|Year : 2020 | Volume
| Issue : 3 | Page : 972-980
Role of magnetic resonance sialography in diagnosis of salivary gland diseases: a meta-analysis
El-Sayed El-Mekkawy El-Sayed1, Rehab M Habib1, Mohamed I. M. Elhalal2
1 Department of Radiology, Faculty of Medicine, Menoufia University, Menoufia, Egypt
2 Department of Radiology, Misr Radiology Center, Cairo, Egypt
|Date of Submission||21-Dec-2018|
|Date of Decision||03-Feb-2019|
|Date of Acceptance||10-Feb-2019|
|Date of Web Publication||30-Sep-2020|
Mohamed I. M. Elhalal
El-Khadra, El-Bajour, El-Menoufia
Source of Support: None, Conflict of Interest: None
To establish the diagnostic accuracy of magnetic resonance sialography (MRS) when compared with the gold standard conventional sialography (CS) in patients with suspected salivary gland disease.
This is a systematic review and meta-analysis to compare both studies. CS is still considered the standard modality for assessing ductal abnormalities. MRS has been applied as an alternative.
We searched PubMed, Cochrane CENTRAL, Scopus, Science Direct, and Web of Science for all studies that compared MRS versus CS regarding radiological outcomes and diagnostic performance as primary outcomes. Combined outcomes were pooled as risk ratios in a fixed model or diagnostic odds ratio (DOR) in a random model, using Reviewer Manager 5.3 and Meta-DiSc 1.4 for Windows.
Eight studies (n = 285 patients) were included in the final analysis. Our results showed that MRS was effective in the detection of main duct according to patients more than CS [risk ratio = 0.84, 95% confidence interval (CI), 0.71–1, P = 0.05). However, it showed no significant difference between the two groups in detection of main duct according to glands. Moreover, there are no significant differences between MRS and CS regarding detection of sialolithiasis, sialectasis, stenosis with stones, or strictures. The diagnostic performance in the detection of sialolithiasis according to glands was as follow: the pooled sensitivity was 0.88 (95% CI, 0.75–0.95), the pooled specificity was 0.98 (95% CI, 0.93–1.00), and the pooled DOR under random effect was 181.64 (95% CI, 39.420–837). Moreover, the diagnostic performance in the detection of the stenosis with stones according to patients was as follows: the pooled sensitivity was 0.727 (95% CI, 0.498–0.893), the pooled specificity was 0.979 (95% CI, 0.889–0.99), and the diagnostic odds ratio under random effect was 78.73 (95% CI, 2.18–508.81).
There is no difference between MRS and CS in radiological findings except in case of detection of main duct according to patients, which indicates that MRS is more effective in revealing the proper information about main salivary ducts. Moreover, MRS provides a noninvasive alternative to CS especially if acute sialadenitis is present or cannulation is not successful.
Keywords: conventional sialography, magnetic resonance sialography, salivary gland diseases
|How to cite this article:|
El-Sayed ESE, Habib RM, Elhalal MI. Role of magnetic resonance sialography in diagnosis of salivary gland diseases: a meta-analysis. Menoufia Med J 2020;33:972-80
|How to cite this URL:|
El-Sayed ESE, Habib RM, Elhalal MI. Role of magnetic resonance sialography in diagnosis of salivary gland diseases: a meta-analysis. Menoufia Med J [serial online] 2020 [cited 2021 May 15];33:972-80. Available from: http://www.mmj.eg.net/text.asp?2020/33/3/972/296691
| Introduction|| |
Saliva reflects the physiologic state of the body. Salivary gland diseases may be inflammatory, noninflammatory, non-neoplastic, or neoplastic lesions .
Advances in imaging have led to improved sensitivity in the diagnosis of diseases that involve the major salivary glands. Ultrasound, plain radiography, sialography, MRI, computed tomography (CT), and nuclear scintigraphy/positron emission tomography all play a part .
Conventional or digital subtraction sialography has long been considered the gold standard for imaging the salivary ductal system. Although this technique offers excellent anatomical detail and spatial resolution, it carries significant disadvantages . Owing to the complex nature of the procedure, patient cooperation and operator experience are required for success. The reported failure rate is as high as 4%, especially in patients with a small submandibular duct opening or papillary stenosis.
