| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||
ABSTRACT |
Thursday, May 4, 1:30 PM-3:30 PM
Abstracts 261-272
Moderators: Janet Baum, MD and Gary J. Whitman, MD
1:30 PM
261. Current Use of MRI in a Dedicated Breast Imaging Practice: Indications and Outcome Analysis
Philpotts L.E.; Andrejeva L.*; Fan J.; Hooley R.; Horvath L.; Tocino I.; Department of Diagnostic Radiology, Yale University School of Medicine, New Haven, CT.
Address correspondence to L. Andrejeva (liva.andrejeva{at}yale.edu)
Objective: To examine the indications, interpretations, and outcomes of patients currently undergoing breast MRI in a dedicated breast imaging program.
Materials and Methods: Prospective data was collected over a 9-month period of all breast MRI's performed at our institution. There were 241 breast MRI's performed in 217 patients. The reasons for the exam, MR findings, recommendations, BIRADS assessment, and interpreting physicians were recorded at the time of the MR. All studies were co-read by a breast imaging and MR specialist. The follow up information was obtained from the radiology and pathology records. Outcomes were assessed by determining the need for biopsy/surgery and follow-up exams. Recommendations were compared by reason for exam categories.
Results: The main reasons for performing breast MRI were: equivocal mammogram (n = 70), high risk patient screen (n = 48), extent of disease/known cancer (n = 41), contralateral breast screen (n = 34), follow-up prior MRI (n = 20) or biopsy (n = 15), and other (n = 12). The recommendations were: mammographic follow-up (n = 94), MRI follow-up (n = 57), biopsy (n = 41), and ultrasound (n = 32). The greatest number of biopsy recommendations were in patients with known cancer/extent of disease. The greatest number of follow up MRI exams were in high-risk patients. The BI-RADS assessments in descending order were: 3, 2, 1, 0, 6, 4, 5. The positive biopsy rate was 32%, and was highest for extent of disease and equivocal mammogram patients. Follow-up imaging was recommended in 148 patients, mammography in 94, and MRI or both MR and mammography in 54.
Conclusion: The main reasons for breast MRI currently are for patients with known cancers, high-risk patients, and those with equivocal mammographic findings. The positive biopsy rate is similar to those found from mammographic abnormalities. A large number of follow up exams are required, including MRIs. This data will aid in further studies of cost effectiveness of current practices.
262. What is the True Sensitivity and Specificity of Breast MRI: an MRI/Mastectomy Pathology Correlation Study
Gabriel H.*; Wiley E.; Hendrick E.; Mendelson E.; Wolfman J.A.; Radiology, Northwestern University, Chicago, IL.
Address correspondence to H. Gabriel (hgabriel{at}nmff.org)
Objective: To study the true accuracy of breast MRI by performing direct mastectomy pathology MR imaging correlation on a lesion by lesion basis as well as on a semiquadrant basis to determine sensitivity, specificity, as well as positive and negative predictive values of breast MRI.
Materials and Methods: In an IRB-approved prospective study, women contemplating mastectomy underwent contrast-enhanced breast MR exam and detailed pathologic analysis. The MR lesions were interpreted using standard features (morphology and contrast kinetics) and a BIRADS classification assigned to each. The mastectomy specimens were specially processed and cut into 1 mm thick sagittal slices to approximate the MR images. Each slice underwent detailed pathologic assessment listing all lesions. The radiologist and pathologist then by consensus did a lesion-by-lesion correlation, listing each lesion as having either concordant or discordant detection and interpretation. To determine the specificity, each breast was divided into nine areas and each was assessed as to the presence or absence of lesions on MRI and pathology.
Results: A total of 29 breast in 22 patients underwent both MR imaging and mastectomy resulting in a total of 196 lesions identified. 191 lesions could be correlated with concordant detection (lesion seen on both MR and pathology) in 153 (80%) and discordant detection (lesion seen only on either MRI or pathology) in 38 (20%). Of the discordantly detected lesions, the majority were either benign (20/38, 53%) or lobular neoplasia (12/38, 32%). Only 6/38, (16%) represented malignancy, usually small foci of disease. The sensitivity of MR interpretation for malignancy based on BIRADS assessment was 96% and the specificity for benign lesions was 75%. With respect to absence of disease based on the semiquadrant analysis, there was a specificity of 99% and a NPV for the absence of malignancy of 99%.
