HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Figures Only Full Text (PDF) Submit a response Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Jäger, L. Articles by Reiser, M. Search for Related Content PubMed PubMed Citation Articles by Jäger, L. Articles by Reiser, M.

Sialolithiasis: MR Sialography of the Submandibular Duct-An Alternative to Conventional Sialography and US?¹

¹ From the Institute of Diagnostic Radiology (L.J., N.H., V.S., M.R.) and the Department of Otorhinolaryngology (F.M., G.G.), Klinikum Grosshadern, Ludwig Maximilians Universität, Marchioninistrasse 15, 81366 Munich, Germany. Received September 28, 1999; revision requested November 1; revision received February 15, 2000; accepted February 22. Address correspondence to L.J. (e-mail: jaeger@ikra.med.uni-muenchen.de).

	ABSTRACT

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
PURPOSE: To determine the value of magnetic resonance (MR) sialography for the diagnosis of sialolithiasis by comparing results prospectively with those of ultrasonography (US) and digital sialography.

MATERIALS AND METHODS: MR sialography was prospectively performed with T2-weighted three-dimensional (3D) constructive interference in steady-state (CISS) and rapid acquisition with relaxation-enhancement (RARE) sequences in 24 patients suspected of having sialolithiasis. Evoked salivation was used as contrast material. T1-weighted spin-echo and T2-weighted turbo spin-echo MR imaging also were performed. The results were then compared with those of US and digital sialography, with the latter as standard of reference.

RESULTS: The 3D CISS images were significantly (P < .05) superior to RARE images for demonstrating the submandibular ductal system, followed by T2-weighted turbo spin-echo images (P < .01) and T1-weighted spin-echo images (P < .001). The sensitivity and specificity were 100% and 80%, respectively, for CISS MR sialography and 80% and 100%, respectively, for RARE MR imaging. The sensitivity and specificity of US were both 80%.

CONCLUSION: MR sialography with evoked salivation is noninvasive and allows delineation of the submandibular ductal system and detection of sialoliths with accuracy that is similar to that of digital sialography and superior to that of US.

Index terms: Magnetic resonance (MR), comparative studies • Salivary glands, diseases, 2642.818 • Salivary glands, MR, 2642.121411, 2642.121412 • Salivary glands, radiography, 2642.1215 • Salivary glands, US, 2642.12981 • Ultrasound (US), comparative studies

	INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
Imaging of the large salivary glands can be performed with conventional sialography, ultrasonography (US), computed tomography (CT), magnetic resonance (MR) imaging, digital sialography, and digital subtraction sialography. However, a reliable technique that generates reproducible results similar to those of conventional (x-ray) sialography has yet to emerge.

The submandibular duct (Wharton duct) extends from the submandibular gland to the posterior edge of the mylohyoid muscle, curves around the muscle, then enters the sublingual space on the surface of the mylohyoid muscle and finally drains into the sublingual papilla. The submandibular duct is approximately 5–6 cm long and has an intraluminal cross-sectional diameter of about 1–3 mm on conventional sialographic images (1–3). The salivary ducts of the third and fourth orders have an intraluminal cross-sectional diameter of 0.4–0.2 mm (2,3).

Sialolithiasis is a disease of adulthood. The submandibular gland is affected in 85% of patients (2). Intraductal sialoliths may cause total obstruction of the ectatic duct and result in painful swelling of the submandibular gland because of secretion and stasis of saliva. With regard to the diagnosis of sialolithiasis, conventional radiography has various limitations. Approximately 10%–20% of sialoliths in the submandibular gland or duct are not radiopaque and, therefore, are not visible on radiographs. In addition, phlebolithiasis and hemangiomas with calcifications or calcified lymph nodes may mimic sialoliths on radiographs. Although US has proven value with regard to the diagnosis of sialolithiasis, sialoliths smaller than 2–3 mm may be overlooked because an acoustic shadow may be absent (4,5). Digital sialography and digital subtraction sialography are the favored techniques for help in the detection of sialolithiasis of the submandibular duct. Even nonradiopaque sialoliths are detectable with this technique. Another advantage of digital sialography and digital subtraction sialography is the ability to make small adjustments in the positioning of the patient’s head during imaging, thus ensuring more accurate positioning and collimation. Motion artifacts, however, are a severe problem with digital subtraction sialographic images.

Sialography has inherent disadvantages. First, it is an invasive method that is associated with complications such as bleeding, traumatic perforation or rupture of the submandibular duct, and side effects from contrast material (6). Second, sialography is contraindicated in patients with an active infection, because the infection could be disseminated into the salivary gland during the procedure.

Because sialolithiasis is often accompanied by sialadenitis, a new sialographic technique is needed that would make it possible to image the salivary ductal system without the risk of bacterial transmission. MR sialography may overcome the limitations of current sialographic techniques. Therefore, in our prospective blinded study, the results of high-spatial-resolution MR sialography with evoked salivation to provide a natural contrast material are compared with those of digital sialography and US to assess the diagnostic accuracy of these methods.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
For this prospective blinded study, informed consent was obtained from 24 consecutive patients who were clinically suspected of having of sialolithiasis of the submandibular duct. These patients were referred to our department for conventional sialography. The study was performed in accordance with the guidelines of the Helsinki II declaration.

All 24 patients underwent US with a 7.5-MHz phased-array sector transducer (Elegra 5000; Siemens Medical Systems, Erlangen, Germany). The submandibular duct and gland were examined on both sides in longitudinal and transverse planes, with the probe in a submental position. In the submandibular duct or in the intraglandular ductal system, small echogenic lesions with an acoustic shadow were regarded as indicative of an intraductal sialolith if the lesion with the acoustic shadow was reproducible in longitudinal and transverse planes.

Digital sialography was performed in 20 patients by using a digital x-ray system with a C arm (Polystar; Siemens Medical Systems). In four patients, cannulation of the submandibular duct was not successful; thus, digital sialography could not be performed in these patients. After cannulation of the submandibular duct with a catheter, 1.5 mL of nonionic water-soluble contrast material (iotrolan, Isovist; Schering, Berlin, Germany) was injected. Posteroanterior and lateral digital radiographs were acquired. Sialoliths in the submandibular duct and in the intraglandular ductal system were regarded as intraductal, and sialoliths in the submandibular gland but with an extraductal location were regarded as parenchymal.

If air bubbles are inadvertently injected with the contrast material, the bubbles may mimic sialolithiasis at sialography. In MR imaging, air bubbles within the duct manifest as signal voids that cannot be differentiated from sialoliths. Therefore, MR imaging was performed at least 4 hours after contrast-enhanced digital sialography to allow normal salivation to wash away air bubbles. MR imaging was performed in 20 of 24 patients (four patients in whom digital sialography was not performed were excluded) by using a 1.5-T whole-body imager (Vision; Siemens Medical Systems) and a circularly polarized head coil. For the purpose of this study, the results of digital sialography were used as the standard of reference for the evaluation of MR sialographic results. The MR system was equipped with a 600-V gradient amplifier that provided a gradient amplitude of 25 mT/m at a ramp time of 600 µsec.

