Accessory AV connections (accessory pathways) are remnants of atrial myocardium in the AV groove that failed to resorb when the atria and ventricles were partitioning and separating during embryonic development. Most accessory pathways are capable of conducting electrical activity both antegrade from the atrium to the ventricle and retrograde from the ventricle to the atrium. However, there is no slowing of the speed of electrical propagation across accessory pathways like there is through the AV node where there is a delay of approximately 100 ms from the atria to the ventricles that appears as the “P-R interval” on a standard electrocardiogram (ECG). Therefore, during normal sinus rhythm (NSR), the atrial wavefront propagates across the accessory pathway faster and arrives in the ventricle earlier than it does through the AV node ( Fig. 9.1 ). This early arrival of the atrial impulse in the ventricle is called “ventricular preexcitation,” and it causes an early deflection of the QRS complex off the baseline that results in a “delta wave” appearing on the ECG and with resultant shortening of the P-R interval. Fusion of the two sites of activation in the ventricles from the impulse arriving both through the AV node and across the accessory pathway results in a widened QRS complex.

Wolff-Parkinson-White (WPW) syndrome during normal sinus rhythm (NSR).

Left upper panel, Antegrade conduction across both the atrioventricular (AV) node and the accessory pathway results in the wavefront reaching the ventricles earlier across the accessory pathway than through the AV node, where there is a normal slowing of conduction by about 100 ms. This results in “preexcitation” of the ventricles causing an early deflection off baseline on the standard electrocardiogram (ECG) (“delta wave”) and an obligatory shortening of the P-R interval.

Right upper panel, As the wavefront continues to propagate, the velocity of conduction in the His bundle and bundle branches becomes faster than the velocity of conduction through the ventricular myocardium, and the wavefront becomes fused, resulting in a widened QRS complex on the ECG.

Lower panel, The combination of a delta wave, a shortened P-R interval, and a widened QRS complex on a standard ECG during normal sinus rhythm identifies the problem as WPW syndrome because no other abnormality causes these specific ECG findings.

As long as the patient remains in sinus rhythm, heart function remains normal despite the abnormal sequence of ventricular activation ( Fig. 9.2 , left panel ). The problem arises when an atrial impulse is blocked from crossing the accessory pathway antegrade ( Fig. 9.2 , middle panel ), which can be caused by factors such as a premature atrial beat, a sudden change in the vagal nerve input to the heart, or sudden exposure of the heart to hot, cold, or irritating agents in the esophagus. The lack of antegrade conduction across the accessory pathway results in a normal ECG for that single beat. Because the accessory pathway was not activated during that cardiac cycle, it is not refractory, and the electrical activity in the ventricles can conduct retrograde across the pathway, resulting in a reciprocating tachycardia ( Fig. 9.2 , right panel ). Some patients have what is called a “concealed” accessory pathway, which is one that conducts in the retrograde direction only. The lack of antegrade conduction across these accessory pathway means that these patients have an entirely normal ECG during NSR, hence the designation as a “concealed pathway.”

Reciprocating tachycardia caused by Wolff-Parkinson-White (WPW) syndrome. (A) As long as a patient with WPW syndrome is in a normal sinus rhythm, there is no tachycardia. (B) If antegrade conduction of atrial activity across the accessory pathway is blocked for a single beat, the electrocardiogram for that one heartbeat will be normal. (C) However, when the ventricular activation reaches the base of the ventricle, it can propagate retrograde across the nonrefractory accessory pathway and establish a stable reciprocating tachycardia with antegrade atrioventricular node conduction and retrograde accessory pathway conduction. This is the classic “reciprocating tachycardia” associated with the WPW syndrome.

Ventricular fibrillation with sudden death can occur in patients with the WPW syndrome who have the unfortunate combination of an antegrade-conducting accessory pathway with an abnormally short refractory period and who then develop an episode of atrial fibrillation (AF). In this case, the ventricles are no longer protected from the rapid and irregular bombardment of the AV node during AF, so the AF can be conducted from the atrium across the accessory pathway and cause the ventricles to fibrillate. One would expect this life-threatening combination to occur more often in older adult patients because of the higher incidence of AF associated with advancing age. However, most patients in our series who experienced this particular problem and were successfully resuscitated were young, perhaps because older patients with the condition experienced sudden death without successful resuscitation. Over the years, we operated on 14 patients with the WPW syndrome who had had a prior ventricular fibrillation cardiac arrest. Their average age was 21 years and included a 14-year-old male who was in the lead entering the final lap of the US Junior Amateur Indoor Bicycle Championship race. As he crossed the finish line with one lap to go (the “pistol lap”), the starter fired a pistol near his left ear, and he developed sudden AF that fibrillated his ventricles. He immediately collapsed on the track with sudden cardiac arrest. Fortunately, he was successfully resuscitated and subsequently underwent successful surgery for a left free-wall accessory pathway.

