|Year : 2019 | Volume
| Issue : 2 | Page : 173-181
Role of diffusion-weighted MRI in evaluation of patients with salivary gland tumors
Abdallah M Al-Kheshen1, Mohammed S Elfeshawy1, Mohammed E Abd-ElGhani2, Mahmoud M Abomosalam Ali1
1 Department of Radiology, Faculty of Medicine, Al Azhar University, Cairo, Egypt
2 Department of General Surgery, Faculty of Medicine, Al Azhar University, Cairo, Egypt
|Date of Submission||15-Mar-2019|
|Date of Decision||06-Apr-2019|
|Date of Acceptance||02-Jun-2019|
|Date of Web Publication||23-Oct-2019|
Mohammed S Elfeshawy
Department of Radiology, Faculty of Medicine, Al Azhar University, Cairo
Source of Support: None, Conflict of Interest: None
Background and objectives Diffusion-weighted imaging (DWI) and apparent diffusion coefficient (ADC) values give appreciable information about tumor cellularity with tissue contrast between the active and necrotic areas within the tumor.
Aim of the work The aim of this study was to investigate the capacity of diffusion MRI to predict benign and malignant salivary gland tumors, using ADC value and DWI.
Patients and methods This study included 40 patients (24 male and 16 female). Sixteen patients had malignant lesions and 24 had benign ones. The DWI was obtained with b values including 0 and 800 mm2/s. The ADC was generated by measuring identical images at different b values and represented as ADC map, from which the ADC value was calculated. This was a prospective randomized controlled trial. Informed consent was obtained from parents or guardians.
Results The absolute ADC value of lesions was significantly different between benign and malignant salivary gland tumors (P<0.001). The sensitivity of ADC in differentiating benign from malignant lesions in our study was 93.7%, indicating a high true positive rate. Hence, if the average ADC is below 0.85×10−3 mm2/s, there is high probability that the mass will be malignant with high specificity of 95.8%. Results revealed that the mean ADC values of benign and malignant salivary gland tumors were 1.33±0.46×10 and 0.65±0.21×10−3 mm2/s, respectively. The mean ADC value of benign lesions was significantly higher than that of malignant lesions.
Conclusion The use of DWI and ADC values can provide better assessment of salivary gland tumors and predict benign and malignant lesions.
Keywords: diagnostic imaging, head and neck neoplasms, salivary glands
|How to cite this article:|
Al-Kheshen AM, Elfeshawy MS, Abd-ElGhani ME, Abomosalam Ali MM. Role of diffusion-weighted MRI in evaluation of patients with salivary gland tumors. Al-Azhar Assiut Med J 2019;17:173-81
|How to cite this URL:|
Al-Kheshen AM, Elfeshawy MS, Abd-ElGhani ME, Abomosalam Ali MM. Role of diffusion-weighted MRI in evaluation of patients with salivary gland tumors. Al-Azhar Assiut Med J [serial online] 2019 [cited 2019 Dec 7];17:173-81. Available from: http://www.azmj.eg.net/text.asp?2019/17/2/173/269762
| Introduction|| |
The major salivary glands are the parotid glands, submandibular glands, and sublingual glands. Minor salivary glands are widely distributed throughout the oral and palatal mucosa, per tonsillar area, pharynx, larynx, and paranasal sinuses. Tumors affecting salivary glands may be benign or malignant and are diverse in their pathology. Approximately 80% of salivary gland tumors occur in the parotid gland .
Most salivary neoplasms are benign (65–70%). Nearly 80% of parotid gland tumors are benign. Malignant tumors are rare. Malignancy typically presents after the age of 60 years, whereas benign lesions usually occur after the age of 40 years .
Only 5% of all salivary gland tumors occur in childhood .
Surgical approach to parotid tumors is different for benign and malignant neoplasms, but the clinical symptoms do not correlate well with histology, and only few symptoms, such as facial palsy, allow the diagnosis of malignancy. Therefore, preoperative imaging has assumed a major role in surgical planning for assessing the location and malignancy of the tumor .
Preoperative diagnosis of the type of salivary gland mass can help the surgeon to determine the most suitable surgical procedure. MRI is the method of choice for the characterization of salivary gland tumors. MRI is a noninvasive technique that can provide morphological information of the mass and allow correct diagnosis and proper staging. It can accurately determine the deep or superficial location of the mass, extension, contour, and signal pattern .
