A comparison of two fat suppressed magnetic resonance imaging pulse sequences to standard T2-weighted images for brain parenchymal contrast and the identification of lesions in dogs with inflammatory intracranial disease

Abstract

T2-weighted sequences are commonly relied upon in magnetic resonance (MR) imaging protocols for the detection of brain lesions in dogs. Previously the effect of fluid suppression via fluid attenuated inversion recovery (FLAIR) has been compared to T2-weighting with mixed results. Short tau inversion recovery (STIR) has been reported to increase the detection of some CNS lesion in people. The purpose of the current study was to evaluate the effect of fat suppression on brain parenchymal contrast resolution and lesion detection in dogs. We compared three sequences: T2-weighted images (T2w), STIR, and T2-weighted fluid attenuated inversion recovery with chemical fat suppression (T2-FLAIR-FS) in dogs with meningoencephalitis. Dogs with meningoencephalitis and dogs with idiopathic epilepsy were retrospectively identified and anonymized. Evaluators recorded the presence or absence of lesions within 12 predetermined brain regions on randomized sequences, viewing and scoring each sequence individually. Additionally signal to noise ratios, contrast to noise ratios, and relative contrast were measured in a reference population. STIR sequences had the highest relative contrast between grey and white matter. While descriptively more lesions were identified by evaluators on T2-FLAIR-FS images, there was no statistical difference in the relative sensitivity of lesion detection between the sequences. Nor was there a statistical difference in false lesion detection within our reference population. STIR may be favored for enhanced anatomic contrast depiction in brain imaging. No benefit of the inclusion of a fat suppressed T2-weighted FLAIR sequence was found.