Atrial Pacing In Sleep Apnea Syndrome

Abstract & Commentary

Synopsis: Higher rate atrial pacing reduces the severity of sleep disturbance in patients with sleep apnea syndrome.

Source: Garrigue S, et al. N Engl J Med. 2002;346: 404-412.

Garrigue and colleagues from bordeaux, France, report on a novel approach to the management of patients with sleep apnea syndrome. They identified patients who had reported symptoms consistent with sleep apnea syndrome who already had dual chamber pacemakers that had been implanted for standard clinical indications. Twenty-six patients were invited to participate and, in 15 of them, polysomnographic recordings revealed sleep apnea syndrome. These patients became the study group. The patients spent 3 consecutive nights in the sleep laboratory. Baseline recordings were performed on night one during which a basic lower rate limit of 55 bpm to 65 bpm was set on their pacemakers. By random assignment, on one of the next 2 nights, the pacemaker was programmed to a VVI mode at a lower rate limit of 40 bpm in hopes of maintaining spontaneous rhythm most of the night. This will be called the "no pacing" study. During the other night, atrial overdrive stimulation was provided at a rate 15 bpm faster than the mean nocturnal heart rate recorded during the baseline assessment. The total duration and numbers of central or obstructive sleep apnea or hypopnea episodes were analyzed and compared.

Among the 15 patients in the study group, 9 patients had a history of only sinus node disease and 6 had a bradycardia-tachycardia syndrome. All patients presented with evidence for episodes of both obstructive and central apnea. The central apnea index was 12 ± 14 and the obstructive apnea index was 7 ± 4. Seven patients had a predominantly obstructive apnea pattern whereas 8 patients had apnea that was predominantly central. Eleven of the 15 patients had mild depression of left ventricular ejection fraction, but no patient had an ejection fraction below 40%. The percentage of total sleep time during which oxyhemoglobin saturation was below 90% was 12 ± 8% and the mean arousal index was 21 ± 12% during the qualifying, baseline study.

During the no pacing phase, pacemaker stimulation at the lower rate limit of 40 bpm was used during 1.0 ± 0.5% of the sleep study. The mean total sleep time during the no pacing study was 321 ± 49 minutes, the mean apnea-hypopnea index was 28 ± 22, and the mean arousal index was 18 ± 9. Atrial pacing at a rate of 15 bpm faster than the mean nocturnal heart rate resulted in a significant reduction in all types of apnea. In 13 of 15 patients, the apnea-hypopnea index was reduced by more than 50%. Both the obstructive apnea index and the central apnea index were reduced in every patient. There was a reduction in the amount of sleep time during which oxygen desaturation was observed from 10 ± 6% to 6 ± 4%. No adverse events of nocturnal overdrive pacing were noted.

Garrigue et al conclude that higher-rate atrial pacing reduces the severity of sleep disturbance in patients with sleep apnea syndrome. Garrigue et al postulate that this is due to alterations in autonomic nervous system tone in the patients with sleep apnea syndrome but consider the improvement in patients with primarily obstructive sleep apnea to be unexplained. They suggest further studies to assess the precise role of cardiac pacing in the general population of patients with sleep apnea syndrome.

Comment by John P. DiMarco, MD, PhD

This interesting paper should stimulate further studies on the use of pacing in patients with sleep apnea. Transient nocturnal bradycardia is common in patients with sleep apnea syndrome. Depending upon the severity of the condition, periods of sinus arrest or AV block may be documented in between 15% and 35% of patients with the sleep apnea. In most of these patients, bradyarrhythmias are not noticed during waking hours. Since purely nocturnal bradycardia does not result in symptoms in the conventional sense, bradycardia during sleep has not been considered a standard indication for pacing. Other studies have shown that therapy for sleep apnea with continuous positive airway pressure decreases the incidence and severity of bradycardia, and it has been felt that the apnea produced reflex changes that caused the bradycardias observed. This paper suggests that the interaction between heart rate and apnea may be 2 directional.

In this paper, all of the patients were recruited from a population of patients who had already received pacemakers. Since this was a preliminary study, having patients who already had pacemakers was probably the only way Garrigue et al could conduct the trial. Purely nocturnal bradycardia that is asymptomatic, the typical situation in patients with sleep apnea syndrome, is usually not thought to be an indication for permanent pacemaker insertion. However, the observation here that among these patients who fortuitously had a pacemaker, sleep disturbances related to sleep apnea decreased with pacing, raises the hope that similar observations would be made in patients without some other primary indication for pacing.

The benefits of pacing may be even greater than is shown here. Even during the "non pacing" part of the study, back-up ventricular pacing at a rate of 40 bpm was available. Frequently, pauses of between 2 and 10 seconds are noted during sleep in patients with sleep apnea. If Garrigue et al had been willing to completely turn off the pacemakers, they may have seen even more dramatic effects.

Dr. DiMarco is Professor of Medicine, Division of Cardiology, University of Virginia, Charlottesville.