Breast cancer (BC) is the most prevalent cancer among women worldwide [1], and its incidence rates have been increasing by approximately 0.5% per year since the mid-2000s, attributed at least in part to continued declines in the fertility rate and increases in excess body weight [2]. The main clinical treatments for BC include surgery, radiotherapy, and chemotherapy, among which surgery is the most common for early-stage BC [3]. In a cohort of 1,216,820 women with early-stage BC, 35.5% of them were reported to undergo resection of the primary tumor [4]. However, postoperative recurrence and distant metastasis are still major challenges [5]. There are still some evolving risks for postoperative patients to develop recurrence and metastasis and may progress to advanced disease [6]. Locoregional recurrence especially ipsilateral breast tumor recurrence is associated with a significantly increased risk of distant disease and death [7,8]. Hence, for mastectomy patients with BC, early detection of locoregional recurrence at follow-up review and timely treatment play key roles in improving the survival rate.

In the current state of BC detection, MRI is the most sensitive imaging modality [9,10], which has led to a rapid increase in the use of postoperative follow-up examinations [11,12]. High-resolution contrast-enhanced T1-weighted imaging (CE-T1WI) is crucial for diagnosis [13,14], and has been widely used in breast MRI with high sensitivity for lesion detection. To accurately characterize and delineate lesion morphology, high-quality fat suppression is acquired in CE-T1WI by eliminating the signals of fat tissue [9,15].

Conventionally, frequency-selective fat suppression with spectral attenuated inversion recovery (SPAIR) is routinely applied in clinical practice, but it suffers from inhomogeneous fat suppression, especially on the postmastectomy breast side. This is related to (a) an uneven skin surface with contracted scars due to surgical excision [16], (b) field inhomogeneity in the broad contact area between the skin surface and surrounding air, and (c) SPAIR relying on frequency-selective excitation with lipid protons due to its intrinsic feature. Thus, poor fat suppression with SPAIR on the postmastectomy breast side makes precise evaluation difficult. Another fat-suppression method, Dixon, has been introduced since 1984 [17], and can achieve water-only and fat-only images by calculation from in- and out-of-phase images, in which the water-only image could serve the purpose of fat suppression. It is insensitive to magnetic field inhomogeneity [17,18] and has great advantages by providing uniform fat-suppression [19], attracting great interest in various clinical applications in heart, head and neck, spine, musculoskeletal system, whole-body MRI [[20], [21], [22], [23], [24]] and breast imaging [[25], [26], [27], [28]]. Those studies have supported the capability of Dixon for excellent uniform fat-suppression in phantoms and patients; however, the application of Dixon in breast CE-T1WI for mastectomy BC patients has not been previously explored.

Therefore, the purpose of the current study was to investigate the image quality and performance of 3D CE-T1WI with two different fat-suppression techniques, Dixon and SPAIR, on 3 T breast MRI for patients who had undergone mastectomy in clinical settings.