Cardiac resynchronization therapy and the relationship of percent biventricular pacing to symptoms and survival

Cardiac resynchronization therapy and the relationship of percent biventricular pacing to symptoms and survival David L. Hayes, MD, FHRS,* John P. Boehmer, MD,† John D. Day, MD, FHRS,‡ F.R. Gilliam III, MD, FHRS,§ Paul A. Heidenreich, MD,e Milan Seth, MS,¶ Paul W. Jones, MS,¶ Leslie A. Saxon, MD, FHRS# From the *Mayo Clinic, Rochester, Minnesota; †Penn State College of Medicine, Hershey, Pennsylvania; ‡Intermountain Heart Rhythm Specialists, Salt Lake City, Utah; §Cardiology Associates of Northeast Arkansas, Jonesboro, Arkansas; e Stanford University, Palo Alto, California; ¶Boston Scientific Corporation, St. Paul, Minnesota; #University of Southern California, Los Angeles, California. BACKGROUND With the advent of cardiac resynchronization therapy, it was unclear what percentage of biventricular pacing would be required to obtain maximal symptomatic and mortality benefit from the therapy. The optimal percentage of biventricular pacing and the association between the amount of continuous pacing and survival is unknown. OBJECTIVE The purpose of this study was to assess the optimal percentage of biventricular pacing and any association with survival in a large cohort of networked patients.

all ventricular beats. Atrial fibrillation and native atrial ventricular condition can limit a high degree of biventricular pacing. Incremental increases in mortality benefit are observed with an increasing percentage of biventricular pacing. CONCLUSION Every effort should be made to reduce native atrioventricular conduction with cardiac resynchronization therapy systems in an attempt to achieve biventricular pacing as close to 100% as possible.

METHODS A large cohort of 36,935 patients followed up in a remote-monitoring network, the LATITUDE Patient Management system (Boston Scientific Corp., Natick, Massachusetts), was assessed to determine the association between the percentage of biventricular pacing and mortality.

ABBREVIATIONS AF ⫽ atrial fibrillation; CRT ⫽ cardiac resynchronization therapy; HF ⫽ heart failure; ICD ⫽ implantable cardioverter-defibrillator; VRR ⫽ ventricular rate regulation

RESULTS The greatest magnitude of reduction in mortality was observed with a biventricular pacing achieved in excess of 98% of

(Heart Rhythm 2011;8:1469 –1475) © 2011 Heart Rhythm Society. All rights reserved.

Introduction

association with bundle branch block.4 – 6 The highest response rates to CRT correlate with maintaining a high percentage of biventricular pacing.7 Effective delivery of continuous CRT may be hindered in the presence of native ventricular condition, due to long AV delay programming, atrial tachycardia, or atrial fibrillation (AF).8 –10 Although AF itself is associated with worsened outcomes in HF patients, CRT seems to improve outcomes in AF.11–15 No clinical trial has evaluated overall mortality and HF symptoms after device implantation in relation to percent biventricular pacing in large numbers of patients with CRT devices that regularly transmit device data and health status information over a network. Earlier studies of CRT therapy in normal sinus rhythm and in the presence of AF have suggested that a greater degree of biventricular pacing may be required to realize all of the benefits of CRT.13,16 Technology advances now enable remote transmission of data collected and stored in CRT devices to be transmitted to protected networks. In the United States, an estimated 13,000 newly implanted implantable cardioverter-defibrillator (ICD) and CRT de-

Heart failure (HF) currently affects an estimated 5.8 million Americans and is one of the leading causes for hospitalization and deaths in the United States.1–3 Cardiac resynchronization therapy (CRT) has been shown to improve outcomes in patients with reduced left ventricular function in

Conflicts of interest: D.L.H. has received consulting fees from Medtronic Inc., St. Jude Medical, Biotronik Inc., Wiley Publishing, Sorin Medical, and Boston Scientific Corp. and royalty fees from Wiley Publishing; L.A.S. has received institutional grant support and institutional fellowship support from Boston Scientific Corp., St. Jude Medical, and Medtronic Inc. and consultant fees from Boston Scientific Corp. and St. Jude Medical; J.D.D. has received consulting fees and lecture honoraria from Boston Scientific Corp. and St. Jude Medical; F.R.G. has received consulting fees from Boston Scientific Corp.; P.A.H. has received consulting fees from Boston Scientific Corp.; J.P.B. has received consulting fees and institutional grant support from Boston Scientific Corp.; M.S. and P.W.J. are employees of Boston Scientific Corp. Address reprint requests and correspondence: Dr. David L. Hayes, Mayo Clinic, 200 First Street SW, Rochester, MN 55905. E-mail address: [email protected]. (Received March 16, 2011; accepted April 6, 2011.)

