Hybrid Pediatric Cardiac Surgery

Pediatr Cardiol 26:315–322, 2005 DOI: 10.1007/s00246-005-8648-0 Hybrid Pediatric Cardiac Surgery E.A. Bacha, Z.M. Hijazi, Q-L. Cao, R. Abdulla, J.P. Starr, J. Quinones, P. Koenig, B. Agarwala Congenital Heart Center, University of Chicago Hospitals, 5841 S. Maryland Avenue, MC 5040, Chicago, IL, 60637, USA Abstract. Minimally invasive strategies can be expanded by combining standard surgical and interventional techniques. We performed a longitudinal prospective study of all pediatric patients who have undergone hybrid cardiac surgery at the University of Chicago Children’s Hospital. Hybrid cardiac surgery was defined as combined catheter-based and surgical interventions in either one setting or in a planned sequential fashion within 24 hours. Between June 2000 and June 2003, 25 patients were treated with hybrid approaches. Seventeen patients with muscular ventricular septal defects (mVSDs) (mean age, 4 months; range, 2 weeks–4 years) underwent either sequential Amplatzer device closure in the catheterization laboratory followed by surgical completion (group 1A, n = 9) or one-stage intraoperative offpump device closure (group IB, n = 8) with subsequent repair of any concomitant heart lesions. Eight patients with branch pulmonary artery (PA) stenoses (group 2) underwent intraoperative PA stenting or stent balloon dilatation along with concomitant surgical procedures. All patients survived hospitalization. Complications from the hybrid approach were mostly confined to groups 1A and 2. At a mean follow-up of 18 months, 2 group 1A patients died suddenly several months after discharge. All other patients are doing well. Hybrid pediatric cardiac surgery performed in tandem by surgeons and cardiologists is safe and effective in reducing or eliminating cardiopulmonary bypass. Patients with mVSDs who are small, have poor vascular access, or have concomitant cardiac lesions are currently treated in one setting with the perventricular approach. Key words: Hybrid congenital heart surgery Although surgery remains the treatment of choice for most congenital cardiac malformations, interventional cardiology approaches are increasingly being Correspondence to: E.A. Bacha, email: ebacha@surgery.bsd. uchicago.edu used in simple and even complex lesions [18]. However, the percutaneous approach can be challenging due to low patient weight or poor vascular access. In addition, in small infants the passage of large delivery catheters may result in rhythm disturbances and hemodynamic compromise. Furthermore, unusual septal planes, such as in double-outlet right ventricle (DORV) or D-transposition of the great arteries (DTGA), or acute turns or kinks in the pulmonary arteries of tetralogy of Fallot (TOF) patients can make percutaneous procedures challenging if not impossible. On the other hand, surgery also has its limitation. Examples are operative closure of multiple apical muscular ventricular septal defects (mVSDs), adequate and lasting relief of peripheral pulmonic stenosis, or management of a previously implanted stenotic stent [10, 15, 16]. Furthermore, in some complex malformations such as DORV, the presence of multiple mVSDs has been found to be an independent risk factor for early mortality [2, 11]. Thus, combining both venues into a single therapeutic maneuver or in short succession makes sense in terms of reduction of complexity, bypass time, and risk, and it eventually improves the outcome. This report summarizes our experience with hybrid techniques in the management of congenital heart disease. Material and Methods All patients undergoing a hybrid pediatric cardiac procedure—defined as combined catheter-based and surgical interventions in either one setting or in a planned sequential fashion within 24 hours—between June 2000 and June 2003 were prospectively entered into a database. A total of 25 patients were treated with hybrid approaches. Other procedures performed via limited partial lower sternotomies and robotic and thoracoscopic procedures were not included in this study. Informed consent was obtained from the patients’ guardians. The study was approved by the hospital investigation review board. A portion of this study (group 1B) was part of a Food and Drug Administration Investigation Device Exemption clinical trial. 