Abstract
Background
The predictive efficacies of parameters related to P‐wave amplitude (PWA) for atrial fibrillation (AF) recurrence after catheter ablation are unclear.
Methods
We measured multiple PWA parameters using an automated system in 126 consecutive patients with persistent and long‐standing persistent AF who underwent catheter ablation. The relationships between AF recurrence and various PWA parameters were examined, including the association with P‐wave vector magnitude (calculated as the square root of the sum of lead II PWA squared, lead V6 PWA squared, and a one‐half lead V2 PWA squared).
Results
Atrial fibrillation did not recur in 87 patients (69%) during 32 ± 15 months of follow‐up. The maximum PWA, mean PWA, and P‐wave vector magnitude were lower in patients with AF recurrence than those without (maximum PWA, 0.14 ± 0.05 mV vs. 0.16 ± 0.05 mV, p = 0.017; mean PWA, 0.05 ± 0.02 mV vs. 0.06 ± 0.02 mV, p = 0.003; P‐wave vector magnitude, 0.09 ± 0.03 mV vs. 0.13 ± 0.04 mV, p < 0.001). A multivariate Cox regression analysis revealed that the predictive ability of P‐wave vector magnitude for AF recurrence was independent of other clinical properties (hazard ratio: 0.153, 95% confidence interval: 0.046–0.507, p = 0.002). Atrial fibrillation freedom rates of patients with P‐wave vector magnitude higher and lower than 0.13 mV were 93% and 57%, respectively. P‐wave vector magnitude weakly correlated with left atrial dimension (R = −0.280, p = 0.004).
Conclusions
P‐wave vector magnitude can predict AF recurrence after catheter ablation in patients with persistent AF.
Keywords: atrial conduction heterogeneity, atrial fibrillation, catheter ablation, P‐wave duration, surface electrocardiogram
1. INTRODUCTION
Since the first report revealing the essential role of pulmonary vein triggers in atrial fibrillation (AF) (Haïssaguerre et al., 1998), catheter ablation has become one of the primary treatments for paroxysmal AF. Furthermore, recent technological advances have extended the indication of catheter ablation to persistent AF (January et al., 2014); however, the long‐term success rate is still unsatisfactory (Ganesan et al., 2013; Verma et al., 2015). Therefore, prediction of AF recurrence risk is important for postprocedure care, especially for patients with persistent AF.
Atrial fibrillation is a final common end point of atrial remodeling caused by cardiac diseases, and AF itself causes remodeling that contributes to the progressive nature of the arrhythmia (Nattel & Harada, 2014). The main histological changes in atrial remodeling are loss of myocardium and interstitial fibrosis (Burstein & Nattel, 2008). These changes cause a decrease in the magnitude of atrial electrical activation, which is manifested by a decrease in the voltage of the atrial electrogram (Malcolme‐Lawes et al., 2013). Furthermore, low voltage of the atrial electrogram is associated with AF recurrence after catheter ablation (Vlachos et al., 2017).
A correlation has been reported between P‐wave amplitude (PWA) and voltage of the atrial electrogram (Park et al., 2016). Thus, PWA may indicate the severity of atrial remodeling. In addition, PWA may be a predictor of AF recurrence after catheter ablation in patients with persistent AF. However, while previous studies have revealed a relationship between PWA and AF development in patients with paroxysmal AF (Kizilirmak et al., 2016; Park et al., 2016), this relationship has not been assessed in patients with persistent AF. Furthermore, PWA may not accurately reflect the magnitude of atrial electrical activation because direction of propagation influences the PWA measured at each electrocardiographic lead. On the other hand, P‐wave vector magnitude (Cortez, Barham et al., 2017; Cortez, Baturova et al., 2017) , which indicates the P‐wave magnitude in three‐dimensional space according to vectorcardiographic principles, may be a more accurate reflection of atrial electrical activation because it is not affected by direction. Thus, the present study aimed to assess the predictive efficacies of PWA parameters including P‐wave vector magnitude for AF recurrence after catheter ablation of persistent AF.
2. METHODS
2.1. Study population
In the present study, we retrospectively enrolled 126 consecutive patients with persistent and long‐standing persistent AF who underwent catheter ablation at the Toyama University Hospital (Toyama, Japan) from January 2010 to December 2016. Persistent AF was defined as AF lasting ≥7 days but <1 year, and long‐standing persistent AF as continuous AF lasting ≥1 year (Calkins et al., 2018). We excluded patients with previous ablation, prior heart surgery, thyroid diseases, and pulmonary diseases. Clinical characteristics including comorbidities, echocardiographic data, and laboratory data were obtained from medical records. This study protocol was approved by the Research and Ethics Committee of the University of Toyama (Toyama, Japan) and conducted in accordance with the principles of the Declaration of Helsinki. All patients provided written informed consent.
