| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||
ABSTRACT |
Thursday, May 4, 10:00 AM-12:30 PM
Abstracts 239-250
Moderators: Roberta A. Jong, MD and Lonie R. Salkowski, MD
10:00 AM
239. A Survey of Patient Beliefs about Screening Mammography
Roubidoux M.A.*; Tang T.; Patterson S.; Yang L.; Helvie M.A.; Radiology, University of Michigan Health Systems, Ann Arbor, MI.
Address correspondence to M.A. Roubidoux (roubidou{at}umich.edu)
Objective: To assess patient expectations and beliefs about screening mammography.
Materials and Methods: We administered a survey to women who came to an outpatient clinic for screening mammography, and 397 women voluntarily agreed to participate. Institutional Review Board approval was obtained beforehand. Demographic and clinical data (age, educational level, current breast problems, reason for the mammogram, and prior mammography) were obtained. Survey Questions in the survey to assess patient beliefs were: (1)'Mammograms can detect all breast cancers, even if they are extremely small'; (rated on a 5 point agreement scale) (2)'What do you think the chances are that your mammogram will detect a breast cancer today?'(;choose one of 6 categories) (3) `What do you think your personal risk of breast cancer is?'; (fill in the blank).
Results: The reason for mammography was screening in 97.7% of respondents, and 2.3 % did not answer. Age range was 40-83 years, mean 52 years. Eighty-four percent of respondents had > 1 year of post high school education, 55% were college graduates, and 23 % had graduate degrees. Prior mammograms had occurred in 94% of respondents; 14.1% had had a prior benign breast biopsy. The responses to the survey questions showed that 16% thought their personal risk was 50% or higher, and 14.1% believed that the likelihood of detecting breast cancer that day was at 50% or higher. 20.6% agreed with the statement `mammograms detect all breast cancers' and 11.4% were neutral about this statement. Answers to questions did not significantly correlate with patient educational level.
Conclusion: Overall, the educational level of the respondents was high and the vast majority had had prior mammography. Nevertheless, among a substantial proportion of these women, beliefs about their personal risk of breast cancer, the likelihood of having breast cancer detected by screening mammography, and expectations about the performance of mammography were abnormally high or unrealistic. These expectations could contribute to patient anxiety and to medico-legal risk.
240. The Hidden Costs to High Patient Compliance in Recall from Screening Mammography
Blane C.E.*; Pinsky R.W.; Joe A.I.; Blajan M.R.; Helvie M.A.; Radiology, University of Michigan, Ann Arbor, MI.
Address correspondence to C.E. Blane (cblane{at}umich.edu)
Objective: To document the hidden costs achieving high recall patient compliance from an off-line screening mammography program.
Materials and Methods: This study was IRB approved. Scheduling for our off-line screening program includes informing patients of the possibility for recall for a diagnostic study. No patient is placed in final BIRADS assessment 3, 4, or 5 without a diagnostic study at our institution. Each Category 0, an incomplete study, is flagged on the day sheet by the interpreting radiologist and computer coded for recall. From the day sheets, a clerk notes the case and initially contacts the patient by phone within 2 working days of the screening interpretation to schedule the diagnostic study. In addition a letter is sent to both the patient and referring physician indicating the need for additional imaging. Diagnostic slots are purposely left open to accommodate these cases. Knowing our recall rate allows us to calculate the average number of diagnostic slots required on a weekly basis. An ongoing computer tickler file of patients without a diagnostic code provides a further check.
Results: From 2002 through 2004, 4025/30,286 screening patients were recalled (13%), a mixture of baseline and interval screening cases. 3977/4005 patients returned for a Diagnostic study after an average of 2.0 telephone calls, average 3.65 minutes total clerical time (60 patient recalls prospectively measured time data) costing $1.03 per case (clerical salary + benefits averages $17 per hour). 48/4025 initially noncompliant patients receive an average of 6 telephone calls (4.7 minutes total time). Clerical costs for this group are $2.27 with the additional cost of $7.50 for a registered letter totaling $9.77 per case. 28 of this group returned for additional imaging. It should be noted one of these initially noncompliant patients went on to biopsy revealing a breast cancer. Patient compliance for 2002 - 2004 was 4005/4025 or 99.5 %. Total cost for this program over the 3 years was $4581.84 or $0.15 per screening patient.