316 Pediatric Cardiology Vol. 26, No. 4, 2005 The patients were divided into three groups: Group 1A: patients with mVSDs treated sequentially (n = 9) Group 1B: patients with mVSDs treated concomitantly via perventricular approach (n = 8) Group 2: patients with branch pulmonary artery stenoses (n = 8) Technical Aspects Patients with Muscular VSDs. From June 2000 to October 2002, nine patients with mVSDs (group 1A) were treated in sequential fashion. The VSDs were closed in the catheterization laboratory using the Amplatzer congenital MVSD device (AGA Medical, Golden Valley, MN, USA), a self-expandable double-disk device made from Nitinol wire mesh. The patients were then taken to the operating room (OR) for completion of the repair (Table 1). Since October 2002, any infant with mVSD or any child with mVSD who also needed an additional surgical procedure was taken to the OR for perventricular closure of the mVSD and repair of concomitant lesion (group 1B, n = 8; Table 2). One patient also had an atrial septal defect (ASD) closed via peratrial approach (Fig. 1). The heart was approached via a median sternotomy, or via a subxyphoid minimally invasive incision without sternotomy if there were no other lesions. Under continuous transesophageal echocardiography (TEE) guidance, the best location for right ventricular (RV) puncture was chosen, making sure that it was away from any papillary muscles but far enough from the septum so as to be able to approach it from perpendicular angle with the needle and wire (Fig. 2). A 5–0 polypropelene purse string was placed at the chosen location. An 18-guage needle (Cook, Bloomington, IN, USA) was introduced in the RV cavity and directed toward the defect to be closed. A 0.035-inch angled guidewire (Boston Scientific, Medi-Tech, Natick, MA, USA) was passed through the needle and manipulated into the left ventricular (LV) cavity through the defect. A 7- to 10-Fr short (8–13 cm) introducer sheath with a dilator was fed over the wire and carefully advanced into the LV cavity. The dilator was removed and the sheath tip positioned in the LV cavity (Fig. 1C and 2C). The appropriate device size was chosen to be 1 or 2 mm larger than the VSD size as assessed by TEE. The device was presoaked in nonheparinized blood for 10 minutes to allow for the tiny fenestrations of the nitinol mesh to thrombose. The device was then screwed to the cable and pulled inside a 6- to 9-Fr loader under blood seal to prevent air bubbles. The device was advanced inside the short delivery sheath until it was seen by TEE to be close to the tip of the delivery sheath. The LV disk was deployed in the LV cavity by gentle retraction of the sheath over the cable. The entire assembly (cable/sheath) was withdrawn gently until the LV disk abutted the septum. Further retraction of the sheath over the cable would deploy the waist inside the septum. Continuous TEE to confirm the device position is of paramount importance. After the position was confirmed, further retraction of the sheath to expand the RV disk was performed. If device position was satisfactory, the device was released by counterclockwise rotation of the cable using the pin vise. A complete TEE study in multiple planes was done to confirm device placement, assess for residual shunting, and determine if there was any obstruction or regurgitation induced by the device. Patients with Branch PA Stenoses. Intraoperative dilatation of an existing stent (n = 5) was done if a stenosed PA stent could not be successfully instrumented via percutaneous route (mostly because of acute angles of branch PA vis-à-vis the main PA). Placement of a stent in a branch PA (n = 3) was performed for diffuse stenosis of the retroaortic portion of the PA during a Fontan procedure (n = 2) and for diffuse LPA stenosis in an older TOF patient (n = 1). (Table 3) All patients had a preoperative angiogram that demonstrated the narrowing (Fig. 4). The stent and balloon sizes were chosen by the cardiologist and were based on measurements made during catheterization. Stents and balloons were placed under direct vision, taking great care not to dissect the branch PA in order to maintain