Abstract
Background:
Atrial fibrillation (AF) causes atrial electrical and structural remodeling, which are linked to recurrence of AF after ablation. Atrial fibrillatory cycle length (AFCL) and AF wall motion velocity (AFW-V) obtained by tissue velocity imaging (TVI) might characterize such atrial electrical and structural remodeling. The purpose of this study was to assess the clinical and electrophysiologic correlates of these parameters and their relation to ablation outcomes.
Methods and Results:
The study group comprised 80 patients who underwent transthoracic echocardiography followed by AF ablation. Atrial TVI traces were used to determine AFCL-tvi and AFW-V-tvi at the left atrial septal wall. AFCL that was measured from intracardiac electrograms correlated well with AFCL-tvi (R=0.6094; P=0.0002). AFW-V-tvi was significantly lower and AFCL-tvi was significantly shorter in patients with non-paroxysmal AF than in those with paroxysmal AF (1.63±0.76 cm/s vs. 2.85±1.00 cm/s, respectively, P
<
0.0001; and 118.2±23.0 ms vs. 145.0±35.0 ms, respectively, P=0.0001). These findings held true for patients with and without post-ablation recurrence. Upon multivariate analysis, a reduced AFW-V-tvi remained the strongest predictor of post-ablation recurrence (hazard ratio for +1-cm/s change, 0.573; 95% confidence interval, 0.337–0.930; P=0.0234).
Conclusions:
TVI of atrial fibrillatory wall motion might enhance the non-invasive characterization of atrial remodeling in patients with AF and thus be used for predicting AF recurrence after ablation. (Circ J 2014; 78: 1619–1627)
Over the past 10 years, catheter-based pulmonary vein isolation (PVI) has been used as an effective therapy for atrial fibrillation (AF).1
Additional aggressive ablation strategies, that is, complex fractionated atrial electrogram (CFAE) ablation and/or atrial linear ablation, are required in patients with progressive atrial electrical, contractile, and structural remodeling. Despite extensive ablation, the post-ablation AF recurrence rate is higher in patients with advanced atrial remodeling than in patients with modest atrial remodeling.2,3
Atrial remodeling is assessed on the basis of the left atrial (LA) size, as measured by transthoracic echocardiography, of shortening of the atrial refractory period, or of decreased atrial fibrillatory cycle length (AFCL) and prolonged atrial conduction time during electrophysiologic study.3–5
Recent studies have shown that AFCL, as seen on the intracardiac electrogram, correlates well with AFCL measured on the basis of atrial mechanical contraction assessed by tissue velocity imaging (TVI).6–8
Furthermore, atrial wall motion velocity, ascertained by TVI during AF, might reflect atrial contractile function.7,8
However, the effects of changes in these echocardiography-derived parameters on the extent of atrial remodeling and the subsequent ablation outcome remain to be clarified. Therefore, we sought to determine whether LA myocardial fibrillatory wall motion assessed by TVI is useful for predicting progression of atrial remodeling and recurrence of AF after ablation.
Methods
Study Patients
The study group comprised 80 consecutive patients with symptomatic drug-refractory AF (70 men, 10 women; mean age, 59.0±10.1 years; median duration of AF, 45.5 months (interquartile range [IQR], 12.3–72.0 months) who were referred to Nihon University Itabashi Hospital for radiofrequency catheter ablation. The group included 27 patients with paroxysmal AF (spontaneous termination within 7 days) and 53 patients with non-paroxysmal AF (AF lasting over 7 days). All patients provided written informed consent for electrophysiologic study and the ablation procedure. Adequate oral anticoagulation was given for at least 1 month before the procedure. All antiarrhythmic drugs were discontinued for at least 5 half-lives prior to ablation. Upon hospital admission, each patient’s medical history was obtained, and a physical examination, 12-lead ECG, chest X-ray, and transesophageal and transthoracic echocardiography were performed. In addition, all patients underwent multi-slice computed tomography, which was performed with a 320-detector row, dynamic volume scanner (Aquilion ONE; Toshiba Medical Systems, Tokyo, Japan) for 3D reconstruction of the LA and PVs.3,9
Transthoracic Echocardiographic Evaluation and TVI Recordings
Comprehensive transthoracic echocardiography was performed during AF 1 day before the ablation procedure, with the use of a Vivid q echocardiography system (GE Medical Systems, Milwaukee, WI, USA) with a 3.5-Hz cardiac transducer. If the patient was in sinus rhythm at the time of echocardiographic evaluation, AF was induced by rapid atrial pacing, and measurements were obtained 5 min later during the electrophysiologic study. Standard echocardiographic measures, that is, LA volume (LAV) measured at the end of the T wave by the prolate-ellipsoid method,10
and left ventricular ejection fraction (LVEF) assessed by M-mode echocardiography (Teichholz’s method), were obtained. Measurements from 3 consecutive beats were averaged.
