| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Department of Pediatric Cardiac Surgery, Collegium Medicum, Jagiellonian University, Krakow, Poland
Accepted for publication June 7, 2007.
* Address correspondence to Dr Malec, Department of Pediatric Cardiac Surgery, Collegium Medicum, Jagiellonian University, 11F Jesionowa St, Krakow, 30-222, Poland (Email: mimalec{at}cyf-kr.edu.pl).
 |
Abstract |
|---|
 Top Abstract IntroductionPatients and MethodsResultsCommentReferences |
|---|
Methods: Between September 1995 and November 2006, a consecutive series of 50 children with hypoplastic left heart syndrome from a single institution underwent a fenestrated Fontan operation (lateral tunnel technique): group 1 (n = 19) after the modified Blalock-Taussig shunt, and group 2 (n = 31) after RV–PA shunt during the Norwood procedure. Hemodynamic, echocardiographic, electrocardiographic, and clinical operative and perioperative data were analyzed.
Results: Children after the RV–PA shunt were characterized by higher preoperative partial oxygen tension in pulmonary arteries (p = 0.018) and the aorta (p = 0.028), as well as lower systolic, diastolic, and mean aortic pressure (p = 0.005, p = 0.004, p = 0.019). After administration of 100% oxygen, this group additionally showed a lower value for systemic resistance (p = 0.013). The analyzed angiograms revealed a higher incidence of systemic–to–pulmonary collateral vessels (p = 0.003) in group 2. At the discharge after Fontan operation, children after the RV–PA shunt demonstrated higher arterial partial oxygen tension (p = 0.004). The two groups did not differ significantly with respect to the mortality, ventricular function, incidence of pleural effusions or rhythm disturbances, intensive care unit stay, and hospitalization time.
Conclusions: The Norwood procedure with the RV–PA shunt provides satisfactory late hemodynamics. Children who underwent this method of palliation were more prone to the development of systemic–to–pulmonary arterial collaterals.
|
Introduction |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentReferences |
|---|
For neonates with hypoplastic left heart syndrome, the Norwood procedure has become the most frequent first-stage management. Although the operative mortality has steadily decreased in recent years, it is still high and the technique of this operation is evolving. The right ventricle–to–pulmonary artery (RV–PA) shunt, first reported by Norwood and colleagues [3] and recently popularized by Sano and colleagues [4, 5], has many favorable effects on early postoperative hemodynamics that result in higher survival [4, 6–8].
The aim of this study is to compare late hemodynamics and operative outcomes at the time of Fontan operation in patients undergoing Norwood procedure with an RV–PA shunt versus a Blalock-Taussig (BT) shunt. We present an outcome review of Fontan patients initially palliated with the Norwood procedure and an RV–PA shunt.
|
Patients and Methods |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentReferences |
|---|
In the neonatal period, all the patients underwent the Norwood procedure (atrial septectomy, pulmonary artery to ascending aorta association with homograft reconstruction of the aortic arch). Group 1 comprised 19 children with the application of an expanded polytetrafluoroethylene (ePTFE) tube IMPRA Vascular Graft (Bard, Tempe, AZ), which was 3.5 mm in 4 children and 4 mm in 15 children. Group 2 comprised 31 patients in whom the Norwood procedure was completed by the construction of an RV–PA shunt with 5-mm PTFE graft.
At a median age of 7.2 months (range, 6.0 to 10.2 months) in group 1 and 6.4 months (range, 4.9 to 10.7 months) in group 2, all the children underwent the hemi-Fontan procedure. The surgical technique was consistent between groups: removal or closure of a systemic–to–pulmonary artery shunt, augmentation of central pulmonary arteries using a homograft patch, association of the superior vena cava with the branch pulmonary arteries, revision of the interatrial communication (if necessary, it was enlarged), and separation of the atrium from the cavopulmonary anastomosis using a PTFE patch.
