Usefulness of Echo-Guided Cardiac Resynchronization Pacing in Patients Undergoing “Ablate and Pace” Therapy for Permanent Atrial Fibrillation and Effects of Heart Rate Regularization and Left Ventricular Resynchronization†

Usefulness of Echo-Guided Cardiac Resynchronization Pacing in Patients Undergoing “Ablate and Pace” Therapy for Permanent Atrial Fibrillation and Effects of Heart Rate Regularization and Left Ventricular Resynchronization† Michele Brignole, MDa,*, Carlo Menozzi, MDb, Gian Luca Botto, MDc, Lluís Mont, MDd, Joaquín Osca Asensi, MDe, Dolores García Medina, MDf, Daniele Oddone, MDa, Alessandro Navazio, MDb, Mario Luzi, MDc, Saverio Iacopino, MDg, Giuseppe De Fabrizio, MDh, Alessandro Proclemer, MDi, and Panos Vardas, MDj An acute comparative study of right ventricular (RV) pacing and echocardiographically guided cardiac resynchronization pacing (CRP) was performed in patients who underwent “ablate and pace” therapy for permanent atrial fibrillation. It was hypothesized that optimized CRP guided by tissue Doppler echocardiography would exert an additive beneficial hemodynamic effect to that of rate regularization achieved through atrioventricular junction ablation. An acute intrapatient comparison of echocardiographic parameters was performed between baseline preablation values and RV pacing and CRP (performed <24 hours after ablation) in 50 patients. Optimized CRP configuration was defined as the modality of pacing corresponding to that of the shortest intra–left ventricular (LV) delay among simultaneous biventricular pacing, sequential biventricular pacing, and singlechamber pacing. The intra-LV delay was defined as the difference between the longest and the shortest activation time in the six basal segments of the left ventricle. Compared with preablation measures, the ejection fraction increased by 10.8% during RV pacing (19% in patients with intra-LV delays <47.5 ms and 3% in those with intra-LV delays >47.5 ms). Compared with RV pacing, CRP caused a 9.2% increase in the ejection fraction, a 6.8% decrease in LV systolic diameter, and a 17.3% decrease in mitral regurgitation area; LV dyssynchrony was reduced from 52 ⴞ 27 to 21 ⴞ 12 ms. Similar results were observed in patients with and without depressed systolic function and in patients with and without left bundle branch block. In conclusion, rate regularization achieved through atrioventricular junction ablation and RV pacing provides a favorable hemodynamic effect that is inversely related to the level of LV dyssynchrony. Minimizing LV dyssynchrony by means of optimized CRP yields an additional important benefit. © 2008 Elsevier Inc. All rights reserved. (Am J Cardiol 2008;102:854 – 860) Atrioventricular (AV) node ablation and permanent pacing from the right ventricular (RV) apex provide highly efficient rate control and regularization of atrial fibrillation (AF) and improve symptoms in selected patients.1– 4 Nevertheless,

a Department of Cardiology, Ospedali del Tigullio, Lavagna; bDepartment of Cardiology, Ospedale S. Maria Nuova, Reggio Emilia; cDepartment of Cardiology, Ospedale S. Anna, Como, Italy; dDepartment of Cardiology, Hospital Clínic i Provincial, Barcelona; eDepartment of Cardiology, Hospital Universitario La Fe, Valencia; fDepartment of Cardiology, Hospital Universitario Virgen de Valme, Sevilla, Spain; gDepartment of Cardiology, Clinica S. Anna, Catanzaro; hDepartment of Cardiology, Ospedale Moscati, Avellino; iDepartment of Cardiology, Ospedale Della Misericordia, Udine, Italy; and jDepartment of Cardiology, Heraklion University Hospital, Heraklion, Greece. Manuscript received April 7, 2008; revised manuscript received and accepted May 14, 2008. *Corresponding author: Tel: 39-0185-329569; fax: 39-0185-306506. E-mail address: [email protected] (M. Brignole). † A list of participating institutions and investigators appears in the Appendix.

0002-9149/08/$ – see front matter © 2008 Elsevier Inc. All rights reserved. doi:10.1016/j.amjcard.2008.05.024

RV pacing is not considered to be optimal, because it provides a nonphysiologic asynchronous contraction that might partly counteract its beneficial effects.1,5 On RV pacing, the ventricular activation sequence resembles that of left bundle branch block; that is, the right ventricle is activated before the left ventricle (interventricular dyssynchrony) and the left ventricular (LV) septum before the LV free wall (intraventricular dyssynchrony). RV pacing has been seen to induce LV dyssynchrony in an acute setting6 and after long-term pacing therapy7 in about 50% of patients. The Ablate and Pace in Atrial Fibrillation (APAF) study is a prospective, randomized, single-blind, acute and chronic comparison between RV pacing and optimized cardiac resynchronization pacing (CRP) guided by tissue Doppler echocardiography for patients with permanent AF who undergo ablation and pacing therapy.8 In this report, we focus on the acute comparison between RV pacing and CRP. We hypothesized that optimized CRP would exert an additive beneficial hemodynamic effect to that of rate regularization achieved through AV junction ablation. Moreover, we investigated whether the changes in cardiac perwww.AJConline.org