Conclusion: The radiology department assumed responsibility for contacting patients who needed additional imaging. With strict documentation of the patients requiring additional imaging, computer checks, clerical support and prompt scheduling we were able to achieve 99.5% compliance. The hidden cost of this compliance is small, 15 cents per screening patient. As a side benefit to good patient care we feel this also reduces our risk exposure.
241. Contribution of Computer Assistance to Cancer Detection in 105,336 Screening Mammograms
Leung J.W.1*; Margolin F.R.2; Jacobs R.P.2; Denny S.R.2; Schrumpf J.D.2; 1. Radiology, University of California San Francisco, San Francisco, CA; 2. Breast Health Center, California Pacific Medical Center, San Francisco, CA.
Address correspondence to J.W. Leung (Jessica.Leung{at}ucsfmedctr.org)
Objective: To determine how computer-aided detection contributed to cancer detection at screening mammography in a high-volume hospital-based community practice.
Materials and Methods: This is a prospective study of consecutive screening mammograms interpreted at a single site. During a 46-month period after the implementation of computer-aided detection (CAD), 105,336 screening mammograms were interpreted by 9 radiologists. At the time of screening mammographic interpretation, each radiologist was required to enter into a computerized database whether the finding prompting recall was attributable to CAD only, radiologist only, or both. Audit programs enabled linkage of each recall to BI-RADS assessment at subsequent diagnostic evaluation and to tissue diagnosis if biopsy was performed. Audit results from the study period were also compared with those collected during the 46-month period immediately prior to the implementation of CAD (N=100,689), which served as a historical control. Minimal cancer was defined as invasive cancer or ductal carcinoma insitu (DCIS) equal to or less than 1 cm, and early-stage cancer as stage 0 or 1 cancer. Positive predictive values were defined in the BIRADS Atlas.
Results: After the implementation of CAD, the recall rate increased by 14%, from 5.8% (5,793 of 100,689) to 6.6% (6,819 of 105,336). Nine cancers were attributed to detection by CAD only at time of screening. Of these 9 cancers, 6 were DCIS and seen as calcifications, 2 were invasive ductal carcinomas and seen as masses, and 1 was invasive lobular carcinoma and seen as architectural distortion. In total, 341 cancers were detected during the study period. Therefore, CAD increased cancer detection by 3% (from 332 to 341 cancers). The cancer detection rate remained the same at 3.2 per 1,000 screening mammograms both before and after CAD. The detection of DCIS increased by 17% (from 29 to 34%), minimal cancer increased by 4% (from 56 to 58%), and early-stage cancer increased by 2% (from 84 to 86%). Positive predictive values before and after CAD were as following, respectively: PPV1 6% and 5%; PPV2 27% and 25%; PPV3 29% and 30%.
Conclusion: A 3% increase in cancer detection was contributed to detection by CAD, but comparison with historical cohort showed no difference in cancer detection (un-changed at 3.2 cancers per 1,000 screens). Use of CAD was accompanied by a 14% increase in recall rate. Detection of favorable-prognosis cancers were increased, with the largest increase in the detection of DCIS.
242. Prospective Assessment of Computer-aided Detection in Interpretation of Screening Mammography
Ko J.M.1; Nicholas M.J.2; Mendel J.B.3; Slanetz P.J.3*; 1. N/A, Tufts University School of Medicine, Boston, MA; 2. Radiology, Hospital of Saint Raphael, New Haven, CT; 3. Radiology, Caritas St. Elizabeth's Medical Center, Boston, MA.
Address correspondence to P.J. Slanetz (priscilla_slanetz{at}cchcs.org)
Objective: To prospectively assess the clinical utility of computer-aided detection (CAD) in the interpretation of screening mammograms in a community-based teaching hospital.
Materials and Methods: Over a 26-month period, 5,016 screening mammograms were interpreted without and subsequently with the assistance of the iCAD Mammo-Reader detection system. Data collected for actionable findings included dominant feature (calcification, mass/asymmetry/architectural distortion), detection method (radiologist only, CAD only, or both radiologist and CAD), BI-RADS assessment code, associated histopathology for those undergoing biopsy, and tumor stage for malignant lesions. Other data selectively collected included breast density, number of CAD marks per image and presence/absence of prior films. The study population was cross-checked against an independent reference standard to identify false negative cases.