the supportive tissue that surrounds the vessel. The technical difficulty was in knowing how far to advance the stent into the branch PA without injuring the lobar branches take-off. Careful preoperative angiogram-based measurements are extremely important in planning how far to advance the stent inside the branch PA. Fluoroscopy was not used. Balloons were placed without a guidewire. The balloons were inflated to the manufacturer’s recommended pressure for approximately 30 seconds. Results Survival All patients survived their hospital stay. Two patients in group 1A died suddenly at home at 8 and 12 months after discharge, respectively. Autopsies were not obtained. Both had presented with significant biventricular dysfunction in addition to other complex malformations (Ebstein’s and pulmonary hypertension in one and atrial flutter in the other). One of them had been discharged with a significant residual apical VSD. Complications from the Hybrid Technique Group 1A had the largest number of complications. Complications from percutaneous device closure of mVSDs included tricuspid regurgitation in two patients (one from leaflet impingement and one from chordal rupture), RV disk malposition in one, device embolization into the aorta, and ventricular tachycardia. Surgical complications included incomplete apical VSD closure in one. Group IB had few complications, most notably difficulties in deployment of the RV disk due to a heavily trabeculated RV apex in one patient. On discharge echocardiogram, only one patient had significant residual shunting across a mVSD. At 5month follow-up, the residual VSD was insignificant. Two patients in group 2 had significant complications. One was LPA tear occurring from overdistention of LPA stent. This was noticed intraoperatively and repaired by suturing a pericardial patch onto the tear. Another patient with Intraoperative complications Status at follow-up (months) VSD closure, PA plasty, pulmonary valvotomy No Well (34) Closure of residual VSDs, PA plasty and debanding Ligation central shunt, closure membranous VSD, division RVOT muscle bundles Apical VSD closure via apical right ventriculotomy Cavopulmonary shunt No Well (35) No Well (32) No Well (36) 7 months/6.9 kg D-TGA, pulmonary atresia, Device closure apical VSD inlet VSD, apical VSD, (in preparation for sp BT shunt two-ventricle repair) 4 years/18 kg Swiss cheese septum, sp PA band, Multiple VSD device TR (leaflet impingement PA plasty and debanding, pulmonary HTN, tracheostomy placement (complete from device) TV repair closure) TR (chordal rupture), Device retrieval, VSD 3 years/10 kg Swiss cheese septum, sp PA band, Multiple VSD device placement (partial closure) embolized device in aorta closure, TV repair, PA sp debanding, PA aneurysm, plasty, Maze atrial flutter 4 years/24 kg Swiss cheese septum, sp PA band Multiple VSD device placement No PA plasty and debanding (complete closure) No Well (18) No Well (36) 3 months/3.4 kg Biliary atresia, inlet and anterior muscular VSD, Ebstein’s, pulmonary HTN No Age/weight Diagnosis Intervention procedure Complications Surgical procedures DORV, apical VSDs, sp repair, residual VSDs, supravalvular PS 9 months/4.1 kg Anterior muscular midmuscular and inlet VSDs, sp PA band 2 years/12 kg Membranous VSD, apical muscular VSD, PS, sp central shunt Attempt at device closure residual VSDs No Device closure midmuscular VSD Device closure muscular VSD, pulmonary balloon valvotomy No 4 months/4.5 kg Single apical VSD Device closure RV disk malposition with residual shunting No 2 years/13 kg Attempted device closure muscular VSD No V-tach VSD closures, TV repair Bacha et al: Hybrid Pediatric Cardiac Surgery Table 1. Patients with muscular VSDs who underwent successive interventional and surgical procedures Residual Sudden death VSD shunting at home (8) No Moderate shunting across apical septum (16) Sudden death at home (12) DORV, double-outlet right ventricle; HTN, hypertension; PA, pulmonary artery; PS, pulmonary stenosis; RV, right ventricle; RVOT, right ventricular outflow tract; sp, status post; t/d, take-down; TR, tricuspid regurgitation; TV, tricuspid valve; VSD, ventricular septal defect. 