Conflict of interest: Nothing to report.
E-ONLY: Pediatric and Congenital Heart Disease
Stenting an aortopulmonary conduit with peripheral cardiopulmonary bypass support
Article first published online: 27 JUN 2013
Copyright © 2013 Wiley Periodicals, Inc.
Catheterization and Cardiovascular Interventions
Volume 83, Issue 1, pages E77–E81, 1 January 2014
How to Cite
Incani, A., Lee, J. C., Nicolae, M. J. and Walters, D. L. (2014), Stenting an aortopulmonary conduit with peripheral cardiopulmonary bypass support. Cathet. Cardiovasc. Intervent., 83: E77–E81. doi: 10.1002/ccd.24952
- Issue published online: 18 DEC 2013
- Article first published online: 27 JUN 2013
- Accepted manuscript online: 16 APR 2013 02:17PM EST
- Manuscript Accepted: 7 APR 2013
- Manuscript Revised: 27 FEB 2013
- Manuscript Received: 28 NOV 2012
- congenital heart disease in adults;
- percutaneous coronary intervention;
- intravascular ultrasound
Although surgically created aortopulmonary (AP) shunts are uncommon in the adult congenital heart disease population, they are often used in patients with pulmonary atresia. For these patients, the shunt is a vital supply of pulmonary blood flow and thus obstruction of the shunt may lead to pulmonary hypoperfusion and hypoxia thereby increasing morbidity and mortality. This report describes a safe and effective method of stenting the conduit with the hemodynamic support of peripheral cardiopulmonary bypass (PCB). Prior to the procedure, a multimodality assessment of a stenosis in a kinked AP conduit using computed tomography, angiography, intravascular ultrasound (IVUS), and pressure wire assessment (PWA) was utilized. While PCB, IVUS, and PWA have all been used to great effect in various clinical scenarios, the combined use of these techniques has not been previously been described in the setting of intervention in adult congenital heart disease. © 2013 Wiley Periodicals, Inc.
A 47-year-old woman was admitted with two days of progressive breathlessness (on minimal exertion), palpitations, and intermittent syncope. She had congenital pulmonary atresia with ventricular septal defect (PA-VSD), hypoplastic pulmonary arteries (PAs), arborization defects, and multiple aortopulmonary (AP) collateral arteries (MAPCAs). She also had coronary artery fistulae to her left ventricle and PA and known dysfunction of the right ventricle (RV). Several previous electrical cardioversions for paroxysmal atrial flutter had been unsuccessful. Hence, she was treated with a combination of flecainide and digoxin. There was no other significant medical history.
Although our patient had undergone a number of surgical procedures as a child at an interstate hospital, her documented history at our institution started at the age of 21 years. At this time, she underwent a modified Blalock-Taussig (BT) shunt (8 mm GORE-TEX®, Gore and Associates, Belrose, NSW) from the right subclavian artery to a diminutive right PA. Three years later, she developed a shunt stenosis, which was successfully managed with percutaneous balloon angioplasty. Eight years after that, she developed worsening dyspnea and fatigue. Angiography demonstrated that the left lung was largely supplied by MAPCAs. The right lung was predominantly supplied by the right BT shunt with support from some collateral arteries arising from the aorta. The right upper lobe MAPCAs were embolized with a view toward unifocalization and trying to connect the PA branches. The right subclavian to the right PA shunt was surgically divided; the left PA was divided at its aortic end. A 16 mm Hemashield® (Boston, Natick, MA) graft was applied between the left and the right PAs as an on lay graft through the aortic concavity. Due to the critical coronary anatomy overlying the RV, a definitive repair with a pulmonary conduit from the RV to the conjoined PA system was not deemed safe and thus, a 10 mm GORE-TEX® graft was applied from the ascending aorta to the 16 mm Hemashield® conduit.
Over the following 10 years, she developed significant stenoses of the interpulmonary Hemashield® conduit (central PA conduit) as well as the GORE-TEX® AP conduit. This was demonstrated by a cardiac catheterization study using IVUS for more objective estimation. The central PA conduit was stenosed to a 2–3 mm opening and the AP shunt was down to 5 mm in size. It was felt that percutaneous intervention to the central PA conduit would not be technically possible, so she was referred again for surgery.