Prior to MR imaging, the patients were given lemon juice to stimulate salivation. MR sialography was performed by using using a rapid acquisition with relaxation enhancement (RARE) sequence (7) and a three-dimensional (3D) constructive interference in steady-state (CISS) sequence (8). The RARE sequence was performed in a plane with an oblique sagittal orientation parallel to the submandibular duct. Transverse images were then obtained with the 3D CISS sequence, and coronal and oblique sagittal images were reformatted by using a multiplanar reconstruction procedure.

The RARE sequence (2,000/10.7 [repetition time msec/echo time msec], 150° flip angle, 240 x 256 matrix, 190-mm field of view, 0.79 x 0.74-mm in-plane resolution, 40-mm section thickness, 7-second acquisition time) generated a train of 240 spin echoes with 180° refocusing pulses after a single-spin excitation with a selective 90° radio-frequency pulse. Heavy T2 weighting was achieved by placing the phase-encoding zero point in the center of the 240 echoes. This resulted in an effective echo time of 1,200 msec and strong T2 contrast. Therefore, soft tissues with a short or medium T2 exhibited very low signal intensity.

The 3D CISS sequence is a gradient-echo sequence consisting of two true fast imaging with steady-state precession, or FISP, sequences (8). All gradients were focused in true FISP sequences (9). Because the signals from both FISP sequences are summed, fluids such as saliva have high signal intensity. Because the 3D CISS sequence is susceptible to patient motion, the acquisition time should be short. Even with a very short repetition time and the application of alternating and nonalternating radio-frequency pulses, there still remain slight frequency offsets in areas with strong susceptibility changes and at the borders of the 3D slab. When using the 3D CISS sequence, various factors must be balanced to maintain image quality. In other words, it is important to prevent motion artifacts by reducing the acquisition time while maintaining image quality by using a high signal-to-noise ratio and in-plane resolution and reducing off-resonance effects. In our study, therefore, a 3D CISS sequence (12.57/5.9, 70° flip angle, 256 x 512 matrix, 160 x 320-mm field of view, 0.63 x 0.63-mm in-plane resolution, 1.25-mm section thickness, 4-minute 12-second acquisition time) with a high in-plane resolution and a short acquisition time was used. The 3D CISS data were reformatted with a 3D multiplanar reconstruction procedure to generate oblique sagittal images parallel to the submandibular duct and oblique coronal images perpendicular to the submandibular duct.

Soft-tissue MR imaging was performed in the transverse plane with a T2-weighted turbo spin-echo sequence (4,500/99 [effective echo time], 242 x 256 matrix, 170-mm field of view, 0.70 x 0.66-mm in-plane resolution, 3-mm section thickness, 0.45-mm intersection gap, 3-minute 58-second acquisition time) and a T1-weighted spin-echo sequence (580/15, 256 x 256 matrix, 170-mm field of view, 0.66 x 0.66-mm in-plane resolution, 3-mm section thickness, 0.45-mm intersection gap, 5-minute 59-second acquisition time).

Three radiologists (L.J., N.H., V.S.) experienced in head and neck radiology evaluated the MR images independently and were blinded to the results of digital sialography and US. A five-point anatomic quality rating was performed for demonstration of the submandibular duct and ductal branches in the submandibular gland: score of 1, definitely not present; score of 2, probably not present; score of 3, unsure; score of 4, probably present; and score of 5, definitely present. This rating system was also used for recording whether the presence of sialoliths and fibrous stenosis, as depicted on images obtained with the four MR techniques, was incongruous. To evaluate the concordance of the image readings, the number of sialoliths were recorded, with a score of 1 for one sialolith, 2 for two sialoliths, 3 for three sialoliths, 4 for four sialoliths, and 5 for more than four sialoliths. Location of sialoliths and fibrous stenosis was also rated: score of 1, extraductal (for sialoliths only); score of 2, submandibular duct; score of 3, major duct of the submandibular gland; score of 4, intraglandular first-order branches; and score of 5, intraglandular second-order branches.

The statistic was used to evaluate the concordance among interpretations by the three image readers. There was significant concordance (P < .001) for the detectability of the submandibular ductal system and the presence and location of sialoliths on images obtained with all MR sequences. Because individual readings showed little divergence, the individual ratings for detectability of the submandibular duct and of the ductal branches in the submandibular gland, as well as for the presence, number, and location of sialoliths and the presence and location of fibrous stenosis, were averaged for the three readers and rounded to the next whole number. These averaged results for MR images were used in a further evaluation to compare MR, digital sialographic, and US images for the presence, number, and location of sialoliths and the presence and location of fibrous stenosis. Thus, a score of 1–3 was indicative of the absence of the submandibular duct, ductal branches of the submandibular gland, sialoliths, and fibrous stenosis, and a score of 4 or 5 was indicative of their presence. In addition, demonstration of the submandibular duct and of the intraglandular ductal branches was evaluated on a patient-by-patient basis by applying the Wilcoxon signed rank test, with a P value of less than .05 indicative of a significant difference. Sensitivity and specificity were calculated on a lesion-by-lesion basis for number and location of the sialoliths.

A retrospective consensus review of the MR images was performed by the readers and by the ear, nose, and throat specialist (F.M.) who performed US and/or surgery or lithotripsy, to be sure of determining the same findings. Final diagnoses were based on findings at digital sialography (20 patients), surgery (eight patients), and lithotripsy (12 patients), as well as all available clinical data.

	RESULTS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
Sialoliths were observed at digital sialography in 15 (75%) of 20 patients. Two sialoliths were detected in four of these 15 patients. In five (25%) of 20 patients, a stenosis of the submandibular duct was found after recurrent sialolithiasis with inflammation or after slitting of the submandibular duct.

On 3D CISS MR images, saliva manifested with very high signal intensity, whereas the normal submandibular gland had low signal intensity. Therefore, the intraglandular salivary ductal system and the submandibular duct were depicted as hyperintense ramifications on the transverse images and the coronal and sagittal reconstruction images. First- and second-order intraglandular branches were clearly depicted in all cases (Table 1) (Fig 1). Sialoliths were depicted as round or oval hypointense structures surrounded by hyperintense saliva (Fig 1). In cases of obstruction of the submandibular duct or the intraglandular ductal system, dilatation of the salivary duct proximal to the sialoliths was found. In one patient (Fig 2), a fibrous stenosis of the submandibular duct was interpreted as a sialolith. With T2-weighted 3D CISS images, the specificity for detection of sialoliths in the submandibular duct was 80%, and the sensitivity was 100% (Table 2). Stenosis of the submandibular duct was detected in four of five patients.

