Fine-needle aspiration is limited in the classification of benign breast diseases

Cytomorphological comparison of ThinPrep and SurePath liquid-based cytology in thyroid fine-needle aspiration

BACKGROUND

The two widely established systems for liquid-based cytology (LBC), ThinPrep and SurePath, employ different principles. The aim of this study was to compare the cytomorphology of thyroid lesions prepared by the two techniques.

METHODS

We retrospectively reviewed 44 thyroid FNA specimens prepared by LBC, including 20 ThinPrep and 22 SurePath. Cytologic diagnoses were made according to the Bethesda system and cytomorphologic parameters were evaluated.

RESULTS

Acellular smears were significantly frequent in ThinPrep than SurePath (10% vs. 0%). Both techniques produced a clean background, well cell preservation, and not apparent cell shrinkage. ThinPrep showed significantly lower cellularity than SurePath (25% vs. 4.3%). ThinPrep produced considerable flattening and fragmented clusters, while SurePath contained larger clusters in a three-dimensional configuration. Colloid was significantly reduced in amount and fragmented in ThinPrep, and was easily observed in SurePath. In cases of Hashimoto's thyroiditis, ThinPrep produced much less leukocytes in background than SurePath. Aggregates of fibrin and leukocytes were frequently present in 10/16 cases (62.5%) processed by ThinPrep. Air-dry artifact at periphery of the ring was present in 6/16 cases (37.5%) processed by ThinPrep. The nuclear features of papillary carcinoma were similarly evident in both LBC preparations.

CONCLUSION

SurePath seems to be superior to ThinPrep for diagnosing benign entities based on adequate representation of colloid and lymphocytes. The cell quality of both techniques in thyroid FNA was comparable, while each method introduces its own unique cytologic artifacts related to its methodology. We should recognize the cytomorphologic alterations to avoid misinterpretations.

FNA of breast fibroadenoma: observer variability and review of cytomorphology with cytohistological correlation

OBJECTIVE

To determine the observer variability in reporting fibroadenoma of the breast by fine needle aspiration (FNA) and to review the cytomorphological features of the lesion with cytohistological correlation.

METHODS

Retrospective analysis of FNA smears from 110 cases diagnosed as fibroadenoma of which surgical pathology follow-up was available in 33. Two pathologists were asked to categorize smears from 67 cases of breast lesions while blinded to the clinical finding as fibroadenoma, epithelial hyperplasia (usual and atypical) and malignant. All fibroadenoma (33) and cancer (15) cases were biopsy-proven. The same set of slides was re-circulated to one of the pathologists, and his first and second round results were compared.

RESULTS

Pre-review cytohistological correlation was attained in 32 of 33 cases of fibroadenoma (97%). The overall agreement between the two observers was 87% [Kappa = 0.74, 95% confidence interval (CI) 0.72-0.76]. Cytohistological correlation was achieved in 26 of 33 (79%) cases. Intra-observer agreement was 91% (Kappa = 0.82, 95% CI 0.89-0.93) with cytohistological correlation in 29 of 33 (87%) cases. Causes of diagnostic errors included marked dissociation, pleomorphism, poorly cellular smears from hyalinized fibrodenoma, lacational changes and apocrine metaplasia with cystic changes. Multinucleated giant cells were frequently encountered in FNA smears from fibroadenoma (31.8%), but in none of the lumpectomy specimens. Their histiocytic nature was suggested by immunohistochemistry.

CONCLUSION

FNA was a highly sensitive method for the diagnosis of fibroadenoma. Current cytological criteria were reliable and gave high inter- and intra-observer reproducibility.

Can hamartoma of the breast be distinguished from fibroadenoma using fine-needle aspiration cytology?

In an attempt to determine if it is possible to distinguish hamartoma of the breast from fibroadenoma using fine-needle aspiration cytology, we reviewed the cytological slides of 13 histopathologically confirmed cases of hamartoma of the breast and compared them with the cytological features of 13 histologically confirmed fibroadenomas. In each case, we studied the epithelial and stromal features. Cytologic characteristics were retrospectively evaluated in a semiquantitative manner. In conclusion, the finding of intact lobular units and a relative paucity of stroma may suggest the diagnosis of hamartoma.

