The Multiple Faces of Cerebral Small Vessel Diseases

This edition of focused updates in cerebrovascular disease of the journal is devoted to cerebral microangiopathies. Research into cerebral small vessel diseases (cSVD) has blossomed over the past 2 decades. While there are many reasons for such heightened interest in these conditions, one major factor is the visibility of many brain lesions attributable to cSVDs on clinical neuroimaging. Cerebral microbleeds, cortical superficial siderosis, white matter disease, lacunar infarcts, cerebral microinfarcts, and enlarged perivascular spaces are all visible on commonly obtained brain MRI sequences. Studies aimed at validating these radiologic lesions showed good correlation with the actual pathologies. NIH identified CSVD as a major contributor to dementia, therefore a top research priority.¹ The result of the improved funding and development of multidisciplinary collaborations is an exponentially increasing number of high-quality research publications. The current focused updates aim to cover some of the key developments in the field. The progress in our understanding of the mechanisms of disease through neuroimaging and genetics, the latest on therapeutic trials, and finally the clinical relevance of cSVDs are discussed in 5 review articles.^2–6 This introductory article will not restate what is already been discussed in detail in these papers but we will cover three specific areas that present challenges in clinical care and research related to cSVDs, therefore offering opportunities for improvement. We will specifically emphasize the importance of cSVD in multidisciplinary efforts aimed at detection and interpretation of neuroimaging findings in the clinic, prediction of risk for hemorrhagic strokes, and relationship with cognitive impairment in middle aged to older adults.

Understanding of cSVD within the Medical Community

Despite remarkable attempts to standardize the terminology, the lack of clarity even in basic concepts used in clinical and research settings remains an important issue. Prominent examples are the interchangeable use of cSVD and white matter disease in research and the extremely variable descriptive wording used in radiology reports for leukoaraiosis, i.e. “age-related non-specific white matter hyperintensities on FLAIR MRI that might be related to chronic cerebral microangiopathies”. It is not uncommon to see anxious patients who are concerned about the significance of neuroimaging findings such as microbleeds or white matter disease. Some physicians, less familiar about the interpretation of these findings, may unnecessarily alarm patients by informing them that they will definitely develop a brain bleed or dementia in the presence of such imaging findings. On the other hand, the presence of even the better known microangiopathic findings such as microbleeds or cortical superficial siderosis is commonly not mentioned in radiology reports or overlooked by the clinicians when they do not review the appropriate MRI sequence such as gradient recalled echo (GRE) or susceptibility weighted imaging (SWI). Missing such imaging markers of high brain bleeding risk can especially be an issue for patients with concomitant conditions such as atrial fibrillation as their presence can affect management. About 2% of people younger than age 65 have atrial fibrillation whereas approximately 9% of people aged 65 years or older have atrial fibrillation, the latter representing a rapidly growing segment of the population. A broader discussion of the impact of cSVD on accurate detection and management of patients with atrial fibrillation is undertaken in subsequent parts of this text.

Improving the systematic use of standard terminology to classify cSVDs, as well as enhancements in physician education to better detect cSVD-related imaging lesions and understand the relevance of cSVDs overall is an important goal. Better implementation of these data in medical education and residency and fellowship training in neurology and radiology programs might help close the current gaps. Closer collaborations between the neurological and radiologic societies to standardize the cSVD terminology and in-depth discussions of clinically relevant aspects of these microangiopathies in post-graduate education courses are also key methods to improve the brain health of middle aged to older adults.

