| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1 Department of Radiology, Hospital of the University of Pennsylvania, 3400
Spruce St., Philadelphia, PA 19104.
2 Department of Radiology, Massachusetts General Hospital, Institute of
Technology Assessment, 101 Merrimac St., 10th Fl., Boston, MA 02214.
3 Department of Surgery, Hospital of the University of Pennsylvania,
Philadelphia, PA 19104.
4 Department of Radiation Oncology, Hospital of the University of Pennsylvania,
Philadelphia, PA 19104.
Received June 18, 2003;
accepted after revision August 7, 2003.
Address correspondence to J. M. Lee
(jlee45{at}partners.org).
Abstract
 Top Abstract IntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
SUBJECTS AND METHODS. Of 101 potentially eligible patients, 80 candidates for breast conservation therapy who had primary breast cancer in 82 breasts diagnosed by excisional biopsy with close or positive margins were included in the study group. All patients underwent contrast-enhanced MRI before further surgery and subsequently underwent either reexcision lumpectomy or mastectomy with histopathologic correlation.
RESULTS. Residual carcinoma, either invasive or in situ, was present in 59.8% of the breasts. The sensitivity and specificity of MRI for detecting residual disease were 61.2% and 69.7%, respectively. Twenty-three additional lesions distant from the biopsy site were identified in 19 breasts, and 18 suspicious lesions underwent biopsy. Histology results indicated that six lesions were malignant, so the overall prevalence was 7.3%. The positive predictive value of identifying an additional suspicious lesion was 33.3%. In 24 breasts, MRI changed which procedure would be performed next from reexcision lumpectomy to mastectomy (n = 9), biopsy of an additional lesion in the ipsilateral (n = 12) or contralateral (n = 2) breast, or neoadjuvant chemotherapy (n = 1). Approximately 25% of the breasts underwent mastectomy as definitive surgical treatment.
CONCLUSION. Overlap in the appearances of benign and malignant lesions limits MRI evaluation for residual disease. MRI can show additional suspicious lesions that are likely to be multicentric or multifocal disease. These findings changed the original treatment plan for approximately 30% of breasts.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
At our institution, reexcision lumpectomy is routinely performed in patients with close (< 2 mm) or positive margins at initial excisional biopsy who desire breast conservation therapy. MRI of the breast has emerged as a promising adjunctive imaging aid in planning surgical treatment by depicting the presence and extent of residual disease and revealing additional foci of cancer that are clinically and mammographically occult. The purposes of this study were to assess the diagnostic accuracy of MRI in evaluating patients for residual cancer, identify the prevalence of multicentric or multifocal disease, and evaluate the impact of MRI on surgical treatment planning in candidates for breast conservation therapy who have close or positive margins at initial excisional biopsy.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Ten patients were excluded because they did not meet inclusion criteria at presentation. The reasons for exclusion were extensive tumor burden or multicentric cancer in one breast referred for MRI evaluation of the other breast (n = 5), axillary lymph nodes positive for metastatic breast cancer with occult primary breast cancer (n = 2), underlying collagen vascular disease precluding radiation therapy (n = 1), negative margins at initial excisional biopsy (n = 1), and a desire to not undergo breast conservation therapy (n = 1). Eleven additional patients were excluded because follow-up was incomplete. This group consisted of patients who had definitive surgical treatment at another institution (n = 3); patients who decided not to undergo further surgery after an MRI did not show evidence of residual or additional disease (n = 2); a patient who began neoadjuvant chemotherapy after MRI examination but did not have definitive surgery because distant metastatic disease developed during the interval (n = 1); and patients for whom full surgical, pathologic, or clinical information was not available to assess the impact of MRI on clinical decision-making (n = 5). The remaining 80 patients were included in this study. Two patients had bilateral primary breast cancers, bringing the total number of breasts to 82.