2.2. Catheter ablation
All antiarrhythmic drugs were discontinued for at least five half‐lives before catheter ablation, and no patients received oral amiodarone before catheter ablation. We used either the NavX System (St. Jude Medical Inc., St. Paul, MN, USA) or the CARTO System (Biosense Webster, Inc., Diamond Bar, CA, USA) for three‐dimensional mapping. Intravenous dexmedetomidine was used for sedation, and thiopental sodium was added depending on the patient's sedation level. Sheath introducers were inserted through the right femoral vein. In addition, the trans‐septal procedure was performed, and three 8‐F SL0 sheaths (St. Jude Medical, Inc.) or two 8‐F SL0 sheaths and a steerable sheath (Agilis; St. Jude Medical, Inc.) were advanced into the left atrium. Pulmonary vein isolation was performed for all patients. Furthermore, additional radiofrequency application targeting the complex fractionated atrial electrogram (CFAE; n = 44) or left atrial posterior wall isolation with mitral isthmus line ablation (n = 52) was performed if the AF did not terminate or was induced with atrial burst pacing following pulmonary vein isolation.
For ablation targeting the CFAE, we performed fractionation analyzes using the NavX system. The mapping parameter (CFAE‐mean) was defined as an interval‐analysis algorithm that evaluated the average index of the fractionation. Moreover, continuous CFAE was defined by an average fractionated index ≤50 ms, suggesting the high temporal stability of the fractionated electrograms maintaining AF (Lin et al., 2009). We targeted all continuous CFAE sites for ablation.
For left atrial posterior wall isolation, we created the left atrial roof line at the most cranial aspect of the left atrial roof and a floor line joining the most inferior margin of the pulmonary vein isolation line. If the left atrial posterior wall could not be isolated, we performed additional radiofrequency applications targeting the earliest activation site on the isolation lines. Entrance block was confirmed by voltage mapping using a 3D mapping system and exit block using high‐output pacing within the left atrial posterior wall isolation line. If AF continued despite additional ablation, external cardioversion was performed. The procedure was completed after cavotricuspid isthmus ablation. Of note, each radiofrequency application was performed for 30–50 s. Temperature was maintained at 42°C and power at 30 W except that a maximum power of 25 W was used while delivering energy to sites near the esophagus.
2.3. Assessment of surface 12‐lead electrocardiogram
A standard 12‐lead electrocardiogram was recorded on the day following ablation in a physiological laboratory. The electrocardiogram was recorded with a paper speed of 25 mm/s and a gain setting of 0.1 mV/mm using digital electrocardiograph (Fukuda Denshi, Tokyo, Japan). The quality of electrocardiographic recording was confirmed before measurement of PWAs. The PWA was measured automatically using a computerized measurement system (Fukuda Denshi). Detection of the P wave was performed from the onset point of the QRS complex toward the offset point of the preceding T wave. The detection threshold of the P wave was decided according to the baseline noise level. The offset point of the P wave was defined as the point where the amplitude of the potential first exceeded the detection threshold, and the onset point was defined as the point where the amplitude of the potential fell below the detection threshold. Furthermore, the onset point of the P wave was reexamined in the opposite direction. The PWA was measured from baseline to the peak of the positive deflection or the nadir of the negative deflection. In biphasic P waves, the PWA was measured by adding the amplitude of the positive deflection and that of the negative deflection.
To assess the magnitude of atrial electrical activation, we evaluated the highest PWA among 12 electrocardiographic leads (maximum PWA) and the average value of PWA of all 12 electrocardiographic leads (mean PWA). Moreover, we adopted the P‐wave vector magnitude as an indicator of atrial electrical activation magnitude. P‐wave vector magnitude was calculated from the PWA measured from orthogonal electrocardiographic leads II, V2, and V6. The root‐mean‐square formula is based on Kors' quasi‐orthogonal transformation (Cortez, Barham et al., 2017; Cortez, Baturova et al., 2017; Kors, van Herpen, Sittig, & van Bemmel, 1990):
Furthermore, we calculated a coefficient of variation of PWA (CV‐PWA) across 12 electrocardiographic leads as an indicator of PWA spatial variation by dividing the SD of PWA by the mean PWA (Nakatani et al., 2016). Examples of PWA parameter determinations are shown in Figure 1. The predictive efficacies of maximum PWA, mean PWA, P‐wave vector magnitude, and CV‐PWA for AF recurrence after ablation were then assessed.
Figure 1.
Measurements of various parameters related to P‐wave amplitude (PWA). Electrocardiograms were recorded on the day following ablation. (a) Examples from a 59‐year‐old female patient without AF recurrence. Maximum PWA was 0.15 mV from lead II. Mean PWA and standard deviation of PWA were 0.07 mV and 0.06 mV, respectively. Therefore, a coefficient of variation (CV‐PWA) was calculated as 0.06/0.07 = 0.86. PWA values from leads II, V2, and V6 were 0.15 mV, 0.06 mV, and 0.09 mV, respectively. Therefore, P‐wave vector magnitude was calculated as . (b) Examples from a 69‐year‐old male patient with AF recurrence after ablation. Maximum PWA, mean PWA, P‐wave vector magnitude, and CV‐PWA were 0.15 mV, 0.06 mV, 0.09 mV, and 0.83, respectively
2.4. Postprocedure care and follow‐up
A clinical interview and surface 12‐lead electrocardiogram were performed on the day following ablation and monthly thereafter during visits to the outpatient clinic. 24‐hr Holter monitoring was performed as needed during the follow‐up period. Atrial fibrillation recurrence was identified from symptoms along with documentation of an AF episode lasting ≥30 s on a surface 12‐lead electrocardiogram or Holter monitoring after a 3‐month blanking period from ablation (Calkins et al., 2018). Antiarrhythmic drugs were resumed at the discretion of the treating physician.
