Skip to main content
The British Journal of Radiology logoLink to The British Journal of Radiology
. 2012 Jun;85(1014):709–713. doi: 10.1259/bjr/55639432

MRI findings of mucoepidermois carcinoma of the parotid gland: correlation with pathological features

N Kashiwagi 1, K Dote 2, K Kawano 3, Y Tomita 4, T Murakami 1, K Nakanishi 5, Y Araki 6, K Mori 7, N Tomiyama 8
PMCID: PMC3474098  PMID: 21896663

Abstract

Objective

The aim of this study was to correlate the MRI appearance and pathological findings in each grade of the mucoepidermoid carcinoma (MEC) of the parotid gland.

Methods

We reviewed surgically proven MECs of parotid glands in 20 patients. Pathologically, 5 tumours were high-grade, 3 were intermediate and 12 were low-grade. MR images were evaluated with emphasis on signal intensities on T2 weighted images, margin characteristics and lymph node metastasis, correlating these with pathological features.

Results

Among the high-grade MECs, four out of five tumours showed inhomogeneous low to intermediate signal intensity on T2 weighted images, reflecting high cellularity. All tumours had an ill-defined margin, reflecting invasive tumour growth. Among the intermediate-grade MECs, all three tumours showed intermediate signal intensity on T2 weighted images and two tumours had an ill-defined margin. Among the low-grade MECs, 11 of the 12 tumours had a hyperintense area on T2 weighted images because of the existence of abundant mucin secreting cells. Seven tumours had an ill-defined margin, reflecting peritumoural inflammatory changes rather than invasive tumour growth. Lymph node metastasis was seen in three high-grade MECs.

Conclusion

MECs of the parotid gland show variable MRI findings reflecting their histological nature, which seems to have certain tendencies depending on the tumour grade.


Mucoepidermoid carcinoma (MEC) is the most common malignancy of the salivary gland, accounting for 3–15% of all salivary gland tumours. Their most favoured site is the parotid gland, which accounts for over 80% of MECs [1-4]. Histologically, MEC has been classified as low, intermediate or high grade according to intracystic components, mitotic figures, neural invasion, necrosis and cellular anaplasia. This histological grading has been reported as correlating with clinical behaviour [1-4]. Although clinical and pathological features of MEC have been well described, radiological features of MEC have not been systematically reported. In this study, we delineate MRI findings in each grade of MEC and correlate these with histological findings.

Methods and materials

20 patients with MEC of the parotid gland who underwent MR examination before resection between 2002 and 2009 were reviewed. There were 9 males and 11 females with an age range of 13–83 years (mean 60 years). The histological grades were classified on the basis of the well-known criteria proposed by Goode et al [1] and were as follows: 5 tumours were high grade, 3 were intermediate and 12 were low grade. MR images were obtained using a 1.5-T machine (Symphony, Siemens, Erlangen, Germany; Signa, GE Healthcare, Milwaukee, WI) in 17 patients and a 3-T machine (Trio, Siemens) in 3 patients. In all patients, transverse T1 weighted spin-echo (SE) images (repetition time/echo time (TR/TE) 466–680 ms/14–20 ms) and T2 weighted fast SE images (TR/TE 2500–5200 ms/80–120 ms) were obtained. In some cases, transverse fat-suppressed T2 weighted, coronal T1 weighted or fat-suppressed T2 weighted images were obtained. In 11 cases, post-contrast transverse and coronal T1 weighted images were acquired with or without a fat-suppression technique. Section thickness was 4–6 mm and pixel size was 0.4×0.7 mm–1.0×1.0 mm.

Two experienced radiologist (KN and YA) investigated the MRI findings with an emphasis on the presence of hyperintense areas on T2 weighted images, boundary definition and the presence of lymph node metastasis. The signal intensity was defined in relation to muscle and fat. Low signal intensity was defined as less than that of muscle, intermediate signal intensity was greater than or equal to that of muscle but less than that of fat and high signal intensity was greater or equal to that of fat. Boundary definition was classified as well defined or ill defined. The lymph node was defined as pathological when it had a minimal axial diameter of >10 mm or central necrosis. Consensus was always obtained in cases of discrepancy between the readers. An experienced radiologist (NK) and pathologists correlated the MR images and pathology.

