Impact of atrioventricular delay on heart rate variability of paced patients with and without heart failure

Intwnational Journalof

CARDIOIBGY ELSEVIER

International

Journal of Cardiology

52 (1995) 235-239

Impact of atrioventricular delay on heart rate variability of paced patients with and without heart failure E. Simantirakis, E. Skalidis, F. Parthenakis, S. Chrysostomakis, E. Manios, G. Kochiadakis, P. Vardas* Cardiology Department, University Hospital of He&lion, Received 22 February

P.O. Box 1352. Heraklion,

Crete, Greece

1995; revision received 27 September 1995; accepted 27 September 1995

Abstract

The aim of the study was to investigate whether the optimisation of gtrioventricular (AV) delay in patients with complete AV block, with or without heart failure, paced under VDD mode, has an effect on heart rate variability and consequently on the autonomic nervous systemin thesepatients. We studied 10 patients (Group I: 7 men, aged 68 =t 9 years) with normal left ventricular function and 9 patients (Group II: 6 men, aged 70 f 6 years) with systolic left ventricular dysfunction (NYHA heart failure Class II or III). Each patient was paced for 24 h with the optimal and 24 h with the worst AV delay in random order and ambulatory electrocardiograms (ECGs) (Marquette) were recorded. Spectral heart rate variability was analysed for each 24-h period on a Holter analysis system(Marquette Series8000).The optimal and worst AV delay were determined by echo-Doppler as those which produced the greatestand least cardiac output, respectively. For the patients in Group I, there was no difference between the two AV delays as regards indices of heart rate variability. In contrast, for Group II the total frequency (TF) was significantly higher and the low frequency (LF) and LF/high frequency (HF) ratio were significantly lower when the patients were paced with optimum AV delay. Furthermore, this AV delay resulted in significantly higher mean NN and SD. In conclusion, in patients with normal left ventricular function, changes in AV delay do not appear to affect the activity of the autonomic nervous systemin the heart. In patients with mild to moderate heart failure, optimisation of the AV delay causesa significant drop in sympathetic nervous tone. Keywords: Autonomic nervous system; Atrioventricular delay; VDD pacing

1. Introduction Several studies have examined the importance of atrioventricular (AV) delay in the haemodynamic result and endocrine balance of patients with dual * Corresponding author. Tel.: +30 81 269 421; Fax: +30 81 542 055; e-mail: [email protected].

chamber pacing systems [l- 151. However, so far there has been no investigation into whether optirnisation of the AV delay has any impact on the activity of the autonomic nervous system, and partitularly on heart rate variability, in patients with such pacing systems. The analysis of heart rate variability has been considered in recent years to be an indirect, but reliable, method for the evalu-

0167-5273/95/$09.50 0 1995 Elsevier Science Ireland Ltd. All rights reserved SSDI 0167-5273(95)02497-K

236

E. Simantirakis et al. /International

ation of autonomic nervous system activity in the heart, from which inferences may be drawn about the overall autonomic nervous system activity. In this study, we evaluated heart rate variability in paced patients with complete AV block, divided into two groups: those with and without heart failure. The aim of the study was to determine whether optimisation of the AV delay, based on echocardiographic findings, would have a specific effect on the activity of the autonomic nervous system in these patients. 2. Materials and methods 2.1. Patients We studied 10 patients (Group I: 7 men, aged 68 f 9 years) with normal left ventricular function and 9 (Group II: 6 men, aged 70 f 6) with systolic left ventricular dysfunction (NYHA heart failure Classes II and III). All were paced with dual chamber pacing systemsbecauseof complete AV block. The inclusion criteria were: age below 80 years, pacemaker implanted at least three months previously and normal sinus node chronotropy. The last criterion was evaluated by an exercise stress test and the determining factor was a maximum heart rate higher than the product: 0.8 x (220 - age) [16]. Patients with chronic obstructive pulmonary diseaseand inability to exercise were excluded. All patients gave their written informed consent to the study. For the patients in Group II, a sufficient washout period was allowed for them to be drug-free for the duration of the study (Metoprolol, 3 patients, 3 days; Enalapril, 6 patients, 3 days; digoxin, 5 patients, 7 days; frusemide, 5 patients, 3 days). 2.2. Methods For the study of heart rate variability, the pacemakers were programmed in DDD mode, with a basic rate below 40 beats per minute (bpm) in order to avoid atria1 pacing. The optimum and worst AV delay were initially determined using echo-Doppler as being those which resulted in the greatestand least cardiac output. The cardiac output was measuredwith a Sonos 1500system,using a methodology which has been described else-

