| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Breast Imaging |
1 From the Advisory Board and Collaborative Group of Readers of the Screening Program, Public Health Office of the Health Department of the Valencia Community, Valencia, Spain. Received January 12, 2000; revision requested February 12; final revision received November 7; accepted November 22. Address correspondence to I.V., Department of Radiology, Dr Peset University Hospital, C/Gaspar Aguilar 90, 46017 Valencia, Spain (e-mail: vizcaino_isi@gva.es).
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODS RESULTSDISCUSSIONREFERENCES |
|---|
MATERIALS AND METHODS: Of 13,790 women aged 45–65 years who underwent first-round screening, 795 (5.8%) underwent short-term mammographic follow-up (every 6 months for 2 years) of nonpalpable probably benign lesions (eg, masses, focal asymmetric densities, and calcifications) previously assessed at an additional imaging evaluation, including ultrasonography. When no changes were found at short-term mammographic follow-up, women were assigned to the 2-year screening interval. Needle localization and surgical biopsy were performed when the lesion progressed (was enlarged or had an increased number or size of calcifications or modification of their initial characteristics). The effectiveness of this approach was evaluated with statistical analysis.
RESULTS: Of 795 lesions, 788 (99%) remained stable, and seven (1%) had changes prompting surgical biopsy. Two cancers (0.3%), one microinvasive intraductal carcinoma and one 7-mm invasive ductal carcinoma without positive nodes, were found. Four of the five benign histologic results were probably benign calcifications with progression at short-term follow-up. The sensitivity, specificity, accuracy, and positive and negative predictive values were 100%, 99%, 99%, 29%, and 100%, respectively.
CONCLUSION: The benign nature of most nonpalpable probably benign lesions can be typified with short-term mammographic follow-up. This approach permitted identification of a few low-stage carcinomas, but progression in the probably benign calcifications was usually unrelated to malignancy.
Index terms: Breast neoplasms, diagnosis, 00.112, 00.30 • Breast radiography, utilization, 00.112 • Cancer screening, 00.112, 00.30
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTSDISCUSSIONREFERENCES |
|---|
On the basis of the preceding recommendations and previous reports (2–10), the advisory board of the 2-year population-based Breast Cancer Screening Program in the Valencia Community (Spain) included a short-term mammographic follow-up protocol (every 6 months between rounds) for the assessment of nonpalpable probably benign lesions (11). In spite of scientific evidence supporting this approach, there are limited data regarding the percentage of malignancies found when probably benign lesions have progressed at surveillance (4,7–9). Also, a variable interpretation of benign and probably benign radiologic concepts is commonly accepted, and overuse of short-term mammographic follow-up could occur (5). Therefore, a retrospective review of nonpalpable probably benign mammographic lesions prospectively diagnosed at the beginning of our program was performed to evaluate the use of short-term follow-up in a 2-year screening program.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTSDISCUSSIONREFERENCES |
|---|
One nonradiologist physician (L.G.), one radiographer, one computer operator, one auxiliary nurse, and one secretary (all full-time employees) were employed by the screening facility. The physician was responsible for the daily activities of the screening facility and technically supervised all mammography, recommended additional projections, and performed physical examination online (at the time of screening), in accordance with patient symptoms or mammographic findings. He performed the official reading and was specially trained in mammography to perform the first reading of the mammograms (11,12). A second reader, a radiologist (L.A.), read the mammograms on a part-time basis at the screening facility, since he is a member of the radiology department at the referral hospital. The second reader performed the US examinations at the referral hospital by using a high-frequency US unit (Sonolayer SSA-250A; Toshiba Medical Systems Europe, Zoetermeer, the Netherlands) with a 7.5-MHz linear-array transducer. The US protocol included transverse and longitudinal real-time imaging, with representative hard-copy images. Patient mammograms and medical history were provided to the second reader for the US examination and report.