For purposes of WPW surgery, the AV groove was divided into four arbitrary anatomic spaces, the left free-wall space, right free-wall space, posterior septal space, and anterior septal space ( Fig. 9.3 ). Accessory pathways are located in the left free-wall space in approximately 70% of patients with the WPW syndrome, the posterior septal space in 15%, the right free-wall space in 10%, and the anterior septal space in 5%. The objective of surgery is to divide the accessory pathway, which can been accomplished by dividing it at its atrial end or its ventricular end. Originally, intraoperative epicardial mapping of the atrium and ventricles was conducted with a single handheld electrode ( Fig. 9.4 A) to identify the precise location of the accessory pathway within one of the four anatomic spaces of the AV groove. The AV groove fat pad was then dissected away from the epicardial surface of the ventricle at the site of the accessory pathway and for 1 to 2 cm on either side of that site. This intentionally limited dissection left open the possibility that other nearby and unsuspected accessory pathways could be present that could become functional after surgery ( Fig. 9.4 B). In the first 200 patients who had this procedure between 1968 and 1980, the success rate for the initial operation was 61%. The overall success rate was 87% after approximately 25% of patients had a second, third, or fourth surgical procedure, and the incidence of heart block was 10.5%. In the majority of these patients, the accessory pathways were divided via the endocardial approach with limited dissection, although epicardial dissection and cryoablation were used occasionally.

Four anatomic spaces of the atrioventricular (AV) groove. The locations of accessory pathways in patients with Wolff-Parkinson-White syndrome are categorized as being located in one (or more) of four anatomic spaces in the AV groove. The most common site (70%) for accessory pathways to be located is in the left free-wall space (yellow shading) that is bordered by the mitral valve annulus and the epicardial reflection off the left ventricle (LV) and extends from the left fibrous trigone to the posterior superior process of the LV. The second most common site (15%) is the posterior septal space (red shading) that is bordered by the mitral and tricuspid valve annuli and the posterior portions of each ventricle and extends from the central fibrous body to the epicardial reflection off both posterior ventricles. In this space, the AV node artery penetrates the posterior ventricular septum and travels to the AV node in the atrial septum. (See text for further discussion.) The right free-wall space (purple shading) is the next most common site for accessory pathways to occur (10%) and is bordered by the tricuspid valve annulus and the epicardial reflection off the right ventricle (RV) and extends around the complete free wall of the RV. The least common site for accessory pathways (5%) is in the anterior septal space (green shading), which is bordered by the tricuspid annulus, aortic root, and epicardial reflection off the anterior right ventricle.

(A) The original single-point, analog mapping system used to map patients with Wolff-Parkinson-White syndrome. One fixed reference electrode was sutured onto the epicardial surface of the right ventricle and a roving handheld exploring electrode was used to determine the activation sequence of the ventricles and the precise site of ventricular preexcitation. After documenting the insertion site of the accessory pathway in the ventricle, reciprocating tachycardia was induced if possible. If not, ventricular pacing near the pathway was performed. Both of these cause retrograde conduction across the accessory pathway. The atrial side of the atrioventricular (AV) groove was then mapped to confirm that site of atrial insertion of the pathway corresponded anatomically with the ventricular insertion site, and the subsequent surgical dissection was carried out in that vicinity of the AV groove.

(B) The AV groove was exposed either from the endocardium or the epicardium at the mapped site of the accessory pathway and a plane of dissection was established between the AV groove fat pad and the top of the ventricle. This dissection plane was extended for 1 to 2 cm on each side of the pathway as determined by the intreoperative mapping. This limited dissection resulted in other unknown nearby accessory pathways sometimes being missed with the initial surgical procedure, requiring repeat surgery in approximately 25% of patients.

