Hamartomas of the breast: clinical, radiologic, and pathologic manifestations.

Between 1986 and 1996, 35 patients with a diagnosis of hamartoma of the breast were seen at the Mayo Clinic. One patient had two lesions. The mean age was 50 years (range 21-86 years). Hamartomas were clinically identified as a palpable lump in 18 cases (11 were detected by the patient and 7 by a physician). The other 18 were identified mammographically. Twenty-four lesions were in the left breast and 12 were in the right breast; 39% were located in the upper outer quadrant. Mammographically most hamartomas were ovoid, and the lesions were well circumscribed. Sonographically they were all solid, but 24% showed cystic areas. Pathologically the mean greatest diameter was 3.2 cm (range 1.0-7.5 cm). All but one lesion showed circumscription. The mean percentage of fibrous tissue was 78% (range 5-95%), fat 13% (range 0-95%), and epithelium 9% (range 1.0-60%). Calcifications were seen in four lesions. Ductal hyperplasia was present in 27% and adenosis in 70% of lesions. Twelve percent of patients had coexistent fibroadenomas.

Tamoxifen in the management of pseudoangiomatous stromal hyperplasia.

Pseudoangiomatous stromal hyperplasia (PASH) is a relatively uncommon histologic finding in breast specimens. The clinicopathologic spectrum of this disease entity can range from a focal nonsignificant microscopic finding to a dominant palpable breast mass. To confirm the diagnosis, a biopsy is required primarily to distinguish PASH from a low-grade angiosarcoma. The mammographic description of PASH is a round or ovoid, circumscribed or partially circumscribed mass. The sonographic feature is a hypoechoic mass. PASH is similar to a fibroadenoma in clinical and imaging features. Progressive breast enlargement associated with engorgement, cyclical breast pain, and burning sensation is of significant concern for some women. The management of the palpable mass and associated symptoms has included excisional biopsy, often leading to recurrent excisions and even mastectomy. This report documents an impressive response to tamoxifen in a patient with PASH presenting with breast enlargement, pain, and breast masses. To our knowledge, there are no reports on the use of tamoxifen or other selective estrogen receptor modulators in the management of this benign breast condition.

Sarcoidosis presenting as spiculated breast masses.

A 67-year-old woman sought medical treatment of cardiomyopathy, which had been diagnosed 2 years earlier; the causative factor was sarcoidosis. A screening mammogram revealed multiple spiculated masses in both breasts. A review of previous films obtained elsewhere showed that these masses had been increasing in prominence during the past 3 years. The patient had no visible axillary nodal abnormalities. Sarcoidosis was considered a diagnostic possibility, and a large-core needle biopsy was done with stereotactic guidance. The histological diagnosis was non-necrotizing granulomatous inflammation, consistent with sarcoidosis.

Ductal carcinoma in situ of the breast.

PURPOSE: The increasing incidence and biological heterogeneity of ductal carcinoma in situ (DCIS) of the breast have made the management of this entity challenging and controversial. This paper reviews data on the natural history of the disease and results obtained with various management approaches. DATA SOURCES: Computerized MEDLINE search of articles related to DCIS published since 1966. STUDY SELECTION: Randomized trials were given higher value; however, because these were relatively scarce, retrospective studies and data published in abstract form were also included. DATA EXTRACTION: The authors reviewed all sources critically. No formal statistical calculations were made. DATA SYNTHESIS: The incidence of DCIS is increasing, and a greater proportion of diagnoses are being made in asymptomatic patients. No data from randomized trials compare mastectomy and breast-conserving therapy for the treatment of DCIS. A large randomized trial comparing lumpectomy with lumpectomy plus radiotherapy showed lumpectomy plus radiotherapy to be effective for management of this disease. The presence of comedo necrosis and surgical margin status are frequently used as predictors of subsequent recurrence, although this practice is controversial. The risk for in-breast recurrence at 5 years after lumpectomy and radiotherapy is approximately 8%. With more refined molecular analysis, the relation of DCIS to invasive breast cancer will be better defined. CONCLUSIONS: Treatment strategies for DCIS have evolved, and lumpectomy followed by radiotherapy is an appropriate alternative for most patients. The use of lumpectomy alone in selected patients remains controversial.

Reduction of breast biopsies with a modified self-organizing map.

A modified self-organizing map with nonlinear weight adjustments has been applied to reduce the number of breast biopsies necessary for breast cancer diagnosis. Tissue features representing texture information from digital sonographic breast images were extracted from sonograms of benign and malignant breast tumors. The resulting hyperspace of data points was then used in a modified self-organizing map that objectively segments population distributions of lesions and accurately establishes benign and malignant regions. These methods were applied to a group of 102 problematic breast cases with sonographic images, including 34 with malignant lesions. All lesions were substantiated by excisional biopsy. The system can isolate clusters of purely benign lesions from other clusters containing both benign and malignant lesions. The hybrid neural network defined a region in which about 60% of the benign lesions were located exclusive of any malignant lesions. The experimental results also suggest that the modified self-organizing map provides more accurate population distribution maps than conventional Kohonen maps.