Results: Of the 5,016 cases, the recall rate increased from 12% to 14% with the addition of CAD. One hundred and seven patients underwent biopsy of which 101 (94%) were prompted by the radiologist and 6 (6%) were prompted by CAD. Of the biopsies, 79 (64%) were benign and 45 (36%) were in-situ or invasive carcinoma. The radiologist detected 43 (89%) of 48 malignancies and 45 (93%) of 48 malignancies with the assistance of CAD. CAD missed eight cancers that were detected by the radiologist, presenting as architectural distortions (3), irregular masses (4) and a circumscribed mass (1). CAD detected two in-situ cancers as a faint cluster of calcifications that had not been perceived by the radiologist and one mass that was dismissed by the radiologist, accounting for at least a 4.7% increase in cancer detection rate and 4% increase in sensitivity.
Conclusion: Routine use of CAD while interpreting screening mammograms significantly increases recall rates, has no significant effect on positive predictive value for biopsy and can increase cancer detection rate by at least 4.7% and sensitivity by 4%.
243. Developing Asymmetry Identified at Mammography: Correlation with Pathology and Imaging Outcome
Leung J.W.*; Sickles E.A.; Radiology, University of California San Francisco, San Francisco, CA.
Address correspondence to J.W. Leung (Jessica.Leung{at}ucsfmedctr.org)
Objective: A developing asymmetry at mammography is an asymmetry that has appeared or increased in size or conspicuity since a prior exam. It may be a benign finding or a nonspecific sign of cancer. We determined the frequency, imaging outcome, and pathological significance of developing asymmetry.
Materials and Methods: This is a retrospective cohort study of data collected over a 20-year period. At time of interpretation, the principal mammographic finding (if any) of each mammographic exam was recorded by the radiologist in a database. We identified our study cohort by searching our database for all cases with the finding of developing asymmetry. Also included in the database were demographic factors including patient age, menopausal status, and family or personal history of breast cancer. We examined radiology records to obtain information on the use of ultrasound and magnetic resonance imaging as adjunctive diagnostic tools and pathology records to determine tissue diagnosis.
Results: The finding of developing asymmetry was present in 290 of 180,801 consecutive screening (0.16%) and 34 of 27,330 consecutive diagnostic exams (0.12%). Three screening cases were excluded because of failure to return for diagnostic imaging. Nine other cases were excluded because of failure to comply with recommendation for biopsy (7 screening, 2 diagnostic). Thus, the study consisted of 280 screening and 32 diagnostic cases. For the 280 screen-detected developing asymmetries, biopsy was recommended and performed in 63 (23%). Thirty-four cancers were identified, resulting in a positive predictive value (PPV1) of 12% (34/280). Among the 32 diagnostic cases, biopsy was recommended and performed in 28 (88%). Nine cancers were identified, resulting in a PPV1 of 28% (9/32). Histologic details were available for 30 of 43 cancers: invasive ductal (19), lobular (5), mucinous (2), tubular (1), micropapillary (1), and ductal carcinoma in-situ (2). The most common benign histologic diagnoses were fibrocystic change (12), fibrosis (10), and pseudoangiomatous stromal hyperplasia (7). Ultrasound was performed for 28 of the 43 cancer cases: solid mass (18), no sonographic correlate (8), hypoechoic focus (1), hypoechoic ducts (1).
Conclusion: Developing asymmetry is an uncommon finding at mammography. It is more likely to be malignant when identified at diagnostic mammography than screening (PPV1 of 28% vs. 12%). However, even when detected at screening mammography, the likelihood of malignancy is sufficiently high to justify tissue diagnosis.
244. Recurrent Cancer Following Breast-Conserving Surgery and Radiation for Ductal Carcinoma In Situ: Mammographic Features, Method of Detection, and Stage of Recurrence
Pinsky R.W.1*; Rebner M.2; Pierce L.J.3; Vicini F.A.4; Helvie M.A.1; 1. Radiology, University of Michigan, Ann Arbor, MI; 2. Radiology, William Beaumont Hospital, Royal Oak, MI; 3. Radiation Oncology, University of Michigan, Ann Arbor, MI; 4. Radiation Oncology, William Beaumont Hospital, Royal Oak, MI.