All mammograms were independently double-read. The readers described breast composition and mammographic findings (eg, fatty or dense breast tissue, mass, calcification, architectural distortion, asymmetric densities, skin or nipple retraction, skin or trabecular thickening, axillary adenopathies, and associated findings), which were classified into assessment categories (2,11). Each reader independently recommended a specific additional imaging examination (eg, spot compression, magnification, or special mammographic views) and/or US examination (BI-RADS category 0, incomplete assessment). If one reader recommended recall for additional imaging evaluation and the other did not, then the final decision for recall was reached by means of discussion between the two readers, although some additional projections were obtained online if indicated by the first reader.
When the imaging assessment was completed, the findings were classified into final categories: routine screening every 2 years (BI-RADS category 1, negative finding; or category 2, benign finding), short-term interval follow-up every 6 months for 2 years (BI-RADS category 3, probably benign finding), and surgical biopsy (BI-RADS category 4, suspicious abnormality; or category 5, highly suggestive of malignancy).
From April 1992 to April 1994, 17,012 women aged 45–65 years were scheduled for screening by using a population census. The screening facility accommodated 40–50 women per day, 5 days per week, and 13,790 women underwent screening. The overall compliance rate was 81% (13,790 of 17,012). The mammographic findings classified into BI-RADS categories 0, 1, 2, and 3 were analyzed separately for the objectives of the present study.
Benign Mammographic Findings
The following mammographic findings were included as characteristically benign and excluded from short-term follow-up (2–4,7,9): fat-containing (radiolucent) lesions and mixed lesions (eg, lipoma, lipid cyst, hamartoma, galactocele, hematoma, and intramammary lymph node) and sharply defined masses with coarse or dense calcification. Summation shadows due to fibroglandular structures and asymmetric breast tissue or to the greater density of breast tissues, containing fat (radiolucent areas within) and usually located in the subareolar area or upper-outer quadrants, with no mass effect or central density, also were considered benign. Calcified fat necrosis and dystrophic calcification, a rim of calcium surrounding a lucent center, milk of calcium in the cyst, coarse or popcornlike calcification, dermal and vascular calcification, and calcification of plasma cell mastitis were classified as benign findings. Round or oval sharply circumscribed masses without fat that were 10 mm or less in diameter and generally located in the axillary tail or outer breast, with homogeneous low density that was transparent to fibroglandular structures and with completely well-defined contour, were also classified as benign (2,13). Well-defined circumscribed masses smaller than 5 mm that had low density and were located anywhere in the breast also were classified as benign.
Benign US Findings
The US criteria for a simple cyst included well-circumscribed margins, a bright posterior wall, round or oval contours, a smooth border, a thin wall, absence of internal echoes, and good posterior acoustic enhancement with slight edge shadows (14,15). In addition, the anterior wall of some cysts may appear less well defined because of reverberation artifact (15). On occasion, septated wall, compressibility by the transducer, and sedimentation within the cyst that changes with the patient’s position may be depicted in a simple cyst. Enlarged ducts were also included as a benign finding. Masses with US characteristics of intramammary lymph nodes—ellipsoid or round, well circumscribed, with echogenic capsule and a hyperechoic central focus—were included as benign findings. US findings correlating with the mammographic diagnosis of hamartoma—smooth margins; hypoechoic, with fat islands; and thin anechoic pseudocapsule—were also classified as benign.
Mammographic and US Characteristics of Probably Benign Lesions
The most important role of US was in determining the cystic or solid nature of a circumscribed mass seen at mammography (16). Also, we performed US to confirm the normalcy of tissues in densities, better define the contour and characteristics of a well-defined circumscribed mass, or exclude a mass in a focal asymmetric density. A strong correlation between mammographic and US findings was necessary. It was important to ensure that the size, shape, and location of a mass or density corresponded at mammography and US. The following mammographic and US characteristics were defined in probably benign lesions (3–9,14–18).