Imaging
MRI was performed using a 1.5-T magnet (Signa, General Electric Medical
Systems, Milwaukee, WI). Patients were placed in the prone position with each
breast in a four-coil compression breast array. Compression was applied gently
to avoid causing patient discomfort. The imaging protocol and parameters were
as follows: axial T1-weighted localizer sequence through both breasts followed
by a sagittal fat-saturated T2-weighted fast spin-echo acquisition through the
affected breast (TR/TE, 5,000/120). Next, images were obtained using a 3D slab
inter-leaved spoiled gradient-echo sequence with intermittent fat-selective
partial inversion for fat suppression. Images were obtained over a 512 x
256 x 32 matrix in the sagittal plane. The typical section thickness was
2.0–2.5 mm, depending on the size of the breast. The volume was adjusted
so that the 28 sections completely filled the dimensions of the gently
compressed breast. The field of view varied from 16 to 18 cm. A TR of 9.2 msec
and a TE of 2.1 msec were used. A fat-selective inversion pulse was applied
twice every pass through the z-axis of k space or every 16 TRs.
Images were acquired before administration of the contrast agent. Contrast-enhanced imaging was initiated in both breasts with completion of an injection of 20 mL of gadopentetate dimeglumine (Magnevist, Berlex Laboratories, Wayne, NJ). The injection was then followed by a saline flush, which occurred during contrast-enhanced imaging. Total imaging time was 2 min 30 sec. Two sequential contrast-enhanced studies were performed.
The examinations were interpreted prospectively before reexcision surgery by one of two radiologists who knew the mammographic findings but not pathologic findings. Both radiologists have extensive experience in interpreting MRI examinations. One radiologist is a breast imager who is proficient in all techniques of breast imaging. The other radiologist specializes in MRI with a focus on breast MRI. A regular thin enhancing rim (1–2 mm) around the seroma cavity was considered negative for residual disease. MRI findings were considered positive if focal, thick, or irregular enhancement was identified around the postoperative biopsy cavity and was reported as suspicious for residual disease or if enhancement was seen to extend away from the biopsy cavity and was reported as suspicious for residual disease. The location and extent of suspicious enhancement were described in the examination report. Localization of suspicious enhancement in the surgical bed before reexcision was not requested by our referring surgeons.
Additional lesions were identified in some breasts separate from the biopsy cavity. They were characterized by the interpreting radiologist as benign, probably benign, or suspicious using an interpretation model that incorporates both architectural and kinetic features of breast lesions, which has been described previously [12, 13]. Briefly, architectural features that are highly predictive of malignant disease included spiculated borders and peripheral rim enhancement in the presence of central lesion enhancement. Architectural features that were highly predictive of benign disease included smooth or lobulated borders, the absence of mass enhancement, and nonenhancing internal septations. Classification of the enhancement curve and pattern was qualitative.
All suspicious lesions underwent histopathologic correlation. The biopsy methods were fine-needle aspiration; core needle biopsy guided by sonography or MRI; excisional biopsy guided by mammography, sonography, or MRI; or mastectomy. The method of MRI-guided needle localization of suspicious breast lesions performed at our institution has been previously described [14]. Scattered punctate (1–3 mm) foci of enhancement or areas of scattered patchy enhancement without architectural distortion were considered typical of benign glandular tissue and were not considered a positive result.
Data Analysis
Relevant radiology, operative, pathology, and clinical reports were
retrospectively examined after the patient had undergone definitive surgical
treatment. Data abstracted included characteristics of the study population
such as patient age, initial pathology results, and information relating to
the decision regarding definitive surgical treatment. The period of clinical
follow-up ranged from 14 to 42 months.
After reviewing the radiology and pathology reports regarding the evaluation for residual disease, the observers placed all cases into one of four categories: true-positive, true-negative, false-positive, or false-negative. A case was considered a true-positive if the prospective interpretation reported enhancement suspicious for residual disease that was also found at histologic examination. Similarly, an examination was considered a true-negative if no suspicious enhancement was reported and no residual disease was seen at histologic examination. An examination was classified as a false-positive if enhancement suspicious for residual disease was reported, but only benign breast disease was reported at pathology. An examination was classified as a false-negative if no suspicious enhancement was reported, but residual intraductal or invasive carcinoma was identified at histologic examination. On the basis of these four subgroups (Figs. 1, 2, 3, 4), we calculated the accuracy of MRI in evaluating for residual disease.
|
|
|
|
When an additional lesion separate from the excisional biopsy cavity was identified on MRI, it was reported as benign, probably benign, or suspicious. Benign lesions were not biopsied, and only routine clinical and mammographic follow-up was performed. Probably benign lesions were not biopsied. In addition to routine clinical and mammographic follow-up, a repeated MRI examination was performed approximately 6 months later. All suspicious lesions were biopsied, either before or at the time of reexcision. Pathology reports were reviewed to correlate the additional suspicious lesions seen on MRI with their histologic appearance (Figs. 5 and 6).
