- Email: [email protected]
Improvement of left atrial function is associated with lower incidence of atrial fibrillation and mortality after cardiac resynchronization therapy
Improvement of left atrial function is associated with lower incidence of atrial fibrillation and mortality after cardiac resynchronization therapy Jeffrey W.H. Fung, MD, Gabriel W.K. Yip, MD, Qing Zhang, PhD, Fang Fang, PhD, Joseph Y.S. Chan, FHKAM, Chun Mei Li, BM, Li Wen Wu, BM, Gary C.P. Chan, MRCP, Hamish C.K. Chan, FHKAM, Cheuk-Man Yu, MD From the Division of Cardiology, Department of Medicine and Therapeutics, Prince of Wales Hospital, The Chinese University of Hong Kong, Hong Kong SAR, China. BACKGROUND Left atrial (LA) volume is a predictor of cardiovascular events in patients with heart failure. Improvement of LA function and reverse remodeling was observed after cardiac resynchronization therapy (CRT). OBJECTIVE The purpose of this study was to explore the clinical significance of improvement in LA function after CRT. METHODS Echocardiographic studies were performed before and 3 months after CRT in 97 patients (72 men and 25 women; age 63.8 ⫾ 13.3 years) with standard CRT indication but no history of atrial fibrillation (AF). LA active emptying fraction based on the change in volumes (LAV-EF) were calculated, and significant improvement in LA function (LA responder) was defined as a relative increase ⱖ50% from baseline LAV-EF. The primary end-points were newly developed AF detected by ECG or device and all-cause mortality.
40%, P ⫽ .002) and mortality (17% vs 44%, P ⫽ .004) than did LA nonresponders. In Cox proportional hazard analysis, LA responders was the only independent predictor of lower risk of new-onset AF (hazard ratio 0.22, 95% confidence interval 0.08 – 0.61, P ⫽ .003), whereas both LA responders (hazard ratio 0.22, 95% confidence interval 0.09 – 0.53, P ⬍.001) and left ventricular reverse remodeling (⬎10% reduction in left ventricular end-systolic volume at 3 months; hazard ratio 0.96, 95% confidence interval 0.93– 0.99, P ⫽ .03) were independent predictors of lower risk of death after CRT. CONCLUSION Improvement of LA function after CRT was associated with a lower incidence of AF and mortality in AF naïve patients with severe heart failure. KEYWORDS Atrial fibrillation; Heart failure; Cardiac resynchronization; Echocardiography; Prognosis
RESULTS After 1,200 ⫾ 705 days of follow-up, LA responders (n ⫽ 47 [48.5%]) had a significantly lower incidence of AF (12.8% vs
(Heart Rhythm 2008;5:780 –786) © 2008 Heart Rhythm Society. All rights reserved.
Introduction
Thus, the benefit of CRT on AF development might have been underestimated. A favorable effect on left atrial (LA) function and reverse remodeling has been reported after CRT.12,13 Moreover, the susceptibility to atrial arrhythmia seemed to be reduced after upgrading to CRT in another study.6 LA volume is a powerful predictor of cardiovascular events.14 –17 Its prognostic value seems to be independent of symptom status, left ventricular (LV) ejection fraction, and peak oxygen consumption and may even predict the risk of sudden death in heart failure.18,19 The clinical significance of the improvement of LA function after CRT is undetermined. The purpose of the current study was to explore the prognostic implication of the improvement of LA function after CRT on new-onset AF and survival in patients with severe heart failure but no history of AF.
Cardiac resynchronization therapy (CRT) is an established therapy for selected patients with advanced heart failure and cardiac dyssynchrony.1 The benefit of CRT in these selected patients is well proven, but its impact on atrial fibrillation (AF) is controversial.2– 6 AF and heart failure often coexist and have been associated with worse clinical outcomes.7,8 Nonpharmacologic control of AF in heart failure patients may be beneficial.9 Post hoc analysis of the Cardiac Resynchronization in Heart Failure (CARE-HF) trial suggested that the device had no favorable impact on AF incidence.3 However, the asymptomatic AF episodes, which seem to have equally important prognostic implication,10,11 could not be completely elucidated in the medical therapy group.
Address reprint requests and correspondence: Dr. Jeffrey W.H. Fung, Division of Cardiology, Department of Medicine and Therapeutics, Prince of Wales Hospital, The Chinese University of Hong Kong, Hong Kong, SAR, China. E-mail address: [email protected]. (Received December 14, 2007; accepted February 29, 2008.)
Methods Patients Consecutive patients with advanced congestive heart failure and standard CRT indication1 but no history of ECG-doc-
1547-5271/$ -see front matter © 2008 Heart Rhythm Society. All rights reserved.
doi:10.1016/j.hrthm.2008.02.043
Fung et al
Improvement of LA Function After CRT
umented AF who had received CRT for at least 3 months were recruited into the study. Serial standard echocardiography with tissue Doppler imaging and clinical assessment, including quality of life score,20 were performed within 1 week before and 3 months after CRT. The first patient was enrolled in January 2001, and the last follow-up was performed in June 2007. A total of 148 patients received CRT, and 51 had ECG-documented AF prior to CRT implantation. Thus, 97 patients were selected and regularly followed-up in the clinic every 3 months or when symptomatic with dyspnea or palpitations. Clinical examination and ECG with device interrogation, including percentage of atrial pacing, was performed at every clinic visit and during hospitalization. The study protocol was approved by the Ethics Committee and conformed to the guiding principles of the Declaration of Helsinki. Written informed consent was obtained from each participant.
Biventricular device implantation Biventricular devices were implanted as previously described.21 Right atrial and ventricular leads were positioned transvenously to the atrial appendage and apex, respectively. The LV pacing lead was inserted by a transvenous approach through the coronary sinus to target the lateral or posterolateral cardiac vein. Choice of CRT device included biventricular pacemaker in 89 patients (InSync or InSync III, Medtronic Inc., Minneapolis, MN, USA; Contak TR or Contak TR II, Guidant Inc., St. Paul, MN, USA) and biventricular defibrillator in eight patients for secondary prevention of ventricular tachyarrhythmia (InSync ICD, InSync Marquis, or InSync Sentry, Medtronic; Contak CD or Contak Renewal, Guidant). All devices were programmed to DDDR mode with lower rate of 60 bpm and atrial high rate detection cutoff ⬎180 bpm. Atrial bipolar sensitivity was programmed to half of the P wave detected by the device, with the lowest value ⬎0.5mV. Onset detection number of beats was 10 and termination number was 10 consecutive beats at a rate above and below the detection rate, respectively.
Echocardiographic assessment Echocardiography with tissue Doppler imaging was performed (Vivid 5 or Vivid 7, Vingmed-General Electric, Horten, Norway) serially before and 3 months after CRT as previously described.13,21 LV volumes and ejection fraction were assessed by biplane Simpson equation. Patients with ⬎10% reduction of LV end-systolic volume were defined as LV responders of CRT, whereas patients with a lesser degree of reduction were labeled as LV nonresponders (Table 1).22 Diastolic dysfunction was graded as abnormal relaxation, pseudonormal, or restrictive filling pattern.23 LA function was assessed by biplane methods as previously described.13,24 In brief, the LA long-axis diameter at end-diastole was measured in the four-chamber view. Using the modified Simpson rule, LA volumes, which were indexed to body surface area, at ventricular end-systole (LAV-max), just before (LAV-pre), and after atrial systole
781 Table 1
Definitions of LA and LV responders
LA responder LA nonresponder LV responder LV nonresponder
Parameter
Cutoff value
Relative increase in indexed LAV-EF before and 3 months after CRT Relative reduction in LV end-systolic volume before and 3 months after CRT
ⱖ50% ⬍50% ⬎10% ⱕ10%
CRT ⫽ cardiac resynchronization therapy; LA ⫽ left atrial; LAV-EF ⫽ left atrial active emptying fraction based on change in volumes; LV ⫽ left ventricular.
(LAV-post) and atrial active emptying fraction (LAV-EF ⫽ [LAV-pre – LAV-post]/LAV-pre) were calculated. Intraobserver and interobserver variability for LAV were 7 ⫾ 9 mL and 5 ⫾ 6 mL, respectively, comparable to a previous report.24 Significant improvement in LA function after CRT (LA responder) was defined as a relative increase ⱖ50% from baseline LAV-EF (Table 1), which could differentiate volumetric responders and nonresponders to CRT.13 Tissue Doppler imaging was performed in the apical four-chamber view for the long-axis motion of the heart as previously described.13 At least three consecutive beats were stored, and the images were digitized and analyzed offline by dedicated software (EchoPac-PC 6.0.1, VingmedGeneral Electric). LA Doppler velocity profile signals were reconstituted offline by placing a 3 ⫻ 12 mm sampling window at the mid levels of LA. Peak regional atrial contraction velocities at atrial systole (after onset of P wave of ECG) were measured. Atrial strain () was measured in the same atrial locations as described previously.13 The cardiologist responsible for echocardiographic data analysis was blinded to the information derived by device interrogation, occurrence of AF, and mortality.
