 $1,025 $1,492 $1,941 $2,407 13-16 $742 $770 $1,333 $1,943 $2,556 $3,167 17-20 $913 $941 $1,641 $2,412 $3,169 $3,929 21-24 $1,072 $1,100 $1,946 $2,867 $3,785 $4,703 25-28 $1,246 $1,274 $2,254 $3,318 $4,398 $5,463 29-32 $1,405 $1,433 $2,561 $3,788 $5,014 $6,237

Covers $148 $168 $308 $463 $615 $768 Tax Due Residents of Virginia, Maryland, Pennsylvania, and the District of Columbia are required to add the appropriate sales tax to each reprint order. For orders shipped to Canada, please add 7% Canadian GST unless exemption is claimed. Ordering Reprint order forms and purchase order or prepayment is required to process your order. Please reference journal name and reprint number or manuscript number on any correspondence. You may use the reverse side of this form as a proforma invoice. Please return your order form and prepayment to: Cadmus Reprints P.O. Box 751903 Charlotte, NC 28275-1903 Note: Do not send express packages to this location, PO Box. FEIN #:541274108 Please direct all inquiries to:

Rose A. Baynard 800-407-9190 (toll free number) 410-819-3966 (direct number) 410-820-9765 (FAX number)

baynardr@cadmus.com (e-mail)

Reprint Order Forms and purchase order or prepayments must be received 72 hours after receipt of form.

Page 2 of 2