Surgical treatment of left free wall accessory pathways of atrioventricular conduction of the Kent type

J

THORAC CARDIOVASC SURG

81:698-706, 1981

Surgical treatment of left free wall accessory pathways of atrioventricular conduction of the Kent type This report relates the surgical experience with 79 patients with arrhythmias who had 82 left free wall accessory pathways of atrioventricular conduction of the Kent type. The current operation for the interruption of the pathway consists of two steps. First. the localization of the pathway is done by mapping the epicardial excitation sequence of the atria and the ventricles. This is followed by a set group of maneuvers needed to divide the pathway. which is not visible or palpable. First. an endocardial atrial incision is made just above the anulus fibrosus extending 1.5 to 2.0 em on each side of the pathway's crossing point. Then. a wide separation of the coronary sulcus fat from the atria and ventricles is done. Finally. superficial fibers of the ventricular myocardium are separated from the anulus fibrosus . The surgical problems such as the possible coronary sinus origin of a pathway and the proximity of pathways to the left fibrous trigone are discussed. both from the electrophysiological and the surgical standpoints. The overall success rate for interruption of left free wall pathways has been 90% with no failures in the last 34 patients.

Will C. Sealy, M.D., and John J. Gallagher, M.D., Durham. N. C.

Left free wall accessory pathways of atrioventricular conduction of the Kent type were found in 79 of the 160 patients at the Duke University Medical Center treated by direct operation for arrhythmias resulting from accessory pathways. Preoperative assessment of the patient, identification of the left free wall pathway at operation, and the systematic dissection needed for its interruption differ in several respects from the maneuvers needed in the surgical treatment of accessory pathways crossing elsewhere on the atrioventricular junction. The purpose of this paper is to record this clinical experience, to point out the problems unique to left free wall pathways, and finally to give the current approach to their surgical treatment. From the Divisions of Thoracic Surgery and Cardiology, Duke University Medical Center, Durham. N. C. Supported in part by Grant RR-30 from the General Clinical Research Centers Branch, Division of Health Resources and Grants HLI5190 and HLI3920 from the National Institutes of Health, Bethesda, Md. This work was done during Dr. Gallagher's tenure as an Established Investigator of the American Heart Association. Received for publication Aug. 28, 1980. Accepted for publication Oct. 7, 1980. Address for reprints: Will C. Sealy, M.D., Professor of Thoracic Surgery, P.O. Box 3093, Duke University Medical Center, Durham, N. C. 27710.

698

Anatomic considerations Left free wall accessory pathways are the anomalous atrioventricular connections that cross from the left atrium to the left ventricular free wall. The left ventricular free wall begins anteriorly at the left fibrous trigone and extends posteriorly to the posterior superior process of the left ventricle (Fig. 1). The latter segment of the left ventricle, which is within the area described as the crux, connects the anulus fibrosus to the posterior portion of the muscular ventricular septum. In this report the free wall is further divided into the anterior sector, which extends from the left fibrous trigone to one half the distance between the trigone and the obtuse angle. The lateral sector, which includes the obtuse angle of the heart, is between the anterior and the posterior sectors. This lateral sector contains the junction of the coronary sinus with the great cardiac vein and posterior auricular vein (oblique vein of Marshall). The posterior sector begins at the junction of the free wall with the posterosuperior process of the left ventricle and includes the posterior ventricular free wall from this point to one half the distance to the obtuse angle. This sector lies beneath most of the course of the coronary sinus. Some points in the surgical anatomy of the area containing left free wall pathways that have significance,

0022-5223/81/050698+09$00.90/0 © 1981 The C. V. Mosby Co.

Volume 81

Left free wall accessory pathways of AV conduction

Number 5 May, 1981

699

Table I. Clinical findings Age (yr)