Definition of end-points Only patients without a history of documented AF were recruited in the study. New-onset AF was defined as any episode of AF detected by ECG during clinic visit and hospitalization or as atrial high rate episodes that lasted ⬎10 minutes with maximum atrial rate at least ⬎220 bpm detected by the device3,11 unless specified. The reliability of AF detection by devices compared to ambulatory Holter monitoring using the same cutoff value has been demonstrated with sensitivity 100%, specificity 97.6%, and falsepositive rate 2.4%.11 Classification of AF was based on published guidelines.25 The electrophysiologist responsible for device interrogation and detection of AF was blinded to the echocardiographic data. The primary end-points of the present analysis were new-onset AF and all-cause mortality. The cause of death was ascertained by reviewing the clinical record, report of close relatives, and postmortem findings.
Statistical analysis Statistical analysis was performed using SPSS for Windows (version 11.0, SPSS, Inc., Chicago, IL, USA). For compar-
782
Heart Rhythm, Vol 5, No 6, June 2008
Table 2 Comparison of clinical and echocardiographic parameters at baseline and after cardiac resynchronization therapy for 3 months
Demographic New York Heart Association class Six-minute hall walk (m) Quality-of-life score Hemodynamic Blood pressure (mmHg) Heart rate (bpm) Echocardiography LV end-systolic volume (mL) LV end-diastolic volume (mL) LV ejection fraction (%) Mitral regurgitation (% LA area) Diastolic dysfunction (% patients) Abnormal relaxation pattern Pseudonormal filling pattern Restrictive filling pattern LA diameter (cm) Indexed LAV-max (mL/m2) Indexed LAV-pre (mL/m2) Indexed LAV-post (mL/m2) LAV-EF (%)
Baseline
3 months after CRT
P value
3.09 ⫾ 0.38 318 ⫾ 104 30 ⫾ 21
2.22 ⫾ 0.59 359 ⫾ 99 20 ⫾ 19
⬍.001 ⬍.001 ⬍.001
105 ⫾ 18 68 ⫾ 10
108 ⫾ 22 71 ⫾ 12
137 183 26.1 33 30 35 35 5.22 44.7 43.3 33.1 23.3
⫾ ⫾ ⫾ ⫾
⫾ ⫾ ⫾ ⫾ ⫾
.341 .411
62 70 7.3 21
111 162 33.6 23
⫾ ⫾ ⫾ ⫾
60 72 9.8 20
⬍.001 .004 ⬍.001 ⬍.001 ⬍.05
0.87 17.4 18.3 22.4 11.0
43 36 21 5.15 44.5 42.7 28.0 31.0
⫾ ⫾ ⫾ ⫾ ⫾
0.93 16.2 17.1 17.1 15.4
.442 .487 .321 .124 ⬍.001
CRT ⫽ cardiac resynchronization therapy; LV ⫽ left ventricular; LA ⫽ left atrial; indexed LAV ⫽ left atrial volume indexed to body surface area; LAV-EF ⫽ left atrial emptying fraction by volume change.
ison of continuous parametric variables between baseline and 3 months after CRT, paired sample t-test was used. Comparison of baseline demographic and echocardiographic parameters between patients with and without newonset AF, survivors and nonsurvivors, and LA responders and LA nonresponders was performed by unpaired t-test or Chi-square test as appropriate. Time to AF onset and death between LA responders and nonresponders was analyzed by Kaplan-Meier method. A Cox proportional hazard model was constructed with AF and death as dependent variables and age, sex, etiology of heart failure, New York Heart Association (NYHA) class, indexed LAV-max and severity of LV diastolic dysfunction, and LA and LV responders as independent variables to predict clinical events. These variables were selected because of their clinical relevance to the outcomes and research question of interest (LA responder). To minimize the risk of overfitting, factors with no significant difference between patients with or those without the outcomes (AF and mortality) in the univariate analysis were not entered into the multivariate model. Thus, for the Cox proportional hazard model with AF as dependent variable, age, sex preponderance, NYHA class, and etiology of heart failure were not entered into the model (see Table 3). Similarly, for the model with death as dependent variable, sex, NYHA class, and etiology of heart failure also were excluded from entering the model. All parametric data are expressed as mean ⫾ SD. P ⬍.05 was considered significant.
Results Among the 97 patients (72 men and 25 women; age 63.8 ⫾ 13.3 years), 82 (85%) were in NYHA class III and 41 (42%)
had ischemic etiology. Medications included diuretics in nearly all patients, angiotensin-converting enzymes inhibitors or angiotensin receptor blockers in 91%, beta-blockers in 72%, and spironolactone in 52%. Amiodarone was prescribed for three patients with defibrillators at baseline for ventricular arrhythmia control. No other antiarrhythmic drug was used by the other patients. Baseline clinical and echocardiographic characteristics of the whole group are given in Table 2. Significant improvements in symptom, exercise capacity, quality of life, and LV reverse remodeling were observed 3 months after CRT. For LA function, significant improvement in peak LA contraction velocity (3.49 ⫾ 1.84 cm/s to 4.11 ⫾ 1.98 cm/s, P ⫽ .009), maximal LA during ventricular systole (16.8% ⫾ 11.0% to 20.73 ⫾ 9.47%, P ⫽ .004), maximal LA during ventricular early diastole (6.38% ⫾ 4.60% to 9.29 ⫾ 5.92%, P ⬍.001), and indexed LAV-EF were observed (Table 2). However, no significant change was observed in indexed LAV-max before and after CRT. Forty-seven patients (48.5%) were LA responders and 57 (58.8%) were LV responders using the definitions given in Table 1. There was no difference in demographic and echocardiographic characteristics between LA responders and nonresponders at baseline (Table 3). The percentage of atrial pacing at 3 months was 36% ⫾ 8% and 41% ⫾ 10% among LA responders and nonresponders, respectively (P ⫽ NS).
New-onset AF After follow-up of 1200 ⫾ 705 days, new-onset AF with mean duration of 401 ⫾ 417 min (range 30 –1680 min) was detected in 26 (26.8%) patients. Six (12.8%) patients among the LA responders and 20 (40%) among the LA nonresponders
Fung et al Table 3
Improvement of LA Function After CRT
783
Baseline patient characteristics
Demographic Age (years) Sex (M/F) New York Heart Association class Ischemic etiology (%) Six-minute hall walk (m) Hypertension (%) Diabetes mellitus (%) Biochemistry Serum potassium level (mmol/L) Medications Angiotensin-converting enzyme inhibitor/angiotensin receptor blocker (%) Beta-blocker (%) Spironolactone (%) Echocardiography LVEF Indexed LAV-max (mL/m2) LAV-EF (%) Mitral regurgitation (% LA area) E/A ratio Abnormal relaxation pattern (%) Pseudonormal pattern (%) Restrictive filling pattern (%)
LA responders (n ⫽ 47)
LA nonresponders (n ⫽ 50)
AF (n ⫽ 26)
No AF (n ⫽ 71)
Survivors (n ⫽ 67)
Nonsurvivors (n ⫽ 30)
62.5 ⫾ 14.6 38/10 3.12 ⫾ 0.41 25 (52) 305 ⫾ 110 18 (38) 19 (40)
65.5 ⫾ 11.7 35/14 3.05 ⫾ 0.39 19 (39) 324 ⫾ 102 22 (44) 26 (52)
66.1 ⫾ 11.7 21/5 3.10 ⫾ 0.35 9 (35) 328 ⫾ 108 10 (38) 16 (62)
63.4 ⫾ 13.7 52/19 3.08 ⫾ 0.42 36 (51) 315 ⫾ 101 30 (42) 29 (41)
61.9 ⫾ 13.2 50/17 3.09 ⫾ 0.41 28 (41) 325 ⫾ 104 28 (42) 30 (45)
69.0 ⫾ 11.9* 23/7 3.07 ⫾ 0.37 17 (57) 310 ⫾ 96 12 (40) 15 (50)
4.2 ⫾ 0.5
4.4 ⫾ 0.6
4.5 ⫾ 0.5
4.3 ⫾ 0.6
4.1 ⫾ 0.4
4.3 ⫾ 0.6
42 (89)
46 (92)
24 (92)
64 (90)
62 (93)
26 (87)
35 (74) 24 (51)
35 (70) 26 (52)
22 (84) 15 (58)
48 (68) 35 (49)
49 (73) 34 (51)
21 (67) 16 (53)
26.3 ⫾ 6.7 44.6 ⫾ 20.0 23.1 ⫾ 10.7 29.4 ⫾ 17.2 1.20 ⫾ 0.59 18 (38) 14 (29) 16 (33)
26.2 ⫾ 8.0 44.6 ⫾ 15.3 23.6 ⫾ 10.1 34.4 ⫾ 24.0 1.35 ⫾ 0.89 11 (22) 20 (41) 18 (37)
26.2 ⫾ 8.1 45.8 ⫾ 16.5 22.1 ⫾ 10.1 33.2 ⫾ 23.1 1.23 ⫾ 0.56 9 (35) 8 (30) 9 (35)
26.2 ⫾ 7.2 44.3 ⫾ 17.8 24.6 ⫾ 9.2 31.9 ⫾ 20.4 1.35 ⫾ 0.66 20 (28) 26 (37) 25 (35)
27.3 ⫾ 6.8 45.1 ⫾ 18.0 24.0 ⫾ 10.2 31.0 ⫾ 16.2 1.44 ⫾ 0.75 19 (28) 22 (33) 26 (39)
24.7 ⫾ 7.1 43.1 ⫾ 16.2 22.7 ⫾ 10.4 36.0 ⫾ 23.0 1.26 ⫾ 0.94 10 (33) 12 (40) 8 (27)
AF ⫽ atrial fibrillation; LA ⫽ left atrial; indexed LAV-max ⫽ left atrial volume measured at left ventricular end-systole indexed to body surface area; LAV-EF ⫽ left atrial emptying fraction by volume change; LVEF ⫽ left ventricular ejection fraction. *P ⬍.05 survivors vs nonsurvivors.
