| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1
Department of Obstetrics and Gynecology, Kantonales Frauenspital Fontana,
Lürlibadstr. 118, 7000 Chur, Switzerland.
2
Department of Radiology, Hopiteaux Universitaires de
Genève, Hopital Cantonal, Rue Micheli/du
Crest, Ch-1211 Genève 14, Switzerland.
3
Department of Radiology, University of California, Magnetic Resonance Science
Center, Box 1290, 1 Irving St., San Francisco, CA 94143-1290.
Received March 22, 2000;
accepted after revision May 15, 2000.
Supported in part by grants from the United States Department of Defense
and the National Institutes of Health.
Abstract
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
MATERIALS AND METHODS. Sixty-eight patients who had undergone excisional biopsy with positive resection margins underwent MR imaging for evaluation of residual breast cancer and possible breast conservation. Patients were retrospectively stratified according to the time interval between lumpectomy and MR imaging. Dynamic and morphologic enhancement features were used for lesion characterization. Imaging findings were correlated with results of histopathology. Sensitivity, specificity, positive predictive value, and negative predictive value were calculated for patients waiting 7, 14, 21, 28, 35, and 42 days after initial surgery before undergoing MR imaging of the breast.
RESULTS. The time interval between lumpectomy and MR imaging of the breast had the greatest influence on the specificity and negative predictive value of MR imaging, increasing progressively over time. A plateau of highest values of 75% specificity and 86% negative predictive value was reached at 28 and 35 days after surgery, respectively. Although the sensitivity and positive predictive value showed smaller variations over time, peak values of 95% sensitivity and 92% positive predictive value were obtained at 35 and 28 days after surgery, respectively.
CONCLUSION. We recommend scheduling patients with positive resection margins no earlier than 28 days after initial surgery for evaluation of residual cancer using MR imaging of the breast.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
MR imaging has been suggested as an adjunctive imaging technique to identify the extent of breast carcinoma and thereby guide treatment planning [7]. MR imaging has therefore been proposed to assess the extent of residual disease in women after excisional biopsy with positive resection margins, particularly in women who have not undergone MR imaging preoperatively to identify unsuspected multifocal disease [8].
The diagnostic evaluation of the postoperative breast with mammography or sonography is frequently difficult because inflammatory changes may obscure or mimic malignancy [8,9,10]. In MR imaging, postsurgical contrast enhancement resulting from inflammatory changes at the site of surgery is known to limit the positive predictive value and specificity for the diagnosis of residual disease [8]. The optimal time interval for evaluation of residual disease between surgery and MR imaging is disputed in the literature [8, 11, 12].
The purpose of this study was to evaluate the influence of the time interval between lumpectomy and MR imaging of the breast on the performance of MR imaging in the diagnosis of residual disease in patients with positive resection margins and to suggest a feasible waiting time interval for scheduling patients after lumpectomy.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
MR Imaging Technique
All MR imaging examinations were performed on a Signa scanner (General
Electric Medical Systems; Milwaukee, WI) at 1.5 T. All patients signed an
informed consent form and then were imaged in a prone position with a
dedicated double breast coil. The breast with the known breast cancer was
imaged. No compression device was applied. An IV cannula was connected to an
automatic injector before the examination. A transverse T1-weighted spin-echo
sequence was acquired for localization purposes followed by a sagittal
fat-suppressed T2-weighted fast spin-echo sequence with the following
parameters: TR/TE, 4600/84; field of view, 18 cm; matrix size, 256 x
192; slice thickness, 3.0 mm with a 0.5-mm gap; and signals averaged, two. A
three-dimensional sagittal fat-suppressed T1-weighted fast gradient-recalled
echo sequence (8.2/4.2; field of view, 18 cm; matrix size, 256 x 192;
slice thickness, 2.0 mm with no gap; acquisitions averaged, two; spectrally
selected inversion recovery-prepared fat suppression) was obtained before, at
2 min 30 sec, and at 7 min 30 sec after a bolus injection of 0.1 mmol/kg of
body weight of gadolinium dimeglumine (Magnevist; Berlex Laboratories, Wayne,
NJ). Maximum-intensity-projection images were created from unenhanced and
early and late contrast-enhanced three-dimensional fast gradient-recalled echo
data sets, creating projection images analogous to mammograms.
Image Interpretation
Sixty-eight MR imaging examinations were retrospectively reviewed with
consensus by two radiologists experienced in the interpretation of breast MR
imaging examinations. The image interpretation was performed without awareness
of histopathologic findings. Lesion characteristics at the site of surgery
were assessed and reported as positive or negative for residual cancer.
