Author + information
- Angel R Leon, MD, FACC*,* (, )
- Jeffrey M Greenberg, MD*,
- Narendra Kanuru, MD*,
- Cindy M Baker, MD*,
- Fernando V Mera, MD, FACC*,
- Andrew L Smith, MD, FACC*,
- Jonathan J Langberg, MD, FACC* and
- David B DeLurgio, MD, FACC*
- ↵*Reprint requests and correspondence:
Dr. Angel R. Leon, Carlyle Fraser Heart Center, Crawford Long Hospital of Emory University, 550 Peachtree Street, NE, Atlanta, Georgia 30365, USA.
Many patients with severe congestive heart failure (CHF) develop chronic atrial fibrillation (AF). The onset of AF eliminates regular atrioventricular (AV) transport, worsening cardiac performance in these patients with ventricular dysfunction. Radiofrequency ablation of the AV junction and right ventricular (RV) pacing improves ventricular function, symptoms and quality of life in patients with AF refractory to drug therapy for ventricular rate control (1,2). However, some patients may have persistent or progressive CHF symptoms after junction ablation and conventional RV apical pacing. The abnormal left ventricular (LV) activation sequence produced by cardiac pacing from the RV apex may contribute to the lack of improvement observed in some patients after ablate-pace therapy.
Right ventricular apical pacing produces a LV conduction pattern similar to a left bundle branch block (LBBB), with a posterior and superior initial and total QRS vector with a prolonged QRS duration. Consequences of RV pacing include paradoxical septal motion and reduction in ejection fraction with a detrimental effect on systolic pressure and cardiac output (3–5). Chronic RV apical pacing produces geometric changes that appear similar on echocardiography to changes associated with intrinsic LBBB. Because RV apical pacing induces the same vector electrocardiographic patterns and functional abnormalities present during intrinsic LV conduction delay, apical pacing may worsen symptoms and functional status when used in the patient with pre-existing ventricular dysfunction.
Cardiac resynchronization with biventricular pacing (BVP) reverses the mechanical abnormalities associated with severe ventricular conduction delay and improves patient symptoms, functional capacity, exercise time and quality of life (6,7). The mechanisms by which BVP improves symptoms in CHF include normalization of ventricular wall motion, optimization of AV filling and reduction of mitral regurgitation in the patient with LV dysfunction and conduction delay with a QRS interval exceeding 130 ms (8–14). Published clinical trials of BVP exclude patients with chronic RV pacing or AF. Cardiac resynchronization for ventricular dysfunction and conduction delay should theoretically extend to the patient with chronic AF. However, to our knowledge, no study to date has described the impact of the therapy in these patients.
Testing the effectiveness of BVP in patients with AF and chronic RV pacing after AV junction ablation provides an opportunity to determine whether cardiac resynchronization improves symptoms of CHF and ventricular function independent of any atrial contribution to cardiac output. This investigation evaluated whether BVP improves ventricular function, functional status, quality of life and decreases hospitalization in patients with CHF and chronic AF.
The authors evaluated 20 patients at Crawford Long Hospital and Emory Hospital of Emory University. The group includes consecutive patients with severe CHF, permanent AF, prior AV junction ablation and RV pacing for at least six months prior to enrollment. Inclusion criteria identified subjects with LV dysfunction and ejection fraction ≤0.35, CHF with severe functional impairment of New York Heart Association (NYHA) functional class III or IV and CHF symptoms refractory to standard medical therapy. Baseline patient evaluation included a history and physical examination, 12-lead electrocardiogram, and an assessment of LV function by echocardiography. We identified the frequency of unscheduled hospitalization for worsening CHF during the year prior to and after enrollment. Ambulatory visits were not included. Patients also completed the Minnesota Living with Heart Failure Quality of Life Survey (15). Table 1displays the demographic and clinical characteristics at entry.
All subjects underwent a procedure to revise the RV pacing system to BVP in the cardiac electrophysiology laboratory. A conscious sedation protocol approved by Emory University provides patient comfort, and local infiltration of lidocaine (1%) provided anesthesia during the procedure. The surgical implant procedure involved incision into and dissection of the chronically implanted pacemaker pocket to expose the generator and lead. We described in detail the technical aspects of the upgrade procedure in a previous publication (16).
The authors predetermined a preference for a lateral, anterolateral or posterolateral LV vein as a target for the final pacing lead position to separate the RV and LV electrodes as much as possible. Introduction of a commercially available, bipolar (Medtronic 4024, Minneapolis, Minnesota) or unipolar (Medtronic 4023) (65 or 85 cm length) pacing lead through the sheath and into the desired vein provided LV free-wall stimulation. We sequentially connected the RV and LV leads to a bipolar Y-adapter (Medtronic 2872) and inserted the Y-adapter into a single chamber rate-responsive pulse generator. We programmed the single chamber generator with the Y-adapter to the VVIR mode. We replaced any chronically implanted pulse generator estimated to have less than half of its battery service life remaining to prevent early pulse generator depletion by potentially increased current loads from parallel connection of two ventricular leads into the Y-adapter.
