Exploring the Link between Anticoagulation, Cognitive Impairment and Dementia in Atrial Fibrillation: A Systematic Review

Background: The impact of oral anticoagulants (OACs) on cognitive impairment and dementia in patients with atrial fibrillation (AF) is not well characterized. This systematic review aims to address this knowledge gap. Methods: SCOPUS and PubMed searches were conducted to identify articles in the English language investigating the association between the use of OACs and cognitive impairment and dementia. We excluded non-original research studies and studies that did not report data on cognitive impairment or included patients who underwent open heart surgery or had psychiatric illnesses or cancer. Results: Out of 22 studies (n = 606,404 patients), 13 studies (n = 597,744 patients) reported a reduction in cognitive impairment/dementia in those undergoing thromboprophylaxis. Using direct oral anticoagulants (DOACs) was associated with a lower incidence of cognitive impairment in 10 studies (n = 284,636 patients). One study found that patients undergoing dual therapy (n = 6794 patients) had a greater incidence of cognitive impairment compared to those undergoing monotherapy (n = 9994 patients). Three studies (n = 61,991 patients) showed that AF patients on DOACs had a lower likelihood of dementia diagnosis than those on vitamin K antagonists (VKAs). Dementia incidence was lower when VKAs were under good control. Conclusions: The use of oral anticoagulants has the potential to prevent cognitive impairment and dementia in patients with AF. Since most of the published research on this subject is observational in nature, more randomized controlled trials are needed to fully understand the effect of anticoagulants on cognitive function.


Introduction
The prevalence of atrial fibrillation (AF) has been increasing worldwide, especially in the older population.It has been suggested as a risk factor for developing dementia and other cognitive problems even without a prior history of stroke [1,2].The high prevalence and possible cognitive complications of AF pose a significant health and economic burden on the general population.Multiple mechanisms have been proposed as an explanation for the association between AF and cognitive impairment.This includes reduced blood flow, inflammation, and microemboli [1,[3][4][5][6][7].Shared risk factors such as old age, arterial hypertension, diabetes, hyperlipidemia, sleep apnea, coronary artery disease, heart failure, chronic kidney disease, obesity, inactivity, and excessive alcohol use [8][9][10][11] could also explain the link between cognitive impairment and AF.Therefore, it has been speculated that preventive strategies for AF complications, namely the use of anticoagulants, could also target cognitive impairment by lowering the risk of microemboli and stroke.However, excessive or prolonged anticoagulant usage may also raise the risk of brain microbleeds [12], which could negatively influence cognitive function.Therefore, further research is needed to understand the impact of medications used to prevent AF complications on cognitive impairment and dementia.In this systematic review, we investigated whether different oral anticoagulants, adherence to treatment and time in therapeutic range (TTR) impact cognitive impairment or dementia among patients with AF.

Methods
In conducting this systematic review, we followed the methods outlined in the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) [13].
Eligibility Criteria: Only studies published in the English language were deemed eligible if they met the following criteria:

■
Studies focusing on the pathophysiology of AF or that do not report data on cognitive impairment or only provide results from brain imaging, biomarkers, or genetic markers.

■
Studies that do not include original data, or are editorials, case reports, case series, systematic reviews, or meta-analyses.

Study Identification
We reviewed articles based on their titles and abstracts, applying predetermined inclusion and exclusion criteria along with a standardized data collection form.Articles that did not meet the inclusion criteria based on the abstract were not considered for fulltext review.The decision to include full texts was made through consensus.All findings were transferred to Zotero, an open source research tool used for organizing and analyzing data, and any duplicate entries were removed.

Data Extraction and Outcomes
A structured form for data collection was used to obtain details from every study, such as the study's design, patient demographics, initial variables, method of cognitive assessment, length of follow-up in years, AF detection methods, maximum adjusted variables, hazard risks (HR)/odds ratios (OR) with a 95% CI, and the treatment administered.
Our primary objective was to assess the effects of OACs on cognitive impairment in AF patients.Studies reporting the use of specific OACs, such as VKAs (warfarin), as well as more recent non-vitamin K-dependent novel oral anticoagulants (NOACs), were also included.In cases where the article did not identify them precisely, oral anticoagulants were collectively referred to as OACs.

Data Synthesis and Analysis
We refrained from conducting a meta-analysis due to notable variations in methodologies and heterogeneity among the studies we reviewed.These differences include the approaches used in conducting the studies, the characteristics of patients involved, the tools utilized for cognitive function assessment, the outcomes observed, and the statistical techniques employed.This rendered the pooling of data unsuitable and potentially prone to generating misleading conclusions.

Risk of Bias Assessment
The quality of each article was assessed using the Newcastle-Ottawa Scale (NOS) items and The Cochrane Collaboration's tool for evaluating the risk of bias in randomized trials.Articles with an NOS score of ≥6 stars were classified as being of high quality, while those with an NOS score of <6 stars were classified as being of low quality.

