Abstract
Objectives
To validate an MRI algorithm characteristic of pleomorphic adenoma (PA).
Study design
Cross-sectional analysis.
Setting
Academic tertiary-care medical center.
Methods
A radiologic algorithm for the MRI diagnosis of PA was developed on the basis of five “high probability” criteria that all must be fulfilled for the MRI to qualify as a positive test result: bright T2-signal, sharp margins, heterogeneous nodular enhancement, lobulated contours, T2-dark rim. We then identified MRI images from our institutional database to test the diagnostic accuracy of the proposed algorithm.
Results
A total of 103 parotidectomy cases with adequate MRI studies were identified (pleomorphic adenoma n = 41, mucoepidermoid carcinoma n = 11, Warthin’s tumor n = 8, adenoid cystic carcinoma n = 6, oncocytoma n = 6, acinic cell carcinoma n = 5, salivary duct carcinoma n = 5, and other n = 21). Eighteen of 21 cases that met all five “high probability” MRI criteria were consistent with PA on final histopathology; 3 were consistent with carcinoma. MRI had a specificity of 95.1% [95% confidence interval: 85.6–98.7%] and sensitivity of 43.9% [95% C.I.: 28.8–60.1%] for PA. The positive predictive value was 85.7% [95% C.I.: 70.4–100%] and the negative predictive value was 71.9% [95% C. I.: 62.0–81.9%]. The overall diagnostic accuracy was 74.8% [95% C.I.: 66.2–83.3%].
Conclusion
A “high probability” MRI is about 95% specific for pleomorphic adenoma. A subset of patients with MRI imaging that is highly suggestive of PA may reliably avoid further workup. The value of MRI in this setting is especially useful if preoperative fine needle aspiration is not readily available. A significant proportion of PAs, however, have indeterminate imaging features that overlap considerably with other benign and malignant lesions.
1
Introduction
Parotid tumors account for about 75% of salivary gland neoplasms and 2–5% of head and neck tumors overall. History and physical examination are the most important tools for diagnosis and the initial diagnostic strategy includes differential diagnosis between tumor and other benign conditions, such as cysts, inflammatory processes, and lymph node hyperplasia . Further preoperative workup of major salivary gland neoplasms often includes fine needle aspiration biopsy (FNAB) and magnetic resonance imaging (MRI) of the head and neck .
Knowledge of whether a parotid tumor is benign or malignant can be helpful to the surgeon preoperatively. This information could influence the planned surgical approach and planned margins (including extent of resection, management of the facial nerve, and the likelihood of a neck dissection) or potentially justify conservative management in a patient who is a poor surgical candidate . Fine-needle aspiration biopsy (FNAB) has been recommended as the diagnostic modality of first choice for characterization of parotid mass lesions . However, FNAB has limited diagnostic yield and has even been shown to potentially misdirect surgical management . The results of fine-needle aspiration cytology are not always conclusive because insufficient specimens are sometimes obtained because of a small sample size or because of the deep location of a tumor . As such, there are a subset of patient for whom FNAB is either inconclusive or less than ideal as a primary modality for evaluation of parotid tumors.
Imaging is often performed to assess the location and extent of a parotid mass and MRI is particularly useful for demonstrating the interface of tumor and surrounding tissues . Interestingly, recent studies seem to suggest that for a subset of cases, it may also be possible to use MRI alone (without FNAB) to narrow the histological differential diagnosis of a parotid gland tumor . Certain imaging features, particularly on MRI, have been described as more suggestive of benign or malignant histology . However, previous studies have shown only modest overall ability to predict benign or malignant histology, with sensitivities and specificities ranging from 40 to 67% and from 81 to 89% . For this reason, many surgeons continue to pursue fine needle aspiration biopsy (FNAB) pre-operatively, in order to optimize the treatment plan .
