Background: Atrial flutter (AFL) is a common postoperative sequela of the modified Fontan operation, or total cavopulmonary connection. We hypothesized that injury to the crista terminalis (CT) by the lateral tunnel suture line contributes to the development of AFL in this setting. This study was designed to determine the effects of alteration of the lateral tunnel suture line, relative to the CT, on the inducibility of AFL in an acute canine model of the modified Fontan operation.
Methods: Adult mongrel dogs (n = 25) underwent a median sternotomy and normothermic cardiopulmonary bypass. In groups 1, 2, and 3, through a right atriotomy, a suture line was placed to simulate the lateral tunnel of the modified Fontan operation (n = 20). The lateral aspect of the suture line ran along the CT in group (n = 10), 5 mm medial to the CT in group 2 (n = 5), and 10 mm anterior to the CT, incorporated into the atriotomy closure, in group 3 (n = 5). In group 4 (n = 5), only the lateral portion of the suture line, along the CT, was placed. Form-fitting 253-point unipolar endocardial mapping electrodes were inserted in the left and right atria via bilateral ventriculotomies. Induction of AFL was then attempted using atrial burst pacing. If sustained AFL could not be induced, isoproterenol was administered and the pacing protocol repeated. Endocardial activation sequence maps of spontaneous rhythm and AFT were constructed.
Results: Under baseline conditions, after placement of the suture line, sustained AFL could reproducibly be induced in 8/10 dogs in group 1, 0/5 dogs in group 2, 0/5 dogs in group 3, and 5/5 dogs in group 4 (p < 0.001). After isoproterenol administration, sustained AFL was reproducibly inducible in the remaining 2 dogs in group 1, 4/5 dogs in group 2, and 0/5 dogs in group 3 (p = 0.01). The mean cycle length of AFL was 189 +/- 25 ms in group 1, 136 +/- 8 ms in group 2, and 182 +/- 20 ms in group 4 (p < 0.001). Atrial activation sequence maps, during sinus rhythm, demonstrated a line of conduction block along the lateral portion of the suture line in all cases in groups 1 and 4 and in only those cases in group 2 in which sustained AFL was inducible. During AFL this block facilitated unidirectional conduction, permitting propagation of the reentrant wavefront. Mean conduction velocity along the CT during sinus rhythm was 0.63 +/- 0.10 m/s in group 1, 1.04 +/- 0.17 m/s in group 2, 1.01 +/- 0.12 m/s in group 3, and 0.44 +/- 0.13 m/s in group 4 (p < 0.01).
Conclusions: In an acute canine model of the modified Fontan operation, conduction block imposed by the lateral tunnel suture line is an essential component of the AFL circuit. The inducibility of AFL is increased by suture line placement along the CT. Slow conduction, resulting from injury to the CT, promotes this increased inducibility. Avoidance of the CT may reduce the incidence of AFL in children undergoing the modified Fontan operation.