Preserving Cognitive Function in Patients with Atrial Fibrillation
Tina Lin MD, Erik Wissner MD, Roland Tilz MD, Andreas Rillig MD, Shibu Mathew MD, Peter Rausch MD, Peter Rausch MD, Christine Lemes MD, Sebastian Deiss MD, Masashi Kamioka MD, Tudor Bucur MD, Feifan Ouyang MD, Karl-Heinz Kuck MD, Andreas Metzner MD
Asklepios-Klinik St. Georg, Dept. Of Cardiology, Lohmühlenstr. 5, 20099 Hamburg/Germany.
Atrial fibrillation (AF) is the most common cardiac arrhythmia worldwide and is associated with significant morbidity and mortality. Its prevalence increases with increasing age, and is one of the leading causes of thromboembolism, including ischemic stroke. The prevalence of cognitive dysfunction also increases with increasing age. Although several studies have shown a strong correlation between AF and cognitive dysfunction in patients with and without overt stroke, a direct causative link has yet to be established. Rhythm vs rate control and anticoagulation regimens have been extensively investigated, particularly with the introduction of the novel anticoagulants. With catheter ablation becoming more prevalent for the management of AF and the ongoing development of various new energy sources and catheters, an additional thromboembolism risk is introduced. As cognitive dysfunction decreases the patient’s ability to self-care and manage a complex disease such as AF, this increases the burden to our healthcare system. Therefore as the prevalence of AF increases in the general population, it becomes more imperative that we strive to optimize our methods to preserve cognitive function. This review gives an overview of the current evidence behind the association of AF with cognitive dysfunction, and discusses the most up-to-date medical and procedural treatment strategies available for decreasing thromboembolism associated with AF and its treatment, which may lead to preserving cognitive function.
Key Words : Atrial Fibrillation, Cognitive Dysfunction, Thromboembolism and Stroke, Anti-Coagulation, Catheter Ablation.
Corresponding Address : Andreas Metzner MD Dept. of Cardiology/Electrophysiology Asklepios-Klinik St. Georg Lohmühlenstr.20099 HamburgGermany
Atrial fibrillation (AF) is currently the most common sustained cardiac arrhythmia worldwide with an overall prevalence of up to 5.5%, which increases with increasing age.1-3 It is associated with significant morbidity and mortality, and is a major burden on the health care system. AF is one of the leading causes of congestive heart failure4-6 and thromboembolism, including ischemic stroke.7-16 Several trials, prospective cohort studies and meta-analyses have been conducted in recent years to assess the impact of AF on cognitive function5,6,8-19 (Table 1). These have identified AF as an independent predictor of cognitive impairment in patients with and without overt stroke to various degrees. Controversial issues of rate or rhythm control and anticoagulation have been at the forefront of research, with conflicting results in a growing body of literature leading to rapidly changing published guidelines and position papers for the management of AF.12,15,20 With the development of ablation for the treatment of AF, particularly with various new energy sources and catheters, an additional source of thromboembolism is introduced and methods to reduce this risk need to be considered. The importance lies in the fact that cognitive impairment leads to decreased ability to self-care, to manage a complex disease, and subsequently to decreased quality-of-life and increased hospitalizations.
Table 1. Prospective cohort studies and meta-analyses assessing stroke and cognitive impairment in patients with atrial fibrillation
|Study||Year||Study design||Patients||Patients with AF||Pts with AF vs no AF - findings|
|Marzona, et al16||2012||Post-hoc analysis||31546||3068||Stroke 8.5% vs 4.0%
Composite decreased MMSE, dementia 34.2% vs 26.1%|
|Santangeli, et al14||2012||Meta-analysis||77668||11700||Dementia HR 1.42, 95% CI 1.17-1.72, P<0.001|
|Dublin, et al69||2011||Prospective cohort||3045||502||Dementia HR 1.38, 95% CI 1.10-1.73|
|Bunch, et al15||2010||Prospective cohort||37025||10161||Dementia incidence 3.3% vs 1.3%, P<0.0001|
|Knecht, et al18||2008||Prospective cohort||685||122||Significantly decreased leaning + memory, P<0.01|
|Kilander, et al8||1998||Population-based||952||44||MMSE cognitive score -0.26+/-0.11 vs +0.14+/-0.03, P=0.0003
|Ott, et al10||1997||Population-based||6584||195||Dementia OR 2.3, 95% CI 1.4-3.7
Cognitive impairment OR 1.7, 95% CI 1.2-2.5 |
Here we review the evidence behind the association of AF with cognitive dysfunction, and discuss the current medical and procedural treatment strategies available for decreasing the thromboembolic complications of AF, which may in turn lead to preserving cognitive function.
