Management of the single ventricle and potentially obstructive systemic ventricular outflow tract Bahaaldin Alsoufi a,⇑ a Division of Cardiothoracic Surgery, Department of Surgery, Children’s Healthcare of Atlanta, Emory University School of Medicine, 1405 Clifton Road, NE, Atlanta, GA 30322 a USA Multi-stage palliation is the current management strategy for the treatment of children with various single ventricle (SV) cardiac malformations. The success of this strategy depends on the presence of favorable anatomic and hemodynamic criteria. Several SV anomalies have the potential of developing systemic ventricular outflow tract obstruction (SVOTO) that might be evident early on or progress later after palliative surgeries. SVOTO could result in ventricular hypertrophy, impaired diastolic function and subendocardial ischemia with subsequent deleterious effects on the SV and disturbance of some of those criteria for a successful multi-stage palliation strategy. Careful identification of SV patients at risk of developing SVOTO and proper planning of the optimal palliation sequence beginning at the 1st stage procedure are vital factors that would affect long-term outcomes in those patients. In the current review, we describe the morphology of SV patients with potential SVOTO risk, surgical procedures that address potential or present SVOTO, and optimal timing of those procedures within the multi-stage palliation chain. We attempt to provide a treatment algorithm for various patients taking into consideration their unique anatomic and physiologic characteristics. Ó 2013 Production and hosting by Elsevier B.V. on behalf of King Saud University. Keywords: Single ventricle, Norwood, Pulmonary artery band, Hybrid Contents Introduction ................................................................................................. 192 Patients with potential risk of SVOTO ............................................................................ 192 Treatment strategies for SV patients with potential SVOTO risk ........................................................ 192 The strategy of initial PAB followed by DKS or VSD/bulboventricular foramen enlargement at the time of BCPC ............ 193 PAB ............................................................................................... 193 BCPC ............................................................................................. 194 DKS............................................................................................... 194 VSD/bulboventricular foramen enlargement ............................................................... 195 The strategy of initial DKS plus shunt at time of 1st stage palliation ................................................ 196 Other treatment strategies .................................................................................. 198 P.O. Box 2925 Riyadh – 11461KSA Tel: +966 1 2520088 ext 40151 Fax: +966 1 2520718 Email: [email protected]URL: www.sha.org.sa REVIEW ARTICLE Received 7 March 2013; revised 19 May 2013; accepted 21 May 2013. Available online 31 May 2013 ⇑ Tel.: +1 404 785 6330; fax: +1 404 785 6266. E-mail address: balsoufi@hotmail.com 1016–7315 Ó 2013 Production and hosting by Elsevier B.V. on behalf of King Saud University. Peer review under responsibility of King Saud University. URL: www.ksu.edu.sa http://dx.doi.org/10.1016/j.jsha.2013.05.003 Production and hosting by Elsevier
Management of the single ventricle andpotentially obstructive systemic ventricularoutflow tract
1016–7315 � 2013 Production and hosting by Elsevier B.V. on behalf of King Saud University.
Peer review under responsibility of King Saud University.
URL: www.ksu.edu.sa
http://dx.doi.org/10.1016/j.jsha.2013.05.003 Production and hosting by Elsevier
Bahaaldin Alsoufi a,⇑
a Division of Cardiothoracic Surgery, Department of Surgery, Children’s Healthcare of Atlanta, Emory University Schoolof Medicine, 1405 Clifton Road, NE, Atlanta, GA 30322
a USA
Multi-stage palliation is the current management strategy for the treatment of children with various single ventricle(SV) cardiac malformations. The success of this strategy depends on the presence of favorable anatomic andhemodynamic criteria. Several SV anomalies have the potential of developing systemic ventricular outflow tractobstruction (SVOTO) that might be evident early on or progress later after palliative surgeries. SVOTO could result inventricular hypertrophy, impaired diastolic function and subendocardial ischemia with subsequent deleterious effectson the SV and disturbance of some of those criteria for a successful multi-stage palliation strategy. Carefulidentification of SV patients at risk of developing SVOTO and proper planning of the optimal palliation sequencebeginning at the 1st stage procedure are vital factors that would affect long-term outcomes in those patients. In thecurrent review, we describe the morphology of SV patients with potential SVOTO risk, surgical procedures thataddress potential or present SVOTO, and optimal timing of those procedures within the multi-stage palliation chain.We attempt to provide a treatment algorithm for various patients taking into consideration their unique anatomic andphysiologic characteristics.
