|
|
||||||||
Eur J Cardiothorac Surg 2007;32:195-200. doi:10.1016/j.ejcts.2007.03.039
Copyright © 2007, European Association for Cardio-Thoracic Surgery. Published by Elsevier B.V. All rights reserved
King Faisal Specialist Hospital and Research Centre, Riyadh, Saudi Arabia
Received 4 September 2006; received in revised form 9 March 2007; accepted 23 March 2007.
* Corresponding author. Address: MBC 16, PO Box 3354, King Faisal Specialist Hospital & Research Center, Riyadh 11211, Saudi Arabia. Tel.: +966 1 442 7470; fax: +966 1 4427791. (Email: asallehuddin{at}kfshrc.edu.sa).
| Abstract |
|---|
|
|
|---|
Key Words: Fontan Cavopulmonary connection Transcatheter intervention Stenting Non-surgical Fontan Hybrid procedure
| 1. Introduction |
|---|
|
|
|---|
| 2. Materials and methods |
|---|
|
|
|---|
2.1 Surgical technique
The surgical procedure is essentially a combination of the hemi-Fontan operation and a lateral tunnel Fontan procedure with a fenestration (aperture) larger than the diameter of the inferior vena cava in the medial wall of the tunnel and patch closure of the mouth of the SVC in the right atrium. The construction of the BCPC is carried out under cardiopulmonary bypass during the cooling phase. Cold blood cardioplegic arrest at moderate hypothermia (32 °C) is instituted for the establishment of the lateral tunnel. The medial wall of the tunnel is constructed using two-thirds of the circumference of a Gore-Tex vascular tube (W.L. Gore & Assoc. Flagstaff, Arizona) of 14 or 16 mm diameter. The Gore-Tex wall is fenestrated with a single aperture 10–14 mm in diameter, as high as possible on its anterior aspect. The size of the aperture is guided by the preoperative echocardiographic diameter of the inferior vena cava (IVC) at the level of the diaphragm (10 mm if IVC < 8 mm; 12 mm if IVC 8–10 mm; and 14 mm if IVC > 10 mm). The aperture is made bigger than the IVC diameter as we anticipate the former to narrow in time from possible endothelialization or peel deposition on its edges. Next, the orifice of the superior vena cava at its junction within the right atrium is patched with autologous pericardium. The aperture and the patch are positioned as close together as possible. Fig. 1
illustrates the product of this modification where a blind pouch of the cardiac end of the superior vena cava was created directly opposite the bidirectional cavopulmonary anastomosis. This is to facilitate the placement of the RF catheter in the subsequent interventional stage.
|
|
|
|
| 3. Results |
|---|
|
|
|---|
|
Ten months later (range 5–16 months), nine patients underwent transcatheter Fontan completion (mean age 20 months, mean weight 10.6 kg). Demographics of these patients are shown in Table 2 . The mean diameter of the aperture had decreased from 11.8 mm to 7.4 mm; p < 0.001 (Fig. 2). The mean fluoroscopy time was 41 min (range 27–81 min). The stents across the RF perforated pericardial patch were dilated to 14.4 mm average diameter (range 12–16 mm). Except in one patient, all apertures were closed with Amplatzer devices (sizes 6–8 mm). Average oxygen saturations increased from 85 to 94% (p = 0.001) but the pulmonary artery pressures were not significantly altered (16 vs. 19 mmHg, p = 0.12). These findings were similar to those reported by Galantowicz and Cheatham [6]. During the catheterization procedure, no patient had hemodynamic compromise and none required inotropic support. There were no early or late deaths. No patient required blood transfusions. No patient required mechanical ventilation and none developed significant pleural effusions or arrhythmias. This is in contrast to an incidence of 15–16% pleural effusions and 7–8% of dysrhythmias in patients following the modified Fontan operation previously reported from our institution [9]. All were discharged from hospital within 6 days of the non-surgical Fontan completion. On average, our surgical patients remain for 14–15 days in hospital after the modified Fontan operation [9].
