G Model EURR-6627; No. of Pages 7

ARTICLE IN PRESS European Journal of Radiology xxx (2013) xxx–xxx

Contents lists available at ScienceDirect

European Journal of Radiology journal homepage: www.elsevier.com/locate/ejrad

Magnetic resonance imaging features of papillary breast lesions Ozgur Sarica ∗ , Fatih Uluc, Deniz Tasmali Taksim Education and Research Hospital, Department of Radiology, Istanbul, Turkey

a r t i c l e

i n f o

Article history: Received 3 November 2012 Received in revised form 27 November 2013 Accepted 9 December 2013 Keywords: Papilloma Papillary carcinoma Breast imaging Ductal findings Magnetic resonance

a b s t r a c t Purpose: This study was aimed to assess the role of magnetic resonance imaging (MRI) in the evaluation of the papillary lesions of the breast and their morphological relationship with the mammary ducts. The potential diagnostic contributory role of ductal oriented protocols to conventional dynamic magnetic resonance examination was also explored. Materials and methods: Retrospective data were collected from 46 patients who had been diagnosed with papillary breast lesions and undergone magnetic resonance examination. The presence of dilated ducts and their morphological relation with the lesion were recorded. Lesions were classified as follows: papilloma, papillomatosis and malignant papillary lesion. Statistical difference between groups was studied for each morphological and dynamic lesion characteristic. Results: Dilated ducts and characteristics of intraductal material can be identified by magnetic resonance imaging. Certain MRI findings such as a mass with crescentic peripheral fluid or focal intraductal mass on T2 weighted images may suggest the presence of an intraductal/papillary lesion. In this respect, non-fatsat T2 weighted images appear particularly useful. There was a significant difference between papilloma and papillomatosis with regard to segmental and heterogeneous contrast enhancement (p < 0.05 for both comparisons). In addition, there was a significant difference between papillomas and carcinomas with regard to homogenous, heterogeneous and segmental contrast enhancement (p < 0.05 for all). On the other hand, papillomatosis and carcinoma did not differ significantly in terms of any of the morphological or dynamical MR criteria compared. Conclusion: Papillary lesions can be detected by MRI. Despite some overlaps in MRI findings between carcinoma, papilloma and papillomatosis, MRI may help differentiate these lesions. Major benefit of retroareolar imaging appears to arise from its ability to demonstrate ductal relation and extension of contrast enhanced regions. © 2013 Elsevier Ireland Ltd. All rights reserved.

1. Introduction Papillary lesions of the breast may pose several diagnostic challenges due to diversity of radiological imaging findings, their small size, non-palpability, multi-site occurrence, and absence of nipple discharge. Symptoms of nipple discharge and palpable mass have been reported to be absent in 60.4% and 38.5% of papilloma patients, respectively [1,2]. Hence, detection of intraductal papillomas mainly depends on imaging findings. The standard diagnostic work-up for papilloma includes mammography, ultrasonography and galactography, while some radiologists also perform directed sonography for the retro-areolar region.

∗ Corresponding author at: Prof. Hifzi Ozcan cd. Adalet Sitesi A Blok Daire 13, Kucukbakkalkoy, Istanbul, Turkey. Tel.: +90 5337359181. E-mail addresses: [email protected] (O. Sarica), [email protected] (F. Uluc), [email protected] (D. Tasmali).

Breast magnetic resonance imaging (MRI) demonstrates a high sensitivity (between 40% and 100%) for noninvasive breast disease [3], while sensitivity may reach up to 87% for high grade DCIS [4]. These figures provide evidence for the potentially high sensitivity of MR imaging in the detection of intraductal disease. In the present study, the role of magnetic resonance imaging in the evaluation of the papillary lesions of the breast and their morphological relationship with the mammary ducts have been examined, as well as the potential diagnostic contribution of ductaloriented MRI to conventional dynamic MRI examination in these conditions. 2. Materials and methods Retrospective data on a total of 51 patients who had a suspicious papillary lesion and underwent an MRI examination were collected. Of the five cases excluded from our study, four had other medical conditions such as cardiac pathology that precluded surgery and they were managed by follow-up only. The fifth patient did not attend to follow-up visits despite the cytological documentation

