Comparison of 1-year Effects of Left Ventricular and Biventricular Pacing in Patients With Heart Failure Who Have Ventricular Arrhythmias and Left Bundle-Branch Block: The Bi vs Left Ventricular Pacing: An International Pilot Evaluation on Heart Failure Patients with Ventricular Arrhythmias (BELIEVE) Multicenter Prospective Randomized Pilot Study
Maurizio Gasparini, MD; Mario Bocchiardo, MD; Maurizio Lunati, MD; Pier Antonio Ravazzi, MD; Massimo Santini, MD; Marco Zardini, MD; Silvia Signorelli, MS; Micaela Passardi, MS; Catherine Klersy, MD, MS 

Am Heart J.  2006;152(1):155.e1-155.e7.  ?2006 Mosby, Inc.
Posted 08/22/2006

Abstract and Introduction

Abstract

Background: Litle is known on the chronic effects of left ventricular pacing (LV) in heart failure.
Methods: Seventy-four patients with LBBB, QRS >130 milliseconds, New York Heart Association class (Bradley DJ, Bradley EA, Braughman KL, et al. Cardiac resynchronization and death from progressive heart failure: a meta-analysis of randomized controlled trials. JAMA 2003;289:730-40.) II, LV ejection fraction (LVEF) <35%, and a class I cardioverter/defibrillator indication were implanted with CRT-D devices and were randomized to either LV or biventricular (BiV) pacing. Response (defined as increases of >5 points increase of LVEF and/or ≥10% 6-minute walking test [6MWT]) between LV and BiV pacing were compared in an attempt to define the number of patients needed to claim noninferiority of LV pacing. In addition, absolute change in LVEF at 12 months in heart failure patients treated with LV pacing was evaluated. The safety of LV pacing was assessed comparing the total number of ventricular arrhythmia episodes, of hospitilazations, and of deaths between the two pacing modes.
Results: The percentage of responders was comparable for both groups (LV = 75%, BiV = 70%, P = .788); based on the 95% CI of the difference between the groups, 1100 patients would be needed to claim noninferiority of LV pacing (with a 5% CI lower limit). LV pacing induced siginificant LVEF increase (5.2%, P = .002). These results remained unchanged after performing adjustment analyses. There were no differences in the numbers of ventricular arrhythmias, hospitalizations, and death events between the 2 pacing modes.
Conclusions: At 12 months, percentage of responders to LV pacing was similar to BIV pacing. Furthermore, LV pacing achieved a significant increase of ejection fraction. LV pacing is both safe and feasible.

Introduction

In patients with ischemic or primary dilated cardiomyopathy and conduction delays, cardiac resynchronization therapy (CRT) has shown to provide significant improvements in quality of life, functional status, exercise tolerance, reducing hospitalizations,[1,2] and mortality.[3,4,5]

Left ventricular (LV) pacing has been proposed[6,7,8,9,10,11,12,13] as an alternative approach to apply cardiac resynchronization. It has been shown that LV pacing induces short-term hemodynamic benefits, but these studies only evaluated acute and short-term effects in a limited number of patients.[7,8,9,12] Two nonrandomized studies[10,11] suggested that the benefits conferred by LV pacing persist for 6 and 12 months, respectively, and are equivalent to those derived from biventricular (BiV) pacing.

This study investigated the midterm effects of LV compared with conventional BiV pacing on a selected group of patients with heart failure (HF) with left bundle-branch block (LBBB) who met indications for an implantable cardioverter/defibrillator (ICD) implantation.

The main aim of the present study was to evaluate the midterm effects of LV pacing and to verify the feasibility of a comparative prospective randomized study between LV and conventional BiV pacing.

Methods

Patients

Patients were eligible for enrolment if they presented with symptomatic HF (NYHA [New York Heart Association] II-IV), with LBBB, LV ejection fraction (LVEF) of ≤35%, a QRS duration of ≥130 milliseconds, and a class I indication for an ICD. All patients were receiving optimal medical therapy; given the particular "arrhythmic" character of the patient population, most patients were also treated with amiodarone. Eligibility criteria are reported in Table I .