KEYWORDS Heart failure, Pacemakers, Survival, Registries

1547-5271/$ -see front matter © 2011 Heart Rhythm Society. All rights reserved.

doi:10.1016/j.hrthm.2011.04.015

1470 vice patients are added to a remote network per month and over 350,000 patients transmit data regularly. These transmitted data are accessible to the physicians and provide a unique opportunity to assess device-related data in large patient cohorts.2 The ALTITUDE study was designed to address prospectively defined questions utilizing de-identified data obtained from the LATITUDE remote monitoring system (Boston Scientific Corp., Natick, Massachusetts). We sought to investigate the impact of percent biventricular pacing on mortality.

Methods The LATITUDE project is an independent clinical science initiative launched in 2008 that was formed to prospectively analyze data from implanted ICD and cardiac resynchronization therapy with defibrillation (CRT-D) devices manufactured by Boston Scientific that regularly transmit information remotely. The system, LATITUDE, gained U.S. Food and Drug Administration approval in 2005 and was market released in 2006 for CRT-D and then ICD devices. Compatible devices implanted before system availability (“legacy devices”) were also approved for addition to the LATITUDE system. After 2006, all patients receiving Boston Scientific ICD and CRT-D devices were eligible for enrollment in the LATITUDE remote follow-up network. The system includes a remote communicator that wirelessly interrogates the device where the interrogation rates are programmable by the physician. A patient may initiate a transmission on demand. The data are transferred over the telephone line. Daily and weekly device monitoring can be used to ensure that device function is normal and to alert the clinician of clinically significant events. The collected data are downloaded to designated device implant follow-up clinics from a secure web site that is managed by Boston Scientific.14 Centers using remotely collected data participated in a data use agreement that allows for the use of the data for research purposes in accordance with The Health Insurance Portability and Accountability Act regulations. Six percent of centers participating in the LATITUDE network did not contribute patients to this analysis because they elected not to share data. The decision to place a patient in the remote follow-up system is made by the physician, either at the time of device implantation or at any time later during clinical follow-up. The LATITUDE study group consists of an independent physician leadership panel. The panel prospectively identifies key clinical questions on a yearly basis for analysis and subsequent publication. A charter governs the conduct and relationship of the LATITUDE leadership panel and Boston Scientific personnel.17 For the study, a total of 36,935 patients, followed up at 1,243 U.S. centers, were evaluated. Each was implanted with Boston Scientific LATITUDE-compatible CRT-D devices. Remote monitoring data were collected until the follow-up interval was closed for data analysis on September 24, 2009. An additional 6 months of data collection was allowed for the reporting of deaths occurring before this

Heart Rhythm, Vol 8, No 9, September 2011 date. The median time to first device interrogation was 139 days (mean 367, range 7 to 2,197 days). Patients were implanted for a mean of 954 ⫾ 530 days and followed up for a mean of 587 ⫾ 324 days on the LATITUDE network.

Survival status Survival status was obtained by cross-reference to the Social Security Death Index provided to Boston Scientific for implanted patients on a quarterly schedule. Follow-up for vital status data was continued for 6 months after collection of study data was closed to allow for lag time in reporting. Patients without Social Security numbers were excluded from the analysis and totaled 10% of the total population.

Percent pacing and AF burden Percent biventricular pacing and AF burden were calculated using pacing and sensing histograms. biventricular pacing was calculated as the percent of ventricular beats paced. The burden of AF was estimated as the percent of atrial sensed beats occurring at rates greater than 180 beats/min.18,19 Patients with an average daily burden of ⬎0.5% were considered to have paroxysmal AF.

Symptom status A subset of patients (n ⫽ 10,344) completed weekly patient HF symptom status questions that were transmitted as part of the weekly device remote follow-up. The questions are displayed on the remote communicator, and patients answer the same questions weekly (Table 1). Symptom data were compared with percent biventricular pacing and AF burden for each week in which symptoms were reported (Table 2).

Table 1

Patient symptom classification

Question

Answer

Analysis

Are you feeling unusually fatigued? Have you felt faint or dizzy over the past few days?