2.5. Statistical analysis
Values are presented as the mean ± SD together with 95% confidence intervals. Continuous variables were compared by unpaired Student's t test and categorical variables by chi‐square test. Receiver operating characteristic (ROC) curve analyses were performed to determine the optimal parameter cutoff values for predicting AF recurrence. Cox regression analysis was performed to adjust for potential confounding factors. All variables with P‐values <0.20 in the univariate analysis were included in the multivariate analysis. The outcome of ablation over time was plotted using a Kaplan–Meier survival curve and compared between parameter groups (below vs. above cutoff) by log‐rank test. The correlations between parameters were analyzed using Pearson's correlation coefficient. A p < 0.05 (two tailed) was accepted as statistically significant for all tests.
3. RESULTS
3.1. Patient characteristics and outcomes of ablation
Patient demographic and clinical characteristics are summarized in Table 1. Mean age was 64 ± 10 years, and 85% of patients were male. Mean AF duration was 21 ± 26 months, and 48% of patients had long‐standing persistent AF. Structural heart disease and congestive heart failure were observed in 31% and 23% of patients, respectively. Antiarrhythmic drugs were administered in 26% of patients before and 62% of patients after ablation. The left atrium was enlarged, and left atrial appendage flow velocity was lower than normal. Although left ventricular systolic function was preserved, B‐type natriuretic peptide was elevated.
Table 1.
Patient characteristics
| All patients (n = 126) | AF‐recurrence group (n = 39) | AF‐free group (n = 87) | p value | |
|---|---|---|---|---|
| Age, years | 64 ± 10 | 64 ± 11 | 63 ± 9 | 0.950 |
| Male gender | 107 (85) | 34 (87) | 73 (84) | 0.838 |
| Duration of AF, months | 21 ± 26 | 21 ± 30 | 22 ± 24 | 0.722 |
| Long‐standing persistent AF | 60 (48) | 19 (49) | 47) | >0.999 |
| Structural heart disease | 39 (31) | 14 (36) | 25 (29) | 0.552 |
| Congestive heart failure | 29 (23) | 11 (28) | 18 (21) | 0.485 |
| Hypertension | 69 (55) | 24 (62) | 45 (52) | 0.407 |
| Diabetes mellitus | 18 (14) | 5 (13) | 13 (15) | 0.969 |
| Past history of stroke | 13 (10) | 3 (8) | 10 (11) | 0.740 |
| Antiarrhythmic drugs before ablation | 33 (26) | 13 (33) | 20 (23) | 0.316 |
| Antiarrhythmic drugs after ablation | 78 (62) | 28 (72) | 50 (57) | 0.183 |
| Left atrial dimension, mm | 46 ± 6 | 47 ± 5 | 45 ± 7 | 0.096 |
| Left atrial appendage flow velocity, cm/s | 37 ± 24 | 36 ± 20 | 37 ± 25 | 0.825 |
| Left ventricular ejection fraction, % | 60 ± 10 | 60 ± 8 | 61 ± 10 | 0.930 |
| B‐type natriuretic peptide, pg/ml | 147 ± 114 | 150 ± 110 | 146 ± 116 | 0.843 |
AF: atrial fibrillation. Data are mean ± SD or number (%) of patients.
Atrial fibrillation did not recur in 87 patients (69%) during 32 ± 15 months of follow‐up. Accordingly, the study subjects were divided into a 39‐patient AF‐recurrence group and 87‐patient AF‐free group. Patient characteristics did not differ between groups (Table 1).
3.2. Electrocardiographic Parameters
Table 2 summarizes electrocardiographic parameters measured one day after ablation. Sinus rate, PQ interval, QRS duration, and QT interval did not differ between AF‐recurrence and AF‐free groups. However, PWAs measured from leads II, III, aVF, V5, and V6 were significantly lower in the AF‐recurrence group, while PWAs measured from leads aVR and aVL were higher (Figure 2). Maximum PWA, mean PWA, and P‐wave vector magnitude were also significantly lower in the AF‐recurrence group than the AF‐free group one day after ablation (maximum PWA, 0.14 ± 0.05 mV vs. 0.16 ± 0.05 mV, p = 0.017, Figure 2a; mean PWA, 0.05 ± 0.02 mV vs. 0.06 ± 0.02 mV, p = 0.003, Figure 2b; P‐wave vector magnitude, 0.09 ± 0.03 mV vs. 0.13 ± 0.04 mV, p < 0.001, Figure 2c); however, CV‐PWA did not differ between groups (1.14 ± 0.77 vs. 1.04 ± 0.47, p = 0.369, Figure 2d).
Table 2.