Arrhythmias and Conduction Disturbances/CRP in Patients With Permanent AF

formance produced by RV pacing and CRP were correlated with the extent of LV dyssynchrony and with the severity of systolic dysfunction. Methods The study population consisted of 50 patients with severely symptomatic, drug-refractory, permanent AF who underwent biventricular pacemaker implantation and AV junction ablation; pacemaker implantation and ablation were allowed to take place at different times (⬍4 weeks apart), but simultaneous procedures were recommended. The following patients were eligible for enrollment in the study: (1) those with permanent AF in whom clinical decisions were made to undertake AV junction ablation and ventricular pacing because of drug-refractory, severely symptomatic, uncontrolled high ventricular rates, and (2) those with permanent AF, drug-refractory heart failure, and depressed LV function in whom clinical decisions were made to undertake CRP (AV junction ablated to avoid interference of the intrinsic rhythm with pacing). Patient exclusion criteria were as follows: (1) New York Heart Association class IV heart failure or systolic blood pressure ⱕ80 mm Hg despite optimized therapy, (2) severe concomitant noncardiac disease, (3) a need for surgical intervention, (4) myocardial infarction within the previous 3 months, (5) primary hypertrophic cardiomyopathy, (6) primary valvular heart disease, (7) a previously implanted pacemaker or implantable cardioverter-defibrillator, (8) an inability to obtain reliable RV and LV pacing or persistent AV block, and (9) an inability to obtain reliable echocardiographic images. All patients received a biventricular pacing system able to deliver sequential pacing, with a programmable VV interval that allowed RV or LV pacing to be advanced up to 80 ms. The VV delay was optimized individually by tissue Doppler echocardiography. The RV leads were positioned in the RV apex. The target for LV leads was the basal or mid portion of the posterolateral free wall. The anatomy of the coronary sinus enabled this target to be reached in 34 patients (84%). The pacemaker was programmed at a rate of 80 beats/min. An acute intrapatient comparison of echocardiographic parameters was made between baseline preablation values and RV pacing and CRP with an individually optimized VV delay (performed ⬍24 hours after ablation). The RV pacing and CRP studies were performed during the same session; the operator who performed the test and analyzed the records was not informed of the mode of pacing. Each echocardiographic evaluation consisted of (1) the measurement of standard echocardiographic parameters (M-mode LV end-diastolic and end-systolic diameters; the biplane ejection fraction [EF] from the apical view; the degree of mitral regurgitation, evaluated by means of color Doppler by measuring the largest area of the regurgitation jet; and diastolic filling time by pulsed-wave Doppler as the duration of the mitral flow, according to the guidelines of the American Society of Echocardiography9), (2) the study of interventricular dyssynchrony (defined as the difference between aortic and pulmonary ejection delays), and (3) the study of LV dyssynchrony.

855

Table 1 Patient characteristics at enrollment (n ⫽ 50) Variable Mean age (yrs) Men Duration of atrial fibrillation (yrs) Patients hospitalized for heart failure before enrollment No. of hospitalizations for heart failure per patient Symptoms and physical capacity* New York Heart Association functional class Minnesota Living With Heart Failure Questionnaire (score) Specific Symptoms Scale (total score) 6-min walking distance (m) Systolic arterial blood pressure (mm Hg) Body weight (kg) Standard electrocardiography Mean heart rate (beats/min) Mean QRS width (ms) Left bundle branch block Holter monitoring Minimum heart rate (beats/min) Mean heart rate (beats/min) Maximum heart rate (beats/min) Medical history Coronary artery disease Dilated cardiomyopathy Other cardiopathies Hypertension Diabetes mellitus Concomitant medications Digoxin Diuretics Nitrates Angiotensin-converting enzyme inhibitors Angiotensin II inhibitors ␤ blockers Calcium antagonists ␣ antagonists Antiarrhythmic drugs

Value 73 ⫾ 8 38 (76%) 3.1 ⫾ 3.4 33 (66%) 3.5 ⫾ 3.5 2.5 ⫾ 0.5 38 ⫾ 16 23 ⫾ 12 350 ⫾ 120 124 ⫾ 15 77 ⫾ 14 96 ⫾ 23 123 ⫾ 38 24 (48%) 63 ⫾ 18 93 ⫾ 18 144 ⫾ 41 16 (32%) 32 (64%) 2 (4%) 25 (50%) 8 (16%) 32 (64%) 41 (82%) 10 (20%) 32 (64%) 10 (20%) 26 (52%) 26 (52%) 2 (4%) 2 (4%)

Data are expressed as mean ⫾ SD or as number (percentage). * Symptoms at enrollment were measured by means of New York Heart Association classification,17 the Minnesota Living With Heart Failure Questionnaire,18 and the Specific Symptoms Scale.1,2,19 Exercise capacity was assessed by the 6-minute walking distance (average of 2 consecutive tests).20

The intra-LV electromechanical delay of the 6 basal segments was used to measure LV dyssynchrony at baseline, after RV pacing, and after sequential CRP with an individually optimized VV delay. Intra-LV electromechanical activation was evaluated by echocardiography, using pulsed-wave tissue Doppler imaging. From 3 apical views (4-chamber, 2-chamber, and long-axis views), the sample volume was placed in the middle of the basal segment of each wall (the posterior septum and lateral wall in the 4-chamber view, the inferior and anterior walls in the 2-chamber view, and the anterior septum and posterior wall in the long-axis view), and the interval between the onset of the QRS complex and the onset of each respective positive component of the regional systolic velocity on tissue Doppler spectral tracing recorded at 100 mm/s was measured. For each basal LV segment, 3 consecutive beats were measured, and the average value was taken. The analysis of systolic velocities was limited to the LV ejection period; postsystolic peaks were excluded. To mark the LV ejection period, the

