ADVERTISEMENT
Obliteration of a Competitive Forward Flow from the Ventricle After a Bidirectional Cavopulmonary Shunt with an Amplatzer Duct O
February 2003
The early and mid-term results of the Amplatzer Duct Occluder (ADO) (AGA Medical Corporation, Golden Valley, Minnesota) for obliterating flow through a patent arterial duct are highly encouraging.1,2 High occlusion rates, low complication rates and application even in small children have made the percutaneous occlusion of arterial ducts with this device an attractive alternative to surgery.
The ADO has also been used in other clinical situations, such as closure of coronary artery fistula,3 surgical conduit,4 left ventricle to aorta tunnel5 and muscular apical ventricular septal defect.6 All of these procedures demonstrate the inherent potential versatility of the device.
We present a case in which an ADO was implanted at a pulmonary artery banding site in the setting of a post-operative bidirectional Glenn (BDG) shunt in order to obliterate the competitive forward flow from the ventricle to the pulmonary circulation.
Case Report. The patient was a 7-month-old boy (weight, 6.5 kg) with complete transposition of the great arteries and multiple ventricular septal defects (VSD). There was one large inlet, one large muscular trabecular and several small apicals. Due to the complexity of the interventricular septum, a univentricular palliation was carried out. At the age of 3 months, he underwent pulmonary artery (PA) banding in order to decrease pulmonary blood flow, and 4 months later he had a BDG shunt implanted. Although the left ventricular outflow tract (LVOT) was left open, allowing pulsatile forward flow to reach the PAs, the PA banding was tightened at the time of the second operation. Hemodynamic data before surgery were: mean PA pressure, 21 mmHg; pulmonary vascular resistance (PVR), 1.6 units Wood/m2 (assumed VO2); and systemic saturation, 81%. The PAs were free of distortions and were of good size. In the early post-operative period, he presented with low systemic cardiac output, requiring vigorous inotropic support. His systemic saturation was in the low 70’s and progressive and massive edema of the head and neck was noted. Two-dimensional color Doppler transthoracic and transesophageal echocardiography demonstrated pulsatile, high-velocity forward flow in the PAs through the PA banding. There was also flow reversal in the superior vena cava (SVC) during both inspiratory and expiratory phases on mechanical ventilation. In addition, it ruled out obstructions and thrombus formation at the surgical anastomosis. Despite optimization of medical treatment (fluids, inotropic support, hyperventilation and inhaled nitric oxide), clinical conditions deteriorated, requiring further investigation. On the sixth post-operative day, the patient was taken to the catheterization laboratory. A 7 French (Fr) sheath was inserted in the right internal jugular vein. A 6 Fr end-hole balloon catheter (Arrow, Reading, Pennsylvania) was advanced to the pulmonary arteries and subsequently to the left ventricle (LV) through the PA band with the aid of a hydrophilic guidewire (Road-runner; Cook Cardiology, Bloomington, Indiana). Hemodynamic data showed no pressure gradient across the BDG circuit (mean SVC = mean PAs = 23 mmHg). Left ventricular pressure was 80/13 mmHg. SVC and PA angiograms confirmed the echocardiographic findings (Figure 1, left panel). Left ventricular angiogram with a high-flow pig-tail catheter (Cook Cardiology) demonstrated good qualitative function and a “loose” PA band, measuring around 6 mm at the narrowest site (Figure 1, right panel). A standard 0.032´´ curved J-wire (Cook Cardiology) was left inside the left ventricle and the 6 Fr end-hole catheter balloon (Arrow) was pulled through the pulmonary valve over this wire. The inflated balloon anchored at the PA band site, completely occluding the forward flow from the left ventricle (Figure 2, left panel). The side-arm of the 7 Fr sheath was used to monitor the pressure in the SVC, which dropped from 23 mm Hg to 15 mmHg (mean) after LVOT occlusion, with no significant change in the systemic saturation (73–76%). A decision was made to implant an Amplatzer duct occluder at the banding site. After multiple catheter and wire exchanges, a looped 0.035´´ exchange guidewire (Cook Cardiology), 260 cm, was left inside the left ventricle. A 7 Fr-long sheath and dilator used for patent ductus arteriosus (PDA) occlusion (AGA Medical Corporation) was manually shaped (straightened at its distal portion) and advanced over this wire to the left ventricular cavity. An 8-6 mm ADO was loaded in the usual manner on the delivery cable used for PDA occlusion,1,2 which was then advanced through the long sheath. The distal disc (skirt) was opened in the left ventricular cavity by pushing the delivery cable and pulling the long sheath. The whole system (sheath plus delivery cable) was then pulled back, crossing the pulmonary valve with no resistance at this level. After the distal disc anchored at the PA banding site (Figure 2, right panel), the sheath was retracted over the delivery cable, allowing the opening of the body of the device, which became lodged within the PA banding site. A pulmonary angiogram performed through the side arm of the long sheath confirmed a good device position, with the proximal portion lying just above the banding level and not protruding into the bifurcation of the PA. The device was released and a final angiogram through the sheath demonstrated that the flow from the left ventricle was blocked due to lack of the “washing out” phenomenon in the PAs (Figure 3). Pressures in the circuit remained in the mid 15’s (mmHg).
Under low-dose heparin sulfate infusion (10 U/kg/hour), the head and neck congestion resolved in 12 hours and the patient was weaned off mechanical ventilation and inotropic support in 48 hours. Echocardiography demonstrated complete occlusion of the LVOT with no flow abnormalities in the SVC and pulmonary arteries. The patient was discharged home 10 days later on coumadin. He has been doing well for 6 months, with systemic saturation in the low 80’s and normal systemic perfusion.
