Journal of the American College of Cardiology Vol. 48, No. 4, by the American College of Cardiology Foundation ISSN /06/$32.

Transcription

1 Journal of the American College of Cardiology Vol. 48, No. 4, by the American College of Cardiology Foundation ISSN /06/$32.00 Published by Elsevier Inc. doi: /j.jacc Four-Year Efficacy of Cardiac Resynchronization Therapy on Exercise Tolerance and Disease Progression The Importance of Performing Atrioventricular Junction Ablation in Patients With Atrial Fibrillation Maurizio Gasparini, MD,* Angelo Auricchio, MD, PHD, François Regoli, MD,* Cecilia Fantoni, MD, Mihoko Kawabata, MD, Paola Galimberti, MD,* Daniela Pini, MD,* Carlo Ceriotti, MD,* Edoardo Gronda, MD,* Catherine Klersy, MD, MSC, Simona Fratini, MD, Helmut H. Klein, MD Milan and Pavia, Italy; Magdeburg, Germany; and Lugano, Switzerland OBJECTIVES BACKGROUND METHODS RESULTS CONCLUSIONS The goal of this study was to investigate the effects of cardiac resynchronization therapy (CRT) in heart failure patients with permanent atrial fibrillation (AF) and the role of atrioventricular junction (AVJ) ablation. Cardiac resynchronization therapy has been proven effective in heart failure patients with sinus rhythm (SR). However, little is known about the effects of CRT in heart failure patients with permanent AF. Efficacy of CRT on ventricular function, exercise performance, and reversal of maladaptive remodeling process was prospectively compared in 48 patients with permanent AF in whom ventricular rate was controlled by drugs, thus resulting in apparently adequate delivery of biventricular pacing ( 85% of pacing time), and in 114 permanent AF patients, who had undergone AVJ ablation (100% of resynchronization therapy delivery). The clinical and echocardiographic long-term outcomes of both groups were compared with those of 511 SR patients treated with CRT. Both SR and AF groups showed significant and sustained improvements of all assessed parameters (model p for all parameters). However, within the AF group, only patients who underwent ablation showed a significant increase of ejection fraction (p 0.001), reverse remodeling effect (p 0.001), and improved exercise tolerance (p 0.001); no improvements were observed in AF patients who did not undergo ablation. Heart failure patients with ventricular conduction disturbance and permanent AF treated with CRT showed large and sustained long-term (up to 4 year) improvements of left ventricular function and functional capacity, similar to patients in SR, only if AVJ ablation was performed. (J Am Coll Cardiol 2006;48:734 43) 2006 by the American College of Cardiology Foundation Cardiac resynchronization therapy (CRT) is an effective therapy in symptomatic, drug-refractory heart failure patients with prolonged QRS and low ejection fraction (1 8). Long-term results of CRT on exercise tolerance and disease progression as evaluated by reversal of maladaptive remodeling process are rather limited (4,9,10), and mostly re- See page 744 ported in patients with sinus rhythm (SR). Permanent or persistent atrial fibrillation (AF) is a rather common supraventricular arrhythmia in heart failure patients and is associated with increased morbidity and mortality (11 13). Some studies have reported significant acute (14 16) and short-term (4,9,10) benefits of CRT in patients with AF and advanced heart failure, while the long-term effects of From the *Department of Cardiology IRCCS Istituto Clinico Humanitas Rozzano-Milano, Milan, Italy; Service of Biometry & Clinical Epidemiology, IRCCS Policlinico San Matteo, Pavia, Italy; Division of Cardiology, University Hospital, Magdeburg, Germany; and the Department of Cardiology, Fondazione Cardiocentro Ticino, Lugano, Switzerland. Manuscript received November 11, 2005; revised manuscript received February 10, 2006, accepted March 16, CRT on ventricular function and dimensions in this patient subgroup are nearly unknown. Beta-blocking agents, digoxin, and amiodarone usually achieve adequate heart rate control at rest, and, thus, relieve symptoms in patients with permanent AF. Heart rate may be controlled at rest but not entirely during exercise (17). The ablate and pace strategy is considered the treatment of choice in those patients with rapidly conducting AF in which pharmacologic control of heart rate cannot be successfully achieved. This approach provides immediate relief of symptoms and large increases in exercise tolerance (18 21). Whether CRT provides similar benefits in AF patients treated with medication for heart rate control (permitting 85% of the biventricular pacing time) or whether performing atrioventricular junction (AVJ) ablation may provide additional benefits is still unknown. The aim of this study was to prospectively investigate the long-term effects of CRT on ventricular function, magnitude of reverse remodeling, and exercise tolerance in patients with AF by comparing patients who underwent AVJ ablation with those in whom adequate rate control was achieved by combining negative chronotropic drugs with

2 JACC Vol. 48, No. 4, 2006 August 15, 2006: Gasparini et al. Cardiac Resynchronization in AF 735 Abbreviations and Acronyms AF atrial fibrillation AVJ atrioventricular junction CRT cardiac resynchronization therapy LVEDD left ventricular end-diastolic diameter LVEF left ventricular ejection fraction LVESV left ventricular end-systolic volume NYHA New York Heart Association SR sinus rhythm VRR ventricular rate regularization appropriate programming of device settings. These longterm echocardiographic and functional changes in patients with permanent AF were compared with those in SR treated with CRT. METHODS Patient population. Between August 1995 and December 2004, 673 consecutive symptomatic heart failure patients were treated with CRT at 2 European centers. All patients had a poor systolic left ventricular ejection fraction (LVEF) 35%, ventricular conduction delay (QRS duration 120 ms), and were in New York Heart Association (NYHA) functional class II despite optimal drug therapy, which included beta-adrenergic blockers, angiotensin-converting enzyme