There is compelling evidence that traumatic brain injury (TBI) can trigger neurodegeneration, and that this is a major determinant of long-term outcome (Smith et al., Reference Smith, Johnson and Stewart2013). A single significant injury such as a road traffic accident or exposure to a bomb blast can predispose an individual to Alzheimer's disease (AD), and here Gilbert and colleagues show for the first time that a history of TBI also alters the progression the disease.

The key finding is that TBIs occurring in late life are predictive of a more rapid progression of AD. The data are taken from a comprehensive cohort of older adults (>65 years old) in Utah, recruited while healthy to a study of dementia progression. During follow-up 325 patients developed AD, and retrospective information about the number of broad type of TBI was available. Individuals who had experienced TBI less than 10 years before the onset of their dementia had a more rapid progression of functional impairment. Patients had been followed up for between 1 and 11 years, but the study is limited by a large drop-off in numbers as the length of follow-up increased. This provides a complex statistical challenge, and makes interpretation of parts of the data problematic. Nevertheless, the study provides evidence that, in the first few years after dementia diagnosis, functional outcome is worse if there has been a relatively recent TBI.

Interestingly, previous work from the same cohort suggests that the type of cognitive impairment seen in AD patients can be influenced by the presence of existing brain injury (Rao et al., Reference Rao2010). Disinhibition was more common in dementia patients with a history of TBI. Dysexecutive problems such as disinhibition are often seen after TBI in the absence of AD, and we have shown they relate to particular patterns of frontal white matter damage (Bonnelle et al., Reference Bonnelle2012). Taken together these results suggest that existing patterns of cognitive impairment produced by TBI may be accentuated by the development of AD.

These findings motivate a more detailed investigation of interactions between TBI and its late neurodegenerative effects. Key pathophysiological factors that may link TBI to AD include the formation of amyloid plaques and neurofibrillary tangles, and the presence of persistent neuroinflammation after injury. Amyloid plaque deposition is increased in models of TBI in transgenic mice, and recent human neuroimaging provides evidence for amyloid plaque deposition in the first year after a TBI (Hong et al., Reference Hong2014). Other key pathological features of AD are seen after TBI. In particular, hyperphosphorylated tau accumulates in neurofibrillary tangles after both single and repetitive TBI (McKee et al., Reference McKee2013; Smith et al., Reference Smith, Johnson and Stewart2013). The type of tau seen after TBI is similar to AD with respect to isoform ratio and phosphorylation state (Schmidt et al., Reference Schmidt, Zhukareva, Newell, Lee and Trojanowski2001), and has also been shown to be present after a single significant TBI in around one-third of patients (Johnson et al., Reference Johnson, Stewart, Begbie, Trojanowski, Smith and Stewart2013). Finally, we have shown that neuroinflammation in the form of microglial activation is present many years after TBI (Ramlackhansingh et al., Reference Ramlackhansingh2011), which is associated with the remnants of axonal injury (Johnson et al., Reference Johnson, Stewart, Begbie, Trojanowski, Smith and Stewart2013). These observations suggest that TBI can trigger or accelerate neurodegeneration and that neuroinflammation may play a role in these effects.

Obviously, not everyone develop neurodegenerative complications after TBI. Therefore, understanding why some individuals are susceptible to late effects is important. Gene/environment interactions are likely to be critical. The most important gene identified to date is APOE, and this is known to influence the risk of developing AD after TBI (Mayeux et al., Reference Mayeux1995), as well as the progression of AD after stroke (Mielke et al., Reference Mielke2011). APOE status was available on the participants in this study, but the effects of genotype were either not examined because of small numbers or were negative, perhaps because of lack of power. It is important that larger studies examine the interaction of APOE and other genes with neurodegeneration after TBI, as the identification of genetic risk factors will allow individuals at most risk to be identified prior to the development of neurodegenerative conditions.

Longitudinal studies of the interaction between TBI and neurodegenerative conditions such as AD are needed. This is a challenging but important research area. The studies are made possible by the availability of sensitive neuroimaging and CSF biomarkers of pathology, which should allow the early stages of the neurodegenerative cascade to be identified after TBI.