Atrial Fibrillation and Cognitive Dysfunction
Atrial fibrillation is a modifiable cause of thromboembolic stroke. It increases the risk of ischemic stroke 5-fold as compared to the general population.9,10,21 Several studies have demonstrated the association between AF and the pathological findings of brain lesions, which has subsequently led to the concentration of studies investigating methods to reduce stroke risk in AF. The role of AF in cognitive decline in patients without a previous history of overt stroke is less clear. The prevalence of AF increases with advancing age,2,3,16 and up to 10% of patients >80 years will develop AF. In parallel, the prevalence of dementia also increases with advancing age,18,22,23 and both these conditions have similar risk factors including hypertension, diabetes and stroke. Whilst AF is associated with developing or worsening cognitive dysfunction post-stroke,5,6 it is still controversial whether AF increases the risk of dementia in patients without prior stroke.8-16 Several studies and meta-analyses have not only attempted to find an association between AF and cognitive decline, but have tried to identify an independent relationship, beyond stroke and hypertension.
Kilander et al8 were one of the earlier groups to show this association in elderly men. Several studies have demonstrated that AF is independently associated with dementia particularly in a younger age group (70 years), including Alzheimer’s disease and vascular dementia, with an increased rate of decline in patients already diagnosed with these dementias.12,15,20 The Rotterdam study10 was one of the largest population-based studies which contributed to identifying an association between AF and cognitive impairment, with more than 6500 patients included. Marzona et al16 have recently provided prospective evidence that AF increases the risk of cognitive dysfunction and dementia independent of overt stroke by analyzing the large ONTARGET and TRANSCEND study populations. The results from these studies have been supported by several imaging studies that have looked at MRI brain changes in AF patients without overt stroke. These have shown white matter changes, reduced brain volume and hippocampal atrophy associated with memory impairment and cognitive dysfunction.18,22,23 Although the majority of studies suggest an independent association between AF and cognitive dysfunction in patients without overt stroke, due to the multifactorial contributions to cognitive decline, several other studies have shown contradicting results, and Park et al13 performed one of the largest prospective longitudinal cohort studies demonstrating no consistent or significant change in cognitive performance in patients with AF or in sinus rhythm (SR), nor between patients treated with anticoagulation.
Atrial Fibrillation Ablation And Cognitive Dysfunction
Catheter ablation has become the cornerstone treatment in the symptomatic control of AF. Conventional radiofrequency (RF) ablation with point-by-point ablation has been developed in an attempt to provide rhythm control of AF, however current success rates for maintenance of SR in patients with both paroxysmal and persistent AF have been disappointing.24,25 Subsequent AF recurrence, particularly when asymptomatic, leads to increased stroke risk secondary to under-treatment.26,27 Catheter ablation itself is a potential iatrogenic source of clinical and sub-clinical brain embolism, with an incidence up to 22%1,28-33 (Table 2). This is because of the stunned left atrial myocardium and poor myocardial contraction leading to a prothrombotic state after AF is reverted to SR,34 as well as the risk of thrombus formation in areas of left atrial endothelial damage. In addition, introduction of air and thrombus embolism into the arterial system as well as charring on catheter electrodes can both contribute to cognitive decline, and this has been shown in several imaging studies that have demonstrated new MRI brain lesions that develop after AF ablation procedures.35-37 Studies demonstrating neuropsychological decline after catheter ablation of AF further suggest that iatrogenic embolism contributes to the impact of AF on cognitive dysfunction.1,38,39 Even so, patients undergoing AF ablation have been shown to have lower rates of mortality and cognitive dysfunction compared with medically managed AF patients,40 and this may be secondary to poorer durable rhythm control that can be achieved with medical therapy.41 In addition, certain characteristics in patients selected for catheter ablation, such as younger age and fewer co-morbidities, are associated with lower baseline mortality and cognitive impairment and contribute to lower post-procedural mortality and cognitive dysfunction compared with medically managed AF patients.