View larger version (104K): [in this window] [in a new window] [Download PPT slide]   Figure 1a. Sialolithiasis and recurrent painful swelling of left submandibular gland in a 41-year-old man. (a) Nonenhanced digital sialogram (lateral projection) shows a sialolith (large arrow) with a tail-like posterior edge (small arrow). (b) Digital sialogram (lateral projection) obtained during injection of contrast material. Because the sialolith (white arrow) occludes the submandibular duct, flow of the contrast material is stopped (black arrow). (c) Transverse T2-weighted 3D CISS MR image (12.57/5.9) shows the intraductal sialolith (straight arrow) as a hypointense structure in the left submandibular duct. The saliva-filled ectatic submandibular ductal system (curved arrow) is hyperintense. (d-f) Multiplanar reconstructions from T2-weighted 3D CISS MR images (12.57/5.9). (d) Sagittal reconstruction shows tail-like posterior edge (curved arrow) of the sialolith (long straight arrow). Proximal ectatic submandibular duct (short straight solid arrow) can be differentiated from the distal submandibular duct (open arrows), which has a normal intraluminal diameter. (e) Oblique sagittal reconstruction shows the sialolith (large straight solid arrow) obstructing the submandibular duct close to the gland. The major duct of the gland (open arrow) is demonstrated, as are first-order (small straight solid arrow) and second-order (curved arrow) branches. (f) Oblique sagittal reconstruction of contralateral submandibular gland shows the major duct of the gland (open arrow) and first-order (straight solid arrows) and second-order (curved arrow) branches. (g) Sagittal T2-weighted RARE MR image (2,000/10.7) shows a hypointense sialolith (long straight arrow). The major duct of the submandibular gland (curved arrow) and first-order (arrowhead) and second-order (short straight arrow) branches are hyperintense. (h) Transverse T2-weighted turbo spin-echo MR image (4,500/99) shows course of the submandibular duct (curved arrows), which is hyperintense on both sides. The sialolith (straight white arrow) in the left submandibular duct is hypointense. The major duct of the submandibular gland (open arrow) and first-order branches (short black arrow) are hyperintense. Because the sialolith obstructs the left submandibular duct, there is stasis of saliva and an increase in signal intensity (long black arrow). (i) Transverse T1-weighted spin-echo MR image (580/15) does not clearly show the submandibular duct, but the sialolith is depicted as hypointense (thick straight arrow). Intraglandular branches of the gland cannot be visualized. Because of the stasis of saliva, the signal intensity of the obstructed left submandibular gland (thin straight arrow) is lower than that of the contralateral duct (curved arrow).

View larger version (98K): [in this window] [in a new window] [Download PPT slide]   Figure 1b. Sialolithiasis and recurrent painful swelling of left submandibular gland in a 41-year-old man. (a) Nonenhanced digital sialogram (lateral projection) shows a sialolith (large arrow) with a tail-like posterior edge (small arrow). (b) Digital sialogram (lateral projection) obtained during injection of contrast material. Because the sialolith (white arrow) occludes the submandibular duct, flow of the contrast material is stopped (black arrow). (c) Transverse T2-weighted 3D CISS MR image (12.57/5.9) shows the intraductal sialolith (straight arrow) as a hypointense structure in the left submandibular duct. The saliva-filled ectatic submandibular ductal system (curved arrow) is hyperintense. (d-f) Multiplanar reconstructions from T2-weighted 3D CISS MR images (12.57/5.9). (d) Sagittal reconstruction shows tail-like posterior edge (curved arrow) of the sialolith (long straight arrow). Proximal ectatic submandibular duct (short straight solid arrow) can be differentiated from the distal submandibular duct (open arrows), which has a normal intraluminal diameter. (e) Oblique sagittal reconstruction shows the sialolith (large straight solid arrow) obstructing the submandibular duct close to the gland. The major duct of the gland (open arrow) is demonstrated, as are first-order (small straight solid arrow) and second-order (curved arrow) branches. (f) Oblique sagittal reconstruction of contralateral submandibular gland shows the major duct of the gland (open arrow) and first-order (straight solid arrows) and second-order (curved arrow) branches. (g) Sagittal T2-weighted RARE MR image (2,000/10.7) shows a hypointense sialolith (long straight arrow). The major duct of the submandibular gland (curved arrow) and first-order (arrowhead) and second-order (short straight arrow) branches are hyperintense. (h) Transverse T2-weighted turbo spin-echo MR image (4,500/99) shows course of the submandibular duct (curved arrows), which is hyperintense on both sides. The sialolith (straight white arrow) in the left submandibular duct is hypointense. The major duct of the submandibular gland (open arrow) and first-order branches (short black arrow) are hyperintense. Because the sialolith obstructs the left submandibular duct, there is stasis of saliva and an increase in signal intensity (long black arrow). (i) Transverse T1-weighted spin-echo MR image (580/15) does not clearly show the submandibular duct, but the sialolith is depicted as hypointense (thick straight arrow). Intraglandular branches of the gland cannot be visualized. Because of the stasis of saliva, the signal intensity of the obstructed left submandibular gland (thin straight arrow) is lower than that of the contralateral duct (curved arrow).

View larger version (125K): [in this window] [in a new window] [Download PPT slide]   Figure 1c. Sialolithiasis and recurrent painful swelling of left submandibular gland in a 41-year-old man. (a) Nonenhanced digital sialogram (lateral projection) shows a sialolith (large arrow) with a tail-like posterior edge (small arrow). (b) Digital sialogram (lateral projection) obtained during injection of contrast material. Because the sialolith (white arrow) occludes the submandibular duct, flow of the contrast material is stopped (black arrow). (c) Transverse T2-weighted 3D CISS MR image (12.57/5.9) shows the intraductal sialolith (straight arrow) as a hypointense structure in the left submandibular duct. The saliva-filled ectatic submandibular ductal system (curved arrow) is hyperintense. (d-f) Multiplanar reconstructions from T2-weighted 3D CISS MR images (12.57/5.9). (d) Sagittal reconstruction shows tail-like posterior edge (curved arrow) of the sialolith (long straight arrow). Proximal ectatic submandibular duct (short straight solid arrow) can be differentiated from the distal submandibular duct (open arrows), which has a normal intraluminal diameter. (e) Oblique sagittal reconstruction shows the sialolith (large straight solid arrow) obstructing the submandibular duct close to the gland. The major duct of the gland (open arrow) is demonstrated, as are first-order (small straight solid arrow) and second-order (curved arrow) branches. (f) Oblique sagittal reconstruction of contralateral submandibular gland shows the major duct of the gland (open arrow) and first-order (straight solid arrows) and second-order (curved arrow) branches. (g) Sagittal T2-weighted RARE MR image (2,000/10.7) shows a hypointense sialolith (long straight arrow). The major duct of the submandibular gland (curved arrow) and first-order (arrowhead) and second-order (short straight arrow) branches are hyperintense. (h) Transverse T2-weighted turbo spin-echo MR image (4,500/99) shows course of the submandibular duct (curved arrows), which is hyperintense on both sides. The sialolith (straight white arrow) in the left submandibular duct is hypointense. The major duct of the submandibular gland (open arrow) and first-order branches (short black arrow) are hyperintense. Because the sialolith obstructs the left submandibular duct, there is stasis of saliva and an increase in signal intensity (long black arrow). (i) Transverse T1-weighted spin-echo MR image (580/15) does not clearly show the submandibular duct, but the sialolith is depicted as hypointense (thick straight arrow). Intraglandular branches of the gland cannot be visualized. Because of the stasis of saliva, the signal intensity of the obstructed left submandibular gland (thin straight arrow) is lower than that of the contralateral duct (curved arrow).