Study Design

This study was a prospective, multicenter, randomized, single-blind, parallel, controlled pilot trial (Figure 1) with end-point evaluation at 12 months.

Figure 1. 

Study flowchart. The diagram includes information on bradycardia and ATP for each randomization arm. RV, Right ventricular stimulation only; LV, left ventricular stimulation only; BIV, biventricular stimulation; ATP, ATP mode.

     

End Points. The primary end point aimed to obtain estimates of the effects of LV and of BiV pacing in response to treatment to calculate the size of a study to assess noninferiority of LV as compared with BiV pacing. Responders were considered patients with an absolute increase in LVEF of >5 points and/or a ≥10% relative increase in distance at 6-minute walking test (6MWT) at 12 months. Noninferiority would be claimed if the proportion of responders in LV pacing would be at most 5% lower than BiV.

The first secondary end point aimed to assess absolute change in LVEF after 12 months in patients with LV stimulation. Another secondary end point aimed to assess the safety of chronic LV pacing with respect to BiV by comparing, at 12 months, the total number of ventricular arrhythmia episodes (as detected by the implanted system), hospitalizations, and mortality rates.

Participants. There were 15 European participating centers. All consecutive patients referred at the participating centers and satisfied the eligibility criteria listed in Table I were included. The study protocol, which included a dedicated consent form, was approved by the institutional review board of each participating center. All patients enrolled gave their informed consent.

Randomization and Device Programming. Patients fulfilling the inclusion criteria were implanted with an InSync ICD Model 7272 device (Medtronic Inc, Minneapolis, MN) and an Attain LV lead (Medtronic Inc). The randomization scheme is represented in Figure 1.

A computer-generated simple randomization scheme was used with 1:1 allocation of treatment with either LV or BiV CRT. Randomization was centralized, and treatment assignment was performed by telephone. A single-blind design was applied, with patients not aware of the type of stimulation used.

Although simpler methods for optimizing atrioventricular delay may be used,[6] this study programmed optimal atrioventricular delay based on echocardiographic evaluation of transmitral filling patterns.[14] Ventricular tachycardia/ventricular fibrillation detection and treatment were enabled for both patient groups immediately after implant. Cutoff for ventricular fibrillation detection was set in all patients at >220 beat/min to reduce the possible occurrence of inappropriate shocks. The ventricular tachyarrhythmia cycle-length cutoff and the programming of anti-tachycardia pacing (ATP) settings were individualized based on the particular arrhythmic history of each patient, whereas the chamber from which ATP was delivered was programmed according to study protocol (Figure 1). Investigators were requested to maintain the bradycardia and ATP modes to which the patients were randomized.

Study Procedures. Echocardiographic Procedures. The echocardiographic videotapes, which documented all the required measurements, were centrally analyzed blindly in a core laboratory (IRCCS, Policlinico San Matteo, Pavia, Italy). The variability of analysis in the core laboratory has been previously reported.[15] The evaluation of LVEF using Simpson's method and other conventional echocardiographic measurements strictly conformed to the echocardiography guidelines.[16] In addition, interventricular mechanical delay (IVMD) was evaluated as a marker of interventricular dyssynchrony.

Six-minute Walking Test. The 6MWT was conducted as previously described by Guyatt et al.[17]

Definition of Responders. The percentage of responders to CRT was based on the proportion of patients in each pacing mode group, having reached the end point of an absolute increase in LVEF of 5 points[15] and/or ≥10% relative increase of 6MWT.[18]

Safety Assessment. During follow-up visits to assess the safety of each pacing mode, the number of ventricular arrhythmic episodes (as detected by device counters), hospitalizations, and cardiac deaths were recorded. No safety assessments were made evaluating pacing effectiveness between the 2 pacing modes because the patients enrolled did not present any degree of heart block.

