Article Text

Download PDFPDF

Dual chamber pacing in patients with severe heart failure on β blocker and amiodarone treatment: preliminary results of a randomised study
  1. H Nägele,
  2. R Schomburg,
  3. B Petersen,
  4. W Rödiger
  1. Department of Cardiovascular Surgery, University Clinic Eppendorf, Hamburg, Germany
  1. Correspondence to:
    Dr. Herbert Nägele, Klinik für Herz-und Gefäβchirurgie, Universitätsklinikum Hamburg-Eppendorf, Martinistr 52, D-0246 Hamburg, Germany;

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Standard dual chamber pacing was first used as adjunctive treatment for severe congestive heart failure in the early 1990s.1 It was proposed that a short atrioventricular delay reduced presystolic mitral regurgitation.2,3 It also may correct chronotropic incompetence and protect against fatal bradyarrhythmias.

We performed a randomised study to test whether DDDR pacing with optimised atrioventricular (AV) delay and reversal of drug induced bradycardia by rate responsive pacing was beneficial in heart failure patients who were receiving combined treatment with amiodarone and β blockers as empirical prophylaxis against sudden tachyarrhythmic death.4

In 82 patients with severe heart failure submitted for heart transplantation and without a conventional pacemaker indication between 1996 and 1998 (85% male; 52% idiopathic dilated cardiomyopathy), treatment with low dose amiodarone (1000 mg/week) plus titrated doses of carvedilol (target 50 mg/day) was instituted. In addition, a dual chamber pacemaker was installed with a minute ventilation sensor (Chorus RM).

After a three month wash out period (no pacing), the devices were randomly programmed in VVI backup mode with a basal rate of 40 impulses/min and a hysteresis of 25% (n = 41, “no pacing”), or in DDDR mode with heart rates from 65 to 120 beats/min and haemodynamically optimised AV delay (n = 41). The two groups (VVI, DDDR) showed similar baseline values in New York Heart Association functional class (mean (SD): 3.24 (0.6) v 3.18 (0.7)), heart rate (88 (14) v 88 (16) beats/min), and ejection fraction (24.8 (9)% v 24.4 (11)%) and were observed for 2.8 (1.1) years.

In contrast to VVI inhibition mode, DDDR pacing shortened the PR interval from 217 (55) to 172 (19) ms (p < 0.001) but prolonged the QRS interval from 143 (38) to 174 (30) ms (p < 0.001). The mean resting and 24 hour heart rates after one year were 60 (8) and 65 (8) beats/min in the VVI group, and 69 (9) and 74 (8) beats/min in the DDDR group (p < 0.01 between the groups). In the VVI patients (without pacing), the left ventricular ejection fraction increased more than in the DDDR patients after one year (40 (12)% v 32 (11)%, p = 0.04 between the groups). Changes in left ventricular ejection fraction (LVEF) were negatively correlated with heart rate changes (r = −0,47, p < 0.001). In other words, heart rate lowering seemed to be closely associated with an increase in LVEF. This may partly be explained by the fact that when cardiac output and volumes were unchanged, the ejection fraction has to be increased. Also, left ventricular end diastolic diameter was slightly higher and left ventricular stroke work significantly lower in paced patients. There was no effect on mitral regurgitation.

In contrast to our hypothesis, there were more cardiac events such as death, the need for heart transplantation, or the need for upgrade to biventricular pacing in the DDDR group than in the VVI group (17 v 8, p = 0.03, table 1). On the other hand nine of the 41 patients in inhibited VVI mode (21%) had to be reprogrammed to DDDR mode because of symptomatic bradycardia (crossover).

Table 1

Cardiac events during the observation period


We conclude that conventional DDDR pacing with optimised AV delay is not superior to VVI pacing and may be deleterious in patients with severe heart failure treated with β blockers and amiodarone, and who do not have a conventional indication for pacing. In particular, no effect on mitral regurgitation could be observed. These data contrast with the findings in patients undergoing biventricular pacing.5

Two mechanisms may be responsible for our negative result: first, the increased, possibly inappropriate, heart rate induced by the minute ventilation sensor may have impaired ventricular filling; second, the direct negative effects of right ventricular pacing in the DDDR group may have induced ventricular dyssynchrony. Our findings raise the possibility that heart failure patients may improve steadily over at least one year without continuous pacing. Backup pacing was necessary in a substantial number of patients treated with a combination of amiodarone and β blockers, and so patients on such treatment should be monitored closely. Pacing may be an adjunct that can maximise medical heart failure treatment, obviating the need to withdraw negative chronotropic drugs.