Patients

The present study was approved by the human research committee of the institutional review board of our hospital and complied with the guidelines of the Health Insurance Portability and Accountability Act of 1996 and the Declaration of Helsinki. Due to the retrospective nature of the study, the requirement for informed consent was waived. We identified patients with histopathologically confirmed myoepithelioma of the head and neck and pleomorphic adenoma of the parotid gland from March 2009 to December 2024 using our hospital’s electronic medical records. Inclusion criteria were as follows: (a) preoperative MRI, (b) diagnosis based on surgical resection, and (c) primary tumor. Exclusion criteria included: (a) diagnosis based on biopsy and (b) recurrent tumors. A total of 11 patients with head and neck myoepithelioma and 103 patients with parotid pleomorphic adenoma who underwent preoperative MRI were included. Among the 11 patients, tumor locations in myoepithelioma included the parotid gland (n = 6), palate (n = 3), submandibular gland (n = 1), and oropharynx (n = 1).

MRI imaging

MRI images were obtained from all 114 patients using a 1.5-T MRI system (Intera Achieva 1.5 T Pulsar or Ingenia Prodiva 1.5 T CS, Philips Healthcare, Best, The Netherlands) or a 3.0 Tesla MRI scanner (Intera Achieva 3.0 T Quasar Dual, Philips Healthcare, Best, The Netherlands). All images were obtained at a section thickness of 3–4 mm with an intersection gap of 1 mm. Axial non-fat-suppressed T2-weighted fast spin-echo (TR/TE, 3000–5710/90 ms), axial and coronal non-fat-suppressed T1-weighted spin-echo (TR/TE, 688–778/15–18 ms), and coronal fat-suppressed T2-weighted fast spin-echo (TR/TE, 3421–5329/60–80 ms) images were obtained from all patients. Axial short-tau inversion recovery single-shot spin-echo echo-planar diffusion-weighted images (TR/TE, 4000–6710/65–68 ms) with b value of 0 and 1000 s/mm were obtained from 103 patients (7 myoepitheliomas and 96 pleomorphic adenomas). Axial and coronal fat-suppressed contrast-enhanced T1-weighted spin-echo (TR/TE, 576–764/14–18 ms) images were obtained from 41 patients (5 myoepitheliomas and 36 pleomorphic adenomas) after the intravenous injection of 0.1 mmol/kg of gadopentetate dimeglumine (Magnevist; Bayer HealthCare, Leverkusen, Germany) or gadobutrol (Gadavist; Bayer HealthCare). Axial fat-suppressed dynamic contrast-enhanced T1-weighted gradient-echo (TR/TE, 3.7–5.9/1.8–3.9 ms) images were obtained from 21 patients (3 myoepitheliomas and 18 pleomorphic adenomas) before and 30, 60, 90, 120, 150, and 180 s after intravenous administration of contrast material at a rate of 2 ml/s. When the contrast enhancement was heterogeneous, the signal intensities (SIs) of solid components were measured, and time-intensity curves were constructed.

CT imaging

CT imaging was performed for 36 patients (7 myoepitheliomas and 29 pleomorphic adenomas) using an 8-slice CT scanner (LightSpeed Ultra; GE Healthcare, Milwaukee, WI, USA), or 16-slice CT scanner (LightSpeed 16; GE Healthcare, Milwaukee, WI, USA), or 64-slice CT scanner (Brilliance 64; Philips Healthcare, Best, The Netherlands or Discovery CT750 HD; GE Healthcare, Milwaukee, WI, USA), or 256-slice CT scanner (Revolution CT with Apex Edition; GE Healthcare, Milwaukee, WI, USA). All transverse CT images were reconstructed with 2.5-mm section thickness and no overlap. Unenhanced CT images were obtained from all patients. Contrast-enhanced CT images were obtained for 13 patients (3 myoepitheliomas and 10 pleomorphic adenomas). Single-phase contrast-enhanced CT images were obtained 45 s after initiating an intravenous bolus injection of 100 mL of nonionic iodine contrast material containing 240 or 300 mg iodine/mL at an injection rate of 2 mL/s.