Masses.—Mammographic findings were noncalcified well-defined circumscribed masses that had a round, oval, or slightly lobular contour that was totally visible, with or without a halo sign, and had no size limitations but were nonpalpable. Well-defined circumscribed masses with partially obscured contour due to superimposed or adjacent normal tissue were also included as probably benign findings. Probably benign masses may be isodense or have high- or low-density tissue to fibroglandular breast tissue but are not fat containing.
The following criteria were used to define the US characteristics of probably benign solid breast masses: shape (round, oval, or macrolobulated); margins (circumscribed or well defined); width–anteroposterior dimension ratio (ellipsoid or with a width greater than height, with the longitudinal axis parallel to the transducer); echogenicity, as compared with the echodensity of fat (hyperechoic, isoechoic, or mildly hypoechoic); echotexture (homogeneity of internal echoes); posterior echoes (enhanced or normal sound transmission, or acoustic shadowing); and an echogenic pseudocapsule. An anechoic inner ductal pattern (tubules) is seen occasionally in well-defined probably benign circumscribed solid masses (14).
Focal asymmetric densities.—Mammographic findings were similar in shape on images obtained in two views but completely lacked borders and the conspicuity of a true mass, with density similar to that of fibroglandular parenchyma or with some difference (slightly more dense or less fat-containing than fibroglandular parenchyma). These densities did not have a counterpart in the opposite breast or had changed on spot mammograms and were generally located outside the subareolar area or upper-outer quadrants, without mass effect, associated structural alteration, or malignant microcalcifications.
Focal asymmetric densities with an echogenic pattern similar to or more echogenic than that of the adjacent fibroglandular structures were included as probably benign lesions. The asymmetric subareolar tubular density that represented enlarged ducts at US also was included. The combination of a focal asymmetric density with probably benign calcifications was assessed as a focal density.
Calcifications.—These included grouped calcifications that occupied less than 2 mL of tissue or had regional distribution that comprised similar elements 1 mm or smaller and were oval, rounded, or spherical; had some size and density differences; contained particles with well-defined margins; and had a number of calcifications with more than five elements. Groups of punctate calcifications of 0.5 mm or smaller and groups of amorphous, indistinct, or hazy elements were included as probably benign findings. A miscellaneous group of indeterminate (not definitively benign) calcifications comprising a mixed variety of particles (a combination of benign calcifications—oval, round, punctate, amorphous, and hazy elements) also were included at short-term mammographic follow-up.
Multiple similar findings.—Two or more well-defined probably benign circumscribed solid masses or groups of probably benign calcifications that were similar in morphology and distribution or diffuse probably benign calcifications that were similar and distributed throughout one or both breasts also were included at short-term follow-up. In multiple or bilateral well-defined probably benign circumscribed solid masses and multiple groups of probably benign calcifications, interpretation was performed for each lesion or group of calcifications.
Management of Probably Benign Lesions
After independent double-reading and complete assessment with additional mammographic evaluation and/or US, the first reader evaluated the mammographic and US findings to finally establish the negative or benign findings and select the patients to undergo short-term mammographic follow-up. In well-defined circumscribed masses and focal asymmetric densities, short-term follow-up was avoided when a simple cyst was visualized at US. The circumscribed masses with mammographic and/or US characteristics of intramammary lymph node or hamartoma were excluded from short-term follow-up. Short-term mammographic follow-up was indicated in probably benign masses when solid probably benign characteristics were found at US. Short-term mammographic follow-up was indicated in probably benign focal asymmetric densities when US was performed to exclude a suspicious mass.