Associations with Oral Anticoagulants
Three cohort studies [15,20,23] indicated that the use of OACs vs. no OACs was associated with a lower incidence of dementia/cognitive impairment.One report revealed a higher risk of cognitive impairment with dual therapy [15].Additionally, one report showed that dual therapy was associated with a higher risk of dementia/cognitive impairment compared to no treatment (Table 1).

Association with Vitamin K Antagonists
Three cohort studies [15,26,31] and one cross-sectional study [27] addressed this question.Overall, while some studies suggest a potential benefit of warfarin over non-OAC treatment and a reduction in dementia risk with DOACs compared to VKAs, other studies showed mixed or non-significant differences.Therefore, the takeaway is that the association between anticoagulant types and dementia/cognitive impairment risk remains complex, and further research is warranted (Table 1).

Direct Oral Anticoagulant vs. Vitamin K Antagonists
Four cohort studies [15,[20][21][22] and two RCTs [18,19] provided evidence suggesting that DOACs may be associated with a reduced risk of dementia or cognitive impairment

Associations with Oral Anticoagulants
Three cohort studies [15,20,23] indicated that the use of OACs vs. no OACs was associated with a lower incidence of dementia/cognitive impairment.One report revealed a higher risk of cognitive impairment with dual therapy [15].Additionally, one report showed that dual therapy was associated with a higher risk of dementia/cognitive impairment compared to no treatment (Table 1).

Association with Vitamin K Antagonists
Three cohort studies [15,26,31] and one cross-sectional study [27] addressed this question.Overall, while some studies suggest a potential benefit of warfarin over non-OAC treatment and a reduction in dementia risk with DOACs compared to VKAs, other studies showed mixed or non-significant differences.Therefore, the takeaway is that the association between anticoagulant types and dementia/cognitive impairment risk remains complex, and further research is warranted (Table 1).

Direct Oral Anticoagulant vs. Vitamin K Antagonists
Four cohort studies [15,[20][21][22] and two RCTs [18,19] provided evidence suggesting that DOACs may be associated with a reduced risk of dementia or cognitive impairment compared to VKAs in patients with AF.However, some studies did not find significant differences in dementia risk between DOACs and VKAs.Additionally, RCTs showed comparable incidence rates of dementia and stroke between DOACs and VKAs, with no significant differences in cognitive function scores between the two groups (Table 1).

Direct Oral Anticoagulant vs. Antiplatelet
Two cohort studies [15,35] compared the use of DOACs with antiplatelet medications in relation to cognitive impairment in atrial fibrillation patients.Their findings suggest that DOACs may have a protective effect against cognitive impairment compared to antiplatelet medications in AF patients (Table 1).

Vitamin K Antagonist vs. Antiplatelet
One cross-sectional study [24] and one RCT [25] found that warfarin, particularly when TTR is controlled well, may be more beneficial in terms of cognitive outcomes compared to aspirin in AF patients (Table 1).

Warfarin/VKAs and Time in Therapeutic Range (TTR): Evaluating the Relationship
Four cohort studies [22,26,28,32], one cross-sectional study [24] and two RCTs [18,33] explored the relationship between warfarin/VKA usage and TTR in relation to cognitive impairment and dementia risk with varying statistical significance observed.Overall, the findings suggest that maintaining good control of INR levels, as reflected by a higher TTR, is associated with a decreased risk of cognitive impairment and dementia in AF patients receiving warfarin or VKA therapy.However, studies showed varying statistical significance regarding this relationship, with some reporting a significantly decreased incidence of dementia with good INR control (TTR > 70%) compared to poor INR control [22].Additionally, AF patients with a TTR of 65% or above had a non-significantly decreased probability of developing new-onset dementia compared to those with a TTR below 65% [26].Lower mean TTR percentages were associated with a higher risk of dementia, with decreasing TTR categories showing increased dementia risk [28,32].Furthermore, using warfarin with a TTR below the median value was linked to a higher risk of cognitive impairment, and incrementally higher risks for dementia were observed as TTR worsened [24,33].Overall, these findings emphasize the importance of maintaining good control of INR levels within the therapeutic range to mitigate the risk of cognitive impairment and dementia in AF patients receiving warfarin or VKA therapy (Table 1).

Association between Cognitive Impairment, Time in Therapeutic Range (TTR) and Anticoagulation Control
Three cohort studies [29,30,34] suggested that cognitive impairment may impact anticoagulation control, potentially affecting the management of AF patients receiving VKAs.The findings indicate that lower MMSE scores were associated with lower TTR values, and patients with cognitive impairment tended to use more VKA-interacting drugs (Table 1).

Association of Non-Adherence to Antithrombotic Therapy and Medication Type with Cognitive Impairment
One cross-sectional [14] and two cohort [16,17] studies showed that non-adherence to antithrombotic therapy was associated with cognitive impairment, while higher medication adherence, particularly with DOACs, may contribute to improved cognitive function in atrial fibrillation patients.Additionally, cognitive function itself can influence medication adherence, with better cognitive function associated with better adherence (Table 1).