The surgical site was differentiated into three morphologic types of appearance: cavity, tract, and scar. The size of the cavity, thickness of the cavity rim, size of the scar, and thickness of the tract were measured. Lesions at the surgical site were classified as having either masslike or non-masslike enhancement. All morphologic features were analyzed on the basis of the first contrast-enhanced images, acquired at 2 min 30 sec after contrast injection, using a recently developed breast MR imaging lexicon (Ikeda DM et al., presented at the Radiological Society of North America meeting November/December 1999). Kinetic parameters, including the enhancement degree and pattern of enhancement, were assessed visually. The degree of enhancement was defined as referring to the most strongly enhancing portion of the lesion identified on the first contrast-enhanced image. Variables of enhancement degree were rated qualitatively and relative to the signal increase in adjacent blood vessels on an ordinal scale as absent, low, mild, moderate, or strong. The enhancement pattern was assessed visually by comparing the signal intensity on the first and second set of dynamic high-spatial-resolution images, acquired at 2 min 30 sec and 7 min 30 sec, respectively, and presented with identical window width and level settings. By definition, any decrease in signal intensity between the first and second contrast-enhanced images was interpreted as a washout enhancement pattern. The term "plateau enhancement" was used if a stabilized enhancement without any change in signal intensity occurred between 2 min 30 sec and 7 min 30 sec. A "progressive enhancement" was considered to be an increase in signal intensity from 2 min 30 sec to 7 min 30 sec. Each lesion was classified according to the strongest enhancement pattern seen over the entire lesion in the order of decreasing strength: washout followed by plateau and progressive.
For the characterization of the surgical site, a regular enhancing rim measuring as large as 0.5 cm was considered negative for residual disease, and an irregular or nodular and thicker rim (measuring >0.5 cm) was considered malignant [8]. Tracts and scars were assessed as positive for residual disease if a distinct lesion was abnormally enhancing within 2 cm of the surgical site. If there was no further enhancement of tissue surrounding the biopsy site, the site was considered negative for residual disease. Abnormal enhancement was defined as any enhancing spiculated focal mass; any nonsmooth focal mass with coexisting washout enhancement pattern according to a recently developed diagnostic algorithm [13]; or any non-masslike enhancement, regional, segmental, or ductal enhancement with coexisting washout enhancement pattern.
Histopathology and MR Imaging Correlation
After MR imaging was performed, 44 patients underwent reexcision and 24
patients underwent mastectomy. Histopathologic correlation between MR images
and histopathologic slides was performed according to standard histopathologic
procedures at our institution after interpretation of breast MR images. All
specimens were anatomically oriented during surgery to distinguish the
mediolateral, inferior-superior, and anteroposterior border. A drawing that
showed the origin of the specimen in the breast was provided during surgery.
Specimens were serially sectioned medially to laterally in a sagittal
orientation. The histopathologic slides were numbered according to their
positions in the specimen. Correlation between MR images and histopathology
was possible for all surgical sites and lesions within an area of 2 cm around
the surgical site, defined by the extent of tissue removal routinely performed
at our institution.
Data Analysis and Statistics
To evaluate the performance of breast MR imaging as a function of the time
interval between lumpectomy and MR imaging of the breast, we calculated the
positive predictive value, negative predictive value, sensitivity, and
specificity, subdividing our patient population into different patient groups:
patients scheduled 7 days or later after initial surgery (n = 68),
patients scheduled 14 days or later (n = 61), 21 days or later
(n = 45), 28 days or later (n = 41), 35 days or later
(n = 34), and 42 days or later (n = 25). Patients imaged
before the cutoff time were excluded from the evaluation of the diagnostic
performance of MR imaging of the breast for each defined patient group.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Residual cancer corresponded to suspicious masslike enhancement with either spiculated margins or washout enhancement pattern in 13 of 47 patients and to suspicious nonmasslike enhancement in 27 of 47 patients (Fig. 2A,2B,2C). Lesions in 16 of the 27 patients with suspicious non-masslike enhancement showed regional enhancement, lesions in three showed ductal enhancement, and lesions in three showed segmental enhancement. All lesions showed a washout enhancement pattern. In two of 47 patients with residual cancer, a cavity with a nodularly enhancing rim measuring more than 0.5 mm in thickness was the only imaging sign of residual cancer.
|
|
|
False-negative MR imaging results were seen in five patients without any enhancing lesion at the surgical site. The surgical site in these five patients was described as a cavity with an enhancing rim measuring less than 5 mm in thickness in three patients and as a mildly enhancing scar in two patients. The time interval between surgery and MR imaging was 16, 17, 22, 34, and 78 days, respectively. Corresponding histopathologic findings in the three patients with thin rim enhancement were invasive lobular carcinoma measuring 1.8 cm and 0.1 cm in diameter, and microscopic foci of ductal carcinoma in situ in an area of 2 cm. A mildly enhancing spiculated scar without evidence of a suspicious lesion in two patients corresponded to infiltrating lobular carcinoma measuring 0.2 cm and to ductal carcinoma in situ measuring 0.5 cm at histopathology.