Group 1 consisted of 5 girls and 14 boys. At the time of the Fontan operation, they were a mean age of 25.6 ± 10.3 months (range, 15.7 to 59.2 months) and weighed a median of 11.3 kg (range, 7.7 to 19 kg). Group 2 included 6 girls and 25 boys. They were a mean age of 29.3 ± 6.4 months (range, 18.9 to 47.3 months) and weighed a median 11.5 kg (range, 10.0 to 18.0 kg; Table 1). The total number of patients seen during the study period (1993 to 2006) is presented in Figure 1.
|
|
Complete hemodynamic and angiographic studies were performed in 18 children (94.7%) from group 1 at the mean age of 25.0 ± 10.1 months (range, 15.4 to 56.7 months) and in all the children from group 2 at the mean age of 27.6 ± 5.2 months (range, 18.5 to 39.3 months). Aortograms and selective arteriograms were reviewed to identify systemic–to–pulmonary arterial collaterals. The systemic–to–pulmonary arterial collateral vessel was defined as one arising from the arterial circulation, having a readily identifiable angiographic origin, supplying the pulmonary parenchyma, and opacifying the pulmonary arteries or veins, or both. In patients who required coil embolization, standard techniques of embolization were used.
The hemodynamic data recorded from each catheterization included superior vena caval, aortic, right ventricular, and pulmonary artery pressures, oxygen saturation, and partial arterial oxygen tension. All hemodynamic measurements were also performed after the administration of 100% oxygen in the inspired mixture. Calculations for the pulmonary–to–systemic blood flow ratio (Qp:Qs) were done according to the Fick principle. Data from routine blood tests performed before catheterization (hematocrit value, hemoglobin concentration) were also collected.
The operation was performed by the same surgeon through a median sternotomy. The hypothermic cardiopulmonary bypass was established by cannulation of the ascending neoaorta and the right atrium below the cavopulmonary anastomosis. At a rectal temperature of 20°C, the circulation was arrested and the ascending neoaorta was cross-clamped. Myocardial protection was achieved with crystalloid cardioplegia (potassium cardioplegia: 0.9% sodium chloride with 20 mmol/L potassium chloride, 4°C, 15 mL/kg).
The surgical procedure was similar to the technique previously described by de Leval and colleagues [9]. The patch separating the atrium from the cavopulmonary anastomosis was completely excised. A new patch cut from a portion of a 10-mm PTFE tube was fenestrated (fenestration was about 3 mm in diameter) and then sewn into the right atrium, creating a lateral atrial tunnel. After closing the atriotomy, cardiopulmonary bypass was reinstituted.
In the postoperative period, the mechanical ventilation was continued until the patient’s hemodynamic and respiratory parameters were stable. For the purpose of this study, prolonged pleural effusions were defined as pleural effusions that required drainage or repeated thoracocentesis for the total time of more than 14 days. Our strategy was to leave the chest tube in place until daily drainage was less than about 3 mL/kg.
The oxygen saturation and partial arterial oxygen tension at the discharge from the intensive care unit, as well as oxygen saturation from the pulse oximetry at the discharge from the hospital, were recorded for all the children.
Operative mortality was defined as death before hospital discharge. The data are represented as median and range. The statistical analysis was done by means of descriptive statistics, 
2 test (with or without the Yates correction) and V2 test for nonparametric variables, and the Mann-Whitney test for continuous variables. Differences were considered statistically significant at a value of p < 0.05.
|
Results |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentReferences |
|---|
The preoperative electrocardiograms revealed sinus rhythm in all the children from both groups. One child in group 1 exhibited premature ventricular contractions. One child in group 2 presented Wolff-Parkinson-White syndrome, and premature atrial beats developed in another child. None of these children received antiarrhythmic treatment. No ventricular arrhythmias were present in the analyzed electrocardiograms of children from group 2.
The echocardiographic evaluation performed before the Fontan operation revealed that the groups did not differ significantly with respect to the incidence of moderate and severe tricuspid regurgitation combined: 4 children (21.0%) in group 1, and 3 children (9.6%) in group 2 (p = 0.481). The incidence of depressed ventricular function was 2 children (10.5%) in group 1 and 7 children (22.5%) in group 2 (p = 0.485). Poor ventricular function was not diagnosed in any of the children. The right pulmonary artery index according to echocardiography was significantly larger in patients after the RV–PA shunt (p = 0.001; Table 1).