Clinical follow-up and end points
Predischarge pacemaker interrogation and 12-lead electrocardiogram confirmed stability of RV and LV pacing thresholds. Initial follow-up evaluations at three to six months included repeat echocardiography, device interrogation with threshold checks, quality of life survey, physical examination and chest radiography. Return visits to the pacemaker clinic were performed at six-month intervals thereafter. End points for analysis included echocardiographic ventricular ejection fraction and dimensions, NYHA functional class, quality of life scores on the University of Minnesota Living with Heart Failure Quality of Life survey and frequency of hospitalization prior to and after pacemaker revision to BVP.
Numerical results are expressed in the text as the mean ± 1 SD. The differences between the continuous variables before and after BVP were evaluated using a paired ttest. Changes in NYHA functional class as a result of the procedure were assessed using contingency table analysis. A p value ≤0.05 was used to indicate statistical significance.
The procedure successfully revised the chronic RV pacing system to BVP in all 20 patients. All 20 patients were alive at the follow-up evaluation and are included in the analysis. All patients completed follow-up evaluation between three to six months after implant. Left ventricular capture threshold was 1.9 ± 1.0 V, with a lead impedance of 694.8 ± 144.6 ohms. Final LV lead position included 4 electrodes in the anterolateral vein, 6 in the lateral vein and 10 in the posterolateral vein. We observed no procedural complications.
Of the 20 patients, 17 remain alive to date, during a mean follow-up period of 17.3 ± 4.5 months. One patient died 18 months after pacemaker revision from bacterial peritonitis complicating long-term peritoneal dialysis. Two other patients died over a year after implantation of worsening CHF. Another patient developed ventricular tachycardia three months after the revision requiring removal of the pacing system and insertion of a transvenous defibrillator with BVP capability.
The NYHA functional classification improved from 3.4 ± 0.5 to 2.4 ± 0.6 (chi square = 20.4, p < 0.001) (Fig. 1). The number of hospitalizations required one year prior to the procedure decreased from an average of 1.9 ± 0.8 to 0.4 ± 0.6 at one year postprocedure (p < 0.001) (Fig. 2). The scores on the Minnesota Living with Heart Failure Quality of Life survey improved from 78 ± 24 to 52 ± 23 (p < 0.01).
Electrocardiographic and echocardiographic outcome
The QRS duration decreased from 213 ± 40 ms to 172 ± 31 ms (p < 0.0001). Repeat echocardiography at three to six months after the upgrade procedure showed a mean LV ejection fraction of 30.9 ± 11.5%, an increase from a baseline of 21.5 ± 6.9% (p < 0.001) (Fig. 3). The mean LV diastolic diameter decreased from 67.9 ± 8.3 mm to 63.5 ± 7.7 mm (p < 0.003) and the end-systolic diameter from 56.3 ± 9.8 mm to 51.5 ± 10.9 mm (p < 0.01).
Conversion to BVP provides cardiac resynchronization to the patients in this study with chronic RV pacing for AF. Our analysis provides evidence that in the patient with CHF, AF and chronic RV apical pacing, conversion to BVP reverses asynchrony, improves ventricular performance and dimensions, quality of life and symptoms of CHF in the same manner as described in patients with sinus rhythm and LBBB who undergo BVP. We detected significant improvement in functional class, increased ejection fraction, a decrease in end-systolic and end-diastolic diameters, a decrease in the number of hospitalizations and improved quality of life scores in this population with severe CHF and chronic RV pacing.
Cause of benefit of BVP in patients with chronic AF and RV pacing
Assessing the benefit of BVP in the patient with AF and complete AV block after junction ablation provides a relevant model to examine the isolated effect of BVP on ventricular function. Absence of any active atrial transport permits an assessment of the isolated impact of resynchronization therapy on cardiac performance independent of any AV timing considerations. Atrioventricular block produced by catheter ablation ensures that all ventricular activation occurs through the pacing leads, without any possible fusion that might contaminate an assessment of the pacing therapy. Obtaining the echocardiographic baseline LV and clinical data prior to upgrade but chronically after junction ablation eliminates the possibility that regularizing ventricular rate and filling interval acutely contributed to the observed improvement in ejection fraction after upgrade to BVP.
Several studies have demonstrated improved LV systolic function in patients with impaired LV function after AV node ablation and RV pacing (17–19). Some investigators have suggested that atrial synchronized BVP with echocardiographic and Doppler-guided programming of the AV delay optimizes left atrial to LV flow (20,21)and improves ventricular output. They attribute a large portion of the benefit of BVP to the impact of optimizing atrial contribution to cardiac output (20,21). Because the patients in our study underwent catheter ablation long before upgrade to BVP, one cannot attribute the improvement in clinical and echocardiographic parameters simply to regularization of the ventricular response to AF, as reported often after junction ablation and standard RV apical pacing (17–19).