Electrocardiographic parameters
| All patients (n = 126) | AF‐recurrence group (n = 39) | AF‐free group (n = 87) | p value | |
|---|---|---|---|---|
| Sinus rate, bpm | 75 ± 10 | 74 ± 10 | 76 ± 10 | 0.481 |
| PQ interval, msec | 183 ± 31 | 183 ± 30 | 182 ± 31 | 0.879 |
| QRS duration, msec | 103 ± 14 | 104 ± 15 | 103 ± 14 | 0.735 |
| QT interval, msec | 384 ± 34 | 392 ± 35 | 380 ± 33 | 0.090 |
| P‐wave amplitude, mV | ||||
| Lead I | 0.03 ± 0.03 | 0.03 ± 0.02 | 0.03 ± 0.03 | 0.313 |
| Lead II | 0.09 ± 0.04 | 0.07 ± 0.03 | 0.10 ± 0.04 | <0.001 |
| Lead III | 0.07 ± 0.04 | 0.05 ± 0.04 | 0.08 ± 0.04 | <0.001 |
| Lead aVR | −0.05 ± 0.02 | −0.04 ± 0.02 | −0.05 ± 0.02 | <0.001 |
| Lead aVL | 0.01 ± 0.04 | 0.02 ± 0.03 | 0.00 ± 0.04 | 0.017 |
| Lead aVF | 0.08 ± 0.04 | 0.05 ± 0.03 | 0.09 ± 0.04 | <0.001 |
| Lead V1 | 0.13 ± 0.08 | 0.12 ± 0.07 | 0.14 ± 0.08 | 0.269 |
| Lead V2 | 0.12 ± 0.06 | 0.11 ± 0.06 | 0.12 ± 0.06 | 0.275 |
| Lead V3 | 0.09 ± 0.05 | 0.08 ± 0.03 | 0.10 ± 0.05 | 0.122 |
| Lead V4 | 0.07 ± 0.03 | 0.06 ± 0.03 | 0.08 ± 0.04 | 0.051 |
| Lead V5 | 0.05 ± 0.03 | 0.04 ± 0.02 | 0.05 ± 0.03 | 0.020 |
| Lead V6 | 0.03 ± 0.02 | 0.02 ± 0.02 | 0.04 ± 0.02 | 0.008 |
AF: atrial fibrillation. Data are mean ± SD.
Figure 2.
Comparison of P‐wave parameters between the atrial fibrillation (AF)‐recurrence group and the AF‐free group. Comparison of maximum P‐wave amplitude (PWA, a), mean PWA (b), P‐wave vector magnitude (c), and coefficient of variation of PWA (CV‐PWA)
3.3. Predictive accuracy of P‐wave parameters for AF recurrence
Table 3 summarizes the accuracies of these P‐wave parameters for predicting AF recurrence after ablation. P‐wave vector magnitude yielded the highest predictive accuracy among the P‐wave parameters according to ROC curve analysis (area under the curve of 0.740; sensitivity of 92%, specificity of 50%, and positive and negative predictive values of 45% and 93%, respectively, for a cutoff value of 0.13 mV). A multivariate Cox regression analysis revealed that the predictive ability of P‐wave vector magnitude for AF recurrence was independent of other clinical properties including the administration of antiarrhythmic drugs and left atrial dimension (hazard ratio: 0.153, 95% confidence interval: 0.046–0.507, p = 0.002; Table 4). The difference in AF‐free survival between groups stratified by a P‐wave vector magnitude cutoff of 0.13 mV was assessed by Kaplan–Meier analysis (Figure 3) and demonstrated that 93% of the high P‐wave vector magnitude group (≥0.13 mV, n = 47) remained AF‐free compared to only 57% of the low P‐wave vector magnitude group (<0.13 mV, n = 79) during follow‐up.
Table 3.
The predictive efficacies of parameters related to the P‐wave amplitude (PWA) for recurrence of atrial fibrillation
| Sensitivity (%) | Specificity (%) | Positive predictive value (%) | Negative predictive value (%) | Area under the curve | Cutoff value | |
|---|---|---|---|---|---|---|
| Maximum PWA, mV | 72 | 47 | 38 | 79 | 0.614 | 0.16 |
| Mean PWA, mV, | 64 | 53 | 38 | 77 | 0.658 | 0.06 |
| P‐wave vector magnitude, mV | 92 | 50 | 45 | 93 | 0.740 | 0.13 |
| CV‐PWA | 56 | 49 | 33 | 71 | 0.501 | 0.95 |
CV‐PWA: coefficient of variation of P‐wave amplitude.
Table 4.