The asymmetric tubular densities that corresponded with enlarged ducts at US were evaluated with short-term mammographic follow-up. Nonpalpable well-defined circumscribed masses that were seen at mammography but not at US or were not definitively cystic were also included at short-term mammographic follow-up. When US depicted a probably benign solid circumscribed mass in a focal asymmetric density, this lesion was excluded from mammographic short-term follow-up and was followed up with US at the referral hospital. Short-term mammographic follow-up was recommended, usually with a single projection that best demonstrated the lesion. The probably benign calcifications were systematically tracked with short-term follow-up during which magnification views were used. Either the demonstration of a probably benign lesion on previous mammograms obviated additional short-term follow-up, or short-term follow-up was completed every 6 months for 2 years and took into account the preceding period of stability.
Women with palpable findings that corresponded with probably benign lesions were excluded from short-term follow-up and further examined with US and/or fine-needle aspiration biopsy. Short-term follow-up was indicated every 6 months for 2 years (with four follow-up examinations, including the 24-month follow-up examination in the second round) and performed at the screening facility. All women who completed at least one follow-up study between the first and second rounds were included as having followed-up and probably benign lesions. The first reader informed the patient of the screening findings and short-term follow-up recommendation, and a personal reminder letter was sent to the patient 2 weeks before her next appointment.
As in previous studies (4,7–9), lesion progression (ie, enlargement or increased number or size of calcifications or modification of their initial characteristics) was considered an indication for needle-guided surgical biopsy, even if lesion examination findings still were probably benign. Nonpalpable lesions were not sampled percutaneously (with mammographically or US-guided fine-needle aspiration biopsy or large-core biopsy) because these techniques were unavailable at the referral hospital at that time. The surgical biopsies of nonpalpable probably benign lesions with progression were performed in accordance with the radiologic-pathologic protocol of the breast cancer multidisciplinary committee of the referral hospital. No interval changes detected at mammographic short-term follow-up every 6 months for 2 years indicated a definitely benign lesion, and the women were assigned to the 2-year screening interval of our program. Lesions that had disappeared or diminished at short-term follow-up were considered benign, and short-term follow-up was concluded.
Calculations
Specific software was designed for data collection. Interobserver agreement was calculated for the benign and probably benign mammographic findings. Each observer independently read the craniocaudal and oblique mediolateral screening mammograms to provisionally classify benign and probably benign findings without using BI-RADS category 0, incomplete assessment. After this provisional classification, each reader judged whether the woman should be recalled for further assessment with additional mammography and/or US. The first reader provisionally classified the mammographic findings before obtaining information from additional projections, previous mammographic examinations, or online clinical examinations; to prevent bias, this information was not made available to the second reader.
If there was no agreement after the independent double-reading, the readers conjointly discussed the mammographic findings, clinical information, and need for further assessment with additional mammographic evaluation and/or US. When complete assessment was performed, the first reader evaluated and correlated the mammographic and/or US findings to establish the definitive classification in final BI-RADS categories. In some cases, the definitive classification required a new consensus between the readers.
Observer variability between the readers for benign and probably benign mammographic findings was calculated by using the weighted 
index. Interobserver agreement for benign and probably benign mammographic findings was obtained from the provisional independent double-readings of 6,810 mammograms without using information from additional mammographic examinations and/ or US. Palpable lesions were excluded from interobserver agreement calculations. A value less than 0.40 indicated poor agreement; 0.41–0.75, fair; and greater than 0.75, excellent (19). The probably benign lesions assessed with short-term follow-up were classified retrospectively and grouped according to the findings at mammography and US, surgical biopsy, and histologic examination, and according to the period in which biopsy was performed. The short-term follow-up protocol (every 6 months for 2 years) for probably benign lesions was evaluated to determine its effectiveness as a diagnostic test. The test results were defined as positive when the lesion had progressed at short-term mammographic follow-up. The histologic diagnosis of cancer was the reference standard.