317 Table 2. Patients with muscular VSDs who underwent perventricular ventricular septal defect device closure Preop symptoms Device size Additional Exposure (mm) procedures Cardioplegic Intraoperative complications CPBa arrest Postoperative QP/QS complications D/C Echo and status day at follow-up (months) 4 months/4 kg Anterior muscular VSD, ASDb FTT, CHF Xyphoid 10 (VSD) No incision 11 (ASD) No No No 1.2 No 3 17 days/3 kg CHF Xyphoid 12 incision Median 6 sternot. No No No No 1.2 No 4 Coarctation repair and arch augmentation Yes No No 1.1 No 20 Coarctation repair and arch augmentation Band removal and PA plasty Yes No No 1.2 No 18 Yes No Difficult deployment of RV diskc 1.5 Band removal and Yes PA plasty Yes Subaortic VSD patch, pulmonary valvotomy RV outflow patch VSD enlargement and Yes LV aortic baffle t/d BDGd No No 1.1 7 POD 1: reintubation forreperfusion pulmonary edema No 4 Yes No 1.2 No 6 Yes 1.1 LV failure, ECMO, t/d of repair, removal of device and Fontan Pleural effsions 20 Age/weight 20 days/3 kg 14 days/3 kg 2.5 years/ 12 kg Diagnosis large anterior muscular VSD Aortic CoA-arch hypoplasia, large anterior muscular VSD, LV hypoplasia Hypoplastic aortic arch + coarctation, midmuscular VSD s/p PA band for multiple apical VSDs Metabolic acidocis, cardiogenic shock Intubated, Median 8 on PGE sternot. No Median 18 sternot. S/p PA band for post. No muscular VSD 5 months/7 kg DORV, subaortic VSD, No subPS/PS, multiple apical VSDs 3 years/20 kg DORV, TGA, inlet VSD, No apical muscular VSD, hypoplasia LV, s/p PA band and BDG Median 14 sternot. Median 8 sternot. 3 years/15 kg Median 14 sternot. No shunt, asymptomatic (5) No shunt asymptomatic (4) tiny apical VSD, asymptomatic (7) No shunt, asymptomatic (2) 2-mm residual VSD (5) No residual VSD (1) No shunt asymptomatic (8) Widely patent ASD and VSD, asymptomatic (3) ASD, atrial septal defect; BDG, bidirectional Glenn shunt, CoA, coarctation of the aorta; CHF, congestive heart failure; CPB, cardiopulmonary bypass; D/C, discharge; FTT, failure to thrive; LV, left ventricle; PA, pulmonary artery; PS, pulmonary stenosis; POD, postoperative day; RV, right ventricle; s/p, status post; t/d, take-down; VSD, ventricular septal defect a Cardiopulmonary bypass was needed only for additional procedures, not for placement of the device, which was done first in patients 3, 4, 5, and 6 and after PA band removal in patient 5. In patient 8, perventricular closure was accomplished on bypass because the decision to do it was made while on bypass. b TEE showed that the atrial septum had multiple separate openings. Using an aortic punch introduced via a right atrial puncture, a central hole was created in the floppy septum primum. Using the technique described, an 11-mm Amplatzer septal occluder was positioned and deployed centrally in the ASD. c The LV disk obliterated the entire left apical septum, resulting in no residual shunting. However, the RV disk was difficult to deploy due to a severely hypertrophied moderator band and apical RV muscle bundles and prevented the expansion of the RV disk. The RV disk was eventually positioned within the apical muscle bundles with the device microscrew protruding through the RV free wall puncture site. The screw was secured to the epicardium with a pledgetted suture. QP/QS at the end of the procedure was 1.5. d On preop studies, the apical muscular VSD had been judged to be insignificant and was left alone initially. Failure to come off CPB after biventricular repair, and evidence of significant left–right shunting across the muscular VSD by TEE, led to the decision to close the VSD using the perventricular technique. This was successful in that the muscular VSD was found to be completely occluded by TEE and the patient was weaned off of CPB. LV hypoplasia prevented successful biventricular repair, and after a 24-hour period of rest on ECMO, the repair was taken down to an extracardiac Fontan. Bacha et al: Hybrid Pediatric Cardiac Surgery 319 Fig. 1. Peratrial atrial septal defect (ASD) device closure in a 4-month-old infant with a large multifenestrated ASD and anterior muscular ventricular septal defect. (A) Transesophageal echocardiogram in the longitudinal plane demonstrating the presence of a large atrial septal aneurysm with a large superior defect (large arrow). RA, right atrium. (B) Peratrial puncture (large arrow) and passage of a wire (small arrow) through a hole created with an aortic punch. (C) The sheath tip is positioned in the left atrium (arrow). (D) Repeat image in the same plane after deployment of an 11-mm Amplatzer septal occluder demonstrating good device position and elimination of the aneurysm. Fig. 2. Perventricular ventricular septal defect (VSD) device closure in the same infant as in Fig . 