Interobserver agreement of a probabilistic approach to reporting breast fine-needle aspirations on ThinPrep®

We have previously demonstrated the accuracy and reproducibility of a probabilistic/categorical approach for reporting breast fine-needle aspiration (FNA). However, the interobserver agreement in the application of this approach has not been assessed. Twenty breast FNA cases (each on one ThinPrep slide) were pulled from the cytology files of Beth Israel Deaconess Medical Center. The cases included benign epithelial proliferative lesions (6), DCIS (4), and infiltrating carcinoma (10), as shown by subsequent histology. Six pathologists with 14-25 yr of experience in interpreting breast FNA and 0-8 yr of experience with ThinPrep preparations rendered diagnoses according to the probabilistic approach. The kappa statistic for the unremarkable/proliferative, atypical, suspicious, and positive categories were 0.64, 0.08, 0.43, and 0.75, respectively (P < 0.001 for all except for the atypical category [P = 0.09]). Spearman's rho correlating the individual pathologist's diagnosis and the histologic diagnosis ranged from 0.51 (P = 0.02) to 0.78 (P < 0.0001). This was not correlated with the pathologists' years of experience interpreting breast FNA (P = 1.0) or with their years using ThinPrep preparations for breast FNA (P = 0.96). In conclusion, the interobserver agreement was excellent for the positive category in the probabilistic approach, poor for the atypical category, and fair to good for the other categories. The specific level of experience interpreting breast FNA or using ThinPrep among experienced pathologists did not seem to influence their accuracy in reporting the cases in our study.

Analysis of routine cytopathologic reports in 1,598 histologically verified benign breast lesions

This retrospective study was designed to evaluate the accuracy of cytopathologic diagnosis and of correct classification of benign breast diseases. A total of 1,598 FNABs were identified to have met the study criteria; of these, 1,258 (78.7%) cases were cytologically benign, 88 (5.5%) suspicious, 3 (0.18%) false-positive, and in 249 (15.6%) cases an inadequate sample was obtained. A specific diagnosis was made in 847/1,258 (67.3%) cases; the other 411 were diagnosed as benign NOS. Out of 847 specific FNABs diagnoses, 451 were fibroadenomas, 27 phyllodes tumors, 289 fibrocystic diseases, 4 proliferative fibrocystic diseases, 38 papillomas, 22 fat necrosis, 9 mastitis, 1 pseudolymphoma, 2 lipomas, 2 duct ecstasies, and 2 atheromas. In our study group the cytopathologic diagnosis of benign breast diseases excluding unsatisfactory aspirates was correct in 93%. Specific diagnosis was correct on average in 50% of cases, only in FA was its accuracy over 60%; in adequately sampled tumor, the predictive value of FA was 86.2%.

Breast hamartoma: fine-needle aspiration cytologic finding

BACKGROUND

Breast hamartoma is an unusual, well-circumscribed, tumor-like mass entering into the differential diagnosis of benign breast disease. To the authors' knowledge, the cytology of these lesions has not been well described. Although fine-needle aspiration is a well established procedure for the detection of breast carcinoma, its utility in classifying benign breast disease is less clear.

METHODS

Fine-needle aspirates from eight patients with histologically proven hamartomas were reviewed. None of the cases had a preoperative fine-needle aspiration diagnosis of hamartoma. Cytologic characteristics were retrospectively evaluated in a semiquantitative manner and compared with the histologic findings.

RESULTS

The aspirates were moderately cellular and contained sheets of both bland ductal cells and lobular units. Adipose tissue was present in varying amounts. Bipolar stromal nuclei were readily apparent, whereas intact stromal fragments were less prominent. Cytologic atypia was uniformly absent.

CONCLUSIONS

The cytology of breast hamartomas shows considerable overlap with other benign breast disease and is unlikely to be interpreted as malignant. The findings of intact lobular units and a relative paucity of stroma in an aspirate from a well circumscribed breast lesion may suggest the diagnosis of hamartoma.