No Yes No Once Twice Several times Decreased Remained same No swelling Increased Increased Remained same No difficulty Decreased None or 1 2 3 or more Slept sitting up No Once Few times More than a few times

Stable Worsened Stable Worsened Worsened Worsened Stable Stable Stable Worsened Stable Stable Stable Worsened Stable Worsened Worsened Worsened Stable Worsened Worsened Worsened

Describe the swelling in your ankles, legs, or abdomen over the past few days: Describe your ability to walk or climb stairs over the past few days: How many pillows did you sleep with last night? How often did you wake up breathless last night?

Hayes et al Table 2

Biventricular Pacing Percentage and Survival

Patient symptoms and percent pacing

Question Are you feeling unusually fatigued? Have you felt faint or dizzy over the past few days? Describe the swelling in your ankles, legs, or abdomen over the past few days: Describe your ability to walk or climb stairs over the past few days: How many pillows did you sleep with last night? How often did you wake up breathless last night?

Number of patients

Number of responses

Responses indicating worsening HF, %

7,119 6,887

469,514 462,745

4,551

Median biventricular pacing (%) during weeks with: Stable symptoms

Worsened symptoms

Difference

P value

26.4% 30.3%

95.87% 95.86%

95.20% 95.45%

0.67% 0.41%

⬍.001 ⬍.001

323,363

8.8%

95.91%

94.71%

1.20%

⬍.001

4,781

326,803

9.3%

95.97%

94.79%

1.19%

⬍.001

3,915

274,340

46.6%

95.52%

95.22%

0.31%

.02

5,840

384,316

21.9%

95.70%

94.85%

0.86%

⬍.001

Statistical analyses Percent biventricular pacing and percent AF at the first remote transmission were included as baseline predictors of subsequent mortality in a Cox regression model. Paroxysmal AF at first transmission was used for stratification of patients for Kaplan-Meier mortality analysis. Percent biventricular pacing was analyzed as quartiles because this demonstrated the best fit to the data using Akaike information criterion.20 A cut-point value dividing the population into 2 pacing groups with a maximally different survival pattern was found by an iterative search to maximize the log-rank test statistic. Logistic regression models were used to identify baseline predictors of biventricular pacing above the median. For analysis of symptom status and percent biventricular pacing, weekly values of biventricular pacing and AF burden were used. A sign test was used for univariate analysis of median biventricular pacing values comparing periods with Table 3

1471

stable versus worsened HF symptoms within each patient. A multivariate Friedman analysis was utilized to test for reduction in biventricular pacing during weeks in which patients reported worsened symptoms. For each patient, the percentile rank of weekly biventricular pace percent values was calculated. A logit transform was applied to this data, and the transformed ranks were used as the dependent variable in a repeated-measures ANOVA model accounting for within patient correlation through an auto regressive covariance structure AR1, which assumes weekly pace percent values are influenced by the value collected in the previous week as well as the independent variables (predictors) included in the model. Predictors evaluated in the ANOVA model included worsening symptoms, follow up time, and the estimated burden of AF by week. The authors had full access to the data and take responsibility for its integrity. All authors have read and agree to the article as written.

Patient characteristics and programmed parameters P value (pacing above vs. below median of 98.5%)

Biventricular pacing quartile at first transmission

N Age LRL AV delay Afib burden Implant duration Male Single chamber (bradycardia mode VVI–VVIR) VRR on (% of patients) Biventricular trigger (% of patients)*

I: ⬍94.8%

II: 94.8%–98.5%

III: 98.5%–99.6%

IV: ⬎99.6%

Univariate comparison

Multivariate model

9,234 71.1 ⫾ 10.8 61.7 ⫾ 10.1 144.5 ⫾ 43.7 20.2% ⫾ 34.8% .97 ⫾ 1.2 7,247 (78.5%) 1,400 (15.2%)

9,234 70.8 60.7 138.6 9.9% 1.03 7,016 701

9,233 70.1 60.4 139.3 7.3% 1.02 6,607 479

9,234 68.8 60.4 139.7 5.9% 1.00 5,751 379

⬍.001 ⬍.001 .11 ⬍.001 ⬍.001 ⬍.001 ⬍.001

⬍.001 ⬍.001 .004 ⬍.001 .060 ⬍.001 .800

⬍.001 ⬍.001

.220 ⬍.001

78.4% 33.6%

⫾ 10.8 ⫾ 9.7 ⫾ 34.0 ⫾ 25.9% ⫾ 1.2 (76.0%) (7.6%)

86.8% 40.9%

AV ⫽ atrioventricular; Afib ⫽ atrial fibrillation; LRL ⫽ lower rate limit. *Only available in a subset of device models (n ⫽ 4,216).

