| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Breast Imaging |
1 From the Department of Radiology, University of Michigan Health Systems, 1500 E Medical Center Dr, Taubman Center 2910N, Ann Arbor, MI 48109-0326. Received July 9, 2001; revision requested August 3; revision received October 15; accepted December 10. Address correspondence to M.A.H.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
MATERIALS AND METHODS: One hundred twenty-three consecutive cases of breast thickening (103 patients) during a 1-year period were reviewed. Experienced breast examiners prospectively identified patients with breast thickening. Results of diagnostic mammographic work-up, breast US, breast biopsy, and clinical follow-up were retrospectively reviewed.
RESULTS: Six (5%) of 123 cases had a diagnosis of breast carcinoma; five (83%) of the six had invasive carcinoma. Mammography was performed in all cases, US in 77 (63%) cases. Mammographic sensitivity for invasive cancer detection was 60% (three of five cases), specificity was 94% (102 of 108 cases), and negative predictive value was 97% (102 of 105 cases). Sensitivity of US alone was 100% (two of two cases), specificity was 96% (65 of 68 cases), and negative predictive value was 100% (65 of 65 cases). The combined negative predictive value of mammography and US was 100%. Patients with prior biopsies at the site of palpable thickening accounted for most false-negative mammograms. Median time to initiate follow-up of patients in whom biopsy was not performed was 14 months.
CONCLUSION: Breast cancer was discovered in 5% of women with palpable breast thickening. Women with negative mammograms and US scans are at low risk for cancer but should, in our opinion, be followed up at short-term intervals with clinical examination and imaging if biopsy is not elected by their surgeon or clinician.
© RSNA, 2002
Index terms: Breast, US, 00.12981 • Breast neoplasms, diagnosis, 00.31, 00.324 • Breast radiography, 00.111, 00.114
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Because the known false-negative rate of mammography is approximately 5%–20%, most clinical breast management guidelines call for surgical consultation for persistent abnormal physical findings even when the mammogram is negative (4,5). Because surgical clinical breast examination is not absolute, many surgeons liberally perform fine-needle aspirations, core biopsies, and open surgical biopsies for abnormal physical findings (1). However, even the combination of a breast physical examination, mammography, and fine-needle aspiration may fail to demonstrate a breast cancer when the physical finding is breast thickening (6). For this reason, the management of breast thickening is especially problematic for clinicians.
The purpose of our study was to determine the frequency of breast carcinoma and ascertain the diagnostic yield of mammography and breast US in the detection of breast carcinoma in women with palpable breast thickening.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
At our institution, all symptomatic patients, as indicated by physician request or patient reporting, undergo a clinical breast physical examination by an experienced and trained mammography technologist before imaging. Patients with recent surgical history of benign breast biopsy, history of breast cancer, or breast implants or at callback for an abnormal screening mammogram also receive a clinical breast examination. The mammography technologist was unaware of the subsequent mammographic or US findings at the time of the clinical breast examination. The technologist coded the breast physical examination findings in the following manner: normal, benign mass, suspicious findings for malignancy, or breast thickening. These data were then prospectively entered into a computer database. In 1998, 172 (17%) of 995 abnormal clinical breast examinations were coded as breast thickening. The frequency of breast thickening relative to abnormal clinical breast examination findings has been stable over time, with 17%–23% of abnormal clinical breast examination findings indicating breast thickening during the 1995–1999 interval. In 1998, of 995 abnormal clinical breast examination findings, 724 (73%) were coded as benign masses and 99 (10%) as findings suspicious for malignancy.
Our mammography technologists have undergone standardized clinical breast examination training and have an average of 8.2 years (range, 2–21 years) of experience performing breast examinations. Breast masses are distinguished from thickening by the ability of the examiner to determine edges around a mass and generate measurements in all directions.
Patient demographic data were recorded, including menstruation status, hormonal replacement status, and patient risk factors such as personal history or first-degree relative with history of breast cancer. We evaluated whether it was the patient, her physician, or the mammography technologist who made the initial discovery of the abnormal physical finding. When a physician or the patient discovered the abnormality, we recorded the original description of the finding. We recorded the time interval from the discovery of thickening to the initial imaging work-up.