No. of patients

10-15 16-30 31-45 46-60 >60

10 28 20 19 2

Totals

79

Mean age at onset of arrhythmia (oldest) 5 (13) 14 (24) 20 (34) 30 (49) 20 yr

yr yr yr yr

Pathways with lethal potential 4 16 8 6 I

35

particularly in localization, are shown in Figs. 2 and 3. The right atrium at the point where the coronary sinus emerges extends to the left of the plane of the sulcus between the two ventricles. Thus the right atrium may be just superior to the junction of the left ventricular free wall and the posterior superior process, a position that offers the possibility for pathways to enter the free wall from the right atrium. The coronary sinus just before its entry into the right atrium may be more than I em from the summit of the ventricle, as noted in Fig. 3, Inset 5. It then courses obliquely across the coronary sulcus in the posterior sector to its junction with the great cardiac vein and the left auricular vein posterior to the obtuse angle. The muscular wall of the coronary sinus is continuous with atrial myocardium, and, from our surgical observations, it is believed that the myocardium in the wall can be the atrial origin of a pathway. The great cardiac vein varies in its distribution over the left side of the heart and its relationship to the circumflex artery and its branches. A common course of this vein is shown in the cross section of the free wall in Fig. 2. The depth of the coronary sulcus varies in different patients as does the configuration of the ventricular summit. In some specimens the summit is thin at its attachment to the anulus fibrosus, whereas in others it is thick and rounded. These geometrical variations, including the thickness of fat between the myocardium and the epicardium, may make the epicardial mappings less precise in localizing a pathway. The relationship of the coronary arteries in the coronary sulcus to the atria and the ventricles may show several different patterns as described by James' and by McAlpine.P The left main coronary vessel as it emerges from the aorta and bifurcates is usually at a distance from the anulus fibrosus, too far away to be injured in the dissection for a pathway interruption at the trigone. The branching pattern of the circumflex branch is such that in most patients it courses for only a short distance in the sulcus but can course around the entire free wall

LEFT FREE WALL PATHWAYS Fig. 1. View of the heart looking down from above showing the location of left free wall pathways. The free wall is divided into three sectors. The anterior (Ant.) one extends from the left fibrous trigone to half the distance from this point to the obtuse angle. The lateral (Lat.) sector includes the obtuse angle and extends to the junction with the posterior (Post.) sector. The posterior sector begins at the junction between the free wall and the posterosuperior process of the left ventricle and continues anteriorly to a point about one half the distance between the point just described and the obtuse angle. This means that the posterior sector contains most of the coronary sinus and, therefore, is more easily mapped. Since it does contain the coronary sinus, which is surrounded with atrial myocardium, the sinus can be the atrial end of an accessory pathway.

and be the origin of the posterior descending artery (9% of patients according to James'). Occasionally, as in one patient in the series, the circumflex branch may course within the atrial muscle in the lateral sector; however, when it does, it is superior to the anulus fibrosus, above the endocardial incision used for pathway interruption. Since the coronary sinus is really partially within the atrial wall, the circumflex coronary artery, when it extends to the posterior sector, is usually below the coronary sinus or between the inferior border of the sinus and the atrial wall. Fortunately, with the exception of the occasional branch with an intraatrial course, the arteries are usually deep within the sulcus fat and are protected from injury during operation. The left fibrous trigone is the anterior limit of the point where the atrium and ventricle are attached to opposite sides of the anulus fibrosus. However, the left trigone, which can be easily identified, can have an accessory pathway along its border. In Fig. 2, Inset 1, the external tip of the trigone is visible at its junction with the anulus fibrosus.

The Journal of

700

Sealy and Gallagher

Thoracic and Cardiovascular Surgery

Fig. 2. Drawing of the lateral view of the heart showing the junction between the coronary sinus, the left auricular oblique vein (shown here surrounding the posterior aspect of the left atrial appendage), and the great cardiac vein. The cross sections of the heart drawn from one autopsy specimen taken at points 1, 2, and 3 are shown in the insets. Inset 1 is made adjacent to the left fibrous trigone, which is shown as an enlargement of the anulus fibrosus. Underneath the epicardium and superiorly one can see the circumflex branch of the coronary artery. Inset 2 includes the great cardiac vein, which is represented by a large, thin-walled vessel just above the ventricular summit. At this location the small branch of the circumflex artery is above the vein, which is the small, thick-walled vessel close to the appendage. In Inset 3, the great vein is now above and medial to the coronary artery. These drawings show how deep the coronary sulcus can be as well as demonstrating an example of a thick ventricular summit.