0720-048X/$ – see front matter © 2013 Elsevier Ireland Ltd. All rights reserved. http://dx.doi.org/10.1016/j.ejrad.2013.12.007

Please cite this article in press as: Sarica O, et al. Magnetic resonance imaging features of papillary breast lesions. Eur J Radiol (2013), http://dx.doi.org/10.1016/j.ejrad.2013.12.007

G Model EURR-6627; No. of Pages 7

ARTICLE IN PRESS O. Sarica et al. / European Journal of Radiology xxx (2013) xxx–xxx

2

Table 1 Distribution of ductal signs among the groups (three cases that had DCIS and invasive carcinoma were excluded from the papillomatosis group).

Discharge Dilated duct and precont low t1 signal Dilated duct and precont high t1 signal Dilated duct and intraductal focal mass-like T2 signal Solid mass with crescentic peripheral fluid Contrast enhanced area related with dilated duct

Papilloma (n = 22)

Papillomatosis (n = 10)

Papillary carcinoma–papillary lesion + DCIS (n14)

8 (36%) 1 (5%) 5 (22%) 1 (5%) 1 (5%) 5 (23%)

3 (30%) 0 (0%) 1 (10%) 2 (20%) 2 (20%) 2 (20%)

3 (21%) 0 (0%) 1 (7%) 0 (0%) 2 (14%) 1 (7%)

of papillary cells in nipple discharge specimens. Thus, a total of 46 patients with papillary lesions who underwent MRI examination were included in the study. Reasons for referral included high risk of malignancy, palpable abnormality, mammographic or sonographic abnormality, and nipple discharge. Patients with additional benign or malignant entities that were identified pathologically were not excluded from the study. MRI was performed using a 1.5-T magnet (Magnetom Avanto, Siemens Medical Solution, Erlangen, Germany) with a dedicated bilateral breast surface coil with prone position. The imaging protocol and parameters were as follows: axial T1-weighted image (TR/TE, 450/11) and non–fat-suppressed T2-weighted turbo spinecho image (3740/122) of both breasts were obtained with 3 mm slice thickness. Next, T1-weighted dynamic images were acquired using a 3D FLASH (fast low-angle shot pulse sequence) without fat suppression through both breasts (TR/TE 5.19/2.38 flip angle 10). T2-weighted short-tau inversion recovery (STIR) sequences were obtained in the coronal plane (TR/TE 5900/78; and IR 130 slice thickness 4.5 mm; FOV 40 × 40 cm). Also non-fat-suppressed T2-weighted turbo spin-echo images were obtained, with 30–35 sections and a full 512 × 512 imaging matrix. The contrast-enhanced dynamic series consisted of 60 sections, each T2 weighted image with 3 mm slice thickness being compensated with two dynamic images with a slice thickness of 1.5 mm without gap. Contrast-enhanced dynamic series exactly matched the T2-weighted images to allow an accurate comparison of signal intensity. First T1-weighted 3D FLASH sequence was acquired before administration of contrast material in the axial planes. Sequential postcontrast multisection whole-breast images were obtained five times in the same plane with 20-s intervals.

