| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Ann Thorac Surg 2005;80:2293-2300
© 2005 The Society of Thoracic Surgeons
Cardiothoracic Centre, All India Institute of Medical Sciences, New Delhi, India
Accepted for publication May 17, 2005.
* Address correspondence to Dr Chowdhury, Department of Cardiothoracic Surgery, All India Institute of Medical Sciences, New Delhi, 110029, India (Email: ujjwalchow{at}rediffmail.com).
 |
Abstract |
|---|
 Top Abstract IntroductionPatients and MethodsResultsCommentAcknowledgmentsReferences |
|---|
METHODS: Eighty-four patients undergoing one and one-half ventricular repair between January 1990 and December 2003 were studied. Age was 4 to 504 months (mean, 47.9 ± 57.3 months). Sixty-nine survivors underwent serial echocardiography, radionuclide studies, cardiac magnetic resonance imaging, and cardiac catheterization.
RESULTS: Operative and late mortality were 10.7% and 8%, respectively. Perioperative and postoperative supraventricular arrhythmias were observed in 14.3% and 15.9% of patients, respectively. Risk factors for supraventricular arrhythmias included systemic ventricular dysfunction, heterotaxy syndrome, and Ebstein's anomaly. Mean late postoperative superior vena caval pressure was 14.2 ± 1.52 mm Hg and right atrial pressure was 6.6 ± 0.74 mm Hg. At a median follow-up of 87 months, actuarial survival was 81.9% ± 0.04%, and 89.8% were in New York Heart Association class I or II. Serial cine–magnetic resonance imaging demonstrated significant growth of tricuspid valve and right ventricular cavity in 45% of patients.
CONCLUSIONS: One and one-half ventricular repair can be performed with an acceptable risk. The operation maintains a low pressure in the inferior vena caval tributaries, and reverses the Fontan paradox. Patients with tripartite right ventricles demonstrated a tendency toward enlargement of the pulmonary ventricular chamber commensurable with somatic growth.
|
Introduction |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentAcknowledgmentsReferences |
|---|
|
Patients and Methods |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentAcknowledgmentsReferences |
|---|
(–) 2, McGoon's ratio 2.0, or Nakata index 250 mm/m2) were considered mandatory [2].
|
Physiologic criteria [2] include the following:
Patient Characteristics
Original cohort
Eighty-four consecutive patients (58 males) underwent 1.5VR with pulsatile bidirectional Glenn (BDG) between 1990 and 2003 at our institution. In 79 patients, 1.5VR was planned preoperatively, and in 5 patients, this was done as a salvage procedure to treat acute RV dysfunction after attempted biventricular repair.
The early and midterm outcome of the first 50 patients have been outlined previously [2]. Age ranged from 4 to 504 months (mean, 47.9 ± 57.3 months; median, 36.0 months). Sixty-two patients (73.8%) were younger than 4 years of age (Table 2). Previous interventions are detailed in Table 3. Nine patients with unbalanced complete atrioventricular canal and 2 patients with ventricular septal defect and straddling tricuspid valve had heterotaxy syndrome with persistent left superior vena cava (SVC). The rare variants of tetralogy of Fallot (n = 11) presented late with extreme RV hypertrophy leading to an inlet obstruction, associated tricuspid stenosis, and hypoplastic tricuspid valve. Four of them had heterotaxy syndrome with persistent left SVC. Preoperative RV systolic pressure ranged from 30 to 110 mm Hg (mean, 50 ± 15 mm Hg). Seven patients had mean PA pressure more than 18 mm Hg and had dominant left-to-right shunt. Because of failure of PA entry, pulmonary vascular resistance could not be obtained in all patients.
|
|
Age at follow-up ranged from 18 months to 50 years (mean, 131.1 ± 83.5 months). Follow-up was limited to 69 patients. Postoperative evaluation consisted of quarterly clinical examination, electrocardiogram, 24-hour Holter, two-dimensional echocardiography, radionuclide angiography, magnetic resonance imaging (MRI), and cardiac catheterization.