Address correspondence to R.W. Pinsky (rpinsky{at}umich.edu)
Objective: To determine the mammographic appearance, method of detection, and stage of Ipsilateral Breast Tumor Recurrence (IBTR) in women treated with breast-conserving surgery (BCT) and whole breast radiotherapy (RT) for Ductal Carcinoma In Situ (DCIS).
Materials and Methods: Following IRB approval, the records of women treated with BCT/RT for DCIS without an invasive component who later developed an IBTR from 1981-2003 were reviewed. This was a multi institutional study conducted at a large academic center and a large private hospital. 513 women were studied of whom 42 (8.2%) developed a local recurrence. Complete records were available for 32 women who constitute the study cohort. The mean age at diagnosis was 49 years (range 26-73 years). The mammograms were analyzed by 2 MQSA certified radiologists (1and 2).
Results: The actuarial rates of IBTR at 5 and 10 years were 5.9 and 9.8%. The mean time to recurrence was 4.5 years (range 1-12 years). 31/32 (97%) recurrences were mammographically apparent and 29/32 (91%) were diagnosed exclusively by mammography. Two (6%) were mammographically apparent and palpable. One patient presented with mammographically occult Paget's disease. Mammographic findings at recurrence were: calcifications 24/32 (75%), mass 6/32 (18%), and distortion 1/32 (3%). 90% of initial cancers had manifested as microcalcifications. The mean distance between the initial and recurrent cancer was 3.8 cm. (range 0-9 cm.). Thirteen (43%) had recurrences in a different quadrant than the original cancer. Recurrences were DCIS in 17 (53%), DCIS with microinvasion in 6 (19%), invasive ductal cancer in 3 (9%), invasive lobular cancer in 2 (6%), and mixed DCIS/invasive cancer in 4 (13%). 6/9 (67%) of patients with invasive cancer (excluding microinvasion) had tumors less than 1 cm (median size of invasive tumor was 0.7 cm). Overall, 91% of recurrences were considered minimal cancers, i.e., DCIS and invasive cancer < 1 cm. All DCIS and invasive recurrences were stage 0 or 1. Zero of 9 patients with sampled nodes had axillary lymph node metastases.
Conclusion: Mammography successfully detected IBTR following BCT/RT for DCIS in 97% of cases predominantly as calcifications or masses. The recurrences were located at a variable distance from the lumpectomy site. 91% of recurrences were detected as Minimal Cancers and all were early stage, connoting a high potential for cure.
245. Breast Imaging of Patients Age 20-25 Years
Joe A.I.*; Blane C.E.; Helvie M.A.; Nees A.V.; Adler D.D.; Diagnostic Radiology / Breast Imaging, University of Michigan, Ann Arbor, MI.
Address correspondence to A.I. Joe (ajoe{at}umich.edu)
Objective: To document our Breast Imaging experience in the 20-25 year age group. This cohort is not well represented in the literature.
Materials and Methods: A 4 year retrospective IRB approved search of the Radiology Information System yielded 204 patients (200 women and 4 men) aged 20-25 years evaluated in Breast Imaging. The medical records were reviewed for clinical findings and mammogram and ultrasound (US) for final BIRADS assessment categories. Outcomes were determined by biopsy, FNA or clinical follow-up = 1 year. 77 patients without tissue diagnosis or documented clinical resolution of problem or record of a clinical breast exam > 1 year after the imaging exam were considered lost to follow-up.
Results: Clinical presentation (some patients had > 1 complaint) included 169/204 (83%) palpable mass, 19 (9%) pain, 13 (6%) tenderness, 8 (4%) nodularity, 7 (3%) thickening, 7 (3%) nipple discharge, 2 (1%) rash, 2 (1%) personal history of breast cancer, and 1 each (0.5%) with asymmetry, increase breast size, and redness. 203/204 (99%) patients had US, 38(19%) of these patients also had mammography and one man had only mammography. 12/203 US (6%) were BIRADS U4, 3 of these were cancers (PPV = 25%, sensitivity 100%). 104 (51%) US were BIRADS U1. 15 (7%) were cysts BIRADS U2 and 72 (35%) were BIRADS U3. No cancers were detected in BIRADS U1-3. A total of 39 mammograms were performed. 7/39 (18%) were BIRADS M4 or M5; 3 of these were cancer (PPV = 43%, sensitivity 100%). The 4th patient with a cancer did not have preoperative imaging. She went directly to surgical bi-opsy for a palpable finding. 32/39 (82%) were BIRADS M1, M2, and M3. No cancers were found in BIRADS M1-3. In 59/204 (29%) biopsy or FNA yielded 4 cancers, 39 fibroadenomas, 3 cysts, 2 fibrocystic changes, 5 normal breast tissue, 2 intraductal papillomas, and one each of: inflammatory mass, tubular adenoma, lactating adenoma, lipoma, pseudoangiomatous stromal hyperplasia, and lymph node. 3 abscesses were drained. 3/169 (1.8%) palpable masses were cancer. One cancer was detected by surveillance mammography in a TRAM reconstruction. All 4 men had gynecomastia both clinically and on imaging.