� 2013 Production and hosting by Elsevier B.V. on behalf of King Saud University.
Keywords: Single ventricle, Norwood, Pulmonary artery band, Hybrid
Multi-stage palliation is the current manage-ment strategy for the treatment of children
with various single ventricle (SV) cardiac malfor-mations. The 1st stage neonatal palliative surgeryvaries based on underlying anatomy. However,the general aim is to achieve the following fourobjectives: (1) unobstructed systemic cardiac out-put; (2) controlled source of pulmonary bloodflow; (3) reliable source of coronary blood flow;and (4) unobstructed egress of pulmonary venouseffluent across the atrial septum. At a 2nd stage,the bidirectional cavo-pulmonary connection(BCPC) is performed by suturing the superiorvena cava to the pulmonary artery and is typicallydone at four to six months of age. At a 3rd stage,the Fontan procedure is performed to channelthe remaining systemic venous return from theinferior vena cava to the pulmonary artery and istypically done at two to four years of age.
Good systolic and diastolic ventricular function,stable rhythm, competent atrioventricular valve,and unobstructed systemic ventricle outflow tract(SVOT) are among some of the well-identifiedstringent hemodynamic and anatomic criteria forsuccessful completion of all palliation stages andachievement of a stable and durable Fontancirculation.
Several SV anomalies have the potential ofdeveloping SVOT obstruction (SVOTO) that mightbe evident early on or progress later after pallia-tive surgeries [1–5]. SVOTO could result in ven-tricular hypertrophy, impaired diastolic functionand subendocardial ischemia with subsequentdeleterious effects on the SV and development ofunfavorable hemodynamic conditions that mightnegatively influence the success of palliation strat-egy towards the Fontan circulation [1–5].
Therefore, careful identification of SV patients atrisk of developing SVOTO and proper planning ofoptimal palliation sequence beginning at 1st stageprocedure are vital factors that would affect long-term outcomes in those patients.
Patients with potential risk of SVOTO
Several groups of children with SV are at poten-tial risk of developing SVOTO. The larger group
comprises patients in whom the aorta arises froman outlet chamber that is connected to the domi-nant ventricle via a bulboventricular foramen ora ventricular septal defect (VSD). This group in-cludes patients with double inlet left ventricle(DILV) or tricuspid atresia associated with ventri-culo-arterial discordance. Also included in thisgroup are some infants with double outlet rightventricle (DORV) and remote VSD that are unsuit-able for biventricular repair; and those with dex-tro- or levo-transposition of the great arteriesassociated with VSD and hypoplastic right ventri-cle. In the above listed anomalies, the systemicblood flow traverses through the bulboventricularforamen or VSD from the dominant ventricle tothe rudimentary ventricle that serves as an outletchamber from which the aorta arises. Those pa-tients are at risk of developing SVOTO as a resultof reduction in VSD/bulboventricular foramensize early after pulmonary artery banding (PAB)or late after volume unloading with BCPC [1–6].The second group of patients at potential risk ofdeveloping SVOTO includes those with well-developed subaortic conus that could enlarge withtime to cause subaortic obstruction, as for examplewith patients with DORV, often associated withhypoplastic left ventricle and mitral stenosis/atre-sia [2,4,6]. Another group of patients at potentialrisk of developing SVOTO includes those withunbalanced atrioventricular septal defect of rightdominance in whom flow to the aorta might be-come compromised by a small inlet VSD or re-stricted by atrioventricular valve tissue, inaddition to the presence of elongated left ventric-ular outflow tract and accessory atrioventricularvalve tissue, all of which could contribute to theobstruction of systemic cardiac output towardsthe aorta [6].
Treatment strategies for SV patients withpotential SVOTO risk
Given the impending negative effects of SVOTOon hemodynamic conditions and future outcome,the treatment plan of patients at potential SVOTOrisk should be carefully deliberated beginning atthe time of 1st stage palliative surgery and indi-vidualized to allow the design of the most proper
surgical sequence that is suitable for patient’sanatomy. Variations in management strategiesare related to the type of procedure utilized to ad-dress SVOTO and timing of that procedure duringthe multi-stage palliation chain.
