| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Ann Thorac Surg 2006;81:2243-2249
© 2006 The Society of Thoracic Surgeons
a Nemours Cardiac Center, Alfred I. duPont Hospital for Children, Wilmington, Delaware
b St. Christopher's Hospital for Children, Division of Cardiology, Philadelphia, Pennsylvania
Accepted for publication January 4, 2006.
* Address correspondence to Dr Pizarro, 1600 Rockland Rd, PO Box 269, Wilmington, DE19899 (Email: cpizarro{at}nemours.org).
Presented at the Fortieth Annual Meeting of The Society of Thoracic Surgeons, San Antonio, TX, Jan 26–28, 2004.
| This article has been selected for the open discussion forum on the CTSNet Web Site: http://www.ctsnet.org/sections/newsandviews/discussions/index.html
|
 |
Abstract |
|---|
 Top Abstract IntroductionMaterial and MethodsResultsCommentDiscussionReferences |
|---|
METHODS: Review of all patients (n = 107) undergoing Fontan completion between January 1998 and July 2005 to identify predictors of outcome: early death, prolonged effusions, and prolonged hospital stay.
RESULTS: Median age was 13 months (range, 11 to 35) and median weight was 9.4 kgs (6.7 to 15.1). Hypoplastic left heart syndrome was present in 61 patients, and stage I Norwood was the initial palliation in 69. An interim superior cavopulmonary connection was performed in all. A lateral tunnel was used, and a deliberate right to left shunt was created in 99 patients. Mean transpulmonary gradient and pulmonary artery pressure were 5.7 ± 1.5 mm Hg and 11.6 ± 2.2 mm Hg, respectively. Median time to extubation was 5 hours (range, 2.5 to 184). Median duration of pleural effusion was 3 days (1 to 58) and was greater or equal to 14 days in 13 patients (12%). Overall mortality was 4.5% (5 of 107). Variables associated with poor outcome included associated noncardiac diagnosis (p < 0.05), elevated transpulmonary gradient (p = 0.03), and pulmonary artery pressure (p < 0.02). Hypoplastic left heart syndrome was the only variable associated with prolonged effusive complications.
CONCLUSIONS: Fontan completion can be performed with good results in the first year of life independent of anatomic diagnosis. Significant noncardiac pathology, and a higher pulmonary artery pressure were predictive of worse outcome. Hypoplastic left heart syndrome was associated with prolonged effusions.
The surgical management of patients with a cardiac malformation characterized by a functional single ventricle rests on the premise that systemic venous blood can flow passively through a mature pulmonary vascular bed. Since the contributions of Fontan and Baudet [1] and Kreutzer and colleagues [2] to the surgical treatment of tricuspid atresia, the management strategy for patients with a functional single ventricle has evolved into a sequence of staged procedures with a final goal of achieving normal volume and pressure work for the single ventricle and normal or near normal oxygen saturation of the systemic blood.
Although it is known that pulmonary vascular development and maturation are essential conditions for a successful cavopulmonary connection, the timing of the surgical staging has been set arbitrarily, particularly the transition from superior cavopulmonary connection to Fontan. If this transition can be accomplished at an early age, the effects of persistent cyanosis and the potential for paradoxical embolization can be minimized.
The purpose of this manuscript is to review our experience using a strategy aimed to complete the Fontan circulation at about 1 year of age. Of particular interest were the impact of age and anatomic diagnosis on outcome.
|
Material and Methods |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentDiscussionReferences |
|---|
Data collection included demographics, preoperative, operative, and postoperative variables (Appendix). Institutional Review Board approval for the chart review was obtained on April 23, 2003, and individual consent was waived.
Anatomic diagnosis was based on two-dimensional echocardiography, angiography, and operative inspection. Atrioventricular valve competence and ventricular function were evaluated qualitatively using angiography and two-dimensional echocardiographic views (subcostal and apical four chambers). Preoperative echocardiographic data were complete in all, and cardiac catheterization data were available in 105 patients (98.1%). Ventricular and atrioventricular valve dysfunction was considered significant if graded moderate or greater by either method. Cardiac rhythm was assessed by review of 12-lead surface electrocardiogram in the preoperative period, first day after surgery, and at the time of discharge. New onset of dysrhythmia was defined as any rhythm change from the preoperative electrocardiogram, which was present before or at the time of hospital discharge. Criteria for sinus node dysfunction have been previously described [3]. Review of cardiac catheterization data before Fontan completion included hemodynamic and angiographic evaluation and angiographic assessment. Pulmonary and systemic blood flows were calculated using the Fick method.
