Prediction of post-operative atrial fibrillation in patients after cardiac surgery using heart rate variability

Abstract

Abstract Purpose Post-operative atrial fibrillation (PoAF) occurs in ~ 30% of patients after cardiac surgery. The etiology of PoAF is complex, but a disbalance in autonomic systems plays an important role. The goal of this study was to assess whether pre-operative heart rate variability analysis can predict the risk of PoAF. Methods Patients without a history of AF with an indication for cardiac surgery were included. Two-hour ECG recordings one day before surgery was used for the HRV analysis. Univariate and multivariate logistic regression, including all HRV parameters, their combination, and clinical variables, were calculated to find the best predictive model for post-operative AF. Results One hundred and thirty-seven patients (33 women) were enrolled in the study. PoAF occurred in 48 patients (35%, AF group); the remaining 89 patients were in the NoAF group. AF patients were significantly older (69.1 ± 8.6 vs. 63.4 ± 10.5 yrs., p = 0.002), and had higher CHA2DS2-VASc score (3 ± 1.4 vs. 2.5 ± 1.3, p = 0.01). In the multivariate regression model, parameters independently associated with higher risk of AF were pNN50, TINN, absolute power VLF, LF and HF, total power, SD2, and the Porta index. A combination of clinical variables with HRV parameters in the ROC analysis achieved an AUC of 0.86, a sensitivity of 0.95, and a specificity of 0.57 and was more effective in PoAF prediction than a combination of clinical variables alone. Conclusion A combination of several HRV parameters is helpful in predicting the risk of PoAF. Attenuation of heart rate variability increases the risk for PoAF.

Keywords

• Post-operative atrial fibrillation
• Cardiac surgery
• Heart rate variability
• Non-linear analysis