had new-onset AF detected by device or ECG (P ⫽ .002). There was a trend for shorter duration of new-onset AF episode among LA responders than LA nonresponders (117 ⫾ 80 min vs 486 ⫾ 440, P ⫽ .055). All ECG-documented AF episodes were detected by the device as atrial high-rate episodes. Eleven patients had new-onset AF episodes detected by ECG during clinic visit or hospitalization (symptomatic). Six of these patients required intravenous amiodarone for AF conversion; the other five patients received rate-controlling agents. Two patients had undergone electrical cardioversion. All 11 patients reverted back to sinus rhythm after new-onset AF. Three patients in the LA responder group and 12 in the LA nonresponder group had the first episode of AF detected by device during routine interrogation (asymptomatic). At the last follow-up, three patients in the LA responder group and five in the LA nonresponder group were noted to have permanent AF. Kaplan-Meier estimate of the time to new-onset AF detected by device or ECG between LA responders and nonresponders is shown in Figure 1. Kaplan-Meier estimate of AF-free survival at 5 years was 0.85 and 0.55 in LA responders and nonresponders, respectively. Significantly earlier onset of AF was observed among the LA nonresponders (log-rank Chi-square ⫽ 8.44, P ⫽ .004). There was no significant difference in baseline demographic and echocardiographic characteristics between patients with and those without AF (Table 3). By multivariate analysis, LA responder was the only predictor of lower risk
of new-onset AF after CRT (hazard ratio 0.22, 95% confidence interval 0.08 – 0.61, P ⫽ .003).
Mortality Thirty (30.9%) patients died during follow-up. Nine (34.6%) patients developed AF, whereas 21 (29.6%) had no AF before
Figure 1 Kaplan-Meier estimate of time to first onset of atrial fibrillation (AF). First onset of AF was significantly earlier in left atrial (LA) nonresponders (dotted line) than in LA responders (continuous line), defined as ⱖ50% relative increase from baseline LA active emptying fraction after 3 months of cardiac resynchronization therapy.
784
Figure 2 Kaplan-Meier survival curves for left atrial (LA) responders (continuous line) and LA nonresponders (dotted line).
Heart Rhythm, Vol 5, No 6, June 2008 ever, the incidence of AF after CRT is under debate. In an early nonrandomized study,2 three of 36 CRT recipients without a history of AF developed AF over 36 months. In the CARE-HF trial,3 the incidence in patients randomized to CRT was 16.1% for symptomatic episodes and as high as 39.4% when including the device-detected episodes over 29.4 months. In the current study, the Kaplan-Meier estimate of AF-free survival at 5 years was 0.85 and 0.55 in LA responders and nonresponders, respectively. The discrepancy in reported AF incidence probably was related to the fact that only patients without a history of AF were recruited for analysis in the current study and up to 20% of patients in the CRT arm of CARE-HF trial had a history of AF. Therefore, it is not unexpected that the incidence of AF would be higher in patients with paroxysmal AF than in those without a history of AF.
death (P ⫽ NS). Eight deaths (17%) occurred in the LA responder group vs 22 (44%) in the LA nonresponder group (P ⫽ .004). Among the LA responders, 4 patients developed sudden death, 2 died of heart failure exacerbation, and 2 died of noncardiac causes. Among the LA nonresponders, 6 patients developed sudden death, 10 died of heart failure, and 6 died of noncardiac causes. There was no significant difference in baseline clinical and echocardiographic characteristics between survivors and nonsurvivors, except for advanced age noted in nonsurvivors (Table 3). The survival curves between the two groups are shown in Figure 2. LA responders had significantly lower mortality than did LA nonresponders (log-rank Chi-square ⫽ 9.08, P ⫽ .003). LA responders also had significantly lower mortality than LA nonresponders for both LV responders (log-rank Chi-square ⫽ 4.89, P ⫽ .027) or nonresponders (log-rank Chi-square ⫽ 6.06, P ⫽ .013; Figure 3). By multivariate analysis, LA responder (hazard ratio 0.22, 95% confidence interval 0.09 – 0.53, P ⬍.001) and LV reverse remodeling (hazard ratio 0.96, 95% confidence interval 0.93-0.99, P ⫽ .03) were independent predictors of survival after CRT.
Discussion The key finding of the current study is that improvement of LA function after CRT was associated with a lower incidence of AF, which was independent of LV reverse remodeling after CRT. Both LA responder and LV reverse remodeling at 3 months were independent predictors of survival after CRT. LA responder seems to be a novel marker for better outcome after CRT and possibly is incremental to the favorable structural changes in the LV in predicting survival in patients with severe heart failure.
Incidence of AF after CRT Development of AF in patients with chronic heart failure may lead to deterioration in hemodynamic performance, heart failure exacerbation, or even hospitalization. How-
Figure 3 A: Kaplan-Meier survival curves for left atrial LA responders (continuous line) and LA nonresponders (dotted line) among left ventricular (LV) responders, defined as reduction of LV end-systolic volume ⬎10% after 3 months of cardiac resynchronization therapy. B: KaplanMeier survival curves for LA responders (continuous line) and LA nonresponders (dotted line) among LV nonresponders.
Fung et al
Improvement of LA Function After CRT
785
Prevention of AF by CRT
LA and LV responders after CRT
Another question that draws much attention is whether CRT has any favorable impact on AF development. In the CARE-HF trial, there was no difference in the time to first episode of AF after CRT between the medical therapy alone group and the CRT plus medical therapy group.3 However, to avoid ascertainment bias only symptomatic episodes were used for analysis. The clinical importance of asymptomatic AF episodes detected by device was demonstrated in the Mode Selection Trial,11 and atrial high-rate events detected by pacemakers were predictors of total mortality and death or nonfatal stroke.11 The effect of CRT upgrading on susceptibility to atrial tachyarrhythmia detected by device was examined in 28 patients, and significant reduction of tachyarrhythmia episodes was observed.6 In the current study, patients with improvement of LA function after CRT had a significantly lower incidence of AF than did nonresponders. After adjustment for other potential risk factors, such as baseline LAV-max and LV reverse remodeling, improvement of LA function 3 months after CRT was the only independent predictor of AF free survival, suggesting that the physiologic improvement in LA active contractile function is vital for maintenance of sinus rhythm in patients with severe heart failure. Such an improvement in LAV-EF may be considered as an important surrogate marker for favorable response to novel AF therapy, such as AF ablation in heart failure patients, in the future.
LV reverse remodeling is an important marker of favorable prognosis after CRT.21 LA size is a powerful predictor of adverse clinical events, including mortality in patients with cardiovascular diseases.14 –17 LAV is a more robust marker of cardiovascular events than is LA size or area in patients with sinus rhythm27 and was predictive of AF development after cardiac surgery.28 The prognostic value of LAV in patients after CRT has not been explored. In addition, although reversal of LA dilation or recovery of LA function has been observed after medical therapy, whether such improvement translates into clinical benefit has not been established.26 The current study demonstrated the prognostic implication of improvement in LAV-EF after CRT, but not indexed LAV-max alone, as an independent predictor of survival after CRT. The favorable prognosis in terms of survival in LA responders was observed in both LV responders and nonresponders, indicating the importance of improvement of LA function probably was complementary to LV reverse remodeling in determining the prognosis after CRT.