New MRI techniques including dynamic contrast-enhanced MRI and diffusion-weighted MRI (DW-MRI) have shown promising results in the differentiation between benign and malignant gland tumors in addition to the static MRI .
Diffusion-weighted imaging (DWI) provides functional information related to the random water diffusion of the mass and has a high ability to determine different histologic subtypes of salivary gland tumors. Apparent diffusion coefficient (ADC) values calculated from DWI can provide quantitative information related to random diffusion of water molecules in tissues and functionally complement conventional MRI and has been reported as helpful for narrowing the differential diagnosis of salivary gland masses .
The aim of this study was to investigate the capacity of DW-MRI to predict the benign and malignant salivary gland tumors, using ADC value and DWI.
| Patients and methods|| |
This cross-sectional study included a total of 40 patients (24 male and 16 female), where 16 had malignant lesions and 24 had benign ones. Approval of the ethical committee, and a written informed consent from all the patients were obtained. This study was conducted between March 2017 and September 2018.
The following were the inclusion criteria:
- Any age group and sex.
- Patient with clinical and radiological findings suggestive of salivary gland tumors.
- The lesion has solid portions suitable for region-of-interest (ROI) analysis.
The following were the exclusion criteria:
- Totally cystic lesions.
- Patients with metallic implants.
- Patients with claustrophobia.
- Patients on electrically programmed infusion pump.
All patients were subjected to complete history taking, full clinical examination, ultrasound examination, and MRI examination.
Machines used was Philips Achieva 1.5 T (Philips, Netherlands).
Parameters of diffusion
Axial DWI was performed by using a single-shot T2-weighted echo planar spin-echo sequence with the following parameters: 1600/107; diffusion gradient encoding in three (x, y, z) orthogonal directions; b values of 0 and 800 s/mm2; field of view, 220 mm; matrix size, 128–128; section thickness, 4 mm; section gap, 0 mm; and one signal acquired. At each b value, x, y, and z single-direction DWI and a baseline image (b 0 s/mm2) were acquired; combined ([x _ y _ z]/3) DWI was calculated and performed automatically by the MR instrument.
Determination of the signal intensity and ADC value is done mathematically by postprocessing the DWI data using a computer workstation using the signal acquired.
The ADC values were measured by manually placing ROI in tumor regions on the ADC map. Whenever possible, ROI were placed in at the site of enhanced lesions on contrast-enhanced T1-weighted MRI, also taking into consideration other MRI to carefully place the ROI only in the solid tumor components. Thus, cystic, necrotic, and hemorrhagic tumor areas were excluded.
| Results|| |
This study included 40 cases (24 male and 16 female) patients.
Patients’ ages ranged from 10 to 80 years, with mean of 45 years ([Table 1]).
Location of the lesions
Overall, 80% of our cases were found in the parotid gland, 10% in the submandibular, and 10% in the minor salivary gland (hard palate).
Patients were distributed according to final pathological diagnosis into two groups: the benign group and the malignant group. The benign group included 24 patients, whereas the malignant group included 16 patients.
The diagnoses in the benign group as determined by pathological analysis were pleomorphic adenoma, Warthin tumor, lipoma, and basal cell adenoma.
Pleomorphic adenoma showed to be the most frequent among benign lesions, representing 50% of all benign lesions.
However, malignant group shows the following entities: lymphoma, adenoid cystic carcinoma, and basal cell carcinoma.
The most common malignant lesions found in our study were lymphoma, representing 50% of all malignant tumors ([Table 2]).
All benign tumors showed low T1, high T2, and high short tau inversion recovery (STIR), except lipoma. Lipoma showed high T1 and T2 and low STIR.
Malignant tumors showed low or intermediate T1 and high T2, except lymphoma, which showed low T2.
All of malignant tumors were enhanced ([Table 3]).
Benign tumors were seen to be facilitated in diffusion and T2 shine through, except Wharton’s tumor and basal cell adenoma, which were seen restricted.
ADC values of benign tumors were high (0.7–1.8). The highest ADC value was for pleomorphic adenoma (1.8), and the lowest ADC value was for basal cell adenoma (0.7).
Malignant tumors were restricted in the diffusion.
ADC values of malignant tumors were low (0.49–1). The lowest ADC was for lymphoma (0.4) ([Table 4]).