Several options are available at the surgeon’sdisposal to alleviate SVOTO. These include theDamus-Kaye-Stansel operation (DKS), VSD/bulboventricular foramen enlargement, resectionof subaortic conus and palliative arterial switchoperation [4,5,7]. Each procedure is associatedwith distinctive advantages and disadvantages.The decision to address SVOTO at the time of1st palliation stage or to delay that until the 2ndstage depends on many factors including the pres-ence of arch obstruction, size of VSD/bulboven-tricular foramen, and presence of subaorticobstruction.
In general, VSD/bulboventricular foramenenlargement is not performed in neonates dueto difficulties with exposure and tissue fragility.Therefore, strategies that address SVOTO inneonates generally include DKS or palliativearterial switch operation. On the other hand,strategies that address SVOTO at the time ofBCPC involve neonatal palliation with a PABfollowed at 2nd stage BCPC combined witheither DKS or VSD/bulboventricular foramenenlargement.
The strategy of initial PAB followed by DKS orVSD/bulboventricular foramen enlargement at thetime of BCPC
PABThose SV anomalies are usually associated with
excessive pulmonary blood flow that requiresearly control with a PAB to prevent developmentof pulmonary vascular disease [4–6,8–14]. The
Figure 1. Left thoracotomy view of a single ventricle patient with normalarch and patent ductus arteriosus (A). The patient underwent extended endarteriosus, and main pulmonary artery banding as a 1st stage palliation
main advantage of the strategy of early PABfollowed by BCPC and DKS or VSD/bulboven-tricular foramen enlargement is avoidance of car-diopulmonary bypass and deep hypothermiccirculatory arrest in the neonatal period. PABcan also be performed via a left thoracotomythereby facilitating 2nd stage procedure (Fig. 1).Nonetheless, there are several drawbacks ofPAB related to the difficulty of obtaining ade-quate pulmonary vascular protection with PAB,possibility of pulmonary artery distortion if thePAB was placed too distally, risk of distortion ofpulmonary valve with subsequent regurgitationif the PAB was placed too proximally and, mostimportantly, occurrence of ventricular hypertro-phy that might accelerate the rate of VSD/bulbo-ventricular foramen narrowing with thesubsequent early development of SVOTO[1,3,15]. Another complicating issue in initialPAB strategy is the frequent incidence of aorticcoarctation at original presentation. While thereare several series reporting aortic coarctation re-pair and PAB via a left thoracotomy as 1st stagepalliation, that approach might not be applicablein patients with hypoplastic aortic arch necessi-tating a sternotomy and cardiopulmonary bypassfor adequate arch augmentation. In addition, itmight be challenging to carry out adequate band-ing when the PAB is performed via a left thora-cotomy because of patient’s position and lungcollapse.
A landmark study by Freedom et al. from Toron-to reviewed 43 patients with SV who did not haveevidence of SVOTO at time of PAB done at a mean0.21 year of age. The study noted that 31/43 pa-tients (72%) developed SVOTO at a mean age of2.52 years, with an especially increased risk in84% of patients with DILV or tricuspid atresiaassociated with ventriculo-arterial discordance.
ly related great vessels who has aortic coarctation, hypoplastic aortic-to-end repair of aortic coarctation, ligation and division of the ductusprocedure (B).
The study concluded that PAB produces myocar-dial hypertrophy with resultant decrease inVSD/bulboventricular foramen size that acceler-ates potential SVOTO and they cautioned againstPAB in those patients with potential SVOTO risk[1]. Another study by Matitiau et al. from Bostonexamined the indexed bulboventricular foramenarea in patients who had undergone various initialpalliative procedures. The authors found that allpatients who did not have an initial DKS (themajority had PAB) and subsequently developedearly SVOTO had an initial bulboventricular fora-men area index less than 2 cm2/m2. They alsonoted that patients with aortic coarctation hadsmaller bulboventricular foramen area index attime of initial presentation, and were likely to beat higher risk of developing SVOTO after PAB[3]. Accordingly, initial VSD/bulboventricularforamen size and presence of aortic coarctation/hypoplastic aortic arch should be taken intoconsideration when deciding on the type of 1stpalliative surgery.
