| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Breast Imaging |
1 From the Department of Radiology, University of British Columbia, 750 W Broadway, Suite 505, Vancouver, British Columbia, Canada V5Z 1H4. From the 2000 RSNA scientific assembly. Received January 8, 2001; revision requested February 15; final revision received February 17, 2003; accepted March 10. Address correspondence to P.B.G.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODS RESULTSDISCUSSIONREFERENCES |
|---|
MATERIALS AND METHODS: For 1,070 consecutive patients with breast lesions diagnosed as fibroadenoma at FNAB, three measurements of each mass were performed at the initial visit when FNAB was performed and at each follow-up ultrasonographic examination. Changes in volumes were calculated. At one or more visits, 194 masses showed an increase in volume. Nonfibroadenomas were excluded, and the data were used for comparison. Percentiles (90th and 95th) for percentage change in volume per month were used to determine acceptable changes in dimensions (specifically, greatest anteroposterior, parallel-to-skin, and perpendicular-to-skin dimensions).
RESULTS: There were 567 interval measurements of 179 masses in 173 patients younger than 50 years and 50 measurements of 15 masses in 14 patients 50 years or older at the time of FNAB. The 95th percentile for percentage change in volume per month was approximately 16% for patients younger than 50 years; the 90th percentile was approximately 13% for patients 50 years or older. The 95th percentile mean change in dimension in a 6-month interval for those younger than 50 years was 20%; the 90th percentile change for those 50 years or older was also 20%. All excised masses with slower growth proved benign at histologic examination.
CONCLUSION: Solid breast masses diagnosed as fibroadenomas at FNAB may be safely followed up if volume growth rate is less than 16% per month in those younger than 50 years and less than 13% per month in those 50 years or older. Acceptable mean change in dimension for a 6-month interval is 20% for all ages.
© RSNA, 2003
Index terms: Breast, biopsy, 00.1261 • Breast neoplasms, diagnosis, 00.30 • Breast neoplasms, US, 00.1298 • Fibroadenoma, 00.311
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTSDISCUSSIONREFERENCES |
|---|
Criteria for probably benign lesions defined for US (11,12) and other modalities such as magnetic resonance (13) and radionuclide imaging (14,15) have been demonstrated to be of value in contributing to the differentiation of benign from malignant solid masses without tissue acquisition. The diagnostic standard procedure is surgical excision, but even preoperative image-guided localization does not ensure successful removal of nonpalpable lesions in 100% of cases, and surgical failure rates as high as 17% have been reported (16).
Cytologic examination of specimens obtained at fine-needle aspiration biopsy (FNAB) yields some false-negative results but can increase diagnostic confidence (17–19). Because it is possible that a cytologic diagnosis may be incorrect, that a fibroadenoma may progress (though rarely) to a phyllodes tumor (20), or that a cancer may (rarely) develop in a fibroadenoma (21), imaging follow-up is necessary, even after FNAB yields benign cytologic findings, and lesions that increase in size are of concern. Masses with a histologic diagnosis of fibroadenoma at large-core needle biopsy also cause anxiety when they enlarge, and most clinicians recommend excision if this occurs (22–24). We believe that when breast lesions diagnosed as probably benign lesions at US are also diagnosed as fibroadenoma at FNAB, the likelihood of malignancy is extremely low, and, therefore, even if such lesions enlarge, they rarely require excision because of the possibility of malignancy.