Conclusion: Detailed MRI pathology correlation suggests that MRI was a highly accurate modality. MRI was highly specific in excluding disease in areas of the breast that were pathologically negative with a negative predictive value of 99%, which has never before been reported. In addition, breast MRI was highly sensitive for the detection of malignancy (96%). In 20% patients, there was disagreement in detection between the MR and the pathology but the majority of discordant lesions represented benign disease or lobular neoplasia.
263. MRI BI-RADS Morphology and Enhancement Patterns of Ductal Carcinoma In Situ
Smith-Foley S.A.1,2*; Lehman C.D.1,2; DeMartini W.B.1,2; Eby P.R.1,2; Peacock S.1,2; Rosen E.L.1,2; 1. Radiology, University of Washington, Seattle, WA; 2. Breast Imaging, Seattle Cancer Care Alliance, Seattle, WA.
Address correspondence to S.A. Smith-Foley (ssf2{at}seattlecca.org)
Objective: To identify the MR imaging morphology and enhancement patterns of ductal carcinoma in situ (DCIS) using the BI-RADS lexicon and to compare with invasive carcinoma.
Materials and Methods: Following IRB approval, a medical records review identified 242 consecutive MRI exams performed for the clinical indication of known carcinoma between 1/1/03 and 9/12/05. 225 patients with pathologic confirmation of either DCIS alone, invasive carcinoma alone, or mixed invasive and in-situ disease were identified following core needle biopsy but prior to surgery. MRI BI-RADS lesion morphology and enhancement patterns were compared. Fisher's exact test was used to assess differences among the groups.
Results: 41/225 (18%) patients had DCIS alone, 61/225 (27.1%) had invasive carcinoma alone, and 116/225 (54.6%) had mixed invasive/in situ carcinoma. MRI detected 36 of 41 cases (87.8%) of DCIS alone, 59 of 61 cases (96.7%) of invasive carcinoma alone, and 120 of 123 cases (97.6%) of mixed invasive and in-situ carcinoma. For pure DCIS lesions, 25/41(61.0%) exhibited non-mass-like enhancement (NMLE), 6/41(14.6%) appeared mass like, 5/41 (12.2%) were a focus and 5/41 (12.2%) had no finding. For pure invasive carcinomas 48/61(78.7%) were mass like, 9/61 (14.8%) were NMLE, 2/61 (3.3%) were a focus, and 2/61 (3.3%) had no finding. For mixed lesions 93/123 (75.6%) were mass like, 23/123 (18.7%) demonstrated NMLE, 4/123 (3.3%) were a focus, and 3/123 (3.3%) had no finding. Of the pure DCIS cases presenting as NMLE the most common BI-RADS lesion subtypes were segmental and clumped. There was a statistically significant difference between the BI-RADS lesion morphology and enhancement patterns of pure in situ, invasive and mixed invasive/in situ, carcinoma (p < 0.0001).
Conclusion: Although there is overlap in the MRI morphology and enhancement pattern of in situ and invasive breast carcinoma, DCIS more frequently manifests as NMLE compared to invasive carcinoma.
264. Parenchymal Enhancement on Breast MRI May be a Marker for Cancer Risk: Correlation of Parenchymal Enhancement with Mammographic Density.
Arkani-Hamed S.1*; Newstead G.1; Abe H.1; Chen V.1; Crawford C.1; Schmidt R.1; Medved M.1; Olopade O.2; 1. Radiology, University of Chicago, Chicago, IL; 2. Hematology/Oncology, University of Chicago, Chicago, IL.
Address correspondence to S. Arkani-Hamed (sarkani{at}uchicago.edu)
Objective: To determine whether the parenchymal enhancement pattern at MR imaging is a marker for cancer risk by correlating it with mammographic breast density in a cohort of high risk women.
