Left ventricular function in atrial fibrillation during overdrive pacing

Left ventricular function in atrial fibrillation during overdrive pacing Viren Vankawala, MD, Bashar Markabawi, MD, Sara Dainiak, MD, Morgan Werner, MD, W. David Hager, MD, William Farrel, MD, Arnold M. Katz, MD, and David I. Silverman, MD Farmington, Conn

Objective Our purpose was to measure the effect of ventricular pacing in patients with atrial fibrillation (AF) on stroke volume and cardiac output. Background Unceasing variation in cycle length in AF decreases stroke volume and cardiac output. Because ventricular-inhibited pacing after atrioventricular node ablation has been reported to improve left ventricular performance, we tested the hypothesis that overdrive pacing would produce a similar benefit by regularizing cycle length. Methods and Results We studied 18 patients with chronic AF and permanent pacemakers. The aortic time velocity integral (TVI) was measured with continuous-wave Doppler and was used as a surrogate measure of stroke volume (stroke volume ⫽ TVI ⫻ aortic valve area, and aortic valve area is constant whether in AF or during pacing). For each patient, the linear relation between preceding cycle length and TVI in AF was used to estimate relative stroke volume (TVI compared within each patient) at a preceding cycle length of 666 ms in AF, and a similar comparison between AF and pacing was made at the minimum allowable pacing rate. Relative stroke volume in AF was then compared with relative stroke volume at both the fixed cycle (666 ms) and the minimum allowable rate. During pacing at 666 ms, relative stroke volume increased significantly by 18% (t ⫽ 2.8, P ⫽ .048), but there was no difference in cardiac output during pacing at the minimum possible rate and the corresponding preceding cycle length in AF. Conclusion Our data suggest that regularization of ventricular rhythm by overdrive pacing in patients with AF only improves stroke volume (and by extension, cardiac output) at pacing rates at the outer limit of and above the normal physiologic range. (Am Heart J 2002;143:827-32.)

Optimal management for patients with atrial fibrillation (AF) remains a major challenge. Although many patients tolerate prolonged AF without apparent difficulty, others remain symptomatic despite adequate rate control, even when ventricular function is normal. Even though AF is associated with decreased survival,1,2 the utility of long-term suppression of AF compared with control of ventricular response remains controversial.3 Regardless of benefit, maintaining sinus rhythm once AF has occurred remains a formidable challenge. The efficacy4-6 and safety7-11 of different antiarrhythmic agents vary widely, and their effect on survival continues to be the subject of ongoing investigation.12 Cardiac pacing to regularize cardiac rhythm may alleviate some of the deleterious effects of AF. Irregular From the Cardiology Division, Department of Medicine, University of Connecticut School of Medicine, Farmington, Conn. Supported in part by National Institutes of Health General Clinical Research Center Grant No. MO1RR06192, and a grant from the Claude Pepper Center for Older Americans, No. 5P60-AG13631– 03 (D.I.S.). Submitted February 13, 2001; accepted October 10, 2001. Reprint requests: David I. Silverman, MD, Cardiology, University of Connecticut Health Center, 263 Farmington Ave, Farmington, CT 06030 –1305. E-mail: [email protected] © 2002 Mosby, Inc. All rights reserved. 0002-8703/2002/$35.00 ⫹ 0 4/1/121272 doi:10.1067/mhj.2002.121272

ventricular pacing at mean heart rates that are similar to sinus rhythm decreases cardiac output,13 whereas both cardiac output and symptoms improve after atrioventricular (AV) nodal ablation followed by fixed-rate pacing at a cycle length equivalent to the mean preablation RR interval.14-16 AV nodal ablation, however, is invasive, expensive, and potentially hazardous and may not have broad applicability to the majority of patients with AF. Because the intraventricular conduction delay caused by right ventricular pacing reduces ejection fraction,17-20 any increased contractility resulting from the regularization of the ventricular response might be outweighed by the decreased cardiac performance caused by abnormal ventricular activation. To test the hypothesis that overdrive pacing might provide a simpler and less expensive alternative to ablation plus pacemaker insertion, we used an established noninvasive method to measure the response to ventricular overdrive pacing in patients with chronic AF and compared stroke volume during AF and ventricular pacing at identical cycle lengths.