Results

The patient characteristics, histological grading and MRI findings are summarised in Table 1. Among high-grade MECs, all five tumours showed inhomogeneous low to intermediate signal intensity on T2 weighted images (Figure 1a) and only one tumour had a small hyperintense area on T2 weighted images. All five tumours had an ill-defined margin (Figure 1b). Regarding pathological correlation, tumours had predominant epidermoid cells and a lack of mucous cells and their products, resulting in solid architectural pattern (Figure 1c), and infiltrative growth of tumour cells with or without capsulation was seen (Figure 1d). Three of the five tumours had pathological lymph nodes, all of which were confirmed on pathological examination as metastatic lymph nodes. Among intermediate-grade MECs, all three tumours had intermediate signal intensity on T2 weighted images (Figure 2a). One tumour had a well-defined margin (Figure 2b) and the other two had an ill-defined margin. Pathological examination showed an increased number of mucous cells (Figure 2c). Regarding pathological correlation, the former showed a fibrous capsule surrounding the tumour (Figure 2d) and the later showed infiltrative tumour growth. None had a pathological lymph node. Among low-grade MECs, 11 out of 12 tumours had a hyperintense area on T2 weighted images (Figure 3a). Five tumours had a well-defined margin and seven tumours had an ill-defined margin (Figure 3b,c). Pathologically, low-grade tumours had abundant mucous cells with their products, resulting in a cystic architectural pattern (Figure 3d). Upon pathological correlation, the well-defined margin reflects the fibrous capsule of the tumour. Tumours with an ill-defined margin also had a fibrous capsule surrounding them. However, infiltration of inflammatory cells outside of the capsule was seen (Figure 3e). None had a pathological lymph node.

Table 1. Histological grade and MRI findings.

Case number Age (years)/sex Histological grade Presence of hyperintense area on T2WI Margin characteristics Lymphadenopathy
1 68/F High None Ill defined No
2 41/M High Present Ill defined No
3 77/M High None Ill defined Yes
4 29/M High None Ill defined Yes
5 83/F High None Ill defined Yes
6 51/F Intermediate None Ill defined No
7 27/M Intermediate None Well defined No
8 60/M Intermediate None Ill defined No
9 18/M Low Present Well defined No
10 71/F Low Present Well defined No
11 34/F Low Present Ill defined No
12 27/F Low Present Ill defined No
13 57/F Low Present Ill defined No
14 72/M Low Present Ill defined No
15 29/F Low Present Well defined No
16 48/F Low Present Ill defined No
17 59/F Low None Ill defined No
18 78/F Low Present Ill defined No
19 57/F Low Present Well defined No
20 13/F Low Present Well defined No

F, female; M, male; T2WI, T2 weighted images.

Figure 1.

Figure 1

An 83-year-old female with a high-grade mucoepidermoid carcinoma (Case 5). (a) Axial T2 weighted image shows a lobulated mass with inhomogeneous low to intermediate signal intensity. (b) Axial T1 weighted image shows an ill-defined margin. (c) Microscopic examination of a central area shows a proliferation of spindle-shaped cells with a high nucleus-to-cytoplasm ratio. This high cellularity seems to contribute to the low to intermediate signal intensity on T2 weighted images (haematoxylin and eosin staining, ×100). (d) Microscopic examination of a peripheral area shows infiltrative growth of the tumour cells (arrows) without capsulation (haematoxylin and eosin staining, ×100).

Figure 2.

Figure 2

A 27-year-old male with an intermediate-grade mucoepidermoid carcinoma (MEC) (Case 7). (a) Axial T2 weighted image shows a mass with intermediate signal intensity. (b) Axial T1 weighted image shows a well-defined margin. (c) Microscopic examination of a central area shows an admixture of mucus cells and epidermoid cells. The cellularity is lower compared with that of high-grade MECs (haematoxylin and eosin staining, ×100). (d) Microscopic examination of a peripheral area shows fibrous capsule (arrows) between the tumour and normal parotid gland (haematoxylin and eosin staining, ×100).

Figure 3.

Figure 3

A 48-year-old female with a low-grade mucoepidermoid carcinoma (Case 16). (a) On T2 weighted image, the lateral component of the mass shows marked high signal intensity, suggesting a cystic portion. (b) On T1 weighted image, the medial margin of the tumour is ill-defined. (c) Post-contrast fat-suppressed T1 weighted image shows an ill-defined margin. (d) Microscopic examination shows a large cavity corresponding to the cystic area on MR images. The arrows indicate secreted mucinous materials (haematoxylin and eosin staining, ×100). (e) On microscopic examination of a peripheral area, although a fibrous capsule exists, infiltration of inflammatory cells (arrows) is seen outside of the capsule, which seems to cause the ill-defined margin on MR images (haematoxylin and eosin staining, ×100).

Discussion

MEC of the salivary gland is believed to arise from pluripotent reserve cells of the excretory ducts and this tumour comprised three different cell types: mucinous, intermediate and epidermoid. As a result of this cellular heterogenity, MECs have a wide range of histological features and are classified as high, intermediate or low grade. This grading reportedly correlates with the clinical behaviour of the tumours [1-4]. Although fine-needle biopsy is a useful and minimally invasive adjunct in the diagnosis of parotid gland tumours, it is often difficult to differentiate between benignity and malignancy or to establish tumour grades because of insufficient specimens, sampling errors and the subjective opinion of the cytopathologist [2,5]. Therefore pre-operative imaging has an important role in surgical planning.