Journal of Cardiology 52 (1995) 235-239

Table 1

Demographic, haemodynamic andpacingcharacteristics from the patientsstudied GroupI Age (years)

68 f 9 71/3F EF% 58 f 9% LVEDD (mm) 49.03 zt 3.6 LVESD (mm) 34.6 f 2.1 Optimum AV delay 170 * 35 (msec) Worst AV delay 50 * 27 (ms=)

Sex

Group II

P

70 + 6 6M/3F 37 f 8 60.66 f 2.5 44.16 ZIE1.9 165 zt 50

NS NS 0.002 0.002 0.004 NS

58 f 32

NS

EF, ejection fraction; LVEDD, left ventricular end diastolic diameter; LVESD, left ventricular end systolic diameter.

where [ 171.Subsequently, each patient was paced for 24 h with both optimal and worst AV delay, in random order, with continuous ambulatory electrocardiogram (ECG) monitoring (Marquette). Heart rate variability was analysed for each 24-h period using a Holter analysis system (Marquette Series 8000). Software from the same company was used to evaluate heart rate variability. The basic principles and algorithms of this software have been described previously [ 181. In the frequency domain we calculated the total frequency (TF: 0.01 to 1 Hz), the high frequency (HF: 0.15 to 0.40 Hz) and low frequency (LF: 0.04 to 0.15 Hz) spectral power and calculated the LF/HF

Table 2 Cardiac output (CO) and heart rate (HR) at selectedAV delays in the two groups of patients AV delay Optimum Group I CO (Vmin) HR 0-w)

P

Worst

6.79 ZIZ3.15 62.20 f 17.42

5.61 f 3.34 63.4 f 16.20

0.008 NS

5.25 f 2.54 71.33 f 38.25

4.4 f 2.20 78.21 * 31.32

0.003 NS

Group II

CO (Iknin) HR @pm)

E. Simantirakis et al. /International

Journal of Cardiology 52 (1995) 235-239

ratio. Variability in the HF band is considered to reflect changes in parasympathetic tone, while variability in the LF band is believed mainly to express alterations in the tone of the sympathetic branch of the autonomic nervous system. The LF/HF ratio indicates the sympathovagal interaction in the heart. In the time domain, the following parameters were derived: mean NN, SDNN, SDANN, SD, rmSSD and pNN50.

Table 3 Results of time and frequency domain analysis of heart rate variability at two different AV delays AV delay Optimum

P

Worst

Group I

TF HF LF LFIHF Mean NN SDNN SDANN SD rmSSD pNN50

7.1 f 0.63 5.2 * 1.1 5.8 zt 0.6 1.2 f 0.1

1026.3f 45.3 118 f 8.2 104.6 f 9.5 50.6 zt 12.05 47.6 f 32.3 12.6 f 10.7

7.2 5.3 5.9 1.3

zt 0.39 zt 0.8 f 0.3

* 0.1 1022 f 71.3 123.6 ZIE22.9 109 f 24.9 54.3 f 5.5 47 + 32.2 14.4 f 12.2

NS NS NS NS NS NS NS NS NS NS

Group II

TF HF LF LFMF Mean NN SDNN SDANN SD rmSSD pNN50

6.6 f 4.2 zt 4.6 -f 1.02 f 869.6 zt 97.3 zt 73.8 f 59 zt 45.8 f 12.8 f

0.18 1.2 0.8 0.2 139 38.8 28.1 29 35.4 11.2

5.9 f 0.91 4.0 f 1.6 5.2 zt 0.6 1.18 f 0.2 805.8 f 134 104.5 f 31.2 81.2 + 45.8 42.5 f 15.8 37.5 + 30.2 12.2 zt 11.8

0.05

NS 0.05 0.05 0.0006

NS NS 0.05

NS NS

SDNN (ms), standard deviation about the mean R-R interval. SDANN (ms), standard deviation of 5-min mean R-R intervals. SD (ms), mean of all 5-min standard deviations of successive RRs. rMSSD (ms), root mean square of difference between successiveRRs. pNN50 (“I&),proportion of adjacent RRs differing by more than 50 ms. LF, HF, TF in In (m&Hz).

