Elsevier

Clinical Imaging

Volume 39, Issue 5, September–October 2015, Pages 791-793
Clinical Imaging

Original Article
Internal mammary lymph nodes as incidental findings at screening breast MRI☆,☆☆,★

https://doi.org/10.1016/j.clinimag.2015.05.011Get rights and content

Abstract

Objective

To evaluate the prevalence of internal mammary lymph nodes (IMLNs) on breast magnetic resonance imaging (MRI) in a screening population.

Materials and methods

We retrospectively reviewed 92 consecutive screening breast MRI exams. Logistic regression was performed to ascertain the risk of IMLNs in cancer-free subjects and to determine whether the risk varies with age.

Results

IMLNs were present in 48.9% of patients. Mean node size was 4 mm (range, 3–10 mm). The prevalence of IMLNs was not related to age. No patients developed breast cancer after a mean follow-up of 3 years.

Conclusion

Subcentimeter IMLNs are common incidental findings at screening breast MRI.

Introduction

Metastasis to the internal mammary lymph nodes (IMLNs) in breast cancer patients has been demonstrated to be a marker of worse prognosis [1], [2], [3]. Identification of IMLN involvement may change clinical management, prompting parasternal radiation and, in some cases, adjuvant chemotherapy. While such therapies have potential benefit for patients with IMLN spread, there may also be significant treatment-related morbidity [4]. As such, care should be taken to avoid overtreatment of patients without verified IMLN metastasis.

Although breast magnetic resonance imaging (MRI) is commonly performed for local staging of breast cancer and depicts the IMLN chains, there are limited data regarding the significance of IMLNs identified at MRI in breast cancer patients. In a series of 16 breast cancer patients, Kinoshita et al. found that using a size threshold of 5 mm, IMLN metastases could be identified with 93.3% sensitivity and 89.3% specificity [5]. However, subcentimeter IMLNs may also be seen in healthy women undergoing screening breast MRI. If small IMLNs are highly prevalent in a screening population, this would suggest that their presence at MRI in breast cancer patients does not necessarily imply disease involvement. Validation of this hypothesis would help to prevent overtreatment of patients without IMLN metastasis and highlight the need for confirmatory tests in cases of suspected IMLN metastasis at breast MRI. Therefore, the goal of this study was to evaluate the prevalence and size distribution of IMLNs at breast MRI in a normal screening population.

Section snippets

Patients

Our study was approved by our Institutional Review Board and was compliant with requirements of the Health Insurance Portability and Accountability Act. Informed consent was not required. We performed a computerized search of our radiology information system (IDXrad, software version 9.7.1; IDX Systems Corporation, Burlington, VT, USA) to identify all screening breast MRI exams obtained at our institution from January 17, 2001, to January 24, 2011. Patients with a prior history of breast

Results

Internal mammary nodes were identified in 45 of 92 [49%; 95% binomial confidence interval (CI): 39, 60] patients without breast cancer at screening MRI (Fig. 1). Thirty-eight percent of patients (95% CI: 28, 49) had left-sided IMLNs, and 22.8% (95% CI: 15, 32) had right-sided nodes. The mean node size (long axis) was 4 mm (range, 3–10 mm) (Fig. 2). Left-sided IMLNs were significantly larger than right-sided nodes with estimated difference in severity rating of − 0.3 (95% CI: − 0.5, − 0.1; P= 

Discussion

Metastasis to the IMLNs in breast cancer patients is associated with increased rates of distant metastasis and lower rates of overall survival [1], [2], [3]. The identification of IMLN metastasis is usually an indication for parasternal radiation therapy. Furthermore, 6%–16% of patients with IMLN metastases do not have axillary node metastases [6], [7], [8], [9], [10], [11]. In this subset of patients, the identification of IMLN metastasis might prompt the initiation of adjuvant chemotherapy.

References (20)

There are more references available in the full text version of this article.

Cited by (9)

View all citing articles on Scopus
☆

Our study was approved by our Institutional Review Board and was compliant with requirements of the Health Insurance Portability and Accountability Act.

☆☆

Grants: Statistical analysis supported by CTSA grant # UL1 RR024131.

★

A.A. supported by NIBIB T32 Training Grant 1 T32 EB001631.

1

Tel.: + 1 707 651 1315.

2

Tel.: + 1 415 833 3700.

3

Tel.: + 1 415 514 8028; fax: + 1 415 514 8150.

4

Tel.: + 1 415 885 7464; fax: + 1 415 885 7876.

View full text
Published by Elsevier Inc.