856

The American Journal of Cardiology (www.AJConline.org)

Table 2 Results Variable

EF (%) LV end-diastolic diameter (mm) LV end-systolic diameter (mm) Mitral regurgitation area (cm2) Diastolic filling time (ms) Interventricular delay (ms) Intra-LV delay (ms)

Baseline

36.2 ⫾ 12.9 62.8 ⫾ 11.5 47.6 ⫾ 11.6 6.5 ⫾ 4.7 307 ⫾ 111 24 ⫾ 19 46 ⫾ 25

RV Pacing

40.1 ⫾ 16.2 63.6 ⫾ 11.6 48.2 ⫾ 12.2 5.2 ⫾ 3.5 321 ⫾ 90 37 ⫾ 22 52 ⫾ 27

CRP

43.8 ⫾ 14.9 61.3 ⫾ 11.2 44.9 ⫾ 11.6 4.3 ⫾ 3.1 328 ⫾ 80 16 ⫾ 12 21 ⫾ 12

RV Pacing vs Baseline

CRP vs Baseline

CRP vs RV Pacing

% Difference

p Value*

% Difference

p Value*

% Difference

p Value*

⫹10.8% ⫹1.3% ⫹1.3% ⫺20.0% ⫹4.6% ⫹54% ⫹12%

0.02 0.90 0.79 0.07 0.32 0.003 0.42

⫹20.1% ⫺2.4% ⫺5.7% ⫺33.8% ⫹6.8% ⫺35% ⫺54%

⬍0.001 0.38 0.01 0.001 0.20 0.02 ⬍0.001

⫹9.2% ⫺3.4% ⫺6.8% ⫺17.3% ⫹2.2% ⫺57% ⫺59%

⬍0.001 0.04 0.02 0.008 0.54 ⬍0.001 ⬍0.001

* Paired Student’s t test.

Table 3 Logistic regression for predictive factors of response* to right ventricular pacing Independent Variable

Figure 1. Comparison between preablation and RV pacing (RVP). Compared with preablation measurements, the EF increased by 19% in the 25 patients who had intra-LV delays ⬍47.5 ms during RVP, whereas it increased by only 3% in those who had intra-LV delays ⬎47.5 ms (p ⫽ 0.045, 2-tailed Mann-Whitney test).

opening and closure of the aortic valve were measured from the pulsed-wave Doppler signals in the LV outflow tract. The intra-LV electromechanical delay was calculated as the time difference between the longest and the shortest interval among the 6 ventricular walls.10 –15 In the previously reported APAF pilot study,13 in a population of 58 controls, the median intra-LV delay was 17 ms, 95% of controls had values ⱕ41 ms, and the interobserver correlation for intra-LV delay was 0.93. The intra-LV delay during simultaneous biventricular pacing was calculated first. If its value was equal to or less than the median of our reference standard (17 ms), LV contraction was considered sufficiently resynchronized, and no other measurement was taken. If longer, the VV interval of the pacemaker was modified by advancing the LV stimulus or the RV stimulus up to 80 ms. If the posterolateral segments were delayed, LV pacing was advanced to a value approximately half that of the delay, and the intra-LV delay was measured again; conversely, if the anteroseptal segments were delayed, RV pacing was advanced in the same manner. Fine adjustments of the VV interval were finally made, if necessary, to achieve the goal of a value ⬍41 ms. If dyssynchrony persisted, LV or RV pacing alone was assessed. The optimized CRP configuration was defined as the modality of pacing corresponding to that of the shortest intra-LV delay recorded during simultaneous

Increasing age (yrs) Men Duration of atrial fibrillation (yrs) Patients hospitalized for heart failure before enrollment New York Heart Association class Minnesota Living With Heart Failure Questionnaire (score) Specific Symptoms Scale (score) 6-min walking distance (m) QRS width (ms) Mean heart rate (beats/min) Left bundle branch block Body weight (kg) Systolic blood pressure (mm Hg) Ischemic disease Dilated cardiomyopathy Hypertension Diabetes mellitus Echocardiographic parameters before ablation LV end-diastolic diameter (mm) Left ventricular end-systolic diameter (mm) Mitral regurgitation area (cm2) Diastolic filling time (ms) Interventricular delay (ms) Intra-LV delay (ms) Intra-LV delay during RV pacing ⬍47.5 ms

Univariate OR

95% CI

p Value

0.92 0.67 0.91 2.14

0.84–1.00 0.18–2.52 0.74–1.11 0.63–7.33

0.06 0.56 0.35 0.22

2.24 1.02

0.76–6.64 0.98–1.06

0.14 0.29

1.02 1.00 0.99 1.00 0.79 1.04 1.00 0.78 0.89 0.54 0.95

0.97–1.07 0.99–1.00 0.98–1.01 0.97–1.03 0.26–2.42 0.99–1.10 0.96–1.04 0.23–2.63 0.27–2.89 0.17–1.72 0.21–4.33