The Fontan connection techniques have been previously reported [4, 5], and included a lateral tunnel technique or an extracardiac conduit. Cardiopulmonary bypass with a period of deep hypothermic circulatory arrest was used in all patients. In cases when an intentional right to left shunt was used depending on surgeon's preference, this consisted of a fenestration of the lateral tunnel or partial exclusion of one hepatic vein. When the latter technique was employed, the lateral tunnel baffle excluded one, usually the anterior hepatic vein from the Fontan pathway, allowing for intentional right to left shunt by decompression of a portion of hepatic venous drainage into the pulmonary venous atrium.
Early mortality was defined as death occurring within 30 days of surgery or before hospital discharge. A patient's outcome was defined as poor if early death, Fontan takedown, or transplantation occurred.
The duration of pleural drainage was defined as the period from the date of surgery until the last thoracentesis or final removal of pleural catheters. Prolonged pleural effusions were those that were drained for a period greater than 14 days after surgery.
Statistical Analysis
Data is presented as medians and ranges or mean and standard deviation where appropriate. Group comparisons included stratification by underlying diagnosis and age at operation (less than or more than 18 months). Univariate analyses were performed for continuous and ordinal variables, using Mann-Whitney U test, Fisher exact test, and 
2. A multivariate analysis was used to study the relationships of individual patient and procedural variables to the specific outcomes: early mortality, prolonged postoperative pleural effusions, new onset of rhythm disturbance, and duration of hospital stay. A p value of 0.05 or less was considered significant.
|
Results |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentDiscussionReferences |
|---|
|
To determine potential influence of age at the time of Fontan completion, besides using age as a continuous varaible, two patient subgroups were defined according to age being equal to or greater than 18 months at the time of Fontan (Table 2). The only significant difference between these subgroups other than age and weight at operation was hematocrit, with higher levels being present in older patients. The distribution of HLHS was similar in each group. Five patients had a mean pulmonary pressure greater than 15 mm Hg, which was associated with end-diastolic ventricular pressure of 10 mm Hg or greater in 4. The Qp/Qs ranged between 0.4 and 0.8, and all patients exhibited an increase in pulmonary blood flow in response to inspired oxygen.
|
|
The mean duration of cardiopulmonary bypass was 65.3 ± 13.7 minutes (range, 37 to 106), and the mean duration of the circulatory arrest was 33 ± 11 minutes (range, 15 to 63). Based on surgeon's preference, modified ultrafiltration was used in 20 patients. The mean volume removed was 444 ± 93 cc, with an associated increase in hematocrit of 15.6% ± 4.6%.
Mortality and Takedown
There were 5 early deaths (4.6%). The preoperative diagnosis and causes of death are shown in Table 4. None of these patients had an anomaly of the Fontan pathway as demonstrated by postoperative catheterization, surgical exploration, or autopsy. Five patients underwent takedown to a hemifontan, including 4 patients who died early. The fifth patient in whom a conversion was used had significant postoperative bleeding and developed a condition characterized by decreased intraventricular cavitary volume and low cardiac output. After conversion to a hemifontan in the first 24 hours, the patient did well and underwent successful Fontan completion 2 years later.
|
The median duration of intensive care unit and hospital stay were, 1 day (range, 1 to 27) and 10 days (range, 3 to 69) respectively. Median duration of postoperative mechanical ventilation was 5 hours (range, 2.5 to 184). Prolonged hospital stay was associated with noncardiac conditions (p < 0.001), the presence of a common atrioventricular valve (p = 0.001), prolonged intensive care unit stay (p < 0.001), and prolonged pleural effusions (p < 0.001). Age at operation did not influence either intensive care unit stay, hospital length of stay, or duration of mechanical ventilation. Outcome by age subgroup is shown in Table 5.
|
Other Complications
Preoperative 12-lead electrocardiograms were available in all patients. Normal sinus rhythm was present in 99 patients (92.5%), 6 had sinus node dysfunction (5.6%), and 2 patients had atrioventricular block, 1 of them had a pacemaker. Discharge electrocardiograms demonstrated new onset of dysrhythmia in 4 patients. Two patients with preexisting rhythm issues received a pacemaker, 1 with sinus node dysfunction and 1 with atrioventricular block. Normal sinus rhythm was documented before discharge in 92 of 102 hospital survivors (86%), 10 patients had sinus node dysfunction (9.3%), and 2 had complete heart block. None of the variables studied correlated with the new onset of dysrhythmia.