The histologic examination results of surgically excised probably benign lesions that had progressed at short-term follow-up were used to determine the true- and false-positive rates. To calculate the true- and false-negative rates, we reviewed the status of probably benign lesions without progression at 2-year short-term follow-up to determine whether there was a cancer diagnosis within 2 years thereafter. A true-positive finding was defined as cancer diagnosed after a biopsy recommendation based on lesion progression at the 2-year short-term follow-up; a true-negative finding, probably benign lesions without a cancer diagnosis within 2 years after 2-year short-term follow-up, without lesion progression; a false-negative finding, probably benign lesions with a cancer diagnosis within 2 years after the 2-year short-term follow-up, without lesion progression; and a false-positive finding, benign histologic findings after a biopsy recommendation based on lesion progression at 2-year short-term follow-up. The sensitivity, specificity, accuracy, and positive and negative predictive values were calculated in accordance with standard methods (2,20–24).
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTSDISCUSSIONREFERENCES |
|---|
values) for benign and probably benign mammographic findings was 0.59 and 0.75, respectively. Of 13,790 women who underwent screening, 1,075 (7.8%) required physical examination and 1,194 (8.7%) required additional imaging with extra views and/or US (BI-RADS category 0). In this group, 499 (3.6%) US examinations were performed. After additional imaging and/or US assessment, mammograms were considered to be negative or show benign findings (BI-RADS final categories 1 and 2, respectively) in 301 (2%) cases, including 176 simple cysts seen at US (35% of 499 US examinations). Eight hundred ninety-three (6.5%) of 13,790 women with nonpalpable probably benign lesions underwent short-term mammographic follow-up (BI-RADS final category 3), and 795 (5.8%) women adhered to short-term follow-up (at least one examination between the first and second rounds). The adherence rate for short-term follow-up was 89% (795 of 893 women). The mean age of the women who underwent follow-up was 50 years (age range, 45–65 years). Ninety-one (11.4% of 795) women had a combination of multiple probably benign findings in one or both breasts.
The overall results for the probably benign mammographic findings are shown in Table 1. Most of the well-defined circumscribed masses had solid probably benign US characteristics; a few cases of nonpalpable probably benign circumscribed masses (nine [3.7%] of 242) were not visualized at US. Five women ([0.6%] of 795) did not accept US examination but were assigned to short-term mammographic follow-up. No fine-needle aspiration biopsy or large-core biopsy was performed in any of the probably benign lesions that were followed up. Calcifications were the most common abnormality (471 [59%] of 795). Of 795 lesions that were followed up, 788 (99%) were stable, and surgical biopsy indicated by lesion progression was performed in seven (1%).
|
|
|
|
|
|
In the group of women with probably benign lesions who adhered to short-term mammographic follow-up, two cancers were found incidentally during the third screening round. However, these lesions were unrelated to the probably benign lesions undergoing initial short-term mammographic follow-up. These malignancies were two irregular masses that appeared as new findings in the third round and were not retrospectively visible on the mammograms obtained during the first and second rounds.
Statistical calculations are presented in Table 2. The sensitivity, specificity, accuracy, and positive and negative predictive values were 100% (two of two cancers), 99% (788 of 793 noncancers), 99% (790 of 795 total lesions), 29%, (two of seven progressed lesions) and 100% (788 of 788 nonprogressed lesions), respectively. The comparison between the pre- and posttest probabilities showed an increased value from 0.003% to 0.2% for breast cancer diagnosis. This increase indicates the value of short-term follow-up as a diagnostic examination for breast cancer in nonpalpable probably benign mammographic lesions. Since no false-negative cases were found within 2 years after the end of short-term follow-up, the sensitivity value was 100%, and consequently it was not possible to calculate the posttest probability for negative results (nonprogression at mammographic short-term follow-up).
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTSDISCUSSIONREFERENCES |
|---|
In our experience, most of the lesions (six of seven—one focal asymmetric density and five cases of grouped probably benign calcifications) had progressed at the 6-month follow-up examination, with one resultant cancer. Because there were few cancers, it was not possible to draw additional conclusions regarding the utility of the 6-month follow-up examination in our study. Because of the low probability of malignancy in the probably benign findings, the advisory board of our screening program suggested a short-term mammographic follow-up protocol limited to two examinations (at 6 and 12 months) between rounds and one final follow-up examination at 24 months, which coincides with the next screening round.
