MR T2* mapping for detecting salivary glands abnormalities in Sjögren's syndrome / 中国医学影像技术

ABSTRACT

Objective: To investigate th value of MR T2* mapping for detecting salivary glands abnormalities in patients with Sjögren's syndrome (SS). Methods: A total of 43 patients with clinically proved SS (SS group) and 40 healthy volunteers (control group) were enrolled and underwent bilateral parotid and submandibular MR examinations (including T2* mapping scanning and T2* values detection of glands). Each side of parotid and submandibular gland were evaluated according to conventional MRI (T1WI, T2WI) and MR sialography images. T2* values of parotid and submandibular were compared between SS group and control group. The diagnostic efficiency of MR morphology, T2* values and the combination of them were analyzed using Logistic regression and ROC curve. The differences of the accuracy for diagnosing same abnormalities with different diagnostic methods and the same diagnostic method in diagnosing different abnormalities were compared. Furthermore, the consistency of T2* values results detected by two observers were analyzed. Results: The average T2* value of parotid ([12.88±3.37]ms vs [10.18±1.88]ms, t=-6.40, P<0.01) and submandibular ([23.58±3.73]ms vs [21.36±1.86]ms, t=-0.49, P<0.01) in SS group were significantly higher than those in control group. The accuracy of combined diagnosis for detecting parotid and submandibular abnormalities was better than independent of MR morphology and T2* values, respectively (parotid gland Z=0.803, 4.471, both P<0.01; submandibular gland Z=8.398, 5.329, both P<0.01). There was no significant difference of diagnostic accuracy between MR morphology and T2* value alone (parotid gland Z=1.388, P=0.165; submandibular gland Z=0.553, P=0.579). Using MR morphology (Z=2.525, P=0.05), T2* value (Z=0.677, P=0.498) and combination of them (Z=0.207, P=0.835), the diagnostic accuracies were not significantly different between parotid gland lesions and submandibular gland lesions. Both intraobserver and interobserver agreements for detection of T2* values were excellent. Conclusion: T2* mapping can be used to detect parotid and submandibular abnormalities in patients with SS, and the combination of T2* values and MR morphology is helpful to improving diagnosis accuracy.