View larger version (111K): [in this window] [in a new window] [Download PPT slide]   Figure 1d. Sialolithiasis and recurrent painful swelling of left submandibular gland in a 41-year-old man. (a) Nonenhanced digital sialogram (lateral projection) shows a sialolith (large arrow) with a tail-like posterior edge (small arrow). (b) Digital sialogram (lateral projection) obtained during injection of contrast material. Because the sialolith (white arrow) occludes the submandibular duct, flow of the contrast material is stopped (black arrow). (c) Transverse T2-weighted 3D CISS MR image (12.57/5.9) shows the intraductal sialolith (straight arrow) as a hypointense structure in the left submandibular duct. The saliva-filled ectatic submandibular ductal system (curved arrow) is hyperintense. (d-f) Multiplanar reconstructions from T2-weighted 3D CISS MR images (12.57/5.9). (d) Sagittal reconstruction shows tail-like posterior edge (curved arrow) of the sialolith (long straight arrow). Proximal ectatic submandibular duct (short straight solid arrow) can be differentiated from the distal submandibular duct (open arrows), which has a normal intraluminal diameter. (e) Oblique sagittal reconstruction shows the sialolith (large straight solid arrow) obstructing the submandibular duct close to the gland. The major duct of the gland (open arrow) is demonstrated, as are first-order (small straight solid arrow) and second-order (curved arrow) branches. (f) Oblique sagittal reconstruction of contralateral submandibular gland shows the major duct of the gland (open arrow) and first-order (straight solid arrows) and second-order (curved arrow) branches. (g) Sagittal T2-weighted RARE MR image (2,000/10.7) shows a hypointense sialolith (long straight arrow). The major duct of the submandibular gland (curved arrow) and first-order (arrowhead) and second-order (short straight arrow) branches are hyperintense. (h) Transverse T2-weighted turbo spin-echo MR image (4,500/99) shows course of the submandibular duct (curved arrows), which is hyperintense on both sides. The sialolith (straight white arrow) in the left submandibular duct is hypointense. The major duct of the submandibular gland (open arrow) and first-order branches (short black arrow) are hyperintense. Because the sialolith obstructs the left submandibular duct, there is stasis of saliva and an increase in signal intensity (long black arrow). (i) Transverse T1-weighted spin-echo MR image (580/15) does not clearly show the submandibular duct, but the sialolith is depicted as hypointense (thick straight arrow). Intraglandular branches of the gland cannot be visualized. Because of the stasis of saliva, the signal intensity of the obstructed left submandibular gland (thin straight arrow) is lower than that of the contralateral duct (curved arrow).

View larger version (123K): [in this window] [in a new window] [Download PPT slide]   Figure 1e. Sialolithiasis and recurrent painful swelling of left submandibular gland in a 41-year-old man. (a) Nonenhanced digital sialogram (lateral projection) shows a sialolith (large arrow) with a tail-like posterior edge (small arrow). (b) Digital sialogram (lateral projection) obtained during injection of contrast material. Because the sialolith (white arrow) occludes the submandibular duct, flow of the contrast material is stopped (black arrow). (c) Transverse T2-weighted 3D CISS MR image (12.57/5.9) shows the intraductal sialolith (straight arrow) as a hypointense structure in the left submandibular duct. The saliva-filled ectatic submandibular ductal system (curved arrow) is hyperintense. (d-f) Multiplanar reconstructions from T2-weighted 3D CISS MR images (12.57/5.9). (d) Sagittal reconstruction shows tail-like posterior edge (curved arrow) of the sialolith (long straight arrow). Proximal ectatic submandibular duct (short straight solid arrow) can be differentiated from the distal submandibular duct (open arrows), which has a normal intraluminal diameter. (e) Oblique sagittal reconstruction shows the sialolith (large straight solid arrow) obstructing the submandibular duct close to the gland. The major duct of the gland (open arrow) is demonstrated, as are first-order (small straight solid arrow) and second-order (curved arrow) branches. (f) Oblique sagittal reconstruction of contralateral submandibular gland shows the major duct of the gland (open arrow) and first-order (straight solid arrows) and second-order (curved arrow) branches. (g) Sagittal T2-weighted RARE MR image (2,000/10.7) shows a hypointense sialolith (long straight arrow). The major duct of the submandibular gland (curved arrow) and first-order (arrowhead) and second-order (short straight arrow) branches are hyperintense. (h) Transverse T2-weighted turbo spin-echo MR image (4,500/99) shows course of the submandibular duct (curved arrows), which is hyperintense on both sides. The sialolith (straight white arrow) in the left submandibular duct is hypointense. The major duct of the submandibular gland (open arrow) and first-order branches (short black arrow) are hyperintense. Because the sialolith obstructs the left submandibular duct, there is stasis of saliva and an increase in signal intensity (long black arrow). (i) Transverse T1-weighted spin-echo MR image (580/15) does not clearly show the submandibular duct, but the sialolith is depicted as hypointense (thick straight arrow). Intraglandular branches of the gland cannot be visualized. Because of the stasis of saliva, the signal intensity of the obstructed left submandibular gland (thin straight arrow) is lower than that of the contralateral duct (curved arrow).

View larger version (127K): [in this window] [in a new window] [Download PPT slide]   Figure 1f. Sialolithiasis and recurrent painful swelling of left submandibular gland in a 41-year-old man. (a) Nonenhanced digital sialogram (lateral projection) shows a sialolith (large arrow) with a tail-like posterior edge (small arrow). (b) Digital sialogram (lateral projection) obtained during injection of contrast material. Because the sialolith (white arrow) occludes the submandibular duct, flow of the contrast material is stopped (black arrow). (c) Transverse T2-weighted 3D CISS MR image (12.57/5.9) shows the intraductal sialolith (straight arrow) as a hypointense structure in the left submandibular duct. The saliva-filled ectatic submandibular ductal system (curved arrow) is hyperintense. (d-f) Multiplanar reconstructions from T2-weighted 3D CISS MR images (12.57/5.9). (d) Sagittal reconstruction shows tail-like posterior edge (curved arrow) of the sialolith (long straight arrow). Proximal ectatic submandibular duct (short straight solid arrow) can be differentiated from the distal submandibular duct (open arrows), which has a normal intraluminal diameter. (e) Oblique sagittal reconstruction shows the sialolith (large straight solid arrow) obstructing the submandibular duct close to the gland. The major duct of the gland (open arrow) is demonstrated, as are first-order (small straight solid arrow) and second-order (curved arrow) branches. (f) Oblique sagittal reconstruction of contralateral submandibular gland shows the major duct of the gland (open arrow) and first-order (straight solid arrows) and second-order (curved arrow) branches. (g) Sagittal T2-weighted RARE MR image (2,000/10.7) shows a hypointense sialolith (long straight arrow). The major duct of the submandibular gland (curved arrow) and first-order (arrowhead) and second-order (short straight arrow) branches are hyperintense. (h) Transverse T2-weighted turbo spin-echo MR image (4,500/99) shows course of the submandibular duct (curved arrows), which is hyperintense on both sides. The sialolith (straight white arrow) in the left submandibular duct is hypointense. The major duct of the submandibular gland (open arrow) and first-order branches (short black arrow) are hyperintense. Because the sialolith obstructs the left submandibular duct, there is stasis of saliva and an increase in signal intensity (long black arrow). (i) Transverse T1-weighted spin-echo MR image (580/15) does not clearly show the submandibular duct, but the sialolith is depicted as hypointense (thick straight arrow). Intraglandular branches of the gland cannot be visualized. Because of the stasis of saliva, the signal intensity of the obstructed left submandibular gland (thin straight arrow) is lower than that of the contralateral duct (curved arrow).