|
|
To assess the impact of MRI on clinical decision-making, we identified cases in which MRI changed the next procedure from reexcision lumpectomy to mastectomy, biopsy, or neoadjuvant chemotherapy. We noted for each breast which definitive surgical treatment was chosen and how many surgical procedures were required.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
|
Of the invasive ductal carcinomas, 15 (23.8%) of 63 had an extensive intraductal component, which was defined as greater than 25% involvement by intraductal carcinoma. Histologic grades of tumors are described in Table 1. The initial histologic grade was not specified in three of 48 cases in which a second review of slides submitted from an outside institution was performed. Of the invasive ductal carcinomas identified at the initial excisional biopsy, size ranged from microinvasive to 4.0 cm. A majority of carcinomas, 50 (79.4%) of 63, were T1 in stage; 12 (19.0%) of 63 carcinomas were T2 in stage. In one case, pathology slides were submitted from an outside institution for review, and the initial size was not specified.
Fifty-seven (69.5%) of 82 lesions had positive margins, with tumor transected at the margin of the specimen. Twenty-two (26.8%) of 82 lesions had close margins, defined as tumor less than 2 mm from the specimen margin. Two (2.4%) of 82 lesions had negative margins but residual suspicious calcifications in the surgical bed on a postbiopsy mammogram. Similarly, one patient also had residual suspicious calcifications in the surgical bed on a postbiopsy mammogram, but the margin status was not described in the available clinical records. Of the 15 lesions with DCIS at initial excision, seven had positive margins, five had close margins, two had negative margins with residual calcifications, and one lesion had an unspecified margin status. As we explained earlier, this patient had residual suspicious calcifications. Of the 63 breasts with invasive ductal carcinoma, 33 had invasive ductal carcinoma transected at the margins, and nine additional lesions had both invasive and DCIS transected at the margins. Six lesions had DCIS transected at the margins; four of these six patients also had invasive ductal carcinoma within 2 mm of the margin. Fifteen patients had close margins with invasive ductal carcinoma, DCIS, or both within 2 mm of the margin. All four patients with invasive lobular or adenoid cystic carcinoma had positive margins at initial excision.
The time from the initial excision to MRI of the breast ranged from 5 to 94 days (mean, 29 days). The exact date of initial excisional biopsy could not be determined in four patients who had their biopsy performed at an outside institution.
Diagnostic Accuracy of MRI in Evaluating for Residual Disease
Forty-nine of 82 breasts contained residual intraductal or invasive cancer
at histologic examination after reexcision lumpectomy or mastectomy, with the
prevalence being 59.8%. Information regarding the diagnostic accuracy of MRI
in evaluating for residual disease is summarized in
Table 2. The sensitivity and
specificity of MRI was 61.2% (30/49) and 69.7% (23/33), respectively. The
positive and negative predictive values were 75.0% (30/40) and 54.8% (23/42),
respectively.
|
Additional Lesions
Twenty-three additional lesions distant from the initial excision cavity
were identified in 19 breasts (19/82 [23.2%]). Five lesions were classified by
the interpreting radiologist as benign or probably benign, and no biopsy was
performed. During clinical follow-up through September 2002 (range,
37–41 months), no interval biopsies were performed and routine
mammographic examinations showed no suspicious lesions. All the lesions
interpreted as probably benign on the initial MRI were stable in appearance on
follow-up MRI. Eighteen lesions in 14 breasts were classified as suspicious by
the interpreting radiologist and underwent tissue sampling. Ten lesions in
eight breasts were seen only on MRI and underwent MRI-guided needle
localization and excisional biopsy. Of these 10 lesions, five were malignant
and five were benign. Four lesions were examined after mastectomy at
pathology, two lesions underwent mammographically guided needle localization
and excisional biopsy, one lesion underwent sonographically guided core
biopsy, and one lesion was sampled by fine-needle aspiration.