Statistical Methods. Sample Size and Power. The sample size of 37 patients per group for this pilot study was defined on feasibility criteria. Further power and sample size calculations based on the results of the study were performed while keeping a type I error of 0.05 for the primary end point. For noninferiority, a proportion of responders in the LV group, at most, 5% lower than in the BiV group, was considered for calculating sample size. Additional calculation was performed by increasing the value to 10%.

Data Analysis. Descriptive statistics were reported as mean and SD for continuous variables and as absolute frequencies and percentages for categorical variables. For the primary end point, the difference in the proportion of responders in both groups and its 95% CI was calculated. The lower limit of the 95% CI was contrasted against the stated limits of noninferiority (5% below the proportion of responders observed in the BiV group): if the 95% CI interval observed is within the range of −5% to 5%, then noninferiority could be claimed. Fisher exact test was used to compare the proportion of responders in both groups. The analysis was based on the intention-to-treat (ITT) principle. The ITT analysis set included all randomized patients who had the baseline measure of LVEF or 6MWT and at least 1 follow-up measure. Data imputation of the measures at 12 months was performed by forwarding the last available measure. The following sensitivity analyses were performed: patients in whom the primary end point was not measurable were assigned to responder or nonresponder status in 2 separate analyses. On-treatment analysis was performed by excluding patients who dropped out from any cause (including death). Two separate analyses were performed where crossovers were included (as randomized) or excluded.

The secondary end point, which evaluated 12 months of LVEF changes induced by LV pacing, compared baseline and 12-month mean LVEF using paired Student t test for continuous variables. The mean change and its 95% CI were computed. Furthermore, the comparison was adjusted for baseline characteristics (center, baseline LVEF, age, sex, and presence of coronary artery disease [CAD]) by means of a general linear regression model.

For the evaluation of LV pacing safety, we used Fisher exact test to compare proportions, negative binomial regression to compare counts (arrhythmic episodes and hospitalizations), and log-rank test to compare survival.

Stata 8 (StataCorp, College Station, TX) and Query Advisor 4 (Statistical Solutions, Cork, Ireland) were used for computation.

Results

Recruitment and Participant Flow

Enrolment occurred between June 2000 and October 2002, with the last 12th-month follow-up performed in November 2003. The study flowchart course is shown in Figure 2, according to CONSORT guidelines.[19] Seventy-four patients were randomized at a median time of 14 days from implant. The primary end point could be evaluated in 69 patients (36 in the LV group and 33 in the BiV group, respectively) after data imputation. They represent the ITT analysis set. For details on patient exclusions, dropout, and crossovers, refer to Figure 2.

Figure 2. 

Study flowchart. The diagram includes information on the intervention-received dropouts, and deaths for each randomization arm.

     

Baseline Data

Main baseline characteristics of the 69 patients were similar for clinical parameters and drug therapy, indicating successful randomization. No change was permitted to the drug regimen during follow-up. The right ventricular (RV) ICD lead was positioned in all cases at the level of the RV apex. In 81% of the patients, the LV lead was positioned in a lateral/posterolateral region of the LV ( Table II ).

Primary End Point

The primary end point was attained by a comparable proportion of patients in the LV (75%, 27/36 patients) and BiV (70%, 23/33 patients) groups. The 95% CI for the 5% difference in proportion ranged from −15% to 25%. The lower limit was less than the stated value of −5% for noninferiority, and, thus, noninferiority could not be claimed. Furthermore, an advantage of 1 of the 2 treatments (Fisher exact test P = .788) could not be stated. To claim noninferiority of LV as compared with BiV, with a lower limit of 5% less than an observed proportion of 70% in the BiV group, 1100 patients per group would be needed. For a lower limit of 10%, 270 patients per group would be required. In both cases, calculations were performed with α = .05 (1-sided) and power = 80%. Results from the sensitivity analyses for the primary end point are summarized in Table III . As shown, similar results were obtained with the different sensitivity analyses strategy. In particular, the size of the effect (difference of percentage of responders between the 2 groups) was of similar magnitude, and noninferiority could never be claimed.