Materials and Methods: The study population consisted of 281 women with either a genetic predisposition, a personal history, or a 5 year relative risk of breast cancer of greater than 1.7 as calculated by the Gail model. All patients were imaged with a 3D bilateral dynamic MR sequence, without consideration of menstrual status. No patient had any abnormal finding. Density on x-ray mammograms was classified independently by three readers according to the BI-RADS categories 1-4. The MR parenchymal pattern of enhancement for each case was classified separately by three readers as minimal, homogeneous, heterogeneous or nodular. This classification was performed by visual analysis of the enhancement pattern on 6 minute post-contrast coronal difference images. Parenchymal signal intensity vs. time curves were generated by manually tracing a region of interest around the total parenchyma visible in a selected precontrast coronal slice. The peak magnitude of parenchymal enhancement relative to the pre-contrast signal intensity E_peak[%] was measured.
Results: Distribution of patients by BI-RADS mammographic density: category 1 = 2% (n = 5), category 2 = 38% (n = 108), category 3 = 46% (n = 129), category 4 = 14% (n = 39). MRI parenchymal enhancement pattern distribution: minimal 45% (n = 127), homogeneous 9% (n = 25), heterogeneous 29% (n = 82), nodular 17% (n = 47). 53% of category 1 and 2 breasts show minimal enhancement while 41% show heterogeneous or nodular patterns. 36 % of category 3 and 4 breasts show minimal enhancement while 58 % show heterogeneous or nodular enhancement. The peak enhancement occurred at the last post contrast minute (6 minute) in 94% of the total population. E_peak was classified as low < 30%, medium 30-100% and high > 100%. The distribution of E_peak by mammographic density: categories 1 and 2:76% low, 23% medium, 1% high; categories 3 and 4: 42% low, 46% medium, 12% high. 72% of women with medium or high E_peak had breast density of category 3 or 4, while 28% were category 1 or 2.
Conclusion: A statistically significant correlation between mammographic density and magnitude (E_peak) and pattern (heterogeneous and nodular) of parenchymal enhancement was found (p < 0.01). Although further study is needed, this observation might lead to an improved method of assessing breast cancer risk based on breast enhancement patterns.
265. Is Breast MRI Helpful in the Evaluation of the Inconclusive Mammogram?
Moy L.1*; Mercado C.1; Kitazono M.2; Toth H.1; 1. Radiology, New York University Cancer Center, New York, NY; 2. Radiology, USC Keck School of Medicine, Los Angeles, CA.
Address correspondence to L. Moy (moyl02{at}med.nyu.edu)
Objective: Rarely, mammography can be inconclusive despite meticulous technique, additional problem-solving views and ultrasound (US). In these cases, the addition of MRI may be helpful. Our objective is to analyze the mammograms to identify the findings for which a MRI is performed. We evaluated our mammograms and MRI to determine the findings that are most frequently associated with an occult cancer.
Materials and Methods: We retrospectively reviewed our radiology database to determine how often the clinical indication for a breast MRI was an abnormal mammogram. We reviewed the mammograms and US to determine the morphologic features, location, and size of the abnormalities. We reviewed the MRI to determine if the corresponding lesion was identified. Radiology and pathology follow up was also performed.
Results: Over 6 years, 115 MRIs were performed for the problematic mammogram. Of these, 43 (37.4%) were for a focal asymmetry seen on two views, most localized in the posterior depths, upper outer quadrants, and not identified on US. Five (4.3%) MRIs were performed for a change in the site of a previous benign biopsy, and 12 (10.4%) for architectural distortion. Fifty-five (47.9%) MRIs were performed for a suspicious lesion seen on one view, 30 (54.5%) were seen on the MLO view only, 32 (58.2%) localized to the superior breast, 8 (14.5%) localized to the medial breast, and none were identified on US. In 70 (60.9%) cases, no mass was seen in the area of the mammographic finding, in 9 (7.8%) a non enhancing mass was seen, and in 36 (31.3%) a regional non-mass like enhancement with benign kinetics was seen that corresponded to the mammographic finding. The latter finding was presumed to be normal fibroglandular breast tissue. Six (5%) cancers were found, 3 asymmetries seen on one view only, 2 focal asymmetry seen on two views, and one with distortion. A targeted US after the MRI was performed identified two of these cancers. All cancers had an irregular shape and margin, with heterogeneous or rim enhancement. The mean size of the cancers was 1.6 cm (range, 1.0-2.9 cm), and they were all invasive ductal carcinomas.