Pleomorphic adenoma (PA), a benign epithelial neoplasm, is the most common parotid tumor, and has a characteristic appearance on MRI that has been well described . Despite the relatively modest overall performance of MRI in distinguishing benign from malignant masses, the characteristic appearance of PA suggests that it may be possible to use MRI to make an accurate preoperative diagnosis and obviate the need for FNAB in a large subset of PA cases. Initial data from Heaton et al. support this idea by showing that certain MRI features have a high positive predictive value for the diagnosis of PA. Many prior authors have similarly identified individual MR features characteristic of pleomorphic adenoma and report on the predictive value of each feature . However, there are no validation studies to date assessing the diagnostic reliability of using MR features to make a diagnosis of pleomorphic adenoma. Here we aimed to assess whether we could validate a highly specific MRI algorithm for reliable identification of pleomorphic adenoma among primary parotid gland tumors.
2
Methods
A radiologic algorithm for the MRI diagnosis of pleomorphic adenoma was developed on the basis of five “high probability” criteria that all must be fulfilled for the MRI to qualify as a positive test result for pleomorphic adenoma: bright T2-signal, sharp margins, heterogeneous nodular enhancement, lobulated contours, T2-dark rim (see Table 1 ). We then identified MRI images from our institutional database to assess the validity of the proposed MRI algorithm as a diagnostic marker for pleomorphic adenoma. This Health Insurance Portability and Accountability Act (HIPPA)-compliant study was performed with Institutional Review Board approval and a waiver of informed consent from the UCLA Office for the Protection of Research Subjects.
| T2 signal | Margins | Enhancement pattern | Contour | T2 rim | |
|---|---|---|---|---|---|
| Criteria in favor of PA | 1 = bright | 1 = sharp | 1 = heterogeneous nodular | 1 = lobulated | 1 = T2 dark rim |
| Neutral criteria | 2 = intermediate | 2 = intermediate | 2 = none 3 = uniform | ||
| Criteria against PA | 3 = dark | 3 = grossly infiltrative | 4 = thick peripheral with necrosis | 2 = not lobulated | 2 = no dark rim |
Search was performed in the pathology database for dates ranging from 2001 to 2012 for all parotid neoplasms. The radiological information system was cross-referenced to identify cases for which MRIs providing pre-operative imaging of the parotid gland were available: 122 cases were encountered. Patients with a history of prior parotidectomy were excluded. MRI studies that were deemed inadequate because of poor technical quality, limited number of axial images through the area of the parotid gland, or because they did not include contrast-enhanced imaging were also excluded. The acquisition parameters varied, as the MRIs were obtained across a number of different scanners, 1.5-T and 3-T, over the course of more than 10 years. All studies selected for the analysis group include high resolution T2-weighted, T1-weighted, and fat-suppression contrast-enhanced T1-weighted images. The final analysis group of 103 cases consisted of 42 males, 61 females (average age 51.5 +/− 16.6 years, range 12–84 years). There were 63 benign tumors and 40 malignant tumors. Distribution of pathology diagnoses was as follows: pleomorphic adenoma n = 41, mucoepidermoid carcinoma n = 11, Warthin’s tumor n = 8, adenoid cystic carcinoma n = 6, oncocytoma n = 6, acinic cell carcinoma n = 5, salivary duct carcinoma n = 5, lymphoma n = 4, basal cell adenoma n = 3, myoepithelioma n = 3, basal cell adenocarcinoma n = 3, carcinoma ex-pleomorphic adenoma n = 2, myoepithelial carcinoma n = 1, melanoma n = 2, lymphangioma n = 1, solitary fibrous tumor n = 1, Merkel cell carcinoma n = 1. The MR images were reviewed by a subspeciality-certified neuroradiologist who was blind to the diagnosis and clinical information. Each scan was assessed qualitatively on the following features: T2-signal (bright, intermediate, or dark); margins (sharp, intermediate, or grossly infiltrative); enhancement pattern (heterogeneous nodular, none, uniform, or thick peripheral with necrosis); contour (lobulated, or not lobulated); T2-rim (dark rim, or no dark rim). Figs. 1 and 2 demonstrate each of these categories.
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