Secondary End Points

Left ventricular ejection fraction increased from 24.6% ? 6.2% to 29.8% ? 9.7% at 12 months in the LV group, with an absolute change of 5.2% (95% CI 3.0-8.1, P = .002). Upon sensitivity on-treatment analysis, a comparable absolute increase of LVEF = 5.9% (95% CI 2.3-9.5, P = .002) was found. After adjustment for center, baseline LVEF, age, sex, and presence of CAD, a similar absolute change of 5.1% (95% CI 1.9-8.2, P = .002) was observed. No interaction between center and treatment was present. For comparison, LVEF increased from 25.8 ? 6.2 to 30.0 ? 9.6 in the BiV group, with an absolute change of 4.2 (95% CI 1.6-7.0).

Safety was analyzed on all 74 randomized patients. A comparable number of patients with ventricular arrhythmia episodes were observed over time: 13 patients (35%) in the LV group and 11 patients (30%) in the BiV group (P = .802); 64 episodes over 441 person-months in LV group versus 65 episodes over 380 person-months, P = .290. No differences were detected in the rate of hospitalizations (number of hospitalizations per person-months: LV 9/441 person months vs BiV 11/381 person months, P = .219). Moreover, there was no difference regarding total and cardiac survival between the 2 pacing modes (total survival LV: 91%, 95% CI 76%-97%, BiV 85%, 95% CI 66%-93%, log-rank test P = .228; cardiac survival LV: 94%, 95% CI 79%-98%, BiV 89%, 95% CI 72%-96%, log-rank test P = .352). When comparing groups for functional and echocardiographic parameters, a beneficial effect was obtained from both stimulation modes with no significant difference between them (see Table IV ).

Discussion

This pilot study failed to demonstrate noninferiority of LV over BiV pacing for long-term percentage of responders; to address this end point, between 540 and 2200 patients would have been required, depending on how restrictive the criterion for noninferiority.

However, this is the first randomized controlled parallel study demonstrating that LV pacing resulted in a significant 12-month improvement of absolute LVEF in patients with HF with LBBB at high risk for sudden death, which was comparable in magnitude to the improvement observed in the BiV group (+5.2% and +4.2%, respectively). Moreover, chronic LV pacing significantly increased EF at 12 months and was shown to be as safe as commonly indicated BiV pacing for CRT.

Noninferiority of LV Pacing

The study population was similar in its baseline characteristics to the patients enrolled in the MIRACLE (Multicenter InSync Randomized Clinical Evaluation)-ICD trial[20] and the InSync Italian Registry,[21] although a consistent proportion of patients were in NYHA class II. The combined end point was attained by a similar proportion of patients, 75% and 70% of enrolled patients for LV and BiV pacing, respectively. The observed percentages of responder patients were similar to those reported in previous series.[1,11,21] However, this was a pilot study and obviously was not powered to show noninferiority of LV pacing with respect to BiV pacing, and, indeed, this could not be claimed. Sample size calculations showed that as many as 1100 patients per group would be needed to demonstrate noninferiority with a limit set 5% below an observed proportion of 76%. This raises questions about the feasibility of such a study, even in a multicenter context.

Left Ventricular Pacing Improves EF After 12 Months

The present study prospectively demonstrated that LV pacing induced a significant absolute increase in EF in a randomized group of patients with HF with LBBB at high risk for sudden cardiac death. Ejection fraction increased by an average of 20% above baseline values. Eight crossovers were observed: although a higher number of crossovers occurred in the LV group, this was not related to a higher occurrence of arrhythmias. The incidence of crossover could theoretically have influenced the results of the study. However, similar significant results were observed after adjustment according to the principal baseline characteristics and performing different sensitivity analyses, thus, strongly supporting the finding that LV pacing induces a significant increase of EF at 1-year follow-up.

