| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Original Research |
1 Department of Radiology and Research Institute of Radiology, University of
Ulsan College of Medicine, 388-1 Pungnap-Dong, Songpa-Gu, Seoul 138-376, South
Korea.
2 Department of Pathology, Asan Medical Center, University of Ulsan College of
Medicine, Songpa-Gu, Seoul, Korea.
Received June 3, 2007;
accepted after revision September 13, 2007.
Address correspondence to H. H. Kim
(hhkim{at}amc.seoul.kr).
Abstract
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
MATERIALS AND METHODS. Surgical or sonographic follow-up (
2
years) was available in 123 women (age range, 21–75 years; mean age, 47
years) with 124 papillary lesions diagnosed at sonographically guided core
needle or vacuum-assisted biopsy during a 7-year period. Surgical excision
results or follow-up sonograms with no change at the 2-year follow-up served
as the reference standard. We reviewed the sonographic features, biopsy
results, and surgical or sonographic follow-up. Statistical analysis was
performed using the Fisher's exact test for the difference of sonographic
features.
RESULTS. Cancer incidence per BI-RADS category was as follows: category 3, zero (0%) of 21; category 4a, 14 (17%) of 85; category 4b, four (36%) of 11; category 4c, one (20%) of five; and category 5, one (50%) of two. On sonography, two features distinguishing benign from malignant papillary lesions were echo pattern and margins of masses. Core needle biopsy gave a sensitivity of 28% and specificity of 100%, whereas both sensitivity and specificity were 100% with vacuum-assisted biopsy. Of 117 lesions with benign biopsy results, upgrade to malignancy or high-risk lesion was found in 17% of benign papillomas, 24% of atypical papillomas, and 0% of multiple papillomas.
CONCLUSION. Vacuum-assisted biopsy was more accurate than core needle biopsy in diagnosing papillary lesions, and both sonographic features and core needle biopsy were not sufficiently accurate. Therefore, surgical excision should be performed for the accurate diagnosis of papillary lesions.
Keywords: breast • breast biopsy • breast cancer • papillary lesion • sonography
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Although percutaneous breast biopsy is a highly reliable method for the diagnosis of breast lesions [3], controversy persists regarding the need for excision of papillomas diagnosed at percutaneous breast biopsy. Theoretic reasons to excise percutaneously diagnosed papillomas include difficulties in pathologic interpretation, possible sampling errors in papillomas that may contain areas of atypia or carcinoma, and the premalignant potential of these lesions [4].
The diagnosis of papillary carcinoma on fine-needle aspiration biopsy can be difficult because fine-needle aspiration biopsy may miss the small foci of carcinoma in situ or the foci that are invasive [5, 6]. A similar limitation might be encountered with core needle biopsy [7–9]. Some authors have suggested that because benign papilloma diagnosed at core needle biopsy can be followed up, immediate surgical excision is not necessary [9]. However, a papillary carcinoma in situ diagnosed at core needle biopsy might be upgraded to invasive carcinoma when the entire lesion is examined; this possibility is a significant limitation of core needle biopsy for the management of papillary lesions [10–12]. In addition, mammographic, galactographic, and sonographic findings cannot accurately distinguish benign from malignant papillary lesions [13–16].
To our knowledge, few reports have been published in the literature regarding the sono-graphic features of papillary lesions based on BI-RADS categorization and the role of sono-graphic features and sonographically guided biopsy in the management of papillary lesions. The purpose of our study was to retrospectively evaluate the sonographic features, biopsy results, and surgical results or sono-graphic follow-up of papillary lesions and retrospectively to assess the potential role of sonographic features, including BI-RADS category and biopsy, in the management of patients with papillary lesions diagnosed at percutaneous sonographically guided biopsy.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
In our institution, we usually perform screening whole-breast sonography in
women with dense breast tissue, high-risk women less than 40 years old with a
family history of breast cancer, and women who desire it. The examinations for
this study were performed by one of seven breast radiologists. Of the
remaining 56 symptomatic patients, 49 mammograms were available for
retrospective review. On mammography, 19 cases showed masses, three showed
masses with calcifications, one showed calcifications, and the other 26 were
negative. Patients without follow-up sonographic findings ( 2 years) or
surgical results were excluded from our study.
