|Home | About | Journals | Submit | Contact Us | Français|
Most individuals in old age who receive a clinical diagnosis of Alzheimer’s disease probably have some pathological features other than (but usually in addition to) plaques and tangles in their brains. Each of the pathologies (Alzheimer’s disease or cerebrovascular pathology) are common, and could have developed mostly independent from each other.1
In The Lancet Neurology, Zoe Arvanitakis and colleagues2 provide important evidence relevant to these complex diseases of the aged brain. By use of data from a large and high-quality autopsy cohort (1143 individuals with median age at death of 88·8 years), they studied the association between cerebrovascular pathology and cognitive changes that had been recorded through longitudinal study of the participants. These researchers have done first-rate studies that capitalise on the longstanding community-based nature of their autopsy cohort. The use of neuropathological data is a strength, along with the detailed neurocognitive assessments. Moving away from the notion that cerebrovascular pathology reflects one disease, the researchers investigated the correlative effects of particular types of large vessel (atherosclerosis) and small vessel (arteriolosclerosis and microinfarcts) brain diseases. Both atherosclerosis and arteriolosclerosis independently increase the likelihood of Alzheimer’s disease. These results emphasise three key ideas: the lack of specificity of clinically defined Alzheimer’s disease, the importance and heterogeneity of cerebrovascular pathology, and the complexity of cerebral multimorbidity.
Previous studies have reported that some cerebrovascular pathology is almost ubiquitous in individuals older than 85 years at death,3 and the presence of cerebrovascular pathology causes a relatively impaired cognitive status to be more likely with a given severity (density and distribution) of Alzheimer’s-type plaques and tangles.4 Clinical guidelines have been revised with the goal of increasing the specificity of the clinical diagnosis of Alzheimer’s disease by incorporating biomarkers—mainly CSF assessments and neuroimaging findings.5, 6 Biomarker studies confirm the prevalence of in-vivo suspected non-Alzheimer’s pathology (SNAP)7, 8—ie, cognitive impairment without amyloidosis as assessed with biomarkers. About a quarter of cases of mild cognitive impairment show the SNAP biomarker signature.7 Cerebrovascular pathology is one of the many subtypes of pathology that can underlie cognitive impairment with or without comorbid Alzheimer’s disease pathology.9
The study by Arvanitakis and colleagues underscores that cerebrovascular pathology is both impactful and heterogeneous. As noted by Arvanitakis and colleagues, previous studies were “not able to control for the mediating or confounding effects of microinfarcts and Alzheimer’s disease pathology”,2 whereas they used statistical modelling to account for potential confounders. The severity of atherosclerosis, noted on gross inspection of the vessels of the circle of Willis, was correlated with increased risk of dementia. Arteriolosclerosis indicates pathological thickening of the walls of brain arterioles and is associated with cognitive status and likelihood of Alzheimer’s disease diagnosis. This finding confirms previous results reported by Ighodaro and colleagues,10 who described an association between brain arteriolosclerosis pathology and global cognitive status in a large autopsy cohort. Still another subtype of cerebrovascular pathology is lacunar infarcts, which refers to cerebrovascular pathology in medium-sized vessels, and which seems to at least partly mediate the eff ect of type 2 diabetes in aged brains.11
Together, these results suggest that the aged brain is complex and we should resist the temptation to ration diagnoses, or to develop overly simplistic hypotheses that mostly ignore the pathogenetic complexity. These mistakes have consequences. The cognitive deficits that occur in Alzheimer’s disease, Alzheimer’s disease plus atherosclerosis, and Alzheimer’s disease plus arteriolosclerosis might overlap extensively, but, if people from all three diagnostic categories are included in the same clinical trial, treatment effects might be diluted. Furthermore, the postulated protective effects of diet and exercise (and other factors) might result from their effects on non-Alzheimer’s disease processes that manifest at autopsy as cerebrovascular pathology. There might also be synergy between cerebrovascular pathology and neurodegenerative mechanisms. All these considerations provide support for the conclusion by Arvanitakis and colleagues2 that the pathophysiological pathways linking large and small cerebral vessel disease to dementia need further characterisation.
I declare no competing interests.