MRS has become an alternative method for imaging the salivary gland and duct . MRS is a noninvasive method for evaluating the ductal structures of the major salivary glands. Although the technique and its applications have evolved greatly from its initial description in the mid-1990s, it remains rooted in the simple concept that stationary fluid, such as saliva, has intrinsic high T2 signal intensity, obviating instillation of contrast material to 'opacify' the salivary ducts .
The purpose of our study was to establish the diagnostic accuracy of MRS when compared with the gold standard of conventional sialography (CS) in a large group of patients with suspected salivary duct disease.
| Methods|| |
We conducted this systematic review and meta-analysis according to PRISMA flow diagram guidelines  and Cochrane Handbook for Systematic Reviews of Intervention .
We performed a comprehensive search of five electronic databases (PubMed, Scopus, ISI Web of Science of clinical trials, Science Direct, and Cochrane Library) for all eligible studies using the following search building: salivary gland disease and magnetic resonance sialography (MRS), and CS. There was no restriction by the year of publication or the language of study.
We included all studies either randomized clinical trials or observational that met our following inclusion criteria: (a) population: patients with salivary gland diseases; (b) intervention: MRS; (c) comparator: CS as gold standard; (d) outcomes: efficacy end points (detection of sialolithiasis, stenosis with stones, sialectasis, stricture, and main duct) and diagnostic performance of sialolithiasis and stenosis with stones; and (e) study design: randomized clinical trials and observational studies. We excluded all papers not related to our inclusion criteria. Eligibility screening was conducted in a two-step wise manner (title/abstract screening and full-text screening). All reviewers preformed each step, and consensus was reached upon discussion among all reviewers.
We extracted data from related studies on standard excel sheet. We abstracted data as follows: list of authors, year of publication, sample size, study design, and baseline characteristics of enrolled patients. Moreover, we extracted outcomes if reported as follows: detection of sialolithiasis, stenosis with stones, sialectasis, stricture and main duct and the diagnostic accuracy of sialolithiasis and stenosis with stones according to glands; in addition, the same outcomes according to patients.
We evaluated the methodological quality of included studies using Quality Assessment of Diagnostic Accuracy Studies (QUADAS)  and Standards for Reporting Studies of Diagnostic Accuracy (STARD) , to assess the risk of bias among included studies. The QUADAS tool includes the following domains: patient selection, index and comparator test, reference standard, and flow and timing. The judgment of authors are 'Yes,' 'No,' and 'unclear risk.' According to Egger and colleagues, the assessment of publication bias using the funnel plot method and the Egger's test is unreliable for less than 10 included studies. Therefore, in the present study, we could not assess for publication bias owing to the small number of included studies ,.
Dichotomous data of efficacy outcomes were pooled as risk ratio (RR) using the Mantel-Haenszel method. We performed all statistical analysis in this study by RevMan software package (version 5.3; Cochrane Collaboration, Oxford, UK) and Meta-DiSc (version 1.4). We assessed the statistical heterogeneity using the χ2 test and its extent measured by I 2 test. We used random effect model between all studies to minimize the heterogeneity if present; otherwise, the fixed-effects model was used. We performed a sensitivity analysis to assess the contribution of each study to the pooled estimation by excluding one trial at a time and recalculating the combined mean difference estimation for the remaining studies.
| Results|| |
Results of the literature search
After searching five databases, we obtained 149 citation after removing duplicates. We started records title and abstract screening, and 37 studies were eligible for full-text screening then we involved about eight studies ,,,,,,,, which met our inclusion criteria (as shown in PRISMA follow diagram (Supplementary File No. 1).
Characteristics of the included studies
The included studies were randomized clinical trials and observational studies with a total number of 285 patients. Included studies compared the diagnostic performance of MRS and CS in detection of radiological findings in salivary gland diseases. The summary of included studies and the baseline characteristics of their population are shown in [Table 1].
