SUMMARY An isoproterenol infusion (1.0-4.0 ,ug/min) was administered to 15 patients with intermittent bundle branch block (BBB) and two patients with apparently fixed BBB. Three main effects were documented: (1) In all patients with phase 3, or tachycardia-dependent, BBB, isoproterenol caused a pronounced shortening of refractoriness in the affected fascicle. (2) In patients showing phase 4, or bradycardia-dependent, BBB, isoproterenol prolonged the phase 4 block range, probably because of enhanced diastolic depolarization. In one patient (four studies) in whom phase 4 block was not present, isoproterenol caused the appearance of a phase 4 block range. (3) In the two patients with fixed BBB, isoproterenol restored conduction, probably as a result of a hyperpolarizing effect. This study shows that isoproterenol tends to restore or improve conduction related to tachycardia-dependent block, but may impair conduction related to bradycardia-dependent block. dependent BBB is favored or provoked, probably through an enhancement of diastolic depolarization.Materials and Methods Fifteen patients with intermittent BBB were studied. Carotid sinus massage was performed in every case during a continuous ECG recording to document the existence and measure the extension of a phase 3 and phase 4 block range. The 15 cases were selected from a much larger series of patients with intermittent BBB and entered the study only when the maneuvers were effective in uncovering these conduction ranges, or in separating a phase 3 block range from a normal conduction range. Two patients who apparently had permanent or rate-independent BBB, but who were known to have intermittent BBB a few weeks before, were also included. During the vagal stimulation, junctional and ventricular escapes were sometimes documented. After this control study, carotid sinus massage was repeated during and after the administration of an i.v. drip of isoproterenol, for which 1 mg of the drug was dissolved in 500 ml of isotonic glucose solution. The infusion was started at a rate of 1 ,ug/min of isoproterenol during 5-10 minutes; if no significant changes in conduction were observed, the rate of infusion was raised in successive steps of 1 gg/min during 5-10 minutes; the infusion was interrupted either when clear changes in conduction were seen or when changes did not occur after the infusion of 4 ,ug/min, or when ventricular arrhythmias developed at any time (the latter occurred only in one patient, who was then excluded from the study). When conduction changes were observed, the vagal stimulation was repeated after the infusion, until the recorded changes returned to the control condition. All the tracings were recorded on a Sanborn direct-writing two-channel machine and conventional measurements were made to determine the variations in cardiac rate and extension of the conduction ranges, as well as in the coupling interval of the escape beats. The statistical significance of the results was assessed using the paired t test. Thirteen patients were studied only 135...