Univariate and multivariate Cox regression analysis for recurrence of atrial fibrillation (AF) after ablation
| Univariate | Multivariate | |||
|---|---|---|---|---|
| Hazard ratio (95% CI) | p value | Hazard ratio (95% CI) | p value | |
| Age, years | 1.003 (0.970 to 1.037) | 0.864 | ||
| Male gender | 0.725 (0.283 to 1.855) | 0.502 | ||
| Duration of AF, months | 0.998 (0.985 to 1.011) | 0.750 | ||
| Structural heart disease | 0.777 (0.404 to 1.495) | 0.450 | ||
| Congestive heart failure | 0.808 (0.401 to 1.627) | 0.550 | ||
| Hypertension | 0.757 (0.397 to 1.444) | 0.399 | ||
| Diabetes mellitus | 1.115 (0.436 to 2.853) | 0.821 | ||
| Past history of stroke | 1.461 (0.450 to 4.748) | 0.528 | ||
| Antiarrhythmic drugs before ablation | 0.697 (0.358 to 1.358) | 0.289 | ||
| Antiarrhythmic drugs after ablation | 0.529 (0.233 to 1.201) | 0.128 | 0.808 (0.518 to 1.260) | 0.347 |
| Left atrial dimension, mm | 1.041 (0.985 to 1.099) | 0.153 | 1.017 (0.966 to 1.071) | 0.527 |
| Left atrial appendage flow velocity, cm/s | 1.000 (0.986 to 1.013) | 0.947 | ||
| Left ventricular ejection fraction, % | 0.997 (0.965 to 1.029) | 0.997 | ||
| B‐type natriuretic peptide, pg/ml | 1.000 (0.997 to 1.003) | 0.995 | ||
| P‐wave vector magnitude ≥ 0.13 mV | 0.135 (0.042 to 0.442) | 0.001 | 0.153 (0.046 to 0.507) | 0.002 |
CI: confidence interval.
Figure 3.
Kaplan–Meier curves for atrial fibrillation‐free survival according to high (≥0.13 mV) or low (<0.13 mV) P‐wave vector magnitude. The cutoff value was determined by receiver operating characteristic curve analysis
3.4. Correlation of P‐wave vector magnitude with echocardiographic parameters
P‐wave vector magnitude was weakly but significantly correlated with left atrial dimension (R = −0.280, p = 0.004, Figure 4a), but not with left atrial appendage flow velocity or left ventricular ejection fraction (left atrial appendage flow velocity, R = 0.032, p = 0.716, Figure 4b; left ventricular ejection fraction, R = −0.118, p = 0.238, Figure 4c).
Figure 4.
Correlations between P‐wave vector magnitude and echocardiographic parameters. Correlation of P‐wave vector magnitude with left atrial dimension (a), left atrial appendage flow velocity (b), and left ventricular ejection fraction (c)
3.5. Influence of ablation strategy on PWA parameters
Ablation strategy had no influence on PWA parameters, as maximum PWA, mean PWA, P‐wave vector magnitude, and CV‐PWA did not differ between the 44 patients who underwent CFAE ablation and the 52 patients receiving left atrial posterior wall isolation with mitral isthmus line ablation (maximum PWA, 0.16 ± 0.05 mV vs. 0.16 ± 0.05 mV, p = 0.970; mean PWA, 0.06 ± 0.02 mV vs. 0.06 ± 0.02 mV, p = 0.317; P‐wave vector magnitude, 0.11 ± 0.04 mV vs. 0.13 ± 0.04 mV, p = 0.091; CV‐PWA, 1.09 ± 0.62 vs. 1.03 ± 0.54, p = 0.603).
4. DISCUSSION
In the present study, we assessed the predictive values of various PWA parameters for AF recurrence after catheter ablation in patients with persistent AF. P‐wave amplitude measured in inferior leads and left precordial leads was lower in patients with AF recurrence compared to those without recurrence. Moreover, lower P‐wave vector magnitude, maximum PWA, and mean PWA were associated with AF recurrence, of which P‐wave vector magnitude yielded the highest predictive accuracy for AF recurrence according to ROC curve analysis. Furthermore, the predictive ability of P‐wave vector magnitude for AF recurrence was independent of other clinical parameters.
P‐wave morphology reflects the complex interplay between electrophysiological and anatomical features of the atria. Thus, P‐wave morphology changes with atrial remodeling (Magnani, Williamson, Ellinor, Monahan, & Benjamin, 2009) which in turn provides the substrate for AF progression. Indeed, significant relationships have been documented among specific P‐wave parameters, atrial remodeling, and consequent AF development. For instance, prolonged P‐wave duration reflects slowing of atrial conduction and is related to the extent of atrial fibrosis (Huo et al., 2014). Accordingly, prolongation of P‐wave duration is related to AF occurrence (Nielsen et al., 2015). Moreover, variation of P‐wave duration among the electrocardiographic leads reflects atrial conduction heterogeneity and is associated with new‐onset AF (Dilaveris et al., 1998) and AF recurrence after catheter ablation (Nakatani et al., 2016). Furthermore, P‐wave terminal force, the product of the duration and amplitude of the P‐wave negative terminal phase from lead V1, is a marker of increased atrial fibrosis, left atrial dilatation, and left atrial function (Eranti et al., 2014; Tiffany Win et al., 2015). Consequently, P‐wave terminal force is associated with AF occurrence (Tereshchenko et al., 2014).