Preoperative transthoracic two-dimensional color-flow Doppler echocardiography was performed using a Hewlett-Packard-Sonos 5500 with 2.7/3.5-MHz transducer. The largest diameter of the tricuspid valve (indexed) was compared with the normal values of King and associates [9], and Z values were computed. The RV volume was also assessed before repair with a modified Tomita method (Table 1) [2, 5, 10]. At operation, the RV was confirmed to be hypoplastic. Z values for the tricuspid valve were determined using data tables [2, 9].
Surgical Techniques
Previous systemic-to-pulmonary artery shunts were interrupted. The SVC and IVC were cannulated directly. A BDG was constructed on cardiopulmonary bypass on a beating, perfused heart. The azygos or hemiazygos vein was ligated. Fifteen patients with bilateral SVC underwent bilateral BDG. Entire repair including BDG, intracardiac repair, and RV outflow tract reconstruction was performed as a single-stage procedure in 79 patients. Five patients had conversion to 1.5VR to manage RV dysfunction. Concomitant cardiac procedures are tabulated in Table 4. Antegrade cold blood cardioplegia was used for intracardiac repair.
|
In our early experience, we placed an RV-to-PA homograft conduit in 3 patients with variants of tetralogy of Fallot with tricuspid valve and RV hypoplasia, and anomalous left anterior descending artery crossing the RV outflow tract, and in 1 patient with pulmonary atresia, intact ventricular septum [2].
Atrioventricular septal defects were repaired using standard two-patch technique [2, 4]. All had postrepair right atrioventricular valve diameters that were at least two standard deviations less than the normal expected value.
We used an adjustable atrial septal fenestration earlier in patients with tricuspid valve diameter less than 50% of normal and in selected patients with PA hypertension. After weaning from bypass, the snare was adjusted, keeping the right atrial pressure between 12 and 15 mm Hg [2]. Since 2000, a 4-mm fenestration is left open in all patients (n = 34).
Five patients with Ebstein's anomaly had concomitant radioablation (2 with Wolff-Parkinson-White syndrome, and 3 with atrial reentry tachycardia).
Postoperative Assessment
The patients' clinical course (survival, need for cardiac medications, late complications) was monitored. The outcome based on New York Heart Association class was then collapsed to form two groups, "good" functional status (class I or II, n = 62) or "poor" functional status (class III or IV, n = 7).
Postoperative echocardiographic studies were performed according to American Society of Echocardiography Criteria [11]. Variables studied included atrioventricular valve regurgitation, flow through the fenestration, biventricular size and function, and tricuspid valve diameter. Superior vena cava and IVC Doppler interrogation was done to determine the pattern of flow during the cardiac cycle.
Twenty-four–hour Holter studies were obtained using a Mortara H-12 Holter unit (Mortara Instruments Inc, Milwaukee, WI). All 69 survivors underwent Holter monitoring in the closing interval. However, if they had symptoms or a 12-lead electrocardiogram showed dysrhythmias, they underwent this earlier. The diagnosis of sinus node dysfunction and arrhythmias were made using standard criteria [2, 5, 10].
Radionuclide studies were performed in 69 patients as per standard protocol [12]. A combination of first-pass and equilibrium radionuclide angiocardiography was used to assess systemic or pulmonary venous pathway obstruction and ventricular function, respectively. Patients with supraventricular arrhythmias (n = 20) were digitalized, and the method of online postbeat filtration was used.
Magnetic resonance imaging was performed using a 1.5-T Siemens Sonata system (Siemens Medical Solutions, Erlangen, Germany) in 69 patients. The standard imaging protocols and sequences were used. Phase-contrast short-axis cine images were used for ventricular volume and myocardial mass analysis. Velocity-encoded phase-contrast sequence was used for direct measurement of forward and reverse flow in the main PA.
Left ventricular and RV end-diastolic and end-systolic volumes, mass, and ejection fraction were analyzed as described by Lorenz [13]. Quantification of flow rates and calculations of pulmonary regurgitation were performed as described by Powell and colleagues [14].
