Prolonged Rhythm Monitoring for the Detection of Occult Paroxysmal Atrial Fibrillation in Ischemic Stroke of Unknown Cause
by Admin | Category Article | Aug 22nd, 2016
Atrial fibrillation (AF), commonly encountered in patients with ischemic stroke and transient ischemic attack (TIA), confers a 5-fold increased risk of ischemic stroke. AF-related strokes are associated with an ≈50% increased risk of disability and 60% increased risk of death at 3 months compared with strokes of other etiologies. Paroxysmal AF (PAF), a self-terminating recurrent form of cardiac arrhythmia that comprises between 25% and 62% of AF cases, may present as a brief single episode of arrhythmia or as clusters of abnormal rhythm of variable duration, sometimes evolving into a more persistent or permanent form.4 The self-terminating nature of PAF may lead to its underdiagnosis and consequent use of less effective treatment strategies (aspirin instead of oral anticoagulation) in poststroke patients.
To address the underdiagnosis of PAF in patients with ischemic stroke and TIAs, several treatment guidelines have singled out the identification of PAF as an important goal after a stroke/TIA. The diagnosis of PAF, however, poses a challenge. Several features of AF (such as its brief duration, episodic frequency, and asymptomatic presentations) make its detection difficult and elusive to bedside screening measures, such as pulse monitoring and routine ECGs. To date, several studies have explored the use of prolonged noninvasive and invasive cardiac monitoring devices to identify AF but with variable success. After detection of AF, a cardioembolic mechanism is often inferred and anticoagulation occasionally prescribed for secondary stroke prevention. The routine use of cardiac monitoring to identify patients with PAF who will benefit from anticoagulation has been reported to be cost-effective. In this review, we provide an overview of the different methods of cardiac monitoring, summarize studies that investigated the incidence of PAF after stroke, and highlight gaps in our understanding of the pathogenic and prognostic significance of AF detected on extended cardiac monitoring after a stroke or TIA.
The Spectrum of Atrial Tachyarrhythmias and Stroke Risks
Atrial tachyarrhythmias are heterogeneous, with etiologies that include rapidly discharging foci (most often in the pulmonary veins), microreentry, macroreentry, and autonomic modulation of the atria, all of which occur in atria with varying degrees of structural abnormality. Because the surface ECG is acquired by recording from sites removed from the atria, it is an integrative recording that “summarizes” the degree of uniformity of atrial activation. Thus, some arrhythmias that may be classified as AF on the surface ECG may have some degree of organization regionally within the atrial myocardium. During AF, intracardiac recordings demonstrate wide regional variation, with complex fractionated electrograms and organized electrograms temporally coexisting. The relationship between atrial arrhythmia rate and complexity and thromboembolic risk is not well established.
Whereas some studies suggest that the stroke risk with atrial flutter (an organized, macroreentrant arrhythmia) is similar to that of AF, others suggest a continuum of stroke risk that increases with greater atrial rate and disorganization. The putative mechanisms of thromboembolism in AF include development of mechanical dysfunction and a proinflammatory and procoagulant state, with resultant thrombus formation in the complex, pectinate-rich structure of the left atrial appendage (LAAalthough the possibility that AF is merely a marker for the presence of stroke risk factors exists. Moreover, the LAA demonstrates great interindividual morphological heterogeneity.2 Whether this anatomic variability affects the risk of thromboembolism associated with atrial tachyarrhythmias is not known.
Evidence exists to indicate that the duration of an atrial tachyarrhythmia predicts its thromboembolic risk. Conversion of AF to sinus rhythm results in transient mechanical dysfunction of the LAA, the duration of which is a function of the length of the antecedent AF episodeThus, although incompletely defined, factors that may play a role in thromboembolism in AF include the rate and duration of the atrial tachyarrhythmia, LAA morphology, and the presence of established risk factors (advanced age, heart failure, diabetes mellitus, hypertension, and previous TIA or stroke) . However, the minimum atrial rate and duration that are thrombogenic are not known. Current guidelines recommend anticoagulation for nonrheumatic AF on the basis of the presence of clinical risk factors, without regard to whether the AF is paroxysmal or permanent; furthermore, they propose treatment of atrial flutter and AF in an identical manner with regard to anticoagulation (class I, level of evidence C).7 For cardioversion, anticoagulation is recommended for episodes lasting more than 48 hours, although evidence to support this strategy is lacking.7 AF ablation guidelines suggest considering any episode lasting at least 30 seconds as a recurrence.