Comparison to other trials
The use of angiotensin-converting enzyme inhibitors or receptor blockers in this study approaches the use in previously reported trials of BVP. The use of beta-blockers in our patients exceeds the use of the drug in early studies such as the Medtronic Inc. InSync study (6)and approaches the use in the Multisite Stimulation in Cardiomyopathy (MUSTIC) (7)study. This study population includes a larger proportion (40%) of patients with NYHA functional class IV symptoms than the other published trials of BVP. Including a larger proportion of patients with severe CHF may increase the probability of detecting a treatment effect from BVP. The other clinical characteristics of the group studied compare favorably with the patients described in previous trials of BVP, except for the presence of AF.
This study evaluates the impact of the upgrade procedure on the same end points of symptomatic improvement and ventricular performance reported in Pacing Therapies for CHF (PATH-CHF) (8), Medtronic Inc. InSync (6), MUSTIC (7)and Multicenter InSync Randomized Clinical Evaluation (MIRACLE) (9). The results find similar improvement in functional status, quality of life scores, echocardiographic assessment of ventricular dimensions and ejection fraction.
Published clinical trials of BVP excluded patients with AF and chronic RV pacing which prevents any objective assessment of the benefit of BVP to the patient without active atrial transport. Therefore, until now, we could not extrapolate the benefit of cardiac resynchronization in CHF to the large patient population with CHF and chronic RV pacing for AF. Two published studies up to now have looked at BVP in patients with AF (22,23). In both of these studies, the AV node was ablated at the time of BVP implantation. All the patients in our study underwent ablation at least six months before BVP implantation, with a mean duration of over two years. Therefore, our results cannot be attributed to the acute beneficial effects of pacing in chronic AF due to the optimization of left atrial to LV filling or to heart rate control (17–19). Other clinical trials will further examine the impact of BVP in the patient with AF. The results from the MUSTIC AF subgroup are preliminary (23). The ongoing Post AV Node Ablation Evaluation (PAVE) (24)study also addresses the role of BVP and LV pacing in improving ventricular function in the patient with AF.
We could not utilize a crossover analysis of BVP similar to our comparison of RV apical pacing to septal pacing (25)in patients with AF after junction ablation. The patients included in the present study underwent chronic RV pacing prior to enrollment that might have produced ventricular structural and functional changes that might not quickly reverse during a short crossover interval. Another possible model utilizing a dual-chamber pulse generator with the LV lead in the atrial port and the RV lead in the ventricular port with random programming to AAIR, DDDR and VVIR modes would permit an elegant comparison of RV, LV and BVP in the same patient. The limited long-term follow-up available on LV pacing discouraged our use of that design. The PAVE study will compare results of the three pacing sites in a larger patient population.
Another limitation of the study is that patients were compared before and after upgrade of the device. The Minnesota Living with Heart Failure questionnaire introduced by Rector et al. (15)is commonly used for the assessment of patients with heart failure. Although its clinical value has been established (26), we cannot rule out the possibility of the placebo effect since there was no control group in our study. In other studies of BVP, quality of life values have shown some improvement in patients who received control therapy (9). Thus, quality of life may show improvement even when no active therapy is delivered.
We conclude that BVP improves symptoms of CHF by improving ventricular function in the absence of AV transport. One cannot attribute the overall benefit of BVP simply to optimization of left atrial to LV filling. One should consider BVP in the patient with AF, LV dysfunction and severe CHF. One should not exclude patients with AF as candidates for BVP when AV conduction can be controlled with medications or junctional ablation.
☆ Dr. Greenberg is a research fellow in cardiac electrophysiology at the Division of Cardiology, Emory University School of Medicine. A grant from the Medtronic Corporation to the Carlyle Fraser Heart Center helps provide economic support for the fellowship.
- American College of Cardiology Foundation
- Wood M.A,
- Brown-Mahoney C,
- Kay G.N,
- et al.
- Lee M.A,
- Dae M.W,
- Langberg J.J,
- et al.
- Auricchio A,
- Stellbrink C,
- Block M,
- et al.
- Abraham W.T
- Kerwin W.F,
- Botnivick E.H,
- O’Connell J.W,
- et al.
- Varma C,
- Callaghan P.O,
- New C,
- et al.
- Liu I,
- Tockman B,
- Belalcazar A,
- et al.
- Etienne Y,
- Valls-Bertault V,
- Mansourati J,
- et al.
- Rector R.S,
- Kubo S.H,
- Cohn J.N
- Edner M,
- Caidahl K,
- Bergfelt L,
- Darpö B,
- Edvardsson N,
- Rosenqvist M
- Nishimura R.A,
- Hayes D.L,
- Holmes D.R,
- Tajik A.J
- Daubert J.-C,
- Linde C,
- Cazeau S,
- Kappenberger L,
- Sutton R,
- Bailleul C
- ↵Left Ventricular-Based Cardiac Stimulation Post AV Node Ablation Evaluation (PAVE) Clinical Investigation Plan (St. Jude Medical, Sylmar, CA) RevC, pp 1–18. November 2000.