Accuracy of diffusion-weighted imaging
Lesions were assessed with DWI and ADC map. They were labeled as either benign-looking ‘B’ or malignant-looking ‘M,’ according to presence of criteria of malignancy, for example, high signal in DWIs (i.e. restricted diffusion pattern) and low ADC value. By DWI and ADC map, 20 malignant-looking lesions were detected, and after pathological analysis, only 16 proved to be truly positive for malignancy, with four false-positive cases. A total of 20 lesions were detected as benign looking, and these 20 proved to be truly negative for malignancy (benign), with no false-negative ones. Thus, using DWI in predicting benign and malignant lesions has the sensitivity of 93.7%, specificity of 95.8%, positive predictive value of 93.8%, and negative predicative value of 95.8% with diagnostic accuracy of 94.4% ([Table 5]).
|Table 5 Diagnostic performance of apparent diffusion coefficient value (×10−3 mm2/s) in discrimination of malignant and benign lesions|
Click here to view
Receiver operating characteristics curve was used to define the best cutoff value of ADC value, which was less than 0.85, with sensitivity of 93.7%, specificity of 95.8%, positive predictive value of 93.8%, and negative predictive value of 95.8% with diagnostic accuracy of 94.4% ([Figure 1]).
|Figure 1 ROC curve for prediction. ROC, receiver operating characteristics.|
Click here to view
Recorded data were analyzed using the statistical package for social sciences, version 20.0 (SPSS Inc., Chicago, Illinois, USA). Quantitative data were expressed as mean±SD. Qualitative data were expressed as frequency and percentage.
The following tests were done:
- Independent samples t test of significance was used when comparing between two means.
- χ2 test of significance was used to compare proportions between two qualitative parameters.
- Receiver operating characteristic curve analysis was used to find out the overall productivity of parameter in and to find out the best cutoff value with detection of sensitivity and specificity at this cutoff value.
- The confidence interval was set to 95% and the margin of error accepted was set to 5%. So, the P value was considered significant as follows:
- P value less than 0.05 was considered significant.
- P value less than 0.001 was considered as highly significant.
- P value more than 0.05 was considered insignificant.
| Discussion|| |
In this study, we included 40 patients (24 male and 16 female). Overall, 16 patients had malignant lesions and 24 had benign ones. The DWI was obtained with b values including 0 and 800 mm2/s. The ADC generated by measuring identical images at different b values and represented as ADC map, from which the ADC value was calculated.
In our study, the absolute ADC value of lesions was significantly different between benign and malignant salivary gland tumors (P<0.001).
The sensitivity of ADC in differentiating benign from malignant lesions in our study was 93.7%, indicating a high true positive rate. Hence, if the average ADC is below 0.85×10−3 mm2/s, there is high probability that the mass will be malignant with high specificity of 95.8%.
Our results revealed that the mean ADC values of benign and malignant salivary gland tumors were 1.33±0.46×10 and 0.65±0.21×10−3 mm2/s, respectively. The mean ADC value of benign was significantly higher than that of malignant lesions.
In agreement with the results of our study is a study done by Salama et al.  that was performed on 25 patients with parotid gland masses, with 11 males and 14 females, having age ranged from 22 to 79 years, with mean age of 53.4±13.6 years. The number of benign masses was 18 (72%). The most common benign tumors were pleomorphic adenomas (61.12% of benign tumors), followed by Warthin tumor (38.88% of benign tumors). The malignant tumors represented seven (28%). The most common malignant tumors were mucoepidermoid carcinomas, representing 71.43% off all malignant tumors, and being more frequently seen in males .
They concluded that ADC values of malignant lesions were significantly lower than that of benign lesions, with the exception of Warthin tumor. Although the mean ADC value of malignant tumors (1.03±0.13×103 mm2/s) has been shown to be significantly smaller than that of pleomorphic adenomas (1.89±0.18×10−3 mm2/s), there was an overlap between the mean ADC values of Warthin tumors (0.92±0.22×10−3 mm2/s) and that of malignant tumors (1.03±0.13×10−3 mm2/s) .
The mean ADC value for malignant tumors was 1.03±0.13×10−3 mm2/s. The malignant masses showed low ADC values (0.6±1.2×10−3 mm2/s) .
In our study, the mean ADC value for malignant lesions was 0.65±0.21×10−3 mm2/s, with range of 0.49–0.21×10−3 mm2/s.
The minute difference in range and mean values in our study may be attributed to the fact that the number of lymphoma cases were more in our study, thus lowering the mean ADC value.
This is familiar to that seen in literature, as lymphoma records particularly low ADC values, with a ratio often inferior to 0.5 .
They concluded that ADC value may provide preoperative tissue characterization of the salivary gland tumors.