0.46 0.70 0.44 0.94 0.67 0.13 0.92 0.69 0.84 0.30 0.95

0.98 0.95

0.93–1.03 0.90–1.01

0.36 0.09

1.01 0.10 0.98 0.98 3.98

0.89–1.16 0.99–1.00 0.94–1.01 0.96–1.01 1.20–13.2

0.84 0.48 0.16 0.19 0.02

* Any improvement in ejection fraction during right ventricular pacing compared with ejection fraction at baseline. CI ⫽ confidence interval; OR ⫽ odds ratio.

biventricular pacing, sequential biventricular pacing, or single-chamber pacing. In general, the time required for individualized programming of the relative contribution of RV and LV pacing was ⱕ60 minutes. The method has been previously described and validated.15 A great variety of VV intervals was selected to optimize

Arrhythmias and Conduction Disturbances/CRP in Patients With Permanent AF

857

Table 4 Comparison between patients with severely symptomatic systolic dysfunction (according to the indications of guidelines on cardiac resynchronization therapy21) and those without Variable

EF (%) LV end-diastolic diameter (mm) LV end-systolic diameter (mm) Mitral regurgitation area (cm2) Diastolic filling time (ms) Interventricular delay (ms) Intra-LV delay (ms)

EF ⱕ35% and NYHA Class III (n ⫽ 18)

EF ⬎35% or NYHA Class ⬍ III (n ⫽ 32)

Comparison Between Groups

RV Pacing

CRP

% Difference

p Value*

RV Pacing

CRP

% Difference

p Value*

p Value†

32.7 ⫾ 11.5 65.3 ⫾ 10.0 51.2 ⫾ 8.5 5.0 ⫾ 2.3 340 ⫾ 87 35 ⫾ 20 46 ⫾ 24

34.7 ⫾ 9.4 64.3 ⫾ 9.8 49.6 ⫾ 8.5 4.5 ⫾ 2.7 323 ⫾ 52 17 ⫾ 12 19 ⫾ 9

⫹6.1% ⫺1.5% ⫺3.1% ⫺10.0% ⫺5.0% ⫺51% ⫺59%

0.05 0.26 0.18 0.27 0.48 0.01 0.001

44.9 ⫾ 16.7 62.0 ⫾ 11.9 45.6 ⫾ 12.4 5.4 ⫾ 4.0 312 ⫾ 92 38 ⫾ 23 55 ⫾ 28

49.0 ⫾ 15.1 60.5 ⫾ 11.6 43.2 ⫾ 12.5 4.3 ⫾ 3.4 331 ⫾ 93 15 ⫾ 13 22 ⫾ 13

⫹9.1% ⫺2.4% ⫺5.3% ⫺20.4% ⫹6.1% ⫺60% ⫺60%

⬍0.001 0.01 0.06 0.02 0.05 ⬍0.001 ⬍0.001

0.98 0.78 0.47 0.55 0.26 0.55 0.83

* Paired Student’s t test. Unpaired Student’s t test.

†

resynchronization. Simultaneous biventricular pacing reached the target in 20 patients (40%). In 30 patients (60%), better resynchronization was achieved through optimization of the VV interval (LV first in 21 and RV first in 4), LV pacing only (in 3), or RV pacing only (in 2). LV anticipation ranged from 20 to 60 ms (mean 34 ⫾ 14); RV anticipation ranged from 20 to 40 ms (mean 35 ⫾ 19). Overall, LV dyssynchrony was reduced from 36 ⫾ 23 ms during simultaneous biventricular pacing to 21 ⫾ 12 ms during optimized CRP (p ⬍0.001). In calculating the sample size, it was assumed, on the basis of a previous study,16 that RV pacing would be able to increase the EF by 10% compared with preablation values. The sample size able to provide 80% power to show an intrapatient difference, with a probability of 95%, was 50 patients. This number was determined to be sufficient to show a difference between CRP and RV pacing if the magnitude of increase in the EF with CRP were also about 10%. Descriptive analyses present continuous variables as mean ⫾ SD and categorical variables as absolute and relative frequencies. Paired Student’s t test was used to compare echocardiographic data at baseline and during RV pacing and CRP. Unpaired Student’s t test or the Mann-Whitney nonparametric test was used as appropriate to compare the distributions of relative differences between subgroups. Univariate and multivariate logistic regression analysis was performed to detect predictors of response to RV pacing and to CRP. All demographic and echocardiographic data with significance ⬍0.10 on univariate analysis were entered into the multivariate model. Calibration of the multivariate model was calculated by Hosmer-Lemeshow test, while the area under receiveroperating characteristic curve showed its discrimination. For all tests, a 2-sided p value ⬍0.05 was considered significant. SPSS (SPSS, Inc., Chicago, Illinois) was used for all analyses. Results The clinical characteristics of the patients are listed in Table 1.17–20 Compared with preablation measurements, the EF increased and mitral regurgitation showed a decreasing trend during RV pacing (Table 2). The EF increased by 19%