In 1980, we changed the surgical technique to a more anatomy-based procedure so that surgical dissection in the AV groove was no longer confined to the immediate site of the accessory pathway ( Fig. 9.5 A). In addition, a more expeditious way to localize the accessory pathway during surgery was created in 1983 using a “band electrode” array ( Fig. 9.5 B and C). After the accessory pathway was localized by preoperative and intraoperative mapping to one of the four anatomicspaces, the AV groove fat pad in that entire anatomic space was dissected free of the epicardial surface of the ventricle. As a result, the only way we could fail surgically was for a previously undetected second accessory pathway to be located in an entirely different anatomic space. Fortunately, that rarely happened in our experience. We reported the first 118 patients undergoing WPW syndrome surgery using this new surgical approach, and we subsequently performed nearly 1,000 such procedures during the 1980s and early 1990s with a 100% success rate after the initial operation.

(A) In 1980, we changed the surgical approach to dividing accessory pathways in patients with the Wolff-Parkinson-White (WPW) syndrome. Rather than continuing the practice of limited dissection in the atrioventricular (AV) groove based on the precise site of the accessory pathway (see Fig. 9.4 ), all pathways within a specific anatomic space received the same complete surgical dissection of that space based on its anatomic boundaries. Thus, intraoperative mapping was used primarily to confirm the findings of preoperative mapping in identifying the correct anatomic space harboring the pathway. In this example (new anatomic approach) and that in Fig. 9.4 (previous map-based approach), the accessory pathways were in the left free-wall space. (See text for further discussion.)

(B and (C) In 1983, a new method for performing intraoperative mapping in patients with WPW syndrome employed the “band electrode” array in which 16 separate bipolar electrodes were imbedded in a cloth band that could be placed around the ventricular side of the AV groove to localize the site of ventricular preexcitation during normal sinus rhythm or atrial pacing (B) and around the atrial side of the AV groove to localize the atrial end of the accessory pathway during induced reciprocating tachycardia (C). This “band electrode” greatly decreased the time required for intraoperative mapping in patients with the WPW syndrome.

(Modified from Cox JL, Gallagher JJ, Cain ME. Experience with 118 consecutive patients undergoing surgery for the Wolff-Parkinson-White syndrome. J Thorac Cardiovasc Surg . 1985;90:490–501.)

Left Free-Wall Pathways

An endocardial incision to expose the AV groove fat pad is placed 2 mm above the posterior mitral valve annulus from the left fibrous trigone to the posterior atrial septum ( Fig. 9.6 A) Video 9-1. A plane of dissection is developed between the AV groove fat pad and the ventricular surface throughout the length of this incision ( Fig. 9.6 B). The anatomic boundaries of the left free-wall space are the mitral annulus, the epicardial reflection off the ventricle, the left fibrous trigone, and the posterior superior process of the left ventricle (LV). The dissection plane is developed in the entire space within those boundaries. Even though surgery for the WPW syndrome became an anatomy-based procedure after 1980, we continued to perform intraoperative mapping primarily to confirm that the preoperative mapping had correctly identified the anatomic space in the AV groove harboring the accessory pathway.

Surgical technique for left free-wall accessory pathways. (A) The atrioventricular (AV) groove fat pad is exposed through a supra-annular incision placed 2 mm above and parallel to the mitral valve annulus extending from the left fibrous trigone to the posterior superior process of the left ventricle. As an added safety measure against residual A-V fibers crossing the mitral annulus, the two ends of this supra-annular incision were “squared off” to the mitral annulus to isolate the supra-annular strip of atrial myocardium from the rest of the heart.

(B) A plane of dissection is then established between the AV groove fat pad and the top of the posterior left ventricle throughout the length of the supraannular incision. The mitral valve annulus is meticulously freed of any visible fibers between the atrium and ventricle, and the dissection is extended all the way to the epicardial reflection off the ventricle.

(Modified from Cox JL, Gallagher JJ, Cain ME. Experience with 118 consecutive patients undergoing surgery for the Wolff-Parkinson-White syndrome. J Thorac Cardiovasc Surg . 1985;90:490–501.)