View larger version (138K): [in this window] [in a new window] [Download PPT slide]   Figure 1g. Sialolithiasis and recurrent painful swelling of left submandibular gland in a 41-year-old man. (a) Nonenhanced digital sialogram (lateral projection) shows a sialolith (large arrow) with a tail-like posterior edge (small arrow). (b) Digital sialogram (lateral projection) obtained during injection of contrast material. Because the sialolith (white arrow) occludes the submandibular duct, flow of the contrast material is stopped (black arrow). (c) Transverse T2-weighted 3D CISS MR image (12.57/5.9) shows the intraductal sialolith (straight arrow) as a hypointense structure in the left submandibular duct. The saliva-filled ectatic submandibular ductal system (curved arrow) is hyperintense. (d-f) Multiplanar reconstructions from T2-weighted 3D CISS MR images (12.57/5.9). (d) Sagittal reconstruction shows tail-like posterior edge (curved arrow) of the sialolith (long straight arrow). Proximal ectatic submandibular duct (short straight solid arrow) can be differentiated from the distal submandibular duct (open arrows), which has a normal intraluminal diameter. (e) Oblique sagittal reconstruction shows the sialolith (large straight solid arrow) obstructing the submandibular duct close to the gland. The major duct of the gland (open arrow) is demonstrated, as are first-order (small straight solid arrow) and second-order (curved arrow) branches. (f) Oblique sagittal reconstruction of contralateral submandibular gland shows the major duct of the gland (open arrow) and first-order (straight solid arrows) and second-order (curved arrow) branches. (g) Sagittal T2-weighted RARE MR image (2,000/10.7) shows a hypointense sialolith (long straight arrow). The major duct of the submandibular gland (curved arrow) and first-order (arrowhead) and second-order (short straight arrow) branches are hyperintense. (h) Transverse T2-weighted turbo spin-echo MR image (4,500/99) shows course of the submandibular duct (curved arrows), which is hyperintense on both sides. The sialolith (straight white arrow) in the left submandibular duct is hypointense. The major duct of the submandibular gland (open arrow) and first-order branches (short black arrow) are hyperintense. Because the sialolith obstructs the left submandibular duct, there is stasis of saliva and an increase in signal intensity (long black arrow). (i) Transverse T1-weighted spin-echo MR image (580/15) does not clearly show the submandibular duct, but the sialolith is depicted as hypointense (thick straight arrow). Intraglandular branches of the gland cannot be visualized. Because of the stasis of saliva, the signal intensity of the obstructed left submandibular gland (thin straight arrow) is lower than that of the contralateral duct (curved arrow).

View larger version (168K): [in this window] [in a new window] [Download PPT slide]   Figure 1h. Sialolithiasis and recurrent painful swelling of left submandibular gland in a 41-year-old man. (a) Nonenhanced digital sialogram (lateral projection) shows a sialolith (large arrow) with a tail-like posterior edge (small arrow). (b) Digital sialogram (lateral projection) obtained during injection of contrast material. Because the sialolith (white arrow) occludes the submandibular duct, flow of the contrast material is stopped (black arrow). (c) Transverse T2-weighted 3D CISS MR image (12.57/5.9) shows the intraductal sialolith (straight arrow) as a hypointense structure in the left submandibular duct. The saliva-filled ectatic submandibular ductal system (curved arrow) is hyperintense. (d-f) Multiplanar reconstructions from T2-weighted 3D CISS MR images (12.57/5.9). (d) Sagittal reconstruction shows tail-like posterior edge (curved arrow) of the sialolith (long straight arrow). Proximal ectatic submandibular duct (short straight solid arrow) can be differentiated from the distal submandibular duct (open arrows), which has a normal intraluminal diameter. (e) Oblique sagittal reconstruction shows the sialolith (large straight solid arrow) obstructing the submandibular duct close to the gland. The major duct of the gland (open arrow) is demonstrated, as are first-order (small straight solid arrow) and second-order (curved arrow) branches. (f) Oblique sagittal reconstruction of contralateral submandibular gland shows the major duct of the gland (open arrow) and first-order (straight solid arrows) and second-order (curved arrow) branches. (g) Sagittal T2-weighted RARE MR image (2,000/10.7) shows a hypointense sialolith (long straight arrow). The major duct of the submandibular gland (curved arrow) and first-order (arrowhead) and second-order (short straight arrow) branches are hyperintense. (h) Transverse T2-weighted turbo spin-echo MR image (4,500/99) shows course of the submandibular duct (curved arrows), which is hyperintense on both sides. The sialolith (straight white arrow) in the left submandibular duct is hypointense. The major duct of the submandibular gland (open arrow) and first-order branches (short black arrow) are hyperintense. Because the sialolith obstructs the left submandibular duct, there is stasis of saliva and an increase in signal intensity (long black arrow). (i) Transverse T1-weighted spin-echo MR image (580/15) does not clearly show the submandibular duct, but the sialolith is depicted as hypointense (thick straight arrow). Intraglandular branches of the gland cannot be visualized. Because of the stasis of saliva, the signal intensity of the obstructed left submandibular gland (thin straight arrow) is lower than that of the contralateral duct (curved arrow).

View larger version (161K): [in this window] [in a new window] [Download PPT slide]   Figure 1i. Sialolithiasis and recurrent painful swelling of left submandibular gland in a 41-year-old man. (a) Nonenhanced digital sialogram (lateral projection) shows a sialolith (large arrow) with a tail-like posterior edge (small arrow). (b) Digital sialogram (lateral projection) obtained during injection of contrast material. Because the sialolith (white arrow) occludes the submandibular duct, flow of the contrast material is stopped (black arrow). (c) Transverse T2-weighted 3D CISS MR image (12.57/5.9) shows the intraductal sialolith (straight arrow) as a hypointense structure in the left submandibular duct. The saliva-filled ectatic submandibular ductal system (curved arrow) is hyperintense. (d-f) Multiplanar reconstructions from T2-weighted 3D CISS MR images (12.57/5.9). (d) Sagittal reconstruction shows tail-like posterior edge (curved arrow) of the sialolith (long straight arrow). Proximal ectatic submandibular duct (short straight solid arrow) can be differentiated from the distal submandibular duct (open arrows), which has a normal intraluminal diameter. (e) Oblique sagittal reconstruction shows the sialolith (large straight solid arrow) obstructing the submandibular duct close to the gland. The major duct of the gland (open arrow) is demonstrated, as are first-order (small straight solid arrow) and second-order (curved arrow) branches. (f) Oblique sagittal reconstruction of contralateral submandibular gland shows the major duct of the gland (open arrow) and first-order (straight solid arrows) and second-order (curved arrow) branches. (g) Sagittal T2-weighted RARE MR image (2,000/10.7) shows a hypointense sialolith (long straight arrow). The major duct of the submandibular gland (curved arrow) and first-order (arrowhead) and second-order (short straight arrow) branches are hyperintense. (h) Transverse T2-weighted turbo spin-echo MR image (4,500/99) shows course of the submandibular duct (curved arrows), which is hyperintense on both sides. The sialolith (straight white arrow) in the left submandibular duct is hypointense. The major duct of the submandibular gland (open arrow) and first-order branches (short black arrow) are hyperintense. Because the sialolith obstructs the left submandibular duct, there is stasis of saliva and an increase in signal intensity (long black arrow). (i) Transverse T1-weighted spin-echo MR image (580/15) does not clearly show the submandibular duct, but the sialolith is depicted as hypointense (thick straight arrow). Intraglandular branches of the gland cannot be visualized. Because of the stasis of saliva, the signal intensity of the obstructed left submandibular gland (thin straight arrow) is lower than that of the contralateral duct (curved arrow).

