Poststroke Cognitive Impairment and Dementia

Stroke is a leading cause of disability worldwide.¹ Cognitive decline is increasingly recognized as a substantial contributor to poststroke disability: an estimated 40% of stroke survivors will experience cognitive impairment within the first year of the cerebrovascular event.² Acute injury to brain parenchyma may cause immediate cognitive dysfunction, but some long-term trajectories are characterized by persistent cognitive decline despite improving physical deficits.³ Therefore, recognition and risk stratification are critical to management of poststroke cognitive impairment (PSCI). There are no proven treatments specific to PSCI, but optimizing lifestyle factors and cardiovascular measures for secondary stroke prevention may have plausible benefit for cognitive health.⁴ In this review, we summarize the classification, epidemiology, evaluation, diagnosis, and management of PSCI and poststroke dementia (PSD).

Definitions

PSCI is a heterogenous term which refers to any cognitive dysfunction developing after an index stroke regardless of etiology or underlying pathologic process.⁴ PSCI and PSD are distinguished from vascular cognitive impairment (VCI), which pertains to cognitive impairment related to any vascular pathology irrespective of whether a clinical stroke has occurred. Therefore, PSCI might encompass deficits directly related to cerebrovascular injury (eg, aphasia, strategic infarcts), cognitive decline after a stroke, or cognitive issues preceding a stroke. Some advocate for a more specific definition of PSCI that excludes premorbid cognitive impairment; however, some deficits might not be recognized until after a stroke.⁵ Generally, the diagnosis of PSCI is delayed for 3 to 6 months after index stroke to allow for some initial recovery.

PSD applies to dementia only, but diagnostic criteria for dementia rely on limitations in activities of daily living that may be difficult to apply to people who have experienced a stroke and have physical impairments.⁶ PSD can be classified further as early or delayed, with the former reserved for dementia diagnosed within the 3- to 6-month time frame poststroke and the latter for dementia diagnosed >6 months poststroke.⁷ The lack of a consensus definition for PSCI and PSD presents a barrier in diagnosis, management, and research of these clinical entities.

Epidemiology

Given inconsistent definitions, epidemiologic data on PSCI and PSD have varied. Prevalence estimates for PSCI within the first year after stroke range from 17% to 55%.^8,9 A recent meta-analysis of 21 studies reported a pooled prevalence estimate of 40% for PSCI, but included studies were highly heterogenous in population, time frame for diagnosis, and whether premorbid cognitive deficits were included.² Efforts to harmonize criteria and data by the Vascular Behavioral and Cognitive Disorders (VASCOG) group and Stroke and Cognition (STROKOG) consortium have yielded similar results of 40% to 50% of stroke survivors having PSCI within 6 months of incident stroke.^10,11 Among those with PSCI, the majority exhibit mild cognitive impairment.¹⁰

Demographic characteristics, clinical characteristics, and stroke features affect the risk of PSCI. For instance, older age (75 vs 65 years), race (Black vs White individuals), lower education, and residence in non–Stroke belt regions in the United States strengthened the effect of incident stroke on cognition.¹² Greater premorbid vascular comorbidity burden may also increase the risk of PSCI. In particular, diabetes and atrial fibrillation may be associated with a 1.5- to 2-fold increased risk of PSCI.¹³ A large 15-year prospective study found an elevated incidence of dementia among those with any cerebrovascular event. However, this rate was lowest among transient ischemic attack (TIA) survivors and progressively increased with stroke severity reflected by the National Institutes of Health Stroke Scale, such that investigators estimated dementia to have been advanced by 2 or 25 years after a TIA or severe stroke, respectively.¹⁴ Previous stroke and subclinical ischemic burden may also factor into the risk of PSCI.¹³

Clinical Features and Assessment

Clinical assessment for PSCI is largely focused on neurocognitive testing. Information processing speed and executive function are commonly affected in both ischemic and hemorrhage strokes; however, poststroke cognitive profile can vary given the multitude of factors that might affect PSCI including demographic and sociocultural elements, stroke location, preexisting vascular pathology burden, and comorbidities.^15,16 For all stroke survivors, international guidelines recommend some form of cognitive assessment to tailor rehabilitation and care needs, but there is no consensus regarding the ideal cognitive battery physicians should utilize.^4,17 In addition, neuropsychologic assessments may be challenging to apply to people who have experienced a stroke due to potential physical or communicative barriers that prevent their full participation. Therefore, a testing model staged by phase of stroke care to account for feasibility has been suggested.^7,18 This model or evaluation schema may need to be tailored according to stroke deficits and syndrome.

