Gaussian Mixture Modeling of MRI-Derived Torus Tubarius and Inferior Turbinate Mucosal Morphometrics: Implications...

Thickness of torus tubarius (TT) surface mucosa overhanging the Eustachian tube (ET) ostium, and mucosa covering the posterior aspects of bilateral inferior turbinates (ITs) are potentially pertinent to ET balloon tuboplasty. Whether TT and IT mucosal thickness is fixed or varies cyclically is potentially important in therapeutic planning but remains unknown. We establish normative MRI morphometrics for mucosa of TTs, posterior ITs (post-ITs), and anterior ITs (ant-ITs) in healthy subjects, and test the hypothesis that post-IT and TT mucosal thickness may vary during the nasal cycle akin to known changes in ant-ITs.

We retrospectively selected 60 subjects with normal nasopharynx MRIs, and measured maximal contrast-enhancing mucosal thickness for bilateral TTs, post-ITs, and ant-ITs. We conceived a novel indirect test for nasal cycle variations by computing likelihood ratio (LR) statistics comparing unimodal versus bimodal Gaussian distributions of left-right mucosal thickness differences, with significance assessed using bootstrapped p values. We validated this test by confirming detection of known nasal cycling in ant-ITs, and then applied it to assess potential cycling in TTs and post-ITs. Significance was set at p < 0.05.

Mean maximal mucosal thicknesses for TTs, post-ITs, and ant-ITs were 3.1 ± 0.8, 7.6 ± 1.9, and 4.5 ± 1.3 mm, respectively. LR testing revealed significant cyclical variations in ant-ITs (LR = 4.599; p = 0.021), thus validating our methodology for inferring nasal cycling. There was no mucosal cycling for TTs (LR = -0.801; p = 0.728) or post-ITs (LR = -1.208; p = 0.799).

TT and post-IT mucosa unlikely exhibit nasal cycle variations. Single MRI views of TTs and post-ITs would suffice for assessment during pre-interventional planning of ET procedures.

4.