5 Fig. 3. The posterior aspect of the heart. The large coronary sinus is seen emerging from the right atrium, both of which extend to the left of the interventricular groove and thus superior to the left ventricle. The coronary sinus, as shown in Inset 5, is usually I em above the left ventricle and extends obliquely downward to the coronary sulcus at the obtuse angle to its termination in the left auricular oblique vein and great cardiac vein. In this specimen, only two small branches of the circumflex artery are present in Inset 4, and none at all are present in Inset 5. The coronary sinus is surrounded by atrial myocardium.

Volume 81

Left free wall accessory pathways of AV conduction

Number 5

70 I

May, 1981

Table II. Postoperative problems

I

Postpericardiotomy syndrome Constrictive pericarditis Hepatitis Pancreatitis Exploration for bleeding Deaths

No. of patients 31 1 1 I 3 3

Patients The extensive preoperative evaluation of patients with accessory pathways of atrioventricular conduction is done to determine the need for operation and to locate the pathway. These electrophysiological examinations have been described in another report from this clinic." Left free wall pathways in the posterior sector and part of the lateral sector are accessible for study since catheter electrodes can be placed in the coronary sinus, the limitations being the course of the first part of the great cardiac vein. Both atrial pacing and recordings as well as left free wall ventricular recordings are possible with the catheter in this location. In this report, series numbers are sometimes used to identify patients with left free wall pathways. The number denotes their sequence in the total of 160 patients with all types of accessory pathways. The age spectrum of the 79 patients with left free wall pathways varied from IO to 72 years (Table I). Ten were between 10 and 15 years, 28 were 16 to 30 years of age, 20 were 31 to 45 years of age, 19 were 46 to 60 years of age, and two were 61 years of age or older. Only four of the IO under 15 years of age had had symptoms since infancy, and only five other patients in the remaining 69 had had symptoms all of their life. The age at onset of symptoms in this group of patients varied. In the 16 to 30 age group, mean onset of symptoms was 14 years with the oldest age at onset being 24 years. In the patients aged 31 to 45 years, the mean was 20 with the oldest 34, and in the 46 to 60 age group the mean was 30 with the oldest 49. Pathways thus can cause symptoms late in life for reasons that are not clear but possibly are due to the higher frequency of premature beats with increasing age. Thirty-five of the 79 patients (44.3%) had a pathway with the potential to cause death, as determined by a history of an episode of malignant ventricular arrhythmia or by the demonstration during preoperative studies of a pathway with a short effective refractory period and with the properties to transmit a rapid atrial rate I: I to the ventricle. Among the 82 pathways in 79 patients, 38 were found at operation to be in the posterior sector, 31 in

Fig. 4. Diagrammatic drawing of a sagittal section through the cardiacwallsshowsby the dotted lines the possiblecourse of pathways from the atrium above to the ventricle. The anulus fibrosus is shown in black with the mitral valve to the left. The anatomically or clinically proved courses, in addition, include an oblique one not represented here. These observations as well as the concept that pathways can be broad are the facts upon which the systematic dissection is based. The coronary sinus is shown with the left atrial wall. the lateral, and 13 in the anterior. The electrophysiological assessment as now carried out allows determination of the exact location of a left free wall pathway before operation in most of our patients. In the last one half of this experience, patient series 81 to 160, 12 patients were thought by electrophysiological studies to have a left free wall pathway in the lateral sector. As determined by surgical interruption, this turned out to be wrong in four and correct in eight. Nineteen patients had a pathway in the posterior sector. The location with preoperative electrophysiological studies was not specified in two, was missed in five, but was correct in 12. Precise preoperative localization in the anterior sector could not be determined because of the limitation of exploration. There were no patients with a single pathway in whom localization to the left free wall was not achieved by these studies. Thirteen patients with 13 left free wall pathways

The Journal of

7 02

Sealy and Gallagher

Thoracic and Cardiovascular Surgery

Fig. S. Operative procedure for correcting left free wall pathways. To the left and above is the left atrial cavity exposed by an incision in the right side of the left atrium. The suture in the atrium marks the external site of the pathway's entry into the left atrium. The extent of the incision is shown by the dotted line. The superior vena cava is to the left, with the left fibrous trigone marked as a solid black elipse-shaped disc on the left. The black elipse on the right is the right fibrous trigone. Inset A, which is the exposure after the atrial incision is made, demonstrates the ventricular summit below. The nerve hook is shown dividing the fibers connecting the sulcus fat to the myocardium. In B is shown the extent of the dissection of the sulcus fat from the atrium and the ventricle. In C is the last step when, with a small sharp nerve hook, all the superficial fibers of the ventricular myocardium are divided as they enter the anulus fibrosus.