Mammography was performed with DMR equipment (GE Medical Systems, Milwaukee, Wis) and 2000 Min R film (Kodak, Rochester, NY) at a mammography site approved by the U.S. Food and Drug Administration. The imaging evaluation of breast thickening was assessed. Final mammographic assessment was prospectively reported by using the Breast Imaging Reporting and Data System (BI-RADS) categorization (7). Other factors recorded included whether a lateral mammographic view or spot view (with or without magnification) was obtained. Breast US was performed at the discretion of the radiologist, with use of a unit (model 700; GE Medical systems) with 7.5–12-MHz transducers and was performed solely by the radiologist. Owing to the paucity of literature on the role of US in breast thickening, no uniform standard for performing US existed during this period. When mammographic abnormalities that warranted biopsy (eg, spiculated mass, suspicious calcifications) were discovered, US generally was not performed. Breast US impressions were categorized prospectively as negative, simple cyst, complex cyst, or solid mass. We (J.S.K., M.A.H.) evaluated retrospectively for the presence of an abnormality on the additional mammographic views in patients who ultimately proved to have a cancer. Specifically, we evaluated with consensus whether the additional views depicted the abnormality to better advantage or were merely additional views of a finding already seen well on the standard two-view (craniocaudal and mediolateral oblique) mammograms.
For the patients who underwent biopsy, the method of biopsy was recorded. Review of pathology reports of all cases subjected to biopsy was conducted (J.S.K.).
For those patients who did not undergo biopsy of the area of breast thickening, initial and long-term follow-up studies (either clinical breast examination or mammography) until January 15, 2001 were assessed. In eight cases, the patient’s physician did not explicitly record the breast physical examination findings or request mammography during our follow-up period. In these cases, follow-up was obtained by researching all patient-visit notes to determine that the patient had not developed interval breast carcinoma at the site of thickening. These patients were considered disease free for our study period.
True-positive mammograms were defined as BI-RADS category 4 or 5 and a histologic diagnosis of cancer. False-positive mammograms were defined as BI-RADS category 4 or 5 examinations with a benign biopsy result. False-negative mammograms were defined as BI-RAD category 1, 2, or 3 with a cancer diagnosis per histologic examination. True-negative mammograms were a BI-RADS 1, 2, or 3 assessment with a benign histologic result or stability at follow-up.
True-positive US scans were defined as a solid mass with a histologic diagnosis of cancer. False-positive US scans were a solid mass with a benign biopsy result or stability at follow-up. False-negative US scans were negative, cyst, or complex cyst findings and a histologic diagnosis of cancer. True-negative US scans were negative, cyst, or complex cyst findings and a benign histologic result or stability at follow-up.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Mammography
In 25 (20%) of the 123 cases, mammographic abnormalities were noted at the site of breast thickening. These abnormalities included two cases of suspicious masses with irregular or spiculated borders with or without associated architectural distortion, 10 cases of indeterminate masses with nonspiculated borders, two cases of architectural distortion, six cases of presumed postsurgical distortion, and five cases of asymmetric breast density. In addition to the standard two-view mammograms, additional mammographic views were obtained in 87 (71%) of the 123 cases. A lateral view alone was obtained in 37 cases, and a lateral and a spot compression view were obtained in 49 cases. A spot view alone was obtained in one case.
Breast US
Breast US of the area of breast thickening was performed in 77 (63%) of the 123 cases. Breast US scans were normal in 60 (78%) of the 77 cases. In the remaining 17 cases, the following entities were found: simple cysts (n = 11), complex cyst (n = 1), cysts and a single solid mass (n = 1), and solid masses (n = 4).
Of the 25 cases that revealed a mammographic abnormality, 14 underwent US examination. Breast US was performed in one of the two cases of suspicious spiculated masses and revealed a solid mass. Breast US was performed in both cases of architectural distortion, and both cases revealed solid masses. Breast US was performed in eight of the 10 cases of masses with nonspiculated borders and revealed simple cysts in five cases, a complex cyst in one case, normal findings in one case, and cysts and a solid mass in another case. Breast US was not performed in two of the 10 cases of nonspiculated masses owing to the long-term stability of the masses at mammography. A breast US scan was obtained in two of the five cases of benign-appearing asymmetric breast tissue and was normal in one case and revealed cysts in the second. Breast US scan was obtained in one of the six cases of benign-appearing presumed postsurgical distortion and was normal.