The AP and central PA conduit arterial grafts were upgraded. The central portion of the conduit between left and right PAs was enlarged and a 10 mm GORE-TEX® shunt was applied from the graft to the ascending aorta. The postoperative course was complicated by multiorgan dysfunction and a ventricular fibrillation cardiac arrest. The patient and her physicians were apprehensive of any further surgery.
Her current presentation of hypoxemia and breathlessness raised the possibility of repeat shunt failure. On examination, her blood pressure was 110/70 mm Hg, the pulse was 72 bpm (in atrial fibrillation). The oxygen saturation level was 70% (on 3L oxygen via nasal prongs). Heart sounds were dual and a continuous murmur was auscultated parasternally. There was marked hepatomegaly. Respiratory examination was unremarkable.
A 6-min walk test (6MWT) 2 months prior to the acute presentation yielded a distance of 235 m. Upon admission, the 6MWT distance was only 54 m. Her oxygen saturation at rest on room air was 67%. The nadir on exertion was 37%.
Computed tomography (CT) of the thorax confirmed the obstruction (Fig. 1) with a loculated mediastinal collection (probably arising from the most recent surgery: white arrow, Fig. 1a and b) surrounding a kinked and stenosed GORE-TEX® AP conduit. The central interpulmonary conduit appears (in Fig. 1b) to be stenosed. This was, in fact, angulated but widely patent (as demonstrated in Fig. 1c). The maximal stenosis of the AP conduit is further demonstrated on a zoomed view (Fig. 1d). The AP conduit stenosis was also visualized on angiography of the shunt using a coronary Amplatz Left 1 (AL1) catheter (Fig. 2).
Figures 1 and 2 demonstrate the difficulty of imaging a “slit-like” stenosis in a kinked conduit. This was the main impetus to investigate further the stenosis anatomically with intravascular ultrasound (IVUS) and hemodynamically with pressure wire assessment (PWA).
Upon selectively engaging the shunt, there was an immediate drop in systemic pressure, consistent with critical hemodynamic significance. Figure 3 shows the shunt diameter was measured as 2 mm at the narrowest portion both on fluoroscopy and IVUS (Atlantis® SR Pro Imaging Catheter, Boston, Natick, MA). The immediate hemodynamic compromise from engaging the AP conduit provided direct evidence that there were few (if any) other systemic collaterals to the pulmonary bed and that indeed her circulation was near-completely dependent on a 2 mm AP conduit minimal luminal diameter. This also alerted us to the need for hemodynamic support during any planned intervention on the AP conduit.
PWA revealed the peak-to-peak gradient across the stenosis as 111 mm Hg (Fig. 4). This peak gradient was felt to be excessive, particularly in light of the IVUS findings, and was consistent with the stenosed AP conduit hypoperfusion the pulmonary bed.
An attempted CT-guided aspiration of the loculated collection was unsuccessful due to a likely organized, solid hematoma. After much discussion between cardiologists, cardiothoracic surgeons, and critical care physicians, it was decided to stent the AP conduit under peripheral cardiopulmonary bypass (PCB) support. Surgery was felt to be prohibitive due to significant morbidity and mortality risk and previous events at surgery a year earlier. The AP conduit was the only major source of pulmonary blood flow and stenting under normal circumstances would have carried significant risks.
Following anesthesia and intubation, femoral PCB was commenced. A formal cutdown was undertaken to insert the sheathes (18 Fr arterial, 24 Fr venous) by Seldinger technique in the left femoral artery and vein, respectively. The venous sheath was advanced to the right atrial junction under guidance of transesophageal echocardiography. This established a stable hemodynamic profile for the duration of the procedure. The flow rate was 3.67 l/min. The total bypass time was 92 min. At the completion of the procedure, PCB was discontinued with decannulation performed, tying down the purse strings to achieve good hemostasis.
The conduit was stented (Fig. 5) with a 10 mm stent in accordance with the IVUS reference measurements. Stent placement and patency was confirmed with subsequent IVUS and CT (Fig. 6). Procedural success was also reflected in stability of hemodynamic parameters off cardiopulmonary bypass. Ideally, PWA of the gradient would be undertaken for comparison with the preprocedure parameters but this would be of secondary academic interest in view of the patient's clinical well-being. It would have been expected that the pulmonary bed would be flooded by systemic pressures and that the gradient across the AP conduit would significantly decline.