Detecting Paroxysmal Atrial Fibrillation in Stroke Patients
Brief and predominantly asymptomatic presentations of AF (often referred to as occult PAF) may remain undetected by traditional methods of arrhythmia screening. Subsequent to earlier reviews in 2000 and 2007 numerous studies have been published that assess the use of cardiac monitoring devices to detect PAF in stroke patients. The incidence of new AF varies widely depending on the choice of cardiac monitoring devices, recruitment criteria of the study population, stroke characteristics, interval of monitoring from stroke onset, and duration of cardiac monitoring.
We performed a systematic review of studies that assessed the incidence of newly diagnosed PAF after stroke. For inclusion, studies were required to fulfill the following criteria: (1) consecutive recruitment of stroke or TIA subjects, (2) cardiac monitoring to detect AF for a minimum duration of 12 hours; and (3) detection of new AF with the use of either a noninvasive or invasive cardiac monitoring device. We identified 19 studies (6 prospective and 13 retrospective) that assessed the ability of cardiac monitoring devices to identify AF in consecutive patients after ischemic stroke, TIA, or both). The methods of cardiac monitoring in these studies were ambulatory ECG, continuous inpatient ECG, combined use of continuous and ambulatory ECG monitoring, automatic event recorder, and implantable loop recorder. Studies varied in their recruitment of subjects for cardiac monitoring. Some studies included subjects on the basis of clinical suspicion of embolic stroke, older age, or unknown stroke mechanism despite extensive investigations (cryptogenic strokes). In some studies, subjects were selected after additional screening for arrhythmia before prolonged cardiac monitoring. The duration of cardiac monitoring also differed significantly across studies, from 24 hours to as long as 14 months.
Significance of Paroxysmal Atrial Fibrillation Detected After Stroke
AF may lead to stroke due to diminished blood flow within the complex anatomy of the trabeculated LAA, promoting thrombus formation by associated inflammation and activation of thromboxane and other rheostatic factors. Whether PAF (with intervening sinus rhythm that may promote thrombus dislodgement via LAA contraction) leads to an elevated risk or lower risk (because of less time in fibrillation) is poorly understood. In patients with PAF determined by standard surface ECG tracings, the risk of stroke in patients with PAF is similar to that observed with chronic and persistent forms of AF, and current guidelines recommend treating PAF on the basis of the concomitant stroke risk factors in a manner identical to persistent forms with regard to stroke prophylaxis. However, new technology now allows identification of very brief episodes of PAF, the significance of which is uncertain. It is not known, for example, whether patients with PAF included in previous trials could represent a selected population with high AF burden (sufficient to be detected by routine ECG) resulting in a stroke risk comparable to that of patients with chronic AF. Whether the very brief episodes of PAF detected by prolonged rhythm monitoring are associated with a similar stroke risk remains to be established. Still, growing data support an independent association between these episodes of PAF and history of cerebral ischemia or future risk of stroke. The brief episodes of PAF could directly contribute to atrial thrombus formation or represent markers of longer episodes of PAF that occur outside of the monitoring period.