Neurologic events included seizures in 3 patients, associated with a febrile episode in one, and one stroke with residual deficit.
The median arterial oxygen saturation at discharge was 87.6% ± 6.1% (range, 77% to 96%).
|
Comment |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentDiscussionReferences |
|---|
The data reported here demonstrates that Fontan completion can be successfully accomplished in infancy regardless of ventricular morphology. Although our cohort presented significant differences for age, weight, and hematocrit between the two age subgroups, the hemodynamic data were remarkably similar and suitable for Fontan completion.
Since overall survival of patients undergoing palliation of hypoplastic left heart syndrome and its anatomic variants has improved considerably, it is not surprising that this group is the predominant diagnostic category among patients who reach Fontan completion at some centers. The issue of ventricular morphology and most particularly the presence of hypoplastic left heart syndrome continues to be associated with the perception of suboptimal outcome following Fontan completion, despite recent reports to the contrary [4, 9, 13]. These clinical perceptions are related to a prejudice about the unsuitability of the morphologic right ventricle to support the systemic circulation, a concept that is not supported by recent data [14, 15]. Mosca and associates reported the outcome of Fontan procedure in 100 consecutive patients with HLHS between 1992 and 1998. In their experience, a number of modifications introduced during the study period resulted in outcomes no different from other forms of single ventricle [16]. A report by Gaynor and colleagues [9] demonstrated an improvement in Fontan outcome over time that was associated with the introduction of a series of modifications in the management of these patients. Neither study identified HLHS to be associated with an increased risk of failure. More recently, McGuirk and associates [17] reported the results of 103 Fontan operations in a cohort with roughly equal numbers of patients with right and left ventricular morphology. Their conclusions reaffirm the idea that in the contemporary era, the Fontan procedure affords low mortality and improved functional class independent of the morphology of the functional single ventricle.
Our experience with a cohort largely composed of patients with hypoplastic left heart syndrome provides additional support for the notion the Fontan procedure can be successfully accomplished in infancy, regardless of ventricular morphology. After an initial palliative strategy dictated by the anatomic diagnosis, a hemifontan type cavopulmonary connection was performed in all patients, followed by completion of the Fontan. Fontan surgery was timed around the first birthday, but was modified by clinical evaluation of qualitative changes in ventricular geometry, volume-mass ratio, and the hemodynamics associated with those changes. From this experience, it appears that a period of approximately 6 months after the superior cavopulmonary connection is sufficient in most cases for the appropriate changes in ventricular geometry and diastolic properties of the single ventricle to occur for a successful Fontan circulation [18]. Patients studied at age 1 year demonstrated no hemodynamic differences compared with peers studied at age 2 years, confirming our premise that based on physiologic data Fontan completion could be successfully performed in infancy, thus reducing the time span these patients are subjected to chronic cyanosis and a risk for paradoxic embolization.
Analysis of the cohort with less satisfactory outcome revealed that significant associated noncardiac disease, a higher pulmonary artery pressure, or transpulmonary gradient had a direct correlation with early death or takedown. The influence of the preoperative hemodynamics on post-Fontan outcome is well documented and has served to identify patients in a "higher-risk" category [19]. Although the presence of important associated noncardiac pathology has been previously associated with poor outcome after stage I Norwood [9, 20], its association with early death at the time of Fontan completion most likely represents the impact of these associated comorbidities on the midterm outcome of these patients in an era when the advances in the management of HLHS allows patients with important comorbidities to progress through the staged surgical sequence beyond the initial stage.
As previously suggested, HLHS appears to be associated with a higher incidence of prolonged pleural effusions [9]. Our data shows a 10-fold increase in the incidence of prolonged effusions among HLHS patients when compared with other anatomic substrates. Additionally, prolonged effusions were also associated with ventricular end-diastolic pressure and aortic saturation. However, it is possible that owing to the limited size of the cohort and the favorable influence of interventions like second stage and obligatory right-left shunt, these surrogate markers for volume load or diastolic dysfunction of the single ventricle only reached borderline significance.