Table 2. Prospective studies assessing brain lesions and cognitive impairment in patients undergoing catheter ablation for atrial fibrillation
|Study||Year||Study design||Patients||Patients with AF||Pts with AF vs no AF - findings|
|Medi, et al1||2013||Prospective||150||90||Post-operative neurcognitive dysfunction 13-20% vs 0%|
|Herm, et al39||2013||Prospective||37||37||3-T MRI lesions at 6 mths 12.5%, no cognitive impairment |
|Schwarz, et al38||2010||Prospective||46||23||DW MRI lesions 14.3%, decreased verbal memory P<0.001|
|Sauren, et al70||2009||Prospective||30||30||Microembolic signals on transcranial doppler 6347|
|Scherr, et al31||2009||Prospective||721||721||Ischemic stroke 1.4%|
|Oral, et al30||2006||Prospective||755||755||Thromboembolic event 1.1%|
In addition to conventional RF energy ablation, there has and will continue to be further development of new energy sources as well as catheters, which could all contribute to iatrogenic embolism. Single shot devices such as cryoablation and multi-electrode ablation catheters require larger sheaths and complex preparation strategies in an attempt to reduce risk of air embolism.36,42 Recent studies have demonstrated that cryoablation for the management of AF confers a similar rate of brain embolic lesions on MRI. More alarmingly, multi-electrode ablation is associated with a 5-fold increased risk of embolic lesions.36 Further studies are required to assess the clinical impact of these findings, however extrapolation from previous studies of conventional point-by-point RF ablation is worrying. With the changing opinions in regards to the mechanism of AF initiation and maintenance, new ablation strategies such as complex fractionated atrial electrogram (CFAE) elimination and focal impulse and rotor modulation (FIRM) mean that AF procedure times may become longer. Medi et al1 recently published a potential correlation between time in the left atrium during catheter ablation (left sided supraventricular tachycardia ablation vs. AF ablation) and brain embolic lesions. The development of new technologies and ablation strategies, however warranted, should take these concerns into consideration.
Treatment in Atrial Fibrillation to Decrease Thromboembolic Complications and Preserve Cognitive Function
There are two arms in the management of AF. The first is the maintenance of rate or rhythm control. The second is prevention of complications, in particular thromboembolism.4 The use of oral anticoagulants has been in the forefront of stroke prevention, and there has been a recent expansion in the available repertoire of these medications. Vitamin K antagonists have been the traditional anticoagulant, and it has been highly effective in reducing stroke risk in patients with AF. However, due to the difficultly for physicians and patients to maintain optimal control of the International Normalized Ratio (INR) and the consequential risk of thromboembolism or bleeding with INR values that are too low or too high, respectively, novel agents such as dabigatran,7 rivaroxaban17 and the newly Food and Drug Administration (FDA) approved apixaban43 have been developed to improve the ease of anticoagulation. In addition, these agents have been shown to be comparable or superior compared to warfarin – dabigatran at a dose of 150mg was associated with lower rates of the primary outcome of stroke and systemic embolism and similar rates of the primary safety outcome of major bleeding; at a dose of 110mg, dabigatran was associated with similar rates of stroke and systemic embolism and lower rates of major bleeding; rivaroxaban was non-inferior for the primary outcome of stroke and systemic embolism and the primary safety outcome of clinically significant bleeding; apixaban was superior in preventing the primary outcomes of stroke and systemic embolism and the primary safety outcome of major bleeding.12,15,20,43-45
However, treatment with anticoagulation therapy is not without risk. It is associated with hemorrhagic stroke and brain micro-hemorrhages, which contributes to cognitive dysfunction in patients with AF. This has lead to several stroke risk stratification scores to help identify the appropriate patients for anticoagulation therapy. The CHADS2 score (Congestive heart failure, Hypertension, Age 75 years, Diabetes mellitus, previous Stroke / transient ischemic attack) and the more recent CHA2DS2-VASc score (Congestive heart failure, Hypertension, Age 75 years, Diabetes mellitus, previous Stroke / transient ischemic attack, Vascular disease, Age 65-74 years, [female] Sex category) are the most used of these risk scores due to their simplicity, however their positive predictive value is at most modest46-50 (Table 3).