An apical 4-chamber view was used for TVI images of the mitral annulus, as previously reported.6–8
The sample volume was obtained from the septal and lateral LA above the mitral annulus. The transducer position and angulation were adjusted to minimize the angle between the insonating beam and the long axis of the atrium. TVI traces during the late diastolic phase between the end of the E’ wave and onset of the S’ wave were evaluated offline with Echopac BT 11 software, version 110.0.2 (GE Medical Systems, Horten, Norway). AFCL-tvi was defined as the time interval between 2 consecutive atrial fibrillatory wall motions (with wall motion defined as the onset of the negative atrial deflection on the atrial velocity curve) obtained via TVI. AF wall motion velocity (AFW-V-tvi) was defined as the amplitude of atrial fibrillatory deflections obtained via TVI (Figure 1A). Numerous reports have shown that AFCL and its related electrophysiologic parameters (such as CFAE and dominant frequency) obtained by averaging 10–30 AF cycles for 3–5 s are temporally stable; thus, they are used for the standard mapping approach.2,11–13
Therefore, we extracted atrial fibrillatory wall motion information from a dataset representing more than a 10-s recording time to determine AFCL-tvi and AFW-V-tvi at the septal and lateral LA sites, and the results were averaged.
To clarify whether AFW-V-tvi reflects the degree of atrial mechanical function, we evaluated the relationship between AFW-V-tvi and atrial mechanical function during sinus rhythm. In patients in whom sinus rhythm was maintained 1 day after ablation (n=73), a’ velocity during sinus rhythm was measured with the sample volume placed slightly above the septal side of the mitral annulus. In addition, we conducted a sub-analysis by measuring LA strain during sinus rhythm in 20 of the 80 patients. Peak LA strain during ventricular systole (LA reservoir phase: LAs) and during late diastole after the P wave on the electrocardiogram (active contraction phase: LAa) were measured by taking the average from 12 segments (annular, mid and superior segments along the septal, lateral, anterior, and inferior LA walls on apical 4-chamber and 2-chamber images).14
All measurements were performed by 2 echocardiographers, each with more than 5 years of experience.
AFCL Recording and Ablation Procedure
An electrophysiologic study was performed as previously described, with each patient under sedation, which was achieved by an intravenous infusion of propofol and fentanyl.3,9
In brief, after vascular access was obtained, single trans-septal puncture was performed and this was followed by extensive ipsilateral PVI, which was guided by double Lasso catheters (15-mm Lasso, 4.5-mm interelectrode spacing; 20-mm Lasso, 6-mm interelectrode spacing; Biosense Webster, Diamond Bar, CA, USA), and a 3D geometric map was generated by a NavX system (St. Jude Medical, St. Paul, MN, USA). A 3.5-mm irrigated-tip catheter (Navistar ThermoCool; Biosense Webster) was used for ablation. LA mapping was performed with a 20-pole circular mapping catheter (1.5-mm interelectrode spacing; Livewire Spiral HP catheter; St. Jude Medical). High-pass and low-pass filter settings were 30 Hz and 400 Hz, respectively. In 23 patients, AFCL derived from the intracardiac electrogram (AFCL-egm) was recorded before ablation. Local atrial electrograms obtained during AF from the septal LA wall near the mitral valve annulus, which corresponded well to the septal LA sites for the TVI data recording, were recorded simultaneously with AF wall motion by transthoracic echocardiography. Eight-second electrogram recordings were stored on a LabSystem PRO EP Recording System (Bard Electrophysiology, Lowell, MA, USA). The mean AFCL measured from a 5-s local atrial electrogram was calculated by the Bard Lab Pro (Bard Electrophysiology). Radiofrequency energy was delivered at a power setting of 30 W, and the upper temperature limit was set to 41°C at a saline irrigation rate of 17–30 ml/min (CoolFlow Pump; Biosense Webster). The endpoint of PVI was the achievement of complete entrance and exit block. In patients in whom AF was not terminated by PVI or in whom sustained AF was inducible after PVI, linear ablation at the LA roof and mitral isthmus and/or CFAE-based ablation within the LA (LA ablation) was performed. The endpoint of these steps was AF termination during the procedure or LA linear ablation and abolition of all CFAEs in the LA. Cavotricuspid isthmus ablation was performed when typical atrial flutter was induced by burst atrial pacing or observed clinically.
Post-Ablation Follow-up
Antiarrhythmic drugs that had been suspended were resumed after the procedure, but were then stopped after a 2-month post-ablation blanking period. All patients underwent routine follow up at our outpatient clinic, where a physical evaluation and 12-lead ECG were performed at 2 weeks and at 1, 3, 6, and 12 months after ablation. Twenty-four-hour Holter recordings were obtained at 3, 6, and 12 months after ablation. Ambulatory ECG event monitoring was initiated for patients who reported any cardiac symptoms. Successful ablation was defined as non-recurrence of AF lasting more than 30 s on the standard ECG, ECG event monitor, or 24-h Holter recording during the 12-month follow-up period after the 2-month post-ablation blanking period.