Imaging assessment

Two radiologists (Radiologists 1 and 2) with 26 and 4 years of post-training experience in head and neck imaging, respectively, individually reviewed MRI and CT images while blinded to clinical and pathological data. The reviewers were unaware of any clinical information or pathological diagnosis. Disagreements between the two reviewers were resolved by consensus.

Qualitative MRI findings included anatomical location of the parotid gland (superficial/deep lobe and superior/inferior pole), lobulated margin, multinodular configuration, capsule formation (entire/partial/absence), intratumoral septa, heterogeneity on T1-, T2-, and fat-suppressed contrast-enhanced T1-weighted images, hyperintense areas on T1-weighted images, hypointense areas on T2-weighted images, predominant SI of solid components on T1-, T2-, and fat-suppressed contrast-enhanced T1-weighted images, and focal unenhanced areas. The parotid gland was anatomically divided into superficial and deep lobes based on the retromandibular vein. The superior and inferior poles were defined as the superior and inferior halves of the parotid gland, respectively. The presence of a capsule formation and intratumoral septa was assessed on T2-weighted images. The SI on T1-weighted images was categorized as hypo-, iso-, or hyperintense relative to the spinal cord; on T2-weighted images as hypo- to isointense, mild hyperintense, or markedly hyperintense relative to the spinal cord; and on fat-suppressed contrast-enhanced T1-weighted images as mild, moderate, or strong enhancement.

Time intensity curves were classified as the following three types:

Type A: time of peak enhancement ≤ 120 s, washout ratio ≥ 30%

Type B: time of peak enhancement ≤ 120 s, washout ratio < 30%

Type C: time of peak enhancement > 120 s.

Washout ratios were calculated using the following formula:

$$\text= \frac-\text}-\text} \times 100 \left(\right).$$

Quantitative MRI measurements included maximum lesion diameter and SIs on axial T1-, T2-, and fat-suppressed contrast-enhanced T1-weighted images, with regions of interest (ROIs) placed on solid components of the lesions (avoiding fluid-filled or artifact-prone areas). ROIs were placed as broadly as possible on the slice with the lesion’s maximum diameter. SI of the spinal cord at the same level as the lesion was measured using 10-mm-diameter circle ROIs to calculate the tumor-to-spinal cord SI ratio (SIR). Apparent diffusion coefficient (ADC) values (× 10−3 mm2/s) were obtained from ADC maps by placing ROIs on solid tumor components.

Qualitative CT findings included hypodense, hyperintense, and calcified areas on unenhanced CT images, and heterogeneity on unenhanced and contrast-enhanced CT images. CT attenuations of hypodense, hyperdense, and calcified areas were defined as < 10HU, 50–200 HU, and > 200 HU, respectively.

Quantitative CT measurements included CT attenuation of solid components on unenhanced and contrast-enhanced CT images. A ROI within the tumor on CT images was placed to include as many solid components as possible while avoiding focal hypodense or unenhanced areas indicating necrosis or cystic change, focal hyperdense areas, and calcification.

Statistical analysis

All statistical analyses were performed using the Statistical Package for the Social Sciences ver. 24.0 (IBM Corp., Armonk, NY, USA) or EZR (Saitama Medical Center, Jichi Medical University, Saitama, Japan). Fisher’s exact test was applied to compare differences in qualitative parameters (gender, location, configuration, heterogeneity, SI, focal unenhanced area, and time-intensity curve type) between the two groups. Interobserver variability in qualitative assessments was evaluated using Kappa statistics, where a Kappa value of ≤ 0.20 was interpreted as slight agreement, 0.21–0.40 as fair agreement, 0.41–0.60 as moderate agreement, 0.61–0.80 as substantial agreement, and ≥ 0.81 as almost perfect agreement. The Mann–Whitney U test was used for comparisons of quantitative variables (age, maximum lesion diameter, SIR, ADC value, and CT attenuation) between the two groups. Interobserver variability in quantitative parameters were assessed using the intraclass correlation coefficient (ICC), where ICC of ≤ 0.5 was interpreted as poor agreement, 0.50–0.75 as moderate agreement, 0.75–0.90 as good agreement, and ≥ 0.90 as excellent agreement. p values < 0.05 were considered significant.