Regarding strategies used to decrease the incidence of effusive complications, we did not identify a clear impact of the type of obligatory right-left shunt; the excluded hepatic vein almost universally was larger than 4 mm and therefore afforded at least the same if not larger capacity to divert blood form the systemic venous pathway to the pulmonary venous atrium, effectively lowering the pressure in the "Fontan circuit." Similarly, we did not observe the beneficial effect of modified ultrafiltration on patients who underwent Fontan completion with an obligatory right-left shunt using a brief period of circulatory arrest. Numerous reports on the favorable effect of modified ultrafiltration on single-ventricle patients have resulted in its widespread application [10, 16, 21]. It is possible that the simultaneous introduction of modified ultrafiltration and the routine use of fenestration in these reports confounded the true benefits of this perfusion tool. On the other hand, the short cardiopulmonary bypass time and minimal weight gain "post-pump" in our cohort could have minimized the well-described benefits of fluid and "bad humor" removal.
Our data demonstrates that single-ventricle patients can transition towards Fontan completion as early as 1 year of age without additional morbidity or mortality. The hemodynamic profile of these patients is no different from the one encountered among patients reaching Fontan at later age, and their outcome is similar. This approach reduces the time single-ventricle patients are exposed to lower levels of arterial saturation and decreases the risk of paradoxic embolization. Additionally, our data support the concept that a favorable outcome after Fontan can be achieved regardless of the anatomic substrate. The incidence of effusive complications was not affected by age at the time of Fontan, but has a higher incidence among patients with HLHS.
Limitations
This study has several limitations. The review is retrospective, and patients were not randomly assigned for timing of surgery. The study period extends between 1998 and 2005; therefore, the follow-up of this cohort is limited. Owing to the small number of patients in the poor outcome group (early death, late death, and Fontan takedown), all preoperative factors contributing to poor outcome may not have been identified, particularly an analysis of the impact of variation in surgical technique, perfusion strategies, or other treatments used to minimize the effusive complications. The routine use of deep hypothermia and circulatory arrest, a superior cavopulmonary connection, and small number of patients with pulmonary artery distortion, atrioventricular valve regurgitation, or without an obligatory right-left shunt did not allow us to determine the potential influence of these variables on outcome.
|
Appendix |
|---|
Age at surgery, sex, weight, gestational age.
Preoperative Variables
Associated cardiac and noncardiac diagnosis, diagnosis (single left ventricle, normally related great arteries, single left ventricular transposition of the great arteries, single right ventricle, hypoplastic left heart syndrome, heterotaxy), atrioventricular valve anatomy (tricuspid, mitral, common), atrioventricular valve regurgitation (none-mild, moderate-severe), pulmonary venous obstruction, anomalous pulmonary veins, pulmonary artery distortion, restrictive atrial septal communication, initial palliation (none, pulmonary artery banding, balloon septotomy, banding and coarct repair, systemic-pulmonary shunt, Norwood), hemodynamic data (pulmonary artery pressure, end-diastolic ventricular pressure, Qp/Qs, transpulmonary gradient, pulmonary artery distortion), systemic-pulmonary collaterals, collateral occlusion, preoperative rhythm.
Operative Variables
Fontan type (lateral tunnel, extracardiac), right-left shunt (fenestration, partially excluded hepatic vein), atrioventricular valve repair, atrial septectomy, duration of circulatory arrest, duration of total circulatory support time, emergency Fontan.
Postoperative Variables
Mechanical circulatory support, Fontan takedown, perioperative dysrhythmia, reintubation, duration of pleural effusions, effusion drainage, oxygen saturation post-Fontan, hematocrit post-Fontan, rhythm post-Fontan, recurrent effusion, recurrent hospital admission, interventional catheterization.
|
Discussion |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentDiscussionReferences |
|---|
The question is whether either of these factors, young age or right ventricular morphology of the single ventricle, increased the risk of this Fontan procedure. Since the authors have adopted a policy of carrying out Fontan procedures at a young age—no patient in the series was more than 2 years old—they cannot address age as a risk factor, and since over half of their patients had hypoplastic left heart syndrome, they cannot assess whether right ventricular morphology is a risk factor for mortality after this procedure.