Leclercq et al[22] previously showed that in subjects with LBBB, efficient mechanical resynchronization could be achieved by LV pacing without achieving electrical synchrony. In our study, patients enrolled in the left stimulation group presented significantly greater-paced QRS duration and QRS-duration lengthening after CRT, compared with the baseline QRS ( Table IV ): nonetheless, results of responders, increase in EF, and the reductions in hospitalizations were all comparable to those observed in the BiV group. Moreover, in the LV-paced patient subgroup, a greater (although nonsignificant) shortening of IVMD occurred because of LV-only pacing CRT ( Table IV ). Taken together, these data reflect the ?difficulties? in adequately evaluating effective electrical or mechanical resynchrony after CRT.

Moreover, the finding that EF increases significantly after 12 months confirms that the favorable hemodynamic changes on LV systolic function by LV pacing in the ?acute? phase[6,7,8] can be maintained at 1-year follow-up. Touiza et al[10] and Blanc et al[11] reported only a modest or nonsignificant improvement of EF with LV pacing. Our data more convincingly show the beneficial effects of LV pacing: at 12 months, in our experience, LV pacing significantly improves EF and, like BiV pacing, produces similar favorable effects on indices of LV performance.

Safety of Chronic LV Pacing

It has been hypothesized that LV pacing may lead to the worsening of electrical dispersion compared with BiV pacing.[22] This, in theory, could increase the risk of malignant ventricular arrhythmias. In our experience, sudden cardiac death occurred in only 2 patients. Moreover, a comparable number of ventricular arrhythmia episodes were observed in LV and BiV groups.

Therefore, data from our study suggest that LV pacing does not appear to have any proarrhythmic effect, although such a finding is mitigated because most patients were treated with Amiodarone. The mortality rate observed in our LV group seemed lower than that reported.[10,11] Because LV pacing alone did not show any significant differences concerning hospitalization rate and survival rate at 1-year follow-up, this pacing mode may be considered ?safe.?

Clinical Implications of the Study

Expanding indications for primary prevention of sudden cardiac death in patients with severely compromised LV function[4] have rendered compulsory the implant of an RV coil and, therefore, limit the practical applicability of LV pacing mode for CRT.

Some possible areas of clinical applicability include patients with HF with important comorbidities, which determine a dismal long-term prognosis (<2-3 years). A ?reasonable? clinical (and ?cost-effective?) choice may be to offer symptomatic relief by delivering CRT through LV-only pacing (without an RV ICD lead); technically, this would imply implanting a conventional double or single (in patients with atrial fibrillation) chamber pacemaker.

Study Limitations

Our study had a single-blind parallel design. The weakness of a single-blind design was partially overcome by the centralized blind core-laboratory evaluation of echocardiographic data.

The main limitation is linked mostly to the low number of patients enrolled in each treatment group, being this a pilot study. Still, we were able to demonstrate a significant increase of EF at 12 months with LV pacing. Because of the limited patient number, the positive results observed in the 2 groups did not allow to make any conclusive statement on the effects of CRT in NYHA class II patients.

Conclusions

This pilot study demonstrated a substantially similar global effect of the 2 pacing modalities in a 12-month follow-up. Also, the study demonstrated with a prospective and randomized design that LV pacing induces a significant absolute increase of LVEF in patients with HF presenting LBBB and at high risk for sudden death. Moreover, 1-year morbidity and mortality were comparable, thus, for the first time, portraying a long-term safety profile supporting LV pacing.

The study also allowed to define a sample size suitable for demonstrating noninferiority of LV pacing, with the knowledge that a comparable proportion of responders could be reached in both groups. Whether it may be worthy to invest in a large confirmatory trial to demonstrate noninferiority, based on the sample size calculated here, remains questionable.

Table I. Eligibility Criteria for Participants


(A) Inclusion criteria
At least 1 episode of cardiac arrest due to VT;
Recurrent spontaneous or inducible sustained VT; or
Prior MI, LVEF ≤35%, documented episode of nonsustained VT and inducible VT; and
NYHA II-IV class HF, LVEF ≤35%, LVEDD ≥55 mm, LBBB pattern, and QRS duration ≥130 ms;
Stable medical regimen for at least 1 m before randomization; no changes in ACE inhibitors and β-blockers regimen in the 12 m after randomization were permitted;
Aged between 18 and 80 y.