They also observed that in contrast to salivary gland cancers, malignant lymphomas arising in the salivary glands were associated with extremely low ADCs (ADC value is lower than 0.6×10−3 mm2/s) throughout the lesions. This was consistent with the homogeneous growth patterns of lymphoma cells .
This agrees with our observations in our study as the mean ADC value for lymphoma was 0.55±0.16×10−3 mm2/s.
In the study by Balçık et al. , a total of 41 parotid gland masses from 40 patients (22 females, 18 males; age ranging between 16 and 85 years; mean age, 51.5±19.1 years) were included. Six of these had been diagnosed by histopathological examination. One mass (lipoma) had been diagnosed via pathognomonic MRI and CT features. Pathological diagnosis of these lesions was also diverse, including pleomorphic adenoma, Warthin tumor, basal cell adenoma, dermoid cyst, cyst that was full of keratinized material, lipoma, mucoepidermoid carcinoma, adenoid cystic carcinoma, carcinoma of the salivary duct, adenocarcinoma, carcinoma ex-pleomorphic adenoma, squamous cell carcinoma, and secondary tumors.
However, the most common benign parotid gland lesion was pleomorphic adenoma, and the secondary tumors were the most determined lesions within the malignant tumors.
The mean ADC value of benign lesions measured as 1.74±0.58×10−3 mm2/s, and that of malignant ones as 1.13±0.13×10−3 mm2/s. The mean ADC values of benign lesions were significantly higher than malignant lesions (P=0.006). This agrees with our study results.
Malignant lesions were fewer in number, thus were not compared among themselves.
In the comparisons that were performed among pleomorphic adenoma (group 1), Warthin tumor (group 2), and malignant tumors (group 3), the ADC values of the pleomorphic adenoma were significantly higher than malignant tumors (P<0.001) or Warthin tumors (P=0.001). The ADC values of malignant tumors were significantly higher than Warthin tumors (P=0.001).
Balçık et al.  concluded no absolute optimal cutoff, preferring to separate the cutoff between dual groups of pleomorphic adenoma, Warthin, and malignant lesion. Thus, they proposed two numbers. Accordingly, the pleomorphic adenomas were differentiated from all other benign and malignant tumors with 1.60×10−3 mm2/s cutoff ADC value, which had a sensitivity of 94.7% [18/19, 95% confidence interval (CI)] and specificity of 100% (22/22, 95% CI). The malignant tumors were differentiated from Warthin tumors with 1.01×10−3 mm2/s cut off ADC value, which had a sensitivity of 92.3% (12/13, 95% CI) and specificity of 100% (4/4, 95% CI).
This agrees with the recorded low ADC value for Warthin tumor in our study which was 0.8×10−3 mm2/s which is considered a false-positive result on the basis of ADC value. However, owing to paucity of cases in this category, no dual comparison between Warthin tumor group and malignant lesions group was done.
This finding might be attributed to the intense lymphoid accumulation in the stroma and proliferation of the epithelial component leading to a decrease in the extracellular extravascular space and therefore a decrease in ADC, even less than malignant lesions .
The major limitation of our study was that it has been done on several salivary glands which offer naturally diverse histological profiles; thus, the numbers of the lesions in each category were few. So further separate studies dedicated to different pathological entities may provide differentiation between tumor types.
In our study, there was a striking false result based on DWI and ADC value. The patient had a parotid Warthin tumor and basal cell adenoma. The calculated ADC values were 0.8×10 and 0.7×10−3 mm2/s, respectively, which may be mistaken as malignant lesions.
This can be explained by the fact that this lesion is composed of a diffused, well-organized lymphoid tissue and lymphocytic interstitial infiltrate .
| Selected cases|| |
The use of DWI and ADC values can provide better assessment of salivary gland tumors and predict benign and malignant lesions ([Figure 2],[Figure 3],[Figure 4],[Figure 5],[Figure 6]). Thus, we recommend using DWI as part of the routine examination of salivary gland tumors as well as integrating the ADC value as part of the regular interpretation by radiologists. As DWI is a noninvasive method and has the advantage of not being time consuming, nor does it require additional special equipment, it represents a valuable addition to the conventional MRI examination adding up considerably to its efficacy as discussed before.