The pathology findings of additional lesions seen in the ipsilateral breast on MRI are as follows: Of the 18 suspicious lesions, six had malignant findings, with one case of DCIS and five cases of invasive ductal carcinoma. The overall prevalence of synchronous malignancy in the ipsilateral breast in this study was six (7.3%) of 82. The remaining 12 lesions were benign on histologic examination: four lesions with a fibroadenoma, two lesions with fibrocystic change, one lesion with fibrocystic change and a fibroadenoma, one lesion with fibrocystic change and an intraductal papilloma, and one lesion with stromal fibrosis. In the remaining three cases, benign breast tissue with no tumor was identified.
In three patients, MRI depicted suspicious lesions in the contralateral breast. One lesion underwent MRI-guided needle localization and excisional biopsy. When histology findings revealed that this lesion was DCIS, the patient underwent bilateral mastectomy for definitive surgical treatment. One lesion underwent MRI-guided core biopsy, which revealed invasive ductal carcinoma; the patient underwent bilateral mastectomy. One lesion underwent fine-needle aspiration at the same time as fine-needle aspiration of an additional lesion seen in the ipsilateral breast. Pathology revealed atypical ductal hyperplasia. The patient chose to undergo bilateral mastectomy. Histologic examination of the contralateral breast at mastectomy revealed atypical ductal hyperplasia and DCIS.
Impact on Management
Table 3 summarizes how MRI
changed the management of patients from the initially planned reexcision
lumpectomy. In 24 (29.3%) of 82 breasts, MRI resulted in a change of surgical
management from reexcision lumpectomy to mastectomy, neoadjuvant chemotherapy,
biopsy of an additional lesion in the ipsilateral breast, or biopsy of an
additional lesion in the contralateral breast. A change in the next procedure
was most commonly due to the identification of an additional suspicious lesion
in either breast. Of the nine patients whose next procedure was mastectomy,
six underwent mastectomy rather than breast conservation therapy because an
MRI showed extensive residual disease. Two of these nine patients had an
additional suspicious lesion seen on MRI. One patient with a primary tumor in
each breast had an MRI that showed residual disease bilaterally in addition to
residual calcifications on mammography. She underwent bilateral mastectomy as
the next procedure. In one patient, MRI showed two large areas of enhancement
suggestive of residual disease. The patient strongly desired breast
conservation therapy and decided to undergo neoadjuvant therapy with her
surgeon on the basis of MRI findings and clinical factors. Neoadjuvant therapy
was begun to improve the likelihood of successful breast conservation therapy,
which the patient eventually underwent.
|
MRI changed the next procedure from reexcision to biopsy of a lesion in the contralateral breast in two patients. One of these patients had residual suspicious calcifications in the excisional biopsy bed and in another location in the breast and a suspicious enhancing lesion on MRI. All three suspicious lesions underwent needle localization and surgical biopsy. The calcific lesions were DCIS and invasive ductal carcinoma, and the enhancing lesion on MRI proved to be fibrocystic change. One additional patient had needle localization and excisional biopsy of residual calcifications seen on mammography in the ipsilateral breast before excisional biopsy of a suspicious lesion in the contralateral breast seen on MRI and was not included in Table 3. Of the breasts that underwent reexcision lumpectomy only, 13 had close or positive margins requiring an additional surgical procedure, either additional reexcision or mastectomy, for definitive treatment.
The definitive surgical treatment for each breast in our study is summarized as follows: 21 patients (25.6%) underwent mastectomy, 51 patients (62.2%) underwent reexcision lumpectomy, and 10 patients (12.2%) had close or positive margins at reexcision requiring an additional reexcision lumpectomy to obtain adequate negative margins. Of the 21 breasts treated with mastectomy, clinical and MRI findings were the primary factors driving the decision to treat with mastectomy in nine patients, because an interval biopsy was not performed. Five patients underwent mastectomy after residual disease with positive margins was identified at reexcision lumpectomy. Four patients underwent mastectomy for biopsy-proven multicentric or multifocal disease, and three patients underwent mastectomy because of cancer in the contralateral breast. Of these patients, two had biopsy-proven cancer in the contralateral breast that was initially identified on MRI. One patient had atypical ductal hyperplasia on fine-needle aspiration of the contralateral lesion, and she opted for bilateral mastectomy for definitive surgical treatment. DCIS was found in the contralateral breast at pathologic examination.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Frei et al. [21] found that the time interval between lumpectomy and MRI influenced the specificity of MRI, with the greatest specificity of 75% being achieved when MRI was performed between 28 and 35 postoperative days. The specificity in this study of approximately 70% that was attained with a mean interval from surgery to MRI of 29 days is consistent with those findings and reflects an improved specificity compared with that reported by Orel et al. [20]. In that study [20], the excision-to-MRI interval averaged 18 days. The median interval from excision to MRI in our study population was 24 days, and 48 (58.5%) of 82 breasts were imaged within 28 days. It is possible that our specificity could be improved slightly if more patients were imaged at least 28 days after surgery. The time to imaging, however, is also influenced by factors such as the scheduling of reexcision surgery and other decisions made by the patient and her surgeon.
