Intended for healthcare professionals


Dementia after traumatic brain injury

BMJ 2023; 383 doi: (Published 19 October 2023) Cite this as: BMJ 2023;383:p2065
  1. Neil SN Graham,, NIHR academic clinical lecturer in neurology 1 2,
  2. David J Sharp, professor of neurology1 2
  1. 1UK DRI Centre for Care Research and Technology, Imperial College London
  2. 2Department of Brain Sciences, Imperial College London

This is a difficult time for people with concerns about the long term neurological effects of head injury. Increasing evidence links traumatic brain injuries (TBIs) to neurodegenerative diseases and dementia.12 Many high profile diagnoses of chronic traumatic encephalopathy (CTE), a specific post-traumatic dementia, have been made at post-mortem,34567 contributing to general anxiety about the long term effects of even mild injuries.8 However, there are many unanswered questions which make it difficult to advise individuals of their own risk of problems and how best to promote brain health.

A central uncertainty is which head injuries, and how many, confer a significant dementia risk.910 An increased focus in sports injury management on identifying, and preventing symptomatic injuries, termed “concussions” is welcome,10 but a full understanding of post-traumatic dementia risk will not be provided by focusing only on symptomatic injuries. The vast majority of head impacts do not produce symptoms, and animal modelling clearly shows that brain damage and neurodegeneration can be produced without concussive symptoms, during so-called “sub-concussive hits.”11 This observation has important implications.

Alongside public health efforts to reduce exposure to head injury, which must not be delayed, an ambitious programme of translational research is required. Researchers should use new technologies, such as instrumented mouthguards containing sensitive accelerometers, to measure cumulative exposure to all head impacts, including those not causing immediate symptoms, as well as measuring their effect on the brain. In a contact sport like football, this means systematically recording the number and type of impacts, including heading the ball, and investigating the relationship between this kind of exposure and clinical outcomes. As not everybody goes on to develop problems, clarifying individual-level susceptibility and protective factors is vital. Ultimately, we need to provide individuals with the information they need to make informed choices about risk and participation across a range of activities.

Important types of TBI have been missed on standard clinical assessments in the past, but our ability to identify traumatic brain damage is improving.1213 Advanced imaging methods such as diffusion magnetic resonance imaging (MRI) are able to identify damage, such as diffuse axonal injury, that was previously missed.1415 Blood markers of brain injury also promise to assist TBI identification and management.161718 Ultrasensitive blood tests for proteins, such as neurofilament light, released following brain injury can identify the presence and extent of TBI,12 and a blood test for mild TBI has recently received US and UK regulatory approval,19 reducing the need for CT scanning in some patients. It is likely that biomarkers of TBI will be routinely and repeatedly measured in the future, providing an area-under-the-curve readout of cumulative injury “dose” in people with recurrent injury.20

In the long term, biomarkers of TBI will help to reduce our reliance on traditional clinical severity measures such as loss of consciousness or duration of post-traumatic amnesia, which are frequently unreliable. However, work still needs to be done establishing how they perform in real-world settings and how comorbid health problems and traumatic injuries elsewhere in the body influence these biomarkers. Processing and analysis still need to be standardised, and cost effectiveness evaluated. Most clinical guidelines, including the 2023 UK National Institute for Health and Care Excellence (NICE) Head Injury Guidelines do not recommend blood testing for TBI, despite recent regulatory approvals.21 Large-scale international research collaborations, such as the International Initiative for Traumatic Brain Injury Research (InTBIR), and the forthcoming UK-wide MRC-funded TBI data and biomarker platform hold the key to this real-world validation.22 They will provide the infrastructure to systematically study a wide range of injury biomarkers across different injury severities and patient groups, linked to long term data on outcomes.

Not all head impacts carry the same risk of brain injury, because the forces produced by each impact are unique. Impact biomechanics have been difficult to estimate accurately, but this is changing. Accident reconstructions, video footage, and wearable sensors are increasingly providing information which can be combined with computational modelling to predict how impact forces affect the brain. Particularly high rates of strain are seen in the depths of the sulci where tau pathologies of chronic traumatic encephalopathy (CTE) typically develop.2324 Hence, we increasingly have the data and computational methods to unpick the relationships between biomechanics and brain injury, which will be key to accurately estimating the long term risks of dementia.

A further important source of confusion concerns the diagnosis of dementia in patients with post-traumatic cognitive impairments. Dementia is defined by progressive worsening across multiple cognitive domains, with clinical deterioration over time that significantly impairs normal functioning. It is vital to separate the effects of the initial injury, which are not progressive, from secondary problems due to neurodegeneration, which are. It cannot be over-stated that the presence of clinical problems or brain damage on imaging after TBI does not alone indicate a progressive disorder, such as chronic traumatic encephalopathy or Alzheimer’s disease.

Many people exposed to repeated head injuries are concerned about chronic traumatic encephalopathy, but this is currently a challenging diagnosis, lacking validated diagnostic criteria or reliable in-vivo biomarkers.25 While we await sensitive and specific tests, a careful and systematic approach to diagnosis is needed, incorporating longitudinal neurological assessment, complemented by repeated neuropsychological testing and tools such as volumetric MRI brain to look sensitively for progressive brain atrophy. Clinical practice should be aligned with large-scale long term research studies where possible, to clarify the types and quantity of head injuries likely to affect brain health, and helping to inform individual and societal choices about injury risk, and preventative public health measures.


  • Competing interests: The authors currently receive grant support from a range of bodies to investigate traumatic brain injury. These include the National Institute for Health Research, Alzheimer’s Research UK, ERANET Neuron, Medical Research Council, UK Dementia Research Institute, Rugby Football Union, and Football Association.

  • Provenance and peer review: not commissioned, externally peer reviewed.