Although relationships between these P‐wave parameters and AF development have been established, accurate assessments are difficult as they require measurement of P‐wave duration between onset and offset points, which are often unclear. On the other hand, there is less risk of error in measurement of PWA, and PWA is indicative of both the magnitude and direction of electrical activity in the atria. Therefore, PWA reliably reflects the decrease in atrial electrical activation magnitude due to loss of myocardial mass and interstitial fibrosis. This notion is supported by a previous electroanatomical mapping study (Park et al., 2016) that revealed a linear correlation of PWA in lead I with the voltage amplitude of the left atrial electrogram. Furthermore, a relationship between PWA and AF development was observed in a few previous studies. Low PWAs from leads II and V1 were related to immediate recurrence of AF after cardioversion (Gorenek et al., 2003), and low PWA in lead I independently predicted AF recurrence after catheter ablation in patients with paroxysmal AF (Park et al., 2016).
The findings of the present study are consistent with these previous studies (Gorenek et al., 2003; Park et al., 2016) in that low PWA was related to AF development. However, the meaningful leads differ among studies. This discrepancy may be due to differences in atrial morphology. Since the morphology of the atria affects the direction of atrial electrical activation, the P‐wave vector may be altered with atrial dilatation. The subjects of the present study had persistent AF, which is associated with left atrial dilatation, whereas a previous study (Park et al., 2016) included paroxysmal AF patients without left atrial dilatation. Accordingly, the P‐wave vector should be considered for estimation of electrical activity in the atria by PWA. Therefore, we adopted P‐wave vector magnitude as an indicator of the magnitude of atrial electrical activation and demonstrated that the P‐wave vector magnitude yields higher predictive accuracy for AF recurrence than all other measured PWA parameters.
It is unclear whether left or right atrial remodeling is the predominant contributor to P‐wave vector magnitude. Since the correlation of P‐wave vector magnitude with left atrial dimension was weak but significant, left atrial remodeling may have contributed to P‐wave vector magnitude. However, right atrial electrical activation may be associated with P‐wave vector magnitude to some extent because we did not observe a difference in P‐wave vector magnitude between patients who underwent left atrial posterior wall isolation with mitral isthmus line ablation and patients receiving CFAE ablation. This conclusion is further supported by a previous study (Cortez, Barham et al., 2017; Cortez, Baturova et al., 2017) reporting that P‐wave vector magnitude inversely correlated with right atrial pressure and right ventricular ejection fraction.
4.1. Clinical implications
Surface 12‐lead electrocardiogram is the simplest and most routine test for the evaluation of cardiac conditions. Thus, PWA parameters can be assessed at low cost without special equipment. Moreover, PWAs can be measured using computerized automated systems as in this study to reduce time and labor. In addition, automated measurements mitigate intra‐ and inter‐observer variations.
P‐wave vector magnitude provides useful information for the postprocedural care of patients with persistent AF. If P‐wave vector magnitude is below the calculated cutoff from ROC curve analysis (<0.13 mV in this cohort), we may continue antiarrhythmic drug treatment after ablation to maintain sinus rhythm and conduct more frequent follow‐up to detect AF recurrence. Alternatively, if P‐wave vector magnitude is above the cutoff, we might consider discontinuation of anticoagulation. Indeed, in this cohort, the negative predictive value of P‐wave vector magnitude for AF recurrence was 93%.
4.2. Study limitations
The present study was limited in several ways. First, since we assessed PWA parameters after ablation, the predictive utilities of PWA parameters assessed before ablation on AF recurrence are still unclear. Therefore, the results of the present study cannot be applied for patient selection. Second, pulmonary vein isolation may have affected PWA parameters after ablation. However, the effect of pulmonary vein isolation on the PWA may be small because P‐wave duration and terminal part of the P wave are preferentially modified after pulmonary vein isolation (Janin et al., 2010). Third, the additional ablation strategy differed among patients. Thus, we cannot exclude the influence of additional ablation on AF recurrence. Fourth, since antiarrhythmic drugs were resumed at the discretion of the treating physician, we cannot exclude the influence of antiarrhythmic drugs on outcome; however, the influence of antiarrhythmic drugs may have been weak because administration rates were similar between recurrence and AF‐free groups. Fifth, P‐wave vector magnitude was calculated from the PWA measured from leads II, V2, and V6 based on Kors' quasi‐orthogonal transformation (Cortez, Barham et al., 2017; Cortez, Baturova et al., 2017; Kors et al., 1990); however, this method may not reproduce the orthogonal electrocardiogram accurately. Finally, AF recurrence was diagnosed by a clinical interview and clinical examinations, including surface ECG and 24‐hr Holter monitoring, a method that is well known for underestimating the prevalence of AF because it neglects asymptomatic AF.
5. CONCLUSIONS
P‐wave amplitude parameters are associated with AF recurrence after ablation in patients with persistent AF, and P‐wave vector magnitude is an especially useful predictor of AF recurrence.
ACKNOWLEDGMENTS
The authors thank Jun Morita for technical assistance.
Nakatani Y, Sakamoto T, Yamaguchi Y, Tsujino Y, Kataoka N, Kinugawa K. P‐wave vector magnitude predicts recurrence of atrial fibrillation after catheter ablation in patients with persistent atrial fibrillation. Ann Noninvasive Electrocardiol. 2019;24:e12646 10.1111/anec.12646
Funding information
This research received no grant from any funding agency in the public, commercial, or not‐for‐profit sectors.