Methods Study protocol Eighteen patients with previously placed permanent ventricular pacemakers and documented chronic AF were recruited from the John Dempsey and Grove Hill Cardiology

American Heart Journal May 2002

828 Vankawala et al

Figure 1

Figure 3

The linear relation between preceding cycle length and TVI for a typical patient, as well as the increase in TVI seen at identical cycle lengths during pacing.

Figure 2 The change in TVI at a preceding cycle length of 666 ms in AF compared with TVI during ventricular pacing at the same cycle length.

that would sustain at least 6 consecutive beats without AF present, and pacing rate was increased by 10-beat/min intervals for at least 30 seconds until a maximum heart rate of 130 beats/min was obtained. Measurements at each pacing cycle were made after at least 6 beats had been recorded and new steady state achieved. Doppler measurements were recorded for a mean of 30 beats at each pacing cycle length.

Data analysis The change in TVI during pacing at decreasing cycle lengths. TVI falls or remains unchanged with decreasing cycle length for almost all patients, although the overall decrease is not significant. The numbers at the top of the graph indicate mean TVI at each cycle length.

pacemaker services. Patients with left bundle branch block at baseline were excluded from the study. All pacemaker leads were placed within the right ventricular apex. Standard 2dimensional echocardiography was performed with a HewlettPackard Sonos 2500 echocardiograph (Palo Alto, Calif). Left atrial size was estimated with standard techniques.21 Because ejection fraction varies dramatically from beat to beat during AF,22 left ventricular ejection fraction was graded as normal, moderately impaired, or severely impaired, with use of both 4- and 2-chamber views by a consensus panel of 3 experienced echocardiographers. Aortic time velocity integral (TVI) was measured with 2-dimensionally guided continuous-wave Doppler flow measurement from the apical window. All primary TVI measurements were taken by the same investigator (S.D.) and checked for accuracy and reproducibility by the study’s senior investigator. A mean of 40 beats was recorded in AF. Patients were then paced at the longest cycle length

The linear relation between preceding cycle length and TVI in AF23 (Figure 1) was used to estimate TVI at a preceding cycle length of 666 ms in AF (the maximum cycle length necessary for overdrive pacing in all of the patients). The change in TVI at shorter cycle lengths for each patient was then examined (Figure 2). TVI in AF was then compared with TVI during pacing at the same cycle length (Figure 3). A second comparison was made between the maximum allowable cycle length for pacing (minimum rate) for each patient and that preceding cycle length in AF (Figure 4). Relative cardiac output in AF was calculated by summing TVI for 60 seconds of continuous beats and during pacing by multiplying TVI by pacing rate. All data are reported as mean ⫾ SD. The significance of the linear regression between preceding cycle length and stroke volume was confirmed for each patient with standard software (SPSS 9.0, SPSS Inc, Chicago, Ill). Comparison of TVI at decreasing cycle lengths was made with analysis of variance for repeated measures. Comparisons of TVI between AF and pacing at identical cycle lengths were made with use of a paired t test.

Results The clinical characteristics and baseline echocardiographic measurements for each patient are presented