⫾ 11.2 ⫾ 9.7 ⫾ 35.0 ⫾ 22.8% ⫾ 1.2 (71.6%) (5.2%)

89.4% 48.4%

⫾ 11.4 ⫾ 9.7 ⫾ 35.0 ⫾ 20.8% ⫾ 1.2 (62.3%) (4.1%)

89.3% 60.8%

1472

Figure 1 A cut-point value was derived that divided the patient population into 2 pacing groups with a maximally different survival pattern. A cut-point value of 98.47% was identified, P ⬍.001, after adjustment for multiple comparisons.

Results Patient characteristics and device programmed parameters are shown in Table 3. Patients were grouped according to the percent of biventricular pacing and AF burden, expressed as percent of daily atrial beat, ⬎180 beats/min. The median and mean biventricular pacing at the first remote follow-up were 98.5% and 94.4%, respectively. In the multivariate Cox regression model adjusting for estimated AF burden, age, and gender, percent biventricular pacing above the median was significantly associated with mortality (P ⬍.001, hazard ratio [HR] ⫽ 1.30, 95% confidence interval 1.23 to 1.37). Increased percentages of biventricular pacing were associated with a significant mortality reduction. The optimal cut-point value described earlier was 98.47% (Fig. 1). Pa-

Heart Rhythm, Vol 8, No 9, September 2011 tients with a biventricular pacing percentage above 99.6% experienced a 24% reduction in mortality compared with the other quartile groups (HR ⫽ 0.76, P ⬍.001) (Fig. 2). Patients with ⬍94.8% biventricular pacing had a 19% increase in mortality (HR ⫽ 1.19, P ⬍.001). The influence of the percentage of biventricular pacing and AF burden as defined earlier is shown in Figure 3. Percent biventricular pacing was also associated with age, with older patients more likely to pace below 95% (P ⬍.001). Male gender was significantly associated with less than optimal biventricular pacing (P ⬍.001, Table 3). Lower levels of biventricular pacing were found at longer AV delay programming. Lower rate programming did not influence percent biventricular pacing. The highest biventricular pacing percentages were observed in patients programmed in the DDD (median ⫽ 98.59) and DDDR (median ⫽ 98.71) modes, respectively. Use of the ventricular rate regulation (VRR) and biventricular trigger were univariately associated with higher biventricular pacing. In a multivariate model, implant duration, single- versus dual-chamber programming, and VRR were no longer significant predictors of percent biventricular pacing. Worsening HF symptoms were associated with a decrease in median biventricular pacing percentage for all 6 of the reported symptoms, compared with periods of stable symptoms (all P ⬍.001, Table 2). The sum of worsening HF symptoms reported in 1 week was a significant predictor of reduced biventricular pacing (P ⬍.001). Multivariate analysis found a significant trend toward a lower biventricular pacing percentage at longer implant durations (P ⬍.001). In an attempt to establish whether biventricular pacing was reduced due to rates beyond which VRR would be programmed to pace, the percentage of atrial pacing, atrial sensing, and ventricular sensed rate was assessed and stratified by AF burden of ⬍5% or ⱖ5%. This is demonstrated

Figure 2 When assessed by quartiles, patients with biventricular pacing percentage above 99.6% experienced a 24% reduction in mortality compared with the other quartile groups (hazard ratio ⫽ 0.76, P ⬍.001). Patients with ⬍94.8% biventricular pacing had a 19% increase in mortality (hazard ratio ⫽ 1.19, P ⬍ .001).

Hayes et al

Biventricular Pacing Percentage and Survival

1473

Figure 3 Assessment by percentage of left ventricular pacing, ⬍98.5% vs. ⬎98.5%, and the burden of atrial fibrillation. (Atrial fibrillation burden was met and the patient included in this analysis when the average daily burden of atrial fibrillation was ⬎0.5% as determined by atrial sensed beats at a rate ⬎180 on the atrial rate histogram.