Sonography
We performed whole-breast sonography on all 123 study patients using
5-12–MHz transducers on an HDI-3000, HDI-5000, or IU-22 unit (Philips
Medical Systems). We usually perform whole-breast sonography rather than
targeted sonography guided by mammographic or clinical findings. The
examinations were performed by a radiologist. The usual time needed to
complete a bilateral whole-breast sonographic examination was 15–30
minutes, and the time required to perform a biopsy was 10–20 minutes.
The usual time needed to complete targeted sonography was 5–10 min utes.
Sonograms were retrospectively reviewed independently by one of three breast
radiologists with 3, 6, and 15 years of clinical experience, respectively. Any
discrepancy in opinion was resolved by consensus.
The sonographic features were evaluated according to the BI-RADS lexicon and assessment categories [17]—that is, the shape (oval, round, or irregular), margin (circumscribed or not circ umscribed), echo pattern (isoechoic, hyperechoic, hypoechoic, mixed hyperechoic and hypoechoic, or complex cystic), posterior acoustic features (normal, enhancement, or shadowing), surrounding tissue change (none, duct dilatation, or architectural distortion), and vascularity (no, diffuse, or focal or penetrating). Of the descriptors of echo patterns, a mixed hyperechoic–hypoechoic mass was defined as a lesion that has portions hyper-echoic to fat and portions hypo- or isoechoic to fat and that has no cystic components. We also evaluated the size of all papillary lesions on sono graphy.
All of the lesions were classified according to one of the following categories: probably benign lesion (category 3), lesion with 2% or lower probability of malignancy; indeterminate lesion (category 4a), lesion with a low suspicion for malignancy (risk of malignancy, 3–10%); intermediately suspicious for malignant lesion (category 4b), lesion with an intermediate suspicion for malignancy (risk of malignancy, 11–49%); highly suspicious for malignant lesion (category 4c), lesion with moderate concern, but not classic for malignancy (risk of malignancy, 50–95%); and malignant lesion (category 5), lesion with 95% or higher likelihood of malignancy.
Biopsy
We performed biopsies on all study patients using sonographic guidance.
Core needle biopsy was performed on 109 lesions using semiautomatic 14-gauge
core needle devices (Stericut, TSK Laboratory); a mean of five core samples
(range, 4–7 samples) were obtained per lesion. Vacuum-assisted biopsy
was performed on 15 lesions. When biopsy was performed with an 11- or 8-gauge
vacuum-assisted probe (Mammotome, Ethicon Endo-Surgery), we usually removed
the lesion until the lesion had disappeared sonographically. The radiologist
who performed the biopsy assessed whether the imaging target was completely
removed or was sampled. This assessment was made and recorded after biopsy had
been performed. Complete removal of the imaging target was possible in 15
(12%) of the 124 lesions.
The type of biopsy was determined by the radiologist at the time of the procedure or by the referring physician. The biopsy procedures were performed by radiologists with 3–15 years' experience in breast imaging. The results of core needle biopsy and vacuum-assisted biopsy were classified either as benign papillary lesions, such as benign solitary papilloma, atypical papilloma, and multiple papillomas or as malignant papillary lesions, such as invasive papillary carcinoma and papillary DCIS. No clips were placed after sonographically guided biopsies. Radiologists correlated the imaging findings with percutaneous biopsy results to assess whether histology results and imaging findings were concordant or discordant. The radiologist who performed the core needle or vacuum-assisted biopsy determined the concordance or discordance between the imaging and histologic findings of each case.