BCPCThe performance of BCPC results in volume
unloading that might further decrease VSD/bulboventricular foramen size and accelerate therate of SVOTO development. In a study by Donof-rio et al. from Philadelphia, the authors noted adecrease in mean VSD/bulboventricular foramenarea index following PAB or BCPC with greaterdiminution noted after BCPC (41 ± 19%) than afterPAB (25 ± 28%) [15]. While volume unloading oc-curs primarily at time of BCPC, the decrease insystemic ventricle end diastolic volume continuesfor years after the Fontan operation and subse-quently the risk of SVOTO might appear late aftercompletion of the Fontan circulation [6].Therefore, there is a general agreement that pro-cedures addressing potential SVOTO should beperformed at the time of BCPC to avoid lateSVOTO which would be more difficult to manage,would require an unnecessary re-operation, andcould be associated with inferior outcomes.
DKSThe DKS operation was originally described for
the treatment of selected patients with transposi-tion of the great arteries. Subsequently, there havebeen reports of successful management of pa-tients with SVOTO utilizing DKS operation[4,9,13]. Furthermore, DKS use in SV patients ex-panded into a planned procedure at the time ofBCPC in patients with anatomic substrate to de-velop SVOTO, regardless of the presence of the
actual gradient between the systemic ventricleand aorta at time of BCPC [4,6,10–14].
Advantages of performing DKS concomitantwith BCPC include the ability to adequately ad-dress most SVOTO cases, ability to perform archreconstruction if needed, and low risk ofdevelopment of heart block. In addition, the mod-ified-in-series circulation after BCPC and DKS ismore stable than the balanced parallel circulationafter DKS with aortopulmonary shunt with noassociated risk of steal from the coronary and sys-temic circulation. Potential disadvantages of DKSinclude distortion of the semilunar valves withsubsequent regurgitation and increased SV vol-ume load, in addition to the risk of compressionof left pulmonary artery or left main bronchus,especially in patients with great vessels orienta-tion that is not side by side [4]. While there is atheoretical concern that prior PAB could resultin pulmonary valve distortion with consequentregurgitation at time of DKS, results fromnumerous studies of concomitant BCPC andDKS after PAB showed that the incidence ofsemilunar valve regurgitation is low despite priorPAB [4,6,8–14].
While initial descriptions of DKS involved anend-to-side anastomosis between the pulmonaryartery and aorta with or without patch augmenta-tion, a modified side-by-side double barrel DKSanastomosis has been suggested by Lambertiand colleagues and shown to better preserve theshape of pulmonary sinus and native aortic sinusof Valsalva [16] (Fig. 2). Maintenance of originalsemilunar valve and sinus geometry in their na-tive positions might offer the advantage of avoid-ing potential distortion. In a study by Fujii et al.from Okayama, the authors compared 13 patientswho had undergone end-to-side DKS with 34 whohad undergone double barrel DKS. The studyfound that although both techniques wereassociated with 100% freedom from SVOTO re-interventions, patients who had undergone end-to-side DKS had more frequent deterioration inpulmonary regurgitation than double barrel DKS(4/11 vs. 1/34, P = 001) [11].
Results of concomitant BCPC and DKS havebeen published in several series and showed lowmortality without much added operative risk,excellent immediate SVOTO relief and freedomfrom recurrent obstruction, good and durablesemilunar valve function, and good functional sta-tus following Fontan operation [4,6,8–14]. In a ser-ies from the King Faisal Specialist Hospital &Research Center in Riyadh, Saudi Arabia, 36 chil-dren with SV underwent DKS at time of BCPC
Figure 2. View from a single ventricle patient with normally related great vessels at time of 2nd stage procedure following initial pulmonaryartery banding. The band was removed, the main pulmonary artery was transected, the branch pulmonary arteries end was closed with a patch,and an incision at the medial side of the aorta was created (A). The patient underwent an end to side Damus-Kaye-Stansel anastomosis,supplemented with patch material, in addition to Glenn bidirectional cavopulmonary connection (B). View from a patient with single ventricleand normally related great vessels at time of 2nd stage procedure following initial pulmonary artery banding. The band was removed, the mainpulmonary artery was transected at the site of the band, the aorta was transected at almost similar height above the sinotubular junction, and thebranch pulmonary arteries end was closed primarily (C). The patient underwent a side to side double barrel Damus-Kaye-Stansel anastomosis,in addition to Glenn bidirectional cavopulmonary connection (D).