LA reverse remodeling and contractile function LA reverse remodeling has been observed in patients with persistent or permanent AF after CRT.12 However, despite the apparently favorable anatomic improvement, reversal to sinus rhythm was not observed in the study. This might be related to the fact that the diseased LA already was at an advanced stage so that the structural change was not be able to translate into clinical benefit. Another potential explanation is that the improvement of active atrial contractile function was more crucial than reduction in LA size alone. LA- was increased during ventricular systole and early diastole and our results suggest that the atria are more elastic and compliant when subjected to passive stretching. The more dynamic changes in LA and active LA emptying fraction as measured by LAV-EF may reflect structural changes leading to better atrial compliance, contractile function, and maintenance of sinus rhythm after CRT. On the other hand, severe LV diastolic dysfunction was not a predictor of AF development in the current study. Transmitral and pulmonary veins Doppler studies are useful for assessing LV diastolic and phasic LA function, including its role as conduit and reservoir.26 The finding from the current study may suggest that LA active emptying fraction is important in determining AF outcomes in these patients with systolic heart failure, and further studies are warranted to determine its role in AF development.
Study limitations The current study is limited by its small sample size. Our study could not clearly rule out the possibility that the indexed LAV alone after CRT was a predictor of clinical events if more patients were recruited. Not all intracardiac electrograms stored were available for analysis to confirm or refute whether the atrial high-rate episodes were indeed AF. The same cutoff criteria for atrial high-rate detection used in the CARE-HF trial were adopted in the current study. Atrial high-rate episodes lasting ⬎5 minutes could effectively exclude oversensing episodes,29 and atrial rate ⬎220 bpm also could exclude slow atrial tachycardia.11 The real benefit, if any, of CRT on AF incidence could not be clearly determined because the current study was not a randomized comparison trial between CRT and medical therapy. Only a randomized CRT trial with device switches on and off together with monitoring of asymptomatic atrial high-rate episodes could definitively address the controversy about the effect of CRT on AF development.
Conclusion Improvement of LA function after CRT in patients without a history of AF was associated with a lower incidence of AF and mortality than in patients without improvement of LA function. Such an improvement was an independent predictor of survival after CRT regardless of the occurrence of LV reverse remodeling.
References 1. Hunt SA, Abraham WT, Chin MH, et al. ACC/AHA 2005 Guideline Update for the Diagnosis and Management of Chronic Heart Failure in the Adult–Summary Article: A Report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines (Writing Committee to Update the 2001 Guidelines for the Evaluation and Management of Heart Failure): Developed in Collaboration With the American College of Chest Physicians and the International Society for Heart and Lung Transplantation: Endorsed by the Heart Rhythm Society. Circulation 2005;112:1825–1852.
786 2. Fung JW, Yu CM, Chan JY, et al. Effects of cardiac resynchronization therapy on incidence of atrial fibrillation in patients with poor left ventricular systolic function. Am J Cardiol 2005;96:728 –731. 3. Hoppe UC, Casares JM, Eiskjaer H, et al. Effect of cardiac resynchronization on the incidence of atrial fibrillation in patients with severe heart failure. Circulation 2006;114:18 –25. 4. Hugl B, Bruns HJ, Unterberg-Buchwald C, et al. Atrial fibrillation burden during the post-implant period after CRT using device-based diagnostics. J Cardiovasc Electrophysiol 2006;17:813– 817. 5. Adelstein EC, Saba S. Burden of atrial fibrillation after cardiac resynchronization therapy. Am J Cardiol 2007;100:268 –272. 6. Yannopoulos D, Lurie KG, Sakaguchi S, et al. Reduced atrial tachyarrhythmia susceptibility after upgrade of conventional implanted pulse generator to cardiac resynchronization therapy in patients with heart failure. J Am Coll Cardiol 2007;50:1246 –1251. 7. Maisel WH, Stevenson LW. Atrial fibrillation in heart failure: epidemiology, pathophysiology, and rationale for therapy. Am J Cardiol 2003;91:2D– 8D. 8. Fung JW, Sanderson JE, Yip GW, et al. Impact of atrial fibrillation on heart failure with normal ejection fraction: a clinical and echocardiographic study. J Card Fail 2007;13:649 – 655. 9. Hsu LF, Jais P, Sanders P, et al. Catheter ablation for atrial fibrillation in heart failure. N Eng J Med 2004;351:2373–2383. 10. Lin HJ, Wolf PA, Benjamin EJ, et al. Newly diagnosed atrial fibrillation and acute stroke. The Framingham Study. Stroke 1995;26:1527–1530. 11. Glotzer TV, Hellkamp AS, Zimmerman J, et al; MOST Investigators. Atrial high rate episodes detected by pacemaker diagnostics predict death and stroke report of the atrial diagnostics ancillary study of the MOde Selection Trial (MOST). Circulation 2003;107:1614 –1619. 12. Kies P, Leclercq C, Bleeker GB, et al. Cardiac resynchronisation therapy in chronic atrial fibrillation: impact on left atrial size and reversal to sinus rhythm. Heart 2006;92:490 – 494. 13. Yu CM, Fang F, Zhang Q, et al. Improvement of atrial function and atrial reverse remodeling after cardiac resynchronization therapy for heart failure. J Am Coll Cardiol 2007;50:778 –785. 14. Rossi A, Cicoira M, Zanolla L, et al. Determinants and prognostic value of left atrial volume in patients with dilated cardiomyopathy. J Am Coll Cardiol 2002;40:14 –25. 15. Modena MG, Muia N, Sgura FA, et al. Left atrial size is the major predictor of cardiac death and overall clinical outcome in patients with dilated cardiomyopathy: a long-term follow-up study. Clin Cardiol 1997;20:553–560. 16. Moller JE, Hillis GS, Oh JK, et al. Left atrial volume: a powerful predictor of survival after acute myocardial infarction. Circulation 2003;107:2207–2212. 17. Dini FL, Cortigiani L, Baldini U, et al. Prognostic value of left atrial enlargement in patients with idiopathic dilated cardiomyopathy and ischemic cardiomyopathy. Am J Cardiol 2002;89:518 –523.
Heart Rhythm, Vol 5, No 6, June 2008 18. Rossi A, Cicoira M, Bonapace S, et al. Left atrial volume provides independent and incremental information compared with exercise tolerance parameters in patients with heart failure and left ventricular systolic dysfunction. Heart 2007; 93:1420 –1425. 19. Tsang TS, Barnes ME, Gersh BJ, et al. Left atrial volume as a morphophysiologic expression of left ventricular diastolic dysfunction and relation to cardiovascular risk burden. Am J Cardiol 2002; 90:1284 –1289. 20. Rectus TS, Kubo SH, Cohn JN. Validity of the Minnesota Living with Heart Failure questionnaire as a measure of therapeutic response to enalapril or placebo. Am J Cardiol 1993;71:1106. 21. Yu CM, Chau E, Sanderson JE, et al. Tissue Doppler echocardiographic evidence of reverse remodeling and improved synchronicity by simultaneously delaying regional contraction after biventricular pacing therapy in heart failure. Circulation 2002;105:438 – 445. 22. 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;112:1580 –1586. 23. Yu CM, Sanderson JE, Shum IO, et al. Diastolic dysfunction and natriuretic peptides in systolic heart failure. Higher ANP and BNP levels are associated with the restrictive filling pattern. Eur Heart J 1996;17:1694 –1702. 24. Lester SJ, Ryan EW, Schiller NB, et al. Best method in clinical practice and in research studies to determine left atrial size. Am J Cardiol 1999;84: 829 – 832. 25. Fuster V, Ryden LE, Cannom DS, et al; American College of Cardiology; American Heart Association Task Force; European Society of Cardiology Committee for Practice Guidelines; European Heart Rhythm Association; Heart Rhythm Society. ACC/AHA/ESC 2006 guidelines for the management of patients with atrial fibrillation: full text: a report of the American College of Cardiology/American Heart Association Task Force on practice guidelines and the European Society of Cardiology Committee for Practice Guidelines (Writing Committee to Revise the 2001 guidelines for the management of patients with atrial fibrillation) developed in collaboration with the European Heart Rhythm Association and the Heart Rhythm Society. Europace 2006;8:651–745. 26. Abhayaratna WP, Seward JB, Appleton CP, et al. Left atrial size: physiologic determinants and clinical applications. J Am Coll Cardiol 2006;47:2357– 2363. 27. Tsang TS, Abhayaratna WP, Barnes ME, et al. Prediction of cardiovascular outcomes with left atrial size: is volume superior to area or diameter? J Am Coll Cardiol 2006;47:1018 –1023. 28. Osranek M, Fatema K, Qaddoura F, et al. Left atrial volume predicts the risk of atrial fibrillation after cardiac surgery: a prospective study. J Am Coll Cardiol 2006;48:779 –786. 29. Pollack WM, Simmons JD, Interian A, et al. Clinical utility of intraatrial pacemaker stored electrograms to diagnose atrial fibrillation and flutter. Pacing Clin Electrophysiol 2001;24:424 – 429.
