RESUMO
OBJECTIVES: The purpose of this study was to compare the diagnostic performance of the fat-to-lesion strain ratio (FLR) and gland-to-lesion strain ratio (GLR) for patients with indeterminate or suspicious breast lesions on ultrasound (US) imaging under a controlled precompression technique and to see whether the technique improves the reproducibility of FLR and GLR measurement. METHODS: Fifty-three lesions in 39 consecutive patients who had scheduled core biopsy or excision surgery based on US findings were examined by US elastography. Each lesion was acquired under controlled precompression by 2 sonographers independently. Both the FLR and GLR of the lesion were calculated. For diagnostic performance, the sensitivity, specificity, negative predictive value, positive predictive value, and accuracy were obtained. Interobserver reliability between different sonographers was assessed by the intraclass correlation coefficient (ICC). RESULTS: Forty lesions were benign, and 13 lesions were malignant. Both the FLR and GLR were significantly higher in malignant than benign lesions (P < .05). The FLR yielded higher accuracy and specificity compared to the GLR (accuracy, 79.2% versus 60.4%; and specificity, 87.5% versus 50.0%). With the controlled precompression applied at less than 25% during elastography, the interobserver agreement was excellent for FLR measurements (ICC, 0.853; 95% confidence interval, 0.738-0.920) and GLR measurements (ICC, 0.779; 95% confidence interval, 0.619-0.87). CONCLUSIONS: The FLR performed better than the GLR in the detection of breast malignancy; thus, fatty tissue was a better reference tissue for calculating the strain ratio on malignant breast tumor elastography. Keeping precompression to less than 25% will enable different operators to acquire similar elastograms with reproducible FLR and GLR readings.
Assuntos
Neoplasias da Mama/diagnóstico por imagem , Técnicas de Imagem por Elasticidade/métodos , Ultrassonografia Mamária/métodos , Tecido Adiposo/diagnóstico por imagem , Adulto , Idoso , Feminino , Humanos , Pessoa de Meia-Idade , Estudos ProspectivosRESUMO
OBJECTIVE: Complex cystic and solid breast mass (CCSBM) is a radiological diagnosis based on grayscale B-mode sonographic features. Because of potential for malignancy, biopsy is typically recommended. We examined the feasibility of contrast-enhanced US (CEUS) as a tool to identify benign CCSBMs. METHODS: This Institutional Review Board-approved prospective observational study performed targeted CEUS of 14 CCSBMs that were subsequently biopsied. CEUS images were independently reviewed by two readers blinded to other sonographic features, noting presence or absence of enhancement and time to perceived optimal enhancement. Interobserver agreement for presence or absence of enhancement was analyzed using Cohen's kappa coefficient. From retrospective review of initial diagnostic US examinations, descriptive CCSBM sizes, subtypes, and Doppler information were recorded. Histopathologies were categorized as benign, benign with upgrade potential (BWUP), and malignant. Measures of diagnostic accuracy and 95% CIs were calculated for CEUS enhancement. RESULTS: Of 14 CCSBMs, 12 were nonmalignant (9 benign, 3 BWUP) and 2 were malignant. There was perfect interobserver agreement (Cohen's kappa 1.00) between the 2 readers for CEUS enhancement. CEUS was 100% sensitive, 25% specific, with an area under the receiver operating characteristic curve (AUROC) of 0.625 (95% CI, 0.50-0.75) in differentiating nonmalignant from malignant lesions. It was 100% sensitive, 33.3% specific, with an AUROC of 0.667 (95% CI, 0.50-0.85) in differentiating benign from surgically significant (BWUP and malignant) CCSBMs. CONCLUSION: This small feasibility study highlighted the potential of CEUS as a safe noninvasive tool to identify the proportion of CCSBMs that are benign and can avoid tissue biopsy.