US-guided FNAB is minimally invasive and can be performed at the time of the first examination for masses that are visible at US. Although 14-gauge core biopsy and even 11-gauge directional vacuum-assisted biopsies are performed more frequently in other institutions, 20-gauge FNAB is performed frequently at our institution because it is inexpensive and because expert cytologic interpretation is readily available. The purpose of our study was to determine what growth rate in solid breast lesions diagnosed as fibroadenoma at FNAB is acceptable before histologic diagnosis is recommended.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTSDISCUSSIONREFERENCES |
|---|
Study subjects were all 1,070 women who were given a US diagnosis of probably benign solid breast lesion according to the criteria defined by Stavros et al (11) and whose masses, at FNAB performed by one of the authors (P.B.G.), had cytologic features that were consistent with a diagnosis of fibroadenoma. No Breast Imaging Reporting and Data System (BI-RADS) assessments were performed because a report of the BI-RADS categorization system was published after this study was already in progress. In retrospect, all patients had, by definition, BI-RADS category 2, 3, or 4 lesions. Patients for whom biopsy was considered unnecessary (ie, those with BI-RADS category 2 lesions) and for whom follow-up alone is usually recommended (ie, those with BI-RADS category 3 lesions) underwent FNAB either at their own or their referring physician’s request. Any patient with a mass that was considered to be BI-RADS category 4 at combined mammographic and US assessment was advised to undergo surgical excision, even if the results of cytologic examination indicated fibroadenoma. During most of the period when these examinations were performed, core biopsy was not available in our community. Currently, if cytologic examination results are discordant with the radiologic assessment, there is the option to perform a large-core needle biopsy rather than proceed to surgical excision.
For premenopausal women, the phase of the menstrual cycle was not recorded at each visit, but if a mass had increased in size, proximity to the patient’s next menses was considered before it was determined whether follow-up was appropriate. The possibility that fluctuations in size might be related to the menstrual cycle was included in the explanation provided to the patient in the process of obtaining informed consent for biopsy.
Patients were routinely asked to return at 6 months and at 1, 2, and 3 years after FNAB for monitoring, and, if an increase in dimension of the lesion was recorded, additional follow-up at 6-month intervals was recommended. Three lesion dimensions were measured at US at each return visit. All initial measurements and almost all follow-up measurements were performed by P.B.G., although a very small number were performed by one other radiologist. Inter- and intraobserver reliability were not tested. All follow-up measurements were performed by using the same US unit (DRF 250; Diasonics, Milpitas, Calif) (which did not have focal zone adjustment) and the same transducer that were used for the initial examination. For each mass, the greatest dimension parallel to the skin (d1), whether in the standard transverse or sagittal planes or in an oblique direction either radial or antiradial to the nipple, was recorded. The second dimension measured (d2) was orthogonal to d1, and the third dimension measured was the anteroposterior dimension (d3). The volume (V) of each mass was calculated as follows: V = 4/3 
x d1/2 x d2/2 x d3/2 = 0.523d1d2d3.
At the time of this writing, 194 masses in 187 patients had shown an increase in dimensions between visits. For these masses, the change in volume was calculated for each pair of visits (F.A.G.). To compensate for the original range of sizes of the lesions, the percentage change in volume from the previous visit (%V/t) was determined as %V/t = (V1-V2)/V1 x 100/t, where V1 is the volume at the previous visit, V2 is the volume at the subsequent visit, and t is time in months. Patients were asked to return 6 months later for repeat evaluation. Patients vary considerably in their compliance with recommendations for follow-up appointments, and, to allow for the variability of time intervals between visits, the percentage change in volume over time was calculated as the percentage change in volume per month.
Each patient was told that her mass had increased slightly in size and that, although acceptable rates of growth have not been reported in the literature, fibroadenomas are known to fluctuate in size with, for example, hormonal cycles. Each patient was told that she could obtain a surgical consultation, but that it was probably safe to follow up the mass with serial US examinations.
If the mass was palpable, the patient was instructed to check it by performing monthly breast self-examination and to return earlier than 6 months if she thought that the mass was enlarging. If the mass was not palpable, and the patient wanted to undergo follow-up, she was shown exactly where the mass was at the time US was being performed and told to monitor that area monthly with breast self-examination and return earlier than 6 months if she thought she felt a mass emerging at that site.
Some patients did see a surgeon, and many of them then chose to follow up the mass after being reassured by the surgeon. Surgical pathology reports were obtained for those patients who underwent mass removal, and data from those reports were used in this study.
