Wendy B. DeMartini

Quantitative Diffusion-Weighted Imaging as an Adjunct to Conventional Breast MRI for Improved Positive Predictive Value

OBJECTIVE The purpose of our study was to investigate whether adding diffusion-weighted imaging (DWI) to dynamic contrast-enhanced MRI (DCE-MRI) could improve the positive predictive value (PPV) of breast MRI. MATERIALS AND METHODS The retrospective study included 70 women with 83 suspicious breast lesions on DCE-MRI (BI-RADS 4 or 5) who underwent subsequent biopsy. DWI was acquired during clinical breast MRI using b = 0 and 600 s/mm(2). Apparent diffusion coefficient (ADC) values were compared for benign and malignant lesions. PPV was calculated for DCE-MRI alone (based on biopsy recommendations) and DCE-MRI plus DWI (adding an ADC threshold) for the same set of lesions. Results were further compared by lesion type (mass, nonmasslike enhancement) and size. RESULTS Of the 83 suspicious lesions, 52 were benign and 31 were malignant (11 ductal carcinoma in situ [DCIS], 20 invasive carcinoma). Both DCIS (mean ADC, 1.31 +/- 0.24 x 10(-3) mm(2)/s) and invasive carcinoma (mean ADC, 1.29 +/- 0.29 x 10(-3) mm(2)/s) exhibited lower mean ADC than benign lesions (1.70 +/- 0.44 x 10(-3) mm(2)/s, p < 0.001). Applying an ADC threshold of 1.81 x 10(-3) mm(2)/s for 100% sensitivity produced a PPV of 47% versus 37% for DCE-MRI alone, which would have avoided biopsy for 33% (17/52) of benign lesions without missing any cancers. DWI increased PPV similarly for masses and nonmasslike enhancement and preferentially improved PPV for smaller (< or = 1 cm) versus larger lesions. CONCLUSION DWI shows potential for improving the PPV of breast MRI for lesions of varied types and sizes. However, considerable overlap in ADC of benign and malignant lesions necessitates validation of these findings in larger studies.

Differential diagnosis of mammographically and clinically occult breast lesions on diffusion‐weighted MRI

To investigate the diagnostic performance of diffusion‐weighted imaging (DWI) for mammographically and clinically occult breast lesions.

Apparent Diffusion Coefficient Values for Discriminating Benign and Malignant Breast MRI Lesions: Effects of Lesion Type and Size

OBJECTIVE The purpose of this study was to determine whether lesion type and size affect discrimination of benign and malignant breast lesions with diffusion-weighted MRI. MATERIALS AND METHODS This study included 91 women with 116 breast lesions identified with dynamic contrast-enhanced MRI. Diffusion-weighted images were acquired during clinical breast MRI at b values of 0 and 600 s/mm(2). Differences in the apparent diffusion coefficients (ADCs) of benign and malignant lesions were compared by lesion type (mass or nonmasslike enhancement) and size (<or= 1 cm or > 1 cm), and receiver operating characteristics analysis was performed to evaluate diagnostic performance based on ADC thresholds. RESULTS Sixteen of 71 masses and 13 of 45 lesions with nonmasslike enhancement were malignant. The mean ADC was significantly lower for malignant than for benign lesions for both masses (mean difference, 0.49 x 10(-3) mm(2)/s; p < 0.001) and lesions with nonmasslike enhancement (mean difference, 0.20 x 10(-3) mm(2)/s; p = 0.02). The area under the receiver operating characteristics curve (AUC) was greater for masses (AUC, 0.80) than for lesions with nonmasslike enhancement (AUC, 0.66). The mean ADC for malignant masses (1.25 x 10(-3) mm(2)/s) was lower than that for malignant lesions with nonmasslike enhancement (1.41 x 10(-3) mm(2)/s; p = 0.07). The median lesion size was 1.1 cm (range, 0.5-8.3 cm); 45 of 71 masses (63%) measured 1 cm or smaller, and 37 of 45 lesions with nonmasslike enhancement (82%) were larger than 1 cm. There was no relation (p > 0.05) between ADC value and lesion size for benign or malignant lesions, and there were no differences in AUC based on lesion size (p > 0.05). CONCLUSION Diffusion-weighted MRI shows promise in differentiation of benign and malignant masses and lesions with nonmasslike enhancement found at breast MRI and is not affected by lesion size. However, ADC measurements may be more useful for discriminating masses than for discriminating lesions with nonmasslike enhancement.