|Figure 2 A case of right parotid pleomorphic adenoma shows deep-lobe lobulated soft tissue lesion displaying low signal in T1 (a) and high signal T2 (b) and STIR Wls (c) with heterogeneous enhancement in the postcontrast series (d). It appear T2 shine through in DWI (e and f). DWI, diffusion-weighted imaging.|
Click here to view
|Figure 3 White arrow shows well-defined right parotid mass lesion which appears low in T1 (a) and mixed intensity in T2 (b) and STIR (c) (high signal denoting cystic component) with mild heterogeneous enhancement (d). It appears restricted in DWI (e and f). DWI, diffusion-weighted imaging.|
Click here to view
|Figure 4 A case shows sharply demarcated soft tissue lesion within the superficial lobe of the left parotid gland. The lesion displays low T1 (a) and high T2 (b) and STIR (c) signal with homogenous enhancement in the postcontrast series (d). It appears restricted in DW (e and f).|
Click here to view
|Figure 5 A case of left parotid lymphoma shows left parotid gland large lesion, displaying low T1 (a) and T2 (b) and high STIR (c) signal. Heterogeneous enhancement in the postcontrast series (e) with nonenhancing central areas likely breaking down noted. Evident thickening and enhancement of the facial nerve through the stylo-mastoid foramen is seen impressive of perineural spread through the facial nerve ‘white arrow.’ The aforementioned lesion appears restricted in DWI (f and g). DWI, diffusion-weighted imaging.|
Click here to view
|Figure 6 A case of right adenoid cystic carcinoma. White arrow indicates ill-defined mass lesion seen in the hard palate. It exhibits low T1 (a) and high T2 (b) with heterogeneous enhancement (c and d). It appears high in DWI and low ADC (restricted diffusion pattern) (e and f). ADC, apparent diffusion coefficient; DWI, diffusion-weighted imaging.|
Click here to view
Source(s) of support: Department of Radiodiagnosis, Al Hussein Univeristy Hospital.
Organization: Faculty of Medicine for Boys.
Place: Al Hussein Univeristy Hospital.
Financial support and sponsorship
Conflicts of interest
There are no conflcits of interest.
| References|| |
Namboodiripad PC. A review: Immunological markers for malignant salivary gland tumors. J Oral Biol Craniofac Res 2014; 4:127–134.
Zhang Q, Wei YM, Qi YG, Li BS. Early changes in apparent diffusion coefficient for salivary glands during radiotherapy for nasopharyngeal carcinoma associated with xerostomia. Korean J Radiol 2018; 19:328–333.
Iro H, Zenk J. Salivary gland diseases in children. GMS Curr Top Otorhinolaryngol Head Neck Surg 2014; 13:Doc06
Christe A, Waldherr C, Hallett R. MR imaging of parotid tumors: typical lesion characteristics in MR imaging improve discrimination between benign and malignant disease. Am J NeuroradiolR 2011; 32:1202–1207.
Yerli H, Aydin E, Haberal N, Harman A, Kaskati T, Alibek S. Diagnosing common parotid tumors with magnetic resonance imaging including diffusion-weighted imaging vs.fine-needle aspiration cytology: a comparative study. Dentomaxillofac Radiol 2010; 39:349–355.
Terra GTC, Oliveira JXD, Hernandez A, Lourenço SV, Arita ES, Cortes ARG. Diffusion-weighted MRI for differentiation between sialadenitis and pleomorphic adenoma. Dentomaxillofac Radiol 2017; 46:20160257.
Salama AA, El-Barbary AH, Mlees MA, El-Sayed Esheba G. Value of apparent diffusion coefficient and magnetic resonance spectroscopy in the identification of various pathological subtypes of parotid gland tumors. Egypt J Radiol Nucl Med 2014; 46:45–52.
Thiagarajan S, Sudhir NV, Poonam J. A review of salivary gland neoplasms and its management. Tata Memorial Centre Mumbai India 2014; 4:16–36.
Eida S, Sumi M, Sakihama N, Takahashi H, Nakamura T. Apparent diffusion coefficient mapping of salivary gland tumors: prediction of the benignancy and malignancy. Am J Neuroradiol 2007; 28:116–121.
Balçık C, Akan H, İncesu L. Evaluating of parotid gland tumours according to diffusion weighted MRI. Eur J Gen Med 2014; 11:77–84.
Orme IM, Basaraba RJ. The formation of the granuloma in tuberculosis infection. Semin Immunol 2014; 26:601–609.
[Figure 1], [Figure 2], [Figure 3], [Figure 4], [Figure 5], [Figure 6]
[Table 1], [Table 2], [Table 3], [Table 4], [Table 5]