Usefulness of P‐wave duration in patients with sick sinus syndrome as a predictor of atrial fibrillation

Abstract Background This study aimed to clarify P‐wave duration (PWD) ability before pacemaker implantation to predict worsening atrial fibrillation (AF) burden after the procedure. Methods We retrospectively investigated 75 patients who underwent permanent pacemaker implantation due to sick sinus syndrome (SSS) at Komaki City Hospital between January 2006 and May 2019. Worsening AF burden was defined as an increase in the number of AF episodes, each lasting ≥5.5 hours a day. Results In the study population, 17 patients (23%) had worsening AF burden during the follow‐up period. These patients had significantly longer PWD in lead Ⅱ (117.9 ± 19.9 ms vs 101.3 ± 20.0 ms, P = .002) than the patients without worsening AF burden. The best discriminative cutoff value for PWD in lead Ⅱ was 108 ms (sensitivity, 77%; specificity, 67%). In multivariate analysis, PWD in lead II ≥108 ms (hazard ratio, 5.395; 95% confidence interval, 1.352‐21.523; P = .017) was an independent predictor of worsening AF burden. Patients with PWD in lead II <108 ms showed a significantly higher event‐free rate against worsening AF burden than those with PWD in lead II ≥108 ms (81% vs 9%, P = .005). Conclusions Prolonged PWD before pacemaker implantation was the most important independent predictor of worsening AF burden after the procedure. In patients with SSS, prolonged PWD can be a useful marker for predicting worsening of AF burden after pacemaker implantation.

quality of life. 2,3 AF is frequently observed in patients with permanent pacemakers, and it carries the risk of heart failure hospitalization and stroke. 4,5 Managing AF is equally essential in patients with pacemakers and those without pacemakers. In patients with pacemakers, a higher percentage of ventricular pacing increased the risk of AF occurrence. 6 Therefore, we attempted to program AV delay prolongation to avoid the high percentage of ventricular pacing in patients maintained by atrioventricular conduction. However, AF episodes were occasionally detected in patients with a lower percentage of ventricular pacing. Thus, the risk factors of AF occurrence are still unclear in patients with pacemakers.
Atrial structural and electrical remodeling are essential factors in the pathogenesis of AF. Atrial remodeling progression causes atrial conduction heterogeneity, 7 which manifests as changes in the P-wave morphology on electrocardiogram (ECG). P-wave duration (PWD) has been demonstrated to be a reliable and noninvasive marker for predicting the incidence of AF. 8 Kaypakli et al 9 reported that prolonged PWD was associated with AF recurrence after cryoballoon ablation.
This study aimed to clarify the predictive ability of PWD before pacemaker implantation on worsening AF burden after the procedure.

| Patient population
We retrospectively investigated 75 patients who underwent permanent pacemaker implantation due to sick sinus syndrome (SSS) at Komaki City Hospital between January 2006 and May 2019.
Pacemaker implantation was performed for patients with symptomatic SSS, such as sinus bradycardia, sinoatrial block, sinus arrest, or bradycardia-tachycardia syndrome. Exclusion criteria were as follows: (1) second or third-degree atrioventricular block, (2) persistent or permanent AF, (3) junctional rhythm, (4) history of ventricular tachycardia, (5) intake of antiarrhythmic drugs before pacemaker implantation, (6) previous catheter ablation or prior heart surgery, (7) severe valvular heart disease, and (8) left ventricular ejection fraction <35%. Informed consent was obtained from all patients before the procedure, in accordance with our institutional guidelines. This study was performed in accordance with the Declaration of Helsinki.

| Electrocardiographic assessment
Twelve standard surface ECG leads were recorded before the procedure in all patients. Patients who received antiarrhythmic drugs before pacemaker implantation were excluded. Therefore, all ECGs were recorded without the influence of antiarrhythmic drugs. The ECG was digitally recorded with a paper speed and scale at 25 mm/s and 10 mm/mV, respectively (ECG-2550; Nihon Kohden). The PR interval, PWD, and P-wave amplitude were measured manually using a digital caliper in leads V1 and Ⅱ. The P-wave duration index (PWDI) was calculated by dividing the PWD by the PR interval. The P-wave was between the initial upward or downward point from the isoelectric line and the returning point to the isoelectric line. The isoelectric line was defined as the beginning of the P-QRS complex to the end of the T-wave ( Figure 1).

| Procedure of pacemaker implantation
The indication for pacemaker implantation was symptomatic SSS. The pacemaker devices used were manufactured by Medtronic, Inc, St.
Jude Medical, Inc, or Abbott, Inc The right atrial lead was placed in the right atrial appendage, and the right ventricular lead was placed in the low-septum or apex. Devices were programmed with pacing mode DDD and prolonged atrioventricular delay, managed by ventricular pacing (MVP TM , Medtronic) or ventricular intrinsic preference (VIP TM , St. Jude Medical or Abbott) mode to minimize ventricular pacing.

| Patient follow-up
The patients were hospitalized under continuous rhythm monitoring for 4 days after the procedure. After hospital discharge, all patients were scheduled for follow-up visits. Device interrogations were performed 1, 6, and 12 months after pacemaker implantation, and then every 6 months. During device interrogations, atrial/ventricle lead parameters, percentages of atrial and ventricular pacing, automatic mode switch episodes, the burden of AF episodes, and noise episodes were recorded. If patients noticed any rhythm disorders in between follow-up visits, they were recommended to arrange an early visit to the hospital for device interrogation. Worsening AF burden was defined as an increase in the number of AF episodes, with each episode lasting ≥5.5 hours a day. 10 The increase in the number of AF episodes was the comparison of the number of AF episodes in the first follow-up visit after pacemaker implantation and in the last follow-up visit. Patients received antiarrhythmic drugs, catheter ablation, or anti-tachycardia pacing (ATP), if necessary.