Conclusion: Fibroadenoma was the most common pathologic diagnosis in the 20-25 year age group and cancer was rare. 3 women in this age group with a breast cancer presented with a palpable mass. All cancers imaged were BIRADS 4 or 5. US is a reasonable initial imaging study for symptomatic young women referred for breast imaging.
246. Features of Breast Cancers Identified at Mammmography in Women Younger than Age 40
Chiang S.H.*; Leung J.W.; Sickles E.A.; Radiology, University of California San Francisco, San Francisco, CA.
Address correspondence to S.H. Chiang (cshwa{at}yahoo.com)
Objective: To determine the features of breast cancers identified at mammography in women younger than age 40.
Materials and Methods: This is a retrospective cohort study of consecutive mammograms performed in women younger than age 40 at our institution during a 20.2-year period. Among 183,223 screening and 28,447 diagnostic mammograms, 15,731 (9%) screening and 2,464 (9%) diagnostic mammograms were performed in women younger than age 40. These 18,195 mammograms constitute this study. We extracted from a computerized database the following parameters: age, number of exams in which cancer (invasive carcinoma and/or ductal carcinoma in situ [DCIS]) was identified, TNM cancer stage, principal mammographic finding, clinical presentation, and personal or family history of breast cancer.
Results: Ninety-four cancers were identified, among which 31 were diagnosed at screening and 63 at diagnostic mammography. Hence, the cancer detection rates were 2.0 per 1,000 for screening and 25.6 per 1,000 for diagnostic mammography. For the same study period, the cancer detection rates at our institution among all age groups were 5.2 per 1,000 for screening and 49.4 per 1,000 for diagnostic mammography. Staging of the 31 cancers detected at screening mammography was: 20 (65%) stage 0 (DCIS), 4 (13%) stage I, 6 (19%) stage II, 1 (3%) stage III. Staging of the 63 cancers detected at diagnostic mammography was: 14 (22%) stage 0 (DCIS), 18 (29%) stage I, 29 (46%) stage II, 2 (3%) stage III. Among the 31 screen-detected cancers, 19 (61%) were seen as calcifications, 5 (16%) as masses, 3 (9%) as asymmetries, 2 (7%) as architectural distortion, and 2 (7%) as masses with calcifications. Among the 63 cancers detected at diagnostic mammography, 26 (41%) were seen as masses, 17 (27%) as masses with calcifications, 16 (25%) as calcifications, and 3 (5%) as asymmetries, and 1 (2%) as architectural distortion. All diagnostic mammography cancer patients except for 2 (61/63, 97%) presented with palpable lumps.
Conclusion: The cancer detection rate at either screening or diagnostic mammography in women younger than age 40 was less than that in women of all older age groups. Screen-detected cancers were less frequent and of earlier stage than cancers detected at diagnostic mammography. Most screen-detected cancers were DCIS and presented as calcifications. At diagnostic mammography, nearly half of cancers detected were stage II, and most were seen as masses with or without calcifications. The most common symptom of cancer was a palpable lump.
247. The Use of BI-RADS 3 Classification in a Symptomatic Patient Population
Walsh S.M.*; Sweeney I.; McCarthy P.A.; Radiology, University College Hospital Galway, Galway, Ireland.
Address correspondence to S.M. Walsh (sinwalsh{at}gmail.com)
Objective: The ACR BI-RADS classification has been widely adopted in the screening population to identify a group of women with such a low probability of malignancy that periodic short term follow up is a safe alternative to core biopsy or surgery. All of the published studies to date have specifically excluded palpable lesions so the use of the "probably benign" BI-RADS 3 categorization in this subgroup is not supported by clinical data. The purpose of our study was to determine by prospective analysis of cases biopsied in a symptomatic population whether application of the BI-RADS 3 category and the suggested follow up guidelines could also be safely applied to symptomatic women.