At the time of the cardiac catheterization before the Fontan operation, the two groups showed no significant differences with respect to age (p = 0.101). The analyzed cardiac catheterization data are presented in Table 2. The hemodynamic measurements performed after the administration of 100% oxygen in the inspired mixture are summarized in Table 3. The children after the RV–PA shunt during the Norwood procedure (group 2) had significantly higher partial oxygen tension in the pulmonary arteries (p = 0.018) and the aorta (0.028), as well as lower systolic, diastolic, and mean pressure in the aorta (p = 0.005, p = 0.004, p = 0.019, respectively). After the administration of 100% oxygen, this group additionally showed a significantly higher value of oxygen saturation (p = 0.015) and partial oxygen tension (p = 0.009) in the superior vena cava and a lower value of the systemic resistance (p = 0.013). In group 2, a nonsignificant trend to demonstrate a higher value of pulmonary blood flow was also noted (p = 0.055).
|
|
The groups did not differ with respect to the cardiopulmonary bypass time (p = 0.800), circulatory arrest time in deep hypothermia (p = 0.343), intensive care unit stay (p = 0.106), time of mechanical ventilation (p = 0.456), or the hospital stay (p = 0.975).
At the time of discharge from the intensive care unit, the children from group 2 showed significantly higher values of arterial partial oxygen tension (p = 0.004) and oxygen saturation (p = 0.024). At discharge from the hospital in patients from this group, a nonsignificant trend was observed to exhibit higher oxygen saturation measured by pulse oximetry (p = 0.073; Table 1). All the children presented sinus rhythm. Episodes of supraventricular tachycardia were observed in 1 child (5.3%) from group 1 and in 2 children (6.4%) from group 2.
The operative survival rate was 100% in both groups. The most important early complications in group 1 included pleural effusions in 7 patients (36.8%) and cardiac failure in 1 (5.3%); in group 2, pleural effusions occurred in 16 (51.6%), ascites in 5 (16.1%), and diaphragmatic paralysis in 2 (6.4%). At a mean follow-up of 47.6 ± 32.3 months (range, 2.8 to 136.1 months), there were two late deaths, all in group 1:1 child had heart failure with massive effusions, and the other suddenly died at home for unknown reasons.
|
Comment |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentReferences |
|---|
Our study has revealed that children with application of the RV–PA shunt during the first-stage palliation are more likely to have systemic–to–pulmonary collateral vessels in late follow-up at the time of the Fontan operation. As a probable consequence of such vessel development, we have noted higher partial oxygen tension in pulmonary arteries and the aorta, as well as lower systolic, diastolic, and mean aortic pressure (due to the runoff of blood to the pulmonary circulation). Measurements performed after administration of 100% oxygen in the inspired mixture also revealed lower systemic resistance and higher pulmonary blood flow in these children.
The incidence of systemic–to–pulmonary collateral vessels in patients on the Fontan pathway is between 20% and 80% [13–17]. Rather than the aberrant vessels occurring in tetralogy of Fallot with pulmonary atresia, which have no equivalent in normal anatomy, these collateral vessels are morphologically normal vessels that have proliferated and enlarged due to an increased flow in response to cyanosis [14, 18]. The development of collateral vessels is probably an adaptive mechanism caused by the preoperative chronic subnormal blood oxygen content [14, 17, 19]. The longer the patient experiences hypoxemia, the higher is the incidence of collaterals [16, 19]. In our study, the children from both groups were at the same age at the time of cardiac catheterization (diagnosis of collateral vessels); however, the hemi-Fontan operation was performed significantly earlier in children after the RV–PA shunt.
In our previous study [20], which was designed to compare children after the Norwood procedure with the RV–PA and BT shunt at the time before the hemi-Fontan procedure, we have observed significantly lower aortic and superior vena cava oxygen saturation as well as a significantly higher hematocrit value in patients with application of the RV–PA shunt. The significantly lower Qp:Qs between the first and the second stage due to flow through the RV–PA shunt only during the systole (and even reversed flow to the right ventricle during the diastole) in comparison with the BT shunt has been noted in many studies [4, 8, 10, 11, 21]. Because of these findings, we would suggest that the stimulus for development of collaterals occurs between the stage 1 and stage 2 procedures, and thus we postulate that the performing the stage 2 palliation earlier or using larger shunts may prevent the development of collateral vessels.