Outcome
In 12 (10%) of the 123 cases, the patient underwent biopsy of the area of breast thickening. In six cases, the diagnosis was benign: localized fibrosis, acellular scar, fibrocystic change, fibroadenoma, acellular fine-needle aspirate, and simple cyst fluid. Six (5%) of the 123 cases of breast thickening in six patients were found to have breast carcinoma at the site of thickening. Cancer types were the following: invasive ductal carcinoma (n = 2), invasive lobular carcinoma with lobular carcinoma in situ (n = 1), mixed invasive lobular and ductal carcinoma (n = 1), ductal carcinoma in situ (n = 1), and inflammatory carcinoma (n = 1). Mean pathologic size of the measurable invasive cancers was 2.2 cm (range, 0.8–4.0 cm). Four (80%) of five patients had lymph nodes negative for axillary metastases (the lymph node status of one patient is unknown). The following methods were used to obtain a tissue diagnosis: stereotactic core biopsy (n = 1), US-guided core biopsy (n = 1), surgical open biopsy (n = 6), skin biopsy (n = 1), and fine-needle aspiration (n = 3).
Mammographic findings in the six cases of breast carcinoma revealed spiculated masses with architectural distortion (n = 2), architectural distortion (n = 1), microcalcifications and presumed postsurgical distortion at a site of prior lumpectomy (n = 1), presumed postsurgical distortion at a site of prior benign biopsy (n = 1), and negative imaging features (n = 1). Three cases were categorized as BI-RADS 4 or 5; two cases, as BI-RADS 3; and one case, as BI-RADS 1. One of the BI-RADS 3 cases demonstrated postsurgical distortion and microcalcifications. Later, surgical biopsy was performed of the microcalcifications, and they were shown to represent a 2-cm span of ductal carcinoma in situ. In this case, it is likely that the thickening was not related to the ductal carcinoma in situ but rather to postsurgical scarring. The other BI-RADS 3 case and the BI-RADS 1 case occurred in patients with a history of prior benign biopsy (range, 1–6 prior biopsies), thereby confounding the interpretation of the architectural distortion. There were no false-negative findings in women with fatty density on the mammogram. Hence, the sensitivity of mammography alone for invasive breast carcinoma at the time of initial imaging was 60% (three of five cases). The specificity of mammography at initial imaging was 94% (102 of 108 cases), and the negative predictive value was 97% (102 of 105 cases) (excludes nine negative cases without follow-up).
Additional mammographic views were obtained in four of the six cancer cases. In one case, the mammographic findings were difficult to see on the initial two-view mammogram (Figure). The additional mammographic views demonstrated suspicious architectural distortion (BI-RADS category 5). In three cases, the findings were seen on the initial two-view mammogram but were better visualized on the additional views. Additional views were not obtained in the negative case nor in the case of presumed postsurgical distortion.
|
|
|
|
A breast cancer diagnosis was rendered at the time of presentation in three cases. Diagnosis of breast cancer occurred at 9–18 months in the remaining three cases, all of which had a negative or probably benign mammographic interpretation at the time of initial presentation.
Mammographic and/or breast physical examination follow-up in 102 (92%) of 111 cases not undergoing tissue diagnosis showed the patients to be disease free. Disease-free follow-up intervals were as follows: less than 12 months (12 cases), 12–24 months (29 cases), and more than 24 months (57 cases). Four patients were seen the same day of their technologist clinical breast examination by their referring physician or by a breast surgeon, were considered to have a normal physical examination, and were considered benign cases. Nine (7%) of the 123 cases were lost to follow-up.
The mean time to initial documented follow-up with either mammography or clinical breast physical examination by the referring physician was 15 months (median, 14 months) after the initial imaging work-up. The mean duration of follow-up with clinical breast examination, mammography, or information in the patient’s chart at her physician’s office was 19 months (median, 23 months).
Demographics
The six patients in whom breast carcinoma was detected ranged in age from 36 to 72 years (mean, 53 years). Five (83%) of six patients were postmenopausal, and three (60%) of the five postmenopausal patients were receiving hormonal therapy. One patient had a family history of breast carcinoma, and another patient had a personal history of breast carcinoma in the ipsilateral breast. The remaining four patients had no known risk factors for breast carcinoma.
