Scanning of paroxysmal atrial fibrillation as an etiological risk factor in patients with acute ischemic stroke: prospective study

Abstract BACKGROUND: Prevention of recurrence of stroke depends on recognition of the underlying mechanism of ischemia. OBJECTIVE: To screen patients who were hospitalized with diagnosis of acute ischemic stroke in terms of atrial fibrillation (AF) with repeated Holter electrocardiography recordings. DESIGN AND SETTING: Prospective study conducted at Konya Education and Research Hospital, Turkey. METHODS: Patients with a diagnosis of acute ischemic stroke, without atrial fibrillation on electrocardiography (ECG), were evaluated. Their age, gender, histories of previous ischemic attack, occurrences of paroxysmal atrial fibrillation (PAF) and other risks were assessed during the first week after acute ischemic stroke and one month thereafter. ECG recordings were obtained from 130 patients through 24-hour ambulatory Holter. Patients without PAF attack during the first Holter were re-evaluated. RESULTS: PAF was detected through the first Holter in 33 (25.4%) out of 130 acute ischemic stroke patients. A second Holter was planned for 97 patients: 53 (54.6%) of them could not attend due to COVID-19 pandemic; while 44 (45.3%) patients had the second Holter and, among these, 4 (9.1%) had PAF. The only parameter associated with PAF was older age. Four (10.8%) of the 37 patients with PAF had also symptomatic carotid stenosis. CONCLUSIONS: Detecting the presence of PAF by screening patients with no AF in the ECG through Holter ECG examinations is valuable in terms of changing the course of the treatment. It should be kept in mind that the possibility of accompanying PAF cannot be ruled out in the presence of other factors that pose a risk of stroke.


INTRODUCTION
Atrial fibrillation (AF) consists of atrial arrhythmia characterized by loss of P waves on electrocardiography, with one or more attacks for at least 30 seconds. Although AF is more common in the elderly or individuals with other cardiovascular risk factors, it is the most common type of arrhythmia, affecting approximately 3% of the adult population. Symptoms associated with AF can be observed in all of its subtypes, seen as paroxysmal, persistent or permanent AF. However, some patients with AF associated with stroke during or after stroke are also asymptomatic. It is known that AF increases the risk of ischemic stroke four to fivefold. 1 It is important to reveal the etiology of stroke and to reduce the risk of stroke recurrence through putting appropriate treatment options into effect. However, in approximately a quarter of all ischemic strokes, the underlying factor cannot be revealed. 2 For example, in the TOAST classification, this group is called cryptogenic stroke.
With developments in the field of neuroradiology and cardiological examination, and widespread access to these examinations, the definition of cryptogenic stroke has been questioned as a result of clarification of the etiology. This has been achieved through advanced examination methods among some of the patients who had previously been diagnosed as presenting cryptogenic stroke. Thus, the term 'Embolic Stroke of Undetermined Source' (ESUS) has been introduced. 3 ESUS is held responsible for 20% of ischemic strokes. 4 By definition, ESUS consists of a non-lacunar brain infarction with no demonstrable proximal arterial stenosis or cardioembolic source, and with a clear indication for anticoagulation. 1

METHODS
The protocol for this prospective study was approved by the Patients for whom it was planned to continue their post-discharge check-ups at an external center and patients whose general condition was bad enough to require monitoring in intensive care were not included in the study group. The examination and treatment plans for any patients who were hospitalized in the neurology clinic with a diagnosis of acute ischemic stroke but did not want to be included in the study were arranged by the neurologist who followed these patients.  Table 2).
The median value on the National Institutes of Health Stroke Scale (NIHSS) was 6 (IQR 4-9). No significant difference was found between a group with NIHSS 0-4 and a group with NIHSS 5 and above, in terms of PAF relationship.
Thrombolytic therapy and/or thrombectomy were applied to all patients who were evaluated in the hyperacute period and for whom these were indicated. In 14 patients, only intravenous thrombolytic therapy was applied, in three patients only thrombectomy was applied and in four patients intravenous thrombolytic therapy and thrombectomy were applied. There was no significant relationship between the patients who underwent thrombolytic therapy and/or thrombectomy and the presence of PAF.

DISCUSSION
If persistent AF is detected through routine ECG evaluation during stroke, it is easy to demonstrate its relationship with stroke. However, it may not always be easy to detect short-term episodes of paroxysmal atrial fibrillation that are expected to end spontaneously within seven days. Some of the cases that are not evaluated with adequate examinations can be considered as included in the cryptogenic stroke group. 7 Although the association of persistent atrial fibrillation with stroke is better known, paroxysmal atrial fibrillation is also blamed for the same risk of ischemic stroke as persistent atrial fibrillation and as a potential source of cryptogenic stroke. 8   Conditions that are known to be risk factors for both AF and stroke, such as age, male gender, hypertension, diabetes mellitus, valvular heart disease, heart failure, coronary heart disease, chronic kidney disease, inflammatory disorders, sleep apnea and tobacco use, have been shown to be responsible for the association between AF and stroke. 10 In our study, no significant associations between presence of PAF and any factors other than advanced age were found.
About a quarter of strokes are recurrent. 11 In our study, no sig- if patients present nonvalvular AF with a CHA2DS2-VASc score of 2 or more. 13 Since all the patients included in our study received a score of 2 points only because of ischemic stroke and this score increased in the presence of other risk factors, patients with AF are considered to be the patient group that will benefit from anticoagulant treatment. It is vital to reveal AF through Holter monitoring in these patients.
Since the 1960s, Holter monitoring has been the cornerstone for diagnosing suspected arrhythmias in patients of all ages. The length of the recording in the most commonly used monitoring systems is limited to 24-48 hours, while newer Holter monitors allow continuous electrocardiogram recording for two weeks.
Prolonging the ECG recording time will increase the diagnostic efficiency of Holter monitoring, especially for rare but recurrent rhythm disorders. 14 However, long-term monitoring has disadvantages such as reduced patient compliance and increased cost. Another issue that we want to draw attention to in this study is the possibility of overlooking PAF if it is not evaluated through further examination, in the presence of other risk factors. Most of the chronic diseases that are considered to be risk factors for stroke, such as essential hypertension and congestive heart failure, are also closely related to atrial fibrillation. Therefore, their presence may play a role in the etiology of stroke, but is not sufficient to rule out the risk of PAF.
In our study, simultaneous PAF was detected in four of the 13 patients with symptomatic carotid stenosis detected through CT angiography and MRI angiography. This patient group was considered to present large-artery atherosclerosis according to the TOAST classification. Although the recommended Holter monitoring time for cases of cryptogenic stroke and ESUS was extended to 72 hours in the updated guidelines, 6 the possibility that stroke might be multifactorial was ignored and no recommendations were made for these patients.

CONCLUSION
Detecting the presence of PAF by scanning patients who did not show AF on ECG, through Holter ECG examination, is valuable in terms of changing the course of the treatment. It should be kept in mind that the possibility of accompanying PAF cannot be ruled out in the presence of other factors that pose a risk of stroke.