Third Ventricle Width is a Reliable and Clinically-Feasible Marker of Brain Atrophy in the Multiple Sclerosis Clinical Care Setting
Sarah Levy1, Jasmin Patel1, James Sumowski1
1Neurology, Icahn School of Medicine at Mount Sinai
Objective:
Examine reliability and clinical utility of third ventricle width (TVW) on standard-of-care MRIs as a marker of cerebral atrophy in real-world multiple sclerosis (MS) clinical care settings.
Background:
Cerebral atrophy is emerging as an important marker of MS disease progression, but clinically-feasible volumetric measurements remain elusive in clinical care settings. Research supports TVW as a reliable, valid, fast, and easy manual two-dimensional MRI estimate of cerebral atrophy that correlates with clinical disability and tracks with volumetric changes. Importantly, most TVW research has utilized standardized research MRIs; it is unclear how findings translate to standard-of-care (SoC) MRIs (i.e., variable field strengths and sequence parameters). We evaluated real-world clinical utility by examining reliability and clinical relevance of TVW measured from SoC (i.e., clinical) MRIs.
Design/Methods:
We measured TVW from SoC brain MRIs in 563 consecutive patients aged 18-65 years. Three trained raters blind to other patient data measured TVW in millimeters from midpoints of left and right boundaries of the third ventricle in axial T1-weighted images. TVWs were adjusted for demographic characteristics. Disability was assessed with the gold-standard MS Functional Composite (MSFC) evaluating cognition (SDMT), upper extremity coordination (NHPT), and gait (T25FW). GLM assessed TVW differences across level of level of impairment (number of MSFC components <5th percentile).
Results:
Interrater reliability of TVW measurement indicated excellent absolute-agreement (two-way mixed effects ICCs: single-rating 0.982; 95%CI: 0.958, 0.994; mean-rating k=3, 0.994; 0.986, 0.998). Medium-to-large difference in TVW across MSFC impairment (p for trend <0.001; ηp2=0.089): TVW (mean [95%CI]) increased across patients impaired on no (n=245; 3.34 [3.14, 3.54]), one (n=155; 3.74 [3.45, 3.99]), two (n=106; 4.30 [4.00, 4.60]), and three (n=57; 4.81 [4.40, 5.22]) MSFC tasks.
Conclusions:
Measurement of TVW from standard-of-care MRIs is a highly-reliable and clinically-relevant marker of brain atrophy in MS that can be easily implemented into routine clinical care.