inhibitors or angiotensin receptor blockers, diuretics, and spironolactone. All patients had at least 1 hospitalization because of heart failure 12 months before device implantation. A total of 162 patients were in permanent AF, and 511 patients were in SR. All permanent AF patients received negative chronotropic drugs at pre-hospital discharge in order to achieve rate control, thus maximizing biventricular pacing delivery. At the 2-month follow-up visit, device interrogation of AF patients was performed, and the percentage of biventricular pacing time since implantation was recorded. An arbitrary cutoff rate of biventricular pacing 85% of pacing time was considered acceptable. When biventricular pacing percentage was 85%, then AVJ ablation was prospectively performed. Subsequently, in these patients digoxin and amiodarone were suspended (amiodarone was continued only if ventricular tachyarrhythmias were present), while beta-blocker therapy was maintained. Due to the potentially serious underlying clinical implications involved in AVJ ablation (i.e., very slow heart rate and pacemaker dependency), the indication for ablation was based on careful examination of effective biventricular stimulation percentage and directly checked by the 2 most expert physicians of each center. Because this approach was not standardized clinical care, formal approval was obtained from the review boards of each institution. All patients gave their written informed consent to undergo the ablation procedure under the understanding that irreversible complete AV block would be created with subsequent pacemaker dependency. Measures of clinical outcome. Before device implantation, all patients underwent 12-lead electrocardiogram (ECG) recording, evaluation of NYHA functional class, assessment of 6-min walked distance (22), or symptomlimited peak oxygen consumption (23). Changes during follow-up of these functional parameters were expressed under a unique parametric index (i.e., the functional capacity score [see Statistical Analysis section for details]). Furthermore, echocardiographic examination was performed according to American Society of Echocardiography guidelines (24). The following parameters were collected: left ventricular end-diastolic volume (LVEDV) and left ventricular end-diastolic diameter (LVEDD), left ventricular end-systolic volume (LVESV) and left ventricular endsystolic diameter (LVESD), LVEF, and degree (if present) of mitral valve regurgitation. In both institutions functional and echocardiographic evaluations were performed in outpatient clinic settings where the physicians were blinded to the AVJ ablation therapeutic scheme. Etiology was assessed in all cases by coronary angiography, and causes of heart failure amenable to surgery or intervention were corrected at least 6 months before device implantation. Because indication for cardioverter-defibrillator therapy changed over time, patients received the most appropriate device based on currently available evidence and guidelines. Device implantation and programming. Conventional ventricular leads were positioned in the right ventricular apex in all patients implanted with CRT systems, and conventional atrial leads were used only in SR patients. In those patients in whom implantable defibrillator was necessary, the defibrillation lead was usually positioned in the right ventricular apex. Transvenous left ventricular lead positioning was guided by coronary sinus venogram, preferring a lateral or posterolateral vein as implantation site. In case of failure or technical difficulties (phrenic nerve stimulation, inability to cannulate the coronary sinus, and so on) of the transvenous approach, epicardial screw-in or steroideluting passive lead was implanted via a limited thoracotomy. The latter was the elective approach used before transvenous lead was routinely available. Follow-up was performed at 2, 6, and 12 months after implantation, and every 12 months thereafter. In patients with preserved SR, device was programmed in atrial-synchronous sequential pacing; programming of the AV delay was optimized either by echocardiographic method or by invasive measurements (25,6). In patients with AF, minimum heart rate was usually set at 70 beats/ min ( 80 beats for 2 weeks after AVJ ablation) (26). Maximum rate was set at 85% of the theoretical maximum heart rate; the rate adaptive response was turned on, and ventricular rate stabilization and/or ventricular rate regulation (VRR) was activated when available. Ventricular rate regulation was activated on the max value that usually determined a pacing rate 8% to 10% faster than intrinsic

3 736 Gasparini et al. JACC Vol. 48, No. 4, 2006 Cardiac Resynchronization in AF August 15, 2006: heart rate. It was assumed that careful programming of device settings combined with prescription of negative chronotropic drugs would allow achievement of adequate heart rate control for effective biventricular pacing. Biventricular pacing percentage was accurately assessed through device counters analysis performed at the 2-month visit. In early generation devices, correct evaluation of the effective percentage of left ventricle stimulation was sometimes difficult due to the presence of dual cathodal systems. The absence of dedicated AF algorithms, in these devices, only rarely permitted to obtain a percentage of biventricular pacing 85%. New and improved device technology, by using separate right ventricular and left ventricular channels, VRR, and the capability of detecting triggered beats, has markedly increased the accuracy of counters, and has permitted the achievement of higher percentages of biventricular pacing. Nonetheless, even in the most recent devices, the possible presence of fusion or pseudofusion beats may overestimate the percentage of true and effective biventricular stimulation. Definition of responders. Responders to CRT were considered those patients who showed a reduction in LVESV 10% relative to baseline (27). Assessments for responder status were performed at 6, 12, and every 12 months thereafter. Repetitive response assessments were obtained in all