Although overlap in the appearances of benign and malignant lesions limits MRI evaluation for residual disease, MRI can depict additional suspicious lesions likely to represent multicentric or multifocal disease; these findings often alter surgical management. In this study, MRI showed clinically and mammographically occult multifocal or multicentric cancer in 7.3% of breasts (6/82). The positive predictive value for an additional suspicious lesion was 33.3% (6/18). In other studies that document the presence of multifocal or multicentric breast cancer detected on MRI, this value ranged from 16% to 45% [20, 22–27]. This wide range may reflect differences in patient populations among studies. For instance, in studies by Harms et al. [22], Boetes et al. [24], and Mumtaz et al. [26], the majority of index tumors were larger, averaging greater than 2.0 cm, and may have been more likely to have been associated with multifocal or multicentric disease. In the study by Mumtaz et al., 66 of 90 patients presented with a palpable mass or skin and nipple changes. Rodenko et al. [25] studied 20 patients, all of whom had invasive lobular cancer.
When examining MRI detection of additional cancers that were mammographically occult, reported detection rates are more comparable to that reported in this study. Mumtaz et al. [26] reported multifocal and multicentric cancer in 25 (27.2%) of 92 patients. This finding was mammographically evident in 11 patients, and MRI revealed mammographically occult foci in 10 additional patients (10/92 [10.9%]). Fischer et al. [27], who studied the addition of MRI to clinical examination, mammography, and sonography in evaluating suspicious lesions, reported that 54 additional malignant lesions in 336 patients (16.1%) were seen only on MRI. Harms et al. [22] reported that MRI showed additional mammographically occult multicentric disease in 10 breasts in 30 patients (33.3%).
Other studies of mastectomy specimens in patients with stage I or II breast cancer have reported rates of multicentric cancer to be 21–43% [28–31]. The current treatment of multifocal cancer is extended excision or mastectomy, and the current treatment of multicentric cancer is mastectomy rather than breast conservation therapy. Thus, the identification of an additional suspicious lesion in the ipsilateral breast can change initially planned surgical treatment. However, 12 (66.7%) of 18 suspicious additional lesions identified in this study proved to be benign disease at histologic examination. In our study, there were three patients whose next procedure after MRI was mastectomy that showed no tumor at pathology. To avoid performing unnecessary mastectomies, we strongly recommend correlation with histology before definitive surgical treatment when MRI depicts additional suspicious lesions.
Although performing percutaneous biopsy of suspicious lesions identified on MRI is optimal, it is not always possible. At our institution, when a previously unsuspected lesion is identified on MRI, correlation with mammography and directed sonography is performed. If the lesion can be seen on sonography, percutaneous biopsy is performed using sonographic guidance. If the lesion is visible only on MRI, we perform MRI-guided needle localization followed by excisional biopsy for histopathologic correlation [14]. At this time, MRI-guided percutaneous core biopsy is not routinely performed. As MRI-compatible equipment is developed and integrated into clinical trials, we anticipate that less-invasive MRI-guided percutaneous biopsies will decrease the number of surgical biopsies performed for benign breast disease, which will result in fewer surgical procedures before definitive surgery and improved cosmetic outcome for patients desiring breast conservation therapy. Decreased costs and morbidity are also likely.
When this study was designed, excisional biopsy was the standard method of obtaining tissue from suspicious breast lesions. Over time, many suspicious breast lesions are undergoing percutaneous biopsy, and more primary breast cancers are being diagnosed in this manner. MRI of the breast in patients with breast cancer diagnosed at percutaneous biopsy before breast conservation therapy is beyond the scope of this study. This topic is currently being studied at our institution.