One hundred six (86%) of the 123 cases of breast thickening occurred in women who were either premenopausal (78 [73%] of 106 cases) or were receiving hormonal replacement therapy (28 [26%] of 106 cases). Seventeen (14%) of 123 cases of breast thickening occurred in women who were postmenopausal and not receiving hormonal replacement. Eighty-six (70%) cases occurred in women who had no personal or family history of breast cancer. Sixteen (13%) cases occurred in patients who had a personal history of breast cancer, and 21 (17%) cases occurred in patients who had a family history of a first-degree relative with breast cancer.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Breast thickening is often a vague breast physical examination finding. Even with more discrete breast abnormalities, there is still a large interobserver variability in detection rates and interpretation. As shown by Boyd et al (1), there was only a 25% interobserver agreement among four experienced breast surgeons that a physical breast abnormality was present at the clinical breast examination in 100 consecutive symptomatic women. Although biopsy was recommended in 28%–39% of these cases, all surgeons agreed that a biopsy should be performed in only 17% of the cases. The concordance for management and the need for biopsy were higher in the 15 patients who ultimately proved to have breast carcinoma. Specifically, all surgeons recommended that a tissue diagnosis be obtained in 73% of the subsequently diagnosed breast cancer cases (1). Given the interobserver variability even among specialists in the field of breast disease and the vagueness of the symptom of breast thickening, it is not surprising that the management of breast thickening is not uniform.
Breast physical examination findings are neither sensitive nor specific for breast carcinoma. Clinical breast examination is even less sensitive in women younger than 50 years. Women younger than 50 years who are found to have cancer often have tumors that are palpable but "ill-defined" (10). Although 83% of the patients with breast cancer in our study were older than 50 years, 64% of patients with breast thickening were younger than 50 years. Our patient population is similar to that in the study by Lannin et al (10), in which 65% of visits for breast symptoms were by women younger than 50 years. In their 10-year retrospective analysis of symptomatic female patients presenting to primary care practitioners, Barton et al (9) found that 21% of patients had thickening noted at breast physical examination, similar to the 17% of patients with thickening in our practice. Hence, our study population appears representative of general clinical studies.
Breast thickening can be a manifestation of breast carcinoma, particularly the infiltrating lobular type (11,12), although it has also been seen in other types of breast cancer including inflammatory breast cancer (13) and Paget disease (14). Breast thickening has also been documented as the sole finding in postsurgical change (15), postirradiation change (15), lymphoma (16), and in combination with other findings in breast infection (17).
Owing to the lack of sensitivity and specificity of the clinical breast examination in patients with breast thickening, we were especially interested in the imaging evaluation of these women. We attempted to provide as much uniformity as possible in our case selection of breast thickening, using the prospective opinion of a group of experienced breast examiners. We found that the sensitivity of mammography alone for the detection of invasive breast carcinoma was 60%, specificity was 94%, and the negative predictive value was 97%. Additional diagnostic mammographic views in the cancer cases proved helpful in determining the presence of a mammographic abnormality in all cases, including clearly defining a questionable two-view finding as a suspicious finding in one case.
The role of breast US in the evaluation of patients with breast thickening and invasive lobular breast carcinoma has been previously described. Butler et al (18) demonstrated the use of breast US in detecting tumor in 71 (87.7%) of 81 cases of mammographically subtle or invisible lesions. Many of these cancers were diagnosed in patients with palpable findings (26% of cases). Whereas 63% of patients in our study underwent breast US, only 33% of breast cancer cases had breast US scans at the time of initial imaging evaluation. The sensitivity of US alone was 100% (two of two cases), specificity was 96%, and negative predictive value was 100%.
The combined negative predictive value of negative mammographic and US findings in our patients with breast thickening was 100%. This value is identical to that in the recent report by Dennis et al (19), who reviewed the outcome of women with negative US scans and mammograms who had different types of palpable findings at presentation. Our study is different in that we reviewed the outcome of a consecutive cohort of symptomatic women with thickening during their usual clinical care, and all US examinations were performed by experienced breast radiologists. The applicability of our results to other methods of breast US practice is uncertain. Given the small number of cancer patients in our cohort who underwent breast US examination, we do not believe that the role of US has been adequately evaluated to form a definitive recommendation against biopsy when mammographic and US findings are negative. It must be remembered that cancer is unusual (1%–5%) in women with abnormal physical examination findings, therefore a very large prospective study would be needed to determine the real risk of cancer in the patients with negative imaging findings (2,9).
The diagnosis of breast cancer and the evaluation of the symptom of breast thickening were difficult in the patients with cancer. Although the frequency of breast cancers in our study was only 5% (six of 123 cases), the diagnosis was not immediately made in 50% of patients owing to several confounding variables. Specifically, false-negative examination findings occurred in patients with multiple prior breast biopsies in the area of thickening and in one patient who had long-standing stable breast thickening. Postsurgical change often causes breast scarring, architectural distortion, fat necrosis, and skin changes. In the context of these mammographic findings and the vague physical examination finding of breast thickening, it may be difficult for the mammographer to distinguish whether the physical findings in question relate to postsurgical scarring versus a breast malignancy. Breast physical examination is also limited in patients after biopsy for benign or malignant causes. In view of our results, any physical examination findings in a patient with a prior biopsy at the site of thickening should still be viewed with some caution even in the presence of imaging findings compatible with those of a prior biopsy.
