Loss of Sinus Rhythm After Total Cavopulmonary Connection
Background Total cavopulmonary connection (TCPC) to repair functional single ventricle involves the sinus node area, in contrast to the Fontan procedure. We compared ECG findings after TCPC and Fontan to evaluate the impact of the cavopulmonary connection on sinus rhythm postoperatively.
Methods and Results The Fontan group consisted of 17 patients repaired at 7.8±3.1 years of age (mean±SD): 11 for tricuspid or pulmonary atresia (TA/PA) and 6 for single ventricle. The TCPC group consisted of 19 patients repaired at 5.1±3.2 years of age (mean±SD) (P<.001): 9 for TA/PA, 4 for single ventricle, and 6 for hypoplastic left heart syndrome. Mean follow-up after Fontan was 7.7±2.7 years versus 2.8±1.6 years for TCPC (P<.001). Preoperative ECGs on all TCPC patients showed sinus rhythm (SR), whereas 16 of 17 Fontan patients had SR and one had nonsinus atrial rhythm (NSAR) since birth. On the first postdischarge ECG, 12 of 19 TCPC patients (63%) were in SR, 4 were in junctional rhythm (JR), and 3 were in NSAR. In comparison, 15 of 17 Fontan patients (88%) were in SR with 1 of 17 in NSAR and 1 in supraventricular tachycardia (P<.05 with χ2 test). By 2 years postoperatively, only 6 of 15 TCPC patients available for follow-up (40%) were in SR, with 7 of 15 in JR and 2 of 15 in NSAR. By contrast, 13 of 17 Fontan patients (76%) remained in SR, with 1 in NSAR and 3 in JR (P<.05 with χ2 test). TCPC patients with loss of SR did not differ from other patients in the group in age at repair, preoperative diagnosis, or surgeon performing the procedure.
Conclusions This significant incidence of loss of SR temporally related to surgery suggests that operative compromise of the sinus node area is common with TCPC.
In recent years, TCPC has replaced the traditional Fontan operation as the procedure of choice for repair of functional single ventricle.1 2 One important difference between these two surgical approaches is the proximity to the sinus node area with TCPC. Historically, surgery for congenital heart disease involving this area has resulted in clinically important bradyarrhythmias due to sinus node dysfunction.3 4 5 In this study, we compared ECG findings after TCPC and Fontan to evaluate the impact of the cavopulmonary connection on sinus rhythm postoperatively.
The records of all hospital survivors of right atrial to pulmonary artery connections who were followed up by the division of Pediatric Cardiology at the State University of New York Health Science Center were reviewed. Seventeen patients were followed up after traditional Fontan and 19 after TCPC. Charts were reviewed for reported arrhythmias and for clinically important events.
For both groups, all ECGs from initial diagnosis to last available postoperative tracing were reviewed. The proximate preoperative tracing, the first postdischarge tracing, the 2-year postoperative tracing, and the last available ECG were analyzed for cardiac rhythm, P wave axis, and P wave morphology.
Ambulatory 24-hour ECGs were analyzed for basic rhythm and for the presence of arrhythmia. Sinus rhythm was diagnosed when present for >90% of the 24-hour period. JR was diagnosed when present for >10% of the 24-hour period. Atrial and ventricular ectopic beats (>50 per hour) and all episodes of arrhythmia were recorded.
Univariate comparisons of descriptive data in the two groups were made with Student’s t tests. Differences in proportion between the two groups were analyzed by use of χ2 analysis.
The study population represents all hospital survivors referred by our division of pediatric cardiology for either procedure between 1982 and 1994. Fontan procedures were performed between 1982 and 1990 by any of three pediatric cardiovascular surgeons. The TCPC group underwent repair between 1989 and 1994. The 6 children with HLHS had all surgery performed at Children’s Hospital of Philadelphia. In the remainder of the TCPC group, surgery was performed by either of two pediatric cardiovascular surgeons. Findings were therefore independent of surgeon or institution.
Clinical characteristics of the patients in the two groups are summarized in Table 1⇓. No Fontan patient had an underlying diagnosis of HLHS, whereas this diagnosis was associated with 31% of those undergoing TCPC. Prior palliative procedures in both groups are shown in Table 2⇓. At ultimate palliation, 2 patients in the TCPC group had placement of an intact intracardiac conduit from inferior vena cava to right pulmonary artery, whereas in the remaining patients, a lateral expanded polytetrafluoroethylene channel including a portion of the right atrial posterior wall was created. TCPC patients were significantly younger than Fontan patients both at repair and at postoperative evaluation.