were found to have only retrograde function over their pathway; thus the patient was subject to re-entry tachycardia but, of course, did not have evidence of pre-excitation on the electrocardiogram. Ten patients had more than one pathway. The preoperative studies demonstrated both pathways in six patients. Three patients had two left free wall pathways. Four patients had an additional posterior septal pathway and two had a right free wall pathway. One patient with three pathways had the additional ones in the anterior septal and right free wall. Thirteen patients had other cardiac disorders. Five patients had a prolapsed mitral valve, one with significant symptoms requiring valve replacement. Two patients had mitral stenosis and needed a valvotomy and valve replacement, respectively. Two had cardiomyopathy. Three had coronary artery disease treated with bypass grafts.

Surgical approach The procedure for interruption of left free wall accessory pathways consisted of two steps. First, epicardial

activation sequences of the atria and ventricles were determined and used to locate the crossing points of the pathways. With this as a guide, a systematic step-bystep dissection was then made for interruption of the nonvisible and nonpalpable structure. The details of the detection method have been reported previously from this clinic." Only some of the unusual problems in finding left free wall pathways will be briefly outlined. The pathways that course from the atrium to the ventricle adjacent to the left fibrous trigone may not show the earliest area of epicardial activation on the ventricle close to the pathway because of the position of the infundibulum of the right ventricle and the pulmonary artery overlying the area of the left fibrous trigone. However, atrial mapping during retrograde conduction over the pathway resulting either from re-entry tachycardia or ventricular pacing will accurately locate the crossing point. Pathways that are found in the free wall close to the crux may occasionally be confused with posterior septal pathways that cause earliest activation to the left of the crux. Differentiation between the two may be possible by noting whether or not there is a time

Volume 81 Number 5 May, 1981

lag between surface activation of the ventricle and the appearance of the delta wave. Mapping during retrograde conduction during re-entry tachycardia will help to separate the two. Because the heart must be displaced upward for mapping through a median sternotomy, left free wall pathway function, particularly in the posterior sector, can be temporarily obtunded. Such trauma may selectively obtund one of two pathways that are close together, as happened in one recent patient. After the heart is opened, retractors causing tension on the coronary sulcus may do the same. A successful operation for interruption of left free wall pathways is based upon the precise localizing of electrophysiological data which define the anatomic area harboring the pathway and then correlating this with the many possibilities of the pathway's route from the atrium to the ventricle within the early activation area. At the time the first operation was done in this series for a left free wall pathway, the description of Ohnell> of a left free wall pathway was the only one available. On the basis of earlier experience with right free wall pathways" and Ohnells description, the access to the coronary sulcus was epicardial, and the pathway was interrupted by incising its ventricular entry just below the mitral anulus. Because of possible problems with control of bleeding from the ventricle, injury to the mitral valve, and injury to the coronary vessels, this was changed to entry into the coronary sulcus by an atrial endocardial incision just above the anulus fibrosus. This latter approach was first done on a patient, series number 13 in our series of 160 patients, the sixth patient operated upon who had a left free wall pathway. Ohnell's sections show the pathway crossing through the fat in the coronary sulcus, a fact that was not at first appreciated. Subsequently, other anatomic descriptions": 8 confirmed Ohnells observation. The pathways were also noted to traverse the coronary sulcus obliquely from the atrium to the ventricle." The fact that more than one strand of myocardium was frequently present suggests that pathways could be wide and composed of multiple strands, as in the patients described by Mann,? Klein," and their associates. Several descriptions of left free wall pathways confirm the fact that the strands of myocardium can be almost on the epicardial aspect of the anulus fibrosus.!" 11 The pathways have not been found to penetrate the anulus as the His bundle penetrates the right fibrous trigone, except in Lunel's'> patients, nor have any been described that course between the anulus fibrosus and the endocardium. Fig. 4 illustrates the possible courses of free wall pathways derived from the anatomic descriptions and our clinical experience.

