Respiratory sinus arrhythmia is thought to be vagally mediated, since it disappears after atropine, but the site of action of the drug (central vs. peripheral) accounting for this effect has not been elucidated. To investigate the effects of anticholinergic agents on respiratory arrhythmia, ten healthy subjects received an intravenous bolus of tropatepine (a presumed central antagonist) at a dose of 0.08 mg per kg of body weight, then, 7 min later, prifinium (a peripheral antagonist) at a dose of 0.1 mg per kg of body weight. Respiratory sinus arrhythmia during controlled breathing was evaluated as the area under the high-frequency peak of the heart rate variability spectrum coinciding with the respiratory frequency +/- 0.02 Hz. The power of this high-frequency peak decreased by 55% after tropatepine (P less than 0.05) with a concomitant increase of the mean RR interval from 930 to 1072 ms (P less than 0.01). When prifinium was added, a further but non-significant decrease of respiratory arrhythmia was observed, while the mean RR interval decreased from 1072 to 714 ms (P less than 0.01). The low-frequency components (0.05 to 0.15 Hz) of the power spectrum, significantly decreased (P less than 0.05) after infusion of both drugs. In conclusion, tropatepine depresses respiratory sinus arrhythmia with a paradoxical concomitant bradycardia. This suggests that tropatepine acts like a pure central muscarinic antagonist, in support of the hypothesis that a central cholinergic receptor is involved in the respiratory modulation of heart rate.