Six months postprocedure, she remained well at outpatient review and has not required further hospitalization. Her dyspnea is stable at New York Heart Association classes II and III and she is awaiting elective radiofrequency ablation for atrial flutter. 6MWT exercise capacity was at approximately the premorbid level, 260 m, with oxygen saturation of 84% at rest and 64% during exertion.
PA-VSD is a condition of unknown genetic etiology although microdeletion of 22q11.2 is commonly associated. It accounts for 1–2% of congenital cardiac disease . It is characterized by a lack of direct communication between the RV and PAs, accompanied by intracardiac shunting and a variable degree of PA arborization anomalies. Patients with PA-VSD typically present with cyanosis, failure to thrive, dyspnea, and heart failure. Isolated segmental pulmonary hypertension is also possible if pulmonary blood flow via large MAPCAs is excessive . Palliative surgeries in these patients are a source of major morbidity and mortality . Management may be complicated by stenosis of a surgically created AP conduit .
This case highlights the typical trajectory of these complicated patients and that once faced with the prospect of stenting a critically important AP conduit, PCB was necessary to establish hemodynamic stability. This controlled hemodynamic environment allowed for precise deployment of the AP conduit stent.
PCB, like extracoropeal membrane oxygenation (ECMO), provides artificial oxygenation and blood pressure support in settings such as severe respiratory failure, severe cardiac failure, and multitrauma . However, PCB and ECMO are usually only considered as a last resort due to the high technical demands, cost, and risk of complications, such as bleeding due to the accompanying anticoagulation  or because of neurological sequelae . To our knowledge, this is the first report of using PCB during stenting of an AP conduit.
The clinical presentation of deteriorating exercise capacity and syncope raised the possibility of shunt failure. It was difficult even with a combination of CT and angiography to determine the clinical significance of the “slit-like” stenosis, particularly given the conduit was also kinked. Thus, the use of IVUS and PWA provided compelling anatomical and physiological evidence that the stenosis was critical.
IVUS is generally used to visualize atherosclerotic plaque, particularly in the coronary arteries, to interrogate the lumen-intima interface and identify areas of under-diagnosed plaque burden . Current clinical uses of IVUS include evaluating complex lesions pre-angioplasty and assessing adequacy of stent deployment. The spatial resolution of IVUS allows for luminal assessment even in large arteries . In our unique case, IVUS was used to confirm a severe conduit stenosis and to plan the intervention using proximal and distal reference diameters and length of lesion measurements.
PWA is almost exclusively employed in determining the functional significance of coronary lesions of intermediate grade in the form of fractional flow reserve . In congenital heart disease, PWA has been described as a diagnostic tool for assessing pressures in distal PAs and across AP collaterals . Such interrogation of a surgically created AP conduit has not been described previously in the literature. This method enabled assessment of poststenotic pressure using the pressure sensor wire without needing to place the guiding catheter through the stenosis, which in this case would have caused severe hemodynamic collapse considering the narrow (2 mm) stenosis and the width (6 Fr) of the guiding catheter. Similar to the Caraballo phenomenon described in aortic stenosis , crossing severely stenosed lesions with catheters may also produce a drop in post-stenotic pressure and thus artificially increase the pressure gradient. This error is minimized by only using the 0.014″ RadiWire (Radi Medical Systems, Uppsala, Sweden) to cross the lesion.
Although surgically created AP shunts are uncommon in the overall adult congenital heart disease population, they are often used in patients with pulmonary atresia. Stenosis of such conduits inevitably leads to complications, management difficulties, and often-significant morbidity and mortality. This report illustrates that not only is stenting a critically stenosed AP conduit feasible, it also vastly improved lung perfusion, effort tolerance, and oxygenation. Conduits receiving the majority of cardiac output between systemic and pulmonary circulations are particularly challenging to stent in view of the potentially devastating sequelae from an ill-planned procedure. Although particular attention to wire and catheter manipulation is important, assessment of conduit size, stenosis dimensions, and hemodynamic significance is crucial. Unique to our case was the multimodality assessment (CT, IVUS, and PWA) of the AP conduit and the use of PCB support.
- 5Anticoagulation and coagulation management for ECMO. Semin Cardiothorac Vasc Anesth 2009;13:154–175..