Materials and Methods: We undertook a prospective study of all cases with BI-RADS 3 characteristics on mammography or ultrasound who subsequently underwent guided core biopsy over an eighteen month period. Women with palpable lesions were separately identified. Histology was reviewed in all cases to determine concordancy.
Results: A total of 113 lesions (including palpable and non palpable) met the criteria for BI-RADS 3 classification. No cancers were found in this group and none found on any repeat biopsy.
Conclusion: Palpable masses have traditionally been excluded from the suggested BI-RADS 3 management guidelines. Our figures, however would indicate that if the BI-RADS 3 features of a lesion are confidently identified, then palpability confers no greater risk of malignancy than that of the screening population and therefore, such a group can be safely followed up on a short term basis with biopsy reserved for cases demonstrating interim growth.
248. Sonography of Complex Breast Cysts: Is Aspiration Necessary to Exclude Malignancy?
Daly C.P.*; Bailey J.E.; Helvie M.A.; Klein K.A.; Department of Radiology, University of Michigan, Ann Arbor, MI.
Address correspondence to C.P. Daly (cpdaly{at}med.umich.edu)
Objective: To determine the incidence of breast cancer in patients presenting for fine needle aspiration of sonographically classified complex cysts.
Materials and Methods: Following IRB approval, we identified 204 consecutive women who presented for fine needle aspiration of 282 complex or painful cysts detected by clinical exam or imaging from Jan. 2002 through Aug. 2003. Mean age of patients was 50 years. A total of 5599 breast ultrasound (US) exams were performed during the study period. Patients recommended for biopsy of sonographically solid masses were excluded. Complex cysts were defined as those meeting some but not all criteria for simple cysts: circumscribed, anechoic, posterior acoustic enhancement, no solid component, and no Doppler signal. All US exams were performed exclusively by breast imaging radiologists. Initial mammography, ultrasound, and procedure reports were reviewed. Pathologic diagnosis was established based on fluid character, cytology, core needle biopsy, or excisional biopsy.
Results: 129/282 (45.7%) aspiration samples with typical cyst fluid were discarded. 5/282 (1.8%) cysts decompressed when punctured, and no fluid was collected. 148 (52.5%) samples were submitted to cytology: 145 (98%) were benign and 3 (2%) had atypia. None of 282 (0%) complex cysts proved malignant. 31 patients underwent core biopsy or surgical excision due to discordant imaging findings: there were 10 fibroadenomas; 9 fibrocystic changes; 5 fibroadenoma and fibrocystic changes; 1 hematoma; 1 lymph node; 1 foreign body reaction. Of 3 patients with cytologic atypia, 1 had atypical lobular hyperplasia, 1 had a myxoid fibroadenoma, and 1 had a fibroadenoma and fibrocystic change. Two malignancies in proximity to the index cysts were identified at the time of initial evaluation. One patient with a sonographically solid mass recommended for core biopsy adjacent to aspirated cysts proved to be invasive ductal carcinoma. The second patient underwent cyst aspiration without resolution of mammographic findings and proved to have a nearby sonographically solid mass diagnosed as invasive ductal carcinoma. Due to geographic discordance, this cancer may have been overlooked if post aspiration mammogram had not been performed.
Conclusion: No sonographic mass prospectively identified as a complex cyst proved malignant. This provides support for the practice of classification of these lesions as probably benign rather than suspicious.
249. Improved Ultrasound Detection of DCIS: A Review of Imaging Criteria
Al Sufayan R.1; Duchesne S.2; Duchesne N.1*; 1. Breast Imaging and Intervention, Ottawa Regional Women's Health Breast Center, Ottawa, ON, Canada; 2. Biomedical Engineering, McGill University, Montreal, QC, Canada.
Address correspondence to N. Duchesne (nathalie_duchesne_22{at}yahoo.ca)
Objective: Pure Ductal Carcinoma In Situ (DCIS) is uncommonly noticed at 2D Ultrasound (US) and has received scanty attention in the literature. The purpose of this study was to review features of pure DCIS in order to improve its detection.
