Diagnostic Imaging Europe
June 2002

Head and Neck

MRI and CT provide salivary gland details

MR sialography matches up with conventional imaging methods

By: Jan W. Casselman, M.D., Ph.D.

The preferred technique for imaging the salivary glands has changed dramatically over the years, although the choice is still primarily based on the findings of a physical examination.

Imaging of the major salivary glands (parotid, submandibular, and sublingual) traditionally relied on conventional radiography and sialography. The advent of ultrasound, CT, and MRI allowed better 3D evaluation of these glands. In the last three years, multislice spiral CT (MSCT) and gradient-echo T2-weighted MR sialography have expanded the options for diagnosing patients with salivary gland disease. Tomographic techniques generally re-sult in better evaluation of lesions of the minor salivary glands. These glands, which number approximately 750, are situated beneath the mucosa of the oral cavity, palate, paranasal sinuses, pharynx, larynx, trachea, and bronchi.

Causes of acute parotitis can be viral (such as mumps, the most common disease of the salivary glands), bacterial (strepto- or staphylococcus), or granulomatous (such as tuberculosis and cat-scratch fever). The condition may also result from sialolithiasis (causing sialadenitis) or an odontogenous abscess (causing salivary gland infection). These patients present with a swollen, painful gland, and imaging is not necessary to make the diagnosis. Conventional sialography is contra-indicated because it can carry the infection back deep into the gland, and the retrograde injection is extremely painful in patients with acute parotitis. Ultrasound is also painful, but it can demonstrate an enlarged gland.

Imaging is necessary only when an abscess is suspected. Although ultrasound and contrast-enhanced MRI can detect and delineate the abscess, CT has some advantages over both modalities. Unenhanced CT images can exclude a sialolith as the cause of the abscess, while contrast-enhanced CT provides better assessment of the nature and extension of inflammatory lesions.1

Sialolithiasis is the second most common disease of the salivary glands. The submandibular gland is involved in 80% to 82% of cases.2 Plain radiography, sialography, ultrasound, CT, and thin-section T2-weighted MRI can demonstrate sialoliths. CT has a 10-fold increased sensitivity over plain films in demonstrating calcifications, and it can depict stones in the ductal system and the gland (Figure 1).3 Conventional sialography is invasive and is often unable to provide information about the duct and gland behind the "obstructing" sialolith. Although ultrasound depicts most of the sialoliths, CT is superior in differentiating multiple clusters of stones from isolated large sialoliths and assessing the deep lobe of the parotid gland.4 High-resolution T2-weighted images also show stones inside the ductal system and provide information on ducts proximal to the sialolith (Figure 1D).

Information about the ductal system and gland behind the sialolith is not often needed but can sometimes help locate the cause of the painful gland. MR sialography with evoked salivation has been shown to detect submandibular duct stones with accuracy similar to digital sialography and better than ultrasound,5 especially when located deep in the gland.6

Unenhanced CT is therefore the imaging technique of choice, especially given that MSCT allows reconstruction in all planes without loss of quality. This results in multiplanar visualization of the stones and the salivary gland and/or the dilated duct proximal to the stone (Figure 1B).

Recurrent bacterial infections, granulomatous processes, and irradiation can cause chronic inflammatory disease of the major salivary gland. The most common cause is sialolithiasis, and at least one calculus is found in two thirds of cases.7 In these patients, strictures occur along the main duct (Figure 2) and dilation of the adjacent parts of the duct are pres-ent (Figure 1D and Figure 2A).

MR sialography shows stenoses and dilations, but cannot demonstrate the normal peripheral ducts and acini. Their presence indicates, at least partially, that the gland has not been completely destroyed and can be saved if the strictures are treated. Absence of the peripheral ducts and acini indicates that the acini have been destroyed and compressed, and that a (partial) gland resection should probably be performed.

Visualization or nonvisualization of the peripheral ducts and acini is one of the last remaining "relative" indications of conventional sialography, although the combination of clinical history and MR sialography findings often leads to the same diagnosis and therapy. When focal abscesses develop, nonuniform globular collections can be seen on conventional or MR sialography. Chronic sialolithiasis often results in a small atrophic gland with intraglandular calcifications (Figure 1A).8

DIFFERENTIAL DIAGNOSES

Primary (isolated) and secondary (associated with connective tissue diseases) Sjogren's syndrome occurs in patients between the ages of 40 and 60, up to 95% of whom are female. The presence of keratoconjunctivitis sicca, xerostomia, rheumatoid arthritis, lupus, or scleroderma leads to the clinical diagnosis. Serology of antinuclear antibodies and a lip or parotid gland biopsy can further establish the diagnosis.

Conventional or MR sialography are used only to stage this disease. The peripheral ducts and acini are involved in the initial stage, and punctate contrast collections can be seen on a sialogram. As the disease progresses, globular contrast accumulations appear on sialography (Figure 3). Retrograde infection also causes sialadenitis changes and dilation of the central ducts (Figures 3D and 3E), resulting in the destructive form of Sjogren's syndrome with formation of peripheral collections and eventually abscesses.

