In this month's “Paper of the Month,” Steffens and colleagues (Steffens et al., Reference Steffens, McQuoid and Potter2014) report on the relationship between depression, mild cognitive impairment (MCI) and future risk of progression to dementia. It has long been known that there is a complex relationship between depression and cognitive impairment, with depression common in those with cognitive impairment and dementia, and subsequent cognitive decline frequent in patients with depression (Wallin et al., Reference Wallin, Boström, Kivipelto and Gustafson2013). This relationship has often been difficult to study, not least because definitions have often been mutually exclusive. For example, many studies of MCI have excluded patients with depression. This is not unreasonable to increase diagnostic certainty and obtain a more “pure” group of those with MCI who are likely to progress to Alzheimer's disease, but of course at the same time limits the ability to determine interactions between MCI and depression. On the other hand, it is known that non-cognitive symptoms, most particularly depression, frequently occur in those with MCI and have been shown to increase the risk of likely future decline.

The study of Steffens and colleagues (Steffens et al., Reference Steffens, McQuoid and Potter2014) uses a large cohort, the Neurocognitive Outcomes of Depression in the Elderly study (NCODE) from Duke University to follow 295 depressed subjects and 161 controls over a mean six year period. They find that, using standard definitions of MCI (but not excluding those with depression), around 21% of the depressed group meet criteria for MCI compared to less than 3% of controls. Over the follow-up period, many more of those in the depressed group (14%) compared to controls (2%) subsequently developed dementia, this being Alzheimer's disease in two thirds of cases. Examining predictors, both age and the presence of amnestic MCI at baseline were strong predictors of progression, and interestingly age was a much stronger predictor in the multi-variate model than the presence of amnestic MCI, though both were significant. The large sample size and the long term follow-up are major strengths of the study.

The study provides important new information about the prevalence of amnestic MCI in a large group of older depressed subjects and confirms that, as with amnestic MCI in the absence of depression, the presence of amnestic MCI is a poor prognostic feature in terms of increasing the risk for progression to future dementia, most often Alzheimer's disease. Of course, one of the key questions is why subjects with late life depression are at increased risk of future cognitive decline and dementia, and whether amnestic MCI in depression has the same biological underpinnings as amnestic MCI in non-depressed subjects. The authors discuss possible mechanisms in the paper, including the fact that vascular factors, known to be more common in those with late life depression (Gonzalez and Tarraf, Reference González and Tarraf2013; Egeto et al., Reference Egeto, Fischer, Ismail, Smith and Schweizer2014), may be important, though the imaging of vascular changes was not examined in this report. Other potential mechanisms include possible neurotoxic effects of raised cortisol levels during depression, or the expression of depression as a prodromal state of Alzheimer's disease. The latter may to some extent be supported by reduced hippocampal volume which has been demonstrated in late-life depression, and its reported relationship with cognitive impairment and decline in depressed subjects. However, caution is needed because structural atrophy on imaging can reflect a number of different neurobiological processes, so hippocampal volume shrinkage may equally well be due to vascular and steroid related changes as degenerative pathology (Ferrarini et al., Reference Ferrarini2014). Unfortunately, there have been very few pathological studies of subjects with late life depression to ascertain the likely cause of the structural brain atrophy which occurs, and further pathological studies would be very informative. In addition, the availability of CSF and imaging biomarkers for degenerative pathology now allow new mechanistic studies to be undertaken in late life depression to help elucidate correlates and mechanisms of cognitive decline. Finally, as with Alzheimer's disease, there is much interest in the possible role for neuroinflammation in depression which may provide a link between depression and cognitive decline. The outcomes of such mechanistic studies are not purely for scientific curiosity, but they will shed light on likely aetiological mechanisms, providing new targets towards which therapeutic interventions can be directed, with the aim of improving outcomes for those with late life depression and preventing future cognitive decline.