Statistical Analysis
Continuous variables are expressed as mean ± SD or as median and IQR. Differences in continuous variables between paroxysmal AF and non-paroxysmal AF patients and between patients with and without AF recurrence after ablation were analyzed by means of a non-paired
t
test or the Mann-Whitney U test, as appropriate. Categorical variables are expressed as percentages, and differences were analyzed by using the chi-squared test. Correlation was evaluated by Spearman’s rank order correlation coefficient. Bland and Altman plots with 95% limits of agreement were used to assess inter- and intra-observer reproducibility of the AFCL-tvi and AFW-V-tvi measurements. Variables that were entered into multivariate logistic or Cox hazard regression analyses were those with a P value of <0.1 in the univariate models (AFCL-tvi was excluded from Cox hazard regression analysis because it was shown to correlate well with AFW-V-tvi [R=0.6003, P<0.0001]). A P
value of <0.05 was considered statistically significant. A receiver operator characteristic (ROC) curve was drawn to evaluate the performance of the best independent predictor of AF recurrence after ablation. The optimal cut-off point was chosen as the point yielding the highest combined sensitivity and specificity. All statistical analyses were performed with JMP 9 software (SAS Institute, Cary, NC, USA).
Results
Reproducibility of the AFCL-tvi and AFW-V-tvi Measurements
In evaluating AFCL-tvi and AFW-V-tvi measured at septal and lateral LA sites, high interobserver congruence was found. For example, the bias (mean difference) between AFCL-tvi measurements was +1.92 ms for septal LA sites and +1.32 ms for lateral LA sites, whereas the minimum and maximum repeatability coefficients for 95% of the patients were −3.09 and +6.93 ms, respectively, and −2.33 ms and +4.97 ms, respectively. Intraobserver reproducibility of AFCL-tvi measurements and interobserver and intraobserver reproducibility of AFW-V-tvi measurements were also high, as shown in Table 1.
Table 1.
Inter- and Intraobserver Reproducibility of AFCL-tvi and AFW-V-tvi Measurements
| |
Interobserver |
Intraobserver |
| |
Mean difference (bias) |
Minimum and maximum
repeatability coefficient |
Mean difference (bias) |
Minimum and maximum
repeatability coefficient |
| AFCL-tvi (ms) |
|
|
|
|
| Septal |
+1.920 |
–3.089 to +6.929 |
+2.320 |
–2.163 to +6.803 |
| Lateral |
+1.320 |
–2.325 to +4.965 |
–0.680 |
–4.635 to +3.275 |
| AFW-V-tvi (cm/s) |
|
|
|
|
| Septal |
–0.0012 |
–0.0567 to +0.0543 |
+0.0054 |
–0.0563 to +0.0672 |
| Lateral |
+0.1043 |
–0.0115 to +0.2202 |
+0.0967 |
–0.0015 to +0.1948 |
AFCL-tvi, atrial fibrillatory cycle length obtained via tissue velocity imaging; AFW-V-tvi, atrial fibrillatory wall motion velocity obtained via tissue velocity imaging.
TVI recording from the septal LA site was completed in all patients, but that from the lateral LA site was not completed in 6 (7.5%) of the 80 patients because of poor image quality. Because of better image quality of AFCL-tvi at the septal sites than at lateral sites, AFW-V-tvi and AFCL-tvi at septal sites were used for further analyses. In the 23 patients for whom AFCL was compared between intracardiac electrograms and simultaneously recorded tissue velocity images, the mean AFCL-tvi at septal LA sites correlated well with the AFCL-egm at corresponding sites in the LA (R=0.6094, P=0.0002; Figure 2).
Relationship Between AF Type, AFCL-tvi, and AFW-V-tvi
Clinical characteristics and baseline echocardiographic variables are shown for the total patient group and per type of AF (paroxysmal vs. non-paroxysmal) in Table 2. The percentage of female patients was greater in the non-paroxysmal AF group than in the paroxysmal AF group, but the difference was not significant. The LAV was significantly larger in the non-paroxysmal AF patients than in the paroxysmal AF patients; the LVEF tended to be smaller and the LV end-systolic diameter tended to be greater in this group. Representative examples of AFCL-tvi and AFW-V-tvi in patients with paroxysmal and non-paroxysmal AF are shown in Figure 1B. The AFCL-tvi was significantly shorter in patients with non-paroxysmal AF than in patients with paroxysmal AF (118.2±23.0 ms vs. 145.0±35.0 ms; P=0.0001). The AFW-V-tvi was significantly lower in patients with non-paroxysmal AF than in patients with paroxysmal AF (1.63±0.76 cm/s vs. 2.85±1.00 cm/s; P<0.0001). After adjustment for the effects of other variables by multivariate logistic regression analysis, decreased AFW-V-tvi (odds ratio [OR] for +1-cm/s change, 0.573; 95% confidence interval [CI], 0.337–0.930; P=0.0234). and female sex (OR, 4.106; 95% CI, 1.488–10.540; P=0.0079) were found to be the strong determinants of non-paroxysmal AF. LAV (OR for +1-ml change, 1.012; 95% CI, 0.990–1.033; P=0.2870) and LVEF (OR for +1% change, 0.978; 95% CI, 0.940–1.017; P=0.2636) as determinants of non-paroxysmal AF did not reach statistical significance.