Studies of patients with implanted devices provide emerging evidence that very brief arrhythmia episodes are associated with stroke risk. In the Mode Selection Trial (MOST), atrial high-rate episodes lasting at least 5 minutes predicted a higher incidence of the composite outcome of death and nonfatal stroke.65 In the Prospective Study of the Clinical Significance of Atrial Arrhythmias Detected by Implanted Device Diagnostics (TRENDS), episodes of atrial tachyarrhythmia/AF were detected by a pacemaker or implantable cardioverter-defibrillator in 28% of patients with previous thromboembolic events and an indication for implantation of the device. In this cohort, a threshold of atrial tachyarrhythmia/AF burden more than 5.5 hours on any of the preceding 30 days of monitoring was associated with an annualized thromboembolic rate of 2.4% (95% confidence interval, 1.2% to 4.5%). The rate of thromboembolic events was low compared with that in patients with AF diagnosed by traditional modalities and similar CHAD2 scores, but the risk was still doubled in patients with an atrial tachyarrhythmia/AF burden more than 5.5 hours compared with those with no atrial tachyarrhythmia/AF (P=0.06).
Evidence for the importance of very brief episodes of atrial tachyarrhythmias comes from the Asymptomatic Atrial Fibrillation and Stroke Evaluation in Pacemaker Patients and the Atrial Fibrillation Reduction Atrial Pacing Trial (ASSERT) (S.J. Connolly, MD, unpublished data, 2010). Hypertensive patients aged ≥65 years with a clinical indication for a dual-chamber pacemaker or implantable cardioverter-defibrillator and no history of AF were enrolled (n=2580). Atrial high-rate episodes were defined as an atrial rate more than190 bpm lasting more than 6 minutes and were adjudicated by expert clinicians. Patients were prospectively followed for stroke or thromboembolism for a mean of 2.8 years, and the cumulative rate of vitamin K antagonists was less than 2% per year. Remarkably, despite the lack of a clinical history of AF, 36% of patients had an atrial high-rate episode during the study. The presence of an atrial high-rate episode conferred a relative risk of 2.49 for subsequent ischemic stroke or systemic embolism. The study has been presented but not yet published; additional insights are certain to emerge once the analysis of these data is completed.
Another study showed that brief bursts of PAF detected by Holter (mean monitoring time 22.6 hours; bursts often less than 30 seconds) were associated with the presence of acute and chronic brain infarcts on brain imaging, especially cortical lesions consistent with embolism. High detection rates have been reported with the use of 21-day monitoring with mobile cardiac outpatient telemetry, but most of these episodes are very brief. In 1 study, 5.3% of cryptogenic stroke patients had episodes of PAF lasting more than 30 seconds, and 23% had episodes lasting less than 30 seconds.
When patients with previous stroke are monitored, it is pertinent to ask if the observed arrhythmia could be a consequence (rather than a cause) of cerebral ischemia. ECG abnormalities, including atrial arrhythmias, have been reported in patients with no underlying cardiac disease. There is considerable evidence indicating that the insular cortex has a substantial role in the regulation of cardiac rate and rhythm and supporting the concept that insular infarctions can be arrhythmogenic. Studies reporting a significantly higher incidence of atrial tachyarrhythmias immediately after an ischemic stroke, especially in patients with total anterior circulation infarction, support the hypothesis of a cerebrogenic source of the cardiac arrhythmia. It remains to be determined whether brief episodes of PAF can be a transient phenomenon after an insular stroke. Moreover, pooled data from the Stroke Prevention in Atrial Fibrillation (SPAF) studies indicate that ≈30% of strokes in AF patients are noncardioembolic in etiology
When Should Paroxysmal Atrial Fibrillation Be Treated?
Once PAF is identified by traditional modalities, oral anticoagulant therapy is indicated to prevent stroke recurrence. However, the optimal management of very brief episodes detected by newer technologies is not known. In some studies, the detection of AF with prolonged monitoring devices after a stroke led to a change in the choice of antithrombotic treatment. In 1 study, oral anticoagulation was started in 28.6% of patients with a new diagnosis of AF detected by ambulatory 7-day ECG monitoring and in all 5 patients with AF detected by loop event recorder. However, caution should be exercised when one extrapolates the potential benefits of anticoagulation to stroke patients with PAF detected by prolonged monitoring with the use of novel devices. At present, there are no data from randomized controlled trials to guide the appropriate treatment of brief episodes of PAF newly detected by prolonged rhythm monitoring after a stroke or TIA. Such trials are needed.