in the 25 patients with intra-LV delay ⬍47.5 ms, whereas it increased by only 3% in the 25 patients with intra-LV delay ⬎47.5 ms (p ⫽ 0.045; Figure 1). Conversely, the 25 responders to RV pacing (defined as those showing any improvement in the EF during RV pacing in comparison with baseline) displayed a significantly lower value of dyssynchrony than the 25 nonresponders (44 ⫾ 23 vs 60 ⫾ 29 ms, p ⫽ 0.045). On multivariate logistic regression analysis, among several variables (listed in Table 3), intra-LV delay ⬍47.5 ms during RV pacing remained a strong predictor of response to RV pacing (odds ratio 6.5, 95% confidence interval 1.6 to 26, p ⫽ 0.008). In contrast, the hemodynamic effect was not influenced by baseline clinical and echocardiographic parameters, except for increasing age (odds ratio 0.89, 95% confidence interval 0.80 to 0.98, p ⫽ 0.02). In particular, baseline dyssynchrony was not predictive of acute outcome. Compared with RV pacing, CRP caused an increase in the EF and decreases in LV systolic diameter, LV diastolic diameter, and mitral regurgitation area; LV dyssynchrony and interventricular dyssynchrony were also reduced (Table 2). On univariate and multivariate logistic analysis, no clinical variable among those listed in Table 3 was able to predict which patients would respond to CRP (defined as those with increases in EFs to above the median). In particular, baseline dyssynchrony was not predictive of acute outcome. Similar results were observed in patients with depressed systolic function and in those without. The patients were subdivided into 2 groups according to the recommendations for the use of biventricular pacing in patients with heart failure with permanent AF of the European Society of Cardiology21: 18 patients had EFs ⱕ35% and were in New York Heart Association class III, thus meeting the requisites for CRP stated in the guidelines, whereas 32 patients did not (Table 4). Similar results were also observed in the patients with and without native left bundle branch block; however, mitral regurgitation was reduced more in patients without native left bundle block (Table 5). Discussion The main result of the study is that in patients who underwent ablate-and-pace therapy for severely symptomatic per-

858

The American Journal of Cardiology (www.AJConline.org)

Table 5 Comparison between patients with baseline left bundle branch block and those without Variable

EF (%) LV end-diastolic diameter (mm) LV end-systolic diameter (mm) Mitral regurgitation area (cm2) Diastolic filling time (ms) Interventricular delay (ms) Intra-LV delay (ms)

Left Bundle Branch Block (n ⫽ 24) RV Pacing

CRP

% Difference

34.6 ⫾ 14.2 66.3 ⫾ 9.1 53.1 ⫾ 9.9 5.7 ⫾ 3.5 318 ⫾ 95 35 ⫾ 21 47 ⫾ 19

38.8 ⫾ 11.8 66.0 ⫾ 9.6 51.7 ⫾ 9.5 5.3 ⫾ 3.3 324 ⫾ 83 16 ⫾ 12 22 ⫾ 10

⫹12.0% ⫺0.5% ⫺2.6% ⫺7.0% ⫹1.9% ⫺53% ⫺52%

No Left Bundle Branch Block (n ⫽ 26) p Value*

RV Pacing

0.002 46.2 ⫾ 16.0 0.55 60.4 ⫾ 12.4 0.11 42.9 ⫾ 10.8 0.37 4.8 ⫾ 3.4 0.70 325 ⫾ 87 0.003 38 ⫾ 23 ⬍0.001 56 ⫾ 32

Comparison Between Groups

CRP

% Difference

p Value*

p Value†

49.4 ⫾ 16.1 58.4 ⫾ 11.1 40.1 ⫾ 10.7 3.5 ⫾ 2.7 331 ⫾ 80 15 ⫾ 13 19 ⫾ 13

⫹6.9% ⫺3.3% ⫺6.5% ⫺27.1% ⫹1.8% ⫺60% ⫺65%

0.009 0.05 0.07 0.004 0.63 ⬍0.001 ⬍0.001

0.15 0.24 0.25 0.03 0.80 0.98 0.26

* Paired Student’s t test. † Unpaired Student’s t test.

manent AF, LV resynchronization after CRP was an acute phenomenon that exerted an additive beneficial hemodynamic effect to that produced by rate regularization achieved by means of AV junction ablation and RV pacing. In brief, AV junction ablation plus RV pacing increased the EF and reduced the magnitude of mitral regurgitation; CRP doubled these effects. Thus, rate regularization and LV resynchronization play a major role in the response to cardiac resynchronization therapy in patients with permanent AF. We have tried to provide a pathophysiologic explanation for the results observed. On RV pacing, about half of the patients reached a sufficient level of synchronicity; in these patients, the effect of rate control and regularization was fully manifested, and cardiac function improved greatly in comparison with the preablation state (Figure 1). In contrast, in those patients in whom some dyssynchrony was still present, the beneficial hemodynamic effect of rate control and regularization of heart rhythm might have been counteracted by the adverse hemodynamic effect of a nonphysiologic RV pacing mode. In other words, rate regularization achieved by means of AV junction ablation and RV pacing provided a favorable hemodynamic effect, which was inversely related to the level of LV dyssynchrony achieved. This hemodynamic effect was independent of baseline clinical parameters, except for age, thus reinforcing the hypothesis of a causal relation between LV synchronicity and cardiac function. The hemodynamic effect was also independent of dyssynchrony evaluated before ablation, probably because AV junction ablation and pacing from the RV apex determine a change in the activation sequence of the left ventricle, which makes any preablation stratification unreliable. Admittedly, the inability to identify other parameters predictive of acute outcome due to the small study size (i.e., type II error) cannot be ruled out. The prerequisite and the rationale for the benefit of cardiac resynchronization therapy is that it is able to resynchronize LV walls that have delayed activation. We designed CRP to be programmed in a way that is not typically performed in clinical practice, with VV optimization guided by intra-LV electromechanical activation; programming was performed not to maximize stroke volume but rather to minimize dyssynchrony. Echocardiographically optimized CRP has been seen to reduce LV dyssynchrony values in most patients, almost to the values observed in normal