Unilateral retroareolar examination was performed on the breast when a lesion had previously been determined by other imaging modalities or the lesion had been associated with clinical complaints. For unilateral retroareolar imaging, pre- and postcontrast coronal fat saturated 2D FLASH (TR/TE 479/5.83 FOV 150 × 150 384 × 230 matrix 3.5 mm slice thickness) and T2 TSE 3D sequence with same FOV and matrix (TR/TE 750/147slice thickness 1 mm) were used. Intravenous gadolinium-gadobenate dimeglumine (DTPA) (MultiHance, Bracco, Milan, Italy) was administered at a rate of 2 ml/s via an automatic injector at a dose of 0.1–0.2 mmol/l per kilogram of body weight, together with a 10-ml bolus of saline solution. All MRI examinations were interpreted by radiologists with special expertise on breast MRI for more than 5 years. However, radiologists were not blinded to the patient data. The largest diameter of the tumor measured on a subtracted image from the series obtained 2 min after contrast injection was recorded as the size of the papilloma. The presence of dilated ducts and their relation with the lesion were also recorded (Table 1). The nomenclature in the American College of Radiology’s Breast Imaging Reporting and Data System (BI-RADS) was used for describing the morphologic characteristics of the lesions [5] (Table 2). For analysis of enhancement kinetics, time–intensity curves were plotted from the signal intensity values obtained from the most enhanced part of the lesion. Time–intensity profile was also classified into the following 3 types according to phase analysis of peak enhancement and washout of contrast enhancement: (a) in the washout type, peak enhancement was seen within 120 s (early phase) and followed by a decrease; (b) in the plateau type, enhancement increased in the early phase and followed by a plateau; and

Table 2 Evaluation of morphologic criteria of groups.

Shape Round Lobulated Irregular Border Well defined Microlobulated Indistinct Contrast pattern Homogeneous Heterogeneous Rim enhancement Nonmass-like Ductal Segmental Regional Multiple mass-like Clustered ring enhancement Focal nodulary-punctate

Papilloma (n = 22)

Papillomatosis (n = 10)

Papillary carcinoma–papillary lesion + DCIS (n14)

3 (13%) 13 (59%) 0 (0%)

2 (20%) 1 (10%) 1 (10%)

2 (14%) 5 (36%) 3 (21%)

12 (54%) 1 (5%) 3 (13%)

3 (30%) 1 (10%) 1 (10%)

9 (64%) 0 (0%) 1 (7%)

15 (68%)+ 1 (5%)*,+ 1 (5%)

2 (20%) 2 (20%)* 1 (10%)

1 (7%)+ 7 (50%)+ 3 (21%)

5 (23%) 0 (0%)*,+ 0 (0%) 2 (9%) 0 (0%) 0 (0%)

1 (10%) 3 (30%)* 0 (0%) 2 (20%) 1 (10%) 1 (10%)

6 (42%) 4 (29%)+ 1 (7%) 0 (0%) 0 (0%) 0 (0%)

The two groups with the sign * – * have statistically significant differences (p < 0.05). The two groups with the sign + – + have statistically significant differences (p < 0.05).

Please cite this article in press as: Sarica O, et al. Magnetic resonance imaging features of papillary breast lesions. Eur J Radiol (2013), http://dx.doi.org/10.1016/j.ejrad.2013.12.007

G Model EURR-6627; No. of Pages 7

ARTICLE IN PRESS O. Sarica et al. / European Journal of Radiology xxx (2013) xxx–xxx

3

Table 3 Evaluation of dynamic criteria and contrast enhancement pattern between groups. Dynamic criteria did not differ significantly between the groups.

Type 1 persistent Type 2 plateau Type 3 washout Nonmass-like enhancement Mass + nonmass-like enhancement

Papilloma (n = 22)

Papillomatosis (n = 10)

Papillary carcinoma–papillary lesion + DCIS (n14)

7 (32%) 6 (27%) 2 (9%) 5 (23%) 3 (14%)

2 (20%) 1 (10%) 1 (10%) 6 (60%) 3 (30%)

3 (21%) 1 (7%) 5 (36%) 7 (50%) 3 (21%)