Left free wall accessory pathways of AV conduction

703

These proved anatomic facts and deductions made from our surgical experience led to the adoption of the following steps in the division of the pathway (Fig. 5). After the left atrium is opened wide, similar to the exposure for mitral valve replacement, the atrial end of the pathway is marked by passing a suture from the epicardium to the left atrial cavity. An atrial endocardial incision is then made 2 to 3 mm above the anulus of the mitral valve; only enough atrium to permit closure of the incision is left attached to the anulus. This avoids injury to the occasional intramural branch of the circumflex artery or to a branch which may be below the coronary sinus. The sinus is also avoided by this same maneuver. The incision is extended 1.5 to 2.0 em on each side of the suture marker. The fat in the sulcus is separated from the external aspect of the upper lip of the atrial incision. This is carried to the coronary sinus in the posterior sector pathways. The sulcus fat is separated from the superior aspect of the ventricle almost to the attachment of the epicardium to the ventricle. Avoidance of the coronary vessels is assured by always keeping the brown myocardium in view and moving the coronary sulcus fat en masse away from the myocardium. The tiny bands of connective tissue that bind the sulcus fat to the myocardium are easily divided by blunt dissection. The third step is the interruption of the superficial fibers of the ventricular myocardium just as they enter the anulus fibrosus. The epicardium is allowed to remain intact. Problems that only occur with left free wall pathways are caused by pathways that cross adjacent to the left fibrous trigone and by ones that originate from the coronary sinus. Pathways can be so close to the left fibrous trigone that the superficial fibers of the trigone have to be interrupted to be certain that the pathway in this location is divided. The coronary sinus, which begins at about the obtuse margin, contains atrial myocardium in its wall, as shown in Fig. 3. This can be the origin of pathways. The pathways that originate from the sinus may be close to the epicardium and thus course inferiorly to the ventricle at a point below the summit and toward the epicardium (Fig. 4). The other details of the conduct of the operative procedure include the use of cold cardioplegia, ascending aortic clamping, and total body hypothermia to 28° C. Exposure may be difficult because of the normal size of the left atrium. In the last 25 patients in the series, the average time of aortic occlusion was 52 minutes, and the pump time was 145 minutes. These figures exclude one patient operated upon successfully in whom the pathway was difficult to find. His pump time was 354 minutes and aortic occlusion time, 108 minutes. After closure of the atrium and rewarming of the

The Journal of

7 04 Sealy and Gallagher

Thoracic and Cardiovascular Surgery

Table III. Failures Series No.

Problem

5

19

RT, syncope

10 21

20 64

RT, syncope RT to atrial fib., congestive failure

23 67

45 50

78

24

88

38

RT, coronary artery disease

89

64

RT to atrial fib.

RT, mitral stenosis RT, mitral stenosis, coronary artery disease, retrograde function only over pathway RT, two left free wall pathways-e lat. sector and ant. sector

Current status Only one episode of RT, controlled on medication; pre-excitation still present RT controlled on medication; Pre-excitation still present No RT; retrograde function returned; His bundle interruption; pacemaker RT controlled on medication; pre-excitation; valve replacement No RT; Op. I, pathway missed, mitral valvulotomy, coronary artery bypass graft; Op. 2, His bundle interruption, pacemaker No RT: Op. I, divided one pathway; Op. 2, His bundle interruption; pre-excitation still present; sinus rhythm by ant. sector pathway; pacemaker RT controlled on medication; pre-excitation still present; coronary artery bypass RT controlled with radiofrequency pacemaker; pre-excitation still present

Legend: RT, Re-entry tachycardia.

heart, the electrocardiogram is checked for the presence of pre-excitation. Incremental pacing of the atria and ventricles is then done to demonstrate the Wenckebach phenomenon, its presence indicating that the accessory pathway has been interrupted. Finally, complete activation maps of the atria and the ventricles are repeated as further assurance that interruption has been accomplished and that other pathways are not present. These studies are done with the patient still on bypass.