Endpoints, in order of power to confirm the FNAB diagnosis, were surgical diagnosis, a decrease in size following the increase in size, and stability in size for at least 24 months after the increase in size.
Statistical Analysis
Patients were grouped by age at diagnosis into those younger than 50 years and those 50 years or older. For patients younger than 50 years, we regarded the 95th percentile of observed volume increases as indicating what degree of growth was safe and acceptable. The 95th percentile was determined with two methods: by rank and by mean plus t(0.95) times the SD, where t is the t distribution. For those patients 50 years or older, the 90th percentile was used instead to err on the side of greater caution in this age group because of their increasing risk of cancer. The CIs around the 90th and 95th percentiles were determined (25). To present the data in more practical terms for the clinical radiologist, the results were also used to determine the 90th and 95th percentiles for change in dimensions for a 6-month interval.
To evaluate whether multiple masses had a growth rate different from that of single masses, measurements of change for multiple masses were compared with measurements for single masses by using a two-sided t test, with a P value of less than or equal to .05 considered to indicate a statistically significant difference.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTSDISCUSSIONREFERENCES |
|---|
|
Five masses were given a surgical diagnosis other than fibroadenoma: Two were benign phyllodes tumors, two were stromal fibroses, and one was a hamartoma. One of the phyllodes tumors decreased in size between the first and second visits but showed a volume increase of 30% per month (ie, an increase in dimensions from 2.4 x 1.3 x 1.7 to 3.1 x 2.7 x 2.2 cm, with an average increase in each dimension of more than 8 mm) between the second and third visits. The percentage increase in volume per month was 5% for the two stromal fibroses and 7% for the hamartoma. There were no carcinomas.
Percentage change in volume per month for masses in patients younger than 50 years is plotted against percentile in Figure 1, and percentage change in volume per month for masses in patients 50 years or older is plotted against percentile in Figure 2. For both groups, 40% of volume changes represented decreases in size.
|
|
|
For practical purposes, a change of 0.2 cm in all dimensions during a 6-month interval in a probably benign solid breast lesion with a FNAB-based diagnosis of fibroadenoma and initial measurements of approximately 1.0 cm in each direction is acceptable for women of all ages.
The mean change in volume per month among multiple masses (30 measurements) was 5.8%. The mean change in volume per month for single masses (587 measurements) was 2.8% (P > .2).
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTSDISCUSSIONREFERENCES |
|---|
The rationale for these recommendations for follow-up rather than biopsy of these lesions is partially financial, and Sickles acknowledges a small but finite risk of false-negative results. Eight hundred forty-two (26.4%) of the probably benign lesions in that series were masses, and 12 of those were subsequently diagnosed as cancer, for a positive predictive value of 2% for masses that appeared benign on the basis of only mammographic criteria. Of the 17 cancers misdiagnosed with Sickles’ criteria at initial work-up, 15 were subsequently correctly diagnosed after mammographic change was identified, and all 17 were stage 0 or 1 cancers, so their prognoses should have been similar to those of cancers diagnosed promptly after initial work-up. However, for some patients and their clinicians, the risk of false-negative results, even without a change in prognosis, is unacceptably high, especially in our location, where a histologic diagnosis can be obtained painlessly and inexpensively.
Varas et al (9) achieved a positive predictive value of 1.7% for probably benign nonpalpable lesions, but their methods were different from those of Sickles: Varas et al did not have access to magnification mammography. Fifty-four percent of the lesions in their series were masses, and 1.4% of those were missed cancers, all of which were eventually diagnosed on the basis of interval growth within 2 years of initial observation.
In addition to the mammographic criteria, US characteristics have proved helpful to the radiologist in differentiating benign from malignant solid masses (11,12). As more sophisticated imaging modalities are developed, the level of diagnostic confidence may increase, but possibly not in proportion to the additional time and cost required. This will vary among sites, depending on the relative costs of needle biopsy and imaging (28,29). In our institution, for patients with US and mammographic diagnosis of a probably benign lesion, obtaining a histologic diagnosis with FNAB is a practical compromise if the patient requests it. US-guided FNAB is used in addition to, rather than instead of, surveillance.