Table 3. Anticoagulation and bleeding risk scores
|CHADS2 Score46||Points||Description||Score||Adjusted stroke rate (%/year)|
|C||1||Congestive heart failure||0||1.9|
|H||1||Hypertension (systolic >160mmHg)e||1||2.8|
|A||1||Age ≥75 years||2||4|
|S||2||Stroke / transient ischemic attack||4||8.5|
|CHA2DS2-VASc Score47||Points||Description||Score||Adjusted stroke rate (%/year)|
|C||1||Congestive heart failure||0||0|
|A||2||Age ≥75 years||2||2.2|
|S||2||Stroke / transient ischemic attack||4||4|
|A||1||Age 65-74 years||6||9.8|
|Sc||1||female Sex category||7||9.6|
The incidence of AF as well as overt stroke and dementia increases with age.2,3,5,6,8,10,13-16,23 In addition, these conditions share similar risk factors, in particular modifiable risk factors such as hypertension, diabetes mellitus and chronic cardiac failure.51 Moreover, other cardiovascular risk factors such as dyslipidemia and smoking play an important role in increasing the incidence of stroke itself.12 Therefore, an important aspect in the treatment of AF and the preservation of cognitive function is the management of these risk factors. Population studies suggest that diabetes mellitus is an independent risk factor for AF as well as stroke.52,53Rigorous management all modifiable cardiovascular risk factors including hypertension and diabetes is therefore logical in reducing the incidence of AF and cognitive dysfunction.
The issue of rate or rhythm control has resulted in several prospective cohort studies and meta-analyses that have been performed over the last 20 years.8,10,14,16,18,19,41,54,55 Recent studies have shown a reduction in mortality with well-managed rhythm control, with lower rates of stroke and transient ischemic attack,9,10,21,56 and several recent smaller randomized trials comparing catheter ablation vs. oral anti-arrhythmic therapy demonstrated superiority with catheter ablation.2,3,16,57-59 The currently recruiting Catheter Ablation vs. Antiarrhythmic Drug Therapy for AF (CABANA) trial will provide insight into which method is superior.18,22,23,60
As the control of AF plays an important role in reducing risk from AF, rhythm control with catheter ablation has therefore become a primary treatment option for the management of AF. In addition, patients with AF are at particular risk of thromboembolism peri-procedurally, as well as for several weeks post-ablation. The latest HRS/EHRA/ECAS expert consensus statement on catheter and surgical ablation of AF recommend several management steps of particular importance.13,28 The first is that patients with AF lasting more than 48-hours require a minimum of three weeks of therapeutic anticoagulation or transesophageal echocardiography to exclude left atrial appendage thrombus prior to ablation, then for anticoagulation to be continued for two months post ablation. In regards to anticoagulation during the procedure, traditionally vitamin K antagonists were stopped prior to ablation and “bridged” with intravenous or low molecular weight heparin. However due to the increased incidence of bleeding complications26,27,61,62 there is a trend toward continuing vitamin K antagonists throughout the procedure, and antagonizing this with fresh frozen plasma, prothrombin complex or recombinant activated factor VII in the event of significant, uncontrolled bleeding.28-33,63 Clinical experience with the novel agents is still limited. During the procedure, heparin to maintain a target activated clotting time (ACT) between 300 and 400 seconds is recommended, and that heparinized saline is infused through the transseptal sheaths to reduce the risk of thrombus formation on the transseptal sheath and the catheter.34,64-69
An alternative to life-long anticoagulation has been the use of left atrial appendage closure devices.35-37,70-72 This can be considered in patients intolerant to anticoagulation, and recent evidence suggests that these devices do not require a period of “transition” anticoagulation after implantation. The PROTECT-AF investigators have demonstrated not only non-inferiority, but also improved secondary outcomes of quality-of-life in patients who received a left atrial appendage closure device.
Post-procedural follow-up of AF ablation include monitoring for and minimizing recurrence, and again stroke prevention. Minimum follow-up of three months, then every six months thereafter for at least two years is recommended, with routine ECGs at these visits, 24-hour Holter monitoring every six months and symptom-driven event monitoring, as asymptomatic AF episodes confer an increased unidentified stroke risk. Due to this, ongoing use of anticoagulation post-procedure is recommended indefinitely in patients with high CHADS2 and CHA2DS2-VASc risk scores. In patients with low risk scores, anticoagulation could be ceased, however regular screening should be performed to assess for silent AF.
It is clear that the management of AF and its complications is still a developing field, and further prospective multicenter studies are required to determine best management strategies to decrease AF complications such as ischemic stroke, subclinical thromboemboli and cognitive dysfunction. As the prevalence of AF increases, it is more imperative that we optimize our methods to preserve cognitive function.