(n = 29) or Fontan (n = 7). Underlying anatomy wasDILV (n = 18), DORV (n = 8), unbalanced atrioven-tricular septal defect (n = 4) and other (n = 6). Priorpalliation included PAB (n = 35), aortic coarctationrepair (n = 11) and atrial septectomy (n = 8). Attime of DKS, 17 patients (47%) had no SVOT gra-dient and 19 patients (53%) had SVOT gradient(median 19 mmHg, range 4–80 mmHg). Overall,survival was 89% and 83% at one month and fiveyears, respectively. None of the deaths were re-lated to SVOTO or DKS complications. Whenpresent, SVOT gradient decreased from23.4 ± 18.7 mmHg pre-operatively to 0 after DKS(p < 0.001). At last follow-up, none of the patientsdeveloped any SVOT gradient; 78% of them hadzero or trivial aortic/neo-aortic valve regurgitationwhile 22% had mild regurgitation. None of the pa-tients had evidence of compression of left pul-monary artery or bronchus; 81% of patients hadreached or were suitable candidates awaiting final
palliative surgery and all survivors were in excel-lent functional status except for one patient withprotein-losing enteropathy [17].
VSD/bulboventricular foramen enlargementSeveral groups have advocated VSD/bulboven-
tricular foramen enlargement at time of BCPC orlater as an alternative strategy for the manage-ment of SV patients with SVOTO [2,4,5,7,18–20](Fig. 3). VSD/bulboventricular foramen enlarge-ment is associated with drawbacks related to thenecessity of performing a ventriculotomy in manycases with subsequent risk of forming ventricularaneurysm. Moreover, the intimate relationship ofthe VSD/bulboventricular foramen with the semi-lunar valves and atrioventricular valves might dis-allow adequate enlargement thus leading toinsufficient SVOTO relief or recurrence ofobstruction. Finally, VSD/bulboventricular fora-men enlargement could result in a high incidence
REV
IEW A
RTICLE
Figure 3. View from a single ventricle patient with ventriculo-arterial discordance who underwent ventriculotomy into the rudimentary ventricleand enlargement of the bulboventricular foramen. The dashed line indicates the safe area for enlargement while the circles demonstrate the likelysite for conduction system (A). Following enlargement of the bulboventricular foramen, the ventriculotomy is closed with a patch (B).
of complete heart block necessitating permanentpacemaker implantation [2,4,18–20]. Approach toVSD/bulboventricular foramen enlargement var-ies based on anatomy and location of VSD, andcan be performed through the aortic valve, atrio-ventricular valve or frequently via a ventriculot-omy. Similarly, site of enlargement depends onventricular looping and dominant ventricle mor-phology and requires knowledge of potentialcourse of conduction system in various morpholo-gies. When the accessory chamber of DILV is lo-cated on the right with a D loop (SDD),bulboventricular foramen could be enlarged supe-riorly as the conduction bundle is located inferi-orly. When the accessory chamber is located tothe left side, with an L loop (SLL), the conductionbundle is supposed to run superiorly and the inci-sion should therefore be done inferiorly. VSDenlargement might be particularly problematic inpatients with unbalanced atrioventricular septaldefect as it might interfere with atrioventricularvalve function following enlargement. In a reportby Pass et al. from New York, nine patients under-went bulboventricular foramen enlargement fol-lowing SV palliation; there was one operativeand one late death, while one patient requiredpermanent pacemaker insertion and three pa-tients (33%) required re-operation for recurrentSVOTO [18]. Therefore, due to the clear advantageof DKS over VSD/bulboventricular foramenenlargement, many surgeons reserve VSD/bulbo-ventricular foramen enlargement to children withpulmonary regurgitation, pulmonary stenosis oratresia, DORV with restricted VSD, or patientswho develop late SVOTO after BCPC with inabil-ity to perform DKS due to sutured pulmonaryvalve cusps [2,4,5,7].