In some lesions, benign and probably benign findings are not clear. The reader’s greater experience increases the diagnosis of benign lesions and thereby reduces the need for additional short-term mammographic follow-up (5). The percentage of short-term follow-up recommendations gradually decreased in our program, and the readers in the present study indicated a short-term follow-up of 2%–3% in the fourth screening round (50% of the initial value of the first round); this recommendation is basically due to the readers’ experience and the availability of mammograms from the first, second, and third rounds.
The agreement between the readers indicates homogeneity of criteria in the classification of benign and probably benign findings. This homogeneity was obtained through meetings before screening began and through the continuous radiologic education process. Although values for benign and probably benign lesions were fair, the lower value in benign findings (0.59 vs 0.75) occurred because the readers randomly included some benign findings (eg, intramammary lymph nodes, dystrophic calcifications, vascular calcifications, and calcifications of plasma cell mastitis) as negative on mammograms (BI-RADS category 1).
There are some unresolved questions due to the complexity of the variables in radiologic imaging (5,7). In our protocol, sharply defined circumscribed masses smaller than 5 mm were considered benign, but low-density well-defined circumscribed masses in the axillary region of the breast that were 10 mm or smaller also were considered benign (13). The size limit of a nonpalpable probably benign mass is an arbitrary decision that should be evaluated conjointly with other findings—shape, margins, and density (2). It is common sense to not miss the opportunity to diagnose a breast cancer before it reaches 10 mm because of the different prognosis with larger tumors.
The contour of a circumscribed mass may be partially obscured by the surrounding fibroglandular parenchyma that simulate ill-defined margins, and this sign may be erroneously interpreted. In definitively benign well-defined circumscribed solid masses in the current study, 81 (34%) of 236 had a partially obscured contour because of the overlapping imaging, but these masses had a smooth border and associated probably benign findings at US.
The detailed description of US findings that are related to well-defined circumscribed masses is not the subject of the present study. The data confirm that certain US features can help differentiate benign from malignant masses (28,29). Some US findings (eg, hypoechogenicity, normal or enhanced transmission, shadowing, and heterogeneous texture) are considered indeterminate when evaluated separately (28). However, we considered these indeterminate features conjointly with the benign mammographic and additional US findings (eg, a well-defined ellipsoid mass with homogeneous hypoechogenicity and shadowing that was depicted at mammography as a probably benign mass was assessed with short-term follow-up). Nevertheless, the inter- and intraobserver variability in defining solid US characteristics may be important (30,31). Correlation of mammographic and US findings is essential to achieve a correct indication for short-term mammographic follow-up in nonpalpable probably benign mammographic masses or densities (15,31).
A mass can be visualized as a focal asymmetric density, although in the present study no cancer was found in focal asymmetric densities. In a previous unpublished review, we detected advanced cancers of TNM stages IIA and B in focal asymmetric densities that were evaluated exclusively with short-term follow-up, without previous clinical and US evaluation. Similar to Wolfe et al (4) and Dawson and Wilson (32), we are opposed to the exclusive use of short-term follow-up in the management of focal asymmetric densities, which include the asymmetric ductal pattern. Palpation, US, and subsequent (if indicated) cytologic examination and/or core biopsy should be performed in all cases. Currently in our screening program, probably benign focal asymmetric densities at US and negative clinical findings are considered definitively benign and are not assessed with short-term mammographic follow-up.
At the present time in our screening program, diffuse calcifications found in one or both breasts and comprising similar elements, multiple similar groups of probably benign calcifications, or grouped calcifications that are oval, round, or spherical, with some size and density difference between particles and with well-defined margins, are considered definitively benign. Calcifications of intermediate concern and a miscellaneous group of indeterminate calcifications are also included at short-term follow-up. To our knowledge, a consensus on how to assess these cases does not exist in the literature.