|
| 4. Discussion |
|---|
|
|
|---|
The modification utilized at our institution has several advantages over the previous adaptations. With the creation of the blind pouch (Fig. 1) directly facing the bidirectional cavopulmonary anastomosis, the interventional catheters are easily guided into the pouch thus facilitating the positioning of the RF wire in the ideal location. This is verified by the short procedure and fluoroscopy time for the interventional procedure, as well as the lack of any major complications such as cardiac perforation. The pouch is also capable of growth thereby diminishing the risk of a stenosis in the Fontan pathway. The suture-line of the autologous pericardial patch facilitates anchoring of the stent. This eliminates the need for insertion of a band during the surgical procedure as have been described by earlier modifications [3,4]. The placement of the whole upper segment of the stent within the pouch avoids having a significant protrusion into the right pulmonary artery. We had established the potential for further dilatation of the stent to accommodate for the growth of the child in one of our patients. The use of a Gore-Tex medial patch that directs IVC flow into the pulmonary arteries makes stenting at this site unnecessary. This eliminates the need for a marking band around the IVC, as well as the risk of either entrapping the hepatic veins, or of excluding it from the Fontan flow as have been described elsewhere [6]. The cardiac motion and diaphragmatic oscillations furthermore, render this site an unrelentingly moving target for stenting. We position the aperture high up in the tunnel to avoid the risk of the ASD closure device impinging on the atrioventricular valve/s and interfering with its function. Its anterior position minimizes the potential for obstruction of the pulmonary veins.
We chose to create a large aperture to minimize the risk of restriction to ventricular inflow. The application of multiple fenestrations in the lateral tunnel as described by Konertz et al. [3] and Sidiropoulos et al. [4] may run the risk of gradual closure of the small holes with potential reduction in cardiac output. This could be catastrophic or may necessitate an earlier Fontan completion where the patient is not hemodynamically ready. The size of the aperture is critical and we have chosen 10 mm as its lower limit as we have noticed gradual restriction of the flow across the aperture with time. We utilized a single ASD occlusion device for this aperture, and the incidence of residual shunting in our patients is acceptably low.
The use of a covered stent as has been previously described [5–7], could simultaneously dilate the connection between the right atrium and the right pulmonary artery. This simpler approach could bring about equally satisfying results and be a cheaper solution. Covered stents are nonetheless custom-made [6] and not as readily available off-the-shelf as their non-covered counterparts. Their ability to completely close the fenestrations, or in our case the aperture, may not be complete and this could potentially lead to a higher rate, and degree of residual shunting across the atrial baffle.
Immediate complications are well-known following surgical Fontan completions, with pleural effusions occurring in the range of 13–54% [9,11,12,16,17]. The interventional Fontan completion reduces the incidence of these complications as had been observed in our experience. The need for ICU care is generally averted. The catheterization process allows for immediate peri-Fontan hemodynamic evaluation that undoubtedly helps in the decision to proceed with the Fontan completion, and the hemodynamics following completion can also be instantly assessed. Additionally, aorto-pulmonary collaterals, which are commonly present in pre-Fontan patients and has significant influence in the incidence of effusions [10,18,19] can be identified and dealt with during the non-surgical Fontan completion.
One may argue that the post-Fontan complications are not eliminated but brought forward to an earlier stage, during the BCPC. This is reasonable as the modifications hereby described entail longer perfusion and ischemic times. Nonetheless, despite this lengthy procedure we have not noticed any significant untoward consequences in our patients. Based on the 100% survival, it may seem intuitive that a more logical approach would be to perform the BCPC without cardiopulmonary bypass and later Fontan operation also without cardiopulmonary bypass. The off-pump Fontan procedure is currently conducted in patients older (mean age 2–9.8 years) and bigger (mean weight 10.8–27.7 kg) than those in our series [20–22].
Undeniably, these lengthy operative times are of concern to us. We foresee nonetheless a possible reduction of bypass time by avoiding cardiopulmonary bypass during the creation of the BCPC. Shortening of ischemic time could also be accomplished by performing the anastomosis between the cardiac end of the SVC and the inferior wall of the right pulmonary artery without aortic cross-clamp as described by Konstantinov et al. [7].
There is however a disadvantage foreseeable, whereby the Fontan has to be completed once the surgical preparation is done. The aperture would eventually be inadequate for the growing child if the Fontan is not completed and may perhaps warrant a surgical take down of the lateral tunnel. We do not foresee that dilating the aperture in the catheterization laboratory is possible, let alone desirable because the Gore-Tex toughens with time and may harbor thrombus that could be released into the systemic circulation.
Another disadvantage is the need for anticoagulation and its attendant problems of inadequate and often difficult control in the very young age group. Investigators had substituted Coumadin with aspirin after 3–4 months [4], or had administered aspirin immediately after the interventional completion [6]. These may diminish the difficulty of this problem to some extent but we continue to prefer the use of Warfarin, as the thromboembolic burden of this new procedure is still unknown. This is even more so in the event of depressed ventricular function later in life. Undeniably, the long-term effects of the early establishment of the lateral tunnel, the impact of growth and the possibility of further stent dilatation as well as the overall outcome of the patients are yet to be discovered as well. A long-term and close follow-up of these patients are essential but we remain confident that this procedure will be beneficial to a large number of patients destined for a multi-staged palliation to establish a Fontan circulation.
