Traumatic Microbleeds in the Hippocampus and Corpus Callosum Predict Duration of Posttraumatic Amnesia.

Of 350 patients in the database, only 65 had usable acute GRE MRI and PTA data, so the analysis may not represent all TBI patients. MRI scans were ordered for clinical care, not study protocol, at variable times after injury; the median time to MRI was 3.0 days, and the effect of scan timing on microbleed visibility is unclear. The study used GRE MRI, not susceptibility-weighted imaging, so results may not apply to centers using SWI. PTA duration was measured by different methods, including chart review in 42 patients, GOAT in 3 patients, and O-LOG in 20 patients. The sample was mostly severe TBI, with 69% severe, 6% moderate, and one quarter complicated mild. The imaging variables only explained 26% of variation in PTA duration, and the authors did not control for admission hemodynamics, ICU complications, pre-morbid illness, or social support. Subgroup analyses were post-hoc and not significant, including the 15 patients with the longest PTA duration and the 42 patients with PTA longer than 30 days. The sample size may have been too small to detect associations in the thalamus and fornix. This was an imaging association study, not a treatment trial, so it cannot show that any rehabilitation intervention improves PTA.

The TBIMS study was funded by the National Institute on Disability, Independent Living, and Rehabilitation Research. The authors declare no competing financial interests.

Do not use hippocampal or corpus callosum microbleed counts as a clinical predictor of PTA duration. The study was retrospective, the sample was limited, and hippocampal microbleeds did not reach significance in the main correlation analysis; in stepwise regression, only admission GCS was significant.