BI-RADS Lesion Characteristics Predict Likelihood of Malignancy in Breast MRI for Masses But Not for Nonmasslike Enhancement

OBJECTIVE The purpose of our study was to evaluate the predictive features of BI-RADS lesion characteristics and the risk of malignancy for mammographically and clinically occult lesions detected initially on breast MRI. MATERIALS AND METHODS We reviewed 1,523 consecutive breast MRI examinations performed from January 1, 2003, to June 30, 2005, to identify all lesions initially detected on MRI and assessed as BI-RADS 4 or 5 for which the patient underwent subsequent imaging-guided needle or excisional biopsy. BI-RADS lesion features were recorded for each case, and the risk of malignancy was assessed using generalized estimating equations. Separate multivariate models were constructed for lesions classified as masses. RESULTS Included in the analysis were 258 suspicious lesions in 196 women. Among all lesions, those of 1 cm or greater were significantly more often malignant (50/147, 34%) than lesions of less than 1 cm (22/111, 20%; odds ratio, 2.09; 95% CI, 1.13-3.83). For masses, size, BI-RADS margin, and enhancement pattern predicted malignancy. In multivariate analysis of combinations of features, masses of 1 cm or greater with heterogeneous enhancement and irregular margins had a 68% probability of malignancy. Masses of 1 cm or greater with smooth margins and homogeneous enhancement had the lowest predicted probability of malignancy of 3%. BI-RADS descriptors and size were not significant predictors of malignancy for nonmasslike enhancement (NMLE). CONCLUSION Combinations of BI-RADS lesion descriptors can predict the probability of malignancy for breast MRI masses but not for NMLE. If our model is validated, masses with a low probability of malignancy may be eligible for short-interval follow-up rather than biopsy. Further research focused on predictive features of NMLE is needed.

Utility of Targeted Sonography for Breast Lesions That Were Suspicious on MRI

OBJECTIVE Data remain sparse to guide the use of targeted sonography for suspicious breast lesions found with MRI. The purpose of our study was to determine the usefulness of sonography in this setting. MATERIALS AND METHODS Retrospective review of our breast MRI database was performed to identify all nonpalpable, mammographically occult MRI-detected suspicious lesions from January 1, 2003, to December 30, 2004, that underwent targeted sonography followed by imaging-guided biopsy. Frequency of sonography detection was determined, and MRI features (BI-RADS lesion type and size), histopathology outcomes, and positive biopsy rates of sonographically detected and occult findings were compared using chi-square analysis. RESULTS Targeted sonography was performed for 167 of 201 MRI lesions. Outcomes were benign in 120 (72%) and malignant in 47 (28%) findings. Lesions assessed on sonography were 84 (50%) masses, 53 (32%) nonmasslike enhancements, and 30 (18%) foci, with mean size of 16 mm. Overall, 76 of 167 (46%) lesions were detected with sonography. Depiction was significantly more frequent for masses (58%) than for nonmasslike enhancement (30%) or foci (37%) (p = 0.003). Lesion size did not affect the frequency of detection. Sonographically depicted findings had a significantly higher positive biopsy rate (36%) than sonographically occult lesions (22%) (p = 0.05). CONCLUSION Targeted sonography detected nearly half (46%) of MRI lesions evaluated, with depiction being most frequent for masses. Sonographically occult lesions had a probability of malignancy (22%), which warrants biopsy despite lack of sonographic detection.

Diffusion tensor MRI: preliminary anisotropy measures and mapping of breast tumors.

