Mechanism of double potentials recorded during sustained atrial flutter in the canine right atrial crush-injury model.

BACKGROUND During atrial flutter, double potentials may be recorded at specific sites in the atria. It has been suggested that double potentials represent sequential activations at the center of the reentrant circuit. An alternative hypothesis is that double potentials represent electrical activity in an area of slow conduction. Understanding their mechanism is important because double potentials have been considered a possible indicator of target sites for catheter ablation. METHODS AND RESULTS We systematically studied double potentials in our canine model of atrial flutter produced by right atrial crush injury using a 64-channel computerized mapping system with 56 electrodes on the right atrium in seven mongrel dogs under general anesthesia. Activation maps were recorded during sinus rhythm before and after crush injury, during rapid pacing above and below the crush injury, and during sustained atrial flutter, entrainment of atrial flutter, and termination of atrial flutter induced with D-sotalol (2 mg/kg). During sinus rhythm before crush injury, activation was uniform, and double potentials were not recorded in any dog. After crush injury, activation proceeded up to and around the crush injury, and narrowly split double potentials were recorded in two of seven dogs. During rapid pacing above and below the crush injury, double potentials were recorded in five dogs. During 14 episodes of atrial flutter (mean cycle length, 140 +/- 16 msec), double potentials were recorded at electrodes along the crush injury. The activation time of the early x component of the double potentials (25 +/- 13 msec) was similar to that of adjacent electrodes above the crush injury (24 +/- 11 msec), and the activation time of the late y component (89 +/- 13 msec) was similar to that of adjacent electrodes below the crush injury (91 +/- 14 msec). The timing of the x and y components was dependent on the location of the recording electrode, with x and y widely spaced at the end of the crush injury near the area of earliest atrial activation during atrial flutter, more equally timed at the center of the crush injury, and more closely timed at the end of the crush injury opposite the area of earliest activation. During transient entrainment, double potentials were accelerated to the pacing rate, but their activation time relative to adjacent electrodes was maintained. During abrupt termination of atrial flutter, the early x component of the double potential was always recorded, but the late y component was not, because of conduction block below the posterior end of the crush injury. CONCLUSIONS This study has shown in our canine model of atrial flutter that double potentials are recorded from the center of the reentrant circuit and that they represent sequential activations as the reentrant wave front passes on either side of the crush injury.