Table 2.
Clinical Characteristics and Echocardiographic Measures in the Total Patient Group and per Study Group
| |
Total (n=80) |
Paroxysmal (n=27) |
Non-paroxysmal (n=53) |
P value* |
| Age (years) |
59.0±10.1 |
57.9±11.0 |
59.6±9.7 |
0.4675 |
| Sex (female) |
10 (13) |
1 (4) |
9 (17) |
0.0638 |
| AF duration (months) |
45.5 [12.3–72.0] |
36.0 [10.0–60.0] |
48.0 [15.5–80.0] |
0.1942 |
| Body mass index (kg/m2) |
24.6±3.7 |
24.7±4.4 |
24.4±3.3 |
0.7193 |
| Casual factors |
|
|
|
|
| Hypertension |
53 (66) |
15 (56) |
38 (72) |
0.1523 |
| Diabetes mellitus |
14 (18) |
4 (15) |
10 (19) |
0.6483 |
| Ischemic heart disease |
3 (4) |
2 (7) |
1 (2) |
0.2357 |
| Heart failure |
16 (20) |
3 (11) |
13 (25) |
0.1402 |
| Stroke or TIA |
6 (8) |
3 (12) |
3 (6) |
0.3678 |
| Echocardiographic measures |
|
|
|
|
| LAD (mm) |
41.5±7.0 |
39.4±7.2 |
42.5±6.8 |
0.0645 |
| LAV (ml) |
51.8±21.5 |
43.8±16.6 |
55.9±22.7 |
0.0162 |
| LVEF (%) |
65.6±10.3 |
67.7±7.5 |
64.1±11.2 |
0.0579 |
| LVEDD (mm) |
49.6±5.7 |
48.7±4.9 |
50.1±6.0 |
0.3144 |
| LVESD (mm) |
31.6±6.4 |
29.7±4.7 |
32.5±7.0 |
0.0681 |
| IVSd (mm) |
10.2±1.9 |
10.1±1.9 |
10.2±1.9 |
0.7574 |
| PWd (mm) |
10.2±1.9 |
9.9±2.2 |
10.3±1.8 |
0.3866 |
| AFCL-tvi (septal) (ms) |
127.4±30.5 |
145.0±35.0 |
118.2±23.0 |
0.0001 |
| AFW-V-tvi (septal) (cm/s) |
2.04±1.03 |
2.85±1.00 |
1.63±0.76 |
<0.0001 |
AF, atrial fibrillation; IVSd, interventricular septal wall end-diastolic width; LAD, left atrial diameter; LAV, LA volume; LVEDD, left ventricular end-diastolic diameter; LVEF, LV ejection fraction; LVESD, LV end-systolic diameter; PWd, posterior wall end-diastolic width; TIA, transient ischemic attack. Other abbreviations as in Table 1.
Values are mean ± SD, median (interquartile ranges), or n (%).
*Paroxysmal vs. non-paroxysmal.
AFW-V-tvi correlated weakly with LAV before ablation (R=−0.2950, P=0.0079) and correlated well with septal a’ velocity during sinus rhythm 1 day after ablation (R=0.5516, P<0.0001; Figures 3A,B), but no correlation was noted between AFCL-tvi and LAV (R=−0.0490, P=0.6658) or septal a’ velocity (R=0.1519, P=0.1832, respectively). In the sub-analysis, AFW-V-tvi was shown to correlate modestly with both LAs strain (R=0.5477, P=0.0109) and LAa strain (R=0.4959, P=0.0266; Figures 3C,D).
Patient Characteristics, Echocardiographic Variables, and Ablation Results per Non-Recurrence and Recurrence of AF
Patients’ clinical characteristics, ablation results, and transthoracic echocardiographic variables are shown in relation to AF non-recurrence and AF recurrence in Table 3. The median follow-up period was 12.1 (11.5–12.9) months, during which time AF recurred in 32 (40%) of the 80 patients. In the recurrence group (vs. the non-recurrence group), female sex (22% vs. 6%, P=0.0400) and non-paroxysmal AF (84% vs. 54%, P=0.0039) were more prevalent, the procedure-based AF termination rate was lower (31% vs. 60%,
P=0.0099), LAV was greater (58.0±22.8 ml vs. 47.6±19.8 ml, P=0.0336), and LVEF was smaller (61.0±11.3% vs. 68.2±9.2%, P=0.0021). AFCL-tvi at the septum was significantly shorter in the recurrence group than in the non-recurrence group (118.1±24.8 ms vs. 133.9±32.6 ms, P=0.0209). AFW-V-tvi was significantly lower in the recurrence group than in the non-recurrence group (1.64±0.9 cm/s vs. 2.31±1.0 cm/s, P
=
0.0033). There were no significant differences for other variables. The variables related to AF recurrence were further analyzed by Cox multivariate regression analysis, with female sex (hazard ratio [HR], 4.106; 95% CI, 1.488–10.540; P=0.0079) and reduced AFW-V-tvi at the septal site (HR for +1-cm/s change, 0.573; 95% CI, 0.337–0.930; P=0.0234) emerging as powerful predictors of AF recurrence (Table 4). The ROC curve (area under the ROC curve: 0.73) revealed that the best cut-off value of AFW-V-tvi at the septal site for prediction of AF recurrence was <1.5 cm/s (sensitivity 63%, specificity 83%; Figure 4).