Conclusion: Although MRI should never be used in place of carefully performed mammography and US, it was helpful in determining that 95% of the problematic mammograms could be assessed as BIRADS 2: benign finding. The inconclusive mammogram is a rare indication for MRI, and the MRI exam is a helpful problem-solving tool.
266. Outcomes of BI-RADS 3 MR Lesions
Eby P.R.1,2; Lauro B.N.1; DeMartini W.B.1,2; Peacock S.1,2; Rosen E.L.1,2; Lehman C.D.1,2; 1. Radiology, University of Washington, Seattle, WA; 2. Breast Imaging, Seattle Cancer Care Alliance, Seattle, WA.
Objective: To evaluate the cancer yield, frequency of use and follow-up compliance of the BI-RADS 3 assessment in breast MR imaging.
Materials and Methods: Following IRB approval, we conducted a retrospective audit of 812 consecutive breast MR examinations performed between January, 2003 and June, 2004. A cohort of 160/812 (20%) index examinations which received a BI-RADS 3 assessment was identified. Data for each patient were collected through June, 2005 via review of clinical records and radiology and pathology reports. The clinical indication, compliance with recommendations, time from the index examination to follow-up imaging, assessment at follow-up examination and pathology findings were recorded. Frequency of use, follow-up compliance and cancer yield were calculated.
Results: Among the index exams, 96/160 (60%) were initially imaged for local staging of known carcinoma, 27/160 (17%) for high risk screening, 17/160 (11%) for problem solving and the remaining 20/160 (13%) for other reasons. 100/160 (63%) women returned for the recommended MR after a mean time to follow-up of 7.4 months (range, 1-25 months). Based on the follow-up examination, the BI-RADS 3 assessment was downgraded to a BI-RADS 1 or 2 in 90/100 (90%), unchanged in 7/100 (7%) and upgraded with biopsy recommended for the remaining 3/100 (3%). Only 1/3 underwent a biopsy which was benign. 60/160 (38%) patients never underwent a second MR despite an average follow-up time of 19.5 months (range 11 - 29 months). Of these, 53/160 (33%) were lost to follow-up or declined and 7 were not imaged secondary to mastectomy for known cancer. 1/7 mastectomy specimens contained malignancy, separate from the primary tumor, corresponding to the BI-RADS 3 findings on MR resulting in a cancer yield of 1/160 (0.6%).
Conclusion: Our data demonstrate a low frequency of malignancy among MR lesions assessed as BI-RADS 3 and suggest that follow-up is a valid alternative to biopsy for these lesions. However, in our series, the BI-RADS 3 assessment was used more frequently than in mammography, and overall patient compliance was only 63%.
2:30 PM
267. Breast MRI in Evaluation of Eligibility for Partial Breast Irradiation
Godinez J.1,2*; Gombos E.C.3,4; Devlin P.M.1,2; Birdwell R.L.3,4; 1. Radiation Oncology, Brigham and Women's Hospital, Boston, MA; 2. Radiation Oncology, Dana Farber Cancer Institute, Boston, MA; 3. Radiology, Brigham and Women's Hospital, Boston, MA; 4. Radiology, Dana Farber Cancer Institute, Boston, MA.
Address correspondence to J. Godinez (jgodinez{at}partners.org)
Objective: The purpose of the study was to retrospectively assess the preoperative MRI exams on women with early stage node negative breast cancer theoretically eligible for partial breast irradiation.
Materials and Methods: Seventy-five cases of women (ages from 29 to 75 years, mean 48 years) who were by clinical and mammographic criteria considered eligible candidates for breast conservation surgery and partial breast irradiation had their breast MRIs retrospectively evaluated. All patients underwent ipsilateral MRI after core needle biopsy diagnosis of breast cancer. Mammograms were compared to MRIs by radiologists based at a university hospital with subspecialty in breast imaging. At review new measurements of the lesions were performed and the presence of occult foci of tumor documented. The quadrant location and relationship to the index tumor were recorded.