Accuracy of frozen-section diagnosis of mammographically directed breast biopsies. Results of 1,490 consecutive cases.

The accuracy of intraoperative frozen-section diagnosis of small in situ and invasive breast cancers is uncertain. We reviewed the results of 1,490 consecutive wire-localized breast biopsies from 1,439 patients with nonpalpable mammographically detected abnormalities examined by frozen section at Mayo Clinic over a 3-year period. The mammographic abnormalities included benign calcifications (61 cases), indeterminate calcifications (422 cases), worrisome calcifications (54 cases), well-circumscribed nodules (115 cases), irregular nodules (473 cases), architectural distortions (52 cases), asymmetry (39 cases), well-circumscribed nodules with calcification (12 cases), irregular nodules with calcification (75 cases), architectural distortions with calcification (35 cases), and asymmetry with calcifications (12 cases). We detected 457 carcinomas, including 135 in situ and 322 invasive cancers. The invasive carcinomas had a mean diameter of 1.07 cm (range, 0.1-3.0 cm), including 191 with diameters of 1 cm or less (59.3% of cases). In 77 cases (5.2% of total), the diagnosis was deferred to permanent sections. Frozen-section accuracy was 97.7%. False-negative diagnoses were rendered in 0.5% of cases, and there were no false-positive diagnoses. There was no correlation of infiltrating carcinoma diameter and error rate. These results indicate that intraoperative frozen section of mammographically directed breast biopsies provides accurate and reliable diagnoses.

Breast biopsy: a comparative study of stereotaxically guided core and excisional techniques.

OBJECTIVE: The major objectives of this prospective study were to compare pathologic findings from stereotaxic core and excisional biopsies performed on patients with impalpable breast lesions and to compare the initial mammographic impression with the final histologic diagnosis. SUBJECTS AND METHODS: All patients referred for preoperative localization of impalpable breast lesions between October 29, 1991, and January 15, 1993, were eligible for the study. If the patient and the lesion, on the basis of mammography, were considered suitable for core biopsy, the patient was asked to participate. Four hundred forty-five excisional biopsies were performed. One hundred sixty lesions were evaluated by core biopsy; for 104 of these lesions, five or more core samples were removed. Core biopsies were done with 14-gauge biopsy needles and were followed by a localization procedure. The pathologic features of core and excisional specimens were compared. RESULTS: Biopsy results were compared for 93 (58%) benign and 67 (42%) malignant lesions. Of 104 lesions evaluated with at least five core specimens, 56 (54%) were benign and 48 (46%) were malignant. Results of core biopsy corresponded to those of excisional biopsy for 96% of benign lesions, 83% of malignant lesions, and 90% overall; sensitivity of core biopsy for malignant lesions was 85%. Of 56 lesions for which fewer than five core specimens were obtained, 37 (66%) were benign and 19 (34%) were malignant. Results of core biopsy corresponded to those of excisional biopsy for 81% of benign lesions, 79% of malignant lesions, and 80% overall; sensitivity of core biopsy for malignant lesions was 84%. Specificity of core biopsy for the entire series of benign lesions was 100%. CONCLUSION: For mammographic lesions that are believed to be not malignant or not very likely malignant, stereotaxic core biopsy decreases the need for excisional biopsy.

Invasive lobular carcinoma of the breast: mammographic findings and extent of disease at diagnosis in 184 patients.

OBJECTIVE: Invasive lobular carcinoma of the breast is a distinct malignant neoplasm consisting of small cells that tend to infiltrate in thin strands that are often only one cell in width. The objectives of this study were to determine if these tumors are more frequently overlooked on mammograms than are other types of invasive carcinoma and to evaluate the specific mammographic and pathologic findings of this disease at the time of diagnosis. MATERIALS AND METHODS: From 1983 through 1991, invasive, purely lobular carcinoma of the breast was diagnosed in 184 women who had screen-film mammography and biopsy at our institution. We examined the initial mammographic reports to determine the principal abnormal mammographic findings that supported or prompted biopsy and to distinguish abnormalities that were overlooked in breasts with false-negative mammographic findings. All mammograms were retrospectively examined for benign or suspicious microcalcification in the vicinity of the neoplasm and for breast parenchymal patterns. Surgical pathology reports were also reviewed. RESULTS: The rate of false-negative findings on initial interpretations of mammograms was 19%. Forty-six percent of the mammograms with false-negative initial interpretation showed no evidence of malignant tumor in retrospect. Ten percent of the neoplasms showed suspicious calcification on mammograms, and 1% of patients had biopsy primarily because of worrisome calcification. Invasive lobular carcinoma tends to produce masses that are of relatively low radiographic opacity, similar to normal fibroglandular breast tissue. Forty-four percent of patients had metastases to axillary lymph nodes at the time of diagnosis. One patient had bilateral invasive lobular carcinoma. CONCLUSION: Our results show that the false-negative rate for the diagnosis of invasive lobular carcinoma is higher than that for other invasive cancers. The reasons for this difference appear to be the low rate of suspicious calcification found with invasive lobular carcinoma and its tendency to be of low opacity. We also found a lower rate of bilateral invasive lobular carcinoma than has been generally reported. Metastases to lymph nodes were common at the time of diagnosis.