Based on our own experience, I have concluded that young age should not be thought of as a precluding risk factor for a Fontan procedure, but if one is to contemplate an elective Fontan operation in the younger patient, there should ideally be no other risk factors present preoperatively. Parenthetically, I believe that in at least some of our analyses, the presence of a single right ventricle, not associated with hypoplastic left heart syndrome, was associated with more favorable outcomes than compared to the entire Fontan population.
I have questions in three areas, and I would like to make one plea. First, since 10% of your patients did not have an intervention to create an intentional right to left shunt, what criteria were used to guide this decision?
Since you used the hepatic vein exclusion technique to create the right to left shunt in the vast majority of the cases, what have been the late consequences of the use of this technique? It was my understanding that many of these patients in previous experiences became progressively more desaturated over time from intrahepatic venovenous collateral formation, and I am interested to know if this is still a problem or, if not, how one avoids it?
Finally, I have a few questions and comments on the statistical analysis. Were your statistical analyses done with Fontan failure as the endpoint or death as the endpoint? Did you carry out separate multivariate analyses based on just preoperative variables or just preoperative and intraoperative variables? Much of the value of these retrospective reviews is to allow your colleagues to use this information to make assessments of risk for individual patients before the operation.
Inclusion of postoperative variables, such as the need for reintubation or circulatory support, is correctly noted in your manuscript as surrogate markers for poor outcome. The question then is whether removal of these surrogate markers from the statistical analysis alters the results in the multivariant analysis?
Finally, the plea that I would like to make is that we need to keep the distinction between a statistical association and a causal relationship clear in our mind. I think too easily we say something caused something because they are statistically associated, and I think the point is made in the manuscript that there is a beneficial outcome of having sinus rhythm on outcome, a finding that is almost intuitively obvious.
I appreciate the opportunity to review the manuscript in advance and for sharing this very important data with us. Thank you.
DR PIZARRO: Thanks, Dr Mayer, for your insightful comments. I will try to answer some of your questions. First, the criteria used to decide which patients would receive an intentional right to left shunt were based on considering aspects of the hepatic venous anatomy, hemodynamics, and the patient's referral origin. If a patient was judged to have favorable hemodynamics, the relationship of the hepatic venous orifices was not amenable to easy placement of a baffle within the inferior vena cava, a patient came from abroad, or a combination of those factors, then a Fontan completion without a fenestration was used, mainly with the idea that with the routine use of a partial hepatic vein exclusion, some form of an intervention would likely be required at a later date.
In this series, a considerable number of patients in whom partial exclusion of one hepatic vein was used developed intrahepatic venovenous collaterals and progressive shunting from the systemic venous pathway through the liver into the pulmonary venous atrium, resulting in arterial desaturation. We observed that at a median followup of 2 years, about two thirds to 75% of the patients have come back for an intervention in order to occlude the excluded hepatic vein, owing to progressive arterial desaturation and polycitemia. This has been generally performed in the cardiac catheterization laboratory. The operative diagram shows the excluded hepatic vein looped around with nonabsorbable suture, which is passed through a tourniquet and placed at the lower end of the sternal incision. This tourniquet then provides an easy way to occlude the excluded hepatic vein if necessary. Occasionally, that has not been sufficient, and an Amplatzer occluder device has been placed to obliterate the orifice of the excluded hepatic vein through a transthoracic approach by directly puncturing the pulmonary venous atrium.
Secondly, the endpoints used for the analysis included operative mortality and Fontan failure. However, owing to the inclusion of the same patients in both groups, no differences were observed. Separate analyses for operative variables, preoperative variables, and patient variable were performed independently and not lumped together. Additionally, we do not know if removal of the surrogate markers for poor outcome would alter the results of this analysis. As Dr Mayer mentioned, with reference to having normal sinus rhythm, the statistical significance doesn't necessarily imply that there is a causal relationship. Thank you.
|
References |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentDiscussionReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  A. Ikai, Y. Fujimoto, K. Hirose, N. Ota, Y. Tosaka, T. Nakata, Y. Ide, and K. Sakamoto Feasibility of the extracardiac conduit Fontan procedure in patients weighing less than 10 kilograms. J. Thorac. Cardiovasc. Surg., May 1, 2008; 135(5): 1145 - 1152. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Sallehuddin, A. Mesned, M. Barakati, M. A. Fayyadh, F. Fadley, and Z. Al-Halees Fontan completion without surgery Eur. J. Cardiothorac. Surg., August 1, 2007; 32(2): 195 - 200. [Abstract] [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 |