Quality of the evidence
The quality of included studies was high quality according to Quality Assessment of Diagnostic Accuracy Studies (QUADAS) and Standards for Reporting Studies of Diagnostic Accuracy (STARD). Quality assessment domain summary is shown in [Figure 1] (authors' judgment with justifications is shown in Supplementary File No. 2).
|Figure 1: (a) Methodological quality summary. (b) Methodological quality graph.|
Click here to view
Our primary outcomes are detection of radiological findings (sialolithiasis, stenosis with stones, sialectasis, strictures, and main duct) and the diagnostic performance of these findings (sialolithiasis and stenosis with stones).
Radiological findings outcomes
Data according to number of glands: four included studies with a total of 61 glands in MRS and 61 glands in CS recorded sialolithiasis. The subgroup analysis showed no significant difference between MRS and CS in term of sialolithiasis [RR = 0.93, 95% confidence interval (CI), 0.83–1.04, P = 0.23]. Although two studies reported stenosis with stones, the subgroup analysis showed no significant difference between the two groups in term of stenosis with stones (RR = 0.93, 95% CI, 0.81–1.08, P = 0.34). The analysis of subgroups showed no significant difference between MRS and CS in term of sialectasis (RR = 0.92, 95% CI, 0.74–1.15, P = 0.47) and in term of stricture (RR = 1, 95% CI, 0.54–1.86, P = 1). Moreover, the overall effect estimate showed no significant difference between the two groups in term of radiological findings according to glands (RR = 0.93, 95% CI, 0.86–1.01, P = 0.11). All pooled studies and subgroups were homogenous (P = 0.76); I 2 = 0% (P = 1); I 2 = 0%; except stenosis with stones, which showed some heterogeneity; therefore, the fixed-effects model was used as shown in [Figure 2]a.
|Figure 2: (a) Radiological findings according to glands. (b) Radiological findings according to patients.|
Click here to view
Two studies reported detection of the main duct. The two studies showed some heterogeneity (P = 0.1), I 2 = 63%; therefore, the random effects model was used. There is no significant difference between MRS versus CS (RR = 0.98, 95% CI, 0.9–1.06, P = 0.62) as shown in [Figure 3]a.
|Figure 3: (a) Detection of main duct according to glands. (b) Detection of main duct according to patients.|
Click here to view
Data according to number of patients: some papers reported the same outcomes but according to the number of patients. The subgroup analysis reported by two papers showed no significant difference between MRS and CS in terms of sialolithiasis (RR = 0.93, 95% CI, 0.81–1.07, P = 0.31). As two papers recorded stenosis with stones, the analysis of subgroup showed no significant difference between MRS and CS (RR = 0.75, 95% CI, 0.34–1.65, P = 0.47). Moreover, two studies reported sialectasis and stricture, and the analysis of subgroup showed no significant difference in term of either sialectasis (RR = 1.23, 95% CI, 0.53–2.86, P = 0.64) or stricture (RR = 1.05, 95% CI, 0.73–1.51, P = 0.79). The overall effect estimate showed no significant difference between the two groups in term of radiological findings according to patients (RR = 1, 95% CI, 0.84–1.18, P = 0.98). These studies were statistically heterogeneous (P = 0.07), I 2 = 47%; therefore, the random effect model was used as shown in [Figure 2]b.
Three studies reported detection of the main duct. The three studies showed no heterogeneity (P = 0.70), I 2 = 0%; therefore, the fixed-effect model was used. There is a significant difference between MRS versus CS favoring the MRS over CS (RR = 0.84, 95% CI, 0.71–1, P = 0.05), as shown in [Figure 3]b.
Sialolithiasis: three studies ,, reported sensitivity and specificity for detection of sialolithiasis according to the number of glands. The sensitivity for Becker et al. , Gadodia et al. , and Kalinowski et al.  was 0.91 (95% CI, 0.75–0.98), 0.86 (95% CI, 0.42–1.00), and 0.8 (95% CI, 0.44–0.97), AbdelMaboud Ibrahim NM  respectively. The pooled sensitivity was 0.88 (95% CI, 0.75–0.95) with no heterogeneity (P = 0.68), I 2 = 0%, whereas the specificity of these three studies was as follows: 0.97 (95% CI, 0.84–1.00), 1.00 (95% CI, 0.29–1.00), and 0.98 (95% CI, 0.92–1.00), respectively. The pooled specificity was 0.98 (95% CI, 0.93–1.00) with no heterogeneity (P = 0.82), I 2 = 0%.