Likelihood of malignant disease for various categories of mammographically detected, nonpalpable breast lesions.

To determine the likelihood of malignant disease for mammographically detected, nonpalpable breast lesions, we defined 11 morphologic categories and retrospectively reviewed the mammograms in 859 cases in which biopsy was performed after a wire localization procedure at our institution during 1989 and 1990. Within each category, the total number of lesions and the percentage of cases with a surgical pathologic diagnosis of malignant involvement were as follows: benign calcification, 25 (0% malignant); indeterminate calcification, 200 (22%); malignant calcification, 39 (92%); smooth mass, 84 (1%); irregular mass, 337 (40%); architectural distortion, 45 (47%); asymmetric breast tissue, 37 (3%, or 1 case of asymmetrically prominent ducts); smooth mass with calcification, 3 (0%); irregular mass with calcification, 68 (66%); architectural distortion with calcification, 14 (57%); and asymmetric breast tissue with calcification, 7 (29%). The overall rate of malignant involvement for the 859 cases was 34%. If follow-up examinations rather than biopsies had been done for the lesions categorized as benign calcification, smooth mass, smooth mass with calcification, and asymmetric breast tissue (excluding asymmetrically prominent ducts), the overall positive predictive value would have increased from 34 to 41%, and 148 biopsies would have been deferred (17% of all biopsies). If morphologic criteria are applied to the evaluation of mammographically detected, nonpalpable lesions, the rate of malignant disease at biopsy may reach 40%. This rate correlates with that in recent large series.

Improvement in specificity of ultrasonography for diagnosis of breast tumors by means of artificial intelligence.

A set of ultrasonograms of lesions from 200 patients between the ages of 14 and 93 years who underwent mammography followed by ultrasonographic examination and excisional biopsy has been studied with computer vision techniques to improve the ultrasonographic specificity of the diagnosis. Selected features representing the texture of the lesion were calculated and then classified by an artificial neural network. This network was biased toward correctly classifying all the malignant cases at the expense of some misclassification of the benign cases. The network diagnosed the malignant cases with 100% sensitivity and 40% specificity (compared with 0% specificity for the radiologists diagnosing the same set of cases in the breast imaging setting), and tests performed with a leave-one-out technique indicate that the network will generalize well to new cases. This suggests that methods based on neural network classification of texture features show promise for potentially decreasing the number of unnecessary biopsies by a significant amount in patients with sonographically identifiable lesions.

A comparison of mammography screen-film combinations.

A comparison study was performed to evaluate the image quality and radiation dose of six mammographic screen-film combinations: a Min-R screen with OM-1, SO-155, and SO-177 films (Eastman Kodak); a Min-R medium screen (Eastman Kodak) with OM-1 film; an HR Mammo medium screen (Fuji Medical Systems USA) with OM-1 film; and a Min-R fast screen with T-Mat M II film (Eastman Kodak). SO-177 films were processed with an extended cycle. Exposures of an acrylic test object with embedded masses, fibers, and specks and of a preserved breast specimen were made, for two paired image comparison tests in which the visibility of diagnostic features, contrast, and noise were judged. In most areas of image quality evaluated, a Min-R screen with OM-1, SO-155, and SO-177 films was superior. These three screen-film combinations had similar imaging characteristics, even though OM-1 film requires a higher radiation exposure. Images produced with a Min-R fast screen and T-Mat M II film were significantly lower in quality.

Imaging of the breast: techniques and results.

Procedures for imaging of the breast have increased in number substantially in the past 30 years. Mammography, currently the most important breast imaging technique, was introduced at the Mayo Clinic in 1961. In 1988, more than 36,000 mammographic procedures were done. Breast ultrasonography is also a valuable examination, particularly for determining whether mass lesions are cystic or solid. The use of mammography for screening is increasing. Screening mammography allows the detection of breast cancer before it is palpable and while it is still in a stage known as minimal breast cancer. Lesions metastatic to axillary lymph nodes are less common in women with nonpalpable breast cancers than in women with palpable lesions. Nondiagnostic mammography should not delay the biopsy of a mass that is suspicious on physical examination. A special breast imaging center for performance of screening studies and evaluation of breast problems has been established at the Mayo Clinic.