in Table 4 by quartiles of biventricular pacing percentage. The percentage of biventricular pacing was calculated using pacing, and sensing histograms and CRT devices with zero paced and sensed beats were therefore excluded. When the P value is calculated above versus below the median of 98.5%, all reached statistical significance. In addition, ventricular rate during atrial tachycardia response (ATR) mode switch was analyzed as shown in Table 5. However, it should be noted that this analysis has limitations because only DDD/DDDR patients can have mode switch episodes. In this population, in the quartile with the lowest percentage of biventricular pacing, 15.2% were in a VVI or VVIR pacing mode. Overall, there was decreasing use of VVI or VVIR pacing modes to 4% in the quartile with the highest percentage of biventricular pacing, for an overall total of 8% in a ventricular pacing mode (2,959 of 36,935). In addition, if the V tachycardia rate is fast enough, and detection enhancements for V ⬎ A rate are not programmed on, the episode will be classified as ventricular, not ATR, and therefore not be included in the query.

ALTITUDE study, mortality was inversely associated with the percentage of biventricular pacing both in the presence of normal sinus or paced atrial rhythm and when the atrial rhythm is fibrillation.7,8,10 A novel finding of this study is that the greatest magnitude of reduction in mortality was observed with a biventricular pacing cutoff in excess of 98%. Prior registry and clinical studies have evaluated percent biventricular pacing at the end of study follow-up and confirmed an association between greater pacing percentages and improved outcomes. Our analysis also indicates that both symptoms and event-driven outcomes are continuously influenced by biventricular pacing percentage. This analysis demonstrates the value of monitoring biventricular pacing both at baseline and as a continuous measure during follow-up to maximize CRT therapy impact on survival. Additionally, the acute association between short-term reductions in biventricular pacing and worsening HF symptoms is an important observation that can provide clinicians with an indication that worsening patient symptom status mandates a review of the percentage of pacing. This type of continuous review of CRT therapy delivery is significantly enabled by the use of the remote follow-up system, which pro-

Discussion Consistent with results from significantly smaller patient populations, in this cohort of ⬎30,000 patients from the Table 4

Percentage of atrial pacing, atrial sensing and ventricular sensed rate stratified by atrial fibrillation burden

Mean atrial pace percent Mean atrial sensed rate Mean ventricular sensed rate Abbreviations as in Table 3.

Afib burden

I: ⬍94.8%

II: 94.8%–98.5%

III: 98.5%–99.6%

IV: ⬎99.6%

P value (pacing above vs. below median of 98.5%)

⬍5% ⬎5% ⬍5% ⬎5% ⬍5% ⬎5%

29% 14% 72 ⫾ 163 ⫾ 95 ⫾ 96 ⫾

31% 21% 75 ⫾ 142 ⫾ 98 ⫾ 98 ⫾

33% 24% 76 ⫾ 143 ⫾ 100 ⫾ 99 ⫾

37% 25% 76 ⫾ 150 ⫾ 102 ⫾ 99 ⫾

⬍.001 ⬍.001 ⬍.001 ⬍.001 ⬍.001 ⬍.001

16 72 14 19

13 70 15 13

12 69 17 14

11 70 20 16

1474 Table 5

Heart Rhythm, Vol 8, No 9, September 2011 Ventricular rate during ATR mode switch

N RV rate at start of ATR episode RA rate at start of ATR episode Single chamber (bradycardia mode VVI–VVIR)

I: ⬍94.8%

II: 94.8%–98.5%

III: 98.5%–99.6%

IV: ⬎99.6%

9,234 108 ⫾ 20 224 ⫾ 61 1,400 (15.2%)

9,234 107 ⫾ 22 220 ⫾ 61 701 (7.6%)

9,233 105 ⫾ 22 216 ⫾ 62 479 (5.2%)

9,234 105 ⫾ 22 216 ⫾ 65 379 (4.1%)

P value (pacing above vs. below median of 98.5%) ⬍.001 ⬍.001 ⬍.001 (univariate model) .800 (multivariate model)

ATR ⫽ atrial tachycardia response; RA ⫽ right atrial; RV ⫽ right ventricular.