(c) in the persistent type, there was no peak with time throughout the examination [6] (Table 3). Patients with ACR BIRADS Category 4 and higher – as visualized by conventional diagnostic modalities – or MRI findings such as contrast enhanced lesions with irregular or ill-defined margins or segmental and asymmetrical distribution or type 2–3 time–intensity curve underwent biopsy due to strong suspicion of malignancy. All patients who had clinical or radiological suspicion a papillary lesion underwent excisional biopsy based on the well-established risk of malignancy upgrade. The patients without such a suspicion underwent 14 G tru-cut biopsy. Histologically, lesions were classified as papilloma, papillomatosis or malignant papillary lesion. Statistical differences between groups were examined for the morphological and dynamic MR BIRADS parameters shown in Tables 2 and 3. Fisher’s exact test and Yates continuity correction were used for statistical analysis. A p-value less than 0.05 was regarded as the indication of statistical significance. 3. Results All patients were female and between 24 and 75 years of age (mean age, 47.9 years). Fourteen of 46 patients (30.4%) had unilateral discharge. Four patients underwent tru-cut biopsy. The remaining 42 patients had a pathological diagnosis documented by excisional biopsy or surgical specimen examination. Two patients diagnosed by tru-cut biopsy did not attend follow-up examinations. In one case, there was no increase in the size of the lesion during 18 months follow up. In the remaining patient who had had unilateral mastectomy, the size of the papilloma on the contralateral breast remained stable during four years follow up. Thirteen patients had papillomatosis. Seven of these had atypical ductal hyperplasia (ADH), one had invasive tumor and the other two of had ductal carcinoma in situ (DCIS). Twenty-two patients had a pathological diagnosis of papilloma, with 6 having concomitant ADH. Eleven patients had papillary carcinoma. Positive signs were detected by MRI in 42 of these 46 lesions. Among 4 cases with a negative MRI, one had papillomatosis, and 3 had papillomas.

Retroareolar MRI examination was performed in 32 patients. In four of the 7 cases with a negative retroareolar examination, lesions were located peripherally (three papillomatosis and one papillary carcinoma). Central lesions in the remaining 3 patients were not visualized by the retroareolar MRI as well as the conventional MRI sequences. MRI indicated the presence of a mass lesion in 31 of 46 patients. The lesion diameter was less than 10 mm in 14 of 31 lesions, while 9 had a diameter between 10 and 20 mm, and 8 had a diameter greater than 20 mm. Of the 14 lesions with a diameter below 10 mm, two cases had papillomatosis and papillary carcinoma, one for each histological type. The diameters of the remaining 12 papillomas were below 10 mm, in 8 (67%) being less than 7 mm. The size of the two smallest lesions detected by MRI was 5 mm × 3 mm. One showed ductal contrast enhancement with an intraductal focal mass on T2 weighed images and other was an intracystic mass with wall irregularity. Major findings in T1 weighted images included duct dilation and high signal intensity, duct dilation and low signal intensity and definable mass lesion. Additional findings in T1W images were isointense mass that could be identified through bulging on parenchymal contours, increase in focal ductal caliber, asymmetrical parenchymal focus, and mass with hemorrhagic areas or parenchymal distortion (Fig. 1). On the other hand, major findings in T2 weighted images were ductal dilatation (isolated dilated duct in subareolar region, diffuse ductal dilatation extending to one or more quadrants periphally, ductal filling defect; Fig. 2), mass lesion (solid mass, solid-cystic mass, intracystic solid mass, solid mass accompanied with peripheral crescentic shaped fluid collection; Fig. 3). Other findings in T2W images were parenchymal distortion or multiple millimetric cystic clusters. Contrast enhancement was present in 39 of 46 patients. In four of 7 patients with no contrast enhancement, there were no demonstrable lesions on MRI. Two of the remaining 3 patients had multiple millimetric cystic clusters. There was no contrast enhancement on MRI in both lesions but intracystic solidified areas were seen on sonography (Fig. 4). The other patient had a dilated duct that had high signal intensity on T1W and T2W. A 6 × 4 mm mass was also visualized by sonography and galactography. One case of papillomatosis co-occurring with ADH could not be detected by

Fig. 1. Filling defect on galactography (a), dilated duct and intraluminal echogenic mass on sonogram (b), high signal intensity on T1W image due to hemorrhagic material filling the duct, but no definable mass (c) in papilloma case.