REFERENCES
- Burstein, B. , & Nattel, S. (2008). Atrial fibrosis: Mechanisms and clinical relevance in atrial fibrillation. Journal of the American College of Cardiology, 51, 802–809. 10.1016/j.jacc.2007.09.064 [DOI] [PubMed] [Google Scholar]
- Calkins, H. , Hindricks, G. , Cappato, R. , Kim, Y. H. , Saad, E. B. , Aguinaga, L. , … Calkins, H. (2018). 2017 HRS/EHRA/ECAS/APHRS/SOLAECE expert consensus statement on catheter and surgical ablation of atrial fibrillation. Europace, 20(1), e1–e160. 10.1093/europace/eux274 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Cortez, D. , Barham, W. , Ruckdeschel, E. , Sharma, N. , McCanta, A. C. , von Alvensleben, J. , … Nguyen, D. T. (2017). Noninvasive predictors of perioperative atrial arrhythmias in patients with tetralogy of Fallot undergoing pulmonary valve replacement. Clinical Cardiology, 40, 591–596. 10.1002/clc.22707 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Cortez, D. , Baturova, M. , Lindgren, A. , Carlson, J. , Shubik, Y. V. , Olsson, B. , … Platonov, P. G. (2017). Atrial time and voltage dispersion are both needed to predict new‐onset atrial fibrillation in ischemic stroke patients. BMC Cardiovascular Disorders, 17, 200–208. 10.1186/s12872-017-0631-1 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Dilaveris, P. E. , Gialafos, E. J. , Sideris, S. K. , Theopistou, A. M. , Andrikopoulos, G. K. , Kyriakidis, M. , … Toutouzas, P. K. (1998). Simple electrocardiographic markers for the prediction of paroxysmal idiopathic atrial fibrillation. American Heart Journal, 135, 733–738. 10.1016/S0002-8703(98)70030-4 [DOI] [PubMed] [Google Scholar]
- Eranti, A. , Aro, A. L. , Kerola, T. , Anttonen, O. , Rissanen, H. A. , Tikkanen, J. T. , … Huikuri, H. V. (2014). Prevalence and prognostic significance of abnormal P terminal force in lead V1 of the ECG in the general population. Circulation: Arrhythmia and Electrophysiology, 7, 1116–1121. 10.1161/CIRCEP.114.001557 [DOI] [PubMed] [Google Scholar]
- Ganesan, A. N. , Shipp, N. J. , Brooks, A. G. , Kuklik, P. , Lau, D. H. , Lim, H. S. , … Sanders, P. (2013). Long‐term outcomes of catheter ablation of atrial fibrillation: A systematic review and meta‐analysis. Journal of American Heart Association, 2, e004549 10.1161/JAHA.112.004549 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Gorenek, B. , Birdane, A. , Kudaiberdieva, G. , Goktekin, O. , Cavusoglu, Y. , Unalir, A. , … Timuralp, B. (2003). P wave amplitude and duration may predict immediate recurrence of atrial fibrillation after internal cardioversion. Annals of Noninvasive Electrocardiology, 8, 215–218. 10.1046/j.1542-474X.2003.08308.x [DOI] [PMC free article] [PubMed] [Google Scholar]
- Haïssaguerre, M. , Jaïs, P. , Shah, D. C. , Takahashi, A. , Hocini, M. , Quiniou, G. , … Clémenty, J. (1998). Spontaneous initiation of atrial fibrillation by ectopic beats originating in the pulmonary veins. New England Journal of Medicine, 339, 659–666. 10.1056/NEJM199809033391003 [DOI] [PubMed] [Google Scholar]
- Huo, Y. , Mitrofanova, L. , Orshanskaya, V. , Holmberg, P. , Holmqvist, F. , & Platonov, P. G. (2014). P‐wave characteristics and histological atrial abnormality. Journal of Electrocardiology, 47, 275–280. 10.1016/j.jelectrocard.2014.01.011 [DOI] [PubMed] [Google Scholar]
- Janin, S. , Wojcik, M. , Kuniss, M. , Berkowitsch, A. , Erkapic, D. , Zaltsberg, S. , … Neumann, T. (2010). Pulmonary vein antrum isolation and terminal part of the P wave. Pacing Clinical Electrophysiology, 33, 784–789. 10.1111/j.1540-8159.2010.02754.x [DOI] [PubMed] [Google Scholar]
- January, C. T. , Wann, L. S. , Alpert, J. S. , Calkins, H. , Cigarroa, J. E. , Jr Cleveland, J. C. , … Yancy, C. W. (2014). 2014 AHA/ACC/HRS guideline for the management of patients with atrial fibrillation: A report of the American college of cardiology/American heart association task force on practice guidelines and the heart rhythm society. Journal of American College of Cardiology, 2014(64), e1–76. 10.1016/j.jacc.2014.03.022 [DOI] [PubMed] [Google Scholar]