View larger version (117K): [in this window] [in a new window] [Download PPT slide]   Figure 2a. Suspected sialolithiasis and recurrent painful swelling of the right submandibular gland in a 28-year-old man with a history of recurrent sialolithiasis. (a) Transverse T2-weighted 3D CISS MR image (12.57/5.9, 256 x 256 matrix, 160 x 320-mm field of view, 1.25-mm section thickness) clearly shows the saliva-filled ectatic submandibular ductal system on the right side (open arrows) and the normal contralateral submandibular duct (solid arrows). (b) Oblique sagittal reconstruction from T2-weighted 3D CISS MR images (same parameters as for a) shows the proximal ectatic section (open arrow) and the distal nonectatic section (thin solid arrow) of the right submandibular duct. A signal void (thick solid arrow) between these two sections is due to fibrous tissue, not to a sialolith. This patient was seen early in the study, before we were aware of this pitfall. Because of susceptibility, the signal intensity of sialoliths is much higher than that of fibrous tissue on 3D CISS MR images. (c) Oblique sagittal T2-weighted RARE MR image (2,000/10.7, 240 x 256 matrix, 190 x 190-mm field of view, 40-mm section thickness) shows the ectatic submandibular duct (open arrows) and the fibrous stenosis (solid arrow) of the submandibular duct. Fibrous tissue does not have a round or oval shape but appears as a long, irregular, low-signal-intensity structure.

View larger version (90K): [in this window] [in a new window] [Download PPT slide]   Figure 2b. Suspected sialolithiasis and recurrent painful swelling of the right submandibular gland in a 28-year-old man with a history of recurrent sialolithiasis. (a) Transverse T2-weighted 3D CISS MR image (12.57/5.9, 256 x 256 matrix, 160 x 320-mm field of view, 1.25-mm section thickness) clearly shows the saliva-filled ectatic submandibular ductal system on the right side (open arrows) and the normal contralateral submandibular duct (solid arrows). (b) Oblique sagittal reconstruction from T2-weighted 3D CISS MR images (same parameters as for a) shows the proximal ectatic section (open arrow) and the distal nonectatic section (thin solid arrow) of the right submandibular duct. A signal void (thick solid arrow) between these two sections is due to fibrous tissue, not to a sialolith. This patient was seen early in the study, before we were aware of this pitfall. Because of susceptibility, the signal intensity of sialoliths is much higher than that of fibrous tissue on 3D CISS MR images. (c) Oblique sagittal T2-weighted RARE MR image (2,000/10.7, 240 x 256 matrix, 190 x 190-mm field of view, 40-mm section thickness) shows the ectatic submandibular duct (open arrows) and the fibrous stenosis (solid arrow) of the submandibular duct. Fibrous tissue does not have a round or oval shape but appears as a long, irregular, low-signal-intensity structure.

View larger version (109K): [in this window] [in a new window] [Download PPT slide]   Figure 2c. Suspected sialolithiasis and recurrent painful swelling of the right submandibular gland in a 28-year-old man with a history of recurrent sialolithiasis. (a) Transverse T2-weighted 3D CISS MR image (12.57/5.9, 256 x 256 matrix, 160 x 320-mm field of view, 1.25-mm section thickness) clearly shows the saliva-filled ectatic submandibular ductal system on the right side (open arrows) and the normal contralateral submandibular duct (solid arrows). (b) Oblique sagittal reconstruction from T2-weighted 3D CISS MR images (same parameters as for a) shows the proximal ectatic section (open arrow) and the distal nonectatic section (thin solid arrow) of the right submandibular duct. A signal void (thick solid arrow) between these two sections is due to fibrous tissue, not to a sialolith. This patient was seen early in the study, before we were aware of this pitfall. Because of susceptibility, the signal intensity of sialoliths is much higher than that of fibrous tissue on 3D CISS MR images. (c) Oblique sagittal T2-weighted RARE MR image (2,000/10.7, 240 x 256 matrix, 190 x 190-mm field of view, 40-mm section thickness) shows the ectatic submandibular duct (open arrows) and the fibrous stenosis (solid arrow) of the submandibular duct. Fibrous tissue does not have a round or oval shape but appears as a long, irregular, low-signal-intensity structure.

On T2-weighted turbo spin-echo MR images, saliva in the salivary ductal system was hyperintense, and sialoliths were hypointense. On transverse turbo spin-echo images, the submandibular duct was depicted in all patients (Fig 1), and first-order intraglandular branches were delineated in 10 of 20 patients (Fig 1, Table 1). Adequate demonstration of second-order branches was achieved with this sequence in only one patient. RARE images were significantly superior (P < .01) to turbo spin-echo images for delineation of intraglandular branches (Table 1). One sialolith with a diameter of 1.5 mm was overlooked on turbo spin-echo images. With regard to detection of sialoliths, T2-weighted turbo spin-echo MR imaging had a sensitivity of 93% and a specificity of 100%. Stenosis of the submandibular duct was diagnosed in two of five patients.

On transverse T1-weighted spin-echo MR images, the submandibular duct manifested as a hypointense band in three patients, and the first- and second-order intraglandular branches could not be detected. T2-weighted turbo spin-echo images were significantly better (P < .001) than T1-weighted spin-echo images for delineation of the anatomy of the submandibular duct and of the intraglandular ductal system (Table 1). Sialoliths were hypointense on T1-weighted spin-echo images (Fig 1). Sialoliths were not demonstrated on T1-weighted spin-echo images in eight of 15 patients. With regard to detection of sialoliths, T1-weighted spin-echo MR imaging had a sensitivity of 47% and a specificity of 100%. Stenosis of the submandibular duct was detected in one of five patients.

US did not demonstrate sialoliths in three patients (diameter of sialoliths, 1.5–2.0 mm), although fibrosis of the submandibular ductal wall with stenosis of the submandibular duct was interpreted as a sialolith in one patient. The sensitivity and specificity of US were both 80% (Table 2). Stenosis of the submandibular duct was correctly diagnosed in four of five patients.

	DISCUSSION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
Conventional (x-ray) and digital sialography allow delineation of the submandibular duct and of the first- through fourth-order side branches. The sensitivity of conventional sialography for detection of sialoliths varies between 64% and 100%, and the specificity varies between 88% and 100% (4,10,11). However, the development of digital sialography and digital subtraction sialography have increased the sensitivity to 96%–100%, although the specificity remains between 88% and 91.1% (11). The major advantage of digital subtraction sialography is the ability to image the submandibular ductal system without superimposition of anatomic structures such as the mandible. One disadvantage of both conventional and digital sialography, however, is the need to use contrast material. In addition, in the case of digital subtraction sialography, because images obtained with contrast material are subtracted from those obtained without contrast material (subtraction mask), involuntary head movement by the patient during the retrograde injection of contrast material may cause motion artifacts that degrade image quality and diagnostic accuracy.