Performing a cytologic or histologic examination is prudent for patients who are at high risk of having breast cancer and for whom additional reassurance is invaluable for peace of mind. In our facility, US-guided FNAB is inexpensive (with a total laboratory cost, including technical, overhead, and pathology fees, of $48.71 in Canadian dollars [$32.00 in U.S. dollars at current exchange rates]) and is offered to any patient with a newly detected solid breast mass. The patient is told, in the appropriate circumstance, that the mass appears to be benign at imaging, and, when possible, an estimate of the level of certainty is provided. Often, FNAB is performed during her first visit if she requests it. In situations in which noncompliance with surveillance recommendations is likely, a histologic diagnosis can reveal cancer in a mass that may appear benign according to imaging criteria (30). At facilities where the cost of FNAB is much higher than that at our facility, the peace of mind provided by this procedure may not justify the additional costs, particularly in view of evidence suggesting that there is no change in prognosis for the few false-negative cases despite a delay in treatment (7,9).
Once the diagnosis of fibroadenoma has been confirmed cytologically, it is important to remember that fibroadenomas in premenopausal women typically fluctuate in size (31) and that fibroadenomas can grow in postmenopausal women whether or not they are receiving hormone replacement therapy (32,33). Data on hormone replacement therapy use were not rigorously collected in our study, and therefore its impact is uncertain.
Previous reports uniformly suggest that histologic diagnosis be recommended for any mass designated as probably benign that increases in size at follow-up (22–24), and, to our knowledge, no author has yet attempted to define a "safe" range of growth. Nevertheless, only 48% of the 31 masses that increased in size in the series of Hermann et al (23) proved to be cancer. In Sickles’ series (1), only 15 cancers (11.5%) were found at the 131 biopsies that were prompted by interval size change; the number of these that were masses is not given. Varas et al (9) recommended biopsy for 16 patients because of interval enlargement, but the number of masses in this group is not given. Nine (56%) of the 16 patients had cancer, and four of the nine patients had cancer diagnosed in benign-appearing masses. Helvie et al (10) observed progressive mammographic change in 10 of 126 patients, but the number of masses in this group was not specified. Only one (10%) of the 10 patients was later given a diagnosis of cancer, and the original mammographic appearance of the lesion in this patient was not that of a mass but that of an asymmetric density. If any of the lesions that grew in the series of Helvie et al (10) were masses, then none was a cancer.
It would be practical to define the amount of growth that is acceptable for managing breast masses with watchful waiting before histologic diagnosis is recommended. The results of this study of 194 masses that enlarged after a cytologic diagnosis of fibroadenoma indicate that fibroadenomas may safely be followed up after an increase in volume of up to approximately 15% per month, the equivalent of an increase of 20% in all three dimensions in a 6-month period.
It is of some interest that the results in women younger than 50 years changed as more data points were collected over time. At the time of preliminary analysis of these data (34), the 95th percentile change in volume per month was 20% for patients younger than 50 years, which is equivalent to a 30% change in dimension in a 6-month interval. The 95th percentile in the current report may be lower because patients who had larger masses and patients with faster lesion growth rates were more likely to have undergone surgery and thus did not contribute growth data over the long-term follow-up period. This difference between patients with surgical diagnoses and patients without is confirmed by the difference in 95th percentiles: higher than 35% among patients younger than 50 years with a surgical diagnosis of fibroadenoma compared with 10% among patients younger than 50 years without a surgical diagnosis. Therefore, the amount of growth that can be safely followed up may be greater than that described here, but this will need to be confirmed with additional research.