Please cite this article in press as: Sarica O, et al. Magnetic resonance imaging features of papillary breast lesions. Eur J Radiol (2013), http://dx.doi.org/10.1016/j.ejrad.2013.12.007

G Model EURR-6627; No. of Pages 7 4

ARTICLE IN PRESS O. Sarica et al. / European Journal of Radiology xxx (2013) xxx–xxx

Fig. 2. Intraductal echogenic mass on sonogram (a) in papilloma case; high resolution T2W image clearly shows intraductal location of the mass (b), on duct oriented imaging the mass and its ductal relation also definitely are visible on contrast enhanced 2D T1W (c) and 3D volumetric T2W image (d).

Fig. 3. Intraductal papilloma patient with peripheral crescentic fluid collection on sonogram (a), high resolution T2W (b), contrast enhanced 2D T1W (c) and 3D volumetric T2W image (d).

groups with regard to homogenous, heterogeneous and segmental contrast enhancement (p < 0.05) Carcinomas showed segmental and heterogeneous contrast enhancement, whereas papillomas showed homogeneous enhancement. There were no significant differences between the groups with respect to dynamic criteria. Comparison of papillomatosis and carcinoma revealed no significant differences in any of the morphological or dynamic MRI criteria (Fig. 6).

Fig. 4. In a papillomatosis case, cluster of cysts and some cysts had 2–3 mm echogenic solid component which visible on sonogram (a) MRI only showed the cysts. (b) No any contrast enhanced solid area in spite of fluid contrast of the cysts (c).

MRI, but showed parenchymal distortion and microcalcifications in mammography, in addition to avascular tubular echogeneity (considered to be intraductal) in ultrasound. As expected, in cases with papillomatosis, non-mass-like contrast enhancement was more marked than in cases with papilloma (p = 0.056). Statistically significant differences were seen with regard to segmental and heterogeneous contrast enhancement according to BIRADS parameters (p < 0.05). There were no significant differences between the groups with regard to dynamic criteria (Fig. 5). Comparison of papillomas and carcinomas in terms of irregular contour parameter, i.e. a morphological BIRADS criteria, yielded a p-value of 0.051. There was a significant difference between the

4. Discussion Currently, differentiation between benign and malignant papillary lesions is not possible with conventional radiological methods as well as with MRI, although MRI has shown some potential for detection of papillary lesions. In this study, more than 90% of papillary lesions had positive signs on MRI. Several studies reported varying MRI-occult papilloma rates. In a study by Son EJ et al. the sensitivity of MRI to detect papillomas was 100%, with no cases having MRI-occult papillomas [7]. On the other hand, occult cases of papilloma comprised approximately 25% of the subjects in Daniel et al.’s study [8], and in the series reported by Rovno et al. [9] the diagnosis of papilloma could not be made by MRI but instead could only be established with excisional biopsy in 16 (69%) of the patients. However, MRI devices with lower resolution capacity were used in these studies as compared to our imaging equipment. The proportion of MRI-occult papillomas in our group was 14% (3/22), and in two of these, the lesion could only

Fig. 5. Different sonographic appearance of a patient with papilloma case (a,b) and a papillomatosis case with DCIS (c,d) both showed ductal contrast enhancement pattern on MRI.

Please cite this article in press as: Sarica O, et al. Magnetic resonance imaging features of papillary breast lesions. Eur J Radiol (2013), http://dx.doi.org/10.1016/j.ejrad.2013.12.007

G Model EURR-6627; No. of Pages 7

ARTICLE IN PRESS O. Sarica et al. / European Journal of Radiology xxx (2013) xxx–xxx

5

Fig. 6. Segmental distribution of disease in two different patients with papillomatosis (sonogram a, contrast enhanced MRI b) and papillary carcinoma case (sonogram c, STIR T2W MRI d).