Hemodynamic implications of collateral pulmonary blood flow have detrimental effects on the Fontan circuit. An analytical description of fluid motion revealed that the mixing of two blood streams with dissimilar velocities causes dissipation of flow mechanical energy [13]. In this way, mixing of systemic arterial with systemic venous blood streams can increase flow energy loss and impair cardiac efficiency [13]. An additional source of pulmonary flow can elevate the pulmonary artery and systemic venous pressure, inhibiting the systemic venous return, and subsequently, through the pulmonary circulation, the filling of the single ventricle [17, 19, 22]. In our study, however, there were no significant differences between the groups with respect to the superior vena caval and pulmonary artery pressures. Decompression of the superior vena cava to the inferior vena cava by the accessory venous pathway was advocated by other reports as the reason for that fact [14]. Some investigators even observed lower superior vena caval pressure in patients after the bidirectional Glenn procedure, with significant collaterals in comparison with patients without these vessels [15].
Collateral circulation is a source of ineffective pulmonary blood flow (admixture of oxygenated arterial blood in pulmonary circulation) and is disadvantageous because of volume overload to the systemic ventricle [13, 14, 16, 22, 23]. The collateral flow may contribute as much as 8% of cardiac output [15] and 55% of flow pump returning to the heart through the pulmonary veins during the cardiopulmonary bypass [19]. Triedman and associates [14] estimated the average total cross-sectional area of collaterals as comparable with the cross-sectional area of a 4-mm systemic-to-pulmonary shunt.
Some reports have shown that patients after the Fontan operation with significant collateral vessels have prolonged pleural effusions, higher incidence of protein-loosing enteropathy, longer stay in the hospital, and even have higher mortality [13, 17, 22, 23]. The duration of the pleural drainage was shorter in patients who had occlusion of collaterals [23]. Our data showed the higher incidence of collateral vessels in group 2 was not associated with adverse outcome, higher incidence of pleural effusions, longer hospitalization time, or higher mortality rate. Our results are supported by other investigators [16].
On the other hand, there is also the advantageous effect of accessory systemic arterial blood flow to the pulmonary circulation, which is the improvement of systemic arterial saturation. It has been even suggested that a small admixture of arterial blood flow into the pulmonary circulation may prevent the development of pulmonary arteriovenous malformation after the stage 2 procedures for the functional single ventricle [14].
The fundamental goal of the staged Fontan palliation is preservation of the single ventricle, thus the collateral vessels should be aggressively diagnosed and effectively occluded to prevent late myocardial dysfunction [17]. However, owing to difficulties related to the identification and quantification, some smaller collaterals are still present after the embolization of the large vessels, what was probably the reason for higher arterial partial oxygen tension and higher oxygen saturation at the time of discharge in our group 2 patients [16]. Aortograms often fail to visualize the presence of collaterals, and selective angiographies of more distal vessels are required [14, 23].
The potential deleterious effect of an incision of single ventricle in the site of the proximal end construction of the RV–PA shunt was the main concern of surgeons who have used this modification of the Norwood procedure. In our experience, the right ventriculotomy was not a substrate for ventricular arrhythmias. The echocardiographic evaluation performed in the preoperative period revealed good ventricular function in almost all patients, with no differences between the groups.
A limitation of this study is that it was a retrospective analysis, and patients could not be randomized strictly to one of the two operation groups, which were consecutive rather than contemporary.
In summary, the results of our study indicate that the Norwood procedure with the RV–PA shunt provides satisfactory late hemodynamics and produces good candidates for the Fontan operation. The higher incidence of systemic–to–pulmonary arterial collaterals emphasized by this study has had no adverse early consequences after the Fontan operation. The long-term effects of the application of the RV–PA shunt during the Norwood operation are unknown and should be clarified by ongoing and future assessments.
|
References |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentReferences |
|---|
Related Article
This article has been cited by other articles:
|
|
 |
|
  S. Sano Invited commentary Ann. Thorac. Surg., November 1, 2007; 84(5): 1617 - 1618. [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 |