Findings on the last outpatient ECG before surgery are shown in Table 3⇓ and Fig 1⇓. All 19 TCPC patients were in sinus rhythm. Sixteen of 17 Fontan patients were in sinus rhythm, and 1 was in NSAR, which had been present since birth. One Fontan patient had manifested intermittent JR on office ECGs before open repair. No NSAR or JR was documented preoperatively on standard ECGs in any TCPC patient.
The first ECG after hospital discharge was recorded a mean of 28 days after TCPC and 24 days after Fontan (P=NS). ECG findings are summarized in Table 3⇑ and Fig 1⇑. Fifteen of 17 Fontan patients (88%) were in sinus rhythm, with one in NSAR; the child with preoperative NSAR presented with supraventricular tachycardia on the first postoperative visit. In comparison, only 12 of 19 TCPC patients (63%) were in sinus rhythm, with 3 in NSAR and 4 in JR (P<.05 with χ2 test) (Fig 2⇓). One patient with tricuspid atresia and 2 with single ventricle presented with NSAR after surgery; JR occurred in 1 patient each with tricuspid and pulmonary atresia and 2 with HLHS.
Analysis of P-wave morphology indicated a significant decrease in P-wave amplitude for those patients in sinus rhythm after TCPC (Fig 3⇓). This difference was statistically significant when compared with preoperative tracings (P<.001). No significant change in P-wave amplitude was noted after Fontan (Fig 4⇓). P-wave changes are summarized in Table 4⇓.
Intermediate-Term ECG Findings: 2 Years Postrepair
Fifteen TCPC patients were followed up for at least 2 years; 1 patient with chronic pleural effusion and hypogammaglobulinemia died of sepsis 5 months after surgery, and 3 patients have been followed up for 1 to 16 months. All 17 Fontan patients were followed up for more than 2 years. ECG findings at 2 years after repair are summarized in Fig 5⇓. Standard ECGs in only 6 of the 15 available TCPC patients (40%) showed sinus rhythm, compared with 13 of 17 Fontan patients (76%) (P<.05 with χ2 test). In the TCPC group, 2 patients who had NSAR early after repair now had JR, 1 patient with early sinus rhythm now had NSAR, and 1 patient with early sinus rhythm now had JR with recurrent atrial flutter. In the Fontan group, 3 patients had developed JR on standard ECGs, including the child with intermittent JR on preoperative tracings; the patient with NSAR since birth remained in this rhythm. Low P-wave amplitude persisted in the TCPC patients (Figs 2⇑, 3⇑, and 4⇑).
No further change in ECGs was recorded in the 13 TCPC patients followed for an additional 1 to 36 months (mean, 18 months) beyond the initial 2-year follow-up period. In the Fontan group, all patients were followed up for a mean of 70 months (range, 12 to 132 months) beyond the initial 2-year period. Over that time period, 1 additional patient developed JR 11 years after repair; 1 with JR developed recurrent atrial flutter/fibrillation and died suddenly 8 years after Fontan; 1 required transplant for intractable protein-losing enteropathy 4 years after Fontan; and 1 died with respiratory failure 3 years after Fontan. On the last ECG, recorded a mean of 9.2 years postrepair, 11 of 14 Fontan survivors (79%) were in sinus rhythm, 1 remained in NSAR, and 2 were in JR. In comparison, the last ECG on 18 of 19 TCPC survivors showed sinus rhythm in only 6 (33%) despite the significantly shorter mean follow-up period of 2.8 years.
Fourteen of 19 TCPC patients and 15 of 17 Fontan patients have undergone at least one postoperative ambulatory ECG. Cardiac rhythm findings are summarized in Fig 6⇓. In no patient did the ambulatory ECG document a rhythm pattern not recorded on standard ECG. In 1 TCPC patient with JR, sinus arrest with prolonged pauses greater than 3 seconds was recorded during sleep after TCPC (Fig 7⇓).
Clinical tachyarrhythmias are recorded in Table 5⇓. One TCPC patient with JR has had recurrent refractory atrial flutter. Four Fontan patients have experienced episodes of atrial flutter/fibrillation; one of these four patients died suddenly 8 years after surgery. One additional Fontan patient has had recurrent supraventricular tachycardia.