Results: Of 75 cases 25 (33%) had mammographically occult cancer foci. Fourteen of the 25 (56%) had additional foci in the same quadrant as the index tumor. Six of the 25 (24%) had additional foci in other quadrants than the index tumor. Five cases of the 25 (20%) had additional foci in both the same and different quadrants.
Conclusion: This retrospective study demonstrated a clinically occult cancer in one third of the women with early stage node negative breast cancer theoretically eligible for partial breast irradiation. For partial breast irradiation eligible patients, MRI should be considered as an important part of their evaluation prior to selection as this treatment does not include the entire breast.
268. Role of Targeted Ultrasound in Evaluating Suspicious Breast MRI Lesions
DeMartini W.B.1,2*; Lehman C.D.1,2; Peacock S.1,2; Russell M.T.1; 1. Radiology, University of Washington, Seattle, WA; 2. Breast Imaging, Seattle Cancer Care Alliance, Seattle, WA.
Address correspondence to W.B. DeMartini (wdemarti{at}u.washington.edu)
Objective: Evaluation and management of suspicious breast MRI lesions vary. Some practice sites perform targeted ultrasound (US) for all such lesions, while some proceed directly to MRI guided biopsy. In addition, some practices elect to follow rather than biopsy suspicious MRI lesions that are occult on US. The purpose of our study was to determine the significance of US detection of suspicious lesions identified on breast MRI.
Materials and Methods: The records of 1146 consecutive breast MR examinations (1/1/03 to 12/30/04) were reviewed. This yielded 205 non-palpable, mammographically occult, MRI-detected suspicious lesions for which targeted US and imaging guided biopsy were recommended and performed. Lesions visible on US were biopsied using US guidance, and those occult on US were biopsied using MRI guidance. ACR BI-RADS MRI features and histopathology results for all lesions were recorded and compared.
Results: Overall, 72/205 (35%) of MRI lesions were identified with targeted US. 26/60 (43%) of malignant lesions were visible on US, and 34/60 (57%) were sonographically occult. The majority of lesions found to be infiltrating ductal carcinoma were visible on US, but 11/29 (38%) had no US correlate. 14/18 (78%) of ductal carcinoma in situ and 6/9 (67%) of infiltrating lobular carcinoma cases were occult on US. The rate of malignancy was slightly higher for lesions detected with US (36.1%) than for lesions which were sonographically occult (25.6%), but the difference was not statistically significant (p = 0.1). Masses were significantly more likely to be identified with US than were other lesion types, with 46/92 (50%) of masses seen sonographically compared to 13/38 (34%) of foci and 13/75 (17%) of non-mass-like enhancement (p < 0.0001). There was no significant correlation between mass size and US detection (p = 0.49), although 4/5 (80%) of masses measuring less than or equal to 5 mm were not seen with US.
Conclusion: The rates of malignancy for US detected (36.1%) and US occult (25.6%) lesions were not significantly different, and both are within the BI-RADS desirable positive biopsy rate range. Our study suggests that lack of US detection does not negate the need for biopsy of suspicious breast MRI lesions. US did identify 43% of malignancies, and detected 50% of mass lesions. Given the benefits of US guided biopsy, targeted US is a reasonable initial approach, and may be particularly useful for masses larger than 5 mm.
269. Utility of Second Look Ultrasound as an Adjunct to Contrast-Enhanced MRI of the Breast
Chellman-Jeffers M.R.*; Listinsky J.; Dinunzio A.; Lieber M.; Rim A.; Breast Imaging Section, Division of Radiology, Cleveland Clinic Foundation, Cleveland, OH.
Address correspondence to M.R. Chellman-Jeffers (chellmm{at}ccf.org)
Objective: To estimate the efficacy of "second-look ultrasound," i.e., directed ultrasound for examining breast findings of indeterminate significance as disclosed by contrast-enhanced MRI of the breasts.