Table 3.
Patient Characteristics, Ablation Results, and Echocardiographic Measures per Study Group
| |
Non-recurrence (n=48) |
Recurrence (n=32) |
P value* |
| Age (years) |
57.7±10.3 |
61.0±9.7 |
0.1593 |
| Sex (female) |
3 (6) |
7 (22) |
0.0400 |
| AF duration (months) |
36.0 [12.0–71.5] |
48.0 [27.8–80.0] |
0.1522 |
| Body mass index (kg/m2) |
24.7±3.9 |
24.3±3.5 |
0.7093 |
| AF type (non-paroxysmal) |
26 (54) |
27 (84) |
0.0039 |
| Casual factors |
|
|
|
| Hypertension |
31 (65) |
22 (69) |
0.6988 |
| Diabetes mellitus |
9 (19) |
5 (16) |
0.7171 |
| Ischemic heart disease |
2 (4) |
1 (3) |
0.8081 |
| Heart failure |
8 (17) |
8 (25) |
0.3649 |
| Stroke or TIA |
3 (6) |
3 (9) |
0.6251 |
| Previously prescribed medications |
|
|
|
| ARBs or ACE inhibitors |
19 (40) |
13 (41) |
0.9859 |
| β-blockers |
24 (50) |
18 (56) |
0.5831 |
| Calcium channel antagonists |
14 (29) |
11 (34) |
0.6234 |
| Statins |
5 (10) |
3 (9) |
0.8787 |
| Class I antiarrhythmic drugs |
25 (52) |
17 (53) |
0.9272 |
| Class III antiarrhythmic drugs |
6 (13) |
8 (25) |
0.1535 |
| Ablation result |
|
|
|
| Terminated by PVI |
29 (60) |
10 (31) |
0.0099 |
| Terminated by PVI and LA ablation |
19 (39) |
16 (48) |
0.4846 |
| Not terminated by ablation |
0 (0) |
6 (18) |
<0.0001 |
| Echocardiographic measures |
|
|
|
| LAD (mm) |
40.2±6.7 |
43.3±7.2 |
0.0484 |
| LAV (ml) |
47.6±19.8 |
58.0±22.8 |
0.0336 |
| LVEF (%) |
68.2±9.2 |
61.0±11.3 |
0.0021 |
| LVEDD (mm) |
49.7±5.8 |
49.5±5.5 |
0.9224 |
| LVESD (mm) |
30.7±6.5 |
33.0±6.1 |
0.1192 |
| IVSd (mm) |
10.0±1.8 |
10.4±1.9 |
0.3993 |
| PWd (mm) |
10.1±1.6 |
10.3±2.3 |
0.6380 |
| AFCL-tvi (septal) (ms) |
133.9±32.6 |
118.1±24.8 |
0.0209 |
| AFW-V-tvi (septal) (cm/s) |
2.31±1.0 |
1.64±0.9 |
0.0033 |
ACE, angiotensin-converting enzyme; ARB, angiotensin receptor blocker; PVI, pulmonary vein isolation. Other
abbreviations as in Table 2.
Values are mean ± SD, median (interquartile ranges), or n (%).
*Non-recurrence vs. recurrence.
Table 4.
Results of Cox Hazard Analysis for AF Recurrence After Ablation
| |
Hazard ratio |
95% CI |
P value |
| Non-paroxysmal AF |
1.215 |
0.414–4.092 |
0.7315 |
| Female sex |
4.106 |
1.488–10.540 |
0.0079 |
| LAV +1-cm change |
1.012 |
0.990–1.033 |
0.2870 |
| LVEF +1% change |
0.978 |
0.940–1.017 |
0.2636 |
| AFW-V-tvi (septal) +1-cm/s change |
0.573 |
0.337–0.930 |
0.0234 |
CI, confidence interval. Other abbreviations are as in Table 2.
Discussion
Main Findings
The mean AFCL-tvi at septal and lateral LA sites correlated well with the mean AFCL-egm measured at corresponding sites. A relatively short mean AFCL-tvi and low AFW-V-tvi were strongly associated with non-paroxysmal AF. The mean AFCL-tvi was shorter and the mean AFW-V-tvi was lower in patients who experienced AF recurrence after ablation than in those who did not, and low AFW-V-tvi was found to be a powerful predictor of post-ablation AF recurrence.