This study documents loss of sinus rhythm in the immediate postoperative period in 37% of patients after TCPC. By 2 years postrepair, only 40% of TCPC patients remained in sinus rhythm compared with 76% of Fontan patients. In contrast to Fontan, the TCPC involves the sinus node area. In previous years, the Mustard procedure, which also involved the sinus node region, was found to be associated with similar postoperative rhythm changes and with a significant incidence of late sudden death.3 4 Electrophysiologically, the post-Mustard arrhythmias were shown to be related to sinus node dysfunction.5 The temporal relation of the rhythm change to surgery in this series suggests that damage to the sinus node area may be the cause of the rhythm changes after TCPC. No TCPC patient in this series has had significant clinical problems with bradycardia related to these rhythm changes, but symptoms including sudden death developed late postoperatively in the Mustard group, and duration of follow-up for this study group is still short.3 6
The TCPC was developed to reduce right atrial pressures and volumes known to contribute to tachyarrhythmias that occur in as many as 50% of patients after Fontan.7 8 9 Hemodynamic results early after TCPC have been favorable, and tachyarrhythmias in the early postoperative period have occurred at lower frequency.10 In this series, postoperative tachyarrhythmias in the intermediate term also occurred at a lower frequency after TCPC than after Fontan, although this difference did not reach statistical significance. No other study documenting loss of sinus rhythm after TCPC has been reported, but most series have looked primarily at the immediate postoperative period,11 focused on tachyarrhythmias10 and/or included a combination of Fontan and TCPC patients.9 11 References to sick sinus syndrome10 11 and to the need for pacemakers because of sinus node dysfunction12 have appeared in reports of surgical results. In our series, loss of sinus rhythm was not associated with any specific diagnosis, younger age at repair, or any one of the five surgeons who performed the operations. The proximity of the sinus node area to the TCPC repair represents a biologically plausible basis for the implication of the surgical repair itself in causation.
The decrease in P-wave amplitude recorded on ECGs in patients after TCPC is also reminiscent of changes seen after Mustard. The etiology of this is unknown, but it may be related to alterations in conduction in the presence of intra-atrial prosthetic material and/or to extensive atrial suture lines. After Mustard, low P-wave voltage often preceded loss of sinus rhythm, and rhythm abnormalities occurred at increased incidence as duration of follow-up increased.13 This association and the progressive loss of sinus rhythm in the first 2 years after surgery in this series suggest that prospective follow-up for atrial conduction abnormalities as well as tachyarrhythmias is important after TCPC.
In patients with functional single ventricle, TCPC confers important hemodynamic advantages over the traditional Fontan procedure. If the findings in this series are documented in a larger group of patients, consideration should be given to modification of the TCPC procedure to avoid the sinus node area.
Selected Abbreviations and Acronyms
|HLHS||=||hypoplastic left heart syndrome|
|NSAR||=||nonsinus atrial rhythm|
|TCPC||=||total cavopulmonary connection|
The authors are grateful to Ellen Percival for her secretarial help in preparation of the manuscript.
Presented in part at the 67th Scientific Sessions of the American Heart Association, Dallas, Tex, November 14-17, 1994, and published in abstract form (Circulation. 1994;90[pt 2]:I-421).
- Copyright © 1995 by American Heart Association
Scott LP, Saalouke MG, Shapiro SR, Rios JC, Perry LW. Sudden unexpected death following Mustard’s procedure for d-transposition of the great arteries. Circulation. 1976;54(suppl II):II-89. Abstract.
Gewillig M, Wyse RK, deLeval MR, Deanfield JE. Early and late arrhythmias after the Fontan operation: predisposing factors and clinical consequences. Br Heart J. 1992;67:72-79.
Balaji S, Gewillig M, Ball C, deLeval MR, Deanfield JE. Arrhythmias after the Fontan procedure: comparison of total cavopulmonary connection and atriopulmonary connection. Circulation. 1991;84(suppl III):III-162-III-167.
Kurer CC, Tanner CS, Norwood WI, Vetter VL. Perioperative arrhythmias after Fontan repair. Circulation. 1990;82(suppl IV): IV-190-IV-194.
Farrell PE, Chang AC, Murdison KA, Baffa JM, Norwood WI, Murphy JD. Outcome and assessment after the modified Fontan procedure for hypoplastic left heart syndrome. Circulation. 1992;85:116-122.