Posterior Septal Pathways

An endocardial incision to expose the fat pad in the posterior septal space is placed 2 mm above the tricuspid valve annulus beginning well posterior to the membranous atrial septum and extending counterclockwise onto the posterior right atrial free-wall ( Fig. 9.7 A). The boundaries of the posterior septal space are the central fibrous body, the posteromedial mitral valve annulus, the epicardial reflection off the posterior ventricles, and the posterior right ventricular free wall, and the dissection plane is developed in the entire space within these boundaries ( Fig. 9.7 B). The floor of the posterior septal space is the posterosuperior ventricular septum, and it is often penetrated by the AV node artery. Prior to 1980, every effort was expended to avoid dividing the AV node artery out of fear of creating heart block. However, that approach resulted in an exceptionally high failure rate. We subsequently reasoned that because accessory pathways often travel in one bundle with an artery, the AV node artery should be identified, skeletonized, and purposely divided in all patients with a posterior septal pathway. Division of the AV node artery did not cause heart block because it is not an end artery. I purposely divided the AV node artery in 114 consecutive patients with posterior septal accessory pathways without a single case of postoperative heart block.

Surgical technique for posterior septal accessory pathways:

(A) The atrioventricular (AV) groove fat pad in the posterior septal space was exposed by a supra-annular incision placed 2 mm above and parallel to the tricuspid valve annulus extending from the posterior central fibrous body to the posterior free-wall of the right ventricle.

(B) A plane of dissection was then established between the fat pad and the top of the posterior ventricular septum, and the dissection was extended throughout the boundaries of the posterior septal space, including the posteromedial mitral valve annulus and the posteromedial tricuspid valve annulus and the epicardial reflection off the posterior left and right ventricles. The AV node artery penetrates the posterior ventricular septum in the mid-portion of this space and prior to 1980, it was spared for fear of creating heart block. However, the failure rate of the approach was excessively high. In the revised more anatomic approach, the AV node artery was specifically identified and purposely divided between two vascular clips because of a suspicion that the accessory pathway was accompanying the AV node artery between the ventricular septum and the atrial septum where the AV node is located. Elective division of the AV node artery was performed in 114 consecutive patients with posterior septal pathways, and not one case of heart block occurred. (See text for further discussion.)

(Modified from Cox JL, Gallagher JJ, Cain ME. Experience with 118 consecutive patients undergoing surgery for the Wolff-Parkinson-White syndrome. J Thorac Cardiovasc Surg . 1985;90:490–501.)

AV node reentry tachycardia (AVNRT) results from the presence of two pathways of conduction through the AV node, a fast pathway and a slow pathway. During NSR, antegrade conduction travels down both pathways, but because it reaches the His bundle earlier through the fast pathway, antegrade conduction in the slow pathway is blocked at its distal end ( Fig. 9.8 A). However, if antegrade block occurs halfway through the fast pathway, the impulse in the slow pathway will not only activate the ventricles, but it can also conduct retrograde in the fast pathway to the site of block ( Fig. 9.8 B). Normally, this will not cause reentry because the refractory period in the fast pathway is not short enough to allow it to develop. However, if antegrade conduction block occurs in the proximal end of the fast pathway ( Fig. 9.8 C), the retrograde conduction coming from the slow pathway can cause reentry within the AV node ( Fig. 9.8 D). This AV node reentrant circuit then activates both the ventricles and the atria, and it does so quite rapidly because of the relatively small size of the reentrant circuit.

(A) Atrioventricular node reentry tachycardia (AVNRT) is caused by the presence of two separate conduction pathways through the atrioventricular (AV) node, one fast pathway and one slow pathway. Because conduction is faster through the fast pathway, conduction is normally blocked at the distal end of the slow pathway.

(B) If antegrade block occurs in the distal or middle portion of the fast pathway, there are no adverse sequelae because most of the fast pathway is refractory even though it is also blocked.

(C) However, if antegrade block occurs in the proximal portion of the fast pathway, this entire pathway will not be refractory, and the impulse conducted through the slow pathway can reenter it in the retrograde direction.

(D) This can result in a reentrant circuit forming within the AV node and a tachycardia that activates the atria and the ventricles simultaneously.

In 1979, Dr. Sealy and I were performing an elective open His bundle ablation in a patient with intractable AVNRT. As described in Chapter 11 , it was surprisingly difficult to block the His bundle surgically in these cases, so surgical dissection in the region of the AV node–His bundle complex was routinely more extensive than one might expect. During the dissection, the patient’s AVNRT suddenly terminated but without heart block . The sudden AVNRT termination with restoration of NSR, rather than heart block, suggested that we had inadvertently blocked only one of the two conduction pathways in the AV node. This meant that the two conduction pathways were physically separated by enough distance that one of them could be divided without injuring the other.