patients who reached 1-year follow-up or more. Statistical analysis. Descriptive statistics were reported as mean and SD for continuous variables, unless skewed when median and quartiles were computed. Categorical variables were summarized as absolute and relative frequencies (%). Baseline characteristics of patients with SR and AF, with and without AVJ ablation, were compared by means of Student t test or Mann-Whitney U test for continuous variables and Fisher exact test for categorical variables. Changes over time of cardiac performance, and role of AF or of AVJ ablation, were assessed by fitting a generalized linear regression model. Intrapatient correlation due to the repeated measures design was accounted for by calculating Huber-White robust standard errors. The following possible confounders were also included in the models: center, age, gender, diagnosis of coronary artery disease, and baseline left ventricle ejection fraction. To evaluate whether changes in time were different between groups, interactions between time and AF or between time and AVJ ablation were also assessed. Functional assessment was performed either by means of 6-min walking test or by means of symptom-limited peak oxygen consumption. The choice of performing one test over the other was mainly determined by the availability of the equipment for cardiopulmonary testing: in the earlier phase of the study, 6-min walking was prevalently performed, while the more complete and sophisticated cardiopulmonary test was performed in the later phase. In the single patient, the same criteria to assess functional capacity at baseline were utilized throughout subsequent follow-up. In order to obtain a unique functional assessment index for the entire case series, both variables were converted to a 1 to 100 scale, by applying the following transformation: functional capacity score 100 (observed value minimum observed value)/range of observed values (28). All tests were 2-sided. A p value 0.05 was retained as indicative of statistical significance. Bonferroni correction for multiple tests was used for post-hoc comparisons over time (5 comparisons). Stata 8 (Stata Corp., College Station, Texas) was used for computation. RESULTS Baseline characteristics of AF and SR patients enrolled at each center did not statistically differ. Some statistically significant, yet probably clinically less relevant, differences appeared when the populations of the 2 centers were combined. However, the parameters accounting for differences were evenly distributed between the 2 centers. Age ( years vs years, p ), NYHA functional class III to IV (98% vs. 86%, p ), LVEF ( % vs %, p 0.001), and LVEDD (69 9 mm vs mm, p ) differed between the 2 centers. After 2 months of CRT, careful evaluation of device counters revealed continuous biventricular pacing activity (mean %) in SR patients. A low proportion of spontaneous intrinsic ventricular beats was detected, most likely due to ventricular premature beats. In no case was the proportion of ventricular premature beats found to be 5%. In 114 patients with AF, a global biventricular percentage of 85% (mean %) was detected which, by prospective study design, led to AVJ ablation. At the following 6-month control after ablation, device counters revealed full biventricular pacing effectiveness, with percentages nearing 100% (mean %). In the remaining 48 patients with AF, the combination of programmed device features (minimum and maximum heart rate, rate adaptive response up to 85% of the theoretical maximum heart rate, and VRR) and negative chronotropic drug therapy determined 85% biventricular pacing percentage (mean %) and satisfactory heart rate control; mean resting heart rate was beats/min in this patient subgroup. Clinical outcome of SR and AF patients. Table 1 shows the baseline characteristics of patients with SR and AF. Overall, AF patients were significantly older (p 0.002), more frequently men (p 0.019), clinically more compromised, with a higher mean NYHA functional class, and poorer functional capacity (functional capacity score was different, p 0.014), but presented less frequently with coronary artery disease (p 0.029). More SR patients were implanted with a device combining CRT and defibrillation (p 0.036). Pharmacologic agents with negative chronotropic effect were significantly more utilized in AF patients (p ) (Table 1).

4 JACC Vol. 48, No. 4, 2006 August 15, 2006: Gasparini et al. Cardiac Resynchronization in AF 737 Table 1. Baseline Characteristics of SR and AF Patients SR (n 511) AF (n 162) p Value Follow-up (months)* 26.1 ( ) 24.6 ( ) Age (yrs) 63.4 (10.0) 66.0 (8.3) Male gender (%) 395 (77.3) 139 (85.8) Functional NYHA functional class III IV (%) 452 (89.7) 154 (96.9) QRS duration (ms) (29.5) (35.5) Coronary artery disease (%) 241 (47.2) 60 (37.1) LVEF (%) 26.0 (7.0) 26.3 (6.7) LVEDD (mm) 67.0 (9.0) 66.0 (8.8) LVESV (ml) 196 (53) 186 (45) Peak oxygen consumption (ml/min/kg) 13.1 (2.8) 12.1 (2.2) min walking distance (m) 332 (117) 311 (107) Functional capacity score 33.4 (7.3) 29.6 (6.5) CRT with a defibrillator (%) 299 (58.5) 79 (48.8) Beta-blockers (%) 424 (83.1) 135 (83.3) ACE inhibitors (%) 460 (90.4) 152 (93.8) Antialdosteronic agents (%) 267 (52.5) 85 (52.5) Diuretics (%) 449 (88.2) 148 (91.4) Negative chronotropic drugs (%) 435 (85.1) 160 (98.7) Lead position: lateral or posterolateral region (%) 454 (89) 142 (88) *Median and 25th 75th percentile (Mann-Whitney U test); mean and (SD) (Student t test); number (%) (Fisher exact test). ACE angiotensin-converting enzyme; AF atrial fibrillation; CRT cardiac resynchronization therapy; LVEDD left ventricular end-diastolic diameter; LVEF left ventricular ejection fraction; LVESV left ventricular end-systolic volume; NYHA New York Heart Association; SR sinus rhythm. During a mean follow-up of months, SR and AF patients showed highly significant and similar improvements in functional capacity, magnitude of reverse remodeling, left ventricular systolic function, and NYHA functional class (Table 2). The greatest changes in LVEF, reverse remodeling, and NYHA functional class occurred within