Age-related white matter hyperintensities: a contribution of cervical venous dys

Anish Kapadia

Department of Medical Imaging,University of Toronto,Toronto,Canada

Abstract

Keywords: All Neurology,Pathogenesis,WMHs,Microangiopathy

Introduction

Cerebral white matter hyperintensities (WMHs) are a common finding on MRI imaging of the brain.It is found with increasing frequency with age and is nearly ubiquitous in the elderly1.It is characterized by multiple foci of periventricular and deep white matter signal change on imaging best seen on T2-weighted fluid inversion recovery (FLAIR) sequences,without significant hypointensity on T1 images.Lesion can also involve the white matter of the brainstem,thalami and basal ganglia2.In more advanced disease,progression to confluent,symmetric,areas of periventricular white matter signal change is observed and extending to the subcortical white matter in the most severe cases.Clinically,this manifests as age-related cognitive decline and in the most severe cases,dementia [1,2].Given the prevalence of the findings,and their potential clinical implications it is important to understand the pathophysiology of the process.Currently it is thought to represent pathologic changes secondary to intracranial small vessel disease,specifically arteriosclerosis; consequently,small vessel lacunar infarcts develop,predominantly in the white matter and basal ganglia,and progress over time to form confluent changes.

Medical Hypothesis

Although WMHs are often lumped into the umbrella of small vessel arterial disease,along with lacunar infarcts,evidence from a range of clinical,imaging and pathologic studies suggest a potential contribution from a different,likely concurrent,underlying mechanism.In addition to the conventionally accepted arterial pathology,a venous etiology is proposed to play a significant role in the development of these changes,one arising due to venous dysfunction at the level of the large cervical veins.With aging there is progressive degeneration of the venous wall with increased stiffness.This results in loss of venous compliance as well as venous valvular dysfunction.Venous reflux and increased pressure in the intracranial venous system secondarily develops.Over time draining cervical veins and of large intracranial veins become dilated.Small venules which are exposed to increased pressures become thickened and narrowed.Ultimately the capillary bed is exposed to higher pressures,through a combination of 1) increased venous pressure and 2) exposure to increased arterial pressure pulsation owing to distal venular stenosis.This results in bloodbrain barrier (BBB) disruption at the capillary bed and proximal small venules,leading to progressive white matter injury.Once BBB disruption occurs,a local secondary neuroinflammatory response develops.Clinically,these changes manifest with cognitive decline and,in more severe cases,dementia.

Evaluation of Hypothesis

Clinical relevance of WMHs

Radiologic and pathologic severity of periventricular and deep WMHs,and clinical deficits are not linearly correlated,and this is corroborated by the loose association between MRI findings and clinical deficits.With that said it is accepted that WMHs are associated with generalized cognitive decline [3].WMHs have been implicated in the earliest stages of cognitive decline which is characterized by deficits in executive function and speed of processing [4,5].Multiple large studies including the Cardiovascular Health Study,Leukokariosis and Disability Study (LADIS) and the Rotterdam trial have linked the severity of WMHs to clinical deterioration including decline in activities of daily living (ADLs),increased disability and even increased risk of death [5-7].In addition to severity of WMHs,progression of WMHs is related to dementia risk [5].The mechanism of cognitive decline is postulated to be related to disruption of in white matter tracts [8,9].Estimated incidence of WMHs in the general population ranges widely based on study population,ranging from 39% to 96%,however,more commonly quoted in the higher end of this range in older clinically relevant cohorts1.In addition,there is an association between presence of WMHs and various dementias (including Alzheimer’s,vascular dementia and Lewy body dementia) and vascular disease [10].Several studies have assessed risk factors which contribute to burden of WMHs and their progression.WMHs are in themselves associated with progression,amongst a host of other factors that include presence of lacunar infarcts,elevated blood pressure,smoking and age [5,6,11,12].There is some evidence suggesting that the risk factors for small vessel disease,which includes WMHs,and large vessel atherosclerosis may differ [13].Notably the burden of WMHs has not been strongly associated with carotid atherosclerosis,diabetes or atrial fibrillation [14-16].WMHs have been shown to correlate with hypertension in some studies,and not others; and,there is lack of efficacy of blood pressure control in slowing progression of cognitive decline [17].In addition,treatment with common interventions for vascular disease such as statins,asprin and clopidogrel have not resulted in slowing progression of white matter disease,and in fact in the case of antiplatelet therapy resulted in increased risk of hemorrhage and death [18,19].Hyperhomocysteinaemia is a known vascular risk factor associated with stroke risk and associated with number of infarcts and total volume of infarcts,however,it is not associated WMHs [20].These findings suggest that white matter disease may not be directly related to conventional risk factors for arterial atherosclerotic/arteriosclerotic disease.