American Heart Journal Volume 143, Number 5

in Table I. Mean age was 78 ⫾ 12 years. Etiology of AF followed a typical distribution, with a majority of patients having hypertension. Most patients had at least some degree of left ventricular dysfunction. Figure 1 demonstrates a typical linear relation between preceding cycle length and TVI (y ⫽ 0.024x ⫹ 67, r ⫽ 0.48, t ⫽ 4.15, P ⬍ .0001); this relation was statistically significant in all 18 patients. The figure also shows TVI at all permissible pacing cycle lengths (mean of 6 beats) and demonstrates that for this patient in AF, as for most patients, TVI during overdrive pacing is greater than that predicted by the regression line at similar cycle lengths. The change in TVI from any cycle length to the next was not significant (F ⫽ 15.47 for analysis of variance for repeated measures, P ⫽ .97; Figure 2). There was also no significant difference in the change in TVI as cycle length shortened when patients with normal baseline LV function were compared with those with decreased LV function, although patients with normal baseline function showed slightly greater increases in TVI at longer cycle lengths. Relative cardiac output during pacing increased significantly with decreasing cycle length from 750 ms to 666 ms (from 1853 ⫾ 816 cm2/min to 2070 ⫾ 929 cm2/min, P ⫽ .012) but remained constant thereafter (2159 ⫾ 666 cm2/min at 600 ms, 2145 ⫾ 927 cm2/min at 545 ms, and 2241 ⫾ 1549 cm2/min at 500 ms). TVI during AF at a preceding cycle length of 666 ms was then compared with TVI during pacing at the same cycle length (Figure 3). Mean TVI was significantly greater during pacing than during AF, increasing from 21.0 ⫾ 9.7 cm2 to 25.0 ⫾ 13.5 cm2 (t ⫽ 2.19, P ⫽ .042). Relative cardiac output during pacing at 666 ms was significantly greater than cardiac output in AF: 1595 ⫾ 681 cm2/min compared with 2249 ⫾ 1211 cm2/min (t ⫽ 3.73, P ⫽ .008). At maximum allowable pacing cycle length for each patient, the difference in TVI between pacing and AF declined and was no longer significant (from 22.3 ⫾ 9.8 cm2 in AF to 25.2 ⫾ 13.8 cm2during pacing; t ⫽ 1.4, P ⫽ .17; Figure 4); therefore the difference between relative cardiac output in AF and at maximum possible cycle length was also not significant (1595 ⫾ 681 cm2/min in AF vs 1730 ⫾ 886 cm2/min at maximum pacing cycle; t ⫽ 1.4, P ⫽ .18).

Discussion We observed 3 differences between cardiac performance during AF and overdrive pacing. First, stroke volume, as estimated by TVI, is greater during cardiac pacing than AF at the same cycle length (Figure 1). Second, the difference between TVI during AF and pacing narrows as cycle length increases. Third, decreasing cycle length during pacing does not increase

Vankawala et al 829

Figure 4

The change in TVI between AF at a preceding cycle length identical to the maximum allowable cycle length for ventricular pacing (mpc [maximum pacing cyle]).

TVI, regardless of baseline left ventricular function. These data agree with previous studies that report that cardiac output is increased during overdrive pacing of AF, despite the decreased stroke volume and reduced ejection fraction associated with diminished ventricular synchrony.17-20 Our data suggest, however, that such an increase is modest. The increased TVI during regular pacing, compared with that at the same cycle length during AF (Figure 1), can be explained by different cycle length– dependent alterations in the availability of intracellular Ca2⫹ for release during excitation-contraction coupling. These changes influence the direction of the forceinterval relationship.22,23 A positive force-interval relationship, in which contractility increases with decreasing cycle length, is seen when the rhythm is regular because plasma membrane Ca2⫹ channels open more frequently at faster rates; the result is an increased Ca2⫹ entry that, by augmenting intracellular Ca2⫹ stores, increases contractility. In AF, in which cycle length changes incessantly, the ability of short cycles to reduce contractility and that of long cycles to increase contractility are a manifestation of the negative force-interval relationship.22 The mechanism by which single short cycles decrease contractility is not clear; this could reflect incomplete recovery (refractoriness) of sarcoplasmic reticulum Ca2⫹ release channels after a short cycle or “trapping” of Ca2⫹ within intracellular membranes when this activator does not have enough time to diffuse from uptake sites in the sarcotubular network to release sites in the subsarcolemmal cister-

American Heart Journal May 2002

830 Vankawala et al

Table I. Patient characteristics Patient 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18

Age

Sex

74 74 82 72 90 90 75 85 82 89 76 73 40 83 73 95 74 76

F M M M M M M M F M M M M F F F M F

AF etiology

Medications

HTN HTN HTN, HTN HTN HTN, HTN, HTN HTN HTN, HTN HTN, HTN HTN HTN HTN HTN, HTN,

D A,C,D A,B,D C,ARB A,D A,D,G B A,C A,D,G A,B,D

CAD

CAD CAD

CAD CAD

CAD CAD

A,B,C,D C,D C,D B,G A,C,D A,D,G B,G

LA size (cm)

