Diagnostic accuracy of breast MRI for patients with suspicious nipple discharge and negative mammography and ultrasound: a prospective study

The Use of MRI to Detect Malignancy in a Patient Presenting With Unilateral Bloody Nipple Discharge

Nipple discharge presents as either physiological, which is green, white, or yellow, or is considered pathological, which is typically unilateral, spontaneous, and bloody. Bloody nipple discharge (BND) can be due to underlying malignancy or premalignant lesions. Mammogram (MMG), ultrasound (US), MRI, and ductography are all used to evaluate BND, but different modalities offer greater value in the diagnostic process. Here, we present a case that demonstrates the ability of MRI to detect abnormalities not seen on MMG and US in the setting of BND due to underlying malignancy. The use of MRI earlier in the diagnostic process allows for the use of breast-conserving measures and decreases the possibility of metastasis. This would result in less of a need for more aggressive treatments.

What is the role of breast MRI in the management of women with pathologic nipple discharge and normal conventional imaging?

BACKGROUND/AIMS

To investigate the diagnostic performance of breast MRI in revealing mammographically and sonographically occult lesions requiring excision in patients with pathologic nipple discharge.

MATERIALS AND METHODS

In this retrospective study, 57 women with pathologic nipple discharge who had normal or inconclusive mammography and ultrasonography results and underwent breast MRI were determined. Patients who had histopathological diagnosis or ≥ 1-year imaging follow-up were included. MRIs were classified as positive and negative according to final BI-RADS assessment categories. Diagnostic performance of MRI, including sensitivity, specificity, negative predictive value, and positive predictive value, was calculated for detecting both malignancy and lesions requiring surgery.

RESULTS

Abnormal contrast enhancement on the pathologic nipple discharge side was detected in 29 MRIs (50.8%), categorized as BI-RADS 4. Abnormal findings were solid masses in 17 cases (58.6%) and non-mass enhancement in 12 cases (41.3%). Despite normal conventional imaging results, 4 malignant lesions and 16 lesions requiring surgery were detected with MRI. The sensitivity and specificity of MRI for detecting lesions requiring surgery were 100% and 68.2%, respectively. The positive predictive value (PPV) and negative predictive value were 55.1% and 100%, respectively.

CONCLUSION

In conclusion, this study confirmed that MRI is a reliable tool to detect lesions requiring surgery in patients with pathologic nipple discharge. MRI should be used in routine workup in patients with normal conventional imaging and imaging follow-up can be safely applied in patients with negative MRI.

Deep learning for real-time detection of breast cancer presenting pathological nipple discharge by ductoscopy

Objective

As a common breast cancer-related complaint, pathological nipple discharge (PND) detected by ductoscopy is often missed diagnosed. Deep learning techniques have enabled great advances in clinical imaging but are rarely applied in breast cancer with PND. This study aimed to design and validate an Intelligent Ductoscopy for Breast Cancer Diagnostic System (IDBCS) for breast cancer diagnosis by analyzing real-time imaging data acquired by ductoscopy.

Materials and methods

The present multicenter, case-control trial was carried out in 6 hospitals in China. Images for consecutive patients, aged ≥18 years, with no previous ductoscopy, were obtained from the involved hospitals. All individuals with PND confirmed from breast lesions by ductoscopy were eligible. Images from Beijing Chao-Yang Hospital were randomly assigned (8:2) to the training (IDBCS development) and internal validation (performance evaluation of the IDBCS) datasets. Diagnostic performance was further assessed with internal and prospective validation datasets from Beijing Chao-Yang Hospital; further external validation was carried out with datasets from 5 primary care hospitals. Diagnostic accuracies, sensitivities, specificities, and positive and negative predictive values for IDBCS and endoscopists (expert, competent, or trainee) in the detection of malignant lesions were obtained by the Clopper-Pearson method.

Results

Totally 11305 ductoscopy images in 1072 patients were utilized for developing and testing the IDBCS. Area under the curves (AUCs) in breast cancer detection were 0·975 (95%CI 0·899-0·998) and 0·954 (95%CI 0·925-0·975) in the internal validation and prospective datasets, respectively, and ranged between 0·922 (95%CI 0·866-0·960) and 0·965 (95%CI 0·892-0·994) in the 5 external validation datasets. The IDBCS had superior diagnostic accuracy compared with expert (0.912 [95%CI 0.839-0.959] vs 0.726 [0.672-0.775]; p<0.001), competent (0.699 [95%CI 0.645-0.750], p<0.001), and trainee (0.703 [95%CI 0.648-0.753], p<0.001) endoscopists.

Conclusions

IDBCS outperforms clinical oncologists, achieving high accuracy in diagnosing breast cancer with PND. The novel system could help endoscopists improve their diagnostic efficacy in breast cancer diagnosis.

Nipple Discharge Imaging Evaluation with Mammography, Ultrasound, Galactography, and MRI

RATIONALE AND OBJECTIVE

To determine the diagnostic yield of various imaging tests used to evaluate nipple discharge.

MATERIALS AND METHODS

A single institution, IRB-approved, retrospective study was performed of 320 consecutive patients presenting with nipple discharge. Imaging and pathology were reviewed to determine the yield for malignancy, atypical high-risk lesions (HRLs), and intraductal papillomas (IDPs).

RESULTS

Of the 320 patients, pathology or follow up confirmed 40 breast malignancies (40/320, 12.5%),14 atypical HRLs (14/320, 4.4%), 71 IDPs (71/320, 22.2%), 48 other benign pathologies (48/320,15.0%), and 147 unknown but benign cases (147/320, 45.9%). Physiologic discharge characteristics were observed in a minority of malignant cases: nonspontaneous (4/40, 10.0%); neither bloody nor clear (4/40, 10.0%); bilateral (3/40, 7.5%). Malignancy was associated with older age (p < 0.001) and bloody discharge (odds ratio 6.5, p < 0.0001). The combination of digital mammography and ultrasound had a 93% sensitivity and a 98% NPV, while contrast enhanced MRI (CE-MRI) had a 100% sensitivity and a 100% NPV for malignancy. Only three galactography examinations were performed among the malignant cohort, with minimal contribution (1 of 3) to the diagnostic evaluation. In this case, galactography findings helped determine imaging-pathology discordance, prompting a recommendation for surgical excision and subsequently a malignant diagnosis.

CONCLUSION

The combination of mammography and ultrasonography detected 93% of breast malignancies associated with nipple discharge and had a 98% NPV for malignancy. The value of CE-MRI is its ability to detect the remaining malignancies, not detected on mammography or ultrasound, and its ability to obviate the need for surgical duct excision.