Introduction: Cellular mechanisms of adenosine include a direct effect on the activation of the adenosine-sensitive potassium current (I<sub>K,Ado</sub>) and an indirect effect on antagonism of catecholamine-stimulated adenylate cyclase activity. However, previous studies evaluating the influence of catecholamine activity on the electrophysiologic effects of adenosine have yielded conflicting results. We tested the hypotheses that if adenosine exerts its atrioventricular (AV) nodal blocking effects directly by activating the I<sub>K,Ado</sub> potassium current, rather than indirectly by reversing the catecholamine effects, then pretreatment with β-adrenergic blockade would not potentiate the effects of adenosine in terminating AV nodal-dependent supraventricular tachycardia (SVT). Methods and Results: During sustained AV nodal reentrant tachycardia (AVNRT) or AV reentrant tachycardia (AVRT) in 28 patients, adenosine was rapidly injected in incremental doses of 1.5, 3, 6, 9, 12 and 18 mg to determine the lowest effective dose required for tachycardia termination before and immediately after the end of esmolol infusion. Esmolol infusion was started with loading doses of 500 µg/kg/min for 1 min and 150 µg/kg/min for 4 min, followed by a maintenance infusion of 50–100 µg/kg/min. Esmolol infusion was continued until the tachycardia was terminated or the maximal dose of 100 mg was reached. Adenosine was effective in terminating SVT in all 28 patients with a mean lowest effective dose of 96 ± 54 µg/kg before esmolol. During esmolol infusion, tachycardia was reproducibly terminated in 8 patients (6 with AVNRT, 2 with AVRT) with a mean dose of 67 ± 23 mg. In the other 20 patients with persistent tachycardia after 100 mg of esmolol infusion, the lowest effective dose of adenosine could be determined in 19 patients. In the remaining patient with AVRT, the maximal dose of adenosine (18 mg) was unable to terminate the tachycardia immediately after the end of esmolol infusion. In these 19 patients, esmolol infusion caused significant lengthening of the tachycardia cycle length from 338 ± 36 to 372 ± 51 ms (p < 0.0001) and reduction of the mean arterial blood pressure from 96 ± 15 to 88 ± 18 mm Hg (p = 0.034). Compared to the dosage that was determined before esmolol infusion, the lowest effective dose of adenosine remained the same in 13 patients after the end of esmolol infusion, whereas the dose increased in 5 and decreased in 1 patient. The mean lowest effective dose of adenosine was not significantly different before (98 ± 54 µg/kg) and immediately after (115 ± 56 µg/kg) the end of esmolol infusion (p = 0.054). Conclusions: Intravenous esmolol infusion (up to 100 mg total dose) usually fails to terminate AV nodal-dependent SVT. In the esmolol-resistant tachycardia, esmolol pretreatment does not produce a positive synergistic effect on the efficacy of adenosine-induced termination of SVT. Therefore, in this tachycardia adenosine may exert its effects on AV nodal conduction directly by activation of the I<sub>K,Ado</sub> potassium current, rather than by antagonizing the β-adrenergic system.