Is dementia more likely following traumatic brain injury? A systematic review.

Background: The association between traumatic brain injury (TBI) and dementia is controversial, and of growing importance considering the ageing demography of TBI.

Objective: To review the scope and quality of the existing literature investigating the relationship between TBI and dementia.

Methods: We conducted a systematic review following PRISMA guidelines. Studies that compared TBI exposure and dementia risk were included. Studies were formally assessed for quality with a validated quality-assessment tool.

Results: 44 studies were included in the final analysis. 75% (n = 33) were cohort studies and data collection was predominantly retrospective (n = 30, 66.7%). 25 studies (56.8%) found a positive relationship between TBI and dementia. Clearly defined and valid measures of assessing TBI history were lacking (case-control studies-88.9%, cohort studies-52.9%). Most studies failed to justify a sample size (case-control studies-77.8%, cohort studies-91.2%), blind assessors to exposure (case-control-66.7%) or blind assessors to exposure status (cohort-3.00%). Studies that identified a relationship between TBI and dementia had a longer median follow-up time (120 months vs 48 months, p = 0.022) and were more likely to use validated TBI definitions (p = 0.01). Studies which clearly defined TBI exposure (p = 0.013) and accounted for TBI severity (p = 0.036) were also more likely to identify an association between TBI and dementia. There was no consensus method by which studies diagnosed dementia and neuropathological confirmation was only available in 15.5% of studies.

Conclusions: Our review suggests a relationship between TBI and dementia, but we are unable to predict the risk of dementia for an individual following TBI. Our conclusions are limited by heterogeneity in both exposure and outcome reporting and by poor study quality. Future studies should; (a) use validated methods to define TBI, accounting for TBI severity; (b) follow consensus agreement on criteria for dementia diagnosis; and (c) undertake follow-up that is both longitudinal, to determine if there is a progressive neurodegenerative change or static post-traumatic deficit, and of sufficient duration.