We classified the findings to be discordant when BI-RADS category 5 (highly suspicious for malignancy) was given to the lesion at imaging and the corresponding histologic finding was benign. In addition, when a lesion manifested as a new finding in a postmenopausal woman (considered moderately suspicious with BI-RADS category 4b or 4c) and had a benign histologic diagnosis at biopsy or when a BI-RADS category 3 or 4a (considered probably benign or low suspicions for malignancy) was given to the lesion at imaging and the corresponding histologic finding was malignant, we also classified these findings as discordant.
On the contrary, we classified the findings to be concordant when a BI-RADS category 3 or 4a was given to the lesion at imaging and the corresponding histologic finding was benign. In addition, when BI-RADS category 5 (highly suspicious for malignancy) was given to the lesion at imaging and the corresponding histologic finding was malignant, we also classified this finding to be concordant. Although most patients with discordant findings were referred for surgical excision, the management of papillomas classified as benign at core needle or vacuum-assisted biopsy was not uniform and a recommendation of short-term imaging follow-up was given in some cases.
Follow-Up Findings
All 124 lesions were either excised or followed up sonographically for at
least 24 months. All of the lesions with atypical or malignant findings on
biopsy were surgically excised, and the surgical results were collected from
pathology reports and compared with the biopsy results. The other benign
lesions were managed by either delayed surgical excision or follow-up
sonography. Delayed surgical excision was performed when the lesion increased
in size or altered in shape, when patients required removal of the lesions
because of symptoms such as nipple discharge or a palpable lump, or when the
patient wanted or the referring physician preferred lesion removal.
For the lesions that were excised, the surgical pathology reports were reviewed, and the lesions were classified in the following categories: benign papilloma, sclerosing papilloma, atypical papilloma, multiple papillomas, and other benign lesions; high-risk lesions, such as ADH, lobular carcinoma in situ (LCIS), or atypical lobular hyperplasia (ALH); and in situ or invasive carcinoma including papillary and invasive ductal carcinoma. Lesions with follow-up sonograms with no change at 2 years were considered benign. Surgical pathology results or follow-up sonograms with no change at 2 years served as the reference standard.
Statistical Analysis
Sonographic and pathologic findings and information from the medical
records were recorded. Data were entered into a computerized spreadsheet
(Excel, Microsoft). Benign and high-risk lesions on surgical excision or
follow-up sonographic findings with no changes after 2 years were classified
as benign for the following analysis. On the basis of these results, the
sensitivity, specificity, positive predictive value (PPV), negative predictive
value (NPV), and accuracy in detecting malignant papillary lesions were
calculated for each biopsy method and sonographic categorization, as were 95%
CIs. The cancer detection rate and incidence of high-risk lesions such as ADH
or lobular neoplasia in each category were calculated from lesions that were
both surgically excised and followed up with sonography. The Fisher's exact
test was used for the differences in the sonographic findings between benign
and malignant papillary lesions. Findings with a p value of less than
0.05 were considered statistically significant.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
|
|
On the comparison of sonographic features between benign and malignant or high-risk lesions, the sonographic findings except two showed no significant difference between the two groups. One finding was echo pattern and the other was the margin of a mass. Echo pattern was one significantly distinguishing finding between the two groups (p = 0.003): Benign papillary lesions were mainly homogeneous iso-, hyper-, or hypoechoic, whereas malignant or high-risk lesions showed mixed hyperechoic–hypoechoic or complex cystic echogenicity (Table 2). Most of the high-risk or malignant papillary lesions were not circumscribed (20/27, 74%), whereas 50 (52%) of 97 benign papillary lesions were circumscribed in margin, which showed a significant difference between the two groups (p = 0.028). The feature of angular or spiculated margins was absent in all papillary lesions. The mean size of all papillary lesions was 1.45 cm, and the range was 0.5–11.0 cm. The mean size of benign papillary lesions was 1.3 cm (range, 0.5–11.0 cm), and that of malignant papillary lesions was 1.8 cm (range, 0.5–5.9 cm). The mean size of asymp tomatic papillary lesions was 1.2 cm (range, 0.5–4.0 cm), and that of symptom atic papillary lesions was 1.7 cm (range, 0.5–11.0 cm).