1. (A) Large anterior muscular VSD (arrow). RA, right atrium; RV, right ventricle. (B) Perventricular puncture (large arrow) and passage of a wire (small arrow) across the VSD. (C) The sheath tip is positioned in the left ventricle cavity and the device (arrow) is introduced. (D) Both disks have been deployed by pulling back the sheath. The device is still attached to the cable and can still be removed at this stage. unbalanced atrioventricular canal and ventricular outflow obstruction with ventricular dysfunction underwent lateral tunnel Fontan, Damus–Kaye– Stanzel, and stent placement in the retroaortic PA. He was supported on ECMO for postcardiotomy syndrome and underwent a successful heart transplant after 3 weeks of ECMO support. We speculated that the retroaortic stent may have compressed the left main coronary because a slight improvement in ventricular function occurred after tacking the PA superiorly. Subsequent catheterization failed to show coronary compression. with DORV/hypoplastic LV, TGA, and inlet VSD who had undergone a bidirectional Glenn and a PA band at another hospital had failure of biventricular repair. In this patient, a small apical VSD became significant after pressurizing the LV and was device closed via a perventricular approach prior to weaning off cardiopulmonary bypass. The repair failed because of LV dysfunction and hypoplasia. After support on ECMO for 24 hours, the repair was successfully taken to a single ventricle repair (extracardiac Fontan), during which time the mVSD device was removed. Discussion Complications Not Related to the Hybrid Approach One patient in group IB had to be reintubated for reperfusion pulmonary edema that was successfully treated with inhaled nitric oxide. Another patient This study summarizes our institutional experience with hybrid cardiac surgery, in which surgeons and interventional cardiologists work in concert toward reducing the operative trauma and improving BD LPA stent LPA stent Bilateral BD PA stents 2 years/15 kg 6 years/30 kg Truncus arteriosus, sp repair, conduit stenosis, sp bilateral PA stents Taussig–Bing heart/IAA, sp repair, sp LPA stent TOP, sp repair, multiple PA stenoses 2 years/11 kg BD central PA stent Central PA stent placement 9 years/22 kg 2 years/11 kg AVVR, atrioventricular valve regurgitation; BD, balloon dilatation; DORV, double-outlet right ventricle; IAA, interrupted aortic arch; LPA, left pulmonary artery; PA, pulmonary artery; PS, pulmonary stenosis; PV, pulmonary valve; RV, right ventricle; TOF, tetralogy of Fallot. Well (12) Well (32) No No Well (19) Well (24) Well (27) No No LPA tear, repaired with pericardial patch No Postcardiotomy syndrome, ECMO, heart transplant No PV replacement PV replacement Fontan, left AVV plasty Fontan, Maze Fontan, Damus– Kaye–Stanzel RV–PA conduit replacement RV outflow plasty MPA plasty TOF, sp repair, sp LPA stent, PR TOF, sp repair, sp LPA stent, PR Tricuspid atresia, pulmonary atresia, severe AVVR DORV, flutter, sp central PA stent Unb.AVC, sp Glenn and PA band 7 years/25 kg 15 years/60 kg 2 years/18 kg BD LPA stent BD LPA stent LPA stent Intraoperative complications Additional procedures Intraoperative intervention procedure Diagnosis Age/weight Table 3. Patients who underwent intraoperative stent insertion or stent balloon dilatation for branch PA stenoses Well (1) Well (18) Well (30) Pediatric Cardiology Vol. 26, No. 4, 2005 Status at follow-up (months) 320 outcomes. This is in line with the recent characterization of minimally invasive cardiac surgery [6], and we believe that children should and will profit from such advances. Efforts to combine catheterization and surgical techniques have been made before. For example, intraoperative balloon occlusion of Blalock–Taussig shunts and