Papillary lesions are a heterogeneous group of breast lesions that include benign papillomas, atypical papillomas, and papillary carcinomas. Intraductal papilloma usually presents with nipple discharge or a palpable mass, and surgical management is the common procedure. However, an increasing number of papillomas are occasionally detected by image screening in the preclinical stage, which usually presents with no clinical symptoms. Some of these lesions are seen sonographically as solid nodules that are not distinguishable from other solid lesions in the breast. It has been suggested that these preclinical papillary lesions require additional surgical excision because they tend to cancerate [9]. In the present study, we found that 56.5% of patients with papilloma were asymptomatic, and 54.2% were categorized as BI-RADS 4a on ultrasonographic images. Only 32.3% of intraductal lesions were identified on an ultrasonographic check-up. However, patients with nipple discharge were more easily identified with ultrasonography (63.6%), possibly because the papilloma was visible in the dilated duct, which was full of fluids. The average papilloma size was 12 mm (5-30 mm) in our study, so it was difficult to differentiate small, impalpable papillary lesions. It was especially difficult to excise lesions with such a small size, but CNB guided by ultrasound was the applicable method to diagnose these lesions. However, because of insufficient tissue collected at biopsy, it was sometimes difficult to distinguish malignant from benign papillary lesions using CNB [10].
Some studies have recommended that papillary lesions diagnosed by CNB, especially those with atypia, should be subjected to open surgery for an accurate diagnosis because the rate of upgrading to malignancy is high in such situations [11 12]. Regarding benign IDP without atypia, previous studies have reported that the rates of upgrading vary widely following excision. Some studies have shown that IDPs are significantly associated with higher-grade lesions, and open excision is recommended in all cases [11 13]. However, other reports have suggested clinical and imaging follow-up rather than surgical excision because the rate of upgrading is low in IDP without atypia [14 15]. Therefore, the management of benign IDPs diagnosed by CNB remains controversial. Several studies have shown that the upgrading rate after open excision is associated with the adequacy of samples in biopsy lesions, even though the needle gauge used in CNB plays an important role in upgrading the rate of IDP [16 17]. In our study, we found an underestimation of 2/26 (7.7%) following open excision after VAE. IDP with AHP was diagnosed in one patient initially diagnosed with IDP on VAE, and the other was diagnosed with intraductal papillary carcinoma after an initial diagnosis of IDP with sclerosing adenosis on VAE. Our underestimation rate was lower than those of other authors. For example, Tatarian et al found that 21.3% of patients who were initially diagnosed with benign papilloma with CNB had IDP with atypia following surgical excision, and the majority of the atypical lesions were obtained from the tissue surrounding the papilloma [18]. The reason for the low rate of underestimation in our patients may be the sufficient samples (mean 8 tissue cores) obtained with VAE (8 gauge needle). Cassano et al [19] also believed that further verification with open excision was not necessary in patients diagnosed with benign lesions by VAE. They showed that patients diagnosed with IDP with VAE exhibited no recurrence or progression when followed up for several months via imaging check-ups [19]. Obviously, the several-month follow-up was too short in the above study; however, a follow-up was conducted for 38 months in our patients who underwent VAE, and suspicious alterations were not found on images, which means VAE is an applicable method of diagnosing and treating breast papillary lesions. However, of the two downregulated patients, one was diagnosed with intraductal papillary cancer after open surgery. This patient suffered simultaneously from contralateral invasive ductal cancer, and her initial diagnosis was IDP with sclerosing adenosis on VAE. Consequently, attention should be paid to such patients when VAE is applied.
In the present study, we found that the average lesion diameter in IDPs with AHP (21 mm) was greater than that in IDPs (13 mm), which means that AHP was associated with lesion size. It has been reported that the BI-RADS category is associated with upgrade rates in benign IDP diagnosed by CNB [20]. In our study, 10 patients with BI-RADS category 4b and 4c on US images underwent open surgery. The histopathological examinations were IDPs in 6 patients and IDP with APH in 4 patients. We could not find the relation between the upgrading rate and BI-RADS category, which might be because of fewer upgrading cases. However, we found that 54.2% of papillary lesions were categorized as BI-RADS 4a on images, and the C3, and C4b, C4c category only accounted for 11.7% (10 cases) of the total patients. Therefore, papillary lesions must be kept in mind when breast lesions are categorized as 4a on ultrasonographic images.
Several reports have revealed that in more than 95% of patients, ultrasound-guided VAE can entirely remove papillomas that measure less than 15 mm and therefore has therapeutic value to avoid open surgery [8 21]. In our patients, the largest diameter of the lesion was 30 mm, and we completely excised the lesions via VAE guided by ultrasonography. In the follow-up period, 8.5% (5/59) of patients presented with hypoechoic lesions in the primary site on ultrasonogram, 2 patients underwent VAE again, and the histopathological diagnoses were IDP and adenosis. The other 3 patients were followed up, and no suspicious alterations were found on images. Because only two patients underwent VAE again, the recurrence rate could not be determined. We might not have entirely removed all lesions with VAE in these patients because the lesion sizes were larger than the needle groove, and tiny residual lesions could not be detected with ultrasonography. There were 5 cases of multifocal lesions on the initial ultrasonogram in our study, and lesions less than 5 mm in diameter were not excised. After an average follow-up of 38 months, we found that 3 in 5 patients presented no lesions, and no enlargement process was found in the other 2 patients’ lesions. Donaldson et al [22] found that the 7-year cumulative breast cancer incidence rate was only 10% in patients who had an initial diagnosis of AHP on CNB, so they believed that close imaging follow-up was applicable to these patients. Furthermore, breast papillary carcinoma is rare, representing only 1–2% of all breast malignancies [23]. Therefore, we consider that it is not necessary to excise lesions less than 5 mm in diameter, especially in patients with multifocal papillomas.