Only 40 patients consented for postoperative cardiac catheterization. This showed a mean mid-SVC pressure of 14.2 ± 1.5 mm Hg (range, 12 to 17 mm Hg). Mean right atrial pressure ranged between 6 and 8 mm Hg (mean, 6.6 ± 0.7 mm Hg), and RV systolic pressure ranged between 28 and 38 mm Hg (mean, 30 ± 6.2 mm Hg). There was phasic SVC flow reversing in systole with a mean peak pressure of 17 mm Hg (range, 16–20 mm Hg). There were no venovenous collaterals.
Statistical Analysis
Data were analyzed with SPSS 10.0 statistical package (SPSS Inc, Chicago, IL). Interval-related data was expressed as the mean ± standard deviation. The 
2 test with Yates' correction was used to analyze the association between arrhythmia and related categorical variables.
Logistic regression was used to represent supraventricular arrhythmias using related categorical variables. There were few events of supraventricular arrhythmias in this study population, and data analysis did not reveal any variable with statistical significance. Hence, the multivariate data for the above variables are not presented. Two-tailed probability was used for all statistical tests. Paired Student's t test was used to compare preoperative and postoperative RV volume measurements. A p value of less than 0.05 was considered statistically significant. Analysis of time-related survival was performed using the Kaplan-Meier method.
|
Results |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentAcknowledgmentsReferences |
|---|
The remaining 75 patients had an uncomplicated course. The arterial, right atrial, and SVC pressures were continuously monitored. Supraventricular arrhythmias were present in 14.3% of patients. There was significant pleural effusion in 22.7% of survivors. All patients were routinely started on oral angiotensin-converting enzyme inhibitors. Digoxin, diuretics, and angiotensin-converting enzyme inhibitors were discontinued within 6 months in the majority of patients (n = 62).
Late Mortality and Morbidity
There were 6 late deaths (8%) attributable to new-onset ventricular arrhythmias (n = 4), pulmonary infection (n = 1), and cerebral abscess (n = 1). Of the children dying of ventricular arrhythmias, 2 had pulmonary atresia and intact ventricular septum, and 2 had Ebstein's anomaly.
Sixty-nine survivors have been followed for 12 to 178 months (mean, 87.7 ± 56.6 months). Actuarial survival was 81.9% ± 0.04% at a median follow-up of 87 months (Fig 1). Sixty-two patients (89.8%) were in New York Heart Association functional class I and II. Only 7 patients (10.2%) were on diuretics and vasodilators late postoperatively and were in class III.
|
Holter studies demonstrated normal sinus rhythm (n = 51), first-degree atrioventricular block (n = 6), left bundle-branch block (n = 5), complete right bundle-branch block (n = 3), incomplete right bundle-branch block (n = 3), and complete heart block (n = 1).
Perioperative and late postoperative supraventricular arrhythmias were observed in 14.3% (12 of 84 patients) and 15.9% (11 of 69 patients) of patients, respectively. Risk factors for supraventricular arrhythmias were heterotaxy syndrome with anomalous systemic venous drainage [p < 0.001; relative risk, 7.16 (95% confidence intervals (CI) 3.83 to 13.31)], systemic ventricular dysfunction [p < 0.001; relative risk, 6.08 (95% CI 3.63 to 10.20)] and Ebstein's anomaly [p < 0.001; relative risk, 6.41 (95% CI 3.22 to 12.78)]. Sinus pause with junctional escape developed in 3 patients with bilateral SVC requiring enlargement of cavopulmonary connections.
Echocardiographic Evaluation
The left ventricular function was normal (ejection fraction > 0.5, n = 62), and mildly depressed (ejection fraction, 0.45 to 0.5, n = 7). Thirty-two patients (46.3%) had mild and 6 (8.7%) had moderate pulmonary regurgitation. Right ventricular outflow tract gradient ranged between 15 and 18 mm Hg. Thirty patients (43.4%) had mild and 4 patients (5.8%) had moderate tricuspid regurgitation. Postoperatively, only 42% of late survivors (29 of 69) demonstrated significant change in the tricuspid valve diameter and RV volume.
