In our examination of the effects of WMH pathology and Aβ deposition on white matter microstructural integrity, we found that (1) WMH pathology was significantly associated with reduced FA in projection and association fibers. (2) Aβ deposition was significantly associated with reduced FA in the fornix and splenium of the corpus callosum. (3) Although there were positive interactions between WMH pathology and Aβ deposition on FA in the internal capsule and parahippocampal white matter, accounting for apoE genotype rendered these interactions only marginally significant. Possible mechanisms and interpretations of the interaction are discussed below.
As hypothesized, WMH pathology was associated with reduced FA in projection and association fibers. This finding is consistent with earlier reports of an inverse relationship between FA and WMH volume 
, reduced FA in the presence of gliosis or infarcts 
, extending beyond normal appearing white matter in subjects with cerebral amyloid angiopathy 
, cerebral autosomal dominant arteriopathy with subcortical infarcts and leukoencephalopathy (CADASIL) 
, and cognitive impairment with and without vascular dementia 
. These results are also in agreement with prior reports that ischemic brain disease 
and vascular risk 
alter FA in normal appearing white matter. The projection and association fiber tracts where we detected significant WMH pathology effects also correspond to the cerebral white matter regions with intermediate-high to low baseline anisotropy that Lee and colleagues 
previously reported were most vulnerable to vascular risk.
As hypothesized, Aβ deposition was significantly associated with reduced FA in the fornix and selenium of the corpus callosum. Several studies have demonstrated that Aβ deposition can lead to DTI changes in mice 
. One previous study examined DTI changes associated with cerebral spinal fluid (CSF) Aβ levels in humans and found that Aβ42 levels were related to lower FA in medial frontal cortex and underlying white matter 
. To the best of our knowledge, this is the first study to examine the relationship between cerebral Aβ deposition and FA. In the current study PiB positivity was associated with lower FA in the fornix, a predominant outflow tract of the hippocampus and a brain region affected early in the course of AD, consistent with previous DTI studies of MCI and AD 
. PiB positivity was also associated with lower FA in the splenium of the corpus callosum, in accordance with previous ROI-based DTI studies of MCI and AD 
There is evidence suggesting that vascular brain injury exerts subtle deleterious effects on brain structure and function beyond frank infarction or hemorrhage 
. For example, epidemiological studies have linked vascular risk factors (e.g., hypertension, hyperlipidemia, diabetes mellitus) to increased incidence of clinically-diagnosed AD 
. Therefore, we hypothesized that there would be an interaction between WMH pathology and Aβ deposition on FA. In the current study, we found an interaction between WMH pathology and PiB positivity in the internal capsule and parahippocampal white matter, where PiB-related reductions in FA was greater among participants who also had WMH pathology. However, these interactions were no longer significant after controlling for apoE ε4 genotype.
One possible mechanism by which WMH pathology and Aβ deposition might interact positively to degrade white matter microstructure is through the reduction of brain reserve. Reserve is the concept that there are characteristics of the brain that buffer the impact of pathology on brain performance 
. The characteristics may be structural (e.g., ‘extra’ neurons and/or synapses) or functional (e.g., a degree of excess capacity or compensatory mechanisms such that the brain could continue to perform well despite damage 
). There is suggestive evidence that WMHs have a negative effect on structural brain reserve. For example, a population-based study of 1,077 non-demented elderly (60–90 years old) individuals found that greater periventricuar WMH severity was associated with a greater risk for developing dementia. Moreover, the association between WMH and dementia risk was independent of vascular risk factors, cerebral infarcts, and brain atrophy 
. Hypertension, one of the strongest risk factor for WMHs, has also been linked with increased risk of developing AD 
and treatment of hypertension has been found to be protective against the development of dementia 
. Thus, in the context of the present study, it could be possible that WMH pathology was related to reduced structural brain reserve, which in turn rendered individuals with WMH pathology more vulnerable to the effects of cortical Aβ and AD pathology. Although it is not clear how precisely WMH pathology degrades brain reserve, some possible mechanisms include impaired neurogenesis and inflammation. Adult stem cells are located in the subventricular region 
, which is frequently affected by WMHs. Thus, the ischemic changes that lead to the development of WMHs could also cause damage to adult stem cells, impairing neurogenesis. Because there is microglial activation in brain tissue affected by WMHs 
and higher C-reactive protein levels in patients with WMHs 
, WMH pathology may also negatively impact reserve through inflammatory cytokines. It is noteworthy that the interaction between WMH pathology and PiB positivity were no longer significant after controlling for apoE ε4 genotype. This suggests that another explanation for the interaction between WMH pathology and PiB may be apoE ε4. Presence of the apoE ε4 allele has been associated with increased amyloid deposition in the parenchyma 
and amyloid positivity. Presence of the apoE ε4 allele may also be associated with amyloid deposition in the blood vessels, which could be associated withCAA and WMH pathology 
. Future studies with a longitudinal design will be needed to examine if increased amyloid deposition in each of these regions is in fact associated with decreased white matter microstructural integrity in different regions of the brain.
The current findings should be considered in the context of several study limitations: First, the study sample was the small and heterogeneous. Second, apoE genotyping was only available in a subset of subjects. Therefore, our interpretation of the role of apoE in the interaction between PiB positivity and WMH pathology on FA are speculative at best and should be regarded with caution. Other study limitations include the low angular resolution of DTI (
, the fact that we did not account for the location of vascular lesions in our analyses of FA values obtained from 9 white matter tracts that covered most of the supratentorial white matter, and the
cross-sectional nature of the study design. Future studies with a longitudinal design will be able to better resolve the nature of the relationship between WMH pathology and PiB positivity on white matter microstructural integrity in different regions of the brain.