Materials and Methods: We reviewed the final reports of contrast-enhanced MRI cases performed between 11/21/03 and 12/31/04 on a 1.0T scanner with CAD. For those cases in which ultrasound evaluation of one or more indeterminate findings were recommended ("second-look ultrasound"), we obtained and tabulated the reports of subsequent ultrasound imaging, as well as pathologic diagnoses from all ultrasound-guided biopsies which were performed as a consequence of the second-look ultrasound studies.
Results: Of 612 total contrast-enhanced MRI studies, recommendations for second-look ultrasound were offered for 80 patients (13.1%) with 104 individual findings. Of these, 73 (91.3% of recommendations) of the patients actually returned to our department and 95 indeterminate findings (91.3%) were evaluated. One or more corresponding sonographic findings were found in 38 patients (52.1%). Biopsies were carried out for 19/38 patients with findings (26.0% of the patients who returned). Of the 19 biopsied patients, 9 were shown to have one or more biopsies positive for breast cancer (47.4% of biopsied patients or 11.2% of patients recommended for second-look ultrasound). Of the 22 biopsied findings, 10/22 were cancer (45.5% or 9.6% of total findings).
Conclusion: Breast cancer lesions, either new cancers or additional foci in patients with a known cancer, were proven in about 10% of patients for whom second-look ultrasound was recommended after breast MRI at our institution. The ability to appropriately revise the therapy in those individual patients was obtained without resort to MR-guided biopsy. This approach may be useful for centers which lack the ready availability of MR-guided breast biopsy.
270. Characteristics of Breast MRI Screen-detected Lesions that Should be Sent for Targeted Ultrasound
Moscovici O.1*; Causer P.A.1; Jong R.A.1; Hill K.2; Warner E.2; Plewes D.B.3; 1. Medical Imaging, Sunnybrook and Women's College Health Sciences Centre, Toronto, Ont, Canada; 2. Medical Oncology, Sunnybrook and Women's College Health Sciences Centre, Toronto, Ont, Canada; 3. Medical Biophysics, Sunnybrook and Women's College Health Sciences Centre, Toronto, Ont, Canada.
Address correspondence to O. Moscovici (oanamoscovici{at}yahoo.com)
Objective: To determine which MRI lesions detected in a high risk screening population should be referred for second look ultrasound.
Materials and Methods: During the period from July 2001 to May 2005, breast MRI screening detected 159 BI-RADS categories 0, 4 and 5 lesions in 108 women at high risk for breast cancer. Targeted ultrasound was recommended as a part of the diagnostic workup. Included are 112/159 (70%) BI-RADS 0, 4 or 5 lesions that underwent targeted ultrasound. Diagnostic follow-up MRI was performed on the same date for 103/112 (92%) of these lesions. Additional inclusion criteria included: tissue sampling of the lesion or a minimum 1 year follow-up MRI documenting stability or resolution of the presumed benign lesion yielding 88/112 (78%) lesions. The screening breast MRIs and targeted sonograms were retrospectively reviewed and the lesions classified using MRI BI-RADS descriptors and size. The number of MRI lesions which were sonographically visible and the number of cases with a histopathologic diagnosis were determined. MRI lesion characteristics, size, benign and malignant pathology and benignity based on follow-up were correlated to the presence or absence of a sonographically visible lesion using logistic regression analysis.
Results: The MRI morphologic characteristics of the 88 lesions were: 38 (44%) mass, 39 (45%) focal area and 11 (13%) linear/ductal. Mean lesion size was 12.2 mm. 31 (35%) were sonographically visible. 36 (42%) were benign based on follow-up studies, 29 (35%) benign and 21 (24%) malignant based on histopathology. Lesions either characterized as a focal area enhancement or due to benign disease based on follow-up were significantly unlikely to be sonographically visible (p = 0.01 and 0.001). Although not significant, there was a trend for linear/ductal enhancement (p = 0.06) and biopsy proven benign disease (p = 0.8) to be non-visible and increased visibility with increasing lesion size (p = 0.7).
Conclusion: Targeted ultrasound can be omitted from the diagnostic algorithm of breast MRI screen-detected lesions characterized as a focal area, and most linear/ductal lesions. Also lesions which either resolve or are classified as likely benign after evaluation with diagnostic MRI do not require focused sonographic evaluation.
