• sonography
• sialography
• Sjögren's syndrome

Recently, it has been suggested that sonographic evaluation of the salivary glands is useful in the diagnosis of Sjögren's syndrome. Kawamura et al and, more recently, Ariji et al, showed that descriptive and quantitative assessment of the salivary glands by sonography efficiently differentiated between diseased and normal glands in patients with Sjögren's syndrome.^1,2 They showed that the proposed sonographic gradings correlated well with the sialographic gradings. These findings suggest that sonography might be an alternative diagnostic tool for Sjögren's syndrome.

Here, we attempted to determine whether sonography can take the place of sialography as an alternative technique for the assessment of salivary gland involvement in Sjögren's syndrome. Sialography and sonography were performed on 294 patients who presented with sicca syndrome (171 positive and 123 negative for Sjögren's syndrome). We diagnosed patients with Sjögren's syndrome on the basis of the criteria of the European Community Study Group.³ Sonographic features characteristic of Sjögren's syndrome are heterogeneous echogenicity with hypo- and hyperechoic signals throughout the affected gland (fig 1⇓).²

Table 1⇓ shows the performance of each of the diagnostic criteria. Sialography performed best among the five diagnostic criteria—that is, sialography, functional tests (Saxon and Schirmer), and serological tests (SS-A and SS-B). Interestingly, when used instead of sialography, sonography provided a good performance, comparable with that of sialography (McNemar test, p=0.067).⁴ In contrast, the other diagnostic criteria did not perform as well as the two imaging criteria.

Logistic regression analysis was performed to identify diagnostic criteria that might be used as predictive indicators for differentiating between patients with and without Sjögren's syndrome.⁵ Univariate logistic regression analysis showed that the six diagnostic criteria assessed (sialography, sonography, Saxon's test, Schirmer test, SS-A, and SS-B) did correlate with a positive diagnosis of Sjögren's syndrome, indicating that these six criteria, if used alone, could effectively predict the presence of Sjögren's syndrome (table 1⇓).

On multivariate analysis, however, only sialography and sonography showed significant correlations with a positive diagnosis of Sjögren's syndrome (table 1⇓); when sialography was used together with the functional and serological criteria, only sialography showed a significant correlation. If sonography was used instead of sialography, only sonography displayed a significant correlation with a positive diagnosis of Sjögren's syndrome (table 1⇓). Collectively, these findings suggest that the sonography performs as well as sialography in differentiating between parotid glands affected by Sjögren's syndrome and normal glands. In contrast, the other diagnostic criteria did not perform as well as the two imaging criteria.

Some discrepancies were found between the diagnostic performance in the present study and that in previous studies. For example, Schirmer's test in our study performed poorly compared with the performance reported by Vitali et al.³ SS-A and SS-B displayed high sensitivity and low specificity in our study, whereas low sensitivity and high specificity were found in the previous study.³ These inconsistencies may be due to the differences in patient groups or in techniques, or both. Despite these differences, the performance by sialography was similar, consistent with the notion that the imaging techniques, including sialography, provide reliable results in the diagnosis of Sjögren's syndrome.

In conclusion, a diagnosis of Sjögren's syndrome can be made on the basis of a wide range of diagnostic tests, and not merely on fixed combinations of these tests. Evaluation of salivary gland involvement contributes significantly to the performance of the criteria. Thus the availability of different imaging techniques, such as Doppler sonography⁶ and magnetic resonance imaging,⁷ to assess salivary gland involvement allows clinicians to classify patients with sicca syndrome correctly.

• Download figure
• Open in new tab
• Download powerpoint

Figure 1

Sialography (A and B) and sonography (C and D) of the parotid glands in patients who presented with sicca syndrome (dry eyes and dry mouth). Normal glands (A and C), and glands affected by Sjögren's syndrome (B and D) are shown for comparison. Sialography of the parotid glands with Sjögren's syndrome shows characteristic globular (B) staining patterns. Sonography of the parotid glands with Sjögren's syndrome shows irregular echogenicity and multiple hyperechoic bands and hypoechoic areas in the gland (D).