the first 6 months and continued for up to 4 years in both groups. Moreover, there was an increase in walked distance by around 140 m and peak oxygen consumption by around 3 ml/kg/min at 6 months that remained stable during the follow-up, reflected by significant improvements of functional capacity score in both groups over 48 months. The proportion of responders to CRT at 1 year was similar for both groups (SR 69% vs. AF 60%, p 0.166) and remained fairly stable over time. Over 4 years of follow-up LVEF of non-responders decreased by a mean value of points compared with baseline. In responder patients, in contrast, a consistent and significant increase of mean LVEF ( points) was observed. Comparison of effects within the AF group. The baseline characteristics of AF patients treated with and without AVJ ablation are shown in Table 3. No significant difference was present at baseline between the 2 groups. Antiarrhythmic drugs as well as other pharmacologic agents, such as beta-blockers (no AVJ ablation 81.3% vs. AVJ ablation 84.2%, p 0.65) and angiotensin-converting enzyme inhibitors (no AVJ ablation 95.8% vs. AVJ ablation 93.0%, p 0.72), were utilized by a similar proportion of patients in each group. In the 2 months after implantation and preceding the assessment phase of percentage of biventricular pacing time, virtually all AF patients (99%) were in therapy with negative chronotropic drugs (Table 1). After adequate and conventional loading doses, mean daily maintenance dose of amiodarone was 185 mg, of digoxin was mg, and of carvedilol was 17.5 mg. By prospective design, patients who failed the 85% biventricular pacing criteria underwent AVJ ablation, and, subsequently, negative chronotropic drug therapy was discontinued; amiodarone was continued only in cases who presented ventricular tachyarrhythmias, while beta-blocker was maintained. Only the ablated group of patients showed a significant improvement of LVEF (p 0.001), LVESV (p 0.001), and exercise capacity (p 0.001) over the follow-up period. Magnitude of improvement of LVEF, LVESV, NYHA functional class, and of functional capacity were significantly larger for ablated patients (Table 4, Fig. 1). Moreover, there was a significantly higher proportion of responder patients at 12 months in the ablated group (68%) compared with the non-ablated group (18%) (p 0.001); this significant difference was maintained throughout follow-up (p 0.001) (Fig. 2). When considering patients who died, the difference in proportion of responders between the 2 groups remained highly significant (p 0.001; odds ratio 11.12, 95% confidence interval 4.03 to 25.35). DISCUSSION This is the first long-term, prospectively designed study demonstrating the importance of AVJ ablation in AF patients treated with CRT. Our study showed that the

5 738 Gasparini et al. JACC Vol. 48, No. 4, 2006 Cardiac Resynchronization in AF August 15, 2006: Table 2. Sequential Changes of Left Ventricular Function and Dimension, NYHA Functional Class, and Physical Performance as Assessed by Functional Capacity Score in Patients With SR or AF: Analysis Adjusted for Center, Baseline LVEF, Age, Gender, and Etiology SR Mean (SD) AF Mean (SD) Model p Value* AF p Value Time Change p Value and Post-Hoc p Values Time-Rhythm Interaction p Value LVEF (n 673) Baseline (%) 26.0 (7.0) 26.3 (6.7) 6 months 31.6 (10.7) 30.5 (11.7) 6 vs. baseline months 33.5 (12.2) 33.8 (13.3) 12 vs. 6 months months 34.4 (13.5) 36.1 (14.4) 24 vs. 12 months months 35.8 (14.3) 38.0 (14.7) 36 vs. 24 months months 33.2 (13.8) 35.2 (14.7) 48 vs. 36 months LVESV (n 673) Baseline (mm) 196 (53) 186 (45) Change from baseline At 6 months (%) 16 (19) 12 (17) 6 vs. baseline At 12 months (%) 23 (21) 14 (25) 12 vs. 6 months At 24 months (%) 26 (24) 21 (19) 24 vs. 12 months At 36 months (%) 31 (22) 27 (18) 36 vs. 24 months At 48 months (%) 31 (25) 25 (19) 48 vs. 36 months NYHA functional class (n 673) Baseline (% III IV) months vs. baseline months vs. 6 months months vs. 12 months months vs. 24 months months vs. 36 months Functional capacity score (n 673) Baseline 33.4 (7.3) 29.6 (6.5) 6 months 40.7 (8.1) 37.4 (8.2) 6 vs. baseline months 42.6 (8.1) 40.1 (8.2) 12 vs. 6 months months 44.3 (7.3) 42.1 (7.8) 24 vs. 12 months months 45.4 (8.2) 41.4 (8.0) 36 vs. 24 months months 43.0 (7.8) 40.4 (9.1) 48 vs. 36 months *Model p value (overall difference between patients with SR and with AF over time [main effects]); p value for changes over time in the whole case series; p value for differences between patients with SR and patients with AF; for Bonferroni correction, statistical significance for post-hoc tests: p 0.01 for comparison of consecutive assessments (5 comparisons); p value for assessing the interaction of rhythm (SR/AF) and time. Abbreviations as in Table 1. magnitude of benefit with CRT was similar between SR and permanent AF. However, among patients with permanent AF, only those patients undergoing ablation of AVJ demonstrated a significant symptomatic benefit and improvement of left ventricular function. The symptomatic benefit, improvement of exercise tolerance, and reversal of maladaptive remodeling process provided by CRT was maintained up to 4 years after therapy initiation. Whether these beneficial effects will result in different survival outcome remains to be addressed. Table 3. Baseline Characteristics of AF Patients With and Without AVJ Ablation AF No AVJ Ablation (n 48) AF AVJ Ablation (n 114) p Value Age (yrs)* 64.1 (6.3) 66.8 (9.0) Male gender (%) 40 (83.3) 99 (86.8) NYHA functional class III IV (%) 48 (100) 106 (95.5) QRS duration (ms)* (32.0) (36.5) Coronary artery disease (%) 19 (39.6) 41 (36.0) LVEF (%)* 25.1 (5.7) 26.8 (7.1) LVEDD (mm)* 67.9 (8.4) 65.3 (8.9) LVESV (ml)* 196 (50) 182 (49) Peak oxygen consumption (ml/min/kg)* 12.1 (2.2) 12.1 (2.1) min walking distance (m)* 313 (72) 311 (68) Functional capacity score 24.6 (6.9) 31.7 (7.1) Negative chronotropic drugs (%) 48 (100) 112 (98) 1.0 *Mean and SD (Student t test); number (%) (Fisher exact test); adjusted for mode of functional assessment. AVJ atrioventricular junction; other abbreviations as in Table 1.