Immediate Clinical Impact of cSVD: Understanding Brain Hemorrhage Risk

Despite the fact that ischemic strokes are 3–4 times more common, hemorrhagic strokes are associated with 3–4 times higher risk of case-fatality and severe disability in survivors.^{7, 8} The most common cause of intracerebral hemorrhage (ICH) in adults over 55 years of age is cSVD with cerebral amyloid angiopathy and hypertensive arteriolosclerosis being the major etiologic factors. Anticoagulant related ICH is a particularly severe condition, associated with short term-mortality rates of 50% when it happens in the setting of either warfarin or non-vitamin K antagonist oral anticoagulants (NOAC).⁷ Long-term anticoagulation is frequently used for ischemic event prevention, particularly for lowering the risk of embolic strokes in the setting of atrial fibrillation.⁹ ICH is the most feared complication of oral anticoagulants and this concern is the most common reason for the inadequate use of effective ischemic stroke prevention measures in patients with atrial fibrillation. New potential indications for different anticoagulant regimens are emerging. One such new indication is adding rivaroxaban 2.5mg BID to aspirin in patients who have stable coronary artery or peripheral vascular disease without atrial fibrillation, per the results of the Cardiovascular Outcomes for People Using Anticoagulation Strategies (COMPASS) study.¹⁰ One important factor that contributed to the success of the combination rivaroxaban and aspirin regimen in this study was the exclusion of patients who had a past history of lacunar stroke, another cSVD marker that is known to increase the risk of both new lacunar infarcts and ICHs. Anticoagulants do not help prevent lacunar infarcts better than aspirin and anticoagulation increases potentially fatal ICH risk in patients with cSVD. A common occurrence in studies comparing anticoagulants to antiplatelets is that the higher frequency of ICH risk with anticoagulation can outweigh any potential ischemic prevention benefit. Therefore, excluding patients with a higher ICH risk can accentuate the potential net benefit from long-term anticoagulation in randomized controlled trials. Explicit indication of these inclusion and exclusion criteria should be clearly specified in FDA approvals, society guidelines and drug marketing efforts.

When it comes to a condition such as anticoagulant-related ICH with dismal outcomes, it is clear that the best management is prevention. Determining ICH risk accurately is key to prevention. Recent large-scale prospective studies shed further light in the relevance of cerebral microbleeds to stratify ICH risk while on anticoagulation. The “Cerebral microbleeds and intracranial haemorrhage risk in patients anticoagulated for atrial fibrillation after acute ischaemic stroke or transient ischaemic attack” (CROMIS-2) was a multicenter observational cohort study.¹¹ Microbleeds were present in 21% of the 1490 patients enrolled. CROMIS-2 showed significantly higher anticoagulant-related ICH risk in atrial fibrillation patients with brain microbleeds compared to patients without microbleeds (adjusted hazard ratio 3·67, 95% confidence interval [CI] 1·27–10·60). This prospective study also showed the very poor performance of HAS-BLED score in predicting ICH risk (C-index 0·41, 95% CI 0·29–0·53) whereas adding presence of microbleeds improved ICH prediction (0·66, 0·53–0·80). Another recent multicenter study, “Hemorrhage Predicted by Resonance in Patients Receiving Oral Anticoagulants” (HERO), not only confirmed the significantly high risk of incident ICH in patients with microbleeds treated with anticoagulation but also showed that the presence of moderate to severe white matter hyperintensities on FLAIR MRI was also a very significant risk predictor for ICH (hazard ratio 5.7, 95% CI 1.6–20).¹² Overall, the markers of cSVD such as microbleeds and leukoaraiosis can help stratify the risk of ICH in patients considered for lifelong anticoagulation. Using lower bleeding risk regimens, modifying other confounding factors such as hypertension, and treatment alternatives to anticoagulation such as left atrial appendage closure in atrial fibrillation should be considered among patients with cSVD.

The stroke prevention field is developing quickly. Predicting ICH risk through cSVD markers should be an important component of such progress. Factor XI inhibitor studies are currently being planned and these newest generation anticoagulants are thought to impose lower hemorrhagic risk. Left atrial appendage closure (LAAC) with the WATCHMAN device is currently FDA-approved for stroke prevention in non-valvular atrial fibrillation patients who need anticoagulant treatment but who have a reasonable rationale to avoid long-term anticoagulation. A very recent randomized controlled trial, the “Left Atrial Appendage Closure vs. Novel Anticoagulation Agents in Atrial Fibrillation” (PRAGUE-17) study showed identical ischemic prevention results between left atrial appendage closure and NOACs.¹³ New and larger scale studies comparing LAAC to NOACs including long-term outcomes are needed to better understand the role of these approaches in atrial fibrillation patients at different levels of ICH risk. These efforts require strong collaborations between neurology, cardiology and cardiac electrophysiology teams. It is hoped that the upcoming studies of both novel pharmacological and nonpharmacological stroke prevention measures are based on such collaborations and patient groups that can benefit the most from a specific therapeutic measure could be identified.

