Brain atrophy and white matter changes grading agreement on NCCT and MRI in ischemic stroke

Purpose Evaluation of brain frailty measures like atrophy and white matter changes (WMC) are becoming increasingly relevant in stroke outcome prediction but are conventionally thought to be best seen on MRI compared to CT. We assessed agreement between baseline CT vs follow-up MRI ratings for brain atrophy and WMC; and compared their predictive validity of 90-day functional outcomes in acute ischemic stroke. Methods In this post-hoc, observational analysis of baseline CT and follow-up MRI data from the Alteplase compared to Tenecteplase (AcT) randomised-controlled trial, experts assessed brain atrophy as well as periventricular and deep WMC using established scales. Binary agreement (none-mild vs. moderate-severe) and agreement across the full range of scores between atrophy and WMC measures on CT and MRI were calculated using Gwet’s agreement coefficient (AC1). Logistic regression and DeLong’s test were used to compare prediction of 90-day modified Rankin Scale (mRS) 0–1. Results Among 1,577 AcT participants, 491(31.1%) had interpretable CT and MRI. Binary agreement was substantial for periventricular (AC1 = 0.70) and total WMC (AC1 = 0.68) scores, Koedam scale (AC1 = 0.76) and frontal atrophy (AC1 = 0.80). Almost perfect agreement (AC1:0.81–0.97) was found for all other measures. There was no significant difference between NCCT or MRI in predicting 90-day mRS 0–1 for any measures. Conclusions CT ratings of brain atrophy and WMC by experts have substantial to almost-perfect agreement compared to MRI. Both generally achieve similar prediction of 90-day functional outcomes. This implies that it is reasonable to use CT scans to evaluate these brain frailty measures in clinical practice and stroke trials.