be identified by ductoscopy but not by other imaging modalities. The third case had a solid mass (5 mm × 4 mm on ultrasound) with cystic components. Of the 32 papilloma patients reported by Tominaga et al., 4 (12.5%) were MRI-occult [10], and only dilated ducts could be visualized by MRI. However, small papillomas, 1–2 mm in size, were found on pathological specimens. In this study, two of the three MRI-occult papillomas with pathological size of less than 2 mm were associated with nipple discharge and could not be detected by other imaging modalities as well. In Rovno et al.’s study, the high rate of MRI-occult lesions was attributed to smaller size of the lesions and absence of nipple discharge. In contrast, in our study two cases with small papillomas had nipple discharge. Therefore, it might be assumed that the percentage of MRI-occult lesions may correlate with the size of the lesions. In our view, the current resolution level of MRI protocols may not be adequate to visualize small-sized lesions, as exemplified the detection of cysts only in two cases of papillomatosis in whom the lesions were located within the cysts (detected by ultrasound) and had a diameter of approximately 2–3 mm. Despite the fluid contrast, MRI was unable to detect these small lesions. MR galactography utilizing special microscopic coils can detect ducts as small as 0.8 mm in diameter and intraductal lesions as small as 1.0 mm in size [11]. An intraductal radiofrequency microcoil developed by Dr. Strick can provide microscale images [12]. Results of these two studies suggest that smaller lesions can be detected when resolution of MRI is increased. Occasionally papillomas have spiculated borders on MRI [9,13]. In our study, there were no papillomas with spiculated borders. Bhattarai et al. found type 3 enhancement curve in more than 70% of papillomas with the use of a microcoil [13]. In our series, 2 out of 13 (15%) of the contrast enhanced mass-like papillomas showed type 3 enhancement curves. Unfortunately, two thirds of papilloma cases in our study had one or more morphological or dynamic findings that suggested a suspicion of malignancy. When cases with benign clinical criteria (e.g. age below 40 years, lesion size less than 10 mm) are excluded, only 13% of our papillomas were categorized as BIRADS 4 or more, using MRI BIRADS lexicon descriptor. Daniel et al. [8] described the two following image characteristics for papillomas: “Small lumenal mass” papillomas (27%) were small, smooth enhancing masses connected to the nipple by an enlarged duct. “Tumor-like” papillomas were more common than “small lumenal mass papillomas” (54%) and could not be distinguished from invasive tumors. In 6 of the 22 papilloma cases (27%) in our study, the mass or enhanced area was intraductal or there was ductal continuity between the lesion and the nipple. In our 12 patients (54%), the mass had no relation with the ducts. However, in four of these patients, the lesions could only be partly differentiated from the remaining 8 solid masses In one case, signal characteristics suggested an intracystic mass, while in the other three there were solid-cystic components on T2 weighed images. In our view, non-fatsat T2 weighed images appear to be useful in the characterization of masses that have no ductal components. In non-fatsat images, intracyctic or small sized solid-cystic masses may indicate

papillary nature of lesions. On the other hand, Daniel et al. used fat saturated T2 weighed images. Thus, it should be borne in mind that fat saturated images do not reflect the internal characteristics of the lesions. The explanation proposed by Kramer et al. for the inability of MRI to detect most of the papillomas is the lack of contrast medium uptake [14]. These authors concluded that filling defects in galactography that do not uptake contrast medium on MRI could help exclude the possibility of malignancy with a high degree of certainty. However, one of our papilloma cases with ADH had hemorrhagic duct with no contrast enhancement on MRI and there were filling defects at galactography. Thus, we believe that intraductal imaging modalities such as galactography or ductoscopy should always be considered in the presence of hemorrhagic discharge, even with a negative MRI result. Histopathological diagnosis of papillary lesions requires excision of the complete specimen, since differentiation between benign and malignant conditions can only be possible through identification of the myoepithelial cell layer of the basement membrane by immunohistochemistry [15]. Although a clear differentiation between malignant and benign papillary lesions may not be possible using MRI, we believe that MRI can still be useful in identifying the pathological nature of the lesion. A solitary papilloma is a single discrete papillary tumor in one duct. Multiple papillomas usually occur in contiguous branches of the ductal system. In previous reports, the importance of making a distinction between intraductal papilloma and papillomatosis (microscopic foci of intraductal hyperplasia, which have papillary architecture without atypia) has been emphasized [16]. Although 44% of the subjects with papillomatosis showed segmental contrast enhancement in this study, there was no segmental contrast enhancement among papillomas. In the light of this observation, segmental contrast enhancement pattern appears to be a useful criterion for the differentiation between papillomas and papillomatosis. On the other hand, some difficulties can be experienced in the differentiation of these two entities using ductal contrast enhancement pattern. In our view, ultrasound examination may be particularly useful in these cases with ductal contrast enhancement due to markedly different sonographic appearance. In the present study, all of the cases with papilloma with non-mass like contrast enhancement pattern, an intraductal or ductal-extending focal mass could be detected using ultrasound. Most of cases with nonmass like contrast enhancing papillomatosis showed non-mass like heterogeneous hypoechoic areas. Only two cases had an intraductal mass or mass with ductal continuity in the papillomatosis group, and in these patients, multiple millimetric masses were found in sonography. The non-mass like contrast enhancement in MRI of these lesions exhibited multiple mass-like nodular lesions rather than ductal enhancement. In our study, when carcinomas and DCIS developing on a papillomatosis background are excluded, 4 of 11 (36%) papillary carcinomas had non-mass like contrast enhancement and two cases had accompanying mass lesions. In both of these cases, the size