Doppler echocardiogram confirmed unobstructed flow through the BDG, and a pulsatile flow from the RV to the PAs. Superior vena cava Doppler showed systolic reverse flow simultaneous to RV ejection and an antegrade flow during the rest of the cardiac cycle in all patients. Inferior vena cava Doppler interrogation disclosed a normal pattern with low-amplitude reverse flow during atrial systole and an antegrade flow during the rest of the cardiac cycle.
Radionuclide Studies
The mean radiotracer transit time on first-pass angiography through the hypoplastic RV was 8.20 ± 2.42 seconds (range, 5 to 11 seconds; n = 28). All patients with poor functional status (n = 7) and those with echocardiographically documented tricuspid regurgitation (n = 34) had hepatic vein reflux, and the radiotracer remained visible in the right atrium for 14 to 26 seconds (mean, 17.60 ± 2.80 seconds).
Postoperatively, systemic ventricular ejection fraction of patients with good functional status (n = 62) was 0.45 to 0.62 (mean, 0.51 ± 0.04). Patients with poor functional status (n = 7) had low ejection fraction ranging between 0.36 and 0.39 (mean, 0.37 ± 0.02). The preoperative and postoperative differences of ejection fraction of both groups of patients were statistically significant (p < 0.001).
Magnetic Resonance Imaging Results
Magnetic resonance imaging was diagnostic in 60 patients (9 were excluded because of coil embolization–related artifacts). Thirty-eight patients with transannular patch had evidence of pulmonary regurgitation and 34 patients had evidence of mild tricuspid regurgitation.
The mean pulmonary regurgitant fraction derived from velocity-encoded phase-contrast sequence–magnetic resonance imaging was 30.69% ± 4.0% (range, 22.0% to 35.2%) in 32 patients and 42.5% ±3.7% (range, 38.0% to 46.4%) in 6 patients. Before 1.5VR, mean indexed RV end-diastolic volume was 25.0 ± 2.2 mL/m2 (range, 20 to 28 mL/m2) and mean indexed RV end-systolic volume was 11.35 ± 0.65 mL/m2 (range, 10 to 12 mL/m2, n = 60). After operation, the RV end-diastolic volume increased to 30 to 35 mL/m2 (mean, 32.26 ± 1.6 mL/m2) in 33 patients and 37 to 40 mL/m2 (mean, 38.65 ± 1.10 mL/m2) in the other 27 patients. Postoperatively, the RV end-systolic volume increased to 22 to 28 mL/m2 (mean, 25.4 ± 2.0 mL/m2) in 33 patients and 28 to 38 mL/m2 (mean, 35.46 ± 3.2 mL/m2) in the other 27 patients.
|
Comment |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentAcknowledgmentsReferences |
|---|
Optimal hemodynamics in normal circulation consists of a higher PA pressure (mean approximately 15 mm Hg) and a lower caval pressure (mean < 10 mm Hg) to keep the pulmonary vasculature patent. The paradox of the Fontan circulation is that it produces caval hypertension and PA hypotension. Hemodynamically, a device capable of producing a stepdown in pressure of 5 mm Hg in the IVC and a step up in pressure of 5 mm Hg in the pulmonary arteries would reverse this paradox. Incorporating a hypoplastic ventricle in the pulmonary circulation can theoretically restore ventriculoarterial coupling, maintaining a low pressure in the IVC district.
In this study, a pressure difference greater than 6 mm Hg was noted between the SVC and the right atrium. Pressure gradients between the systemic veins could produce collaterals through venovenous shunts [5, 6, 10]. No venovenous collaterals or pulmonary arteriovenous malformations were identified so far. Although we routinely ligate the azygos vein, the issue is debatable and will be apparent on late follow-up [2]. The postoperative hemodynamic data thus compare favorably with the mean SVC and IVC pressure (14 mm Hg) reported after a Fontan operation, and 1.5VR physiology appears superior to Fontan physiology [12, 15].