40%, P ⫽ .002) and mortality (17% vs 44%, P ⫽ .004) than did LA nonresponders. In Cox proportional hazard analysis, LA responders was the only independent predictor of lower risk of new-onset AF (hazard ratio 0.22, 95% confidence interval 0.08 – 0.61, P ⫽ .003), whereas both LA responders (hazard ratio 0.22, 95% confidence interval 0.09 – 0.53, P ⬍.001) and left ventricular reverse remodeling (⬎10% reduction in left ventricular end-systolic volume at 3 months; hazard ratio 0.96, 95% confidence interval 0.93– 0.99, P ⫽ .03) were independent predictors of lower risk of death after CRT. CONCLUSION Improvement of LA function after CRT was associated with a lower incidence of AF and mortality in AF naïve patients with severe heart failure. KEYWORDS Atrial fibrillation; Heart failure; Cardiac resynchronization; Echocardiography; Prognosis
RESULTS After 1,200 ⫾ 705 days of follow-up, LA responders (n ⫽ 47 [48.5%]) had a significantly lower incidence of AF (12.8% vs
(Heart Rhythm 2008;5:780 –786) © 2008 Heart Rhythm Society. All rights reserved.
Introduction
Thus, the benefit of CRT on AF development might have been underestimated. A favorable effect on left atrial (LA) function and reverse remodeling has been reported after CRT.12,13 Moreover, the susceptibility to atrial arrhythmia seemed to be reduced after upgrading to CRT in another study.6 LA volume is a powerful predictor of cardiovascular events.14 –17 Its prognostic value seems to be independent of symptom status, left ventricular (LV) ejection fraction, and peak oxygen consumption and may even predict the risk of sudden death in heart failure.18,19 The clinical significance of the improvement of LA function after CRT is undetermined. The purpose of the current study was to explore the prognostic implication of the improvement of LA function after CRT on new-onset AF and survival in patients with severe heart failure but no history of AF.
Cardiac resynchronization therapy (CRT) is an established therapy for selected patients with advanced heart failure and cardiac dyssynchrony.1 The benefit of CRT in these selected patients is well proven, but its impact on atrial fibrillation (AF) is controversial.2– 6 AF and heart failure often coexist and have been associated with worse clinical outcomes.7,8 Nonpharmacologic control of AF in heart failure patients may be beneficial.9 Post hoc analysis of the Cardiac Resynchronization in Heart Failure (CARE-HF) trial suggested that the device had no favorable impact on AF incidence.3 However, the asymptomatic AF episodes, which seem to have equally important prognostic implication,10,11 could not be completely elucidated in the medical therapy group.
Address reprint requests and correspondence: Dr. Jeffrey W.H. Fung, Division of Cardiology, Department of Medicine and Therapeutics, Prince of Wales Hospital, The Chinese University of Hong Kong, Hong Kong, SAR, China. E-mail address: [email protected]. (Received December 14, 2007; accepted February 29, 2008.)
Methods Patients Consecutive patients with advanced congestive heart failure and standard CRT indication1 but no history of ECG-doc-
1547-5271/$ -see front matter © 2008 Heart Rhythm Society. All rights reserved.
doi:10.1016/j.hrthm.2008.02.043
Fung et al
Improvement of LA Function After CRT
umented AF who had received CRT for at least 3 months were recruited into the study. Serial standard echocardiography with tissue Doppler imaging and clinical assessment, including quality of life score,20 were performed within 1 week before and 3 months after CRT. The first patient was enrolled in January 2001, and the last follow-up was performed in June 2007. A total of 148 patients received CRT, and 51 had ECG-documented AF prior to CRT implantation. Thus, 97 patients were selected and regularly followed-up in the clinic every 3 months or when symptomatic with dyspnea or palpitations. Clinical examination and ECG with device interrogation, including percentage of atrial pacing, was performed at every clinic visit and during hospitalization. The study protocol was approved by the Ethics Committee and conformed to the guiding principles of the Declaration of Helsinki. Written informed consent was obtained from each participant.
Biventricular device implantation Biventricular devices were implanted as previously described.21 Right atrial and ventricular leads were positioned transvenously to the atrial appendage and apex, respectively. The LV pacing lead was inserted by a transvenous approach through the coronary sinus to target the lateral or posterolateral cardiac vein. Choice of CRT device included biventricular pacemaker in 89 patients (InSync or InSync III, Medtronic Inc., Minneapolis, MN, USA; Contak TR or Contak TR II, Guidant Inc., St. Paul, MN, USA) and biventricular defibrillator in eight patients for secondary prevention of ventricular tachyarrhythmia (InSync ICD, InSync Marquis, or InSync Sentry, Medtronic; Contak CD or Contak Renewal, Guidant). All devices were programmed to DDDR mode with lower rate of 60 bpm and atrial high rate detection cutoff ⬎180 bpm. Atrial bipolar sensitivity was programmed to half of the P wave detected by the device, with the lowest value ⬎0.5mV. Onset detection number of beats was 10 and termination number was 10 consecutive beats at a rate above and below the detection rate, respectively.
Echocardiographic assessment Echocardiography with tissue Doppler imaging was performed (Vivid 5 or Vivid 7, Vingmed-General Electric, Horten, Norway) serially before and 3 months after CRT as previously described.13,21 LV volumes and ejection fraction were assessed by biplane Simpson equation. Patients with ⬎10% reduction of LV end-systolic volume were defined as LV responders of CRT, whereas patients with a lesser degree of reduction were labeled as LV nonresponders (Table 1).22 Diastolic dysfunction was graded as abnormal relaxation, pseudonormal, or restrictive filling pattern.23 LA function was assessed by biplane methods as previously described.13,24 In brief, the LA long-axis diameter at end-diastole was measured in the four-chamber view. Using the modified Simpson rule, LA volumes, which were indexed to body surface area, at ventricular end-systole (LAV-max), just before (LAV-pre), and after atrial systole
781 Table 1
Definitions of LA and LV responders
LA responder LA nonresponder LV responder LV nonresponder
Parameter
Cutoff value
Relative increase in indexed LAV-EF before and 3 months after CRT Relative reduction in LV end-systolic volume before and 3 months after CRT
ⱖ50% ⬍50% ⬎10% ⱕ10%
CRT ⫽ cardiac resynchronization therapy; LA ⫽ left atrial; LAV-EF ⫽ left atrial active emptying fraction based on change in volumes; LV ⫽ left ventricular.
(LAV-post) and atrial active emptying fraction (LAV-EF ⫽ [LAV-pre – LAV-post]/LAV-pre) were calculated. Intraobserver and interobserver variability for LAV were 7 ⫾ 9 mL and 5 ⫾ 6 mL, respectively, comparable to a previous report.24 Significant improvement in LA function after CRT (LA responder) was defined as a relative increase ⱖ50% from baseline LAV-EF (Table 1), which could differentiate volumetric responders and nonresponders to CRT.13 Tissue Doppler imaging was performed in the apical four-chamber view for the long-axis motion of the heart as previously described.13 At least three consecutive beats were stored, and the images were digitized and analyzed offline by dedicated software (EchoPac-PC 6.0.1, VingmedGeneral Electric). LA Doppler velocity profile signals were reconstituted offline by placing a 3 ⫻ 12 mm sampling window at the mid levels of LA. Peak regional atrial contraction velocities at atrial systole (after onset of P wave of ECG) were measured. Atrial strain () was measured in the same atrial locations as described previously.13 The cardiologist responsible for echocardiographic data analysis was blinded to the information derived by device interrogation, occurrence of AF, and mortality.