patent ductus arteriosus has been reported [4, 8]; intraoperative balloon dilatation of critical aortic stenosis in neonates and infants [9] and intraoperative stenting have also been described [17]. There are also reports of intraoperative implantation of stent-mounted pulmonary valves via ventricular puncture [19]. Several series of intraoperative device closure, of mVSDs using double-umbrella devices have also been reported [5, 7, 13]. Overall, results were not satisfactory, with high mortality and failure rates. The common approach to all was that the devices were placed under direct vision under cardioplegic arrest. Difficulties in delivering the device or having to suture the device were specifically quoted as a factor in poor outcomes. Interventional approaches are the preferred therapeutic approach for most mVSDs but are limited by several factors, such as patient weight, vascular access, and the need for surgical repair of concomitant lesions. In addition, passing catheters across AV valves introduces the risk of chordal rupture or impingement. A program of sequential management (group 1A) was followed implemented. Later, this was abondoned, and a onetime procedure was adopted. Recently, we reported our initial experience with this simplified technique of off-pump intraoperative device closure of mVSD via a perventricular approach [3]. This technique’s safety has been validated in animal experiments [1], and in fact it has markedly reduced the complication rate (group 1A vs 1B). The one-step approach is simple and required no more than 20 minutes to accomplish in all cases. Many patients with mVSDs present with ventricular dysfunction and do better without prolonged catheterization and operative times. In fact, we believe that both patients in group 1A who died late may have fared better with a the perventricular approach. Advantages over open repair include avoidance of transection of the moderator band or other RV muscle bundles, immediate confirmation of adequate closure, and avoidance of any ventricular incisions. In the absence of associated defects, a minimally invasive approach such as a subxyphoid incision (1–2 cm) can easily be used (two patients in our series). Compared to percutaneous approaches, small patients size and limital vascular access do not pose contraindication to the procedure. By inserting catheters via ventricular puncture, one stays below the subvalvar apparatus of the tricuspid valve and the Bacha et al: Hybrid Pediatric Cardiac Surgery 321 Fig. 3. Intraoperative photograph of a child with subaortic ventricular septal defect (VSD) and muscular VSD. The VSD device is seen in the muscular septum, and some of the sutures and pledgets used to suture the patch are seen at the anteroseptal commissure of the tricuspid valve. Fig. 4. Preoperative angiogram in an older child status post-repair of tetralogy of Fallot and left pulmonary artery (PA) stent. The angle between the main PA (pigtail catheter) and the stented left PA is acute and could not be crossed during catheterization. This child underwent intraoperative balloon dilatation of the stent. risk of chordal damage is markedly reduced. In addition, percutaneous closure of mVSDs in a child palliated with a PA band often results in residual shunting after the PA band is removed. As illustrated here, the current technique offers the possibility to deband the PA and close all VSDs using the perventricular approach in one setting. This technique can also be used for peratrial closure of ASDs. Patients with PA stenoses can also benefit from a hybrid approach. Single-ventricle patients with a large reconstructed ascending aorta, such as occurs after Norwood-style reconstruction, often have compression of the retroaortic portion of the branch PA. It is time-consuming to dissect that area for patching, and dissection can also result in injury to the left main coronary. Delivery of a stent under direct vision via the opened right PA is simple and can be done on the beating heart or even off-pump if a source of pulmonary blood flow such as a cavopulmonary connection is present [14]. However, compression of the left main coronary by a retroaortic PA stent is also well described in the interventional 322 