Atrial Fibrillatory Wall Motion Measured by TVI and Atrial Remodeling
This study revealed significant correlation between AFCL-egm and AFCL-tvi recorded from septal sites in the LA. The well-known electrophysiologic characteristics of the atrial substrate for AF are shortening and dispersion of refractoriness and prolonged conduction time in the LA and PVs.4,5,15–17
Atrial refractoriness correlates with mean AFCL during AF on the surface ECG and intracardiac electrogram.18,19
Measurement of AFCL during AF by TVI was first reported by Duytschaever et al, who found that local mechanical activation of the atria during AF can be identified via TVI.6
They found a significant correlation between AFCL-tvi and AFCL measured on simultaneously obtained intracardiac electrograms. Our results are in accordance with theirs. In our study, non-paroxysmal AF was characterized by a significantly shorter AFCL-tvi and lower AFW-V-tvi. In a canine model of AF, a correlation was found between mean AFCL and the local epicardial refractory period.5
A shortened atrial refractory period and slow conduction have been linked to atrial contractile dysfunction.20,21
Similarly, we found that AFCL-tvi and AFW-V-tvi reflect shortened atrial refractoriness and reduced atrial mechanical function, which are associated with atrial electrical and structural remodeling in AF.16–19
De Vos et al. similarly reported paroxysmal AF to be associated with non-complex electrograms with a longer average AFCL, whereas the complex electrograms with a shorter AFCL, greater AFCL variability, and electrogram fragmentation were seen mainly in patients with permanent AF.7
They also found shortening of the AFCL-tvi and reduced AFW-V-tvi with the progression of AF.7
Effect of Atrial Fibrillatory Wall Motion Measured by TVI on Ablation Outcome
In our patient series, post-ablation AF recurrence was related to female sex and non-paroxysmal AF. These clinical variables are widely reported as strong predictors of post-ablation recurrence of AF.2,3,11-13,22–24
Importantly, AFCL-tvi and AFW-V-tvi were shown to be predictors of AF recurrence. Two factors are reported to be associated with AF recurrence after ablation. The first is PV re-conduction, which has been observed in approximately 80% of study patients who experienced atrial tachyarrhythmia after the initial AF ablation, and the majority of these patients were free of symptoms after re-isolation of the re-connected PVs.24,25
The second factor is progression of atrial remodeling, as suggested by several previous studies.2,11–13
Thus, electrophysiologic properties such as shortening of the refractory period (AFCL-egm) and slowed conduction or low atrial voltage, and decreased atrial contractile function might contribute to the vulnerability of the atrial substrate to AF. In fact, atrial fibrillatory frequency, as obtained by spectral analysis and AFCL, has been reported to predict AF termination and subsequent short- and long-term success of catheter ablation.2,11–13
With regard to the association between atrial mechanical function and atrial remodeling, atrial mechanical function during sinus rhythm assessed by speckle tracking echocardiography has been shown to be an independent predictor of recurrence of AF or LA reverse remodeling after ablation.22,23
Theoretically, the reduced atrial contraction during AF rhythm shown by AFW-V-tvi would predispose to the reduced atrial mechanical function during sinus rhythm. In the present study, AFW-V-tvi correlated well with septal a’ velocity during sinus rhythm 1 day after ablation. The main hemodynamic determinants of a’ velocity during sinus rhythm include LA systolic function and left ventricular end diastolic pressure, such that an increase in LA contractility leads to an increased a’ velocity.26
In addition, we also found a reasonable association between AFW-V-tvi and LA whole mechanical function, as shown by both LAs and LAa strain. Therefore, a lower AFW-V-tvi leading to the poor AF ablation outcomes we observed is in keeping with these reported findings. Importantly, we found superiority of AFW-V-tvi over AFCL-tvi for predicting post-AF recurrence, and we suggest that AFW-V-tvi reflecting structural atrial remodeling is more useful than AFCL-tvi reflecting electrical remodeling for the prediction of AF recurrence after ablation. These results can be explained by the contradictory results of the previous studies regarding the relationship between AFCL and atrial remodeling. Most of these recent studies have suggested that a shorter AFCL is observed with the progression of AF,2,11–13
but other reports have indicated that a somewhat longer AFCL is observed in patients with more advanced atrial electrical remodeling, including the presence of LA scarring.27,28
Therefore, we speculated that the AFCL would be shortened with the progression of atrial electrical remodeling in the initial phase from paroxysmal to persistent AF, but that the shortened AFCL would increase in the advanced phase from persistent to long-lasting AF. That is, because the presence of LA scarring reduces the area of fractionated activity during AF, this results in a trend toward an increase in the AFCL. In contrast, AFW-V-tvi would decrease more linearly with the progression of atrial remodeling. In fact, we found that LAV, as a robust marker of atrial remodeling, did not correlate with AFCL-tvi but correlated well with AFW-V-tvi.
Clinical Implications
Numerous electrophysiologic and echocardiographic parameters such as atrial refractoriness, conduction time, low voltage area, atrial size, and atrial mechanical function are known predictors of atrial remodeling and AF recurrence after ablation.2,3,11–13,22–24
Measurement of most variables must be done in sinus rhythm or via an invasive approach. Both AFCL-tvi and AFW-V-tvi have significant advantages over other parameters because they can be determined non-invasively, and we can simultaneously evaluate not only mechanical function but also the electrical substrate in patients during AF rhythm, even in cases of persistent, long-lasting AF. These TVI parameters aid in identifying patients at high risk for AF recurrence after ablation and also the selection of optimal AF ablation strategies before the procedure.
