MRI-related safety incidents during acute cerebrovascular care: A descriptive analysis of a national incident-reporting database in Japan.

OBJECTIVE

Magnetic resonance imaging (MRI) is integral to acute cerebrovascular care but creates safety challenges beyond metal screening. We characterized MRI-related safety incidents and safety-confirmation breakdown points in a national Japanese incident-reporting database.

METHODS

We conducted a retrospective cross-sectional descriptive analysis of 54,310 adverse-event reports from 2013 to 2024. Candidates were identified by co-occurrence of MRI-related and cerebrovascular terms. Eligible reports described incidents during preparation, transfer, scanning, monitoring, or return transfer in patients evaluated or managed for acute cerebrovascular disease. Incident type and safety-confirmation breakdown point were classified by consensus after re-assessment of a stratified sample by a second neurosurgeon.

RESULTS

Of 1166 candidates, 31 incidents were included. Patient transfer or transport was most frequent (9/31, 29%), followed by deterioration during scanning or monitoring (8/31, 26%) and ferromagnetic or metal objects (6/31, 19%). Inadequate monitoring or observation was the leading breakdown point (12/31, 39%), followed by inadequate information transfer (5/31, 16%). Deterioration during scanning with inadequate monitoring was the most common pairing (7/31, 23%). Ten incidents (32%) were severe and 3 (10%) recorded death, including 2 deaths in the transfer/transport category.

CONCLUSION

In acute cerebrovascular care, reported MRI safety incidents extended beyond ferromagnetic-object events. Transfer-related events and inadequate monitoring were prominent, and breakdown points spanned pre-transfer, pre-entry, and scanning phases. These patterns suggest that safety assessment may need to address clinical stability, transfer readiness, MR Conditional equipment, monitoring, sedation, and emergency response across the MRI workflow. Because these data derive from an incident-reporting database, they do not estimate incidence or establish causality.