A study comparing the accuracy of conventional sialography and MR found an agreement rate of 89%,9 demonstrating that MR sialography can identify the three stages of the disease and can replace conventional sialography as a viable, nonirradiating technique. Ultrasound is negative in the first stage but later demonstrates areas of lower attenuation. Transverse CT and MRI can detect only the globular and destructive changes. The gland first enlarges and becomes denser on CT. A honeycomb glandular appearance then becomes visible in the globular stage on CT, and cystic areas are seen on MRI. Patients are 44 times more likely to develop a non-Hodgkin's lymphoma.10

Any isolated mass in a salivary gland affected by Sjogren's syndrome must be considered a lymphoma unless the contrary can be proven. The punctate or globular pattern is uniform in size and distribution, which is not the case in chronic sialadenitis (see above). This allows identification of the two entities on sialography.

Cysts and lymphoid nodules can be found in the parotid glands of HIV-positive patients. A diagnosis of AIDS-related parotid cysts should be made whenever multiple cystic lesions are found in the parotid gland. Differential diagnosis of Sjogren's disease can be difficult but is helped by the presence of cervical adenopathies and involvement of Waldeyer's ring.11

Fungal infections, TB, sarcoidosis, Wegener's disease, and cat-scratch fever may cause painless focal or gene-ralized salivary gland masses. Ultra-sound, CT, or MR images. Association with cervical adenopathy helps in making the diagnosis.3 Granulomatous disease is more likely in the presence of calcifications in a painless parotid gland mass, and pleomorphic adenoma is the most important differential diagnosis.

Patients with sialosis have a bilateral painless enlargement of the parotid glands, often due to diabetes mellitus, malnutrition, medication, alcoholism, or hypothyroidism. The enlarged glands show increased attenuation on CT and increased signal intensity on T2-weighted MR images. The ductal system is splayed by the increased gland volume; this allows differentiation of Sjogren's disease and chronic sialadenitis.3

Intermittent or incomplete ductal obstruction can result in cyst formation. Sialolithiasis, chronic sialadenitis, biopsy, trauma, and tumors can cause such an obstruction, called a ranula when it occurs in the sublingual gland. A simple ranula is situated above the mylohyoid muscle in the floor of the mouth and is a true cyst with epithelial lining. A plunging ranula develops when a simple one ruptures, often extending below the mylohyoid muscle. It has no epithelial lining and is therefore a pseudocyst. Ranulas can be seen on ultrasound, CT, and MRI, but the lesion and its extension are better identified when multiplanar imaging with good contrast between different tissues is available, as with MRI.

A tumor is suspected when an enlarging painless mass is found. The smaller the salivary gland, the higher the chance that the lesion is malignant.12,13 MRI is the imaging method of choice because it provides the best contrast between the lesion and surrounding tissue and clearly demonstrates bone invasion, meningeal or intracranial extension, and perineural spread. The latter typically follows the branches of the facial and trigeminal nerves. The T2 signal intensity also helps differentiate between malignant and benign lesions. Lesions with low signal intensity on T2-weighted images are frequently malignant (Figure 4), while high-intensity lesions are more often benign. This predictive value is 73%.14 But because the opposite can also occur, tumor characterization is most reliable when aspiration or biopsy is performed. Both ultrasound and CT can be used to guide this procedure.

Contrast media are often necessary to distinguish cystic areas from solid parts of the tumor. CT is less reliable in differentiating between malignant and benign lesions and is less suited to demonstrate the exact tumor extension.15 Because only CT can depict small calculi, the modality is used when a pseudomass, caused by sialo-lithiasis, is suspected clinically.

Ultrasound can evaluate salivary gland tumors but has several drawbacks. It is operator-dependent, unable to assess the deep lobe of the parotid gland and retropharyngeal lymph nodes, and cannot detect parapharyngeal or intracranial extension. Nor do ultrasound images show the relationship between the tumor and the course of the facial nerve. On the other hand, ultrasound is excellent for diagnosing superficial salivary gland lesions.16

Eighty percent of all benign masses are pleomorphic adenomas, which occur most frequently in middle-aged women. They have low signal intensity on T1-weighted images, high signal intensity on T2-weighted images, and enhance after contrast administration. These lesions, although benign, are removed because malignant degeneration can develop.17

The second most common benign tumors are Warthin's, which are nearly always found in the tail of the parotid gland of older men. They can have intermediate to low signal intensity on T2-weighted images, which is atypical for a benign tumor, but the patient's age and the presence of multiple parotid masses often lead to the correct diagnosis. Malignant lesions are suspected when the patient mentions a dull pain, adenopathies are palpated, facial nerve palsy is present, or invasion of the surrounding superficial tissue is seen or palpated.

The most frequently occurring malignant tumor of the parotid gland is a mucoepidermoid carcinoma. Ade-noid cystic carcinomas are the most frequently seen tumors in the submandibular gland, and they have a high frequency (50% to 60%) of perineural spread, leading to paralysis of cranial nerves.18,19 All nerve branches in the area near the tumor must therefore be checked for perineural tumor spread on MRI (Figure 5).