To investigate whether diffusion tensor imaging (DTI) measures of anisotropy in breast tumors are different from normal breast tissue and can improve the discrimination between benign and malignant lesions.

Positive Predictive Value of BI-RADS MR Imaging

PURPOSE To evaluate the positive predictive values (PPVs) of Breast Imaging and Reporting Data Systems (BI-RADS) assessment categories for breast magnetic resonance (MR) imaging and to identify the BI-RADS MR imaging lesion features most predictive of cancer. MATERIALS AND METHODS This institutional review board-approved HIPAA-compliant prospective multicenter study was performed with written informed consent. Breast MR imaging studies of the contralateral breast in women with a recent diagnosis of breast cancer were prospectively evaluated. Contralateral breast MR imaging BI-RADS assessment categories, morphologic descriptors for foci, masses, non-masslike enhancement (NMLE), and kinetic features were assessed for predictive values for malignancy. PPV of each imaging characteristic of interest was estimated, and logistic regression analysis was used to examine the predictive ability of combinations of characteristics. RESULTS Of 969 participants, 71.3% had a BI-RADS category 1 or 2 assessment; 10.9%, a BI-RADS category 3 assessment; 10.0%, a BI-RADS category 4 or 5 assessment; and 7.7%, a BI-RADS category 0 assessment on the basis of initial MR images. Thirty-one cancers were detected with MR imaging. Overall PPV for BI-RADS category 4 and 5 lesions was 0.278, with 17 cancers in patients with a BI-RADS category 4 lesion (PPV, 0.205) and 10 cancers in patients with a BI-RADS category 5 lesion (PPV, 0.714). Of the cancers, one was a focus, 17 were masses, and 13 were NMLEs. For masses, irregular shape, irregular margins, spiculated margins, and marked internal enhancement were most predictive of malignancy. For NMLEs, ductal, clumped, and reticular or dendritic enhancement were the features most frequently seen with malignancy. Kinetic enhancement features were less predictive of malignancy than were morphologic features. CONCLUSION Standardized terminology of the BI-RADS lexicon enables quantification of the likelihood of malignancy for MR imaging-detected lesions through careful evaluation of lesion features. In particular, BI-RADS assessment categories and morphologic descriptors for masses and NMLE were useful in estimating the probability of cancer.

Background Parenchymal Enhancement on Breast MRI: Impact on Diagnostic Performance

OBJECTIVE Breast density is documented to reduce sensitivity and specificity of mammography. However, little is known regarding the effect of normal background parenchymal enhancement on accuracy of breast MRI. The purpose of this study was to evaluate the effect of background parenchymal enhancement on MRI diagnostic performance. MATERIALS AND METHODS A review of our established MRI data identified all women undergoing breast MRI from March 1, 2006, through June 30, 2007. Prospectively reported background parenchymal enhancement categories of minimal, mild, moderate, or marked (anticipated BI-RADS MRI lexicon definitions) and assessments were extracted from the database for each patient. Outcomes were determined by pathologic analysis, imaging, and linkage with the regional tumor registry with a minimum of 24 months of follow-up. Patients were dichotomized into categories of minimal or mild versus moderate or marked background parenchymal enhancement. Associations with patient age, abnormal interpretation rate, positive biopsy rate, cancer yield, sensitivity, and specificity were compared using chi-square and z score tests. RESULTS The study cohort included 736 women. Moderate or marked background parenchymal enhancement was significantly more frequent among patients younger than 50 years compared with those 50 years old and older (39.7% vs 18.9%; p < 0.0001). Moderate or marked background parenchymal enhancement was also associated with a higher abnormal interpretation rate compared with minimal or mild background parenchymal enhancement (30.5% vs 23.3%; p = 0.046). Positive biopsy rate, cancer yield, sensitivity, and specificity were not significantly different according to background parenchymal enhancement category. CONCLUSION Increased background parenchymal enhancement on breast MRI is associated with younger patient age and higher abnormal interpretation rate. However, it is not related to significant differences in positive biopsy rate, cancer yield, sensitivity, or specificity of MRI.