Large-core needle biopsy may have a lower incidence, possibly even approaching zero, of yielding false-negative results than FNAB. The growth limits suggested in the current report should therefore also be applicable to masses sampled with large-core needle biopsy.
One might postulate that these growth limits might even be applied to lesions that are probably benign according to imaging criteria but have not been sampled at percutaneous biopsy or at FNAB. The authors advise against using these growth limits unless results of a FNAB or a percutaneous needle biopsy have provided further substantiation of a benign diagnosis. For lesions designated as BI-RADS category 3, the risk of malignancy is less than 2%, and if a benign-appearing mass enlarges at follow-up, we would suggest that the mass has converted to a BI-RADS category 4 lesion, for which percutaneous biopsy or FNAB would be appropriate. In such a case, if the biopsy results are consistent with a diagnosis of fibroadenoma and the rate of growth is within the range described here as "acceptable," then surgical biopsy could be averted. This would substantially reduce costs.
If results of further studies support these guidelines, the potential exists to further increase the true-positive rate for histologic diagnosis of mammographically detected nonpalpable masses without compromising accuracy, and this would decrease the cost of screening for breast cancer.
Furthermore, with these guidelines, lesions for which cytologic results were benign but not specific for fibroadenoma might also safely be managed expectantly. We plan to investigate such masses in the future.
In summary, solid breast masses diagnosed as fibroadenoma at FNAB or percutaneous needle biopsy may safely be followed up if their volume growth rate is less than 15% per month. Acceptable mean change in each of three dimensions (ie, anteroposterior, parallel-to-skin, and perpendicular-to-skin dimensions) in a 6-month interval is 20% for all ages.
|
FOOTNOTES |
|---|
Author contributions: Guarantor of integrity of entire study, P.B.G.; study concepts, P.B.G., F.A.G., L.L.; study design, P.B.G., F.A.G.; literature research, P.B.G.; clinical studies and data acquisition, P.B.G., L.L.; data analysis/interpretation, P.B.G., F.A.G.; statistical analysis, F.A.G.; manuscript preparation, definition of intellectual content, and editing, P.B.G., F.A.G., L.L.; manuscript revision/review, P.B.G., F.A.G.; manuscript final version approval, P.B.G., F.A.G., L.L.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  A. Vade, V. S. Lafita, K. A. Ward, J. E. Lim-Dunham, and D. Bova Role of Breast Sonography in Imaging of Adolescents with Palpable Solid Breast Masses Am. J. Roentgenol., September 1, 2008; 191(3): 659 - 663. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Irshad, S. J. Ackerman, T. L. Pope, C. K. Moses, T. Rumboldt, and B. Panzegrau Rare Breast Lesions: Correlation of Imaging and Histologic Features with WHO Classification RadioGraphics, September 1, 2008; 28(5): 1399 - 1414. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Sklair-Levy, T. Sella, T. Alweiss, I. Craciun, E. Libson, and B. Mally Incidence and Management of Complex Fibroadenomas Am. J. Roentgenol., January 1, 2008; 190(1): 214 - 218. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 O. Graf, T. H. Helbich, G. Hopf, C. Graf, and E. A. Sickles Probably Benign Breast Masses at US: Is Follow-up an Acceptable Alternative to Biopsy? Radiology, July 1, 2007; 244(1): 87 - 93. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Costantini, P. Belli, R. Lombardi, G. Franceschini, A. Mule, and L. Bonomo Characterization of solid breast masses: use of the sonographic breast imaging reporting and data system lexicon. J. Ultrasound Med., May 1, 2006; 25(5): 649 - 659. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
F. M. Hall and P. Gordon Acceptable Rates of Growth in Fibroadenomas Diagnosed with Fine-Needle Aspiration Biopsy [letter] * Dr Gordon and colleagues respond: Radiology, May 1, 2004; 231(2): 602 - 602. [Full Text] [PDF]
|
|
Read all eLetters
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| RADIOLOGY | RADIOGRAPHICS | RSNA JOURNALS ONLINE |