231

2.3. Statistical analysis

Data were expressedasmean * standard deviation (S.D.). Comparisons between groups of independent samples were made using analysis of variance (ANOVA). Comparisons between groups of related samples were assessedby means of repeated measures one way ANOVA. When ANOVA assumptionswere violated, the non parametric Wilcoxon matched pair signed rank test was used instead. A P-value of co.05 was considered as significant. 3. Results

Table 1 gives the characteristics of the two groups of patients. Of those in Group II, 5 had ischaemic heart disease, 3 had dilated myocardiopathy and 1 had hypertensive heart disease.Six of thesepatients were in Class II and 3 in Class III of the NYHA classification of heart failure. Table 2 shows the cardiac output and heart rate at the optimum and worst AV delays while these were being determined. For the patients in Group I, there was no differencebetween the two AV delays as regards indices of heart rate variability (Table 3). In contrast, for Group II the TF was significantly higher and the LF and the LF/HF ratio were significantly lower when the patients were paced with optimum AV delay. Furthermore, this AV delay resulted in significantly higher mean NN and SD. 4. Discussion

This study shows that short term changesin AV delay in patients with normal left ventricular function have no significant effect on autonomic nervous system activity in the heart. In contrast, in patients with systolic left ventricular dysfunction and mild to moderate heart failure, optimisation of the AV delay is particularly important and ensures a reduction in sympathetic activity. Previous studies have examined the importance of the AV delay in the haemodynamic result and the endocrine balance of patients with dual chamber pacing systems.Videen et al. [l] found that a long AV delay is essential in patients with an ejection fraction <40%. In contrast, Feliciano

238

E. Simantirakis et al. /International

et al. [12] concluded that changing the AV delay does not improve the acute haemodynamic status in patients in NYHA ClassesIII to IV of heart failure. Dritsas et al. [13] found that optimisation of the AV delay is more important in patients with diastolic dysfunction than in those with impaired systolic left ventricular function. Brecker et al. [lo] showed that a short AV delay has therapeutic potential in patients with dilated cardiomyopathy, by decreasingthe diastolic mitral valve regurgitation. Finally, Surdacki et al. [ 141found that plasma levels of natriuretic peptide were lower when the patients were paced with the optimum AV delay. In the literature, there is no study which examines the role played by AV delay in the activity of the autonomic nervous system in the heart. However, it is well known that, in patients with heart failure, there is an increase in sympathetic activity, which is also associated with these patients’ prognosis. In the present study, we found that the optimisation of the AV delay in paced patients with heart failure resulted in a significant decreasein sympathetic activity in the heart. This finding is of particular importance in that the optimisation of the AV delay could have a positive influence on the prognosis of patients with heart failure. In patients with normal left ventricular function, optimisation of the AV delay had no impact on the sympathovagal interaction in the heart. In a previous study [15], we examined the autonomic nervous system activity using three different AV delays (100, 175and 250 ms) and found that, in patients with normal left ventricular function, neither the cardiac output nor the autonomic nervous system activity differed significantly from one AV delay to another. In the present study, although the cardiac output differed significantly from the optimal to the worst AV delay, the autonomic nervous systemactivity showed no significant change. It is possible that the 24-h period allotted for pacing at each AV delay was insufficient to cause changes in autonomic tone. A further study is needed to investigate whether longer term pacing with the optimum AV delay has any impact on the sympathovagal activity in such patients. In conclusion, optimisation of the AV delay, while having no impact on the autonomic nervous

Journal of Cardiology 52 (1995) 235-239

systemactivity in patients with normal left ventricular function, has great importance for patients with systolic dysfunction, since it appears to result in a reduction in the sympathetic activity in these patients. References