The parotid is the only salivary gland in which lymph nodes are found, because it is the only one that encap-sulates after the emergence of the lymphatic system.3 Metastatic lymph nodes, lymph nodes in the setting of lymphoma, and primary lymphoma can therefore occur in the parotid gland. Multiple painless masses may also be Warthin's tumors, pleomorphic adenomas, oncocytomas, acinic cell carcinomas, HIV-related cysts, and metastases.

Malignant tumors are also common in the minor salivary glands, accounting for 50% to 80% of all tumors in these glands. Adenoid cystic carcinoma is the most frequently found malignant tumor in these glands. Typical locations are the palate, paranasal sinuses/nasal cavity, infraglottic area, and trachea. Be-cause these tumors frequently follow adjacent nerves, contrast-enhanced fat-suppressed MRI is the method of choice for staging them (Figure 5).20

When patients present with a painful swelling or mass, unenhanced CT imaging can either rule out or demonstrate sialolithiasis, while contrast-enhanced CT can identify abscess formation. MRI is the method of choice when a painless mass or growing lesion is found at presentation. It is unrivaled in the evaluation of tumor extension (perineural, intracranial, and bone, for example) and is helpful in characterizing these lesions. Sialography is rarely necessary; it is used only to grade lesions in patients with Sjogren's disease and to determine whether the gland can be saved in case of severe chronic sialadenitis. Recent studies show that MR sialography is rapidly replacing conventional sialography.

DR. CASSELMAN is clinical director of MRI and head and neck radiology at A.Z. St.-Jan Bruges and A.Z. St.-Augustinus in Antwerp, Belgium.

---

References

1. Casselman JW, Mancuso AA. Major salivary gland masses: comparison of MR imaging and CT. Radiology 1987;165:183-189.
2. Levy DM, ReMine WH, Devine KD. Salivary gland calculi. JAMA 1962;181:1115-1119.
3. Som PM, Brandwein M. Salivary glands. In: Som PM, Curtin HD, eds. Head and neck imaging. 3rd ed. St. Louis: Mosby Year Book, 1996:823-914.
4. Avrahami E, Englender M, Chen E, et al. CT of submandibular gland sialolithiasis. J Neuroradiol 1996;38:287-290.
5. Jager L, Menauer F, Holzknecht N, et al. Sialolithiasis: MR sialography of the submandibular duct-an alternative to conventional sialography and US? Radiology 2000;216:665-671.
6. Varghese JC, Thornton F, Lucey BC, et al. A prospective study of MR sialography of salivary duct disease. AJR 1999;173:1497-1503.
7. Sigal R. Oral cavity, oropharynx, and salivary glands. Neuroimaging Clin North Am 1996;6:379-400.
8. Yousem DM, Kraut MA, Chalian AA. Major salivary gland imaging. Radiology 2000;216:19-29.
9. Ohbayashi N, Yamada I, Yoshino N, Sasaki T. Sjogren syndrome: comparison of assessments with MR sialography and conventional sialography. Radiology 1998;209:683-688.
10. Grevers G, Ihrler S, Vogl TJ, Weiss M. A comparison of clinical, pathological and radiological findings with magnetic resonance imaging studies of lymphomas in patients with Sjogren's syndrome. Eur Arch Otorhinolaryngol 1994;251:214-217.
11. Som PM, Brandwein MS, Silvers A. Nodal inclusion cysts of the parotid gland and parapharyngeal space: a discussion of lymphoepithelial, AIDS-related parotid and branchial cysts, cystic Warthin's tumors, and cysts in Sjogren's syndrome. Laryngoscope 1995;105:1122-1128.
12. Freling NJ, Molenaar WM, Vermey A, et al. Malignant parotid tumors: clinical use of MR imaging and histologic correlation. Radiology 1992;185:691-696.
13. Batsakis JG. Tumors of the head and neck: clinical and pathologic considerations. Baltimore: Williams & Wilkins, 1979.
14. Som PM, Biller HF. High-grade malignancies of the parotid gland: identification with MR imaging. Radiology 1989;173:823-826.
15. Berg HM, Jacobs JB, Kaufman D, Reede DL. Correlation of fine needle aspiration biopsy and CT scanning of parotid masses. Laryngoscope 1986;96:1357-1362.
16. Gritzmann N. Sonography of the salivary glands. AJR 1989;153:161-166.
17. Som PM, Shugar JM, Sacher M, et al. Benign and malignant parotid pleomorphic adenomas: CT and MR studies. J Comput Assist Tomogr 1988;12:65-69.
18. Seaver PR Jr, Kuehn PG. Adenoid cystic carcinoma of the salivary glands: a study of ninety-three cases. Am J Surg 1979;137:449-455.
19. Vogl TJ, Balzer J, Mack M, Steger W. Salivary glands. In: Vogl TH, ed. Differential diagnosis in head and neck imaging. Stuttgart: Georg Thieme Verlag, 1999:237-253.
20. Parker GD, Harnsberger HR. Clinical-radiologic issues in perineural tumor spread of malignant diseases of the extra-cranial head and neck. Radiographics 1991;11:383-399.