Definition of end-points Only patients without a history of documented AF were recruited in the study. New-onset AF was defined as any episode of AF detected by ECG during clinic visit and hospitalization or as atrial high rate episodes that lasted ⬎10 minutes with maximum atrial rate at least ⬎220 bpm detected by the device3,11 unless specified. The reliability of AF detection by devices compared to ambulatory Holter monitoring using the same cutoff value has been demonstrated with sensitivity 100%, specificity 97.6%, and falsepositive rate 2.4%.11 Classification of AF was based on published guidelines.25 The electrophysiologist responsible for device interrogation and detection of AF was blinded to the echocardiographic data. The primary end-points of the present analysis were new-onset AF and all-cause mortality. The cause of death was ascertained by reviewing the clinical record, report of close relatives, and postmortem findings.
Statistical analysis Statistical analysis was performed using SPSS for Windows (version 11.0, SPSS, Inc., Chicago, IL, USA). For compar-
782
Heart Rhythm, Vol 5, No 6, June 2008
Table 2 Comparison of clinical and echocardiographic parameters at baseline and after cardiac resynchronization therapy for 3 months
Demographic New York Heart Association class Six-minute hall walk (m) Quality-of-life score Hemodynamic Blood pressure (mmHg) Heart rate (bpm) Echocardiography LV end-systolic volume (mL) LV end-diastolic volume (mL) LV ejection fraction (%) Mitral regurgitation (% LA area) Diastolic dysfunction (% patients) Abnormal relaxation pattern Pseudonormal filling pattern Restrictive filling pattern LA diameter (cm) Indexed LAV-max (mL/m2) Indexed LAV-pre (mL/m2) Indexed LAV-post (mL/m2) LAV-EF (%)
Baseline
3 months after CRT
P value
3.09 ⫾ 0.38 318 ⫾ 104 30 ⫾ 21
2.22 ⫾ 0.59 359 ⫾ 99 20 ⫾ 19
⬍.001 ⬍.001 ⬍.001
105 ⫾ 18 68 ⫾ 10
108 ⫾ 22 71 ⫾ 12
137 183 26.1 33 30 35 35 5.22 44.7 43.3 33.1 23.3
⫾ ⫾ ⫾ ⫾
⫾ ⫾ ⫾ ⫾ ⫾
.341 .411
62 70 7.3 21
111 162 33.6 23
⫾ ⫾ ⫾ ⫾
60 72 9.8 20
⬍.001 .004 ⬍.001 ⬍.001 ⬍.05
0.87 17.4 18.3 22.4 11.0
43 36 21 5.15 44.5 42.7 28.0 31.0
⫾ ⫾ ⫾ ⫾ ⫾
0.93 16.2 17.1 17.1 15.4
.442 .487 .321 .124 ⬍.001
CRT ⫽ cardiac resynchronization therapy; LV ⫽ left ventricular; LA ⫽ left atrial; indexed LAV ⫽ left atrial volume indexed to body surface area; LAV-EF ⫽ left atrial emptying fraction by volume change.
ison of continuous parametric variables between baseline and 3 months after CRT, paired sample t-test was used. Comparison of baseline demographic and echocardiographic parameters between patients with and without newonset AF, survivors and nonsurvivors, and LA responders and LA nonresponders was performed by unpaired t-test or Chi-square test as appropriate. Time to AF onset and death between LA responders and nonresponders was analyzed by Kaplan-Meier method. A Cox proportional hazard model was constructed with AF and death as dependent variables and age, sex, etiology of heart failure, New York Heart Association (NYHA) class, indexed LAV-max and severity of LV diastolic dysfunction, and LA and LV responders as independent variables to predict clinical events. These variables were selected because of their clinical relevance to the outcomes and research question of interest (LA responder). To minimize the risk of overfitting, factors with no significant difference between patients with or those without the outcomes (AF and mortality) in the univariate analysis were not entered into the multivariate model. Thus, for the Cox proportional hazard model with AF as dependent variable, age, sex preponderance, NYHA class, and etiology of heart failure were not entered into the model (see Table 3). Similarly, for the model with death as dependent variable, sex, NYHA class, and etiology of heart failure also were excluded from entering the model. All parametric data are expressed as mean ⫾ SD. P ⬍.05 was considered significant.
Results Among the 97 patients (72 men and 25 women; age 63.8 ⫾ 13.3 years), 82 (85%) were in NYHA class III and 41 (42%)
had ischemic etiology. Medications included diuretics in nearly all patients, angiotensin-converting enzymes inhibitors or angiotensin receptor blockers in 91%, beta-blockers in 72%, and spironolactone in 52%. Amiodarone was prescribed for three patients with defibrillators at baseline for ventricular arrhythmia control. No other antiarrhythmic drug was used by the other patients. Baseline clinical and echocardiographic characteristics of the whole group are given in Table 2. Significant improvements in symptom, exercise capacity, quality of life, and LV reverse remodeling were observed 3 months after CRT. For LA function, significant improvement in peak LA contraction velocity (3.49 ⫾ 1.84 cm/s to 4.11 ⫾ 1.98 cm/s, P ⫽ .009), maximal LA during ventricular systole (16.8% ⫾ 11.0% to 20.73 ⫾ 9.47%, P ⫽ .004), maximal LA during ventricular early diastole (6.38% ⫾ 4.60% to 9.29 ⫾ 5.92%, P ⬍.001), and indexed LAV-EF were observed (Table 2). However, no significant change was observed in indexed LAV-max before and after CRT. Forty-seven patients (48.5%) were LA responders and 57 (58.8%) were LV responders using the definitions given in Table 1. There was no difference in demographic and echocardiographic characteristics between LA responders and nonresponders at baseline (Table 3). The percentage of atrial pacing at 3 months was 36% ⫾ 8% and 41% ⫾ 10% among LA responders and nonresponders, respectively (P ⫽ NS).
New-onset AF After follow-up of 1200 ⫾ 705 days, new-onset AF with mean duration of 401 ⫾ 417 min (range 30 –1680 min) was detected in 26 (26.8%) patients. Six (12.8%) patients among the LA responders and 20 (40%) among the LA nonresponders
Fung et al Table 3
Improvement of LA Function After CRT
783
Baseline patient characteristics
Demographic Age (years) Sex (M/F) New York Heart Association class Ischemic etiology (%) Six-minute hall walk (m) Hypertension (%) Diabetes mellitus (%) Biochemistry Serum potassium level (mmol/L) Medications Angiotensin-converting enzyme inhibitor/angiotensin receptor blocker (%) Beta-blocker (%) Spironolactone (%) Echocardiography LVEF Indexed LAV-max (mL/m2) LAV-EF (%) Mitral regurgitation (% LA area) E/A ratio Abnormal relaxation pattern (%) Pseudonormal pattern (%) Restrictive filling pattern (%)
LA responders (n ⫽ 47)
LA nonresponders (n ⫽ 50)
AF (n ⫽ 26)
No AF (n ⫽ 71)
Survivors (n ⫽ 67)
Nonsurvivors (n ⫽ 30)
62.5 ⫾ 14.6 38/10 3.12 ⫾ 0.41 25 (52) 305 ⫾ 110 18 (38) 19 (40)
65.5 ⫾ 11.7 35/14 3.05 ⫾ 0.39 19 (39) 324 ⫾ 102 22 (44) 26 (52)
66.1 ⫾ 11.7 21/5 3.10 ⫾ 0.35 9 (35) 328 ⫾ 108 10 (38) 16 (62)
63.4 ⫾ 13.7 52/19 3.08 ⫾ 0.42 36 (51) 315 ⫾ 101 30 (42) 29 (41)
61.9 ⫾ 13.2 50/17 3.09 ⫾ 0.41 28 (41) 325 ⫾ 104 28 (42) 30 (45)
69.0 ⫾ 11.9* 23/7 3.07 ⫾ 0.37 17 (57) 310 ⫾ 96 12 (40) 15 (50)
4.2 ⫾ 0.5
4.4 ⫾ 0.6
4.5 ⫾ 0.5
4.3 ⫾ 0.6
4.1 ⫾ 0.4
4.3 ⫾ 0.6
42 (89)
46 (92)
24 (92)
64 (90)
62 (93)
26 (87)
35 (74) 24 (51)
35 (70) 26 (52)
22 (84) 15 (58)
48 (68) 35 (49)
49 (73) 34 (51)
21 (67) 16 (53)
26.3 ⫾ 6.7 44.6 ⫾ 20.0 23.1 ⫾ 10.7 29.4 ⫾ 17.2 1.20 ⫾ 0.59 18 (38) 14 (29) 16 (33)
26.2 ⫾ 8.0 44.6 ⫾ 15.3 23.6 ⫾ 10.1 34.4 ⫾ 24.0 1.35 ⫾ 0.89 11 (22) 20 (41) 18 (37)
26.2 ⫾ 8.1 45.8 ⫾ 16.5 22.1 ⫾ 10.1 33.2 ⫾ 23.1 1.23 ⫾ 0.56 9 (35) 8 (30) 9 (35)
26.2 ⫾ 7.2 44.3 ⫾ 17.8 24.6 ⫾ 9.2 31.9 ⫾ 20.4 1.35 ⫾ 0.66 20 (28) 26 (37) 25 (35)
27.3 ⫾ 6.8 45.1 ⫾ 18.0 24.0 ⫾ 10.2 31.0 ⫾ 16.2 1.44 ⫾ 0.75 19 (28) 22 (33) 26 (39)
24.7 ⫾ 7.1 43.1 ⫾ 16.2 22.7 ⫾ 10.4 36.0 ⫾ 23.0 1.26 ⫾ 0.94 10 (33) 12 (40) 8 (27)
AF ⫽ atrial fibrillation; LA ⫽ left atrial; indexed LAV-max ⫽ left atrial volume measured at left ventricular end-systole indexed to body surface area; LAV-EF ⫽ left atrial emptying fraction by volume change; LVEF ⫽ left ventricular ejection fraction. *P ⬍.05 survivors vs nonsurvivors.
