Assessing initial MRI reports for suspected CJD patients

Abstract

Background: MRI is invaluable for the pre-mortem diagnosis of sporadic Creutzfeldt-Jakob disease (sCJD), demonstrating characteristic diffusion abnormalities. Previous work showed these changes were often not reported (low sensitivity), leading to eventual diagnosis at a more advanced state. Here, we reviewed the situation a decade later, on the presumption of improved access and awareness over time. Methods: We reviewed initial MRI scans of 102 consecutive suspected sCJD patients recruited to the National Prion Monitoring Cohort study between 2015 and 2019, assessing for characteristic signal changes in the striatum, thalamus and cortical ribbon. We compared our findings to formal reports from referring centres. Requesting indications were studied to assess if they were suggestive of CJD. Patients were examined and their MRC Prion Disease Rating Scale scores recorded. Results: We identified characteristic MRI abnormalities in 101 cases (99% sensitivity), whilst referring centres reported changes in 70 cases (69% sensitivity), which was a significant improvement in reporting sensitivity from 2012. Reporting sensitivity was associated with signal change in the cerebral cortex, and with the number of regions involved, but not significantly affected by clinical information on request forms, or referring centres being regional neuroscience/non-neuroscience centres. Similar to a previous study, patients with missed abnormalities on initial reporting possessed lower MRC Scale scores when referred to the NPC than those correctly identified. Conclusions: Whilst local MRI reporting of sCJD has improved with time, characteristic abnormalities remain significantly under detected on initial scans. Sensitivity is better when the cerebral cortex and multiple regions are involved. We re-emphasize the utility of MRI and encourage further efforts to improve awareness and sensitivity in the assessment of patients with rapidly progressive dementia.