| Statistical analysis
Continuous variables are presented as mean ± standard deviation. Categorical variables are presented as percentages. A chisquare test was performed to compare categorical variables, and a F I G U R E 1 Schematic representation of P-wave duration in lead V1. A is the P-wave duration, and B is the PR interval Mann-Whitney U test was performed to compare continuous variables. In this study, we used the receiver-operating characteristic (ROC) curve analysis to determine the cutoff value. This method calculates the distance between the point (0, 1) and the point of cutoff value defined as the point on ROC curve where the distance is at a minimum. The factors shown to have a P-value of <.05 in the univariate analysis were further assessed using multivariate analysis. The event-free survival rate was estimated using the Kaplan-Meier method and compared to the recurrence rate using a log-rank test. Statistical analyses were performed using SPSS version 25 (SPSS Inc). A P-value of <.05 was considered statistically significant.

| Patient characteristics
A comparison of the baseline demographic and clinical characteristics between the patients with and without worsening AF burden is presented in Table 1. In the study population, 17 patients (23%) had worsening AF burden during the follow-up period. The patients with worsening AF burden had a significantly higher age than the patients without AF burden (79 ± 6 years vs 74 ± 11 years, P = .016).
They also had a higher proportion of hypertension patients (59% vs 31%, P = .037) and history of AF (88% vs 53%, P = .01). In addition, AF burdens in the first follow-up visit after pacemaker implantation TA B L E 1 Comparison of baseline demographic and clinical characteristics between the patients with and without worsen AT/AF burden in the study population showed significantly higher in the patients with worsening AF burden (23.0 ± 36.9% vs 5.0 ± 12.2%, P < .001). In terms of echocardiographic parameters, the left ventricular ejection fraction (LVEF) was significantly lower in the patients with worsening AF burden (61.0 ± 11.6% vs 68.3 ± 7.3%, P = .021). The left atrial diameter and left atrial volume index were similar between the two groups.
In addition, the number of patients prescribed with β-blockers after pacemaker implantation was significantly higher in the patients with worsening AF burden (71% vs 29%, P = .002). The other clinical and echocardiographic parameters, and the details of antiarrhythmic therapy, such as antiarrhythmic drugs, catheter ablation, and ATP, were similar between the two groups.

| ECG parameters before pacemaker implantation
A comparison of the ECG parameters of the P-wave indices is shown in Table 2. The PR interval, number of patients with first-degree atrioventricular block, P-wave amplitude in leads V1 and Ⅱ, PWD, and PWDI in lead V1 were similar between the two groups. In patients with worsening AF burden, the PWD in lead Ⅱ was significantly longer (117.9 ± 19.9 ms vs 101.3 ± 20.0 ms, P = .002), and the PWDI in lead Ⅱ was significantly larger (0.65 ± 0.14 vs 0.56 ± 0.12, P = .014). ROC curve analysis was performed to evaluate the correlation between PWD in lead Ⅱ and worsening AF burden after pacemaker implantation. We set the cutoff values of PWD and PWDI in lead Ⅱ to 108 ms (sensitivity, 77%; specificity, 67%; Figure 2A) and 0.52 (sensitivity, 88%; specificity, 45%; Figure 2B), respectively.

| The predictors of worsening AF burden after pacemaker implantation
Univariate and multivariate Cox regression analyses revealed the predictors of worsening AF burden after pacemaker implantation.  (Table 3). PWDI was excluded from this analysis to eliminate confounding factors. In multivariate analysis, PWD in lead II ≥108 ms (HR, 5.395; 95% CI, 1.352-21.523; P = .017) was an independent predictor of worsening AF burden (Table 4).

| The association of PWD with worsening AF burden after pacemaker implantation
A Kaplan-Meier analysis was performed to evaluate the event-free rate of patients with worsening AF burden after pacemaker implantation. Patients with PWD in lead Ⅱ <108 ms exhibited a significantly higher event-free rate than those with PWD in lead II ≥108 ms (81% vs 9%; P = .005; Figure 3).

| D ISCUSS I ON
This study aimed to demonstrate the relationship between PWD and worsening AF burden in patients with SSS. The study found that prolonged PWD before pacemaker implantation was the most important independent predictor of worsening AF burden after the procedure.

ECG can be obtained noninvasively. Previous reports have
shown that P-wave indices such as the PR interval, 11 P-wave axis, 12 and P-wave terminal force in V1 13 are related to AF. The PWD is a noninvasive marker of AF recurrence after catheter ablation. 9 Demirtas et al 14  Abbreviations: AF, atrial fibrillation; AT, atrial tachycardia; LVEF, left ventricular ejection fraction; PWD, P-wave duration.
TA B L E 4 Multivariate Cox regression analyses for worsening AT/AF burden after pacemaker implantation F I G U R E 3 Kaplan-Meier curves of the survival-free rate of worsening AF burden after pacemaker implantation between the two groups (PWD in lead Ⅱ <108 ms; PWD in lead Ⅱ ≥108 ms). AF, atrial fibrillation; PWD, P-wave duration

| CON CLUS ION
This study demonstrated the relationship between PWD in lead Ⅱ and worsening AF burden after pacemaker implantation in patients with SSS. Prolonged PWD before pacemaker implantation was the most important independent predictor of worsening AF burden after the procedure. In patients with SSS, prolonged PWD can be a useful marker for predicting the worsening of AF burden after pacemaker implantation.

CO N FLI C T O F I NTE R E S T
Authors declare no conflict of interests for this article.