LV Fn

4.0 4.5 5.5 4.6 4.0 4.0 3.9 4.0 4.6 4.5 4.5 4.3 4.4 4.1 3.4 5.0 5.4 5.5

22 22 22 22 2 22 22 Normal 22 Normal Normal Normal 2 2 Normal 2 Normal Normal

LA, Left atrium; LV, left ventricular; Fn, function; M, male; F, female; HTN, hypertension; CAD, coronary artery disease; A, angiotensin-converting enzyme; B, ␤-blocker; C, calcium-channel blocker; D, diuretic; G, digoxin; ARB, angiotensin-receptor blocker; 2, mild to moderately reduced (ejection fraction 35%–50%); 22, severely reduced (ejection fraction ⬍ 35%).

nae. Whatever the mechanism, the increased contractility associated with each of a regular series of shortpaced cycles, compared with that during a single short cycle during AF, augments stroke volume (Figure 1). This benefit is attenuated at longer cycle lengths (Figure 4) because the net gain in cytoplasmic Ca2⫹ that accompanies the regular rhythm is modified by the negative inotropic effect of asynchronous activation caused when the ventricles are depolarized by stimulation of the right ventricular apex.19,20 At sustained shorter cycle lengths, AF might more closely mimic pacing due to loss of beat-to-beat variability, but such conditions would seldom occur in patients receiving adequate rate control. The decline in TVI at higher pacing rates is similar to data reported for patients in sinus rhythm and who have cardiomyopathy undergoing ventricular pacing.24,25 A previous study has reported increased dp/ dtmax and cardiac output at moderate pacing rates (80110 beats/min) in patients with normal LV function but not in patients with cardiomyopathy.25 In that previous study, contractility and cardiac output reached a plateau at higher rates (similar to those used in our study) and then fell with further rate increases in both groups.25 Taken together, these data confirm that in the living human heart, an increase in the intervalforce relation is restricted largely to physiologic heart rates. Pacing from a variety of sites, with or without AV nodal ablation, has been used for both prevention and treatment of both chronic and paroxysmal AF.26-29 In sinus rhythm, neither AV nor ventricular pacing of patients in sinus rhythm has been shown to decrease

the incidence of subsequent AF in high-risk patients,30 but dual-chamber pacing after AV nodal ablation has been reported to reduce symptoms associated with paroxysmal AF, to increase cardiac output, and to improve quality of life measures.26-28,30-32 The addition of mode switching appears to improve symptoms further33 but is paradoxically associated with increased likelihood of recurrent AF.34,35 By contrast, biatrial pacing, pacing from selected sites, and selected pacing algorithms appear to be promising in terms of AF suppression in a variety of settings.36-38 The study design produced several limitations. First, the presence of various AV nodal blocking agents reduced contractility at baseline, as well as during pacing. However, because each patient was compared with only himself or herself, the effect of baseline medications for rate control should be neutral. Calcium channel blockers and ␤-blockers are unlikely to produce significantly greater effects on cardiac contractility or stroke volume in AF as compared with pacing. Second, the inability to pace at longer cycle lengths limited the comparison between AF and pacing to a relatively narrow range of rates in this study. For this reason, the increase in TVI at shorter pacing cycle lengths does not exclude the possibility that regularization through pacing at longer, more physiologic cycles might increase stroke volume and simply eliminate the negative inotropic effect of short cycles. The modest improvement seen with pacing in this group of patients, most of whom had moderate-to-severe LV systolic dysfunction, raises the possibility that any improvement with pacing may be limited by underlying

American Heart Journal Volume 143, Number 5

systolic function. Third, even though our data (and that of other authors) demonstrate improved cardiac output at supraphysiologic pacing cycle lengths, prolonged rapid pacing cannot be recommended as a clinical therapy because of its association with tachycardia-induced cardiomyopathy and the possibility of inducing dangerous arrhythmias. Although pacing in the setting of simultaneous heart failure and left bundle branch block improves shortterm cardiac function,39 neither long-term results nor pacing in the absence of a preexistent intraventricular conduction delay has been adequately studied. Pacing after pharmacologic slowing of AV conduction with ␤-adrenergic receptor antagonists might produce similar results and therefore provide a more practical approach. Indeed, long-term evaluation of such an approach represents the next logical step in this research.

