Beta blocker effects on heart rate during sleep: A placebo-controlled polysomnographic study with normotensive males

Abstract

Although the effect of beta blockers on heart rate (HR) at rest, during exercise, and by ambulatory electrocardiography during the day and night has been studied extensively, data on the effect of these drugs on heart rate during the various stages of sleep are not available. We performed overnight polysomnography and exercise testing in a randomized, double-blind, placebo-controlled, Latin square crossover study of four beta blockers with different ancillary properties (atenolol 100 mg daily, metoprolol 100 mg, pindolol 10 mg, and propranolol 80 mg), on 30 healthy men aged 23-40 years (29.4 ± 4.3) (mean ± SD). At rest pindolol increased HR by 4.7 beats/min ± 13.1 (p<0.05), while beta blockers without intrinsic sympathomimetic activity (ISA) decreased HR (p < 0.0001) by 6.1 ± 8.8 (atenolol), 5.8 ± 8.4 (propranolol), and 5.0 ± 9.4 (metoprolol). Exercise at 125 W increased HR on placebo by 76.4 ± 18.4 beats. Compared to placebo all beta blockers were associated with lower (p < 0.0001) exercise HR by 18.3 ± 23.2 atenolol, 21.1 ± 15.5 metoprolol, 16.8 ± 14.1 pindolol, and 20.8 ± 13.1 propranolol [not significant (NS) among beta blockers]. Thus the effect of beta blockers on heart rate was magnified during exercise. Mean and maximum HR were higher in rapid eye movement (REM) than in nonREM (NREM) sleep (p < 0.001). However, the effects of beta blockers on NREM and REM sleep were almost identical (atenolol 7.3 ± 5.9 decline during NREM and 7.4 ± 5.7 during REM, metoprolol 7.7 ± 6.5 NREM and 8.0 ± 6.0 REM, propranolol 6.7 ± 5.7 NREM and 6.8 ± 6.1 REM). A differential effect of pindolol on NREM/REM sleep analogous to rest/increase of HR in the early morning hours was not observed as long as the subjects remained asleep. These data suggest that: 1) the tachycardia of REM sleep is mediated primarily through nonadrenergic (e.g., vagal) mechanisms, and 2) the increase in HR during early morning hours is probably due to awakening and the assumption of erect position. These findings may be clinically relevant in view of the circadian variation of ischemic morbid and mortal events.