For the practicing radiologist, our study findings support the standard use of diagnostic mammography for patients with palpable thickening. Views incremental to the standard craniocaudal and mediolateral oblique projections are recommended. We used the lateral and spot views (with or without magnification) to advantage. Breast US also seems prudent in most situations, unless definite suspicious findings are found on diagnostic mammograms. Since the false-negative rate of mammography is low in women with fat density on mammograms, the incremental use of US when the mammogram is normal in such women is uncertain.
The time to initial follow-up with either mammography or clinical breast physical examination by the referring physician was about 1 year after the initial imaging work-up. Even though some national guidelines call for follow-up clinical examinations in several months (3), our findings suggest that the referring physicians equated a negative mammogram and a negative breast US scan in the setting of a vague symptom of breast thickening to be a negative outcome, and the patients reverted to annual screening. This clinical opinion, although correct in most cases, is not justified by our data since the frequency of cancer in those cases with a negative mammogram was 2.4% (three of 123 cases). Because a typical clinician may personally see only a few cases of breast thickening per year, his or her experience would suggest that cancer is of no issue when imaging findings are negative.
The real clinical question is what to do with the patient with palpable thickening and negative mammograms and US scans. Clinical and surgical judgment in this setting remains essential. Women with negative mammograms and negative US scans are at low risk for cancer but should, in our opinion, be followed up at short-term intervals with physical examination and imaging if biopsy is not elected clinically. Previous recommendations for clinical follow-up in 3 months (3,19) seem reasonable and may prevent delayed diagnosis of malignancy. Short-term imaging follow-up at 6 months, similar to the widely practiced management of cases with BI-RADS category 3 (probably benign) mammograms, also seems prudent for persistent breast thickening with negative imaging findings if nonsurgical management is elected clinically.
|
FOOTNOTES |
|---|
Author contributions: Guarantors of integrity of entire study, M.A.H., J.S.K.; study concepts and design, M.A.H., J.S.K.; literature research, J.S.K., M.A.H.; clinical studies, M.A.H., J.S.K.; data acquisition, J.S.K.; data analysis/interpretation, all authors; statistical analysis, M.A.H., J.S.K.; manuscript preparation, all authors; manuscript definition of intellectual content and editing, M.A.H., J.S.K., M.A.R.; manuscript revision/review and final version approval, all authors.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  E.-K. Kim, K. H. Ko, K. K. Oh, J. Y. Kwak, J. K. You, M. J. Kim, and B.-W. Park Clinical Application of the BI-RADS Final Assessment to Breast Sonography in Conjunction with Mammography Am. J. Roentgenol., May 1, 2008; 190(5): 1209 - 1215. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 W. A. Berg, J. D. Blume, J. B. Cormack, and E. B. Mendelson Operator Dependence of Physician-performed Whole-Breast US: Lesion Detection and Characterization. Radiology, November 1, 2006; 241(2): 355 - 365. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Y. Kwak, E.-K. Kim, H.-L. Park, J.-Y. Kim, and K. K. Oh Application of the breast imaging reporting and data system final assessment system in sonography of palpable breast lesions and reconsideration of the modified triple test. J. Ultrasound Med., October 1, 2006; 25(10): 1255 - 1261. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
O. Graf, T. H. Helbich, M. H. Fuchsjaeger, G. Hopf, M. Morgun, C. Graf, R. Mallek, and E. A. Sickles Follow-up of Palpable Circumscribed Noncalcified Solid Breast Masses at Mammography and US: Can Biopsy Be Averted? Radiology, December 1, 2004; 233(3): 850 - 856. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. K. Shetty, Y. P. Shah, and R. S. Sharman Prospective Evaluation of the Value of Combined Mammographic and Sonographic Assessment in Patients With Palpable Abnormalities of the Breast J. Ultrasound Med., March 1, 2003; 22(3): 263 - 268. [Abstract] [Full Text] [PDF]
|
|
Read all eLetters
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| RADIOLOGY | RADIOGRAPHICS | RSNA JOURNALS ONLINE |