|
Surgical Follow-Up
Surgical excision was performed on 109 (88%) of the 124 lesions. The mean
surgical follow-up interval was 2.1 months, and the range was 2 days–18
months. Eighty-two (75%) of the 109 lesions were benign, seven (6%) were
high-risk lesions, and the remaining 20 (18%) were cancer. Of the 82 benign
lesions, 59 (72%) were papillomas, seven (9%) were atypical papillomas, four
(5%) were multiple papillomas, two (2%) were sclerosing papillomas, and the
remaining 10 (12%) were other benign diseases. Of the 20 malignant lesions, 13
(65%) were DCIS and seven (35%) were invasive cancers. Of the seven high-risk
lesions, four (57%) were ADH, two (29%) were LCIS, and only one (14%) was ALH
(Table 3).
|
Sonographic Follow-Up
Sonographic follow-up consisted of a minimum of 24 months in 15 (12%) of
the 124 lesions. The mean follow-up was 36 months, and the range was
24–60 months. Of the 15 lesions that were followed up, all 15 lesions
had benign, concordant results of benign papillary lesions and were
categorized as probably benign (n = 8) or as indeterminate
(n = 7). Eight (53%) of these 15 lesions did not change on follow-up
sonography for at least 2 years and seven (47%) had decreased in size or
disappeared. Consequently, all 15 lesions were considered to be benign.
Of the 117 lesions with benign biopsy results, malignancy was found in 12 (13%) of 96 benign papillomas (Figs. 2 and 3A, 3B), one (6%) of 17 atypical papillomas, and none of four multiple papillomas. Upgrade to high-risk lesion or malignancy was found in 16 (17%) benign papillomas, four (24%) atypical papillomas, and zero multiple papillomas.
|
|
|
Of the 15 lesions undergoing vacuum-assisted biopsy, 10 (67%) were benign papilloma, two (13%) were atypical papilloma, one (7%) was multiple papillomas, and the remaining two (13%) were papillary carcinoma. One of two atypical papillomas that were diagnosed on vacuum-assisted biopsy was proven to be ADH on surgical excision, and the other one had disappeared on follow-up sonography. Two of 10 benign papillomas were proven to be benign papilloma on surgical excision, and the remaining eight had disappeared on follow-up sonography. One lesion diagnosed as multiple papillomas had disappeared on follow-up sonography. Both lesions diagnosed as papillary carcinoma were proven to be papillary carcinoma.
On the basis of the results of surgical excision combined with follow-up sonographic findings, core needle biopsy gave a sensitivity of 28%, specificity of 100%, PPV of 100%, and NPV of 88%; whereas vacuum-assisted biopsy gave a sensitivity of 100%, specificity of 100%, PPV of 100%, and NPV of 100% for the diagnosis of malignant papillary lesion.
One hundred three lesions had benign, concordant biopsy results; 14 lesions showed benign, discordant results; two lesions showed malignant, concordant results; and five lesions showed malignant, discordant results. Malignancy was found in 10 (10%) of 103 lesions with benign, concordant biopsy results and three (21%) of 14 lesions with benign, discordant results.
Of 19 lesions with malignant (n = 5) or benign (n = 14) discordant results, all 19 discordant lesions underwent immediate surgical excision within 1 month of biopsy. Eight were cancer, and the remaining 11 were benign.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
In our study, of 117 lesions with benign biopsy results, malignancy or high-risk lesions were associated with 16 (17%) benign papillomas, four (24%) atypical papillomas, and none of the cases of ductal papillomatosis. Seventeen (17%) of 103 lesions yielding benign, concordant biopsy results were upgraded to a malignancy or high-risk lesion.
On sonography, papilloma usually presents as a circumscribed, oval-shaped, hypoechoic, or complex cystic mass; the vascularity identified in the fibrovascular core of the papilloma may suggest the diagnosis [14, 16, 18]. Yang et al. [14] found that the presence of dilated ducts was a common sonographic finding that is often associated with a visible intraluminal echo. Lam et al. [16] found that neither mammo-graphic nor sonographic features were sensitive or specific enough to allow accurate differentiation between benign and malignant lesions.