6 JACC Vol. 48, No. 4, 2006 August 15, 2006: Gasparini et al. Cardiac Resynchronization in AF 739 Table 4. Sequential Changes of Left Ventricular Function and Dimension, NYHA Functional Class, and Physical Performance as Assessed by Functional Capacity Score in Patients With and Without AVJ Ablation: Analysis Adjusted for Center, Baseline LVEF, Age, Gender, and Etiology Variables AF no AVJ Ablation (n 48) AF AVJ Ablation (n 114) Model p Value* Ablation p Value Time Change p Value and Post-Hoc p Values Time AVJ Ablation Interaction p Value LVEF (n 162) Baseline 25.1 (5.7) 26.8 (7.1) 6 months 22.4 (8.1) 33.9 (11.4) 6 vs. baseline months 23.7 (9.4) 37.1 (12.7) 12 vs. 6 months months 24.1 (10.8) 39.7 (13.4) 24 vs. 12 months months 25.8 (10.6) 41.7 (13.8) 36 vs. 24 months months 25.0 (12.8) 39.5 (13.6) 48 vs. 36 months LVESV (n 162) Baseline (mm) 196 (50) 182 (49) Changes from baseline At 6 months (%) 2 (11) 16 (18) 6 vs. baseline At 12 months (%) 3 (14) 18 (28) 12 vs. 6 months At 24 months (%) 1 (17) 27 (18) 24 vs. 12 months At 36 months (%) 5 (14) 34 (18) 36 vs. 24 months At 48 months (%) 2 (20) 36 (16) 48 vs. 36 months NYHA functional class (n 162) Baseline (% III IV) months vs. baseline months vs. 6 months months vs. 12 months months vs. 24 months months vs. 36 months Functional capacity score (n 162) Baseline 24.6 (3.5) 31.7 (7.1) 6 months 25.2 (4.1) 42.5 (8.1) 6 vs. baseline months 27.5 (4.8) 44.2 (8.0) 12 vs. 6 months months 28.6 (5.1) 46.3 (7.3) 24 vs. 12 months months 28.8 (4.1) 45.2 (7.9) 36 vs. 24 months months 25.3 (2.5) 46.5 (8.9) 48 vs. 36 months *Model p value (overall difference between patients with and without AVJ ablation [main effects]); p value for changes over time in the whole case series; p value for differences between patients with and without AVJ ablation; for Bonferroni correction, statistical significance for post-hoc tests: p 0.01 for comparison of consecutive assessments (5 comparisons); p value for assessing the presence of interaction of ablation (yes/no) and time. AVJ atrioventricular junction; other abbreviations as in Table 1. Long-term efficacy of CRT. Our findings expand previous knowledge of short-term improvement, showing how the benefits are sustained up to 4 years after the initiation of CRT. Overall, CRT promoted reversal of maladaptive remodeling process of the left ventricle as shown by increases in LVEF and decreases in LVESV. These changes were similar both in patients with AF and SR. Improvements in left ventricular function and exercise capacity observed in our group of patients with SR were similar or even greater than those reported in several prospective, randomized trials (4,10,29). Moreover, our findings showed that, in SR patients, CRT has a favorable effect up to 4 years after therapy initiation. Our long-term data in patients with AF are novel, and showed a continuous increase of LVEF and LVESV reduction over a 4-year time span. Our results in these patients are partially in disagreement with those reported by Leclercq et al. (29) and by Linde et al. (4). Leclercq et al. (29) showed that, despite improvement of exercise tolerance in both patients with AF and SR, a significant amelioration of LVEF was obtained exclusively in AF patients. Linde et al. (4) reported only modest, non-significant changes in left ventricular dimensions 12 months after initiation of CRT despite significant reduction of mitral valve regurgitation in both SR and AF patients. Difference in sample size, longer follow-up, and management strategy of AF patients (AVJ ablation) may account for differences between our results and the findings by others (4,29). One recent study (10) reported some preliminary data on 30 heart failure patients with AF treated with CRT, a proportion of whom underwent AVJ ablation. A greater improvement of quality of life and a trend towards better response at 6 months follow-up in the subgroup that underwent AVJ ablation was observed. Cardiac resynchronization therapy enhanced exercise capacity assessed by 6-min walking distance or peak oxygen consumption. The long-lasting effect of CRT on exercise tolerance cannot be attributed to a placebo effect, given the fact that this effect tends to wane over time. Although none of the patients of this study underwent an additional controlled training program, the continuous increase of

7 740 Gasparini et al. JACC Vol. 48, No. 4, 2006 Cardiac Resynchronization in AF August 15, 2006: Figure 1. This figure presents sequential changes of left ventricular ejection fraction (A), exercise capacity (functional capacity score) (B), and changes in left ventricular end-systolic volume (C) in patients with sinus rhythm (SR) and atrial fibrillation (AF) with and without atrioventricular junction (AVJ) ablation. Significance values within the single groups are given in Tables 2 and 4. Also included (D) below the legend are the total number of patients reaching the different follow-up visits and the number of deaths that occurred in the time interval since last follow-up. exercise capacity over time was strongly suggestive that a training effect occurred after device implantation. Importance of AVJ ablation in AF patients with CRT. Our treatment approach in heart failure patients with AF was novel and different from the standard ablate and pace strategy. This latter approach is usually indicated in patients who have rapidly conducting AF not well controlled by pharmacologic therapy, eventually presenting with heart failure. Patients enrolled in these studies (18 21) were usually older, had an LVEF usually of around 40%, a shorter