Materials and Methods: Subjects -This is an IRB approved single center retrospective study of the breast ultrasounds performed from November 2001 to September 2005. Selection of patients was based on a diagnosis of pure DCIS, excluding cases with UGBBx / surgical results showing an invasive component. Data collection - Reasons for ultrasound and mammographic findings were collected. Two radiologists (one breast expert and one resident) reviewed independently the US films and assessed US occurrence frequency for US criteria such as duct extensions, branching patterns, microlobulations, microcalcifications, enlarged ducts, tumor filled ducts, lobule cancerizations, amorphous sheets of tissue, intraductal/intracystic solid lesions and lobule invasion. Data analysis- Data were analyzed with the Spearman's Rho correlation test.
Results: Our cohort consisted of 21 patients, representing 0.6% of all UGBBx performed at our center during that time period. The most frequent reason for US was the presence of a mass or opacity on the mammogram. Microcalcifications were seen on mammogram in only 5 (23.8%) patients. US-visible microlobulations had the highestoccurrence (28, seen in 19 (86%) patients), followed by duct extension (19, seen in 12 (54%) patients) and lobule cancerization (18, seen in 12 (54%) patients). All other features occurred with lower frequencies. 95% of patients (20/21) had 2 or more co-occurring features. The most statistically significant combination determined using the correlations test (p < 0.0002) were enlarged, tumor filled ducts. Other significant combinations (p < 0.05) included duct extension with either lobules cancerization, amorphous sheets of tissue or branching pattern; amorphous sheets of tissue with tumor filled ducts; and finally branching pattern with lobules cancerization.
Conclusion: Pure DCIS is an uncommon entity encountered at US assessment. Our results indicate that most pure DCIS present with 2 or more US morphological criteria, the most frequent being microlobulations and duct extension. However, the most significant combination of US criteria for DCIS diagnosis is enlarged, tumor filled ducts. Pure DCIS recognition on US is critical to prevent progression to invasive carcinoma and improve patient's management.
250. Clinical Outcomes when Breast Biopsy is Recommended by the Radiologist but Not Performed
Lvoff N.M.*; Leung J.W.; Sickles E.A.; Department of Radiology, University of California San Francisco, San Francisco, CA.
Address correspondence to N.M. Lvoff (natalya.lvoff{at}radiology.ucsf.edu)
Objective: To review clinical outcomes in women who do not undergo prompt biopsies that are recommended by breast-imaging radiologists.
Materials and Methods: This is a retrospective review of 22,018 consecutive diagnostic mammography examinations performed from 1997 to 2003. We studied women who were given BI-RADS category 4 or 5 assessments and biopsy recommendations for whom biopsy was not performed within the next 6 months. The reason for not obtaining biopsy was determined by health-care provider response to a radiology letter requesting follow-up or from the woman's medical record, and was categorized into two groups, "woman refused biopsy" and "health-care provider advised against biopsy." Outcomes were determined by either radiology-pathology concordant biopsy or at least two years of documented follow-up after the initial diagnostic mammography examination.
Results: Of 22,018 diagnostic mammography examinations, 3,045 (13.8%) were assessed as BI-RADS 4 or 5. Ninety cases (0.41%) met study inclusion criteria. Ten women were then excluded because substantial (non-breast) co-morbidity was judged to preclude biopsy, and 17 women were lost to follow-up, leaving a final study group of 63 women. Of these, lack of biopsy was attributed to patient refusal in 19 (30.2%) cases, and to clinician's advice in 44 (69.8%) cases. During subsequent follow-up, biopsy was performed for 23 women (36.5%), with a cancer diagnosis in 11 (17.5%). The remaining 40 women were followed for at least 2 years without cancer diagnosis (mean interval, 30.3 months), with either stable or decreased mammographic findings. The rate of cancer diagnosis was 10.5% (2 of 19) among patient-refusal cases and 20.5% (9 of 44) among clinician-advice cases; p = 0.29.
Conclusion: Non-compliance with radiologist-recommended biopsy is more often attributable to contrary advice from the health-care provider than to patient refusal. In choosing to forego prompt biopsy, clinicians are no more correct than their patients in selecting cases with benign outcomes. The overall 17.5% rate of cancer diagnosis in such cases is sufficiently high to support continued radiologist recommendation for biopsy during follow-up despite patient refusal or contrary clinician advice.
* Will present paper
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