This singular fortuitous event could not be repeated in subsequent patients despite all efforts to do so, but it stimulated me to try to develop a reproduceable way to block only one of AV node conduction pathways in patients with AVNRT while leaving the second pathway intact. I believed that our best option was to use a small-tipped cryoprobe, rather than a scalpel, to “sneak up” on the AV node and selectively block only one of the pathways while leaving the other pathway intact. Frigitronics, Inc. made us a special small cryoprobe with a 2-mm active tip, and I assigned the task of trying to block a single pathway to Drs. William Holman and Masatoshi Ikeshita, who had joined our research laboratory in the summer of 1980. The problem was that dogs do not have AVNRT, but by performing pacing studies on every animal entering our laboratory, Holman was able to find three dogs that had dual AV node conduction pathways. The procedure that we had envisioned of “sneaking up” on the AV node using a small cryoprobe was successful in blocking only one of the pathways in all three of those dogs while leaving the second pathway intact, thereby negating the need for a postoperative pacemaker ( Fig. 9.9 ).

Specific pacing studies can identify the two atrioventricular (AV) node pathways in a patient with AV node reentry tachycardia. (A) By plotting the A ₁ A ₂ interval against the A ₂ H ₂ interval, the slow pathway and fast pathway are easily distinguishable. In this particular study, the pacing studies were performed in a dog with dual AV conduction pathways. (B) After performing the discrete cryosurgery procedure around the AV node (see Fig. 9.10 ), the fast pathway is no longer present. A, Atrium; H, His bundle; PCL, pacing cycle length.

(Reproduced from Cox JL. Surgery for cardiac arrhythmias . Cur Prob Cardiol . 1983;8(4):1–60.)

After institutional review board approval in August 1982, we performed the first “discrete cryosurgery procedure” for AVNRT in a 35-year-old female patient in whom we successfully divided the slow pathway in the AV node while leaving the fast pathway intact. Dynamic teamwork between the surgeon and the electrophysiologist was required during the surgical procedure because of the close monitoring of the A-V interval required during application of the cryosurgery. We started the procedure by placing a series of overlapping cryolesions around the borders of the triangle of Koch ( Fig. 9.10 ). Electrical signals were recorded on a storage oscilloscope from leads I, aVF, and V ₅ R and from epicardial electrodes on the surfaces of the RA and RV ( Fig. 9.11 ). Each sweep of the signals was triggered on the RA electrogram and stored on the oscilloscope screen so that with each heartbeat, the A-V interval could be visualized and compared with the A-V interval of the previous heartbeat. After encircling the triangle of Koch, 2-minute cryolesions were placed inside the triangle beginning in the region closest to the os of the coronary and working toward the apex of the triangle. During that time, if the patient suddenly developed complete heart block (i.e., the ventricular deflection disappeared), the cryothermia was immediately stopped. Within a few seconds, the ventricular deflection would reappear, and the A-V interval would start to shorten with each heartbeat. The endpoint of the surgical procedure was the inability to place a cryolesion anywhere within the triangle of Koch for at least 2 minutes without causing complete heart block.

Discrete cryosurgical procedure for atrioventricular node reentry tachycardia. (A) The boundaries and contents of the triangle of Koch in the atrial septum from the surgeon’s view. (B) The discrete cryosurgical procedure begins by creating a small cryolesion at the lateral end of the tendon of Todaro with a 2-mm cryoprobe. (C) Sequential overlapping cryolesions are placed around the borders of the triangle of Koch while the A-V interval is monitored on a beat-to-beat basis.

(Reproduced from Cox JL, Holman WL, Cain ME. Cryosurgical treatment of atrio-ventricular node reentry tachycardia. Circulation. 1987;76:1329–1336.)

During the discrete cryosurgical procedure, the A-V interval is monitored on a beat-to-beat basis using a storage oscilloscope with a screen sweep on every heartbeat that is triggered on the atrial pacing spike. This not only allows the A-V interval to be stored, but it also graphically demonstrates any prolongation of the A-V interval with every heartbeat. (See text for details.)

(Reproduced from Cox JL, Holman WL, Cain ME. Cryosurgical treatment of atrio-ventricular node reentry tachycardia. Circulation. 1987;76:1329–1336.)

Stay updated, free articles. Join our Telegram channel

Join