Doppler echocardiogram and catheterization revealed phasic SVC flow reversing in systole with a mean peak pressure of 17 mm Hg. Right PA banding has been advocated to decrease the excessive pulsatility in the SVC and to avoid the risk of aneurysmal SVC [7]. There was cessation of reverse flow at long-term follow-up. We have not performed the hybrid type of pulmonary circulation described by Gentles and coworkers [7].
Previous reports have largely concentrated on establishing the anatomic criteria for 1.5VR [1, 3–8]. Right ventricular size ranges from ventricles that are unsuitable to those that are nearly normal. Within this range is a bandwidth of ventricles that are capable of dealing with venous return from the IVC alone [2]. As pointed out by Hanley [8], we should clarify policy and criteria for choosing this intermediate procedure.
Morphologic defects may be defined by tricuspid valve Z value, which reflects the corresponding ventricular volume [1–7]. In our earlier publication, we demonstrated that severe RV hypoplasia with a tricuspid valve Z score lower than –4.8 and mild RV hypoplasia with tricuspid valve Z score higher than –1.5 would predicate a Fontan type or a two-ventricle repair, respectively [2].
Even in the setting of 1.5VR, the hypoplastic RV needs to be reasonably functioning to forward IVC blood with low atrial pressure. The failures in our initial experience indicated that patients with borderline ventricular function, and those requiring intricate intracardiac repair, may be better suited for a univentricular type of repair to reduce the complexity of the procedure [2].
Results for this procedure when used as a salvage operation for RV dysfunction were unsatisfactory. A borderline, dysfunctional, and dilated RV was an indication for this adjunctive approach. Although extracorporeal membrane oxygenation is a useful tool to manage severe postoperative RV failure, it was not used because of the impression that the RV is inadequate to handle the biventricular repair. We feel that decision-making for 1.5VR repair should be done preoperatively.
Only 45% of survivors demonstrated significant growth of the tricuspid valve and RV. These patients are borderline candidates for a biventricular repair and had tripartite RV. However, none have qualified for conversion to biventricular repair so far. The insignificant growth of the tricuspid valve annulus and RV in the great majority is probably because of the existence of dysplastic tricuspid valve and RV. The impact of pulmonic insufficiency in the setting of 1.5VR repair is unclear [2–8]. In our experience, patients with mild pulmonary regurgitation seem to have a smooth postoperative course [2]. However, long-term effects will be apparent only on longer follow-up. Serial cine-MRI evaluation may be of particular importance in this regard.
We added a BDG for selected patients undergoing repair of Ebstein's anomaly (Carpentier's type C and D) to reduce the preload to a dysfunctional RV. This approach allowed hypoplastic and dysplastic RV to adequately handle the reduced preload. Additionally, this enabled a more aggressive tricuspid valve reconstruction downsizing the tricuspid valve annulus to a Z value less than 0 without hemodynamic compromise. In patients with Ebstein's anomaly and dilated right atrium and RV, this operation has a favorable impact on arrhythmias and ventricular function [4–7]. Others have used it as an alternative or a bridge to heart transplantation in patients with severe RV dysfunction [6].
Supraventricular arrhythmias have an unfavorable impact on patients undergoing 1.5VR. Their incidence is 12% to 30% [2–8, 10]. Abnormalities of the conduction system are known to exist in patients with heterotaxy syndrome and may provoke arrhythmia irrespective of the type of surgical repair [1–7, 10, 12, 15].
In our study, patients with systemic ventricular dysfunction, heterotaxy syndrome, and Ebstein's anomaly had a higher incidence of arrhythmia. As a treatment of elevated right atrial pressure or supraventricular arrhythmias subsequent to 1.5VR, we performed atrial septal fenestration. In patients with a failing 1.5VR or high right atrial pressure, the other option is to convert them to BDG alone. None underwent conversion from 1.5VR to Fontan as advised by others [6, 10].