In our study, the small number of patients makes a statistically significant correlation difficult. Nevertheless, the MRI findings seem to have a tendency to be related to the histological grade. Among high-grade MECs, although non-specific, the tumours showed MRI findings suggesting high-grade malignancies [6-9]; low to intermediate signal intensities on T2 weighted images reflect high-cellularity, the ill-defined margin reflects the invasive growth of tumour cells and the frequent nodal metastasis reflects a propensity of lymphatic involvement.

MRI features of intermediate-grade MECs showed a combination of those of low- and high-grade MECs; however, further study with a larger number of intermediate-grade MECs is needed for better characterisation.

Among low-grade MECs, tumours usually had a hyperintense area on T2 weighted images, reflecting a cystic architectural pattern because of the existence of abundant mucin-secreting cells. A similar tendency of increased signal intensity on T2 weighted images in lower-grade malignancies was reported for adenoid cystic carcinomas of the head and neck region [10]. In contrast with the previous literature that describes ill-defined margins on MR images as a predictor of high-grade malignancies [6-9], half of our low-grade MECs had ill-defined margins on MR images, reflecting peritumoural inflammatory changes. This peritumoural reactive inflammation in low-grade MECs has also been described in the pathology literature [2,11]; however, our study is the first to present this radiologically. This may be because previous studies of MECs included only a small number of patients with low-grade MECs, ranging from one to four patients per study [6-9]. In contrast, our study is the first to radiologically evaluate more than ten cases of low-grade MECs. Therefore, for the MECs of the parotid glands, a diagnosis of high-grade MECs should not be concluded based solely on an ill-defined margin on MR images.

References

  • 1.Goode RK, Auclair PL, Ellis GL. Mucoepidermoid carcinoma of the major salivary glands: clinical and histopathological analysis of 234 cases with evaluation of grading criteria. Cancer 1998;82:1217–24 [DOI] [PubMed] [Google Scholar]
  • 2.Brandwein MS, Ivanov K, Wallace DI, Hille JJ, Wang B, Fahmy A, et al. Mucoepidermoid carcinoma. A clinicopathologic study of 80 patients with special reference to histological grading. Am J Surg Pathol 2001;25:835–45 [DOI] [PubMed] [Google Scholar]
  • 3.Nance AM, Seethala RR, Wang Y, Chinosea IS, Mayer EN, Johnson JT, et al. Treatment and survival outcomes based on histological grading in patients with head and neck mucoepidermoid carcinoma. Cancer 2008;113:2082–9 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 4.Pires FR, Almeida OP, Araujo VC, Kowalski LP. Prognostic factors in head and neck mucoepidermoid carcinoma. Arch Otolaryngol Head Neck Surg 2004;130:174–80 [DOI] [PubMed] [Google Scholar]
  • 5.Seethala RR, LiVolsi VA, Baloch ZW. Relative accuracy of fine needle aspiration and frozen section in the diagnosis of lesions of the parotid gland. Head Neck 2005;27:217–23 [DOI] [PubMed] [Google Scholar]
  • 6.Som PM, Biller HF. High-grade malignancies of the parotid gland: identification with MR imaging. Radiology 1989;173:823–6 [DOI] [PubMed] [Google Scholar]
  • 7.Vogl TJ, Dresel SH, Späth M, Grevers G, Wilimzig C, Schedel HK, et al. Parotid gland: plain and gadolinum-enhanced MR imaging. Radiology 1990;177:667–74 [DOI] [PubMed] [Google Scholar]
  • 8.Freling NJM, Molenaar WM, Vermey A, Mooyaart EL, Panders AK, Annyas AA, et al. Malignant parotid tumors: clinical use of MR imaging and histologic correlation. Radiology 1992;185:691–6 [DOI] [PubMed] [Google Scholar]
  • 9.Motoori K, Iida Y, Nagai Y, Yamamoto S, Ueda T, Funatsu H, et al. MR imaging of salivary duct carcinoma. Am J Neuroradiol 2005;26:1201–6 [PMC free article] [PubMed] [Google Scholar]
  • 10.Sigal R, Monnet O, de Baere T, Micheau C, Shapeero LG, Julieron M, et al. Adenoid cystic carcinoma of the head and neck: evaluation with MR imaging and clinical-pathologic correlation in 27 patients. Radiology 1992;184:95–101 [DOI] [PubMed] [Google Scholar]
  • 11.Muller S, Barnes L, Goodurn WJ., Jr Sclrelosing mucoepidermoid carcinoma of the parotid. Oral Surg Oral Med Oral Pathol Oral Radiol Endod 1997;83:865–90 [DOI] [PubMed] [Google Scholar]

Articles from The British Journal of Radiology are provided here courtesy of Oxford University Press

RESOURCES