271. ADH Underestimation at MRI-Guided 9-Gauge Vacuum-Assisted Breast Biopsy
Holland A.E.*; Morris E.A.; Bartella L.; Dershaw D.D.; Liberman L.; Radiology, Memorial Sloan-Kettering Cancer Center, New York, NY.
Address correspondence to A.E. Holland (hollanda{at}mskcc.org)
Objective: The purpose of this study was to determine the frequency of diagnosing atypical ductal hyperplasia (ADH) at MRI-guided 9-gauge vacuum-assisted breast biopsy, and to assess the ADH underestimation rate at subsequent surgical excision.
Materials and Methods: With approval of our Institutional Review Board, retrospective review was performed of medical records of 237 consecutive MRI-detected lesions that had MRI-guided biopsy during a 33-month period. Biopsies were performed using a 9-gauge MRI-guided vacuum-assisted biopsy probe (ATEC, Suros, Indianapolis, IN). Pathology records were reviewed to identify lesions that yielded ADH at vacuum-assisted biopsy. Imaging studies were reviewed to determine MRI lesion characteristics, and medical records were reviewed to evaluate surgical histologic outcome. An ADH underestimate was defined as a lesion yielding ADH at vacuum-assisted biopsy and cancer at surgery. The 95% confidence intervals (CI) were calculated using the Geigy Scientific Tables.
Results: Histologic analysis of material obtained at MRI-guided vacuum-assisted biopsy yielded ADH without cancer in 15 (6%) of 237 lesions. Among 15 patients inwhom vacuum-assisted biopsy yielded ADH, median age was 52 (range, 46-68) years. The median number of specimens obtained, reported in 11 lesions, was 9 (range, 8-18). Median MRI lesion size was 1.3 (range, 0.7-7.0) cm. Among 15 MRI lesions, ten (67%) were non-mass lesions and five (33%) were masses. Surgical excision, recommended in all 15 lesions, was performed in 11 lesions. In these 11 lesions, surgical histology was malignant in four (36%), of all which were ductal carcinoma in situ (DCIS); high-risk in five (45%), including ADH and lobular carcinoma in situ (LCIS) in two, LCIS in two, and ADH in one; and benign in two (18%). These data indicate an ADH underestimation rate of 36% (95% CI, 11%-69%).
Conclusion: ADH without cancer was encountered in 6% of MRI-guided 9-gauge vacuum-assisted biopsies. ADH at MRI-guided vacuum-assisted biopsy is an indication for surgical excision, due to the high (36%) frequency of underestimation in these lesions.
272. MRI of the Breast Post Biopsy
Mehdizade A.*; Curran S.; Morris E.; Radiology, Memorial Sloan-Kettering Cancer Center, New York, NY.
Address correspondence to A. Mehdizade (mehdizadea{at}hss.edu)
Objective: The purpose of this study was to assess the frequency and spectrum of post biopsy changes on MR imaging of breast following percutaneous needle biopsy and their effect on subsequent MR interpretation. This has not been previously described in the literature to our knowledge.
Materials and Methods: From February 1998 to May 2005, the records of 415 patients who had breast MRI was retrospectively reviewed. From these 51 had needle core or stereotactic biopsy prior to the breast MRI. Patients who had lumpectomy, surgical excision or fine needle aspirations were excluded form this study. The average age was 55 ranging from 39 to 73 years. Lesion size ranged from 0.4 to 3.0 cm. median 1.4cm. The mean interval time between the MR and biopsy was 213 days with a range of 1 to 545 days. Of these examinations, 42 had ultrasound guided breast core biopsies, and 11 Stereotactic biopsy, prior to the MRI exam. From these 14 were benign and 37 were malignant. Analyses of MRI was performed by two radiologists blind folded to the results to detect hematomas, distortion, needle track artifacts or abnormal enhancement, limiting subsequent MR diagnosis.
Results: Post biopsy hematoma was observed in one case only. There were no significant enhancements following the MR biopsies, or needle track artifacts obscuring the lesions.
Conclusion: Post biopsy changes do not limit MRI interpretation. MRI of the breast can be performed anytime after stereotactic or US guided core biopsy.
* Will present paper
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