vides both the symptom status and the percent pacing in a weekly report. Alternatively, if the patient experiences a worsening of symptoms, a remote interrogation can be initiated by the patient. An important clinical message from this real-world registry is the need for vigilance on the part of the clinician following up with the patient in monitoring the percentage of biventricular pacing to achieve the highest possible percent pacing. This may involve advancing AV nodal blocking therapies, reprogramming AV delay, or AV nodal ablation. Despite the limitations from registry data (see later), this analysis would favor abandonment of a clinical reflex to be satisfied, for example, with 95% biventricular pacing. The clinician should not be satisfied until assured that the absolute highest percentage has been achieved, aiming for 100% (Fig. 2). The effect of AF and biventricular pacing on mortality appear to be additive.11,12 In the presence of AF, mortality benefit has been demonstrated with biventricular pacing, but loss of biventricular capture has a real potential to curtail benefit of therapy in patients with native AV nodal conduction.15,21 A high degree of biventricular pacing was obtained in some of the patients with chronic AF in our patient population (Fig. 3). The presence of biventricular fusion (hybrid between paced and intrinsic morphologies) and pseudofusion (pacing artifact delivered but intrinsic morphology not altered) beats may be even higher in patients with underlying AF. This requires specific attention when programming devices for individual patients. In addition, this issue needs further evaluation in studies designed in which the degree of fusion and pseudofusion can be accurately quantified. Our observations suggest that aggressive re-establishment of normal sinus or paced rhythm or timely medical therapeutic intervention to limit AV nodal conduction is warranted. This is particularly true as AF impacts the clinical course of up to 25% of patients who receive CRT.6,15 The percentage of atrial pacing and sensing and the mean ventricular sensed rate were reported (Table 4), as well as the mean right atrial and ventricular rates at the start of ATR episodes in an effort to establish whether biventricular pacing was reduced due to rates beyond which ventricular rate regulation would be programmed to pace. Unfortunately there are a number of confounders that make a definitive answer unreliable. Additional analyses including a lower cut-point or using the AF burden as a continuous metric,

especially looking at small changes in percent biventricular pacing, may be helpful in answering these questions. The heterogeneity of the devices included and the differences in which the atrial rate is displayed in the current analysis may be possible to overcome in future studies as the registry data expand. Although multiple programmable parameters appeared to significantly influence the percentage of biventricular pacing in the univariate analysis, only a few programmable parameters remained significant in the multivariate analysis. Specifically, AV delay programming and biventricular triggering parameters may have a role in improving the percentage of biventricular pacing. The use of ventricular rate regulation and triggered biventricular pacing were (univariately) associated with a higher biventricular pacing rate; we do not know whether these programmable parameters independently offer benefit in terms of clinical outcomes, and it is not possible to suggest definitive practices for optimal programming given the heterogeneity of devices and variation of programming. Further prospective studies are needed to determine whether mortality can be reduced by specific programming strategies that increase the percent of biventricular pacing. When patients experienced worsening HF symptoms, left ventricular pacing decreased in the interval prior to symptoms. The mean magnitude of the difference in median pacing for each patient is relatively small, but the correlation may well have clinical significance and explain diminished response rates to CRT over time. The utility of uniform and consistent assessment of symptomatology in CRT patients remains relatively poorly explored and likely underutilized, even when made available in a remote system.

Study limitations The analysis was performed in a retrospective patient cohort. Fusion/pseudofusion beats could not be evaluated in this study, and effective pacing may be underreported by device measures.18 Prospective studies that allow a beat-bybeat analysis of biventricular capture are required to confirm our findings. Premature ventricular contractions may also be responsible for a reduction in biventricular pacing. Although premature ventricular contraction counters are included and reported upon device interrogation, the way in which the premature ventricular contraction counters

Hayes et al

Biventricular Pacing Percentage and Survival

1475

are incremented and classified differ among pulse generators. Given the heterogeneity of pulse generators included in this patient cohort, it is difficult to perform a meaningful analysis of this variable.

7.

Conclusion

8.

