Rationale and objectives: Fluid-attenuated inversion recovery (FLAIR) imaging can detect elevated protein levels in fluid that may be difficult or impossible to detect with T1- or T2-weighted imaging. The purpose of this study is to evaluate the frequency and asses the diagnostic utility of abnormal FLAIR signal in various types of inner ear and internal auditory canal (IAC) pathology.
Materials and methods: A retrospective review of medical records and magnetic resonance images from 187 consecutive patients who underwent imaging of the temporal bones for possible inner ear or IAC pathology over a 1-year period were reviewed for abnormal increased FLAIR signal, increased intrinsic T1 signal, abnormal enhancement after gadolinium administration, and the presence of a mass lesion within the cerebellopontine angle, IAC, or inner ear. Reviewers were blinded to clinical diagnosis.
Results: Twenty-five of 32 (78%) patients with schwannomas restricted to the IAC and cerebellopontine angle demonstrated associated increased FLAIR signal within the ipsilateral inner ear structures. The sensitivity, specificity, positive predictive value, and negative predictive value of inner ear FLAIR hyperintensity for a schwannoma were 80%, 95%, 78%, and 95%, respectively. Inner ear FLAIR hyperintensity was also seen in cases of intracochlear hemorrhage, labyrinthitis, and labyrinthitis ossificans, but these conditions did not occur with sufficient frequency in our study population to determine statistical significance.
Conclusion: Identification of inner ear FLAIR hyperintensity can alert the radiologist to scrutinize pre- and post-gadolinium T1-weighted images and T2-weighted images for subtle IAC and inner ear abnormalities. Although enhancement after the administration of gadolinium contrast media is the gold standard for detection of schwannoma, inner ear FLAIR hyperintensity may be a helpful diagnostic adjunct for vestibular schwannoma. This may be particularly helpful when gadolinium was not administered and the T2-weighted images are equivocal, especially when dedicated thin-section imaging of the IACs and temporal bones was not performed.