References 1. Middlekauff HR, Stevenson WG, Stevenson LW. Prognostic significance of atrial fibrillation in congestive heart failure. Circulation 1991;84:40-8. 2. Wolf PA, Mitchell JB, Baker CS, et al. Impact of atrial fibrillation on mortality, stroke, and medical costs. Arch Intern Med 1998; 158:229-34. 3. Hohnloser SH, Kuck KH, Lilienthal J. Rhythm or rate control in atrial fibrillation: Pharmacological Intervention in Atrial Fibrillation (PIAF): a randomised trial. Lancet 2000;356:1789-94. 4. Jung F, DiMarco P. Antiarrhythmic drug therapy in the treatment of atrial fibrillation. Cardiol Clin 1996;14:507-20. 5. Juul-Mo ¨ ller S, Edvardsson N, Rehnqvist-Ahlberg N. Sotalol versus quinidine for the maintenance of sinus rhythm after direct current conversion of atrial fibrillation. Circulation 1990;82:1932-9. 6. Reimold SC, Cantillon CO, Friedman PL, et al. Propafenone versus sotalol for suppression of recurrent symptomatic atrial fibrillation. Am J Cardiol 1993;71:558-63. 7. Nathan AW, Hellestrand KJ, Bexton RS, et al. Proarrhythmic effects of the new antiarrhythmic flecainide acetate. Am Heart J 1984;107:222-8. 8. Coplen SE, Antman EM, Berlin JA, et al. Efficacy and safety of quinidine therapy for maintenance of sinus rhythm after cardioversion: a meta-analysis of randomized controlled trials. Circulation 1990;82:1106-16. 9. Flaker GC, Blackshear JL, McBride R, et al. Antiarrhythmic drug therapy and cardiac mortality in atrial fibrillation. The Stroke Prevention in Atrial fibrillation Investigators. J Am Coll Cardiol 1992; 20:527-32. 10. DePaola AA, Horowitz LN, Morganroth J, et al. Influence of left ventricular dysfunction on flecainide therapy. J Am Coll Cardiol 1987;91:163-8. 11. Ruskin J. The cardiac arrhythmia suppression trial (CAST). N Engl J Med 1989;321:386-7. 12. The Planning and Steering Committees of the AFFIRM study for the NHLBI AFFIRM investigators. Atrial fibrillation follow-up investigation of rhythm management: the AFFIRM study design. Am J Cardiol 1997;79:1198-202. 13. Clark DM, Plumb VJ, Epstein AE, et al. Hemodynamic effects of an irregular sequence of ventricular cycle lengths during atrial fibrillation, J Am Coll Cardiol 1997;30:1039-45.