Cerebral Small Vessel Diseases and Cognitive Impairment

As a result of the increasing awareness of the contribution of cSVD to dementia, 2 of the 3 highest priority research areas for the next 5–10 years identified by NINDS’ Workgroup on Stroke Prevention Research in 2012 were focused on prevention of vascular cognitive impairment (VCI) and imaging biomarkers in stroke prevention.¹ Increased research support and formation of multidisciplinary collaborations resulted in high quality research output in the field of cSVD. Basic science and clinical investigations showed that both the macro-lesions such as brain infarcts/hemorrhages and the smaller lesions such as microbleeds, microinfarcts and white matter disease increase the brain atrophy and disruption of structural connectivity, and that these factors contribute to cognitive impairment. Despite progress in understanding mechanisms of cSVD, the translation of such research to therapeutic advances for treatment and prevention of cognitive impairment has been slow. Part of the problem emanates from the multifactorial nature of cognitive impairment with different factors contributing to the changes in each individual. Brain amyloid and tau accumulation commonly attributed to Alzheimer’s Disease, the cSVDs, cardiovascular factors (hypertension, heart failure, atrial fibrillation), atherosclerotic large vessel disease, other systemic/metabolic problems (diabetes, hepatic/renal dysfunction) and individual factors (sex, genetic and environmental factors) probably contribute at different levels to cognitive changes in each individual. Failure of multiple agents that targeted amyloid in clinical dementia treatment trials is probably due to the fact that amyloid might only be one of the contributors to cognitive problems or that amyloid might even be a byproduct of less well-known mechanisms including effects of cSVD. Therapeutic efforts in the field of dementia and cognitive aging have been dominated by anti-amyloid and anti-tau treatment trials, an approach that underestimates the contributions of cSVD and other cardiovascular conditions.

The cSVDs are not only relevant to cognitive research because of their very common co-occurrence with Alzheimer’s Disease pathologies but cSVD also commonly overlap with other systemic cardiovascular problems. There is increased interest in cognitive consequences of cardiac conditions such as atrial fibrillation and heart failure. Hypoperfusion and embolism are hypothesized to be the culprits mediating cognitive effects of heart disease. Interestingly enough, the imaging measures classically attributed to cSVD appeared to be major determinants of dementia in patients with atrial fibrillation although detailed characteristics related to atrial fibrillation and patients without atrial fibrillation were not included in this analysis.¹⁴ To further complicate things, such cardiovascular pathologies can also contribute to the cSVD markers on brain MRI such as white matter disease or they both can result from common risk factors. Studies of cognitive impairment in heart diseases may not provide accurate results if they do not include details about both the brain structure/function as well as the features of the cardiac condition studied.

Research efforts aimed at understanding the associations between cardiac diseases, dementia, and cognitive aging are increasing. Going forward, it will be even more important to consider the complex interplay between cardiac pathologies, large vessel diseases and finally cSVD, in the science of vascular contributions to cognitive impairment and dementia (VCID).¹⁵ This approach will require strong collaborations between neurology and cardiology subspecialists, as well as translational neuroscience researchers. Because of the reasons previously discussed, it would also be even more important to take into account the potential role of cSVD in clinical trials of Alzheimer’s Disease and related dementias. Such multidimensional planning can prevent more failures in future cognitive treatment studies.

Conclusions

Cerebral small vessel diseases have become major targets for ischemic and hemorrhagic stroke prevention as well as efforts to prevent and treat cognitive impairment. Clinical practice and research in cSVD should involve multidisciplinary collaborations including not only neuroscience but also radiology, cardiac subspecialties (electrophysiology, interventional, surgical), and basic science experts.