Discussion
Based on this systematic review, we found that the use of OACs in patients with AF is generally associated with a lower risk of cognitive impairment.Only 1 out of the 22 studies included in our review demonstrated a detrimental association between thromboprophylaxis and cognitive impairment.No significant differences were seen between DOACs and VKAs, and both were more effective than antiplatelet medications.A lower risk of cognitive impairment was also linked to maintaining good control of TTR/INR levels.Notably, studies have shown that cognitive function may affect adherence to oral anticoagulants, which may increase morbidity in AF patients.

Mechanism of Cognitive Impairment and Dementia in AF
The relationship between AF and dementia may be explained by shared vascular risk factors [36,37] and AF-induced silent brain infarcts and thromboembolic strokes [38][39][40][41][42][43][44].Stroke, the most severe consequence of AF, is due to blood stasis in the left atrial appendage and subsequent clot formation, which ultimately propagates to intracranial vasculature, leading to cardioembolic stroke [45][46][47][48].It is suspected that the association between microembolism and dementia is related to symptomatic strokes.However, most micro cerebral infarcts discovered using brain MRI do not manifest as apparent strokes [49].This may explain why patients with dementia sometimes have no signs of a stroke.A previous report has shown that nearly one-third of patients with AF exhibit evidence of silent brain infarcts on MRI [50].Anticoagulation can reduce the risk of subclinical strokes, hence potentially reducing the likelihood of cognitive impairment, as suggested by the results of our systematic review.
Cerebral microhemorrhage appears to also be common among AF patients, regardless of whether they are administered OACs or not [51].A meta-analysis involving approximately 7000 patients with AF revealed a prevalence of microbleeds reaching 28% [52].These results are consistent with the proposed mechanism that subclinical cerebral ischemia lesions may increase the risk of dementia in patients with AF [36,49,53,54].

Impact of Oral Anticoagulants on Cognition and Dementia
Our findings of a favorable impact of anticoagulants on cognitive impairment and dementia agree with prior reports [36,44].Similarly, lower TTR has been previously shown to increase the risk of dementia [55,56], as we reported.There is evidence that it is feasible to maintain cognitive function and lower the risk of dementia by properly regulating INRs within the therapeutic range [36].Nevertheless, another report could not provide evidence of cognitive benefit or harm from anticoagulation [57].
It is worth noting that various factors could potentially influence the decision-making process regarding the use of oral anticoagulation, including frailty, increased fall risk, and bleeding diathesis.Additionally, some of these patients may have preexisting conditions that could independently elevate the risk of cognitive impairment, regardless of oral anticoagulation use [58].

Comparing Different Medications
In our review, we found no significant differences between DOACs and VKAs, and both were more effective than antiplatelets.However, in a previous report [44], the antiplatelet treatment showed equivalent efficiency to anticoagulant therapy in preventing dementia among patients with AF.Differences in the characteristics of the patients enrolled in the studies may explain differences in the results.Since the use of anticoagulants in patients with AF is dictated by the guidelines to prevent stroke, it is a challenge to compare the impact of different anticoagulants and antiplatelets in isolation from the comorbidities of the patients.
As ablation has become increasingly popular as a mainstream treatment for patients with longstanding AF in recent years, it is vital to acknowledge its association with a risk of silent ischemic strokes and cerebral microbleeds, which could potentially lead to declining cognitive function in this population [59,60].In our review, we focused on medications only.Future research is needed to investigate the impact of ablation on cognitive function in AF patients.

Conclusions
Our review highlights the potential benefit of oral anticoagulants in reducing the risk of cognitive impairment and dementia in patients with AF.Further research is needed to understand the complex relationship between AF with cognitive impairment and dementia, as well as the potential impact of anticoagulants on prevention.

■
Observational studies or randomized controlled trials.■Studiesenrolling patients with AF, either permanent, persistent or paroxysmal, and who are >18 years old.■ Studies enrolling patients with AF who are receiving anticoagulant therapy.■ Studies whose outcomes of interest are clearly reported and assessed via recognized scoring systems, e.g., MMSE (Mini Mental State Exam), MoCA (Montreal Cognitive Assessment), IQCODE (Informant Questionnaire on Cognitive Decline in the Elderly), ICD codes, etc. Exclusion criteria: ■ Studies enrolling patients who underwent open heart surgery or currently have underlying psychological disorders or cancer.
ADAS = Alzheimer's Disease Assessment Scale; DAD = Disability Assessment for Dementia; ICD = International Classification of Diseases; SPMSQ = Short Portable Mental Status Questionnaire; TICSm = The Modified Telephone Interview for Cognitive Status; IQCODE = Informant Questionnaire on Cognitive Decline In The Elderly; OR = odds ratio; HR = hazard ratio; CI = confidence interval, NA: Not applicable.

Table 1 .
Characteristics and outcomes of studies evaluating the association between anticoagulation and cognitive impairment among AF patients.