111Videen JS, Huang SK, Bazgan ID, Mechling E, Patton DD. Hemodynamic comparison of ventricular pacing, atrioventricular sequential pacing, and atria1 synchronous ventricular pacing using radionuclide ventriculography. Am J Cardiol 1986; 57: 1305-1308. PI Iwase M, Sotobata I, Yokota M et al. Evaluation by pulsed Doppler echocardiography of the atria1 contribution to left ventricular tilling in patients with DDD pacemakers.Am J Cardiol 1986; 58: 104-109. [31 Haskell RJ, French WJ. Optimum AV interval in dual chamber pacemakers. PACE 1986;9/5: 670-675. [41 Ah EU, Von Bibra H, Blomer H. Different beneficial AV intervals with DDD pacing after sensedor paced atria1 events. J Electrophysiol 1987; 113:250-256. [51 PehrssonSK, Hjemdahl P, Nordlander R, Astrom H. A comparison of sympathoadrenal activity and cardiac performance at rest and during exercise in patients with ventricular demand or atria1 synchronous pacing. Br Heart J 1988;60/3: 212-220. 161Wish M, Gottdiener JS, Cohen AI, Fletcher RD. Mmode echocardiogramsfor determination of optimal left atrial timing in patients with dual chamber pacemakers. Am J Cardiol 1986;61: 317-322. 171 Mehta D, Gilmour S, Ward DE, Camm AJ. Optimal atrioventricular delay at rest and during exercise in patients with dual chamber pacemakers: A non-invasive assessmentby continuous wave Doppler. Br Heart J 1989;61: 161-166. 181Sulke AN, Chambers JB, Sowton E. The effect of atrioventricular delay programming in patients with DDDR pacemakers. Eur Heart J 1992; 13/4: 464-472. [91 Ritter P, Daubert C, Mabo P, DescavesC, Goufault J. Haemodynamic benefit of a rate-adapted A-V delay in dual chamber pacing. Eur Heart J 1989; 10/7: 637-646. 1101Brecker SJD, Xiao HB, Sparrow J, Gibson DG. Effects of dual-chamber pacing with short atrioventricular delay in dilated cardiomyopathy. Lancet 1992;340: 1308-1312. 1111Frielingsdorf J, Gerber AE, Dur P, Vuilliomenet A, Bertel 0. Importance of an individual programmed atrioventricular delay at rest and on work capacity in patients with dual chamber pacemakers. PACE 1994; 17: 37-45. WI Feliciano Z, Fisher ML, Corretti MC, Gottlieb SS, Gold MR. Acute hemodynamic effectsof A-V delay in patients with congestive heart failure. J Am Co11Cardiol 1994; lA-574A: 349A (791-793). I131 Dritsas A, Joshi J, Webb S, Lewis L, Oakley C, Nihoyannopoulos P. Optimal atrioventricular interval during DDD pacing: Relation to underlying ventricular function. PACE 1993; 16: 884 (abstr).

E. Simantirakis ef al. /International

[14] Surdacki A, Bednarek J, Radziszewski W et al. Plasma ANP and cyclic GMP levels versus left ventricular performance at different AV delays in AV sequential pacing. PACE 1994; 17: 627-636. [15] Vardas P, Skalidis E, Simantirakis E, Eleftherakis N, Kochiadakis G, Manios E. Autonomic nervous system changes during different atrioventricular delays in dual chamber paced patients. Eur JCPE 1994;4: 81. [16] Katritsis D, Camm AJ. Chronotropic incompetence: A proposal for definition and diagnosis. Br Heart J 1993; 70: 400-402. [17] Darsee JR, Walter PF, Nutter DO. Transutaneous Doppler method of measuring cardiac output. II. Nonin-

.lournal of Cardiology 52 (1995) 235-239

239

vasive measurement by transcutaneous Doppler aortic blood velocity integration and M-mode echocardiography. Am J Cardiol 1980;46: 613-618. [18] Marquette Electronics, Inc. Heart rate variability: Physician’s Guide. Milwaukee USA, 1992. [19] FurlanR, Guzzetti S, Crivellaro W et al. Continuous 24-h asessmentof the neural regulation of systemic arterial pressure and RR variabilities in ambulant subjects. Circulation 1990;81: 537-547. [20] Malliani A, Pagani M, Lombardi F, Gerutti S. Cardiovascular neural regulation explored in the frequency domain Circulation 1991; 84: 482-492.