control subjects.13 The values of synchronicity achieved through optimized CRP were also 40% better than those obtained during simultaneous biventricular pacing (see “Methods”). Interventricular dyssynchrony also normalized as a consequence of biventricular pacing. However, in this study, the results of VV optimization were not compared with simultaneous biventricular pacing, and therefore we were unable to evaluate the incremental contribution of VV optimization on top of biventricular pacing. Whatever the importance of VV optimization, our results show an association between reduction in LV dyssynchrony and improvement in cardiac function. We can therefore suppose that the reduction in LV dyssynchrony (LV resynchronization) achieved with CRP plays a major role in the response to cardiac pacing therapy in patients with permanent AF and that this reduction is independent of the clinical features of the patients. Recently, Bleeker et al22 reported similar results in patients in sinus rhythm; immediately after implantation, the responders to CRP (defined as reduction in LV end-systolic volume) displayed a significant reduction in LV dyssynchrony, whereas the nonresponders did not. An association between reduced intra-LV dyssynchrony after biventricular pacing and improved cardiac function has also been shown by others.12,14,23–26 Interventricular resynchronization also seems to be correlated with a favorable outcome.27 In the present study, most patients underwent AV junction ablation primarily to achieve control and regulation of their irregular heart rates, and only a minority met the criteria for cardiac resynchronization therapy recommended by European guidelines21 (i.e., EF ⱕ35% and New York Heart Association class ⱖ3). Therefore, the study design is somewhat different from the typical studies on cardiac resynchronization therapy and should be regarded mainly as a study on rate control and regularization therapy in which resynchronization was used with the intent to counteract a potential negative effect of RV apical pacing. Does cardiac resynchronization therapy have different indications in AF from those applicable in sinus rhythm? The patients with AF who had conventional indications for CRP according to current guidelines21 showed acute hemodynamic changes of a similar extent to the patients who are not usually considered to be indicated for CRP; moreover, the acute outcomes of patients with narrow QRS complexes were the same as,

Arrhythmias and Conduction Disturbances/CRP in Patients With Permanent AF

or even better than, those of patients with wide QRS complexes. This observation is consistent with that of another previous small study.16,19 In patients in sinus rhythm, the efficacy of resynchronization is questionable in patients with preserved systolic function or with narrow QRS complexes.21,28 Nevertheless, the study size was small, and we might simply have been unable to identify distinct and important components of the total study cohort (i.e., type II error). For these reasons, these data need to be confirmed in large randomized prospective trials before becoming established. The tissue Doppler imaging method for the study of LV dyssynchrony is not established and differs, even markedly, among studies; it is therefore difficult to make comparisons among studies. The level of the ventricle at which to assess dyssynchrony is unclear. The LV base contains the largest myocardial mass and is potentially of most hemodynamic significance. Basal rather than midlevel assessment may therefore be more important in predicting response to cardiac resynchronization therapy.29 The number of ventricular walls requiring assessment is equally unclear. A compromise is necessary between the optimal detection of mechanical delay and the feasibility of examining numerous segments in clinical practice.29 Interrogation of only 2 segments may overlook a significant proportion of delayed myocardium, thus inadequately assessing dyssynchrony. Various software algorithms have been developed to measure dyssynchrony automatically. How these algorithms correlate among themselves and with the manual measurement of the time to the onset of systolic velocity is unclear. Many studies have used the time to peak systolic velocity to assess LV dyssynchrony; we used the time to the onset of systolic velocity, mainly because the onset of systolic velocity may be easier to identify than its peak, especially in patients with severe LV dysfunction. Bordachar et al12 have demonstrated that the 2 indexes are equally sensitive to patients’ hemodynamic status. The study of LV dyssynchrony may be marred by low reproducibility and the difficulty of obtaining reliable measurements in individual subjects. For these reasons, some experts have expressed concern as to its real applicability in clinical practice. Although a good interobserver correlation is usually reported in single-center studies, the intercenter correlation is likely to be worse, arousing concern about the reproducibility of measurements in different hospitals. To limit these intrinsic problems, 2 pilot studies6,13 were previously performed to validate the method and to calculate the normal range of intra-LV delay in normal subjects and its variations in various diseases. Moreover, all participating centers were trained in the method, and quality control was centralized. The results of each single center were compared with those of the center that recruited the most patients and of the pilot studies. When discrepancies were observed, these were corrected when possible. Finally, we found it impossible to obtain reliable echocardiographic images in some patients because of technical problems (e.g., bad signal due to poor chest window or impossibility to detect the regional systolic velocity wave in some segments). These patients were excluded from the analysis. The exclusion rate was 15% (9 patients).

859

Acknowledgment: We wish to thank Nicoletta Grovale, Tiziana De Santo, and Paola Di Stefano of Medtronic Italy SpA (Milan, Italy) for their help in clinical monitoring and assistance in statistical analysis. Appendix Study organization: The APAF study is a self-sponsored study. The Steering Committee designed the study protocol as described in this report. All data were collected and managed electronically by an external company (Demiurg Clinical Technologies, Barcelona, Spain) and processed by the Study Coordination Center in Lavagna, Italy. The analysis of data, including statistical analysis, was performed by the End-Point Committee (Michele Brignole, Carlo Menozzi, Lluís Mont, and Panos Vardas), with the organizational support of Medtronic Italy SpA. Steering Committee: Michele Brignole (principal investigator), Carlo Menozzi (principal investigator), Gianluca Botto, Giuseppe De Fabrizio, Lluís Mont (coordinator for Spain), Alessandro Proclemer, and Panos Vardas (coordinator for Greece). End-Point Committee: Michele Brignole, Carlo Menozzi, Lluís Mont, and Panos Vardas. APAF Acute-Echo study participating centers and investigators (ordered by number of patients enrolled): Ospedali del Tigullio, Lavagna, Italy: R. Bollini, G. Lupi, D. Oddone, R. Maggi, and M. Brignole; Ospedale S. Maria Nuova, Reggio Emilia, Italy: O. Gaddi, A. Navazio, and C. Menozzi; Ospedale S. Anna, Como, Italy: M. Luzi, S. Molteni, and G.L. Botto; Hospital Universitario La Fe, Valencia, Spain: J. Osca Asensi; Hospital Universitario Virgen de Valme, Sevilla, Spain: D. García Medina; Hospital Clinic i Provincial, Barcelona, Spain: L. Mont, M. Sitges, and B. Vidal; Clinica S. Anna, Catanzaro, Italy: S. Iacopino and R. Alemanni; Heraklion University Hospital, Heraklion, Greece: F.I. Parthenakis, A.P. Patrianakos, and P.E. Vardas; S. Maria Della Misericordia, Udine, Italy: L. Badano and A. Proclemer; and Ospedale S. Croce e Carle, Cuneo, Italy: G. Rossetti and A. Vado. 1. Brignole M, Menozzi C, Gianfranchi L, Musso G, Mureddu R, Bottoni N, Lolli G. Assessment of atrioventricular junction ablation and VVIR pacemaker versus pharmacological treatment in patients with heart failure and chronic atrial fibrillation: a randomized, controlled study. Circulation 1998;98:953–960. 2. Brignole M, Gianfranchi L, Menozzi C, Alboni P, Musso G, Bongiorni MG, Gasparini M, Raviele A, Lolli G, Paparella N, Acquarone S. Assessment of atrioventricular junction ablation and DDDR modeswitching pacemaker versus pharmacological treatment in patients with severely symptomatic paroxysmal atrial fibrillation: a randomized controlled study. Circulation 1997;96:2617–2624. 3. Kay GN, Ellenbogen KA, Giudici M, Redfield MM, Jenkins LS, Mianulli M, Wilkoff B; APT Investigators. The Ablate and Pace Trial: a prospective study of catheter ablation of the AV conduction system and permanent pacemaker implantation for treatment of atrial fibrillation. J Interv Card Electrophysiol 1998;2:121–135. 4. Wood MA, Brown-Mahoney C, Kay GN, Ellenbogen KA. Clinical outcomes after ablation and pacing therapy for atrial fibrillation: a meta-analysis. Circulation 2000;101:1138 –1144. 5. Vernooy K, Dijkman B, Cheriex EC, Prinzen FW, Crijns HJ. Ventricular remodeling during long-term right ventricular pacing following His bundle ablation. Am J Cardiol 2006;97:1223–1227. 6. Lupi G, Sassone B, Badano L, Peraldo C, Gaddi O, Sitges M, Parthenakis F, Molteni S, Pagliuca MR, Grovale N, et al. Effects of right

860

7.

8. 9.

10.

11.

12.

13.

14.

15.

16.

17.

