Cognitive deficits and mild traumatic brain injury

BMJ 2013; 346 doi: http://dx.doi.org/10.1136/bmj.f1522 (Published 13 March 2013) Cite this as: BMJ 2013;346:f1522
  1. V F J Newcombe, academic clinical fellow in emergency and intensive care medicine,
  2. D K Menon, professor of anaesthesia
  1. 1Division of Anaesthesia, University of Cambridge, Addenbrooke’s Hospital, Cambridge CB2 2QQ, UK
  1. dkm13{at}wbic.cam.ac.uk

New study identifies risk factors and raises questions about the nature of any implied causal association

Between 7% and 33% of patients who have “mild” traumatic brain injury (sometimes called concussion) develop persistent post-concussion syndrome, which may last weeks to months after injury.1 More than 15% have a measurable cognitive deficit at one year.2 3 There is growing interest in the syndrome of post-traumatic encephalopathy,4 5 which may follow a blast injury or repeated sports related concussion. However, despite this growing literature on the cognitive consequences of mild traumatic brain injury, our knowledge of risk factors that predispose people to sustaining such injury is limited.

In a linked paper (doi:10.1136/bmj.f723), Nordström and colleagues examine the associations and temporal associations between a history of concussion, cognitive function, academic achievement, and measures of social wellbeing in a cohort of more than 300 000 Swedish conscripts.6 Given the paucity of data on premorbid neurocognitive testing in traumatic brain injury, this paper draws on an impressively large dataset that allows comparison of neurocognitive function before and after such injury in a nationwide cohort of Swedish men.

The results complement an earlier study from the same group, which examined the association between cognitive performance and incidence of a subdural haematoma. That study concluded that low global intelligence in adolescence was a risk factor for subsequent development of a subdural haematoma.7 Although the current study investigates a more common diagnosis, case ascertainment was probably less precise than the more clearly definable endpoint of subdural haematoma. The case ascertainment of “concussion” that the authors used was based on the International Classification of Diseases and probably represents the best epidemiological approximation achievable in the administrative databases that were searched. However, a substantial proportion of patients with mild traumatic brain injury are never admitted to hospital or seen in the outpatient setting. Therefore, this study probably underestimated the incidence of this condition in the study population. Conversely, the approaches used may not have fully excluded subjects who sustained a moderate or severe injury. Cross correlating multiple sources of data could mitigate against this source of confounding, which is common when administrative datasets are analysed.8

Despite these caveats related to case ascertainment, Nordström and colleagues’ study provides unique insights into the epidemiology of mild traumatic brain injury. Unsurprisingly, poor cognitive function, low educational status, and other risk factors were associated with mild traumatic brain injury. However, surprisingly, the association between cognitive function and concussion did not depend on the temporal association between the two and was just as common when poor cognitive performance preceded concussion. In addition, similar cognitive scores were seen before and after injury in twins discordant for mild traumatic brain injury, which suggests that both genetic and environmental influences contributed to the low cognitive function found. Other strong independent (but not unexpected) risk factors for development of mild traumatic brain injury included a previous episode of brain injury, hospital admission for intoxication, and low education and socioeconomic status. Surprisingly, the analysis found no significant differences in cognitive performance before and after the index event in men who sustained an injury.

These results are important for several reasons. Firstly, they identify potential risk factors for mild traumatic brain injury and could help guide attempts to investigate prevention strategies, perhaps through education initiatives (particularly in accessible populations such as the military conscripts investigated here). Secondly, they provide a context for interpreting studies that measure cognitive function after injury only and compare it with matched controls from the general population, with the assumption that those with brain injury have similar pre-injury characteristics to the general population. The results of this study suggest that such assumptions may be incorrect. Finally, those who subsequently sustained a mild traumatic brain injury had similar cognitive performance to that of those who had previously sustained such an injury, which implies that the injury itself may not reduce cognitive function. However, the tests used (word recollection; visuospatial geometric perception; logical and inductive performance; and mathematical and physics problem solving) have not been validated as sensitive measures of changing performance in cognitive areas thought to be affected by mild traumatic brain injury. These tests may therefore have missed important changes.

It is important that additional studies attempt to replicate these findings. Suitable populations for such studies include other military cohorts and cohorts of people who practise contact sports, which are associated with a relatively high incidence of mild traumatic brain injury. Such studies must take account of “gaming” by soldiers and sportspeople, who allegedly choose to perform suboptimally on pre-injury cognitive screening to hide evidence of any post-injury cognitive decrement, thus enabling them to stay with their units and teams. Although it may not be easy to control for such confounding, more studies like the current one will increase our understanding of the epidemiology, pathophysiology, and outcome impact of traumatic brain injury.


Cite this as: BMJ 2013;346:f1522

Related links


  • Research, doi:10.1136/bmj.f723
  • Competing interests: Both authors have completed the ICMJE uniform disclosure form at http://www.icmje.org/coi_disclosure.pdf and declare: no support from any organisation for the submitted work; no financial relationships with any organisations that might have an interest in the submitted work in the previous three years, no other relationships or activities that could appear to have influenced the submitted work.

  • Provenance and peer review: Commissioned; not externally peer reviewed.