Vankawala et al 831

14. Edner M, Caidahl K, Bergfeldt L, et al. Prospective study of left ventricular function after radiofrequency ablation of atrioventricular junction in patients with atrial fibrillation. Br Heart J 1995;74: 261-7. 15. Morady F, Hasse C, Strickberger SA, et al. Long-term follow-up after radiofrequency modification of the atrioventricular node in patients with atrial fibrillation. J Am Coll Cardiol 1997;29:11321. 16. Natale A, Kent V, Zimerman L, et al. Impact on ventricular function and quality of life of transcatheter ablation of the atrioventricular junction in chronic atrial fibrillation with a normal ventricular response. Am J Cardiol 1996;78:1431-3. 17. Leclercq C, Gras D, Le Helloco A, et al. Hemodynamic importance of preserving the normal sequence of ventricular activation in permanent cardiac pacing. Am Heart J 1995;129:1133-41. 18. Huikuri HV, Peuhkurinen KJ, Takkunen JT. Relationship between beat-to-beat changes in hemodynamic state and action potential duration of the left ventricle during rapid ventricular pacing in man. PACE 1992;15:878-85. 19. Boucher CA, Pohost GM, Okada RD, et al. Effect of ventricular pacing on left ventricular function assessed by radionuclide angiography. Am Heart J 1983;106:1105-11. 20. DiCarlo LA Jr, Morady F, Krol RB, et al. The hemodynamic effects of ventricular pacing with and without atrioventricular synchrony in patients with normal and diminished left ventricular function. Am Heart J 1987;114:746-52. 21. Sahn DJ, DeMaria A, Kisslo J, et al. Recommendations regarding quantification in M-mode echocardiography: results of a survey of echocardiographic measurements. Circulation 1978;58:1072-83. 22. Raymond RJ, Lee AJ, Messineo FC, et al. Cardiac performance early after cardioversion from atrial fibrillation. Am Heart J 1998; 136:435-42. 23. Schneider F, Martin DT, Schick EC, et al. Interval-dependent changes in left ventricular contractile state in lone atrial fibrillation and in atrial fibrillation associated with coronary artery disease. Am J Cardiol 1997;80:586-90. 24. Bhargava V, Shabetai R, Mathiasen RA, et al. Loss of adrenergic control of the force-frequency relation in heart failure secondary to idiopathic or ischemic cardiomyopathy. Am J Cardiol 1998:81: 1130-7. 25. Hasenfuss G, Holubarsch C, Hermann HP, et al. Influence of the force-frequency relationship on haemodynamics and left ventricular function in patients with non-failing hearts and in patients with dilated cardiomyopathy. Eur Heart J 1994;15:164-70. 26. Proclemer A, Della Bella P, Tondo C, et al. Radiofrequency ablation of atrioventricular junction and pacemaker implantation versus modulation of atrioventricular conduction in drug refractory atrial fibrillation. Am J Cardiol 1999;83:1437-42. 27. Brignole M, Gianfranchi L, Menozzi C, et al. Assessment of atrioventricular junction ablation and DDDR mode-switching pacemaker versus pharmacological treatment in patients with severely symptomatic paroxysmal atrial fibrillation: a randomized controlled study. Circulation 1997;96:2617-24. 28. Natale A, Zimerman L, Tomassoni G, et al. Impact on ventricular function and quality of life of transcatheter ablation of the atrioventricular junction in chronic atrial fibrillation with a normal ventricular response. Am J Cardiol 1996;78:1431-3. 29. Levy T, Walker S, Rex S, et al. Ablate and pace for drug refractory paroxysmal atrial fibrillation. Is ablation necessary? Int J Cardiol 2000;75:187-95. 30. Gillis AM, Connolly SJ, Lacombe P, et al. Randomized crossover

832 Vankawala et al

31.

32.

33.

34.

comparison of DDDR versus VDD pacing after atrioventricular junction ablation for prevention of atrial fibrillation. The atrial pacing peri-ablation for paroxysmal atrial fibrillation (PA (3)) study investigators. Circulation 2000;102:736-41. Edner M, Caidahl K, Bergfeldt L, et al. Prospective study of left ventricular function after radiofrequency ablation of atrioventricular junction in patients with atrial fibrillation. Br Heart J 1995;74: 261-7. Morady F, Hasse C, Strickberger SA, et al. Long-term follow-up after radiofrequency modification of the atrioventricular node in patients with atrial fibrillation. J Am Coll Cardiol 1997;29:113-21. Schuchert A, van Langen H, Michels K, et al. DDD(R) pacing with automatic mode switch in patients with paroxysmal atrial fibrillation following AV nodal ablation. Cardiology 1997;88:323-7. Marshall HJ, Harris ZI, Griffith MJ, et al. Prospective randomized study of ablation and pacing versus medical therapy for paroxysmal atrial fibrillation: effects of pacing mode and mode-switch algorithm. Circulation 1999;99:1587-92.

American Heart Journal May 2002

35. Kay GN, Bubien RS, Epstein AE, et al. Effect of catheter ablation of the atrioventricular junction on quality of life and exercise tolerance in paroxysmal atrial fibrillation. Am J Cardiol 1988;62: 741-4. 36. Israel CW, Lawo T, Lemke B, et al. Atrial pacing in the prevention of paroxysmal atrial fibrillation: first results of a new combined algorithm. Pacing Clin Electrophysiol 2000;11:1888-90. 37. Ramdat Misier AR, Beukema WP, Oude Luttikhuis HA, et al. Multisite atrial pacing: an option for atrial fibrillation prevention? Preliminary results of the Dutch dual-site right atrial pacing for prevention of atrial fibrillation study. Am J Cardiol 2000;86(9 Suppl 1):K20-4. 38. Gillis AM. Pacing to prevent atrial fibrillation. Cardiol Clin 2000; 18:25-36. 39. 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 bundlebranch block. Circulation 2000;102:3053-9.