- Kizilirmak, F. , Demir, G. G. , Gokdeniz, T. , Gunes, H. M. , Cakal, B. , Guler, E. , … Kilicaslan, F. (2016). Changes in electrocardiographic P wave parameters after cryoballoon ablation and their association with atrial fibrillation recurrence. Annals of Noninvasive Electrocardiology, 21, 580–587. 10.1111/anec.12364 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Kors, J. A. , van Herpen, G. , Sittig, A. C. , & van Bemmel, J. H. (1990). Reconstruction of the Frank vectorcardiogram from standard electrocardiographic leads: Diagnostic comparison of different methods. European Heart Journal, 11, 1083–1092. 10.1093/oxfordjournals.eurheartj.a059647 [DOI] [PubMed] [Google Scholar]
- Lin, Y. J. , Tai, C. T. , Kao, T. , Chang, S. L. , Lo, L. W. , Tuan, T. C. , … Chen, S. A. (2009). Spatiotemporal organization of the left atrial substrate after circumferential pulmonary vein isolation of atrial fibrillation. Circulation: Arrhythmia and Electrophysiology, 2, 233–241. 10.1161/CIRCEP.108.812024 [DOI] [PubMed] [Google Scholar]
- Magnani, J. W. , Williamson, M. A. , Ellinor, P. T. , Monahan, K. M. , & Benjamin, E. J. (2009). P wave indices: Current status and future directions in epidemiology, clinical, and research applications. Circulation: Arrhythmia and Electrophysiology, 2, 72–79. 10.1161/CIRCEP.108.806828 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Malcolme‐Lawes, L. C. , Juli, C. , Karim, R. , Bai, W. , Quest, R. , Lim, P. B. , … Peters, N. S. (2013). Automated analysis of atrial late gadolinium enhancement imaging that correlates with endocardial voltage and clinical outcomes: A 2‐center study. Heart Rhythm, 10, 1184–1191. 10.1016/j.hrthm.2013.04.030 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Nakatani, Y. , Sakamoto, T. , Mizumaki, K. , Nishida, K. , Kataoka, N. , Tsujino, Y. , … Inoue, H. (2016). Coefficient of variation of P‐wave duration is a novel atrial heterogeneity index to predict recurrence of atrial fibrillation after catheter ablation. Journal of Cardiovascular Electrophysiology, 27, 542–548. 10.1111/jce.12920 [DOI] [PubMed] [Google Scholar]
- Nattel, S. , & Harada, M. (2014). Atrial remodeling and atrial fibrillation: Recent advances and translational perspectives. Journal of the American College of Cardiology, 63, 2335–2345. 10.1016/j.jacc.2014.02.555 [DOI] [PubMed] [Google Scholar]
- Nielsen, J. B. , Kühl, J. T. , Pietersen, A. , Graff, C. , Lind, B. , Struijk, J. J. , … Holst, A. G. (2015). P‐wave duration and the risk of atrial fibrillation: Results from the Copenhagen ECG study. Heart Rhythm, 12, 1887–1895. 10.1016/j.hrthm.2015.04.026 [DOI] [PubMed] [Google Scholar]
- Park, J. K. , Park, J. , Uhm, J. S. , Joung, B. , Lee, M. H. , & Pak, H. N. (2016). Low P‐wave amplitude (<0.1 mV) in lead I is associated with displaced inter‐atrial conduction and clinical recurrence of paroxysmal atrial fibrillation after radiofrequency catheter ablation. Europace, 18, 384–391. 10.1093/europace/euv028 [DOI] [PubMed] [Google Scholar]
- Tereshchenko, L. G. , Henrikson, C. A. , Sotoodehnia, N. , Arking, D. E. , Agarwal, S. K. , Siscovick, D. S. , … Soliman, E. Z. (2014). Electrocardiographic deep terminal negativity of the P wave in V(1) and risk of sudden cardiac death: The atherosclerosis risk in communities (ARIC) study. Journal of American Heart Association, 3, e001387 10.1161/JAHA.114.001387 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Tiffany Win, T. , Ambale Venkatesh, B. , Volpe, G. J. , Mewton, N. , Rizzi, P. , Sharma, R. K. , … Tereshchenko, L. G. (2015). Associations of electrocardiographic P‐wave characteristics with left atrial function, and diffuse left ventricular fibrosis defined by cardiac magnetic resonance: The PRIMERI study. Heart Rhythm, 12, 155–162. 10.1016/j.hrthm.2014.09.044 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Verma, A. , Jiang, C. Y. , Betts, T. R. , Chen, J. , Deisenhofer, I. , Mantovan, R. , … Sanders, P. (2015). STAR AF II Investigators. Approaches to catheter ablation for persistent atrial fibrillation. The New England Journal of Medicine, 372, 1812–1822. 10.1056/NEJMoa1408288 [DOI] [PubMed] [Google Scholar]
- Vlachos, K. , Efremidis, M. , Letsas, K. P. , Bazoukis, G. , Martin, R. , Kalafateli, M. , … Sideris, A. (2017). Low‐voltage areas detected by high‐density electroanatomical mapping predict recurrence after ablation for paroxysmal atrial fibrillation. Journal of Cardiovascular Electrophysiology, 28, 1393–1402. 10.1111/jce.13321 [DOI] [PubMed] [Google Scholar]