Discussion. Maintenance of pulsatile forward flow to the PAs is an option when a BDG shunt is created.7–9 However, the amount of this additional flow is difficult to control at surgery. Pulmonary artery banding may be too “loose”, resulting in excessive pulmonary blood flow, ventricular volume overload, and high pulmonary artery and SVC pressures with subsequent impairment of the venous drainage from the upper body.8,10 In the case reported herein, the additional flow from the ventricle was clearly excessive, creating an unfavorable hemodynamic situation. Our patient could not be weaned off mechanical ventilation even with the use of aggressive inotropic therapy and selective pulmonary vasodilation (inhaled nitric oxide). Test occlusion with a catheter balloon in the catheterization laboratory suppressed the hemodynamic detrimental effects of the additional forward flow from the LV.
Although the standard approach to manage this condition would be pulmonary artery ligation or division, there have been occasional attempts at percutaneous closure of post-operative residual forward flow to the pulmonary arteries in the setting of Fontan circulation.11 Double-umbrella device types have been also used for this purpose with mixed results (Dr. Lee Benson, personal communication). Due to their versatility, ADO devices have been successfully used in clinical situations other than PDA and ASD occlusions.3–6 Advantages of these devices include smaller sheaths required for implantation, easy-to-load and easy-to-handle systems, acceptable flexibility, self-centering adjustment at implantation, stenting mechanism of occlusion and stability, retrievability, controlled release mechanism, and high rates of complete closure, which make them attractive to use in different clinical circumstances.1,2 In this case, the decision to use the Amplatzer ductal occluder was based on the underlying anatomy. We selected the 8-6 device because we felt it was large enough to be secured at the narrowest banding site, which measured 6 mm. In addition, the 12 mm distal disc (4 mm larger than the body of the device) was not too big and had enough room to be accommodated between the pulmonary valve and the banding site. Because the device is just 7 mm long, it did not protrude significantly into the pulmonary arteries. One can speculate whether an ASD Amplatzer device could have been used. We felt that there was not enough room. In a 6-7 mm ASD device (probably the most suitable for the narrowest site), the left atrial disc would have been too large to fit in the space between the pulmonary valve and the banding site. “Mushrooming” of the device and significant protrusion into the PAs could also have occurred. A similar phenomenon could have happened with the 8-mm long, 10/8 ADO, had it been used.
Anticoagulation policy after BDG and Fontan operations varies among different institutions, with no clear guidelines.12 The decision to put this patient on coumadin was due to the presence of a highly thrombogenic device close to a slow-flow system. After the endothelialization of the device takes place (usually at 6–12 months), it is anticipated that the anticoagulation regimen is switched to low-dose aspirin.
In conclusion, we believe the ADO may have a role in the management of selected patients suffering from impaired upper body drainage due to the competitive flow from the ventricle in the setting of a bidirectional cavopulmonary shunt. In those cases in which the standard medical treatment does not resolve the high transpulmonary gradient, with high central venous pressures and impaired upper body drainage, the implantation of an ADO may be an important tool to abolish the competitive flow through a loose PA banding. Its properties have advantages over other occlusion devices, making transcatheter approach a logical alternative to surgery in these situations.
1. Faella HJ, Hijazi ZM. Closure of the patent ductus arteriosus with the Amplatzer PDA device: Immediate results of the international clinical trial. Cathet Cardiovasc Intervent 2000;51:50–54.
2. Bilkis AA, Alwi M, Hasri S, et al. The Amplatzer duct occluder: Experience in 209 patients. J Am Coll Cardiol 2001;37:258–261.
3. Pedra CA, Pihkala J, Nykanen DG, Benson LN. Antegrade transcatheter closure of coronary artery fistulae using vascular occlusion devices. Heart 2000;83:94–96.
4. Amin Z, Leatherbury L, Moore HV, Strong WB. A novel use of Amplazer duct occluder. Pediatr Cardiol 2000;21:180–182.
5. Chessa M, Chaudhari M, De Giovanni JV. Aorto-left ventricular tunnel: Transcatheter closure using an Amplatzer duct occluder device. Am J Cardiol 2000;86:253–254.
6. Rodes J, Piechaud JF, Oauknine R, et al. Transcatheter closure of apical ventricular muscular septal defect combined with arterial switch operation in a newborn infant. Cathet Cardiovasc Intervent 2000;49:173–176.
7. Albanese SB, Carotti A, DiDonato RM, et al. Bi-directional cavopulmonary anastomosis in patients under two years of age. J Thorac Cardiovasc Surg 1992;104:904–909.
8. Kobayashi J, Matsuda H, Nakano S, et al. Hemodynamic effects of bi-directional cavopulmonary shunt with pulsatile forward flow. Circulation 1991;84(Suppl III):III-219–III-225.
9. Uemura H, Yagihara T, Kawashima Y, et al. Use of bidirectional Glenn procedure in the presence of forward flow from the ventricles to the pulmonary arteries. Circulation 1995;92(Suppl II):II-228–II-232.
10. Migliavacca F, de Leval MR, Dubini G, Pietrabissa R. A computational pulsatile model of the bi-directional cavopulmonary anastomosis: The influence of pulmonary forward flow. J Biomech Eng 118:520–528.
11. Chaudhari M, Wright JGC, Stumper O. Transcatheter closure of residual ventricle to pulmonary artery communication after Fontan operation. Cardiol Young 2001;11(Suppl 1):264.
12. Monagle P, Cochrane A, McCrndle B, et al. Thromboembolic complications after Fontan procedures. The role of prophylactic anticoagulation. J Thorac Cardiovasc Surg 1998;115:493–498.