(B) Exclusion criteria
Unstable angina, acute MI, coronary-aortic bypass graft, or percutaneous transluminal coronary angioplasty within the past 3 m;
Cerebral vascular attacks or transient ischemic attack within the last month;
Chronic atrial arrhythmias, paroxysmal atrial fibrillation events, or cardioversion for atrial fibrillation within the last month;
Pacemaker dependency;
Primary valvular disease not treated by surgical replacement;
≤6 m survival expectancy;
Mechanical right heart valves;
VT associated with reversible causes;
Enrolled in another study;
Pregnancy;
Refusal of study informed consent;
Expected lack of compliance during follow-up.

ACE, Angiotensin-converting enzyme; VT, ventricular tachycardia; MI, myocardial infarction; LVEDD, left ventricular end-diastolic diameter.

 

Table II. Baseline Characteristics


  Left (n = 36) Biv (n = 33) All*
Mean age [y (SD)] 66.5 (7.3) 66.9 (7.6) 66.7 (7.4)
Men [n (%)] 34 (94.4) 29 (87.9) 63 (91.3)
CAD [n (%)] 25 (69.4) 19 (57.6) 44 (63.8)
Mean QRS width in ms (SD) 169 (31) 176 (25) 172 (28)
NYHA III-IV [n (%)] 24 (66.7) 19 (57.6) 43 (62.3)
Mean 6MWT (SD) 350 (113) 346 (120) 348 (115)
Mean LVEF (SD) 24.6 (6.2 ) 25.8 (6.2) 25.2 (6.2)
Mean LVESV (SD) 242 (68) 238 (81) 240 (74)
Mean IVMD in ms (SD) 33.6 (34.3) 36.3 (33.2) 35.0 (33.5)
β-Blockers (%) 49 51 50
ACE inhibitors (%) 81 84 82
Diuretics (%) 89 97 93
Amiodarone (%) 65 57 61
Lateral/posterolateral LV lead (%) 78 84 81

LVESV, Left ventricular end-systolic volume.
*No significant differences between the 2 groups in any of their baseline clinical and demographic characteristics.

 

Table III. Sensitivity Analyses for the Primary End Point


Analysis Responder LV (%) Responder BiV (%) Difference (95% CI) Noninferiority demonstrated
Primary 75 70 5 (−16 to 26) No
Sensitivity 1 (points not evaluable assigned to responder) 73 62 11 (−10 to 32) No
Sensitivity 2 (points not evaluable assigned to nonresponder) 76 73 13 (−17 to 23) No
On-treatment 1 (dropouts from any cause excluded) 72 70 2 (−22 to 25) No
On-treatment 2 (dropouts from any cause and crossover excluded) 70 73 3 (−27 to 21) No

 

Table IV. Comparison of Changes in Global Clinical, Instrumental, and Functional Evaluations at 12 Months for the 2 Pacing Modes


Variable LV (n = 36) BiV (n = 33) Left-BiV difference P
Mean (95% CI) Mean (95% CI) Mean (95% CI)
6MWT (absolute change, m) 48.1 (9.9 to 86.4) 69.7 (35.2 to 104.3) −21.6 (−72.5 to 29.3) .400
LVEF (points) 5.2 (3.0 to 8.1) 4.2 (1.6 to 7.0) 1.0 (−3.2 to 4.9) .701
LVESV (mL) −19.7 (−34.5 to −4.9) −25.5 (−44.2 to −6.8) 5.8 (−17.4 to 29.0) .617
Paced QRS duration (ms) 28 (16 to 40) −23 (−28 to −18) 51 (22 to 79) .0001
IVMD (ms) −27.6 (−45.0 to −10.3) −11.8 (−24.4 to 0.7)* −15.8 (−36.7 to 5.2) .136
Change to lower NYHA [n (%)] 25 (69.4%) 19 (52.9%) 16.5% (−6.0 to 39.0) .220

*Non-significant (P < .05) change from baseline.