Sialography requires cannulation of the submandibular duct and retrograde filling of the submandibular ductal system with contrast material. When intraductal sialoliths are close to the sublingual papilla, edema of the surrounding mucosa or of the wall of the submandibular duct often results in a narrowing of the duct; therefore, cannula insertion may not be possible. Stenosis of the submandibular duct is also found in the presence of intraductal fibrosis due to chronic sialadenitis or previous cannula insertions. Cannulation and contrast material injection into the submandibular duct may also cause the spread of bacteria from the buccal mucosa into the ductal system, with consequent infection. Moreover, if infection of the submandibular gland is already present, sialography is contraindicated.

CT sialography, either without or with 3D image processing, has been used to delineate the submandibular ductal system (5,12–14). CT sialography enables demonstration of both the soft tissues of the gland and the ductal system, which can be displayed with 3D reconstructions. Moreover, there is no superimposition of other structures such as the mandible. Second- and third-order branches, however, are not demonstrated with CT sialography. This CT technique is as invasive as conventional sialography because it also necessitates cannulation of the submandibular duct and injection of contrast material. These disadvantages and the high exposure to x rays, as compared with the exposure at conventional sialography, are reasons for the infrequent use of CT sialography. Recent developments in CT technology, such as multi–detector row CT, may allow further improvement in spatial resolution without an increase in x-ray exposure. This development and the application of intravenous contrast media, which enable the delineation of the widened submandibular ductal system as a hypoattenuating structure, could enable wider use of CT for demonstration of sialolithiasis.

US of the salivary glands is a well-established method for the examination of patients clinically suspected of having sialolithiasis. The major advantage of US is its noninvasiveness. Therefore, US can be performed in patients with acute sialadenitis. Moreover, there is no radiation exposure with US. In sialolithiasis, US has a sensitivity of 59.1%–93.7% and a specificity of 86.7%–100% (4,9,15). These wide ranges of sensitivity and specificity values reported in the literature may be due to various factors. The results of US are highly dependent on the experience and expertise of the examiner. In addition, small (3-mm-diameter) sialoliths may not be detected with US because they may not produce a dorsal acoustic shadow, depending on their chemical composition (4,5).

MR sialography is a relatively new method that has been studied three ways thus far. One technique uses conventional T1- and T2-weighted spin-echo sequences and involves some of the same preparation as for CT sialography: For these MR studies, cannulation of the parotid duct is needed, as well as injection of a gadolinium-based contrast material (12,16). Disadvantages of this technique include invasiveness and limited spatial resolution. Another MR sialographic technique consists of T2-weighted turbo spin-echo imaging in combination with normal salivation as contrast material (17–19). Disadvantages of this method include limited spatial resolution and an acquisition time that is too long to prevent swallowing artifacts. The third MR sialographic technique involves the use of heavily T2-weighted RARE imaging (20). RARE MR sialographic sequences are similar to those used in MR cholangiography (21). This technique overcomes some of the obstacles associated with the conventional and turbo spin-echo techniques. Specifically, it is noninvasive (because saliva is used as contrast material) and has a short acquisition time of 1 minute 36 seconds (which helps reduce motion artifacts). However, spatial resolution is still limited.

Until now, to our knowledge, there was no prospective evaluation to compare MR sialography, conventional sialography, and US. Therefore, it was not clear whether MR sialography was an objective and reliable alternative to conventional sialography. For this reason a systematic evaluation of these techniques was essential. It was especially important to assess the feasibility of MR sialography as a noninvasive technique with evoked salivation as a natural contrast material. Thus, heavily T2-weighted sequences were the first choice in MR sialography.

Heavily T2-weighted gradient-echo MR sequences with high spatial resolution, such as the 3D CISS sequence, were introduced successfully for imaging of the inner ear (22). The 3D CISS sequence has three major advantages: (a) The sequence produces high in-plane resolution while allowing acquisition of thin sections; (b) the sequence is sensitive to fluid-filled structures, which manifest with high signal intensity; and (c) oblique images can be calculated with multiplanar reconstruction techniques.

There are two primary disadvantages to the use of these heavily T2-weighted sequences. First, the acquisition time with such gradient-echo sequences is long relative to that with T2-weighted fast-spin-echo or RARE sequences. Second, T2-weighted gradient-echo sequences are highly sensitive to motion artifacts and to susceptibility artifacts due, for example, to the presence of prosthetic material. Even so, these sequences show promise for imaging of the intraglandular ductal system of salivary glands by allowing more precise delineation than other MR techniques and achieving resolution of the anatomy that is close to that of the digital sialography.

In our study, the MR protocol included two conventional T1- and T2-weighted MR sequences, as well as a RARE sequence and a 3D CISS sequence. To compare these sequences, in-plane resolution was varied within a range of 0.63 to 0.79 mm. Because we used saliva as a natural contrast material, intraductal filling of the subglandular ductal system was increased by stimulating salivation with lemon juice. MR sialography may, therefore, be limited in patients with reduced salivation (eg, patients with chronic or recurrent sialadenitis), because sufficient saliva is necessary in the submandibular ductal system for this method to work. In comparison with imaging of the inner ear, a shorter acquisition time for 3D CISS imaging was achieved by increasing the section thickness to 1.25 mm while maintaining the high in-plane resolution. To render the RARE sequence more resistant to motion artifacts that have been reported in the literature (20), the RARE sequence was designed as a snapshot sequence.

Our results show that 3D CISS MR imaging was significantly superior to RARE MR imaging for delineation of the ductal anatomy of the submandibular gland and that RARE imaging was superior to conventional T1- and T2-weighted spin-echo and turbo spin-echo MR imaging. Second-order intraglandular branches were reliably detected only on the 3D CISS images. Because of their small cross-sectional intraluminal diameter (0.4–0.2 mm), third- and fourth-order intraglandular branches were delineated only with digital sialography and were overlooked on 3D CISS images (spatial resolution, 0.63 x 0.63 mm). Noninvasive MR sialography could reliably delineate first- and second-order branches of the intraglandular submandibular ductal system.

Our sensitivity and specificity results with regard to detection of sialoliths with US are in agreement with those published by other groups (4,15,23). The sensitivity and specificity of RARE MR imaging were as good as or superior to those of US. However, it was not possible with RARE imaging to delineate small sialoliths because of the spatial resolution used. In addition, imaging had to be repeated several times at different oblique angles to cover the entire submandibular ductal system, because curvilinear structures exit two-dimensional projection planes.

The highest sensitivity for detection of sialoliths was found for the 3D CISS MR sequence. No sialolith was overlooked, but one case of fibrous tissue was misinterpreted as a sialolith, which resulted in a reduction in the specificity of 3D CISS imaging to 80%. The major advantage of this sequence was the ability to generate oblique images with the multiplanar reconstruction procedure. This allowed us to easily locate sialoliths in the intraductal system, which is clinically important because lithotripsy is commonly performed for intraductal sialoliths.