Results The complications after operation are listed in Table II. Postpericardiotomy syndrome occurred frequently and was actively treated in 31 of the 79 patients. One patient had to be operated upon for constrictive pericarditis 8 months after pathway interruption. One patient had hepatitis, and another patient had acute pancreatitis possibly related to cardiopulmonary bypass, although the patient had a history of gallbladder disease. Three patients required a second operation because of bleeding, all in the first 40 series. With the endocardial atrial incision there has been no mitral valve or coronary vascular injuries. Three deaths occurred. The patient with pancreatitis eventually died, 30 days after operation. The other two, both of whom had a cardiomyopathy, died at operation or in the immediate postoperative period. Eight pathways were missed in eight patients who had nine pathways (Table III). Four of the patients were in series numbers 5 through 27 in our 160 patients. The remaining four patients were in series numbers 60 through 89. One patient had only the anterograde function of his pathway interrupted by operation, but re-

entry tachycardia recurred because retrograde conduction remained. A second operation was then done to interrupt the His bundle. One of the failures occurred in a patient with two pathways, one of which was adjacent to the left fibrous trigone. Two operations were done on this patient, and even then the second pathway was missed. However, at this same operation the His bundle was interrupted, blocking the re-entry circuit yet leaving the patient with sinus rhythm resulting from antegrade conduction over the bypass tract; however, a demand pacemaker was still required because of the uncertainty of antegrade conduction. Another patient with two pathways had a second operation, but the anterior sector pathway continued to function even after the second attempt; for this reason His bundle interruption was then done and a demand pacemaker inserted. The second pathway was considered to have been missed at operation. Eventually, conduction over the missed accessory pathway ceased. The patient is entirely dependent on her pacemaker. Three of the failures were in patients with other cardiac problems. One had mitral insufficiency necessitating valve replacement. The second, a patient who had a pathway with only retrograde function, had mitral stenosis and coronary artery disease and had to have a valvulotomy and coronary artery bypass grafts. This patient's His bundle was interrupted at a second operation. The third patient had coronary artery disease and required a bypass graft. He still has pre-excitation, but his tachycardia is controlled with drugs. There is no obvious reason why in these three patients, all with other cardiac disease, the pathway was missed. The mitral valve disease might have been a factor, although just how is not evident in review. One

Volume 81

Left free wall accessory pathways of AV conduction

Number 5 May. 1981

26 24 22

~

Totol No

m

24

705

23 23

Kent Divided

20 /8

17

~16 ~

;: 14 c c, 12 c

z 10

8 6 4

2 Series No.1 - 40

Series No.4I-80

Series No.81-120

Series NO.12H60

Fig. 6. Results of operation in 82 left free wall Kent bundles found in 79 patients separated serially into four groups. The total number given on the left bar represents the number of left free wall pathways found in each sequence of 40 patients with all types of accessory pathways.

failure occurred in a 66-year-old obese woman (number 89 in the patient series) and was due to inadequate dissection. Her tachycardia is now controlled with a radiofrequency pacemaker. Four patients had second attempts at division of the anomalous pathways. Two have already been alluded to. In two patients. the second operation was successful. One patient had a second pathway I cm from the limits of the dissection at the first operation. The other very likely had either a second pathway or a pathway composed of a wide strand of myocardium. Another patient had her first operation in another hospital. The failure was due to an inadequate dissection of coronary sulcus fat. The second operation at our hospital was successful. Counted as a success was one patient, number eight in the series, who had pre-excitation for only one year after operation, but no more episodes of tachycardia. An electrophysiological study after this time revealed that the pathway no longer functioned. Among the 10 patients with multiple pathways, three patients had their second pathways on the left free wall; two of the three had only one pathway divided following two operations, although the second pathway eventually ceased to function in one patient. The third patient also required two operations. One patient had additional right free wall and posterior septal pathways, and all three were interrupted. Another patient, series number 36, had only the left free wall pathway divided,

while the posterior septal one was electively not disturbed, as our experience at that time with septal pathways was limited. Two others had the combination of a posterior septal and left free wall pathway successfully divided. The remaining two had both left free wall pathways divided at one operation. The results for pathway interruption are recorded in Fig. 6. There have been no failures to divide pathways in the last 34 patients with left free wall pathways. Discussion These observations have permitted us to draw several conclusions about the surgical anatomy of left free wall accessory pathways of atrioventricular conduction. On the basis of the circumference of the anulus fibrosus and the interruption of the left ventricle- to-left atrial continuity by the aorta, one would expect that pathways would be more common on the right than on the left. Our experience indicates that this supposition is not true, for 79 of our 160 patients with accessory pathways had left free wall pathways. Having examined several excellent anatomic studies and carefully analyzed our surgical problems, we now believe that pathways crossing from the atrium to the ventricle can be widely dispersed through the fat of the coronary sulcus. In the posterior sector of the left free wall, the atrial origin of a pathway can be the coronary sinus. The pathways do not penetrate the anulus fibrosus as the His bundle does the right fibrous trigone, nor do they seem