In 29.3% of breasts (24/82), MRI findings resulted in a change in management from reexcision lumpectomy to mastectomy, biopsy of an additional lesion in the ipsilateral or contralateral breast, or neoadjuvant therapy. Twenty (26.3%) of 76 breasts were eventually managed with mastectomy as the definitive surgical treatment. Fischer et al. [27] also found that MRI findings led to a change in treatment in 19.6% of patients (66/336).
Although MRI can result in changes in surgical management, its ultimate impact on patient outcome is still uncertain. This uncertainty is due to limited understanding of the clinical significance of subclinical disease in breast conservation therapy candidates—that is, additional cancer that is not detectable at physical examination and on conventional mammography at the time of diagnosis of the index cancer. The assumption is that removal of all identifiable cancer at the time of diagnosis improves prognosis. However, subclinical disease is known to exist from pathologic and clinical trials [28–31]. Despite this, when long-term follow-up (as long as 20 years) results of patients who underwent breast conservation therapy were compared with those of patients who underwent mastectomy for the treatment of early-stage breast cancer, no significant differences in survival were found [32, 33]. Long-term follow-up has also been available for patients diagnosed with DCIS [11]. Researchers have found the prognosis in patients treated with breast conservation therapy to be favorable. Both radiation therapy and adjuvant systemic chemotherapy contribute to the relatively low rates of recurrence of cancer in the treated breast, and the extent to which subclinical disease is treated is uncertain. When subclinical disease becomes clinically significant and when subclinical disease is no longer controlled by adjuvant therapy are unknown.
An important clinical role of MRI will likely be to identify a subset of patients with extensive residual or multicentric cancer who would be better served by undergoing mastectomy rather than breast conservation therapy. MRI may also be of greater value in patients with breast cancer diagnosed at core biopsy before definitive surgical treatment in whom the aim is to aid in planning a single definitive surgical treatment procedure. Negative findings on MRI examination have also been suggested to possibly have significant clinical value. Harms [34] noted that in study B-06 of the National Surgical Adjuvant Breast Project (NSABP-06 trial) [32], 60% of patients treated with breast conservation therapy without radiation did not develop recurrence. Harms suggested that MRI of the breast may allow differentiation of women who would benefit from radiation therapy from those who would not. Randomized clinical trials with long-term follow-up that includes MRI integrated into the diagnostic algorithm are necessary before the current treatment paradigms can be reevaluated or revised.
The use of MRI as an aid to surgical treatment planning results in increased short-term costs related to the cost of MRI and costs associated with biopsy of additional suspicious lesions, most of which are have benign histology. Whether the increased short-term costs are offset by decreased costs associated with lower rates of ipsilateral breast recurrence and salvage therapy is uncertain. Esserman et al. [35] performed a preliminary cost analysis of using MRI as a staging tool. These researchers estimated potential cost savings of approximately $1,800 (1995 U.S. dollars) per case when MRI was used. However, their cost analysis did not take into account the increased costs of tissue sampling for additional suspicious lesions before definitive surgical treatment. To our knowledge, no cost-effectiveness analyses to date have examined the role of MRI in staging breast cancer. Additional studies are needed to better define both the optimal clinical role of MRI and the role of MRI in cost-effective detection and treatment of breast cancer.
In summary, we found that the overlap of the appearances of benign and malignant disease limits the accuracy of MRI in evaluating for the presence and extent of residual cancer in patients after excisional biopsy. However, MRI can be used to identify additional lesions that are suspicious for multifocal or multicentric cancer, and these findings can alter surgical treatment planning. Additional research is needed to understand the significance of subclinical disease and to aid in defining the optimal role of MRI in breast cancer treatment.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  C. C. Genson, C. E. Blane, M. A. Helvie, S. A. Waits, and T. L. Chenevert Effects on Breast MRI of Artifacts Caused by Metallic Tissue Marker Clips Am. J. Roentgenol., February 1, 2007; 188(2): 372 - 376. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. K. Shah, S. K. Shah, and K. V. Greatrex Current Role of Magnetic Resonance Imaging in Breast Imaging: A Primer for the Primary Care Physician J Am Board Fam Med, November 1, 2005; 18(6): 478 - 490. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. A. Newman and H. M. Kuerer Advances in Breast Conservation Therapy J. Clin. Oncol., March 10, 2005; 23(8): 1685 - 1697. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
