Characteristics and outcomes of anterior mediastinal cystic lesions diagnosed on chest MRI: implications for management of cystic lesions – Insights into Imaging

This retrospective study was approved by the institutional review boards, and the requirement for informed consent was waived (IRB Numbers: 2021-0402, 2021-02-156).

Study design and patients

In this multicenter study, consecutive patients who underwent contrast-enhanced chest MRI for the evaluation of anterior mediastinal lesion manifesting as a cystic lesion were included from January 2012 to June 2019 at Asan Medical Center (institution A) and from January 2010 to June 2019 at Samsung Medical Center (institution B). All MRI examinations were performed to evaluate incidental anterior mediastinal lesions detected on CT. Exclusion criteria were overt solid tumor on MRI that showed solid enhancement of a nodule or mass, thymic hyperplasia, and poor image quality or at least one sequence unavailable among T2-weighted, T1-weighted (pre- or postcontrast), or diffusion-weighted images (DWIs) (Fig. 1) [10, 13].

Flowchart of study inclusion

The medical record of each patient was reviewed, and the clinical information (patient age, sex, reason for imaging follow-up, follow-up numbers, and intervals after chest MRI) and pathologic findings if the cystic lesion was resected were recorded.

Imaging protocols and scanners

Imaging studies including baseline and follow-up chest CT scans and MRI were performed with multidetector CT and MRI from different vendors and two magnet strengths (1.5 and 3.0 T). MRIs from institution A (n = 170) were obtained with either 3.0 T scanner (MAGNETOM Skyra, Siemens Healthineers, Erlangen, Germany; n = 137) or 1.5 T scanner (MAGNETOM Avanto, Siemens Healthineers, Erlangen, Germany; n = 33). MRIs from institution B (n = 33) were obtained with 3.0 T scanners from different vendors, either Philips (Ingenia, Philips Healthcare, Best, the Netherlands; n = 13) or Siemens (MAGNETOM Skyra, Siemens Healthineers, Erlangen, Germany; n = 20).

The MRI protocol included at least the following sequences: (1) axial T2-weighted images with fat-suppression, (2) pre- and post-contrast axial T1-weighted images with fat-suppression, and (3) DWIs with corresponding apparent diffusion coefficient (ADC) maps. T2-weighted imaging was performed mostly with breath-hold electrocardiography-gated double inversion recovery T2-weighted imaging. Electrocardiography-gating was applied to obtain better and reliable image quality with high-signal and high-resolution images of the lesion of interest. Images were mostly obtained before and after dynamic administration of gadolinium, predominantly with breath-hold three-dimensional ultrafast gradient-echo fat-saturated T1-weighted imaging (93.6%, 191/204) and occasionally with fat-saturated turbo spin-echo imaging (6.4%, 13/204). Breath-hold DWIs were obtained with b values of 0, b values of 100 or 500 s/mm², and b values of 700 or 1000 s/mm². The details of MRI protocols are demonstrated in the Additional file 1: Tables S1 and S2.

Image analysis

All imaging studies were reviewed by two thoracic radiologists (J.C. and S.M.L., with 5 years’ and 10 years’ experience in thoracic radiology, respectively) who were blinded to clinical information. Diametric measurements were performed manually on multiplanar images (axial, sagittal, or coronal) with the largest diameter using electronic calipers. The single largest diameter at each examination was recorded, and a measurable change in the size of anterior mediastinal lesion was defined as a decrease or increase in more than 2.5 mm in the largest size between the two examinations [14, 15].

Imaging features of anterior mediastinal lesions were assessed on MRI, including internal signal intensity on T2-weighted image and pre-contrast T1-weighted image relative to muscle, the presence of diffusion restriction, and the presence of contrast-enhancement of the lesion (smooth thin wall enhancement, nodular enhancing solid portion, eccentric thickening of septa or wall with enhancement). Signal intensities of lesions were measured on representative images with regions of interest covering the entire lesion. Fat-saturated axial pre-contrast T1-weighted images and breath-hold electrocardiography-gated double inversion recovery T2-weighted images were preferentially used for evaluation of T1 and T2 signal intensities, respectively. Based on MRI findings, all lesions were classified into two categories: category 1, probable cysts (or simple cyst) vs. category 2, indeterminate lesion (complex cyst) (Fig. 2). The MRI criteria for probable cyst (or simple cyst) included the following features: 1) no eccentric wall thickening or nodular enhancement except smooth thin wall enhancement, 2) no diffusion restriction, and 3) high signal intensity on T2-weighted image. An indeterminate lesion (complex cyst) was defined when the lesion did not satisfy the above findings. Confidence level of the diagnosis was also rated for probable cysts by 3-point scale (high, intermediate, and low).

MR findings of probable cyst (simple cyst, category 1) and indeterminate lesions (complex cyst, category 2). Chest MRI in a 50-year-old female (a–c) shows 22-mm well-defined, ovoid, and unilocular lesion (arrows) in the anterior mediastinum showing homogeneous high signal intensity on T2-weighted image (a) and no remarkable diffusion restriction on high b-value (b1000) diffusion-weighted image and high apparent diffusion coefficient (ADC) value (b). The lesion shows intermediate signal intensity on pre-contrast T1-weighted image, and no remarkable enhancement on post-contrast T1-weighted image (c). The lesion can be classified as a probable cyst based on chest MRI findings. Chest MRI in a 65-year-old female (d–f) shows a 43-mm lobulated and multilocular lesion in the anterior mediastinum showing internal high signal intensity on T2-weighted image (arrow, d) and no remarkable diffusion restriction (e). The lesion shows intermediate signal intensity on pre-contrast T1-weighted image and diffuse enhancement along cyst wall, and septa with a nodular solid enhancement attached to the right lateral side of the cyst wall on post-contrast T1-weighted image (arrowhead, f). The lesion can be classified as an indeterminate lesion based on chest MRI findings. The patient underwent thymectomy one month after chest MRI. On gross pathology, the cystic mass was well-defined and multiseptated with a focal solid portion and sebaceous material. The cystic mass was compatible with the thymic cyst on histopathology

For the evaluation of follow-up outcomes, the change in size of lesion and development of new nodular solid portion or irregular wall thickening on both CT and MRI were evaluated and events of surgery and incidence of confirmed tumor were recorded.

Statistical analysis

Descriptive statistics were demonstrated as mean ± standard deviation, median with interquartile range, or numbers with percentage. Variables were compared using the Fisher’s exact test, Chi-square test or Mann–Whitney U test, as appropriate. Follow-up strategies among physicians and between institutions were compared using the Kruskal–Wallis test or Mann–Whitney U test. All statistical analyses were performed using SPSS software (version 19.0, IBM Corp., Armonk, NY, USA). P values < 0.05 were considered to indicate statistical significance.