Moreover, the positive likelihood ratio for these studies was 29 (95% CI, 4.2–200.26), 6.5 (95% CI, 0.48–88.76), and 52.8 (95% CI, 7.37–378.46), respectively, and the pooled LR + was 26.35 (95% CI, 7.78–89.25) with no heterogeneity (P = 0.449), I 2 = 0%. However, the negative likelihood ratio was 0.097 (95% CI, 0.03–0.29), 0.21 (95% CI, 0.05–0.95), and 0.2 (95% CI, 0.06–0.7), respectively, and the pooled LR − was 0.149 (95% CI, 0.07–0.304), with no heterogeneity (P = 0.58), I 2 = 0%.
In addition, the diagnostic odds ratio under random effect was 299.67 (95% CI, 29.48–3046.7), 30.3 (95% CI, 0.959–959.68), and 260 (95% CI, 21.12–32.2). These three studies showed no heterogeneity (P = 0.71), I 2 = 0%. The pooled diagnostic odds ratio (DOR) was 181.64 (95% CI, 39.420–837), as shown in Supplementary File No. 3.
There was a paper by Jeger et al.  which reported sialolithiasis according to the number of patients. The sensitivity and specificity were 0.77 (95% CI, (0.55–0.92) and 0.00 (95% CI, 0.00–0.84), respectively.
Stenosis with stones: two studies , reported sensitivity and specificity for detection of stenosis with stones according to the number of patients. The sensitivity for Elhalal et al.  and Varghese et al.  was 0.78 (95% CI, 0.40–0.97) and 0.69 (95% CI, 0.39–0.91), respectively. The pooled sensitivity was 0.727 (95% CI, 0.498–0.893), with no heterogeneity (P = 656), I 2 = 0%. However, the specificity was 1 (95% CI, 0.78–1.00) and 0.97 (95% CI, 0.84–1.00), respectively, and the pooled specificity was 0.979 (95% CI, 0.889–0.99), with no heterogeneity (P = 0.39), I 2 = 0%.
Moreover, the positive likelihood ratio for these studies was 24 (95% CI, 1.532–376) and 22.846 (95% CI, 3.206–162.81), and the pooled LR + was 23.229 (95% CI, 4.697–114.87), with no heterogeneity (P = 0.977), I 2 = 0%. However, the negative likelihood ratio was 0.258 (95% CI, 0.088–0.758) and 0.317 (95% CI, 0.14–0.719), respectively, and the pooled LR − was 0.294 (95% CI, 0.153–0.564), with no heterogeneity (P = 0.764), I 2 = 0%.
In addition, the diagnostic odds ratio under random effect was 93 (95% CI, 3.949–2190) and 72 (95% CI, 7.127–727.34). These two studies showed no heterogeneity (P = 898); I 2 = 0%. The pooled DOR was 78.73 (95% CI, 2.18–508.81), as shown in Supplementary File No. 4.
There was a paper by Becker et al.  which reported the stenosis with stones according to the number of glands. The sensitivity and specificity were 1 (95% CI, 0.83–1.00) and 0.98 (95% CI, 0.88–1.00), respectively.
| Discussion|| |
The common clinical indications of salivary gland imaging are pain and swelling. Imaging methods for the salivary glands include plain radiography, sialography (conventional, CT, and MRI), high-resolution ultrasonography, CT, MRI, and radionuclide scintigraphy . It is important to choose the best imaging modality regarding cost, radiation, accuracy, reliability, and patient satisfaction .
Yet, CS still is the method that provides high-spatial resolution images for examining the anatomical features of the salivary ducts and assessing their ductal abnormalities. CS indications are sialoliths, strictures, diverticula, autoimmune diseases, and sialosis . However, it is an invasive examination that may cause inconvenience and a risk of complications, such as dye retention, inability to duct cannulation, and intraprocedural or postprocedural pain . The examination is contraindicated in acute sialadenitis and in cases of former allergic reactions to iodine-containing contrast agent .
For us too, this explains the need of a noninvasive diagnostic method like MRS.