 



References

  1. Abraham WT, Fisher WG, Smith AL, et al, MIRACLE Study Group. MIRACLE Study Group. Cardiac resynchronization in chronic heart failure. N Engl J Med 2002;46:1845-5.
  2. Cazeau S, Leclercq C, Lavergne T, et al. Multisite Stimulation in Cardiomyopathies (MUSTIC) Study Investigators. Effects of multisite biventricular pacing in patients with heart failure and intraventricular conduction delay. N Engl J Med 2001;44: 29-4.
  3. Bradley DJ, Bradley EA, Baughman KL, et al. Cardiac resynchronization and death from progressive heart failure: a meta-analysis of randomized controlled trials. JAMA 200;289:70-40.
  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;50:2140-50.
  5. Cleland JG, Daubert JC, Erdmann E, et al. The effect of cardiac resynchronization on morbidity and mortality in heart failure. N Engl J Med 2005.
  6. Auricchio A, Stellbrink C, Block M, et al, for the Pacing Therapies for Congestive Heart Failure Study Group. Effect of pacing chamber and atrioventricular delay on acute systolic function of paced patients with congestive heart failure. Circulation 1999;99:299-001.
  7. Blanc JJ, Etienne Y, Gilard M, et al. Evaluation of different ventricular pacing sites in patients with severe heart failure. Results of an acute hemodynamic study. Circulation 1997;96:27-7.
  8. Kass D, Chen-Huan C, Curry C, et al. Improved left ventricular mechanics from acute VDD pacing in patients with dilated cardiomyopathy and ventricular conduction delay. Circulation 1999;99:1567-7.
  9. Nelson GS, Berger RD, Fetics BJ, et al. Left ventricular or biventricular pacing improves cardiac function at diminished energy cost in patients with dilated cardiomyopathy and left bundle-branch block. Circulation 2000;102:05-9.
  10. Touiza A, Etienne Y, Gilard M, et al. Long-term left ventricular pacing: assessment and comparison with biventricular pacing in patients with severe congestive heart failure. J Am Coll Cardiol 2001;8:1966-70.
  11. Blanc JJ, Bertault-Valls V, Fatemi M, et al. Midterm benefits of left univentricular pacing in patients with congestive heart failure. Circulation 2004;109:1741-4.
  12. Auricchio A, Stellbrink C, Butter C, et al. Pacing Therapies in Congestive Heart Failure II Study Group. Long-term clinical effect of hemodynamically optimized cardiac resynchronization therapy in patients with heart failure and ventricular conduction delay. J Am Coll Cardiol 2002;9:2026-.
  13. Garrigue S, Bordachar P, Reuter S, et al. Comparison of permanent left ventricular and biventricular pacing in patients with heart failure and chronic atrial fibrillation: prospective haemodynamic study. Heart 2002;87:529-4.
  14. Dupuis JM, Kobeissi A, Vitali L, et al. Programming optimal atrioventricular delay in dual chamber pacing using peak endocardial acceleration measurements. Pacing Clin Electrophysiol 200;26(1 Pt 2):210-.
  15. Ghio S, Freemantle N, Serio A, et al. Baseline echocardiographic characteristics of heart failure patients enrolled in a large European multicentre trial (Cardiac Resynchronisation Heart Failure study). Eur J Echocardiography 2005 [Epub ahead of print].
  16. Schiller NB, Shah PM, Crawford M, et al. Recommendations for quantification of the left ventricle by two-dimensional echocardiography. American Society of Echocardiography Committee on Standards. Subcommittee on Quantification of Two-dimensional Echocardiograms. J Am Soc Echocardiogr 1989;2:58-67.
  17. Guyatt GH, Sullivan MJ, Thompson PJ, et al. The 6-minute walk: a new measure of exercise capacity in patients with chronic heart failure. CMAJ 1985;12:919-2.
  18. Yu CM, Abraham WT, Bax J, et al, on behalf of the PROSPECT Investigators. Predictors of response to cardiac resynchronization therapy (PROSPECT)? study design. Am Heart J 2005;149:600-5.
  19. Altman DG, Schulz KF, Moher D, et al. CONSORT GROUP (Consolidated Standards of Reporting Trials). The revised CONSORT statement for reporting randomized trials: explanation and elaboration. Ann Intern Med 2001;14:66-94.
  20. Young JB, Abraham WT, Smith AL, et al. Multicenter InSync ICD Randomized Clinical Evaluation (MIRACLE ICD) Trial Investigators. Combined cardiac resynchronization and implantable cardioversion defibrillation in advanced chronic heart failure: the MIRACLE ICD Trial. JAMA 200;289:2685-94.
  21. Gasparini M, Lunati M, Bocchiardo M, et al. Cardiac resynchronization and ICD therapy: preliminary results from the InSync ICD Italian Registry. Pacing Clin Electrophysiol 200;26:148-51.
  22. Leclercq C, Faris O, Tunin R, et al. Systolic improvement and mechanical resynchronization does not require electrical synchrony in the dilated failing heart with left bundle-branch block. Circulation 2002;106:1760-3.