In this study, most papillary lesions (106/124, 85%) were categorized as probably benign (n = 21) or indeterminate (n = 85), and none of the papillary lesions showed the feature of angular or spiculated margin. That is, papillary carcinomas fall within circumscribed cancers including papillary, medullary, and mucinous cancer. In our study, benign papillary lesions had a trend of more circumscribed margins than malignant or high-risk lesions. The overall sensitivity of sonography was 100%; specificity, 20%; PPV, 19%; NPV, 100%; and accuracy, 33%.
Philpotts et al. [8] reported DCIS in one of six papillary lesions with benign histo-logic findings at percutaneous breast biopsy and concluded that surgical excision of papillary lesions with benign histologic findings is warranted only if the imaging and histologic findings are discordant. Mercado et al. [27] reviewed 18 nonmalignant papillary lesions and suggested that benign papillomas may not require surgical excision if the mammographic findings are concordant with the histologic results. Gutman et al. [29] reviewed 95 women with papilloma or papillomatosis and found that in up to 10% of these patients, solitary papilloma was associated with breast carcinoma. An additional 9% of these patients presented with invasive or noninvasive carcinoma in the papilloma. Liberman et al. [30] found that among lesions yielding a benign, concordant diagnosis of papilloma at percutaneous biopsy, surgery revealed cancer in 14% and high-risk lesions in 17% of their patients. More recently, Mercado et al. [31] reported a study of papillary lesions diagnosed at core needle biopsy and concluded that benign papillary lesions should be surgically excised. In that study of 43 lesions in 42 patients, nine patients (21%) had ADH or DCIS on surgical excision. Sydnor et al. [28] concluded that benign papilloma diagnosed at core needle biopsy was only infrequently (3%) associated with malignancy and that mammographic follow-up was sufficient. However, a diagnosis of atypical papilloma at core needle biopsy should prompt excision for definitive diagnosis because of the high association with malignancy (67%).
In our study, core needle biopsy gave a sensitivity of 28% and specificity of 100%, whereas both sensitivity and specificity were 100% with vacuum-assisted biopsy. Although the number of vacuum-assisted biopsied lesions was small, vacuum-assisted biopsy was more accurate than core needle biopsy in the diagnosis of papillary lesions. Therefore, further study is necessary to confirm the role of vacuum-assisted biopsy in the diagnosis of papillary lesions.
Several limitations to our study should be mentioned. First, a small number of vacuum-assisted biopsies were performed. In this study, core needle biopsy had a low sensitivity for the diagnosis of malignant papillary lesion, and vacuum-assisted biopsy had an accuracy of 100%. However, the number of vacuum-assisted biopsies was small, so further study is necessary to confirm the accuracy of sonographically guided vacuum-assisted biopsy. Second, we analyzed only sonographic findings and we did not analyze mammographic findings. However, in previous studies [16, 18], investigators evaluated mammographic findings and found that most papillary lesions were not visible on mammography. Thus, we thought the value of evaluating mammography of papillary lesions would be low and we undertook a limited analysis of mammographic findings. Third, our study was retrospective in design. Only patients with surgical excision or sono-graphic follow-up for 2 or more years were included. As we mentioned earlier, a total of 457 patients who underwent sonographically guided biopsy turned out to have papillary lesions, but only 124 patients met our selection criteria. The more benign nature the papillary lesions had, the higher the possibility to be lost to follow-up because asymptomatic patients gave up regular visits. There would be selection bias in our study. Therefore, a large prospective study should be performed to confirm these results.
In conclusion, vacuum-assisted biopsy was more accurate than core needle biopsy, and both sonographic features and core needle biopsy were not sufficiently accurate in diagnosing papillary lesions. Therefore, surgical excision should be performed for the accurate diagnosis of papillary lesions.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