| 5. Conclusions |
|---|
|
|
|---|
| Appendix A |
|---|
|
|
|---|
Dr C. Pizarro (Wilmington, USA): You certainly demonstrated that this strategy is applicable to patients, and I think that this technique has been described by a German author many years ago. I have a couple of questions. One is that when you did the bidirectional Glenn did you actually divide the azygos vein? Because I think that it has been shown that although dilations have been scheduled in the future in all these patients, it is a fact that every time you take a patient to the cath lab to have further dilations you accomplish maybe 2, maybe 3 mm of increased cross section at most, and over time, the area where the stent has been placed, it sort of gets encased in fibrous tissue which is difficult to dilate. If you have not divided the azygos vein, you would have an alternative route for the blood from the lower compartment to travel towards the branch pulmonary arteries. So, did you divide, the azygos vein or was left intact?
Dr Sallehuddin: It is our practice and our policy to divide all azygos veins during bidirectional cavopulmonary shunt. It is a good point that you bring out in the use of the extra amount of caval tissue from the azygos vein, but the placement of the stent is actually well below where the azygos vein is. The stent is positioned around the area of the lower end of the superior vena cava at its junction with the right atrium. At that point I think the azygos is well away and would not be useful in the future if we have left it behind.
Dr Pizarro: And could you comment regarding the leaks that you have had? Are those leaks that relate largely to a lack of seal across the Amplatzer device or is it around it? Why are those leaks present? When you close an atrial septal defect with an Amplatzer, although you might see a leak initially, those leaks tend to disappear.
Dr Sallehuddin: Those leaks were all around the area of the Amplatzer device. I am not quite sure why they were more significant in this group of patients. Most likely the stents that we have to place are quite close to the aperture and there may be poor apposition of the device with the whole rim of the defect. But we have had no leaks other than around the area of the Amplatzer device itself.
Dr G. Stellin (Padova, Italy): The conduit that you have used to connect the inferior vena cava to the pulmonary artery is between 14 and 16 mm of diameter. That means that most likely those conduits will become restrictive with patient growth. In other words, by catheter dilating a membrane, you are avoiding an operation but you are committing your patients to a further future operation. Could you please comment about this problem?
Dr Sallehuddin: The baffle is created from a conduit. It is only two-thirds of the circumference of the conduit. It is much like the construction of a lateral tunnel using a baffle as the medial wall. So the lateral wall of the tunnel still consists of the atrial tissue and would therefore provide room for growth in that segment of the tunnel. The tunnel is not entirely made of the conduit.
Dr A. Corno (Liverpool, United Kingdom): In your drawing you showed that the anastomosis of the atrial stump of the superior vena cava to the inferior aspect of the pulmonary artery was exactly in front of the superior vena cava anastomosis, and this has been demonstrated to be no good in terms of power loss, vortex formation and flow distribution. Is this really the technique that you are using?
Dr Sallehuddin: I am quite aware of the hemodynamic studies that show that there is some benefit of skewing the anastomosis of the stump towards the right pulmonary artery and the SVC towards the left pulmonary artery. In this situation we are still early in our experience and we want to make sure it facilitates the interventionalist to find the patch, to perforate it, and further dilate it. But I think with time and with experience, the same concept of skewing the anastomosis of the cava to the pulmonary artery can be done, and with increasing experience I think the interventionalist would also easily find the patch.
Dr Corno: The preservation of the ventricular function is very important in these patients for the long-term results. With your approach you avoid the second operation, but for the first stage you need a period of myocardial ischemia. In your experience the average duration of aortic cross clamping was 67 min. The conventional superior vena cava to pulmonary artery anastomosis can be done without any period of myocardial ischemia as well as the second stage with completion of extracardiac Fontan. Can you comment on the advantages with your approach in this regard?
Dr Sallehuddin: Again, this is an early experience and we have combined groups of patients whom we have subjected to Damus-Kaye-Stansel and repair of TAPVD, and those are the patients that require really a long period of aortic cross-clamp. With experience, as has been shown by Konertz in Hannover, who did almost the same kind of reconstruction within the heart, avoiding the use of cross-clamp for the lower anastomosis between the stump and the pulmonary artery; in his publication he could actually reduce the cross-clamp time down to around 30 min, and, in fact, his average time was 31 min.
| Footnotes |
|---|
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
K. Francois, T. Bove, K. De Groote, J. Panzer, K. Vandekerckhove, B. Suys, D. De Wolf, and G. Van Nooten Pleural effusions, water balance mediators and the influence of lisinopril after completion Fontan procedures Eur. J. Cardiothorac. Surg., July 1, 2009; 36(1): 57 - 62. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. R. de Leval Editorial comment: Fontan completion without surgery Eur. J. Cardiothorac. Surg., August 1, 2007; 32(2): 201 - 201. [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 |