In the acute setting, such as the intensive care or stroke unit, experts have recommended a brief screen requiring <5 minutes, such as the Mini-Cog or an abbreviated Montreal Cognitive Assessment (MoCA).^7,18 The National Institute of Neurological Disease and Stroke–Canadian Stroke Network (NINDS-CSN) previously devised cognitive screening protocols of various time lengths to study VCI. This includes a 5-minute protocol that was validated for PSCI consisting of the orientation, memory, and phonemic fluency MoCA subtests.^19,20 After stabilization in the acute stroke phase, a full screening for prestroke and current cognitive impairment may be considered. For prestroke cognition, the Informant Questionnaire for Cognitive Decline the Elderly (IQCODE) is an extensively used standardized survey, but it requires reliable informant participation and might suffer from recall bias.²¹ For cognitive impairment screening in the stroke population, MoCA is often preferred over the Mini-Mental State Examination because of increased sensitivity for mild cognitive impairment, but alternatives like the NINDS-CSN protocols or the Oxford Cognitive Screen might be preferable for use in people with aphasia or neglect.^22,23 Those who screen positive could then be referred for a full neuropsychological battery after initial recovery, which might occur in a rehabilitation center or outpatient setting 3 to 6 months after the index stroke.

Cognitive testing for PSCI can be beneficial not only for initial diagnosis and management but also to follow the trajectory of PSCI over time. Serial screening and, if needed, comprehensive neuropsychologic assessments could be continued during long-term clinic follow-up. Trajectories of PSCI can be variable and currently there is no clear method to accurately predict the natural history of PSCI. Some individuals will have an immediate decline in cognition associated with the stroke, but the magnitude can range from little to severe cognitive change. After a stroke, some individuals may experience a steeper decline in cognition than they would without having had a stroke, while others may experience no difference in cognitive decline as a result of a stroke. In addition, although cognitive domains could transiently improve in the shorter term, a declining course could persist for years after the stroke (Figure).^3,24

Management

No therapeutic intervention has been proven to be beneficial in PSCI. Despite the strong observational association among vascular risk factors, cerebrovascular disease, and dementia, there are limited high-quality data supporting any specific intensive lifestyle or risk factor modification for prevention and treatment of PSCI.¹⁸ Nevertheless, it remains highly plausible that strategies to prevent and treat stroke would also benefit PSCI and cognitive health. In 2017, the American Heart Association (AHA)/American Stroke Association (AHS) issued an advisory for practical cardiovascular goals toward maintaining ideal cognitive health.²⁵ These follow the AHA’s Life’s Simple 7, which include 4 lifestyle goals and 3 risk factor guidelines. Ideal cognitive health goals follow preventative strategies for stroke and have been shown to be associated with improved midlife cognition.²⁶

There also are no recommended or approved therapies for symptomatic treatment of PSCI. Various trials tested treatments for PSCI or VCI that are typically used for Alzheimer dementia (ie, rivastigmine, acetylcholinesterase inhibitors, memantine), but these studies failed to demonstrate significant results.^27-29 Nevertheless, a diagnosis of PSCI should not preclude these treatments for Alzheimer dementia, which can frequently coexist with cerebrovascular pathology.³⁰

Conclusion and Future Directions

PSCI and PSD substantially contribute to poststroke disability. Limited understanding of the ideal cognitive testing and pathophysiology of PSCI is a major barrier that hinders physicians’ ability to formulate a clearer clinical definition of and targeted treatment modalities for PSCI. There are ongoing studies assessing the role of cerebrovascular disease in cognitive health including the MarkVCID Consortium.³¹ The Determinants of Stroke Cognitive Outcomes and Vascular Effects on Recovery (DISCOVERY) study includes participants from the United States with incident stroke without dementia and is designed to evaluate cognitive trajectory and promising biomarkers such as serum inflammatory markers, neurofilament light chain, and longitudinal structural and functional magnetic resonance imaging of the brain. This study will seek to harmonize timing and methods of PSCI assessment.³² Therapeutic trials are also critically needed. The ongoing Nutrition Effects on Brain Outcomes and Recovery in Stroke After Hospitalization (NOURISH, NCT04337255) trial is testing a 2- to 3-year application of the Mediterranean-Dietary Approaches to Stop Hypertension Intervention for Neurodegenerative Delay (MIND) diet vs usual poststroke care on cognitive decline and biomarkers of Alzheimer dementia in older acute stroke survivors without premorbid dementia. Clarification of PSCI and PSD is urgently needed to combat the burden of cognitive impairment that is expected to grow in an aging global population.³³