history of heart failure, and were receiving less medication for heart failure and more antiarrhythmic drugs. Conversely, our population (both SR and AF patients) was younger and had severely depressed left ventricular function; the reason for ablation was not drug-refractory control of ventricular rate during AF, but rather suboptimal biventricular pacing. Before ablation, the mean resting heart rate was about 80 beats/min, which was considerably lower than the ventricular rate of patients who undergo typical ablate and pace strategy, thus greatly reducing the likelihood that heart failure could be due to underlying tachycardiomyopathy. We observed that the effect of biventricular pacing, even when proven to be above 85% of pacing time, was not as effective as 100% of biventricular pacing time achievable only by ablating the AVJ. Drugs such as beta-blocking agents, digoxin, or amiodarone (mostly in combination) were given for the first months to all our AF patients in order to achieve adequate rate control, and, consequently, adequate percentage of biventricular pacing. We prospectively designed to ablate the AVJ in every patient who, at the 2-month follow-up visit, presented a percentage of effective biventricular pacing 85%. Based on this criterion, a minority of patients (only 29%) had satisfactory heart rate control achieved by drugs. These patients were similar in many clinical and echocardiographic parameters to patients who underwent ablation. Digoxin and amiodarone were discontinued in patients who underwent AVJ ablation (amiodarone was continued only if ventricular tachyarrhythmias were present), while beta-blockers were maintained. The percentage of responders in the ablated group was more than 3-fold that of the non-ablated AF group. The definition of responder to CRT differs widely in the literature (27,30 33). The proportion of non-responders in both SR and AF was either similar or somewhat lower than what has been previously reported (2,30). Long-term effects of CRT on exercise tolerance, NYHA functional class, ventricular function, and magnitude of reverse remodeling in patients with AF in whom the mean device-derived percentage of biventricular pacing was apparently well controlled ( 85% of daily time) by drugs was

8 JACC Vol. 48, No. 4, 2006 August 15, 2006: Gasparini et al. Cardiac Resynchronization in AF 741 Figure 2. Percentage of responders to cardiac resynchronization therapy (CRT) at different follow-up times are presented. The open bars refer to sinus rhythm patients, the ruled bars relate to atrial fibrillation with atrioventricular junction ablation, and, finally, the solid bars refer to atrial fibrillation without atrioventricular junction ablation. There was a significantly higher proportion of responders in patients with sinus rhythm or atrial fibrillation with atrioventricular junction ablation compared with atrial fibrillation patients without atrioventricular junction ablation. significantly lower than those of patients in whom ablation was performed (ensuring 100% CRT delivery). In contrast, effective and full biventricular capture achieved through AVJ ablation conferred marked improvements of both LVEF and left ventricular dimensions implying a possible prognostic benefit, because both these parameters have been correlated with favorable prognosis in heart failure patients undergoing CRT (27). Several reasons may account for the better functional outcome of patients who have been rendered pacemakerdependent by ablation. First and foremost, it is likely that complete rhythm regularization obtained through ablation, by optimizing diastolic and systolic phases of the cardiac cycle, improved global cardiac mechanics (21,34). Without ablation, relatively high biventricular pacing percentage was achieved at a cost of higher mean heart rates with deleterious effects on diastolic performance (35,36). Then, during exercise, heart rate control was usually difficult (17). The occurrence of fusion or pseudofusion beats between intrinsic conducted and paced beats (even obtained with the VRR setting) tends to overestimate true effective CRT delivery as determined by the surrogate parameter of percentage of biventricular stimulation time and may provide a less efficient resynchronization effect compared with full, regular, and continuous biventricular pacing. Taking into account the interference of intrinsic beats, a more plausible estimate of effective CRT dose may be lower than the device-derived biventricular pacing percentage, which averaged 88%. In this way, the outcome observed for the AF population not undergoing AVJ ablation may be easier to comprehend. A growing amount of evidence has reported the efficacy of the anatomic (37) or the electrophysiological ablation approach (38) in the treatment of paroxysmal and persistent AF. In particular, the effectiveness of the electrophysiological strategy has been proven in a limited patient population with mild heart failure disease (38). These approaches may be considered in the future among the different possible treatment approaches to AF in heart failure patients. Study limitations. The relatively small number of patients in the non-ablated group may in some way bias the outcome of the study. Nevertheless, the intriguing findings of consistent low benefit in non-ablated AF patients suggests that, even by enlarging the patient cohort, the absolute increase of left ventricular function would likely remain small. The percentage cutoff of biventricular pacing for indicating AVJ ablation was arbitrarily selected. It seemed reasonable to