Atrial septal fenestration in patients undergoing 1.5VR is controversial [1–8, 10, 15]. Forty-four of 84 patients in this study (n = 10, before 2000 publication [2]; and n = 34, after 2000) had concomitant atrial septal fenestration. Four patients in our earlier experience who died early postoperatively of low cardiac output had high right atrial pressure and would probably have benefited from an elective fenestration [2]. Presently, we perform atrial septal fenestration in patients with tricuspid valve between one third and one half of normal.
With this strategy, the clinical outcome has improved, and no long-term survivors developed pleural effusion, ascites, or protein-losing enteropathy. No patient with fenestration had significant desaturation or a cerebrovascular accident.
This study highlights the importance of quantification of pulmonary regurgitation and quantitative evaluation of RV and left ventricular dimension and function with echocardiography, MRI, or radionuclide studies. Varying ventricular morphologies in the univentricular cohort make the calculation of ejection fraction by conventional angiocardiograms unreliable [11–13]. Gated blood pool scintigraphy is relatively independent of geometric assumptions, but spatial resolution is poor in small children, and the atrial blood pool cannot be clearly separated from the ventricular region [12].
Multisection gradient-echocardiographic cine-MRI is probably more accurate because it is noninvasive and unaffected by acoustic windows, and it measures ventricular dimensions and functions independently. However, it may overestimate pulmonary regurgitant fraction in patients with combined pulmonary and tricuspid regurgitation [14, 16]. Therefore, we used velocity-encoded cine-MRI for accurate quantification of pulmonary regurgitation.
Study Limitations
The present study provides some of the longest follow-up information after 1.5VR, but the mean age of patients at last follow-up was only 131.1 ± 83.5 months. Clearly, these data, though encouraging, may not provide reliable projections of late survival.
Exercise testing and MRI in older patients are in progress to examine whether 1.5VR actually augments the cardiac output compared with a Fontan-type connection. It remains to be determined whether the peaks of venous pressure produced with pulsatile assistance result in the late sequelae of chronic systemic venous hypertension.
Conclusions
A 1.5VR can be performed with an acceptable risk provided the anatomy permits and the physiologic criteria appear similar to those for Fontan candidates. The operation maintains a low pressure in the IVC compartment, and reverses the Fontan paradox with no consequences of the "pulsatile Glenn." For severe forms of Ebstein's anomaly, it allows a hypoplastic RV to adequately handle the reduced preload and enables aggressive tricuspid valve repair. Results are unsatisfactory when used in the setting of elevated pulmonary vascular resistance, and when used as a salvage procedure to treat acute postoperative RV dysfunction. We propose utilization of serial MRI for assessment of chamber dimensions and quantification of pulmonary regurgitation.
|
Acknowledgments |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentAcknowledgmentsReferences |
|---|
|
References |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentAcknowledgmentsReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  S. Paranon and P. Acar Ebstein's anomaly of the tricuspid valve: from fetus to adult Heart, February 1, 2008; 94(2): 237 - 243. [Full Text] [PDF]
|
|
|
|
 |
|
 H. Laks, D. Marelli, M. Plunkett, and J. Myers Adult Congenital Heart Disease Card. Surg. Adult, January 1, 2008; 3(2008): 1431 - 1464. [Full Text]
|
|
|
|
 |
|
 Y. Hirata, J. M. Chen, J. M. Quaegebeur, W. E. Hellenbrand, and R. S. Mosca Pulmonary Atresia With Intact Ventricular Septum: Limitations of Catheter-Based Intervention Ann. Thorac. Surg., August 1, 2007; 84(2): 574 - 580. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. P. Graham Jr The Year in Congenital Heart Disease J. Am. Coll. Cardiol., June 20, 2006; 47(12): 2545 - 2553. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ANN THORAC SURG | ASIAN CARDIOVASC THORAC ANN | EUR J CARDIOTHORAC SURG |
| J THORAC CARDIOVASC SURG | ICVTS | ALL CTSNet JOURNALS |