Diagnostic imaging of salivary glands has been revolutionized with the advent of cross-sectional imaging modalities like MRI, avoiding most of the complications and contraindications of CS . MRS is based on the rule of imaging stationary fluids using single-shot RARE sequence as it was first described by Lomas et al. . Since then, various investigators preformed MRS by using other sequences such fast spin echo ,,, constructive interference in steady state (CISS) , and half-Fourier acquisition single-shot turbo spin-echo (HASTE) .
In this systematic review and meta-analysis, we found that there is no significant difference in radiological findings between MRS and CS regarding detection of sialolithiasis. Gadodia et al.  may support this sentence as it showed no significant difference between MRS and CS in sialolithiasis. However, in the same paper when they combined CISS sequence with HASTE sequence, it showed that MRS was superior to CS in seven glands, because it enabled the upstream imaging of the ductal system which was not seen at CS owing to near-total obstruction by distal calculus [Figure 4].
|Figure 4: Sialolith in the proximal main duct of right submandibular gland. (a) Right lateral view of conventional sialogram showing cessation of contrast in the proximal main submandibular duct. (b) Sagittal MR sialogram of the right submandibular gland reveals the mild dilatation with beading of the intraglandular branches with the stone (arrow) in the proximal part of the right submandibular duct .|
Click here to view
Furthermore, Jäger and colleagues compared three techniques in MRS which were 3D CISS, RARE, and T1–T2 turbo fast spin echo, and they conclude that 3D CISS showed superior significance to RARE and T1–T2 turbo spin-echo for detecting the gland sialoliths. The major advantage of 3D CISS is the ability to yield oblique images with the multiplanar reconstruction procedure. Moreover, they reported a higher sensitivity of 3D CISS than other techniques; however, in one case, fibrous tissue was missed resulting in the reduction of specificity . The sensitivity and specificity of detection of calculi using deferent MR sequences by deferent investigators ranged from 80 to 100% and 80 to 94% ,,,, because the diagnosis of sialolithiasis is based on indirect signs such as areas with complete obstruction. Therefore, small nonobstructing ducts will be missed in most cases, especially the sialoliths close to orifice which are difficult to be diagnosed with MR using HASTE. However, Gadodia et al.  combined CISS and HASTE sequences and concluded with 100% sensitivity and 100% specificity for detection of sialolithiasis.
In stenosis with stones, we found that there is no significant difference in revealing stenosis with stones according to either glands or patients in the included studies. However, according to Varghese and colleagues they found that MRS had poor sensitivity of duct stenosis with stones because there was no rim of high signal intensity salvia which makes them stand out against the soft tissue background. Although when they combined the control radiographs to MRS, they revealed stenosis with stones with 100% sensitivity and specificity, because control radiographs showed the calcification but not the localization. Moreover, administration of sialagogues may lead to pooling the salvia around stenosis with stones, hence making them more obvious in MRS .
Regarding strictures and sialectasis [Figure 5], we found no difference between MRS compared with CS in either patients or glands. Gadodia and colleagues were the only ones who reported them according to glands; however, they were able to diagnose all strictures in all patients when they combined CISS with HASTE, where MRS was useful in tight strictures and was superior to CS, as the opacification of the strictures and the proximal dilated duct was difficult on CS. This combination makes detection easier because of CISS three-dimensional nature and the pathology of diseases is confirmed in multiple views. On combining the two sequences, the sensitivity and specificity were 93 and 100%, respectively . These results are in line with Varghese et al. , Elhalal et al. , and Kalinowski et al.  where MRS showed the extent of strictures and excluded the proximal stricture as tandem lesion or calculi. Owing to the better visualization of the upstream portions of salivary glands and strictures involving the distal third of the main duct, MRS was superior to CS. However, CS was superior to MRS in demonstration of the distal ducts, concretions, and changes of sialodochitis .
|Figure 5: A 59-year-old man with a history of chronic sialadenitis of the left parotid gland. (a) Lateral digital subtraction sialogram shows multiple strictures, sialectasis, and prestenotic dilatation of the Stensen duct (arrows) and also of secondary and tertiary branching intraglandular ducts (arrowheads). (b) Lateral MR sialogram shows the same abnormal findings but at lesser spatial resolution. Subtle strictures are more difficult to visualize and sialectasis is not as prominent (arrow). The enlargement of the ductal system is demonstrated up to secondary branching ducts .|
Click here to view
In addition, Varghese and colleagues showed that MRS gave full clear delineation of sialectasis compared with CS; however, CS is highly sensitive in showing sialectasis [Figure 6], but the necessity of injection of contrast materials causes exacerbation of the process of inflammation leading to increase in pain and discomfort. Thus, MRS proves to be a valuable noninvasive alternative to CS in these patients .
|Figure 6: (a–j) MR sialograms (a–e) and conventional sialograms (f–j) of the parotid glands in patients with xerostomia with or without Sjögren's syndrome: grade 0 (a and f), grade 1 (b and g), grade 2 (c and h), grade 3 (d and i), and grade 4 (e and j). Note apparent differences in sialographic features at grade 4 (e vs. j) .|
Click here to view
This is the first meta-analysis that compares between MRS with different sequences and CS. Our limitations in our study are the small number of included studies that might have played an essential role in our outcomes and our heterogeneity, and some outcomes are not measured by some papers.
Moreover, the small sample size in our studies makes a need for more studies in this field. Furthermore, we excluded non-English studies. However, recent evidence suggests that excluding non-English studies does not produce a bias to meta-analysis . Our recommendations are to increase the number of studies comparing different sequences of MRS with CS to conclude better results about that.
| Conclusion|| |
Our study reveals that MRS is a potentially diagnostic tool that can replace invasive imaging methods such as CS.
Moreover, overall CS continues to be the standard technique of imaging. However, when the insertion of cannula is not successful or there is an acute sialadenitis, an alternative way is required, which can be MRS, especially when it is performed with a combination of CISS and HASTE sequences.
Financial support and sponsorship
Conflicts of interest
There are no conflicts of interest.
| References|| |
Krishnamurthy S, Vasudeva SB, Vijayasarathy S. Salivary gland disorders: a comprehensive review. World J Stomatol 2015; 4
Burke CJ, Thomas RH, Howlett D. Imaging the major salivary glands. Br J Oral Maxillofac Surg 2011; 49
Jacob ADC, Billington PD, Strong P. Warrington UK, Bolton/UK. MR sialography in the assessment of major salivary glands: a pictorial review. ECR 2010.
Chellathurai A, Gnanasigamani S, Kumaresan S, Balasubramaniam S, Damodarasamy K, Subbiah K, et al.
MR sialography and conventional sialography in salivary gland and duct pathologies: a comparative study. J Evolution Med Dent Sci 2016; 5
Koontz NA, Kralik SF, Fritsch MH, Mosier KM. MR sialography: a pictorial review. Neurographics 2014; 4
Moher D, Liberati A, Tetzlaff J, Altman DG. Preferred reporting items for systematic reviews and meta-analysis: the PRISMA statement. PLoS Med 2009; 6
Higgins JP, Green S. (2008) Cochrane Handbook for Systematic Reviews of Interventions: Cochrane Book Series. Print ISBN:9780470699515 |Online ISBN:9780470712184 |DOI:10.1002/9780470712184
Whiting P, Rutjes AW, Dinnes J, Reitsma J, Bossuyt PM, Kleijnen J. Development and validation of methods for assessing the quality of diagnostic accuracy studies. Health Technol Assess 2004; 8
:iii, 1 234.
Bossuyt PM, Reitsma JB, Bruns DE, Gatsonis CA, Glasziou PP, Irwig LM, et al.