assume that 85% of biventricular pacing rate was enough; on the contrary, this relatively high pacing time was not sufficient in detecting any substantial benefit from CRT. Percentage of biventricular stimulation probably overestimates effective biventricular capture, particularly because percentage of pacing does not objectively account for the problem of fusion and pseudofusion between intrinsic and paced beats. In fact, 24-h Holter ECG monitoring performed in some non-ablated AF patients implanted with CRT systems confirmed some discordance between true effective biventricular capture and the percentage of biventricular stimulation time derived from device counters; in these cases, while device-derived percentage of biventricular time was 85%, effective biventricular capture detected at 24-h monitoring resulted lower. This limitation, however, does not change the very relevant clinical message of the present work: AVJ ablation in heart failure patients with permanent AF treated with CRT maximized CRT delivery better than any other approach. The populations studied at the 2 centers differed in terms of baseline LVEF. Although some interobserver variability when evaluating LVEF by echocardiography cannot be completely dismissed, 1 center started CRT very early on, when extremely compromised patients with very low value of LVEF were considered ideal candidates for the therapy. The limits posed by the differences within the population were overcome by the statistical analysis that considered possible confounding effects of center, gender, age, etiology, and baseline LVEF in the fitted models. Moreover, the choice of using a general linear model for the analysis allowed the efficient use of all the information available at each time point. Because patients were consecutively enrolled, a few patients in NYHA functional class II were included (10% in SR and 3% in the AF groups). The positive results observed in the global population cannot permit any conclusive statement on the improvement of symptoms and exercise tolerance in the NYHA functional class II patient subgroup. Finally, the declining number of patients over follow-up may represent a limitation. But this is the case of all observational cohort studies. All the available patients and visits were included, and no patient was lost to follow-up. Thus, the possibility of informative censoring was restricted to patients dying, who, however, were considered in the

9 742 Gasparini et al. JACC Vol. 48, No. 4, 2006 Cardiac Resynchronization in AF August 15, 2006: analysis as long as their data were available. We are aware of this possible bias, and in this sense we might have overestimated the magnitude of changes in time of the parameters evaluated, but potentially not of the role of AF and of AVJ ablation. To indirectly deal with this problem, we assessed the role of AF and of AVJ ablation on the response to CRT, both while excluding and including cardiac death as nonresponse, and conclusions did not differ substantially. Conclusions. Cardiac resynchronization therapy induced a remarkable, progressive, and persistent long-term improvement of left ventricular function and functional capacity in heart failure patients with AF. This long-term improvement progressed in parallel to that observed in patients with SR. The combination of CRT and AVJ ablation is a superior treatment strategy in heart failure patients with AF. Large randomized studies would be useful to further confirm these results. Reprint requests and correspondence: Dr. Maurizio Gasparini, IRCCS Istituto Clinico Humanitas, Rozzano-Milano, Italy, Via Manzoni 56, Rozzano, Milano, Italy. maurizio.gasparini@ humanitas.it. REFERENCES 1. Saxon LA, Ellenbogen KA. Resynchronization therapy for the treatment of heart failure. Circulation 2003;108: Abraham WT, Fisher WG, Smith AL, et al. Multicenter InSync Randomized Clinical Evaluation. Cardiac resynchronization in chronic heart failure. N Engl J Med 2002;346: Cazeau S, Leclerq C, Lavergne T, et al. Effects of multisite biventricular pacing in patients with heart failure and intraventricular conduction delay. N Engl J Med 2001;344: Linde C, Leclerq C, Rex S, et al. Long-term benefits of biventricular pacing in congestive heart failure: results from the Multislice Stimulation in Cardiomyopathy (MUSTIC) study. J Am Coll Cardiol 2002;40: Auricchio A, Stellbrink C, Sack S, et al. The Pacing Therapies for Congestive Heart Failure (PATH-CHF) study: rationale, design, and endpoints of a prospective randomised multicenter study. Am J Cardiol 1999;83:130D 5D. 6. Auricchio A, Stellbrink C, Sack S, et al. Effect of pacing chamber and atrioventricular delay on acute systolic function of paced patients with congestive hart failure. The Pacing Therapies for Congestive Heart Failure Study Group. Circulation 1999;99: Auricchio A, Stellbrink C, Sack S, et al. Long-term clinical effect of hemodynamically optimized cardiac resynchronization therapy in patients with heart failure and ventricular conduction delay. J Am Coll Cardiol 2002;39: Gras D, Leclercq C, Tang A, Bucknall C, Luttikhuis HO, Kirstein- Pedersen A. Cardiac resynchronization therapy in advanced heart failure: the multicenter InSync clinical study. Eur J Heart Fail 2002;4: Leclercq C, Victor F, Alonso C, et al. Comparative effects of permanent biventricular pacing for refractory heart failure in patients with stable sinus rhythm of chronic atrial fibrillation. Am J Cardiol 2000;85: Molhoek SG, Bax JJ, Bleeker GB, et al. Comparison of response to cardiac resynchronization therapy in patients with sinus rhythm