Calcifications were frequent (471 [59%] of 795), but the positive predictive value for cancer was low (one [0.2%] of 471). Although most indeterminate calcifications are benign, a few are associated with ductal carcinoma in situ (33); however, surgical treatment of these potentially deadly lesions increases cost and morbidity. This is in addition to the significant screening cost that could be reduced by decreasing the number of surgical biopsies performed in benign calcifications. In indeterminate calcifications, it is not possible to exclude malignancy by means of percutaneous tissue sampling with stereotactic devices, and additional short-term follow-up is recommended in the presence of negative results (34). Large-core biopsy in probably benign lesions has been suggested for women who cannot comply with mammographic surveillance or who personally request histologic examination (35), but the cost-effectiveness of this technique in low-risk lesions is questionable (36).
In Table 3, the results from four series (4,7–9) and the current study are compared. All series included data obtained in the first screening rounds. In the study by Wolfe et al (4), the patients were examined with negative-mode xeromammography, the number of benign biopsies was not available, and US was not performed to select lesions for follow-up. Also, in the Helvie et al study (8), US was not performed. Our percentage of probably benign lesions that was related to overall screening mammograms was 6%, which falls within the reported range of 2%–11% (4,7–9). The foregoing data show a correct indication for short-term mammographic follow-up at the Alcoy City screening facility in the current study and validate the recommendations of the screening program’s advisory board.
|
Our overall biopsy yield value (29%) was higher than that of Sickles (7) (11%) and Helvie et al (8) (10%) but lower than that of Varas et al (9) (56%). Although there are few cases, we found opposite and extreme values in well-defined circumscribed solid masses (one of one) and focal asymmetric densities (zero of one). Most of the biopsies with a benign histologic result were calcifications with progression at short-term follow-up, with a positive predictive value of surgical biopsy of 20% (one of five). US did not demonstrate masses in the followed-up focal asymmetric densities, and no benign well-defined circumscribed solid masses had progressed in our study. In addition, the indication for surgical biopsy depends on extraradiologic aspects, which include medicolegal reasons (5). Some women prefer surgical biopsy as an alternative to living with the uncertainty of having a potentially malignant lesion with "low probability," and this can influence the benign histologic results of surgical biopsy, although it did not influence our results.
Because some cancers may progress slowly even at 2 years, we reviewed the status of the probably benign lesions without progression at the 2-year short-term follow-up to determine whether there was a cancer diagnosis within 2 years after the 2-year short-term follow-up to calculate the true- and false-negative rates. Our accuracy, sensitivity, specificity, and positive and negative predictive values were excellent. Although in our study the rate of adherence to short-term follow-up was 89%, some cancer cases may have been lost to short-term mammographic follow-up and may have influenced overall calculations (Table 2). Screening compliance in the assessment of nonpalpable probably benign lesions with short-term mammographic follow-up remains a significant problem (8).
In conclusion, assessment with short-term interval mammographic follow-up is useful to confirm the benign nature of most nonpalpable probably benign lesions and permits detection of a small number of breast cancers at a low stage. The period considered short-term follow-up must be defined in relation to the screening interval for a specific screening method. The lesion progression demonstrated at short-term follow-up, however, was not accurate for predicting malignancy in probably benign calcifications. Further studies on the radiologic-pathologic correlation of calcifications are warranted to achieve more precision in defining benign radiologic findings.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
Author contributions: Guarantor of integrity of entire study, I.V.; study concepts, D.S.; study design, I.V., D.S.; literature research, I.V., F.R.P.; clinical studies, L.G., L.A.; data acquisition, L.G., D.S., F.B.; data analysis, I.V., D.S.; statistical analysis, D.S.; manuscript preparation, I.V.; definition of intellectual content, D.C., C.H.; manuscript editing, I.V.; manuscript review, L.A., D.S., F.R.P., D.C., C.H.; manuscript final version approval, all authors.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  S. Raza, S. A. Chikarmane, S. S. Neilsen, L. M. Zorn, and R. L. Birdwell BI-RADS 3, 4, and 5 Lesions: Value of US in Management--Follow-up and Outcome Radiology, September 1, 2008; 248(3): 773 - 781. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. J. A. Bowles, D. L. Miglioretti, E. A. Sickles, L. Abraham, P. A. Carney, B. C. Yankaskas, and J. G. Elmore Accuracy of Short-Interval Follow-Up Mammograms by Patient and Radiologist Characteristics Am. J. Roentgenol., May 1, 2008; 190(5): 1200 - 1208. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
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]
|
|
|
|
|
|
C. D. Lehman, C. M. Rutter, P. R. Eby, E. White, D. S. M. Buist, and S. H. Taplin Lesion and Patient Characteristics Associated with Malignancy After a Probably Benign Finding on Community Practice Mammography Am. J. Roentgenol., February 1, 2008; 190(2): 511 - 515. [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]
|
|
|
|
|
|
E. S. Burnside, J. E. Ochsner, K. J. Fowler, J. P. Fine, L. R. Salkowski, D. L. Rubin, and G. A. Sisney Use of Microcalcification Descriptors in BI-RADS 4th Edition to Stratify Risk of Malignancy Radiology, February 1, 2007; 242(2): 388 - 395. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
W. A. Berg Sonographically Depicted Breast Clustered Microcysts: Is Follow-Up Appropriate? Am. J. Roentgenol., October 1, 2005; 185(4): 952 - 959. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
F. M. Hall, E. A. Sickles, C. Lehman, R. Smith-Bindman, B. C. Yankaskas, B. M. Geller, P. A. Carney, and K. Kerlikowske Short-Interval Follow-up in Screening Mammography * Dr Sickles and colleagues respond: Radiology, October 1, 2005; 237(1): 371 - 373. [Full Text] [PDF]
|
|
|
|
|
|
E. A. Sickles, D. L. Miglioretti, R. Ballard-Barbash, B. M. Geller, J. W. T. Leung, R. D. Rosenberg, R. Smith-Bindman, and B. C. Yankaskas Performance Benchmarks for Diagnostic Mammography Radiology, June 1, 2005; 235(3): 775 - 790. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. Kerlikowske, R. Smith-Bindman, L. A. Abraham, C. D. Lehman, B. C. Yankaskas, R. Ballard-Barbash, W. E. Barlow, J. H. Voeks, B. M. Geller, P. A. Carney, et al. Breast Cancer Yield for Screening Mammographic Examinations with Recommendation for Short-Interval Follow-up Radiology, March 1, 2005; 234(3): 684 - 692. [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]
|
|
|
|
 |
|
 K. Kerlikowske, R. Smith-Bindman, B.-M. Ljung, and D. Grady Evaluation of Abnormal Mammography Results and Palpable Breast Abnormalities Ann Intern Med, August 19, 2003; 139(4): 274 - 284. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. Kerlikowske, R. Smith-Bindman, and E. A. Sickles Short-Interval Follow-Up Mammography: Are We Doing the Right Thing? J Natl Cancer Inst, March 19, 2003; 95(6): 418 - 419. [Full Text] [PDF]
|
|
|
|
8) that compose one round, well-defined, high-density particle of 1 mm or smaller, three other elements of similar size but different shape and density, and associated smaller particles of 0.1 mm or smaller. The distance between the calcifications is 2-25 mm. The magnification factor is 1:1.7. (b) Craniocaudal spot magnification image (1:1.7) obtained at 6-month follow-up shows progression of size and number of calcifications (arrows). Histologic examination revealed a microinvasive intraductal carcinoma.
