Background
Atrial fibrillation (AF) is a well-known risk factor for ischemic stroke. Interatrial block (IAB), resulting from a conduction delay in Bachmann's bundle, is an electrical abnormality that, especially in its advanced (A-IAB) form, has been associated with atrial fibrotic cardiopathy and remodeling, sharing a common pathologic substrate with AF. Therefore, IAB has been proposed as a marker of atrial electromechanical dysfunction that may predispose to the development of AF. We hypothesized that IAB could be a marker of AF in patients with cryptogenic stroke (CS).
Methods
We retrospectively screened a cohort of CS patients with no prior history of AF, and collected their demographic, cardiovascular risk factors and other co-morbidities, stroke severity and resultant functional disability (NIHSS and mRS scales), CHA2DS2-VASc score, baseline 12-lead ECG, transthoracic echocardiogram, and neuroimaging studies on admission. All patients underwent continuous ECG monitoring for at least 48h. Forty-eight (64%) patients underwent additional out-patient 24h Holter monitoring following hospital discharge. P-wave (PW) analysis of baseline ECG classified patients into three groups: normal p-wave duration (N, PW <120 ms), partial IAB (P-IAB, PW ≥120 ms) and advanced IAB (A-IAB, PW ≥120 ms and biphasic morphology in leads DII, DIII and aVF). AF episodes (defined as irregular supraventricular arrhythmias without detectable PWs lasting >30 s), frequent premature atrial contractions (PACs) (>1%) and atrial tachyarrhythmia episodes (runs of >3 consecutive PACs) were detected on 24h Holter readings. Ischemic brain lesions were classified as cortical, subcortical or lacunar in relation to their localization and maximal diameter.
Results
Out of the 80 consecutive CS patients identified, 5 were excluded due to poor quality ECG readings. A high prevalence of both P-IAB (n=30, 40%) and A-IAB (n=23, 30.7%) was found in the final analysis with 75 CS patients. No differences were found between demographics, cardiovascular risk factors and co-morbidities, NIHSS, mRS and CHA2DS2-VASc scores, left atrial size and left ventricular ejection fraction among the 3 ECG categories. Cortical ischemic brain lesions were more frequent in patients in the IAB groups than in the N ECG group. After a mean follow-up of 522 days, AF was clinically diagnosed in 13 patients (17%), with an increasing risk in accordance to the severity of the IAB (p<0.05) (Figure 1A). 24h Holter readings showed greater frequency of PACs and atrial tachyarrhythmia episodes in patients with IAB (p<0.05) (Figure 1B).
Figure 1
Conclusions
In CS, IAB is associated with a greater risk to develop AF and its presence identifies a subgroup of patients with probable atrial fibrotic cardiopathy that could potentially benefit from early oral anticoagulation in secondary prevention.