The American Journal of Cardiology (www.AJConline.org) ventricular pacing on intra-left ventricular electromechanical activation in patients with native narrow QRS. Am J Cardiol 2006;98:219 –222. Tops LF, Schalij MJ, Holman ER, van Erven L, van der Wall EE, Bax JJ. Right ventricular pacing can induce ventricular dyssynchrony in patients with atrial fibrillation after atrioventricular node ablation. J Am Coll Cardiol 2006;48:1642–1648. APAF: assessment of cardiac resynchronization therapy in patients with permanent atrial fibrillation. Available at: http://www.clinicaltrials.gov/ ct2/show/NCT00111527?id⫽1893A&rank⫽1. Accessed 2005. Schiller NB, Shah PM, Crawford M, DeMaria A, Devereux R, Feigenbaum H, Gutgesell H, Reichek N, Sahn D, Schnittger I. Recommendations for quantitation of the left ventricle by two-dimensional echocardiography. American Society of Echocardiography Committee on Standards, Subcommittee on Quantitation of Two-Dimensional Echocardiograms. J Am Soc Echocardiogr 1989;2:358 –367. Notabartolo D, Merlino JD, Smith AL, DeLurgio DB, Vera FV, Easley KA, Martin RP, León AR. Usefulness of the peak velocity difference by tissue Doppler imaging technique as an effective predictor of response to cardiac resynchronization therapy. Am J Cardiol 2004;94: 817– 820. Ansalone G, Giannantoni P, Ricci R, Trambaiolo P, Fedele F, Santini M. Doppler myocardial imaging to evaluate the effectiveness of pacing sites in patients receiving biventricular pacing. J Am Coll Cardiol 2002;39:489 – 499. Bordachar P, Lafitte S, Reuter S, Sanders P, Jaïs P, Haïssaguerre M, Roudaut R, Garrigue S, Clementy J. Echocardiographic parameters of ventricular dyssynchrony validation in patients with heart failure using sequential biventricular pacing. J Am Coll Cardiol 2004;44:2157– 2165. Badano L, Gaddi O, Peraldo C, Lupi G, Sitges M, Parthenakis F, Molteni S, Pagliuca MR, Sassone B, Di Stefano P, et al. Left ventricular electromechanical activation in patients with heart failure and normal QRS duration, and in patients with wide QRS complexes. Europace 2007;9:41– 47. Vanderheyden M, De Backer T, Rivero-Ayerza M, Geelen P, Bartunek J, Verstreken S, De Zutter M, Goethals M. Tailored echocardiographic interventricular delay programming further optimizes left ventricular performance after cardiac resynchronization therapy. Heart Rhythm 2005;2:1066 –1072. Brignole M, Oddone D, Maggi R, Lupi G, Bollini R, Corallo S, Robotti S, Solano A, Donateo P, Croci F. Resynchronization of the left ventricular contraction by tailored programming of right and left ventricular pacing. Europace 2008;10:489 – 495. Puggioni E, Brignole M, Gammage M, Soldati E, Bongiorni MG, Simantirakis EN, Vardas P, Gadler F, Bergfeldt L, Tomasi C, et al. Acute comparative effect of right and left ventricular pacing in patients with permanent atrial fibrillation. J Am Coll Cardiol 2004;43: 234 –238. The Criteria Committee of the New York Heart Association. Nomenclature and Criteria for Diagnosis of Diseases of the Heart and Great Vessels. 9th ed. Boston, Massachusetts: Little, Brown, 1994:253–256.

18. Rector TS, Kubo SH, Cohn JN. Patients’ self-assessment of their congestive heart failure. Part II: content, reliability and validity of a new measure, the Minnesota Living With Heart Failure Questionnaire. Heart Fail 1987;3:198 –209. 19. Brignole M, Gammage M, Puggioni E Alboni P, Raviele A, Sutton R, Vardas P, Bongiorni MG, Bergfeldt L, Menozzi C, Musso G. Comparative assessment of right, left and bi-ventricular pacing in patients with permanent atrial fibrillation. Eur Heart J 2005;26:712–722. 20. Opasich C, Pinna D, Mazza A, Febo O, Riccardi PG, Capomolla S, Cobelli F, Tavazzi L. Reproducibility of the six-minute walking test in patients with chronic congestive heart failure: practical implications. Am J Cardiol 1998;81:1497–1500. 21. Vardas P, Auricchio A, Blanc JJ, Daubert JC, Drexler H, Ector H, Gasparini M, Linde C, Morgado FB, Oto A, et al. Guidelines for cardiac pacing and cardiac resynchronization therapy. Eur Heart J 2007;28:2256 –2295. 22. Bleeker G, Mollema S, Holman E, Van de Veire N, Ypenburg C, Boersma E, van der Wall EE, Schalij MJ, Bax JJ. Left ventricular resynchronization is mandatory for response to cardiac resynchronization therapy analysis in patients with echocardiographic evidence of left ventricular dyssynchrony at baseline. Circulation 2007;116:1440 – 1448. 23. Yu CM, Fung WH, Lin H, Zhang Q, Sanderson J, Lau CP. Predictors of left ventricular reverse remodeling after cardiac resynchronization therapy for heart failure secondary to idiopathic dilated or ischemic cardiomyopathy. Am J Cardiol 2002;91:684 – 688. 24. Schuster P, Faerestrand S, Ohm O. Reverse remodeling of systolic left ventricular contraction pattern by long term cardiac resynchronization therapy: colour Doppler shows resynchronization. Heart 2004;90: 1411–1416. 25. Breithard OA, Stellbrink C, Herbots L, Claus P, Sinha AM, Bijnens B, Hanrath P, Sutherland GR. Cardiac resynchronization therapy can reverse abnormal myocardial strain distribution in patients with heart failure and left bundle branch block. J Am Coll Cardiol 2003;42:486 – 494. 26. Vidal B, Sitges M, Marigliano A, Delgado V, Díaz-Infante E, Azqueta M, Tamborero D, Tolosana JM, Berruezo A, Pérez-Villa F, et al. Optimizing the programmation of cardiac resynchronization therapy in patients with heart failure and left bundle branch block. Am J Cardiol 2007;100:1002–1006. 27. Richardson M, Freemantle N, Calvert M, Cleland J, Tavazzi L. Predictors and treatment response with cardiac resynchronization therapy in patients with heart failure characterized by dyssynchrony: a predefined analysis form the CARE-HF trial. Eur Heart J 2007;28:1827– 1834. 28. Beshai J, Grimm R, Nagueh S, Baker JH II, Beau SL, Greenberg SM, Pires LA, Tchou PJ. Cardiac-resynchronization therapy in heart failure with narrow QRS complexes. N Engl J Med 2007;357:2461–2471. 29. Hawkins N, Petrie M, MacDonald M, Hogg1 K, McMurray J. Selecting patients for cardiac resynchronization therapy: electrical or mechanical dyssynchrony? Eur Heart J 2006;27:1270 –1281.