The Journal of

706

Sealy and Gallagher

to cross the endocardial aspect of the anulus. Pathways can cross from the atrium to the ventricle adjacent to the left fibrous trigone and cause the earliest ventricular activation over the ventricle at a distance from the pathway's location. Patients can have multiple pathways in locations at other points on the right and left anuli as well as at more than one point on the left free wall. Pathways can be broad and composed of multiple myocardial strands. Although no anatomic differences are known, pathways may conduct in only one direction, most commonly retrograde. The analysis of the clinical experience outlined here correlated with the few anatomic studies available has led to a systematic dissection of the crossing point of the pathway from the atrium to the ventricle, as illustrated in Fig. 5. These points are (I) atrial endocardial incisions of liberal length, (2) extensive separation of the coronary sulcus fat from the atrial and ventricular myocardium, and (3) division of the superficial ventricular myocardial fibers that enter the anulus fibrosus. The certainty with which left free wall pathways can be divided has led to expansion of the indications for operation. This is a congenital cardiac disorder, and the risk of operation in patients who do not have other cardiac problems is similar to the surgical risk for other simple congenital cardiac anomalies. In the group in which pathways have the potential for causing death, there is now no doubt about the urgent need for operation. Now, in addition, surgical correction is offered to young patients with symptoms, who need daily medication for control of their arrhythmias. Older patients, in whom medical treatment is not satisfactory because of difficulty with patient compliance, current or potential drug toxicity, lack of drug response, and drug expense are now also being offered operation. REFERENCES James TN: Anatomy of the Coronary Arteries. New York, 1961, Paul B. Hoeber, Inc.

Thoracic and Cardiovascular Surgery

2 McAlpine WA: Heart and Coronary Arteries. New York, Heidelberg, Berlin, 1975, Springer-Verlag 3 Gallagher JJ, Gilbert M, Svenson RH, Sealy WC, Kasell J, Wallace AG: Wolff-Parkinson-White syndrome. The problem, evaluation, and surgical correction. Circulation 51:767-785, 1975 4 Gallagher JJ, Kasell J, Sealy WC, Pritchett ELC, Wallace AG: Epicardial mapping in the Wolff-Parkinson-White syndrome. Circulation 57:854-866, 1978 5 Ohnell RF: Patho-anatomical observations. How does auricular excitation elicit the "pre-excitation"? A. Post mortems in WPW and pre-excitation cases. Acta Med Scand Suppl 152:74-93, 1944 6 Sealy WC, Wallace AG: Surgical treatment of Wolff-Parkinson-White syndrome. J THoRAc CARDIOVASC SURG 68:757-770, 1974 7 Mann RB, Fisher RS, Scherlis S, Hutchins GM: Accessory left atrioventricular connection in type A WolffParkinson-White syndrome. Johns Hopkins Med J 132: 242-249, 1973 8 Klein GJ, Hackel DB, Gallagher JJ: Anatomic substrate of impaired antegrade conduction over an accessory atrioventricular pathway in the Wolff-Parkinson-White syndrome. Circulation 61: 1249-1256, 1980 9 Brechenmacher C, Latham J, Iris L, Gerbaux A, Lenegre J: Etude histologique des voies anormales de conduction dans un syndrome de Wolff-Parkinson-White et dans un syndrome Lown-Ganong-Levine. Arch Mal Coeur 67: 507-519, 1974 10 Brechenmacher C, Coumel P, Fauchier JP, Cachera JP, James TN: De subitaneis mortibus. XXII. Intractable paroxysmal tachycardia which proved fatal in type A Wolff-Parkinson-White syndrome. Circulation 55:408417, 1977 II Becker AE, Anderson RH, Durrer D, Wellens HJJ: The anatomical substrates of Wolff-Parkinson-White syndrome. A clinicopathologic correlation in seven patients. Circulation 57:870-879, 1978 12 Lunel AAV: Significance of anulus fibrosus of heart in relation to AV conduction and ventricular activation in cases of Wolff-Parkinson-White syndrome. Br Heart J 34: 1263-1271, 1972