Study Limitations
Study limitations were as follows. First, we recorded the AFCL-tvi and AFW-V-tvi at only 2 atrial sites. Values at other LA sites, PVs, or right atrial sites remain to be determined, so it is unclear whether these TVI parameters can be extrapolated to the whole atria. Because of poor image quality via the transthoracic approach, analysis of atrial fibrillatory wall motion at other sites could be challenging. Instead, we measured LA strain as a surrogate marker of whole LA function during sinus rhythm in 20 patients and found good correlation between AFW-V-tvi and LAs strain. This implies that AFW-V-tvi can be extrapolated to LA mechanical/structural remodeling. Assessment of right atrial mechanical function will be needed to clarify the relationship between the whole atria and TVI parameters. Next, we measured septal a’ velocity and LA strain 1 day after ablation as an indicator of atrial mechanical function during sinus rhythm. It is possible that ablation influenced the values. This effect would have been minimized, however, because all patients underwent a similar intervention, either PVI alone or PVI plus LA ablation, and none underwent ablation at septal sites near the mitral annulus.
Conclusions
Recurrence of AF after ablation can be predicted from AFCL-tvi and AFW-V-tvi. Therefore, TVI of the fibrillating atrial myocardium might enhance the non-invasive characterization of remodeling in patients with AF and be a robust predictor AF recurrence after ablation.
Acknowledgments
The authors wish to thank Nao Iyogi (GE Medical Systems) for technical assistance with the TVI measurements.
Disclosures
This study was supported by departmental resources only. All authors have no conflicts of interest to disclose.
References
- 1.
Ouyang F, Bänsch D, Ernst S, Schaumann A, Hachiya H, Chen M, et al.
Complete isolation of left atrium surrounding the pulmonary veins: New insights from the double-Lasso technique in paroxysmal atrial fibrillation.
Circulation
2004;
110:
2090–2096.
- 2.
Lin YJ, Tai CT, Kao T, Chang SL, Lo LW, Tuan TC, et al.
Spatiotemporal organization of the left atrial substrate after circumferential pulmonary vein isolation of atrial fibrillation.
Circ Arrhythm Electrophysiol
2009;
2:
233–241.
- 3.
Okumura Y, Watanabe I, Nakai T, Ohkubo K, Kofune T, Kofune M, et al.
Impact of biomarkers of inflammation and extracellular matrix turnover on the outcome of atrial fibrillation ablation: Importance of matrix metalloproteinase-2 as a predictor of atrial fibrillation recurrence.
J Cardiovasc Electrophysiol
2011;
22:
987–993.
- 4.
Abhayaratna WP, Seward JB, Appleton CP, Douglas PS, Oh JK, Tajik AJ, et al.
Left atrial size: Physiologic determinants and clinical applications.
J Am Coll Cardiol
2006;
47:
2357–2363.
- 5.
Sih HJ, Zipes DP, Berbari EJ, Adams DE, Olgin JE.
Differences in organization between acute and chronic atrial fibrillation in dogs.
J Am Coll Cardiol
2000;
36:
924–931.
- 6.
Duytschaever M, Heyse A, de Sutter J, Crijns H, Gillebert T, Tavernier R, et al.
Transthoracic tissue Doppler imaging of the atria: A novel method to determine the atrial fibrillation cycle length.
J Cardiovasc Electrophysiol
2006;
17:
1202–1209.
- 7.
De Vos CB, Pison L, Pisters R, Schotten U, Cheriex EC, Prins MH, et al.
Atrial fibrillatory wall motion and degree of atrial remodeling in patients with atrial fibrillation: A tissue velocity imaging study. Transthoracic tissue Doppler imaging of the atria: A novel method to determine the atrial fibrillation cycle length.
J Cardiovasc Electrophysiol
2009;
20:
1374–1381.
- 8.
De Vos CB, Crijns HJ, Tieleman RG.
The fibrillating atrial myocardium visualized: An unexploited source of information.
Heart Rhythm
2009;
6:
1247–1248.
- 9.
Sasaki N, Okumura Y, Watanabe I, Hirayama A.
Pulmonary vein remnant as a trigger site for atrial fibrillation.
Circ J
2013;
77:
494–496.
- 10.
Ujino K, Barnes ME, Cha SS, Langins AP, Bailey KR, Seward JB, et al.
Two-dimensional echocardiographic methods for assessment of left atrial volume.
Am J Cardiol
2006;
98:
1185–1188.
- 11.
Haissaguerre M, Sanders P, Hocini M, Hsu LF, Shah DC, Scavee C, et al.
Changes in atrial fibrillation cycle length and inducibility during catheter ablation and their relation to outcome.
Circulation
2004;
109:
3007–3013.