BI-RADS MRI Enhancement Characteristics of Ductal Carcinoma In Situ

Abstract:  To identify the Breast Imaging Reporting and Data System magnetic resonance imaging (MRI) enhancement characteristics of ductal carcinoma in situ (DCIS). A retrospective review of consecutive patients who underwent breast MRI for newly diagnosed breast carcinoma prior to surgery was conducted. This yielded 381 lesions in 361 patients with pathologic confirmation of either DCIS alone, invasive carcinoma alone, or mixed invasive and in‐situ disease. Presence or absence of a MRI lesion at the site of the documented carcinoma was recorded, and for all identified MRI lesions the Breast Imaging Reporting and Data System morphology patterns were recorded. MRI features of the different malignancy types were compared utilizing Fisher’s exact tests; 64/381 (16.8%) lesions had DCIS, 101/381 (26.5%) had invasive carcinoma, and 216/381 (56.7%) had mixed invasive/in situ carcinoma. A MRI lesion corresponding to the known cancer was identified in 55/64 (85.9%) cases of DCIS, 98/101 (97.0%) cases of invasive carcinoma, and 212/216 (98.1%) cases of mixed invasive and in‐situ carcinoma. For pure DCIS lesions, 38/64 (59.4%) exhibited nonmass‐like enhancement (NMLE), 9/64 (14.1%) were masses, and 8/64 (12.5%) were a focus. For pure invasive carcinomas 79/101(78.2%) were masses, 16/101 (15.8%) were NMLE, and 3/101 (3.0%) were a focus. For mixed lesions 163/216 (75.5%) were masses, 44/216 (20.4%) demonstrated NMLE, and 5/216 (2.3%) were a focus. The most common NMLE patterns of pure DCIS were segmental distribution and clumped internal enhancement. Although there is overlap in the MRI morphology and enhancement pattern of in situ and invasive breast carcinoma, DCIS more frequently manifests as NMLE than does invasive carcinoma.

Risk of Upgrade of Atypical Ductal Hyperplasia after Stereotactic Breast Biopsy: Effects of Number of Foci and Complete Removal of Calcifications

PURPOSE To determine if patients with fewer than three foci of atypical ductal hyperplasia (ADH) who have all of their calcifications removed after stereotactic 9- or 11-gauge vacuum-assisted breast biopsy (VABB) have a rate of upgrade to malignancy that is sufficiently low to obviate surgical excision. MATERIALS AND METHODS An institutional review board-approved, HIPAA-compliant retrospective review of 991 cases of consecutive 9- or 11-gauge stereotactic VABB performed during a 65-month period revealed 147 cases of atypia. One pathologist performed a blinded review of the results of procedures performed to assess for calcifications and confirmed ADH in 101 cases with subsequent surgical excision. Each large duct or terminal duct-lobular unit containing ADH was considered a focus and counted. Postbiopsy mammograms were reviewed to determine whether all calcifications were removed. Upgrade to malignancy was determined from excisional biopsy pathology reports. Upgrade rates as a function of both number of foci and presence or absence of residual calcifications were calculated and compared by using chi(2) tests. RESULTS Upgrade to malignancy occurred in 20 (19.8%) of the 101 cases. The upgrade rate was significantly higher in cases of three or more foci of ADH (15 [28%] of 53 cases) than in cases of fewer than three foci (five [10%] of 48 cases) (P = .02). Upgrade rates were similar, regardless of whether all mammographic calcifications were removed (seven [17%] of 41 cases) or all were not removed (nine [20%] of 45 cases) (P = .77). Upgrade occurred in two (12%) of 17 cases in which there were fewer than three ADH foci and all calcifications were removed. CONCLUSION The upgrade rate is significantly higher when ADH involves at least three foci. Surgical excision is recommended even when ADH involves fewer than three foci and all mammographic calcifications have been removed, because the upgrade rate is 12%.