Among 21 patients with no residual cancer at histopathology, two patients showed nonsuspicious enhancement at the biopsy site, and nine patients showed no enhancement at all. Nonsuspicious enhancement corresponded to regional enhancement with a progressive enhancement pattern in one patient and to segmental enhancement with a progressive enhancement pattern in another patient. The surgical site of the nine patients with no enhancement at the biopsy site was described as a cavity with a thin enhancing rim measuring less than 5 mm in thickness in three patients. One patient had a biopsy cavity with no rim (Fig. 3A,3B,3C). A mildly enhancing scar was described in two patients, and a tract was described in three patients.
|
|
|
Among 10 false-positive imaging findings, histopathology diagnosed foreign body reaction in two patients, fat necrosis in two, and fibrocystic changes in six (Fig. 4A,4B,4C). The time interval between initial surgery and false-positive MR imaging findings was 14 days in two patients, 15 days in two patients, and 17, 19, 23, 27, 43, and 50 days in the remaining patients (Fig. 5A,5B,5C). Corresponding MR imaging findings were masslike enhancement with either spiculated borders or washout enhancement pattern or both in five patients, regional enhancement showing washout in two patients, and segmental enhancement showing washout in one patient. In two patients, a biopsy cavity with a nodularly enhancing rim measuring 3 and 3.8 cm in thickness was described and therefore assessed positive for residual cancer.
|
|
|
|
|
|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
MR imaging of the breast after surgery, unlike mammography, is not limited to the evaluation of lesions that contain microcalcifications. Studies comparing the sensitivity of MR imaging and mammography have shown that MR imaging can depict mammographically visible tumors and mammographically occult tumors [15]. One advantage of MR imaging over mammography in postsurgical patients is that no compression of the breast is necessary, and thus the imaging of the postsurgical breast is painless. Early reports of strong enhancement in postsurgical breasts due to inflammatory changes at the site of surgery did not recommend the use of MR imaging after surgery [11]. More recently, high-spatial-resolution MR imaging has shown persistent false-positive enhancement due to granulation tissue [8].
The optimal time interval between surgery and MR imaging is disputed [11, 12]. Different opinions on how soon postoperative changes can be accurately differentiated from residual disease have been reported in the literature. Heywang-Köbrunner et al. [11] found that MR imaging was not helpful in differentiating benign from recurring lesions before 9 months after surgery, whereas Orel et al. [8] evaluated patients between 6 and 40 days (mean, 18 days) after initial excisional biopsy and achieved a positive predictive value of 82% and a negative predictive value of 61%. Soderstrom et al. [12] evaluated a time interval within 10 months after surgery and assessed the presence or absence of residual disease correctly in 84% of patients. Our hypothesis is that postsurgical MR enhancement and the accuracy of predicting residual disease are related to the time interval between surgery and MR imaging. The logic of this hypothesis is based on the natural history of healing after surgery, of which there are at least two components. First, nonspecific inflammation develops, peaks, and resolves. Second, a pattern of anatomic response to the surgical procedure emerges and becomes distinctive. Better criteria for inflammation and its resolution and criteria for surgically induced anatomic alteration still have to be defined and differentiated from those of residual cancer.
The results of our study show that the positive predictive value reached a peak value of 92% in the patient population waiting 28 days after initial surgery to undergo MR imaging of the breast. In our series with a time interval of more than 28 days between surgery and MR imaging, false-positive results due to postsurgical changes were reduced. Eight false-positive results were reported in patients undergoing MR imaging of the breast before 28 days after surgery. However, false-positive findings due to fibrocystic changes can also be seen independently; two patients had false-positive imaging findings due to fibrocystic changes at 43 and 50 days after surgery. Unfortunately, until better diagnostic MR imaging criteria for the differentiation between fibrocystic changes and residual carcinoma, particularly ductal carcinoma in situ, are available, false-positive findings cannot be avoided.
False-negative results for residual disease were diagnosed in five patients with either invasive lobular cancer or ductal carcinoma in situ. Although the sensitivity of MR imaging is known to be lower for these tumor types than for invasive ductal cancer, another explanation for false-negative findings includes early post-surgical changes seen within 1 and 3 weeks after surgery, which might obscure a small enhancing residual tumor [8].