Fibroadenomas with atypia: causes of under- and overdiagnosis by aspiration biopsy

Fibroadenoma (FA) is a common benign breast lesion frequently sampled by fine-needle aspiration biopsy (FNAB). Although the cytologic diagnosis is straightforward in most cases, cellular discohesion and atypia in FAs may lead to falsely atypical or positive FNAB diagnoses. Conversely, some adenocarcinomas mimic a fibroadenomatous pattern on FNAB, resulting in a false-negative diagnosis. We reviewed the cytologic and histologic findings in 25 cases with a preoperative FNAB diagnosis of FA, wherein excision was recommended based on atypia. Our aim was to analyze the spectrum of changes causing under- or overdiagnosis in such cases. The smears were assessed for cellularity, cellular discohesion, presence of dissociated intact cells and nucleoli, nuclear pleomorphism, oval bare nuclei, and stromal fragments. The histologic findings were correlated with FNAB features. At excision, 88% of FAs classified as atypical on FNAB were benign (FA with ductal hyperplasia and lactational change, myxoid FA, and other fibroepithelial lesions). Differentiating myxoid FA from colloid carcinoma was difficult due to the abundance of extracellular mucin in which the dissociated epithelial cells were floating. Two (8%) cases were carcinomas on excision; the reasons for underdiagnosis in one case reflected sampling, and in the other, interpretative error. There was one (4%) benign phyllodes tumor which lacked stromal fragments and single stromal cells on FNAB smears. The lesion was called atypical, based on the epithelial discohesion on the smears. We conclude that the majority of FAs with atypia on FNAB are benign lesions. Considering the grave consequences of a false-positive cytologic diagnosis, we recommend a conservative approach in interpreting FNAB smears which overall display a fibroadenomatous pattern.

Ultrasound-guided core needle biopsy as an initial diagnostic test for palpable breast masses

PURPOSE

To determine the role of ultrasound (US)-guided core needle biopsy as an initial diagnostic test for palpable breast masses.

MATERIALS AND METHODS

Ninety-eight consecutive patients, each with a palpable breast mass, were referred for US-guided core biopsy by a multidisciplinary team of physicians who specialize in the care of breast diseases. All palpable breast masses were clearly visible on high-resolution US. Ninety-nine core needle biopsies were performed under local anesthesia with use of freehand technique, mostly in an outpatient setting. Core needle path through each mass was documented in two orthogonal sections. A mean of 3.4 tissue core samples (range, 1-7) were obtained in each patient.

RESULTS

Core needle biopsy resulted in the diagnosis of 66 malignancies, two cases of atypical ductal hyperplasia (ADH), and 30 benign diseases of the breast. Surgery with curative intent was performed in 63 breast malignancies and excisional biopsies were performed for 10 benign diseases (two cases of ADH and eight benign lesions). Twenty-five breast masses were managed nonoperatively: chemotherapy was performed in three locally advanced breast cancers and 3-year follow-up was conducted for 22 benign lesions. Malignancies were correctly diagnosed in all cases. No malignancy was discovered at surgery or during clinical follow-up of ADH and no benign breast lesions were diagnosed by core needle biopsy. US-guided core needle biopsy is 100% sensitive and specific for palpable breast malignancies, with no false-positive results. A puncture site ecchymosis was the only morbidity or complication noted.

CONCLUSION

US-guided core needle biopsy is a safe and accurate first diagnostic test for palpable breast masses that require tissue proof.

Human breast lesions: characterization with contrast-enhanced in vivo proton MR spectroscopy--initial results

PURPOSE

To assess the clinical usefulness of localized proton (hydrogen 1) magnetic resonance (MR) spectroscopy in the characterization of contrast material-enhanced breast lesions on the basis of choline detection.

MATERIALS AND METHODS

Examinations were performed at 1.5 T with use of a standard breast coil. Contrast-enhanced MR imaging was performed in 30 consecutive patients (mean age, 50 years; age range, 20--80 years) who had nonspecific lesions (>1.5 cm in diameter) on sonograms or mammograms. Single-voxel (1)H MR spectroscopy was performed in the enhancing lesions by using a point-resolved spectroscopic sequence with echo times of 38, 135, and 270 msec. MR spectroscopic and histopathologic findings were determined in blinded fashion and compared.

RESULTS

Twenty-four patients had carcinoma of the breast (tumor size, 2.0--11.2 cm; mean, 4.7 cm), and six had benign lesions (lesion size, 1.8--3.8 cm; mean, 2.7 cm). Choline was detected in 22 patients with carcinoma. Choline was not detected in five patients with benign lesions and in two patients with carcinoma. The preliminary results indicate that this technique had a sensitivity of 92%, specificity of 83%, and accuracy of 90%.

CONCLUSION

Choline can be reliably detected in less than 45 minutes in large contrast-enhanced breast lesions by using a multiecho point-resolved spectroscopic protocol. The presence of water-soluble choline metabolites obtainable with (1)H MR spectroscopy could complement MR imaging findings to improve specificity and to reduce the number of unnecessary biopsies.