had new-onset AF detected by device or ECG (P ⫽ .002). There was a trend for shorter duration of new-onset AF episode among LA responders than LA nonresponders (117 ⫾ 80 min vs 486 ⫾ 440, P ⫽ .055). All ECG-documented AF episodes were detected by the device as atrial high-rate episodes. Eleven patients had new-onset AF episodes detected by ECG during clinic visit or hospitalization (symptomatic). Six of these patients required intravenous amiodarone for AF conversion; the other five patients received rate-controlling agents. Two patients had undergone electrical cardioversion. All 11 patients reverted back to sinus rhythm after new-onset AF. Three patients in the LA responder group and 12 in the LA nonresponder group had the first episode of AF detected by device during routine interrogation (asymptomatic). At the last follow-up, three patients in the LA responder group and five in the LA nonresponder group were noted to have permanent AF. Kaplan-Meier estimate of the time to new-onset AF detected by device or ECG between LA responders and nonresponders is shown in Figure 1. Kaplan-Meier estimate of AF-free survival at 5 years was 0.85 and 0.55 in LA responders and nonresponders, respectively. Significantly earlier onset of AF was observed among the LA nonresponders (log-rank Chi-square ⫽ 8.44, P ⫽ .004). There was no significant difference in baseline demographic and echocardiographic characteristics between patients with and those without AF (Table 3). By multivariate analysis, LA responder was the only predictor of lower risk
of new-onset AF after CRT (hazard ratio 0.22, 95% confidence interval 0.08 – 0.61, P ⫽ .003).
Mortality Thirty (30.9%) patients died during follow-up. Nine (34.6%) patients developed AF, whereas 21 (29.6%) had no AF before
Figure 1 Kaplan-Meier estimate of time to first onset of atrial fibrillation (AF). First onset of AF was significantly earlier in left atrial (LA) nonresponders (dotted line) than in LA responders (continuous line), defined as ⱖ50% relative increase from baseline LA active emptying fraction after 3 months of cardiac resynchronization therapy.
784
Figure 2 Kaplan-Meier survival curves for left atrial (LA) responders (continuous line) and LA nonresponders (dotted line).
Heart Rhythm, Vol 5, No 6, June 2008 ever, the incidence of AF after CRT is under debate. In an early nonrandomized study,2 three of 36 CRT recipients without a history of AF developed AF over 36 months. In the CARE-HF trial,3 the incidence in patients randomized to CRT was 16.1% for symptomatic episodes and as high as 39.4% when including the device-detected episodes over 29.4 months. In the current study, the Kaplan-Meier estimate of AF-free survival at 5 years was 0.85 and 0.55 in LA responders and nonresponders, respectively. The discrepancy in reported AF incidence probably was related to the fact that only patients without a history of AF were recruited for analysis in the current study and up to 20% of patients in the CRT arm of CARE-HF trial had a history of AF. Therefore, it is not unexpected that the incidence of AF would be higher in patients with paroxysmal AF than in those without a history of AF.
death (P ⫽ NS). Eight deaths (17%) occurred in the LA responder group vs 22 (44%) in the LA nonresponder group (P ⫽ .004). Among the LA responders, 4 patients developed sudden death, 2 died of heart failure exacerbation, and 2 died of noncardiac causes. Among the LA nonresponders, 6 patients developed sudden death, 10 died of heart failure, and 6 died of noncardiac causes. There was no significant difference in baseline clinical and echocardiographic characteristics between survivors and nonsurvivors, except for advanced age noted in nonsurvivors (Table 3). The survival curves between the two groups are shown in Figure 2. LA responders had significantly lower mortality than did LA nonresponders (log-rank Chi-square ⫽ 9.08, P ⫽ .003). LA responders also had significantly lower mortality than LA nonresponders for both LV responders (log-rank Chi-square ⫽ 4.89, P ⫽ .027) or nonresponders (log-rank Chi-square ⫽ 6.06, P ⫽ .013; Figure 3). By multivariate analysis, LA responder (hazard ratio 0.22, 95% confidence interval 0.09 – 0.53, P ⬍.001) and LV reverse remodeling (hazard ratio 0.96, 95% confidence interval 0.93-0.99, P ⫽ .03) were independent predictors of survival after CRT.
Discussion The key finding of the current study is that improvement of LA function after CRT was associated with a lower incidence of AF, which was independent of LV reverse remodeling after CRT. Both LA responder and LV reverse remodeling at 3 months were independent predictors of survival after CRT. LA responder seems to be a novel marker for better outcome after CRT and possibly is incremental to the favorable structural changes in the LV in predicting survival in patients with severe heart failure.
Incidence of AF after CRT Development of AF in patients with chronic heart failure may lead to deterioration in hemodynamic performance, heart failure exacerbation, or even hospitalization. How-
Figure 3 A: Kaplan-Meier survival curves for left atrial LA responders (continuous line) and LA nonresponders (dotted line) among left ventricular (LV) responders, defined as reduction of LV end-systolic volume ⬎10% after 3 months of cardiac resynchronization therapy. B: KaplanMeier survival curves for LA responders (continuous line) and LA nonresponders (dotted line) among LV nonresponders.
Fung et al
Improvement of LA Function After CRT
785
Prevention of AF by CRT
LA and LV responders after CRT
Another question that draws much attention is whether CRT has any favorable impact on AF development. In the CARE-HF trial, there was no difference in the time to first episode of AF after CRT between the medical therapy alone group and the CRT plus medical therapy group.3 However, to avoid ascertainment bias only symptomatic episodes were used for analysis. The clinical importance of asymptomatic AF episodes detected by device was demonstrated in the Mode Selection Trial,11 and atrial high-rate events detected by pacemakers were predictors of total mortality and death or nonfatal stroke.11 The effect of CRT upgrading on susceptibility to atrial tachyarrhythmia detected by device was examined in 28 patients, and significant reduction of tachyarrhythmia episodes was observed.6 In the current study, patients with improvement of LA function after CRT had a significantly lower incidence of AF than did nonresponders. After adjustment for other potential risk factors, such as baseline LAV-max and LV reverse remodeling, improvement of LA function 3 months after CRT was the only independent predictor of AF free survival, suggesting that the physiologic improvement in LA active contractile function is vital for maintenance of sinus rhythm in patients with severe heart failure. Such an improvement in LAV-EF may be considered as an important surrogate marker for favorable response to novel AF therapy, such as AF ablation in heart failure patients, in the future.
LV reverse remodeling is an important marker of favorable prognosis after CRT.21 LA size is a powerful predictor of adverse clinical events, including mortality in patients with cardiovascular diseases.14 –17 LAV is a more robust marker of cardiovascular events than is LA size or area in patients with sinus rhythm27 and was predictive of AF development after cardiac surgery.28 The prognostic value of LAV in patients after CRT has not been explored. In addition, although reversal of LA dilation or recovery of LA function has been observed after medical therapy, whether such improvement translates into clinical benefit has not been established.26 The current study demonstrated the prognostic implication of improvement in LAV-EF after CRT, but not indexed LAV-max alone, as an independent predictor of survival after CRT. The favorable prognosis in terms of survival in LA responders was observed in both LV responders and nonresponders, indicating the importance of improvement of LA function probably was complementary to LV reverse remodeling in determining the prognosis after CRT.