Anatomy and hemodynamics

Anatomically the distribution of the WMHs,in the periventricular and deep white matter,mirrors the course of the venous system following the transmedullary venous channels.Changes are typically symmetric in distribution,likely related to the anatomy of the straight sinus which provides venous drainage to bilateral periventricular/deep venous systems.In patients with WMHs small periventricular veins and venules demonstrate wall thickening with collagen subtype I/III deposition [21].Downstream there is an association between higher burden of WMHs and dilation of the large intracranial veins on imaging,specifically internal cerebral vein and basal vein [22].At the level of the neck,there is evidence for age related hemodynamic dysfunction involving the internal jugular vein (IJV).Ultrasound based evaluations have demonstrated increased IJV diameter,reduced flow and increased reflux with age [23,24].Furthermore,there is generalized decrease in venous compliance with age [25].Increased IJV cross-sectional area has been linked to increased CSF pulsitility,as seen with MRI CSF studies,suggesting transmission of pressure upstream [26].

Imaging and pathology of WMHs

There is evidence for increased BBB-permeability with aging,and this is suspected to progress in an exponential manner [27].The process is global with BBB-disruption seen in normal appearing white matter in patients with WMHs [28].It has been suggested that the increased permeability may precede development of WMHs [29].On MRI,increased BBB-permeability also demonstrates a gravity dependent gradient with increased permeability in the occipital region in supine subjects [30].This may explain the preponderance of most severe changes of WMHs around the lateral ventricular trigone and occipital white matter.There is also evidence for decreased blood flow with age.Some studies have demonstrated association between WMHs and decreased arterial velocities [31,32].It has been suggested that some of the changes may be attributed to brain atrophy; however,hemodynamically brain atrophy would be expected to result in reduced blood volume and not necessarily blood velocity [33].In fact,there is evidence of WMHs are associated with reduced arterial velocities,independent of age [34].Furthermore,reduced brain perfusion in patients with higher burden of WMHs portents an increased risk for dementia [35].Areas of WMHs also demonstrate increased microglial activation suggesting a superimposed inflammatory component [36].This inflammatory process has been observed for a long time but remains poorly understood.These observations of reduced arterial velocities,increased vascular permeability,BBB breakdown and neuroinflammation in patients with WMHs is here posited to represent secondary phenomena from a single pathologic mechanism,specifically,resulting from downstream cervical venous dysfunction.Recent work in mice,provides some evidence along this line,and has shown that vascular congestion promotes BBB breakdown and neuroinflammation [37].

Regions of WMHs are thought to be the end result of a pathologic insult,representing the most severe form of disease.In these patients,regions of normal appearing white matter (NAWM) demonstrate abnormalities on diffusion tensor imaging (DTI),indicating disruption of normal axonal/myelin integrity and increased extracellular fluid [8,9,38].In fact,the DTI changes have been shown to precede and in a step-wise manner result in the development of WMHs [39].Correspondingly pathologic changes can be minimal with only the presence of extracellular matrix disorganization in areas of normal appearing white matter,this is thought to represent the mildest pathologic presentation of the disease [36].Histologically areas of WMHs demonstrate myelin rarefaction,axonal loss and spongiosis; and have been described as incomplete infarcts [36].Regions of WMHs also demonstrate enlargement of perivascular spaces [36]; possibly as a result of brain atrophy or sequalae of chronic increased pulsatility.Although complete white matter infarcts (lacunar infarcts) are also seen in many of these patients,these maybe secondary to a parallel arteriosclerosis.

Consequence and Future Direction

The current evidence based on disparate clinical,imaging and pathologic studies suggests a potential association between WMHs and cervical venous dysfunction.This process is posited to occur in parallel to conventionally accepted atherosclerotic/arteriosclerotic disease,and may underpin significant component of periventricular WMHs.Determining causal relationship requires further clinical and preclinical work.If future studies support a causal relationship between venous dysfunction and WMHs,this will provide a new paradigm in the way cerebrovascular diseases are considered.Particularly the current concepts of age-related cognitive decline will need to be reconsidered in this new context.Given the ubiquity of WMHs with aging,age-related cognitive decline,which is considered "normal" cognitive decline related to brain senescence,may at least in part be due to white matter injury secondary to venous dysfunction.Not only would this reshape our understanding of how age-related cognitive decline occurs,it would also provide a potential therapeutic target to prevent or delay this decline.WMHs are also present concurrently in various dementias including Alzheimer’s and vascular dementia.Understanding the degree to which WMHs contribute to these conditions may add potential new therapeutic targets for slowing cognitive decline in these difficult to treat disorders.In addition to white matter changes small cortical infarcts and cortical thinning is also observed on advanced MRI imaging at high field strengths in patients with WMHs [40].The mechanism for this is proposed to be secondary to disconnection related to white matter injury,or related to infarcts secondary to arteriosclerosis.However,it is possible that venous dysfunction involving the cortical veins may also play a role.As such,age related dilation of the major cortical veins along with findings of venous callogenesis involving their cortical venular tributaries would be expected and could serve as targets of future work.