Interatrial conduction disturbance in postoperative atrial fibrillation: a comparative study of P-wave dispersion and Doppler myocardial imaging in cardiac surgery

OBJECTIVE Disturbances of interatrial conduction have been proposed as one of the contributing mechanisms of postoperative atrial fibrillation (AF). P-wave dispersion has been recognized as a sensitive tool for detecting interatrial conduction disturbances. Doppler myocardial imaging (DMI) has been validated as a non-invasive tool to indirectly reflect electrical atrial activation and therefore is used in this study to detect possible interatrial electromechanical disturbances after cardiac surgery. METHODS 30 patients (23 men, age 62 ± 1 years) admitted for coronary bypass surgery with no prior history of AF were included in this investigation. Echocardiography and electrocardiograms (ECG) were obtained on the day before and after surgery. In addition to standard echocardiography, DMI-loops were acquired from the apical window. The following time intervals were derived off-line from the free right atrial (RA), left atrial (LA) lateral and LA posterior wall: onset P-wave to start (P to A'start), to peak (P to A'peak) and to end of atrial deformation (total electromechanical activity). These intervals were compared to each other and to P-wave dispersion derived from the recorded ECGs. RESULTS All patients were in sinus rhythm during their postoperative assessment, but 11 patients presented episodes of AF within the first three postoperative days. Atrial electromechanical activation was earliest in the RA and latest in the lateral LA. In patients with AF, P-wave dispersion was significantly prolonged postoperatively (mean: +18.6 ms; 95% confidence interval (CI): 12.1-25.2 ms; p < 0.001) compared to non-AF patients (mean: -2.4 ms; CI: -6.6-1.9 ms). P dispersion was closely correlated to P to A'start intervals (from RA to LA lat.: preop.: rho = 0.74, postop.: rho = 0.87; p < 0.001). Prolonged right to left conduction interval was associated with an elevated risk for AF (from RA to LA lat.: odds ratio 1.13 (CI:1.03-1.24); p: 0.007. CONCLUSION DMI enabled detection of interatrial conduction disturbances in concordance to findings of prolonged postoperative P-wave dispersion. Equally effective to P-wave dispersion, this simple and reproducible tool might help to early identify the risk for postoperative AF, thus extending the informative value of routine postoperative echocardiography.