- 12.
Okumura Y, Watanabe I, Kofune M, Nagashima K, Sonoda K, Mano H, et al.
Characteristics and distribution of complex fractionated atrial electrograms and the dominant frequency during atrial fibrillation: Relationship to the response and outcome of circumferential pulmonary vein isolation.
J Interv Card Electrophysiol
2012;
34:
267–275.
- 13.
Matsuo S, Lellouche N, Wright M, Bevilacqua M, Knecht S, Nault I, et al.
Clinical predictors of termination and clinical outcome of catheter ablation for persistent atrial fibrillation.
J Am Coll Cardiol
2009;
54:
788–795.
- 14.
Machino-Ohtsuka T, Seo Y, Ishizu T, Yanaka S, Nakajima H, Atsumi A, et al.
Significant improvement of left atrial and left atrial appendage function after catheter ablation for persistent atrial fibrillation.
Circ J
2013;
77:
1695–1704.
- 15.
Jaïs P, Hocini M, Macle L, Choi KJ, Deisenhofer I, Weerasooriya R, et al.
Distinctive electrophysiological properties of pulmonary veins in patients with atrial fibrillation.
Circulation
2002;
106:
2479–2485.
- 16.
Kistler PM, Sanders P, Fynn SP, Stevenson IH, Spence SJ, Vohra JK, et al.
Electrophysiologic and electroanatomic changes in the human atrium associated with age.
J Am Coll Cardiol
2004;
44:
109–116.
- 17.
Teh AW, Kistler PM, Lee G, Medi C, Heck PM, Spence S, et al.
Electroanatomic properties of the pulmonary veins: Slowed conduction, low voltage and altered refractoriness in AF patients.
J Cardiovasc Electrophysiol
2011;
22:
1083–1091.
- 18.
Misier AR, Opthof T, van Hemel NM, Defauw JJ, de Bakker JM, Janse MJ, et al.
Increased dispersion of refractoriness in patients with idiopathic paroxysmal atrial fibrillation.
J Am Coll Cardiol
1992;
19:
1531–1535.
- 19.
Capucci A, Biffi M, Boriani G, Ravelli F, Nollo G, Sabbatani P, et al.
Dynamic electrophysiological behavior of human atria during paroxysmal atrial fibrillation.
Circulation
1995;
92:
1193–1202.
- 20.
Schotten U, Duytschaever M, Ausma J, Eijsbouts S, Neuberger HR, Allessie M.
Electrical and contractile remodeling during the first days of atrial fibrillation go hand in hand.
Circulation
2003;
107:
1433–1439.
- 21.
Sung SH, Chang SL, Hsu TL, Yu WC, Tai CT, Lin YJ, et al.
Do the left atrial substrate properties correlate with the left atrial mechanical function? A novel insight from the electromechanical study in patients with atrial fibrillation.
J Cardiovasc Electrophysiol
2008;
19:
165–171.
- 22.
Tops LF, Delgado V, Bertini M, Marsan NA, Den Uijl DW, Trines SA, et al.
Left atrial strain predicts reverse remodeling after catheter ablation for atrial fibrillation.
J Am Coll Cardiol
2011;
57:
324–331.
- 23.
Mirza M, Caracciolo G, Khan U, Mori N, Saha SK, Srivathsan K, et al.
Left atrial reservoir function predicts atrial fibrillation recurrence after catheter ablation: A two-dimensional speckle strain study.
J Interv Card Electrophysiol
2011;
31:
197–206.
- 24.
Gerstenfeld EP, Callans DJ, Dixit S, Zado E, Marchlinski FE.
Incidence and location of focal atrial fibrillation triggers in patients undergoing repeat pulmonary vein isolation: Implications for ablation strategies.
J Cardiovasc Electrophysiol
2003;
14:
685–690.
- 25.
Ouyang F, Antz M, Ernst S, Hachiya H, Mavrakis H, Deger FT, et al.
Recovered pulmonary vein conduction as a dominant factor for recurrent atrial tachyarrhythmias after complete circular isolation of the pulmonary veins: Lessons from double Lasso technique.
Circulation
2005;
111:
127–135.
- 26.
Nagueh SF, Sun H, Kopelen HA, Middleton KJ, Khoury DS.
Hemodynamic determinants of the mitral annulus diastolic velocities by tissue Doppler.
J Am Coll Cardiol
2001;
37:
278–285.
- 27.
Yoshida K, Rabbani AB, Oral H, Bach D, Morady F, Chugh A, et al.
Left atrial volume and dominant frequency of atrial fibrillation in patients undergoing catheter ablation of persistent atrial fibrillation.
J Interv Card Electrophysiol
2011;
32:
155–161.
- 28.
Yokokawa M, Latchamsetty R, Good E, Crawford T, Jongnarangsin K, Pelosi FJr, et al.
The impact of age on the atrial substrate: Insights from patients with a low scar burden undergoing catheter ablation of persistent atrial fibrillation.
J Interv Card Electrophysiol
2012;
34:
287–294.