The STARD statement for reporting studies of diagnostic accuracy: explanation and elaboration. Clin Chem 2003; 49
Egger M, Smith GD, Schneider M, Minder C. Bias in meta analysis detected by a simple, graphical test. BMJ 2015; 14
Terrin N, Schmid CH, Lau J, Olkin I. Adjusting for publication bias in the presence of heterogeneity. Stat Med Stat Med 2003; 22
Gadodia A, Seith A, Sharma R, Thakar A, Parshad R. Magnetic resonance sialography using CISS and HASTE sequences in inflammatory salivary gland disease: comparison with digital sialography. Acta Radiol 2010; 2
Jäger L, Menauer F, Holzknecht N, Scholz V, Grevers G, Reiser M. Sialolithiasis: MR sialography of the submandibular duct–-an alternative to conventional sialography and US? Radiology 2000; 216
Varghese JC, Thornton F, Lucey BC, Walsh M, Farrell MA, Lee MJ. A prospective comparative study of MR sialography and conventional sialography of salivary duct diseases. Am J Roentgenol 1999; 173
Becker M, Marchal F, Becker CD, Dulguerov P, Geogakopoulos G, Lehmann W, Terrier F. Sialolithiasis and salivary ductal stenosis: diagnostic accuracy of MR sialography with a three-dimensional extended-phase conjugate-symmetry rapid spin-echo sequence. Radiology 2000; 217
Kalinowski M, Heverhagen JT, Rehberg E, Klose KJ, Wagner HJ. Comparative study of MR sialography and digital subtraction: sialography for benign salivary gland disorders. Am J Neuroradiol 2002; 23
Elhalal M, Khattab Y, El-Sayed E, Habib R. Role of magnetic resonance sialography in diagnosis of salivary gland diseases. Internet J Radiol 2018; 21
Murakami R, Baba Y, Nishimura R, Baba T, Matsumoto N, Yamashita Y, et al.
MR sialography using half-fourier acquisition single–shot turbo spin-echo HASTE sequences. Am J Neuroradiol 1998; 19
AbdelMaboud Ibrahim NM, El Badry A. Diagnostic accuracy of MR sialography in sialolithiasis and salivary ductal stenosis. Egypt J Radiol Nucl Med 2013; 44
Rastogi R, Bhargava S, Mallarajapatna GJ, Singh SK. Pictorial essay: salivary gland imaging. Indian J Radiol Imag 2012; 22
Afzelius P, Nielsen M, Ewertsen C, Bloch KP. Imaging of the major salivary glands. Clin Physiol Funct Imaging 2016; 36:1–10.
Marchal F, Kurt AM, Dulguerov P, Becker M, Oedman M, Lehmann W. Histopathology of submandibular glands removed for sialolithiasis. Ann Otol Rhinol Laryngol 2001; 110(Part 1)
Tonami H, Higashi K, Matoba M, Yokota H, Yamamoto I, Sugai S. A comparative study between MR sialography and salivary gland scintigraphy in the diagnosis of Sjögren's syndrome. J Comput Assist Tomogr 2001; 25
Shah GV. MR imaging of salivary glands. Neuroimag Clin N
Am 2004; 14
Lomas DJ, Carrol NR, Johnson G, Antoun NM, Freer CE. MR sialography. Work Prog Radiol 1996; 200
Fischbach R, Kugel H, Ernst S, Schröder U, Brochhagen HG, Jungehülsing M, et al.
MR sialography: initial experience using a T2W fast SE sequence. J Comput Assist Tomogr 1997; 21
Sartoretti-Schefer S, Kollias S, Wichmann W, Valavanis A. 3D T2-weighted fast spin-echo MRI sialography of the parotid gland. Neuroradiology 1999; 41
Tonami H, Ogawa Y, Matoba M, Kuginuki Y, Yokota H, Higashi K, et al.
MR sialography in patients with Sjögren syndrome. Am J Neuroradiol 1998; 19
Kalk WW, Vissink A, Spijkervet FK, Möller JM, Roodenburg JL. Morbidity from parotid sialography. Oral Surg Oral Med Oral Pathol Oral Radiol Endod 2001; 92
Jungehulsing M, Fishbach R, Schroder U, Kugel H, Damm M, Eckel HE. Magnetic resonance sialography. Otolaryngol Head Neck Surg 1999; 121
Takagi Y, Sumi M, Sumi T, Ichikawa Y, Nakamura T. MR microscopy of the parotid glands in patients with sjögren's syndrome: quantitative mr diagnostic criteria. Am J Neuroradiol 2005; 26
Moher D, Pham B, Klassen TP, Schulz KF, Berlin JA, Jadad AR, et al.
What contributions do languages other than English make on the results of meta-analyses? J Clin Epidemiol 2000; 53
[Figure 1], [Figure 2], [Figure 3], [Figure 4], [Figure 5], [Figure 6]