Introduction

Sleep apnea is surprisingly common. It affects between 2% and 4% of middle-aged US adults -- more if polysomnography criteria are used for diagnosis. It's associated with daytime sleepiness, as well as an increased risk of hypertension and cardiovascular disease. Over 90% of cases are obstructive sleep apnea, treated largely by mechanical means, such as nasal continuous positive airway pressure (nCPAP). Effective nCPAP has recently been shown to reduce blood pressure in such patients ¹. However, this type of therapy is not always well tolerated by the patient, especially if they don't have severe daytime sleepiness. And drug treatment has little value in this condition.

Cycles of nocturnal bradycardia and paroxysmal tachycardia are common in patients with sleep apnea. A recent study from France, reported in the New England Journal of Medicine, was initiated by the observation that some patients who had been given a pacemaker had reduced numbers of breathing problems. The investigators evaluated the polysomnography results after atrial pacing in 15 patients with sleep apnea.

Method

Patients who had received dual-chamber pacemakers at least a year previously were screened for problems suggestive of sleep apnea -- e.g. excessive snoring, daytime somnolence, frequent arousals at night. Of those selected, 15 were found on polysomnography in a sleep laboratory to have sleep apnea syndrome, defined by an Apnea Index² of 5 or above, and an Apnea-Hypopnea Index³ of 15 or above.

These 15 patients had no clinical evidence of heart failure, but 11 of them (73%) had a left ventricular ejection fraction between 40% and 56%. None of them were dependent on their pacemakers, which had been implanted for symptomatic sinus bradycardia.

The patients spent 3 consecutive nights in the sleep laboratory, undergoing polysomnographic recordings. The first night represented the baseline, with a basic pacing rate of 55 to 60 beats per minute (bpm). The next night they were randomly assigned to one of two procedures, which were reversed on the third night.

The first intervention was programming the basic ventricular rate of the pacemaker to 40 bpm, with the purpose of recording spontaneous rhythm for a high proportion of the time, representing a 'no-pacing' phase. The alternative intervention ('pacing phase') involved setting atrial overdrive stimulation at a rate 15 bpm faster than the mean nocturnal baseline heart rate; this level was based on its therapeutic efficacy in vagally induced atrial arrhythmias.

For each night, the total duration of sleep and the number of episodes of sleep apnea or hypopnea were analyzed.

Results

The 15 patients comprised 11 men and 4 women, with an average age of 69. There were 8 cases of the central type of sleep apnea, and 7 of the obstructive type.

The results of the breathing disturbances during sleep are given in the table below:

It can be seen that atrial pacing was not associated with any reduction in total sleep time, whereas there were significant reductions in both apnea and hypopnea episodes, compared with the no-pacing, or spontaneous rhythm phase.

Episodes of atrial arrhythmias were seen in both the no-pacing and pacing phases to an equal degree. Moreover, there were no differences in the number of premature ventricular contractions per hour of sleep between the two phases. No episodes of ventricular tachycardia were seen in either phase.

Conclusions

The authors of this study believe that these somewhat surprising results are related to vagal tone in patients with sleep apnea. Hypoxemia, retention of carbon dioxide, bradycardia, and a decrease in blood pressure, all of which are usually seen in central sleep apnea patients, are associated with increased vagal tone. This can be counteracted by atrial pacing at a high rate relative to the spontaneous sinus bradycardia usually seen in such patients. (In the pacing phase, the mean heart rate was 72 bpm, compared with 51 bpm in the no-pacing phase.) Thus theophylline has a beneficial effect in central sleep apnea, which may be partially due to its ability to reduce vagal tone.

The mechanism of the beneficial effect of atrial pacing in obstructive sleep apnea remains unexplained. It is possible that some of the upper airway muscles whose relaxation may be responsible for the condition are indeed under vagal control. Certainly, the benefits of pacing were equally present in this group of 7 patients, so that further work is clearly needed to clarify the exact mechanism.

An accompanying editorial in the Journal points out that pacemaker insertion is unlikely to become an acceptable treatment for sleep apnea. However, the findings, provided they can be replicated, clearly suggest that there may be other, simpler, ways to achieve the same beneficial effect (possibly by affecting vagal tone) that may help the majority of sleep apnea sufferers.