zation on morbidity and mortality in heart failure. N Engl J Med 2005;352:1539 –1549. Koplan BA, Kaplan AJ, Weiner S, Jones PW, Seth M, Christman SA. Heart failure decompensation and all-cause mortality in relation to percent biventricular pacing in patients with heart failure: is a goal of 100% biventricular pacing necessary? J Am Coll Cardiol 2009;53:355–360. Gasparini M, Regoli F, Galimberti P Ceriotti C, Cappelleri A. Cardiac resynchronization therapy in heart failure patients with atrial fibrillation. Europace 2009;11 suppl 5:v82–v86. Maisel WH, Stevenson LW. Atrial fibrillation in heart failure: epidemiology, pathophysiology, and rationale for therapy. Am J Cardiol 2003;91:2D– 8D. Gasparini M, Cappelleri A. Atrial arrhythmias after cardiac resynchronization therapy: an inverse correlation with achieving 100% biventricular pacing and cardiac resynchronization therapy effectiveness. Europace 2010;12:9 –10. 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;107:2920 –2925. Dries DL, Exner DV, Gersh BJ, Cooper HA, Carson PE, Domanski MJ, Studies of Left Ventricular Dysfunction. Atrial fibrillation is associated with an increased risk for mortality and heart failure progression in patients with asymptomatic and symptomatic left ventricular systolic dysfunction: a retrospective analysis of the SOLVD trials. J Am Coll Cardiol 1998;32:695–703. Khadjooi K, Foley PW, Chalil S, et al. Heart rhythm disorders and pacemakerss: long-term effects of cardiac resynchronization therapy in patients with atrial fibrillation. Heart 2008;48:879 – 883. Gasparini M, Auricchio A, Regoli F, et al. 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. J Am Coll Cardiol 2006;48:734 –743. Hoppe UC, Casares JM, Eiskjaer H, et al. Effect of cardiac resynchronization in the incidence of atrial fibrillation in patients with severe heart failure. Circulation 2006;114:18 –25. Mullens W, Grimm RA, Verga T, et al. Insights from a cardiac resynchronization optimization clinic as part of a heart failure disease management program. J Am Coll Cardiol 2009;53:765–773. Saxon LA, Hayes DL, Gilliam FR, et al. Long-term outcome after ICD and CRT implantation and influence of remote device follow-up: the ALTITUDE Survival Study. Circulation 2010;122:2359 – 67. Epub 2010 Nov 22. Gilliam FR, Ewald GA, Margolis D, Sweeney RJ. Observation of new-onset high-rate atrial rhythms by remote monitoring (abstract). J Card Fail 2008;14:S64. Gilliam FR, Ewald GA, Margolis D, Sweeney RJ. Relationship of high-rate atrial rhythm with heart failure decompensation (abstract). Heart Rhythm 2008; 5:S86. Akaike H. A new look at the statistical model identification. IEEE Trans Autom Control 1974;19:716 –723. Kamath GS, Cotiga D, Koneru JN, et al. The utility of 12-lead Holter monitoring in patients with permanent atrial fibrillation for the identification of nonresponders after cardiac resynchronization therapy. J Am Coll Cardiol 2009;53: 1050 –1055.

In this large cohort analyzed from the ALTITUDE study, mortality was inversely associated with the percentage of biventricular pacing, and small gains in the percentage of biventricular pacing are clinically important. Further prospective studies are needed to determine whether mortality can be reduced by specific programming strategies that further increase biventricular pacing.

References 1. Lloyd-Jones D, Adams RJ, Brown TM, et al, American Heart Association Statistics Committee and Stroke Statistics Subcommittee. Heart disease and stroke statistics—2010 update: a report from the American Heart Association. Circulation 2010;121:e46 – e215. 2. Hunt SA, Abraham WT, Chin MH, et al, American College of Cardiology, American Heart Association Task Force on Practice Guidelines, American College of Chest Physicians,International Society for Heart and Lung Transplantation, Heart Rhythm Society. ACC/AHA 2005 guideline update for the diagnosis and management of chronic heart failure in the adult: 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:e154 – e235. 3. Swedberg K, Cleland J, Dargie H, et al, Task Force for the Diagnosis and Treatment of Chronic Heart Failure of the European Society of Cardiology. Guidelines for the diagnosis and treatment of chronic heart failure: executive summary (update 2005): The Task Force for the Diagnosis and Treatment of Chronic Heart Failure of the European Society of Cardiology. Eur Heart J 2005;26:1115–1140. 4. Bristow MR, Saxon LA, Boehmer J, et al, Comparison of Medical Therapy, Pacing, and Defibrillation in Heart Failure (COMPANION) investigators. Cardiac-resynchronization therapy with or without an implantable defibrillator in advanced chronic heart failure. N Engl J Med 2004;350:2140 –2150. 5. Higgins SL, Hummel JD, Niazi IK, et al. Cardiac resynchronization therapy for the treatment of heart failure in patients with intraventricualr conduction delay and malignant ventricular tachyarrhythmias. J Am Coll Cardiol 2003;42:1454 –1459. 6. Cleland JG, Daubert JC, Erdmann E, et al. Cardiac Resynchronization-Heart Failure (CARE-HF) study investigators. The effect of cardiac resynchroni-

9. 10.

11.

12.

13.

14.

15.

16.

17.

18. 19.

20. 21.