literature and should be carefully avoided by not overdilating the stent [12]. Fluoroscopy was not used, but it may be a very useful tool to ensure correct positioning of the stent prior to dilatation. In conclusion, although not randomized, this study supports the notion that hybrid pediatric cardiac surgery performed in tandem by surgeons and cardiologists is safe and effective in reducing or eliminating cardiopulmonary bypass. We believe that perventricular mVSD closure should be the treatment of choice for any infant with mVSDs, for older children with poor vascular access, and for patients previously palliated with PA bands. In addition, intraoperative PA stenting is a valuable addition to the surgeon’s armamentarium. As experience in hybrid cardiac surgery is gained, thoracoscopic or robotic device delivery into a PA or a septal defect by perventricular puncture may become a reality. References 1. Amin Z, Gu X, Berry JM, Gidding T, Rocchini AL (1999) Perventricular closure of ventricular septal defects without cardiopulmonay bypass. Ann Thorac Surg 68:149–154 2. Aoki M, Forbess JM, Jonas RA, Mayer JEJr, Castaneda AR (1994) Results of biventricular repair for double-outlet right ventricle. J Thorac Cardiovasc Surg 107:338–350 3. Bacha EA, Cao QL, Starr JP, et al. (2003) Perventricular device closure of muscular ventricular septal defects on the beating heart: technique and results. J Thorac Cardiovasc Surg 126:1718–1723 4. Bhati BS, Nandakumaran CP, Shatapathy P, John S, Cherian G (1972) Closure of patent ductus arteriosus during openheart-surgery. Surgical experience with different techniques. J Thorac Cardiovasc Surg 63:820–826 5. Chaturvedi RR, Shore DF, Yacoub M, Redington AN (1996) Intraoperative apical ventricular septal defect closure using a modified Rashkind double umbrella. Heart 76:367–369 6. Chitwood WR, Gulielmos V (2003, March) What is minimally invasive cardiac surgery? Editor’s note, CTSNet.org Pediatric Cardiology Vol. 26, No. 4, 2005 7. Fishberger SB, Bridges ND, Keane JF, et al. (1993) Congential heart disease: intraoperative device closure of ventricular septal defects. Circulation 88:II205–II209 8. Hjortdal VE, Redington AN, de Leval MR, Tsang VT (2002) Hybrid approaches to complex congenital cardiac surgery. Eur J Cardio thorac Surg 22:885–890 9. Hussain A, al Faraidi Y, Abdulhamed J, et al. (2002) Transesophageal echocardiography-guided transventricular balloon dilation of congenital critical aortic stenosis in the neonate and young infant. J Cardiothorac Vasc Anesth 16: 766–772 10. Kitagawa T, Durham LA, Mosca RS, Bove EU (1998) Techniques and results in the management of multiple ventricular septal defects. J Thorac Cardiovasc Surg 115:848–856 11. Kleinert S, Sano T, Weintraub RG, et al. (1997) Anatomic features and surgical strategies in double-outlet right ventricle. Circulation 96:1233–1239 12. Mendelsohn AM, Bove EL, Lupinetti FM, et al. (1993) Intraoperative and percutaneous stenting of congenital pulmonary artery and vein stenosis. Circulation 88:II210–II217 13. Okubo M, Benson LN, Nykanen D, et al. (2001) Outcomes of intraoperative device closure of muscular ventricular septal defects. Ann Thorac Surg 72:416–423 14. Petrossian E, Thompson LD, Hanley FL (2000) Extracardiac conduit variation of the Fontan procedure. Adv Card Surg 12:175–198 15. Seddio F, Reddy VM, McElhinney DB, et al. (1999) Multiple ventricular septal defects: how and when should they be repaired? J Thorac Cardiovasc Surg 117:134–140 16. Serraf A, Lacour-Gayet F, Bruniaux J, et al. (1992) Surgical management of isolated multiple ventricular septal defects Logical approach in 130 cases. J Thorac Cardiovasc Surg 103:437–442 17. Ungerleider RM, Johnston TA, O’Laughlin MP, Jaggers JJ, Gaskin PR (2001) Intraoperative stents to rehabilitate severely stenotic pulmonary vessels. Ann Thorac Surg 71:476–481 18. Waight DJ, Bacha EA, Kahana M, et al. (2002) Catheter therapy of Swiss cheese ventricular septal defects using the Amplatzer muscular VSD occluder. Cathet Cardiovasc Intervent 55:355–361 19. Zhou JQ, Corno AF, Huber CH, Tozzi P, Segesser LKV (2003) Self-expandable valved stent of large size: off-bypass implantation in pulmonary position. Eur J Cardiothorac Surg 24:212– 216