Appendix A. BELIEVE Investigators

Echocardiographic core laboratory: IRCSS Policlinico S. Matteo, Pavia, Italy: S Ghio:

  1. IRCCS Istituto Clinico Humanitas Mirasole, Milan, Italy: P Galimberti, F Regoli;
  2. Ospedale Civile, Asti, Italy: F Gaita, P Di Donna;
  3. Ospedale Niguarda Ca' Granda, Milan, Italy: G Cattafi, G Magenta;
  4. Ospedale SS Antonio e Biagio, Alessandria, Italy: PA Ravazzi, P Diotallevi;
  5. Istituto Cliniche Gavazzeni, Bergamo, Italy: M Zardini;
  6. Ospedale San Filippo Neri, Rome, Italy: R Ricci, C Pignalberi;
  7. IRCSS Policlinico S. Matteo, Pavia, Italy: L Tavazzi, M Landolina;
  8. Istituto Clinico ?Mater Domini,? Castellanza, Italy: T Forzani, M Tritto;
  9. Ospedale San Gerardo dei Tintori, Monza, Italy: A Vincenti, A. Cirò
  10. APSS Ospedale S. Chiara, Trento, Italy: M Disertori, L Gramegna
  11. Ospedale S. Giovanni di Dio, Cagliari, Italy: R Pirisi;
  12. Clinica Medica I Universit?di Firenze?Policlinico Careggi, Firenze, Italy: L Padeletti, A Colella;
  13. H?ital Europ?n Georges Pompidou, Paris Cedex 15, France: T Lavergne;
  14. Hospital General de Valencia, Valencia, Spain: A Quesada;
  15. Clinica Sanitaria La Paz, Madrid, Spain: R Peinado, JL Merino.
Reprint Address

Maurizio Gasparini, MD, IRCCS Istituto Clinico Humanitas, Via Manzoni 56 Rozzano-MILANO ? Italy. E-mail: maurizio.gasparini@humanitas.it
Maurizio Gasparini, MD,a Mario Bocchiardo, MD,b Maurizio Lunati, MD,c Pier Antonio Ravazzi, MD,d Massimo Santini, MD,e Marco Zardini, MD,f Silvia Signorelli, MS,g Micaela Passardi, MS,g and Catherine Klersy, MD, MSh

aIRCCS Istituto Clinico Humanitas, Rozzano, Italy
bOspedale Civile, Asti, Italy
cOspedale Ca Granda Niguarda, Milan, Italy
dOspedale SS. Antonio e Biagio, Alessandria, Italy
eOspedale San Filippo Neri, Roma, Italy
fCliniche Gavazzeni Humanitas, Bergamo, Italy
gMedtronic, Milano, Italy
hIRCCS Policlinico San Matteo, Pavia, Italy