Of interest, the sensitivity and specificity of conventional T2-weighted turbo spin-echo imaging were superior to those of US and were as good as those of the two MR sialography sequences. Differentiation between an intraductal or extraductal position of the sialolith was not always possible on T2-weighted turbo spin-echo images, however, owing to the limited anatomic resolution possible with that sequence.

Overall, digital sialography continues to be the standard technique for imaging of the submandibular duct and the intraglandular ductal system. If acute sialadenitis is present or insertion of a cannula into the submandibular duct is not successful, however, an alternative method is needed. In such cases, noninvasive MR sialography with adequate evoked salivation as a natural contrast material provides an excellent alternative, especially when performed with a combination of RARE and 3D CISS sequences. Furthermore, because MR sialography was superior to US with regard to the detection of sialoliths, it allowed delineation of small anatomic structures and, like US, is noninvasive.

	FOOTNOTES

 
Abbreviations: CISS = constructive interference in steady state, RARE = rapid acquisition with relaxation enhancement, 3D = three-dimensional

Author contributions: Guarantor of integrity of entire study, L.J.; study concepts and design, L.J.; definition of intellectual content, L.J.; literature research, L.J.; clinical studies, L.J., F.M., N.H., V.S.; data acquisition and analysis, L.J., F.M., N.H., V.S.; statistical analysis, L.J., N.H.; manuscript preparation and editing, L.J.; manuscript review, L.J., N.H., G.G., M.R.

	REFERENCES

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 

1. Murakami R, Baba Y, Nishimura R, et al. MR Sialography using half-Fourier acquisition single-shot turbo spin-echo (HASTE) sequences. AJNR Am J Neuroradiol 1998; 19:959-961.[Abstract]
2. Pfeiffer K. Present status and value of sialography. Radiologe 1987; 27:248-254[German].[Medline]
3. Hettwer KJ, Folsom TC. The normal sialogram. Oral Surg Oral Med Oral Pathol 1968; 26:790-799.[Medline]
4. Diederich S, Wernecke K, Peters PE. Sialographic and sonographic diagnosis of diseases of the salivary gland. Radiologe 1987; 27:255-261[German].[Medline]
5. Rinast E, Gmelin E, Hollands-Thorn B. Digital subtraction sialography, conventional sialography, high-resolution ultrasonography and computed tomography in the diagnosis of salivary gland diseases. Eur J Radiol 1989; 9:224-230.[Medline]
6. Cockrell DJ, Rout PG. An adverse reaction following sialography. Dentomaxillofac Radiol 1993; 22:41-42.[Abstract]
7. Henning J, Nauerth A, Friedburg H. A fast imaging method for clinical MR. Magn Reson Med 1986; 3:823-833.[Medline]
8. Deimling M, Laub G. Constructive interference in steady state (CISS) for motion sensitivity reconstruction (abstr) In: Book of abstracts: Society of Magnetic Resonance in Medicine 1989. Berkeley, Calif: Society of Magnetic Resonance in Medicine, 1989; 842.
9. Haacke EM, Wielopolski PA, Tkach JA, et al. Steady-state free precession imaging in the presence of motion: application for improved visualization of the cerebrospinal fluid. Radiology 1990; 175:545-552.[Abstract/Free Full Text]
10. Bohndorf K, Lonnecken I, Zanella F, Lanfermann L. Value of sonography and sialography in the evaluation of salivary gland disease. Rofo Fortschr Geb Rontgenstr Nuklearmed 1987; 147:288-293[German].
11. Landwehr P, Hohmann D, Krahe T, Lackner K. Digital subtraction sialography in comparison with conventional sialography, sonography of the salivary glands and intraoperative findings. Rofo Fortschr Geb Rontgenstr Neuen Bildgeb Verfahr 1992; 156:437-442[German].[Medline]
12. Eracleous E, Kallis S, Tziakouri C, et al. Sonography, CT, CT sialography, MRI and MRI sialography in investigation of the facial nerve and the differentiation between deep and superficial parotid lesions. Neuroradiology 1997; 39:506-511.[Medline]
13. Szolar DH, Groell R, Braun H, et al. Ultrafast computed tomography and three-dimensional image processing of CT sialography in patients with parotid masses poorly defined by magnetic resonance imaging. Acta Otolaryngol 1996; 116:112-118.[Medline]
14. Szolar DH, Groell R, Preidler K, et al. Three-dimensional processing of ultrafast CT sialography for parotid masses. AJNR Am J Neuroradiol 1995; 16:1889-1893.[Abstract]
15. Gritzmann N. Sonography of the salivary glands. AJR Am J Roentgenol 1989; 153:161-166.[Abstract/Free Full Text]
16. Thibault F, Halimi P, Bely N, et al. Internal architecture of the parotid gland at MR imaging: facial nerve or ductal system?. Radiology 1993; 188:701-704.[Abstract/Free Full Text]
17. Jungehülsing M, Fischbach R, Schröder U, et al. Imaging case of the month: magnetic resonance sialography. Ann Otol Rhinol Laryngol 1998; 107:530-535.[Medline]
18. Fischbach R, Kugel H, Ernst S, et al. MR sialography: initial experience using a T2-weighted fast SE sequence. J Comput Assist Tomogr 1997; 21:826-830.[Medline]
19. Sartoretti-Schefer S, Kollias S, Wichmann W, Valavanis A. 3D T2-weighted fast spin-echo MRI sialography of the parotid gland. Neuroradiology 1999; 41:46-51.[Medline]
20. Lomas DJ, Caroll NR, Johnson G, et al. MR sialography. Radiology 1996; 200:129-133.[Abstract/Free Full Text]
21. Holzknecht N, Gauger J, Sackmann M, et al. Breath-hold MR cholangiography with snapshot techniques: prospective comparison with endoscopic retrograde cholangiography. Radiology 1998; 206:657-664.[Abstract/Free Full Text]
22. Casselman JW, Kuhweide R, Deimling M, et al. Constructive interference in steady state-3DFT MR imaging of the inner ear and cerebellopontine angle. AJNR Am J Neuroradiol 1993; 14:47-57.[Abstract]
23. Hohmann D, Landwehr P. Clinical value of sialography in digital and conventional imaging technique. HNO 1991; 39:13-17[German].[Medline]

This article has been cited by other articles:

				B. J. Trumpatori, K. Geissler, and K. G. Mathews Parotid Duct Sialolithiasis in a Dog J. Am. Anim. Hosp. Assoc., January 1, 2007; 43(1): 45 - 51. [Abstract] [Full Text] [PDF]

				E. J. Bialek, W. Jakubowski, P. Zajkowski, K. T. Szopinski, and A. Osmolski US of the major salivary glands: anatomy and spatial relationships, pathologic conditions, and pitfalls. RadioGraphics, May 1, 2006; 26(3): 745 - 763. [Abstract] [Full Text] [PDF]

				F. Marchal and P. Dulguerov Sialolithiasis Management: The State of the Art Arch Otolaryngol Head Neck Surg, September 1, 2003; 129(9): 951 - 956. [Full Text] [PDF]

This Article Abstract Figures Only Full Text (PDF) Submit a response Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services