versus chronic atrial fibrillation. Am J Cardiol 2004;94: Benjamin EJ, Wolf PA, D Agostino RB, Silbershatz H, Kannel WB, Levy D. Impact of atrial fibrillation on the risk of death. The Framingham Heart study. Circulation 1998;98: Middlekauff HR, Stevenson WG, Stevenson LW. Prognostic significance of atrial fibrillation in advanced heart failure. Circulation 1991;84: Wang TJ, Larson MG, Levy D, et al. Temporal relations of atrial fibrillation and congestive heart failure and their joint influence on mortality: the Framingham Heart study. Circulation 2003;107: Garrigue S, Bordachar P, Reuter S, Jais P, Haissaguerre M, Clementy J. Comparison of permanent left ventricular and biventricular pacing in patients with heart failure and chronic atrial fibrillation: a prospective hemodynamic study. Card Electrophysiol Rev 2003;7: Simantirakis EN, Vardakis KE, Kochiadakis GE, et al. Left ventricular mechanics during right ventricular apical or left ventricular-based pacing in patients with chronic atrial fibrillation after atrioventricular junction ablation. J Am Coll Cardiol 2004;43: Hay I, Melenovsky V, Fetics BJ, et al. Short-term effect of right-left heart sequential cardiac resynchronization in patients with heart failure, chronic atrial fibrillation, and atrioventricular nodal block. Circulation 2004;110: Middlekauf HR, Wiener I, Stevenson WG. Low-dose amiodarone for atrial fibrillation. Am J Cardiol 1993;72:75F 81F. 18. Twidale N, Manda V, Nave K, Seal A. Predictors of outcome after radiofrequency catheter ablation of the atrioventricular node for atrial fibrillation and congestive heart failure. Am Heart J 1998;136: Brignole M, Menozzi C, Gianfranchi L, et al. 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: Natale A, Zimerman L, Tomassoni G, et al. AV node ablation and pacemaker implantation after withdrawal of effective rate-control medications for chronic atrial fibrillation: effect on quality of life and exercise performance. Pacing Clin Electrophysiol 1999;22: Kwo-Chang U, Tsung-Po T, Ching-Feng T, et al. Acute and long-term effects of atrioventricular junction ablation and VVIR pacemaker in symptomatic patients with chronic lone atrial fibrillation and normal ventricular response. J Cardiovasc Electrophysiol 2001;12: Guyatt GH, Sullivan MJ, Thompson PJ. The 6 minute walk: a new measure of exercise capacity in patients with chronic heart failure. Can Med Assoc J 1985;132: Opasich C, Pinna GD, Mazza A, et al. Reproducibility of the six-minute walking test in patients with chronic congestive heart failure: practical implications. Am J Cardiol 1998;81: Schiller NB, Shah PM, Crawford M, et al. Recommendations for quantification of the left ventricle by two-dimensional echocardiography. American Society of Echocardiography Committee on Standards. Subcommittee on Quantification of Two-Dimensional Echocardiograms. J Am Soc Echocardiogr 1989;2: Ritter P, Padeletti L, Gillio-Meina L, Gaggini G. Determination of the optimal atrioventricular delay in DDD pacing. Comparison between echo and peak endocardial acceleration measurements. Europace 1999;1: Geelen P, Brugada J, Andries E, Brugada P. Ventricular fibrillation and sudden death after radiofrequency catheter ablation of the atrioventricular junction. Pacing Clin Electrophysiol 1997;20: Yu CM, Bleeker GB, Fung JW, et al. Left ventricular reverse remodeling but not clinical improvement predicts long-term survival after cardiac resynchronization therapy. Circulation 2005;112: Ware JE, Kosinski M, Gandek B. SF-36 Health Survey: Manual and Interpretation Guide. Lincoln, RI: Quality Metric Incorporated, 1993: Leclercq C, Walker S, Linde C. Comparative effects of permanent biventricular and right-univentricular pacing in heart failure patients with chronic atrial fibrillation. Eur Heart J 2002;23: Reuter S, Garrigue S, Barold S, Jais P. Comparison of characteristics in responders versus nonresponders with biventricular pacing for drug-resistant congestive heart failure. Am J Cardiol 2002;89: Nelson G, Curry C, Wyman B. Predictors of systolic augmentation from left ventricular preexcitation in patients with dilated cardiomyopathy and intraventricular conduction delay. Circulation 2000;101: Pitzalis MV, Iacoviello M, Romito R, et al. Cardiac resynchronization therapy tailored by echocardiographic evaluation of ventricular asynchrony. J Am Coll Cardiol 2002;40: Leon AR, Greenberg JM, Kanuru N, Baker C. Cardiac resynchronization in patients with congestive heart failure and chronic atrial

10 JACC Vol. 48, No. 4, 2006 August 15, 2006: Gasparini et al. Cardiac Resynchronization in AF 743 fibrillation. Effect of upgrading to biventricular pacing after chronic right ventricular pacing. J Am Coll Cardiol 2002;39: Melenovsky V, Hay I, Fetics BJ, et al. Functional impact of rate irregularity in patients with heart failure and atrial fibrillation receiving cardiac resynchronization therapy. Eur Heart J 2005;7: Grogan M, Smith HC, Gersh BJ, Wood DW. Left ventricular dysfunction due to atrial fibrillation in patients initially believed to have idiopathic dilated cardiomyopathy. Am J Cardiol 1992;69: Daoud EG, Weiss R, Bahu M, et al. Effect of an irregular ventricular rhythm on cardiac output. Am J Cardiol 1996;78: Pappone C, Rosanio S, Augello G, et al. Mortality, morbidity, and quality of life after circumferential pulmonary vein ablation for atrial fibrillation: outcomes from a controlled non-randomized long-term study. J Am Coll Cardiol 2003;42: Hsu LF, Jais P, Sanders P, et al. Catheter ablation for atrial fibrillation in congestive heart failure. N Engl J Med 2004;351: