Effects of Long-Term Biventricular Stimulation for Resynchronization on Echocardiographic Measures of Remodeling
Background— Long-term ventricular resynchronization therapy improves symptom status. Changes in left ventricular remodeling have not been adequately evaluated.
Methods and Results— Fifty-three patients with systolic heart failure and bundle-branch block underwent implantation of biventricular stimulation (BVS) devices as part of a randomized trial. Echocardiograms were acquired at randomization and at 6-week intervals until completion of 12 weeks of continuous BVS. There were no changes in heart rate or QRS duration after 12 weeks of BVS. Serum norepinephrine values did not change with BVS. After 12 weeks of BVS, left atrial volume decreased. Left ventricular end-systolic and end-diastolic dimensions and left ventricular end-systolic volume also decreased after 12 weeks of BVS. Sphericity index did not change. Measures of systolic function, including left ventricular outflow tract and aortic velocity time integral and myocardial performance index, improved.
Conclusions— Long-term resynchronization therapy results in atrial and ventricular reverse remodeling and improved hemodynamics.
Received November 9, 2001; revision received January 10, 2002; accepted January 12, 2002.
Bundle-branch block leads to abnormal ventricular depolarization and introduces dyssynchrony between right and left ventricular (LV) contraction and relaxation.1–6⇓⇓⇓⇓⇓ In patients with heart failure, most interventricular conduction delay is left bundle-branch block, and the incidence is 20% to 30% in New York Heart Association symptom class III to IV.7,8⇓
Studies have identified interventricular conduction delay as an independent predictor of worsened symptom status and cardiac mortality in patients with heart failure.7,9⇓ The sequelae of ventricular dyssynchrony attributable to bundle-branch block in heart failure include loss of coordination of contraction and relaxation leading to increases in regional and global wall stress, reductions in stroke volume and in the rate of rise of LV pressure, diminished diastolic filling time, prolongation of mitral regurgitation, and diminished effective ejection time.4–6⇓⇓
These observations have led to several acute and uncontrolled nonrandomized chronic studies of LV and biventricular stimulation aimed at restoring synchrony.1,6,10–16⇓⇓⇓⇓⇓⇓⇓⇓ The acute hemodynamic responses to biventricular stimulation (BVS) appear instantaneously at the onset of stimulation, and the predominant effect is on parameters of systolic function.14–16⇓⇓ Acutely, the hemodynamic benefit reported during LV or BVS correlates with baseline QRS prolongation, but acute shortening of the QRS is not correlated with hemodynamic response.14–16⇓⇓ Long-term studies published to date report improved symptom class and exercise duration.10,17⇓
The objective of this study was to assess the effects of long-term ventricular resynchronization therapy with BVS on echocardiographic parameters and neurohormonal surrogates of reverse remodeling in patients with symptomatic systolic heart failure and prolonged QRS duration.
Fifty-three consecutive patients who were enrolled in the VIGOR-CHF trial underwent implantation of BVS devices between October 1996 and April 1998 at 19 centers. The VIGOR-CHF trial (Guidant Inc, St Paul, Minn), was a multicenter, randomized, clinical trial. Inclusion criteria were dilated cardiomyopathy, NYHA functional class III to IV heart failure, QRS duration ≥120 ms, and PR interval ≥160 ms. The study was closed in September 2000 after 53 device implants because of slow patient enrollment. Factors contributing to slow enrollment were the need for a mini-thoracotomy for placement of the epicardial LV lead and the emergence of competing investigational protocols requiring a less-invasive procedure that uses a transvenous coronary sinus approach for placement of the LV epicardial lead.
The primary end point of the study was exercise capacity assessed at 6-week intervals by cardiopulmonary exercise testing. Secondary end points included echocardiographic measures of LV function. We report the echocardiographic findings. All patients provided written informed consent before participating in the study, which was performed under the Food and Drug Administration Investigational Device Exemption No. G960112.
Patients were randomized at 1 to 2 weeks after device implantation in a 2:1 ratio to no BVS or BVS. The first group was assigned to no BVS for 6 weeks and then to 12 weeks of continuous BVS. The second group received BVS continuously for 18 weeks (Figure 1). Data were acquired at 6-week intervals. Twelve-week echocardiographic data are compared with baseline randomization and were obtained at the 18-week visit in patients initially randomized to no BVS and at the 12-week visit in patients randomized to BVS. Biventricular stimulation was performed after atrial sensing (VDD mode). The atrioventricular (AV) delay was programmed short enough to ensure complete biventricular capture with exercise.
The devices (VIGOR-CHF model 1240 congestive heart failure device) were designed with a trilumen lead connector. Right atrial and right ventricular stimulation leads were implanted using a transvenous approach. The epicardial stimulation lead was placed at the LV apex or lateral LV free wall using a limited thoracotomy.
A total of 35 patients had adequate quality echocardiograms for quantitative analysis at the core echocardiographic laboratory at the University of California, San Francisco. Echocardiograms were analyzed with the reader blinded to the QRS. Interobserver variability at the core laboratory for assessment of LV end-systolic and -diastolic volume was 3.0±25.l mL and 3.5±29.4 mL, respectively. At randomization, all study patients had been on a stable medical heart failure drug regimen for at least 3 months before randomization. There were no significant changes in the use of heart failure medical therapies during the course of the trial. A total of 27% of patients were receiving β-receptor blocking agents, 24% were taking low-dose amiodarone, and 94% were on ACE inhibitor therapy. At each visit, in those patients receiving BVS, echocardiograms were acquired during BVS and then reacquired during normal sinus rhythm.
Echocardiographic images of the LV were acquired in the standard views as recommended by the American Society of Echocardiography Committee. Previously validated techniques were used for measurement of myocardial performance index (MPI), transmitral Doppler flow, and pulmonary venous flow analysis.18–23⇓⇓⇓⇓⇓
MPI was computed as the sum of isovolumic contraction and relaxation time divided by left ventricular ejection time (LVET) (Figure 2). Interval “a” is measured as the time from mitral valve closure to subsequent mitral valve opening. Interval “b” is measured as LVET. Sphericity index was calculated by dividing the maximum short-axis by the maximum long-axis dimension.24
Collection of blood for norepinephrine determination was drawn with the patient in the supine resting state for at least 30 minutes after IV access was obtained. Norepinephrine levels were determined by high-performance liquid chromatography electrochemical detection (coefficient of variation, 3.3 to 4.4; Quest Diagnostics Nichols Institute). The normal range for serum norepinephrine is 112 to 658 pg/mL.
The clinical characteristics of the 35 patients are shown in Table 1.
Paired t tests were used to compare data obtained at randomization and at each 6-week follow-up visit. Pearson’s correlation coefficient was used to establish correlates with the MPI and LV volumes. Data are presented as mean±SD.
Baseline Randomization Compared With 6 Weeks of No BVS or BVS
Table 2 compares randomization to 6-week echocardiographic measures of LV remodeling and systolic function. At randomization, there were no differences between those patients randomized to no BVS or BVS.
At the 6-week interval, neither the no BVS nor the BVS patients had a significant change in any echocardiographic measure, except AV velocity time integral, which improved in the BVS group. No measure of diastolic function changed in either group.
Baseline Randomization Compared With 12 Weeks of BVS
Table 3 lists the changes in measured variables associated with 12 weeks of BVS.
Echocardiographic Intervals and Serum Norepinephrine
Neither sinus rate nor QRS duration decreased after 12 weeks of BVS. The extent of QRS delay at baseline correlated with baseline LV end-systolic (r=43, P=0.03) and end-diastolic (r=.11, P=0.02) volume and LV mass (r=.38, P=0.04). However, the extent of QRS delay at baseline or extent of narrowing with BVS did not correlate with improvement in any echocardiographic measure.
Effects of BVS on LV Volumes, Dimension, and Mass
Significant decreases were observed in left atrial volume indexes. LV end-systolic volume decreased from 100.3±36 mL/m2 versus 92.1±40 mL/m2, P=0.02, (Figures 3A though 3D). A >10% decrease was observed in 50% of patients. LV end-systolic dimension index and end-diastolic dimension index decreased. A >10% decrease was observed in 39% and 18% of patients, respectively. When BVS was acutely programmed off and echocardiographic measurements were repeated, the reduction in systolic volume and dimension was maintained.
Sphericity index was markedly abnormal at baseline and did not change from baseline to 12 weeks of BVS. LV mass index did not significantly change after 12 weeks of BVS.
Effects of BVS on Measures of Systolic Function
There was a trend toward an increase in LV ejection fraction. Both aortic and LV outflow tract velocity time integral improved.
Eighteen of 24 patients (75%) demonstrated a reduction in MPI after 12 weeks of BVS. The degree of MPI abnormality at baseline correlated with the magnitude of improvement in MPI with BVS (r=−0.83, P<0.0003). The 6 patients showing no reduction in MPI during therapy each had randomization MPI values in the normal range. The component of MPI, isovolumic contraction time (ICT), significantly shortened with BVS. Neither IRT nor LVET showed significant change.
Effects of BVS on Parameters of Diastolic Function
The only measure of diastolic function that improved was mitral deceleration time (Figure 4).
This study provides the first randomized data from a United States study of long-term BVS to achieve ventricular resynchronization therapy. Improvement in several independent echocardiographic measures of LV systolic function that are well validated and have prognostic significance in heart failure subjects were observed.25,26⇓
Effects of BVS on Electrical and Neurohormonal Correlates of Remodeling
The fact that we did not observe significant decreases in heart rate with BVS differs from uncontrolled European clinical trial data.10 This may be because of the fact that patients were required to be optimized on medical therapy before enrollment and that nearly one third of patients were taking either β-adrenergic receptor blocking agents or amiodarone. QRS narrowing was not consistently achieved with BVS, a finding that in a limited number of patients has been shown to correlate with improvement in clinical status.11 Our data are consistent with those from several acute studies demonstrating that QRS narrowing with BVS may be uncoupled from the beneficial mechanical systolic response.15,16⇓ This suggests that QRS duration at baseline may be a relatively crude measure of dyssynchrony. In addition, in this study, no attempt was made to find the LV stimulation site that would result in the narrowest QRS duration.
The finding that plasma norepinephrine levels did not significantly increase with BVS is reassuring. Many drug therapies that augment systolic function and neurohormonal activation have exhibited deleterious effects on survival when used over a long duration.27–30⇓⇓⇓
Improvement in Correlates of Remodeling with BVS
The finding that 12-week but not 6-week BVS results in significant reductions in left atrial size, LV end-systolic volume, and LV dimensions is a provocative one. The observation that these benefits are maintained with brief cessation of stimulation suggests that ventricular resynchronization therapy favorably influences the remodeling process in a time-dependent manner. Potential mechanisms responsible for reductions in LV end-systolic volume include therapy-induced improvement in contractile response or decreased systemic vascular resistance.27 It seems that resynchronization therapy is chronically changing the contractile response of the ventricle. The lack of change in ventricular geometry, measured by the sphericity index, may reflect the fact that measures were performed after only 12 weeks of therapy and a longer interval may be required to measure change in this variable.24
Some drug therapies for heart failure that reverse remodeling may improve the contractile response but also decrease LV mass and systemic vascular resistance.31–33⇓⇓ This may be in part attributable to beneficial direct and indirect drug effects at the level of the myocyte. The finding that BVS does not result in a decrease in LV mass may be attributable to the small number of study patients. The primary action of BVS to reverse the remodeling stimulus can be attributed to long-term enhancement of the synchrony of contraction.
Effects of BVS on Measures of Systolic Function
The finding that LV ejection fraction did not significantly increase during BVS despite decreases in ventricular volumes is not unexpected. Previous trials of ACE inhibitors and β-blocker therapies for heart failure have not documented consistent improvements in ejection fraction despite volume reductions.26,31,32⇓⇓ The observed improvements in aortic and LV outflow tract volume thickness index, the surrogate for stoke volume, provide convincing data that long-term resynchronization improves forward output.
The myocardial performance index incorporates phases of active contraction and relaxation and has the advantage of nongeometric dependency, reproducibility, and significant correlation with invasive measures of peak (+)dP/dt and peak (−)dP/dt.20,25⇓ Unlike QRS duration at baseline, the degree of baseline MPI abnormality correlates with remodeling benefit to BVS. The component of MPI that improves during therapy, the isovolumic contraction time, correlates with (+)dP/dt. This suggests that the ICT reduction reflects improvement in the systolic response attributable to enhanced ventricular synchrony. Measurement of the ICT may be a useful tool to predict which patients are likely to obtain remodeling benefit. This should be studied and prospectively validated in a larger patient cohort.
In general, the degree of mitral regurgitation in study patients was not hemodynamically significant. Thus, it is not possible to comment on any interaction between BVS and improvement in the severity of mitral regurgitation.
Effects of BVS on Parameters of Diastolic Function
Right atrial pressure and the degree of tricuspid regurgitation were not abnormal at baseline, suggesting that medical therapy had been optimized in these patients. These parameters did not change with BVS, suggesting a neutral effect on right-sided hemodynamics. This is in marked contrast to well-described adverse effects of right ventricular pacing on these parameters in patients with heart failure.1
The deceleration time was markedly abnormal at baseline, consistent with the advanced functional class of this patient group. The deceleration time improved with BVS. The magnitude and direction of change in deceleration time was similar to the improvement noted in a recent trial of therapy with carvedilol.33
The E/A ratio and pulmonary vein systolic/diastolic flow ratio, also reflecting LV filling pressure, were not markedly abnormal at baseline and did not improve in the group as a whole. There was a trend for pulmonary venous flow ratio to become systolic dominant after 12 weeks of continuous BVS, although the change was not statistically significant. The change is reassuring, given that blunted systolic flow, like shortened mitral deceleration time, has been associated with adverse prognosis.34 Although a change in AV delay may have influenced E/A ratio, it should not have influenced the deceleration time.35
The echocardiographic data presented here are taken from a clinical trial terminated before complete enrollment. The insufficient patient enrollment prohibits concurrent analysis of the primary end points of the study. Neither the site of LV pacing nor AV delay were individually optimized, but data to date do not agree on how this is best accomplished. In addition, most patients in the study had nonischemic cardiomyopathy, possibly limiting the applicability of the results to those patients with ischemic cardiomyopathy and bundle-branch block.
Long-term BVS for cardiac resynchronization improves echocardiographic parameters of remodeling. This does not occur at the cost of heightened sympathetic stimulation. The degree of abnormality of the Doppler-derived myocardial performance index predicts a remodeling response with long-term resynchronization.
Support for echocardiographic analysis at the core laboratory was provided by Guidant, Corp, St Paul, Minn.
- ↵Hardarson T, Arnason A, Eliasson GJ, et al. Left bundle branch block: prevalence, incidence, follow-up and outcome. Eur Heart J. 1987; 8: 1075–1079.
- ↵Grines CL, Bashore TM, Boudoulas H, et al. Functional abnormalities in isolated left bundle branch block: the effect of interventricular asynchrony. Circulation. 1989; 79: 845–853.
- ↵Xiao HB, Brecker SG, Gibson DG. Effects of abnormal activation on the time course of left ventricular pressure pulse in dilated cardiomyopathy. Br Heart J. 1992; 68: 403–407.
- ↵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: 3053–3059.
- ↵Aaronson KD, Schwartz JS, Chen TM, et al. Development and prospective validation of a clinical index to predict survival in ambulatory patients referred for cardiac transplant evaluation. Circulation. 1997; 95: 2660–2667.
- ↵Blanc JJ, Etienne Y, Gilard M, et al. Evaluation of different pacing sites in patients with severe heart failure: results of an acute hemodynamic study. Circulation. 1997; 96: 3273–3277.
- ↵Kass DA, Chen CH, Curry C, et al. Improved left ventricular mechanics from acute VDD pacing in patients with dilated cardiomyopathy and ventricular conduction delay. Circulation. 1999; 99: 1567–1573.
- ↵Auricchio A, Stellbrink C, Block M, et al. Effect of pacing chamber and atrioventricular delay on acute systolic function of paced patients with congestive heart failure. Circulation. 1999; 99: 2993–3001.
- ↵Pozzoli M, Traversi E, Cioffi G, et al. Loading manipulations improve the prognostic value of Doppler evaluation of mitral flow in patients with chronic heart failure. Circulation. 1997; 5: 1222–1230.
- ↵Nishimura RA, Appleton CP, Redfield MM, et al. Noninvasive Doppler echocardiographic evaluation of left ventricular filling pressures in patients with cardiomyopathies: a simultaneous Doppler echocardiographic and cardiac catheterization study. J Am Coll Cardiol. 1996; 28: 1226–1233.
- ↵Cohn JN, Ferri R, Sharpe N, et al. Cardiac remodeling-concepts and clinical implications: a consensus paper from an international forum on cardiac remodeling. J Am Coll Cardiol. 2000; 86: 431–438.
- ↵Chatterjee K. Heart failure therapy in evolution. Circulation. 1996; 94: 2689–2693.
- ↵Francis GS. Neurohumoral activation and progression of heart failure hypothetical and clinical considerations. J Cardiovasc Pharmacol. 1998; 32 (suppl 1): S16–S21.
- ↵Mann DL, Kent RL, Parsons B, et al. Adrenergic effects on the biology of the adult mammalian cardiocyte. Circulation. 1992; 85: 790–804.
- ↵Francis GS, Cohn JN, Johnson G, et al. Plasma norepinephrine, plasma renin activity, and congestive heart failure. Circulation. 1993; 87 (suppl 6): 40–48.
- ↵Greenberg B, Quinones MA, Koilpillai C, et al. Effects of long-term enalapril therapy on cardiac structure and function in patients with left ventricular dysfunction. Circulation. 1995; 91: 2573–2581.