With a peak value for the positive predictive value and plateau values for specificity at 28 days after surgery, our results suggest a better performance of MR imaging in patients scheduled for more than 28 days after surgery. MR imaging of the breast performed 28 days after excisional biopsy or lumpectomy appears to represent a feasible time interval, allowing reduction of swelling and consolidation of the biopsy cavity. It seems unreasonable to delay necessary surgical treatment for more than 4-5 weeks considering the marginal gain of test performance of MR imaging of the breast more than 28 days after surgery, especially in patients with fast-growing tumors. Information of the extent and location of residual disease might considerably influence the surgeon's choice between reexcision and mastectomy. Additionally, MR imaging may identify areas for specific surgical attention and offer the surgeon a "road map" at reexcision.
Reexcision at the biopsy site is commonly performed if the initial margins of resection are positive. Neither positive nor negative margins of resection are reliable indicators of the completeness of surgical excision, probably because of the difficulty in assessing all margins of resection [16,17,18]. In cases in which the margins of the initial excisional biopsy are positive, reported rates of finding residual tumor are as high as 69% [4, 5, 6, 16]. The positive predictive value for positive margins in our study was 69% (47/68)—a similar rate—whereas the positive predictive value of MR imaging of 92% for patients who waited 28 days before undergoing MR imaging of the breast was considerably higher and might therefore offer another reliable indicator for residual disease.
One limitation of our study is the relatively small sample size that did not allow us to show statistically significant differences in test performance in the different patient groups. However, there was a trend toward greater specificity in patients who had a longer time interval between initial surgery and MR imaging. This finding suggests a possible relationship between the postoperative time interval and MR accuracy. This relationship confirms our hypothesis that the time interval matters and that MR imaging of the breast too soon after surgery should be avoided.
Because of the overlapping enhancement characteristics of benign and malignant lesions, research efforts have recently concentrated on defining more specific diagnostic criteria for lesion characterization [19]. Whereas the relative importance of either high spatial or high temporal resolution MR imaging of the breast remains controversial, our group used defined diagnostic criteria combining morphologic and kinetic enhancement patterns. In our study, many of a the false-positive findings were related to the assessment of a washout enhancement pattern. Further prospective multicenter studies working with defined diagnostic criteria, including quantitative analysis of the enhancement pattern, could allow a larger sample size and might yield statistically significant results and a higher specificity.
On the basis of our results, we recommend scheduling patients with positive resection margins after excisional biopsy or lumpectomy no earlier than 28 days after the initial surgery for evaluation of residual disease and possible breast-conserving surgery by MR imaging.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  K. J. Macura, R. Ouwerkerk, M. A. Jacobs, and D. A. Bluemke Patterns of Enhancement on Breast MR Images: Interpretation and Imaging Pitfalls RadioGraphics, November 1, 2006; 26(6): 1719 - 1734. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A Khiat, D Gianfelice, M Amara, and Y Boulanger Influence of post-treatment delay on the evaluation of the response to focused ultrasound surgery of breast cancer by dynamic contrast enhanced MRI. Br. J. Radiol., April 1, 2006; 79(940): 308 - 314. [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]
|
|
|
|
 |
|
 F. Sardanelli, G. M. Giuseppetti, P. Panizza, M. Bazzocchi, A. Fausto, G. Simonetti, V. Lattanzio, and A. Del Maschio Sensitivity of MRI Versus Mammography for Detecting Foci of Multifocal, Multicentric Breast Cancer in Fatty and Dense Breasts Using the Whole-Breast Pathologic Examination as a Gold Standard Am. J. Roentgenol., October 1, 2004; 183(4): 1149 - 1157. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. M. Lee, S. G. Orel, B. J. Czerniecki, L. J. Solin, and M. D. Schnall MRI Before Reexcision Surgery in Patients with Breast Cancer Am. J. Roentgenol., February 1, 2004; 182(2): 473 - 480. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. S. Hwang, K. Kinkel, L. J. Esserman, Y. Lu, N. Weidner, and N. M. Hylton Magnetic Resonance Imaging in Patients Diagnosed With Ductal Carcinoma-In-Situ: Value in the Diagnosis of Residual Disease, Occult Invasion, and Multicentricity Ann. Surg. Oncol., May 1, 2003; 10(4): 381 - 388. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
E. A. Morris, L. Liberman, D. D. Dershaw, J. B. Kaplan, L. R. LaTrenta, A. F. Abramson, and D. J. Ballon Preoperative MR Imaging--Guided Needle Localization of Breast Lesions Am. J. Roentgenol., May 1, 2002; 178(5): 1211 - 1220. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
, specificity =
, PPV = 
, NPV = X.
