Prevention of Atrial Tachycardia After a Right Atriotomy

Patient selection Consecutive patients operated in the Department of Cardiothoracic Surgery, Aarhus University Hospital at Skejby were included. Inclusion criteria were: 1. Clinical indication for corrective operation for congenital heart disease with the use of a right lateral atriotomy, 2. Age >18 years, 3. Ability to give an informed consent. The only exclusion criterion was acute operation.

Written informed consent was obtained from all the patients prior to the inclusion in the study. The ethical committee for Aarhus County approved the study September 18, 2003.

Operation and postoperative care The operation was performed in the usual fashion. Additionally, a single linear endocardial 1-minute cryolesion connecting the right lateral atriotomy with the tricuspid annulus anteriorly and with crista terminalis posteriorly was performed before the right atriotomy has been sutured (Figure 1). SurgiFrost® cryosurgical probe (Endocare Inc., Irvine, California) was used. Two pairs of epicardial electrodes were placed on the atrium between the incision and the tricuspid annulus 1.5 cm cranially, and caudally from the cryolesion at the end of the repair. They were brought out together to the thorax on the patient's right side. Subsequently, the thorax was closed in the usual way.

Patients were monitored continuously using bedside monitors or telemetry for the first 16-24 hours after the operation. A 12-lead electrocardiogram (ECG) was obtained at any clinical suspicion of arrhythmia. Atrial tachycardia was defined as a sustained regular monomorphic atrial rhythm at a constant rate >100/min originating outside the sinus node.

Measurement of conduction time On the second postoperative day, the conduction time was measured with the help of pacing the electrode pair on the one side of the cryolesion between the incision and the tricuspid annulus and sensing the unipolar signals on the other side (Figure 2). Signals were recorded at 100 mm/s using a commercially available ECG machine.

Electrophysiology study and coronary angiography Electrophysiology study and coronary angiography were carried out 3 months after the operation. Patients were free of antiarrhythmic agents at the time of the study. Coronary angiography was performed with the usual technique from the femoral approach. Afterwards, a 7-French quadripolar deflectable ablation catheter with a 4-mm tip electrode (Navistar®, Biosense Webster, Diamond Bar, California) was advanced via the right femoral vein into the right atrium for mapping and radiofrequency catheter ablation. An active fixation lead (Sweet Picotip®, Guidant, St. Paul, Minnesota) was screwed in the lateral right atrium near the tricuspid annulus and the performed cryolesion and was used for pacing. Simultaneous recordings of intracardiac electrograms filtered at 50-500 Hz and 12-lead surface ECG were stored digitally (CardioLab®, Prucka Engineering, Inc., Houston, Texas or Prucka CardioLab 2000®, GE Medical Systems, Milwaukee, Wisconsin). An electroanatomic map (CARTO®, Biosense Webster, Diamond Bar, California) was constructed during pacing with special emphasis on the region of surgical scar and cryolesion to be able to judge conduction across the cryolesion. Afterwards, the active fixation lead was repositioned to the opposite side of the cryolesion near the Navistar® catheter, which was placed with the help of the already constructed electroanatomic map. Another electroanatomic map was created under pacing from this second site to judge the completeness of the cryolesion in the other direction. Afterwards, attempts at induction of atrial arrhythmias were made. These included rapid atrial pacing with 20 stimuli starting from a cycle length of 400 ms with 20 ms decrements down to a cycle length of 200 ms, and single and double extrastimuli at 400 ms cycle length down to atrial refractoriness. Isoproterenol was not used during the study.

Safety assessment of the intervention All adverse events were recorded. Likely relationship of adverse events to the study intervention was determined based on the consensus of the participating investigators.

Statistical analysis All calculations were carried out using the statistical software Intercooled Stata® Release 8 (Stata Corporation, College Station, Texas). Comparison of continuous variables was performed using the Student's t-test, or the Wilcoxon rank-sum test, as appropriate. Comparison of proportions was performed using the Fisher's exact test. Median (interquartile range) or mean ± standard deviation are reported, as appropriate.