Comparison of mammography and transillumination light scanning in the detection of breast lesions.

A prospective study was done involving 822 women who had both film screen mammography and a breast transillumination light scan examination. The study population was not randomized; a modified phase-2 study was done. Mammography was superior for detecting malignancy: of the 67 pathologically proved breast cancers, 64 (95.5%) were detected by mammography and 45 (67.2%) were detected by transillumination. Biopsy revealed that 74 patients had benign lesions. In this group of patients, the false-positive rate was 12.2% for light scanning and 79.7% for mammography; however, 117 other patients had false-positive light scans.

Prebiopsy localization of nonpalpable breast lesions.

Invasive localization procedures are indicated preoperatively when breast lesions are nonpalpable. Several techniques are available for locating such lesions. In experience with 343 nonpalpable lesions, 27% of the biopsied lesions were malignant. Calcification was present in 51% of all lesions. A combination needle-hooked wire technique was used. Some problems were encountered, the most frequent being vasovagal reactions. The most serious problem was the failure to remove the located lesion in nine cases.

Ultrasound transmission computed tomography of the breast.

A preliminary clinical study of ultrasound transmission computed tomography of the breast (UTCTB) was undertaken to evaluate its capacity in the detection of breast abnormalities and to establish criteria for distinguishing benign from malignant lesions. Only patients with palpable and/or mammographically evident lesions were selected for study; complete analysis was accomplished in 78 cases. Visual and computer interpretations of reconstructed UTCTB scans were based on changes in speed of sound wave transmission and attenuation between the suspicious area and the surrounding tissue. In the computer-aided classification, discriminant functions were derived to predict the presence or absence of carcinoma. Visual analysis was subject to a low sensitivity. The presence of high speed transmission within a lesion usually indicated malignancy, although the converse was not true. Computer-aided preliminary screening of UTCTB scans by a trained technician may have the potential of contributing to interpretation accuracy; however, this finding must be highly qualified, given the methodologic constraints of the study.

Computed transmission ultrasound tomography.

Beginning with the wave equation, we have derived the classic reconstruction equations, which assume the ultrasonic energy travels in a straight line. The straight line reconstruction methods result in images that are not absolutely quantitative, although they may be useful in delineating speed and attenuation within two-dimensional cross sections, especially in organs such as the breast. Aberrations associated with straight-line reconstruction images are results of the effects of refraction and of diffraction. In addition, these methods assume that the acoustic wave travels within a plane and not in three dimensions; thus the assumed dimensionality of the problem also gives aberrations in the final image. The effects of diffraction are very complex and, given the current methods of measuring arrival time and amplitude, cause aberrations in the image, which result in errors both in geometry and in magnitude of the reconstructed values. Correction of diffraction effects with techniques termed 'diffraction tomography' are being investigated and have resulted in some preliminary data.

Upper gastrointestinal tract: villous tumors.

Villous tumors of the upper gastrointestinal tract are unusual. A review of the surgical experience at the Mayo Clinic disclosed 18 such tumors in 16 patients. Nine were in the stomach; two of these arose near the esophagogastric junction and appeared to be esophageal masses. Nine villous tumors arose in the duodenum and six of these originated near the papilla of Vater and produced obstructive jaundice. The total group of patients consisted of nine men and seven women aged 35-78 years. All tumors were malignant, with changes ranging from carcinoma in situ to invasive grade 3 adenocarcinoma. The malignancy rate in previous reports has been only 55%. The radiographic characteristics of villous tumors appearing as lesions within the esophagus and within the common bile duct have not been previously described in the literature. The basic features of villous tumors of the upper gastrointestinal tract are the same as those of their more common counterparts in the colon.

Computed tomographic mammography (CTM).

Breasts of 724 patients were studied by physical examination, mammography, and computed tomographic mammography (CTM) using a scanner designed for evaluation of the breast. Among cases in which CTM was not accompanied by use of contrast material, there were 60 malignant lesions, of which 10% were missed by mammography, 32% by CTM, and 8% by both. Among cases where CTM was supplemented by a 50 ml injection of 75% contrast material, there were 63 malignant lesions, of which 14% were missed by mammography, 16% by CTM, and 3% by both. Among cases where CTM was supplemented by a 300 ml infusion of 30% contrast material, there were 41 malignant lesions, of which 7% were missed by mammography and 5% by CTM, but none by both. Clinically, 22% of the malignant lesions in the infusion series were occult. There were 44 benign lesions in patients studied with the infusion technique. With mammography 68% were suspicious for malignancy, and with CTM, 56%.