LA reverse remodeling and contractile function LA reverse remodeling has been observed in patients with persistent or permanent AF after CRT.12 However, despite the apparently favorable anatomic improvement, reversal to sinus rhythm was not observed in the study. This might be related to the fact that the diseased LA already was at an advanced stage so that the structural change was not be able to translate into clinical benefit. Another potential explanation is that the improvement of active atrial contractile function was more crucial than reduction in LA size alone. LA- was increased during ventricular systole and early diastole and our results suggest that the atria are more elastic and compliant when subjected to passive stretching. The more dynamic changes in LA and active LA emptying fraction as measured by LAV-EF may reflect structural changes leading to better atrial compliance, contractile function, and maintenance of sinus rhythm after CRT. On the other hand, severe LV diastolic dysfunction was not a predictor of AF development in the current study. Transmitral and pulmonary veins Doppler studies are useful for assessing LV diastolic and phasic LA function, including its role as conduit and reservoir.26 The finding from the current study may suggest that LA active emptying fraction is important in determining AF outcomes in these patients with systolic heart failure, and further studies are warranted to determine its role in AF development.
Study limitations The current study is limited by its small sample size. Our study could not clearly rule out the possibility that the indexed LAV alone after CRT was a predictor of clinical events if more patients were recruited. Not all intracardiac electrograms stored were available for analysis to confirm or refute whether the atrial high-rate episodes were indeed AF. The same cutoff criteria for atrial high-rate detection used in the CARE-HF trial were adopted in the current study. Atrial high-rate episodes lasting ⬎5 minutes could effectively exclude oversensing episodes,29 and atrial rate ⬎220 bpm also could exclude slow atrial tachycardia.11 The real benefit, if any, of CRT on AF incidence could not be clearly determined because the current study was not a randomized comparison trial between CRT and medical therapy. Only a randomized CRT trial with device switches on and off together with monitoring of asymptomatic atrial high-rate episodes could definitively address the controversy about the effect of CRT on AF development.
Conclusion Improvement of LA function after CRT in patients without a history of AF was associated with a lower incidence of AF and mortality than in patients without improvement of LA function. Such an improvement was an independent predictor of survival after CRT regardless of the occurrence of LV reverse remodeling.
References 1. Hunt SA, Abraham WT, Chin MH, et al. ACC/AHA 2005 Guideline Update for the Diagnosis and Management of Chronic Heart Failure in the Adult–Summary Article: A Report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines (Writing Committee to Update the 2001 Guidelines for the Evaluation and Management of Heart Failure): Developed in Collaboration With the American College of Chest Physicians and the International Society for Heart and Lung Transplantation: Endorsed by the Heart Rhythm Society. Circulation 2005;112:1825–1852.
786 2. Fung JW, Yu CM, Chan JY, et al. Effects of cardiac resynchronization therapy on incidence of atrial fibrillation in patients with poor left ventricular systolic function. Am J Cardiol 2005;96:728 –731. 3. Hoppe UC, Casares JM, Eiskjaer H, et al. Effect of cardiac resynchronization on the incidence of atrial fibrillation in patients with severe heart failure. Circulation 2006;114:18 –25. 4. Hugl B, Bruns HJ, Unterberg-Buchwald C, et al. Atrial fibrillation burden during the post-implant period after CRT using device-based diagnostics. J Cardiovasc Electrophysiol 2006;17:813– 817. 5. Adelstein EC, Saba S. Burden of atrial fibrillation after cardiac resynchronization therapy. Am J Cardiol 2007;100:268 –272. 6. Yannopoulos D, Lurie KG, Sakaguchi S, et al. Reduced atrial tachyarrhythmia susceptibility after upgrade of conventional implanted pulse generator to cardiac resynchronization therapy in patients with heart failure. J Am Coll Cardiol 2007;50:1246 –1251. 7. Maisel WH, Stevenson LW. Atrial fibrillation in heart failure: epidemiology, pathophysiology, and rationale for therapy. Am J Cardiol 2003;91:2D– 8D. 8. Fung JW, Sanderson JE, Yip GW, et al. Impact of atrial fibrillation on heart failure with normal ejection fraction: a clinical and echocardiographic study. J Card Fail 2007;13:649 – 655. 9. Hsu LF, Jais P, Sanders P, et al. Catheter ablation for atrial fibrillation in heart failure. N Eng J Med 2004;351:2373–2383. 10. Lin HJ, Wolf PA, Benjamin EJ, et al. Newly diagnosed atrial fibrillation and acute stroke. The Framingham Study. Stroke 1995;26:1527–1530. 11. Glotzer TV, Hellkamp AS, Zimmerman J, et al; MOST Investigators. Atrial high rate episodes detected by pacemaker diagnostics predict death and stroke report of the atrial diagnostics ancillary study of the MOde Selection Trial (MOST). Circulation 2003;107:1614 –1619. 12. Kies P, Leclercq C, Bleeker GB, et al. Cardiac resynchronisation therapy in chronic atrial fibrillation: impact on left atrial size and reversal to sinus rhythm. Heart 2006;92:490 – 494. 13. Yu CM, Fang F, Zhang Q, et al. Improvement of atrial function and atrial reverse remodeling after cardiac resynchronization therapy for heart failure. J Am Coll Cardiol 2007;50:778 –785. 14. Rossi A, Cicoira M, Zanolla L, et al. Determinants and prognostic value of left atrial volume in patients with dilated cardiomyopathy. J Am Coll Cardiol 2002;40:14 –25. 15. Modena MG, Muia N, Sgura FA, et al. Left atrial size is the major predictor of cardiac death and overall clinical outcome in patients with dilated cardiomyopathy: a long-term follow-up study. Clin Cardiol 1997;20:553–560. 16. Moller JE, Hillis GS, Oh JK, et al. Left atrial volume: a powerful predictor of survival after acute myocardial infarction. Circulation 2003;107:2207–2212. 17. Dini FL, Cortigiani L, Baldini U, et al. Prognostic value of left atrial enlargement in patients with idiopathic dilated cardiomyopathy and ischemic cardiomyopathy. Am J Cardiol 2002;89:518 –523.
Heart Rhythm, Vol 5, No 6, June 2008 18. Rossi A, Cicoira M, Bonapace S, et al. Left atrial volume provides independent and incremental information compared with exercise tolerance parameters in patients with heart failure and left ventricular systolic dysfunction. Heart 2007; 93:1420 –1425. 19. Tsang TS, Barnes ME, Gersh BJ, et al. Left atrial volume as a morphophysiologic expression of left ventricular diastolic dysfunction and relation to cardiovascular risk burden. Am J Cardiol 2002; 90:1284 –1289. 20. Rectus TS, Kubo SH, Cohn JN. Validity of the Minnesota Living with Heart Failure questionnaire as a measure of therapeutic response to enalapril or placebo. Am J Cardiol 1993;71:1106. 21. Yu CM, Chau E, Sanderson JE, et al. Tissue Doppler echocardiographic evidence of reverse remodeling and improved synchronicity by simultaneously delaying regional contraction after biventricular pacing therapy in heart failure. Circulation 2002;105:438 – 445. 22. 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;112:1580 –1586. 23. Yu CM, Sanderson JE, Shum IO, et al. Diastolic dysfunction and natriuretic peptides in systolic heart failure. Higher ANP and BNP levels are associated with the restrictive filling pattern. Eur Heart J 1996;17:1694 –1702. 24. Lester SJ, Ryan EW, Schiller NB, et al. Best method in clinical practice and in research studies to determine left atrial size. Am J Cardiol 1999;84: 829 – 832. 25. Fuster V, Ryden LE, Cannom DS, et al; American College of Cardiology; American Heart Association Task Force; European Society of Cardiology Committee for Practice Guidelines; European Heart Rhythm Association; Heart Rhythm Society. ACC/AHA/ESC 2006 guidelines for the management of patients with atrial fibrillation: full text: a report of the American College of Cardiology/American Heart Association Task Force on practice guidelines and the European Society of Cardiology Committee for Practice Guidelines (Writing Committee to Revise the 2001 guidelines for the management of patients with atrial fibrillation) developed in collaboration with the European Heart Rhythm Association and the Heart Rhythm Society. Europace 2006;8:651–745. 26. Abhayaratna WP, Seward JB, Appleton CP, et al. Left atrial size: physiologic determinants and clinical applications. J Am Coll Cardiol 2006;47:2357– 2363. 27. Tsang TS, Abhayaratna WP, Barnes ME, et al. Prediction of cardiovascular outcomes with left atrial size: is volume superior to area or diameter? J Am Coll Cardiol 2006;47:1018 –1023. 28. Osranek M, Fatema K, Qaddoura F, et al. Left atrial volume predicts the risk of atrial fibrillation after cardiac surgery: a prospective study. J Am Coll Cardiol 2006;48:779 –786. 29. Pollack WM, Simmons JD, Interian A, et al. Clinical utility of intraatrial pacemaker stored electrograms to diagnose atrial fibrillation and flutter. Pacing Clin Electrophysiol 2001;24:424 – 429.