Please cite this article in press as: Sarica O, et al. Magnetic resonance imaging features of papillary breast lesions. Eur J Radiol (2013), http://dx.doi.org/10.1016/j.ejrad.2013.12.007

G Model EURR-6627; No. of Pages 7 6

ARTICLE IN PRESS O. Sarica et al. / European Journal of Radiology xxx (2013) xxx–xxx

of the mass was greater than 20 mm, with morphological features suggesting malignancy. Among the lesions with non-mass like contrast enhancement only, there was one with segmental and regional enhancement and there was one asymmetrical segmental and ductal contrast enhancement. Both cases were referred with palpable masses, and one had hemorrhagic nipple discharge with nipple retraction. On the basis of clinical assessment, MRI images, and other imaging signs, the differentiation between non-mass like carcinoma and papilloma–papillomatosis could be readily made. Most of intraductal papillomas with a diameter less than 5 mm may extend along the duct for several centimeters [17]. This may explain why areas with ductal contrast enhancement in MRI had a greater dimension than the focal mass detected by ultrasound. Eight of the 12 papillomas with a diameter below 10 mm were associated with positive focal retroareolar images. There were no cases where a retroaerolar image with small FOV was undetected by conventional contrast enhanced images. A possible advantage of retroareolar imaging is its ability to demonstrate the ductal relations and extension of contrast enhancement. According to Tominaga et al. [10] the luminal space between the papillary fronds and the duct wall on the other side, despite being intact, is too narrow to be detected by their MRI protocol. In their study fat-sat T2W images were used, whereas non fat-sat high matrix (512 × 512) T2W series with better spatial resolution have been used in our study. In 5 of 29 (17%) mass lesions detected on T2-weighted images, crescentic fluid collection sign was detected that indicates the presence of a luminal space between lesion and duct wall. This finding may indicate that the lesion has intraductal localization. Our literature search did not reveal any studies that mentioned this finding. In a study by Baltzer et al., ductal obstruction could be visualized due to high contrasting of fluids against other tissues in T2-TSE sequences, which are clearly superior to TIRM sequences in terms of spatial resolution [18]. Intracystic papillary lesions greater than 3 cm in diameter are generally considered malignant. It is difficult to distinguish between ICP and ICPC using imaging techniques or cytologic examinations preoperatively [19,20]. Three cases in our study were interpreted as intracystic solid lesions on MR images. The diameter of one of these lesions was below the criteria for malignancy (i.e.

Magnetic resonance imaging features of papillary breast lesions.

This study was aimed to assess the role of magnetic resonance imaging (MRI) in the evaluation of the papillary lesions of the breast and their morphol...
1MB Sizes 0 Downloads 0 Views