Introduction Sjögren's syndrome (SS) is an autoimmune disorder involving the salivary and lacrimal glands. The most common clinical presentation of SS is dry eyes and dry mouth. Relevant serum autoantibodies such as anti-SS-A (Ro) or anti-SS-B (La) are often detectable. Currently, minor salivary gland biopsy is recommended to confirm a diagnosis of SS. However, biopsy is invasive and causes discomfort to the patient. Salivary gland abnormalities may be detected by imaging studies such as ultrasonography, which is noninvasive. Published studies disagree on the overall utility of ultrasonography in SS. The purpose of this study was to evaluate the utility of high-frequency ultrasonography in identifying salivary gland abnormalities in SS.
Methods In the subjects of this study, a clinical diagnosis of SS was previously established by the presence of sicca symptoms and autoantibodies such as anti-SS-A (Ro) and anti-SS-B (La). Subjects with other connective tissue diseases, such as rheumatoid arthritis or lupus, were excluded from this study. Ultrasound images of the parotid glands were obtained using an 8 to 16 MHz transducer (Diasus model UME71-8, Scotland, UK). Two rheumatologists, blinded to the subjects' identity and clinical history, interpreted the images for heterogeneity. These readers used an atlas of standardized images for comparison. The images were graded on a 4-point scale: 1 = definitely normal, 2 = probably normal, 3 = probably abnormal, or 4 = definitely abnormal. Subjects also were clinically examined by the sonographer by palpation with two fingers over the area of the parotid gland.
Results We imaged parotid glands of 20 subjects. Ten had SS (9 female and 1 male) and 10 were control subjects (10 female). The mean ages for the control group (55.6 ± 14.5 years) and subjects with SS (53.9 ± 17.5 years) were similar. The readers observed two patterns of hypoechoic areas on ultrasonography in subjects with SS: diffuse and focal. We defined focal as two or more overlapping areas in a 0.25 cm2 area of the gland. In diffuse disease, the hypoechoic areas were spread throughout the gland. The readers observed 2 diffuse and 7 focal ultrasound studies of 10 in the subjects with SS. One ultrasound study of a subject with SS had no ultrasound abnormalities. Four of 10 subjects with SS had palpable bogginess and nodularity of the parotid gland. All of these subjects were rated abnormal on ultrasonography. Also, 5 of 10 subjects with SS had a benign clinical examination but were rated abnormal on ultrasonography. On ultrasonography, 9 of 10 control subjects were rated normal by reader 1 (mean ultrasound score of 1.6), 5 of 10 control subjects were rated normal by reader 2 (mean ultrasound score of 2.5), and 9 of 10 subjects with SS were rated abnormal by both readers (mean ultrasound scores of 3.5 and 3.6). The higher mean scores for reader 2 reflect a higher threshold for calling images normal. The specificity for reader 2 was 50%, and the specificity for reader 1 was 90%. Both readers achieved a sensitivity of 90%.
Conclusion Our findings suggest that abnormalities in subjects with SS can be detected on ultrasonography, even in those with a normal physical examination. We conclude that ultrasound imaging of the salivary gland can be useful in confirming a noninvasive diagnosis of SS.