The strategy of initial DKS plus shunt at time of1st stage palliation
The strategy of performing PAB +/� aorticcoarctation repair as initial palliative surgery forSV with excessive pulmonary blood flow is associ-ated with numerous disadvantages, listed earlier.Results following PAB in those patients have notbeen optimal with significant operative and latemortality risk and a high incidence of poor candi-dacy to progress through palliation stages towardsFontan circulation [4,6,8–14,21–24].
In a study by Franklin et al. from London, theauthors found that survival following PAB was77% at one year and 45% at five years for thosewho had PAB alone and was 44% at one yearand 22% at five years for those who had PABand aortic coarctation repair.[23] In another studyfrom the same authors, suitability for Fontan oper-ation was 42% for those who had PAB alone and8% for those who had PAB and aortic coarctationrepair.[24] In a study by Miura et al. from Osakaexamining 21 SV patients who had undergone ini-tial PAB, the authors noted that the bulboventric-ular foramen area index decreased from a medianof 103–75% necessitating DKS at 2nd stage in themajority of patients; however 6/21 (29%) patientsdied before reaching 2nd palliation stage [14]. Inanother study from Fiore et al. from St. Louisand Indianapolis examining 27 SV patients whohad undergone initial PAB with concomitant aor-tic coarctation in 18/27 (66%) patients, the authorsnoted that all those patients required DKS at 2ndstage. However, an aortopulmonary shunt wasnecessary as a sole or supplemental source of pul-monary blood flow in 11/27 children with highmortality in 6/11 of those patients (55%) [12].
Figure 4. View from a single ventricle patient with ventriculo-arterial discordance who underwent a Norwood-type 1st stage palliation. Themain pulmonary artery and ascending aorta were transected above the sinotubular junction, the hypoplastic aortic arch was opened and thecoarctation area was excised (A). The reconstruction was completed by performing a Damus-Kaye-Stansel anastomosis, arch reconstruction andaugmentation with patch. The source of pulmonary blood flow is provided with an aortopulmonary shunt, or a right ventricle to pulmonaryartery shunt (Not shown) (B).
Due to those disappointing results, severalauthors recommended that SVOTO is addressedat time of neonatal 1st stage procedure with DKSplus aortopulmonary shunt, or a Norwood-type1st stage palliation if concomitant arch reconstruc-tion was needed [25,26] (Fig. 4). Advantages of thisapproach include the ability to address SVOTOearly on and concomitantly address all types ofarch pathology. This approach has the potentialof avoiding development of ventricular hypertro-phy, thereby improving candidacy for a Fontanprocedure. Moreover, detrimental effects of pul-monary regurgitation as a result of proximal dis-placement of the PAB or pulmonary arterystenosis as a result of distal displacement of thePAB are avoided by the strategy of neonatal 1ststage DKS plus aortopulmonary shunt. Finally, ashunt offers an advantage of having a uniformdiameter with more predictable pulmonary bloodflow as compared to PAB that is harder to adjustas that is carried out under non-physiologic condi-tions that can be affected by anesthesia, positiveventilation and open chest, and can also changelater on with a decrease in pulmonary vascularresistance and pulmonary artery remodeling.The obvious disadvantages of this strategy, how-ever, include cardiopulmonary bypass and deephypothermic circulatory arrest needs during theneonatal period, complex post-operative course,shunt related complications, and early and interimmortality risk [27].
In a study by Mosca et al. from Ann Arbor, theauthors reported outcomes of 38 patients with TA
or DILV and ventriculo-arterial discordance whohad undergone Norwood 1st stage palliation. Theyfound on echocardiographic follow-up thatalthough absolute bulboventricular foramen sizeincreased in approximately 50% of the patients,when indexed to body surface area there was anoverall decrease with time. Their operativemortality with the modified Norwood operationwas 8% and five year survival was 71% with noincidence of recurrent SVOTO or significant semi-lunar valve insufficiency. They concluded thattheir outcomes were superior to those reportedwith other treatment strategies in that group ofpatients [25]. Outcomes of the Norwood operationin the current era have appreciably improvedthrough refinements in surgical technique, perfu-sion strategy, pre-operative stabilization, post-operative care and home monitoring; and expectedsurvival has increased, especially in the subset ofpatients with tricuspid atresia or DILV [27]. A morerecent study from Bradley et al. from Charlestonreported outcomes of 21 SV patients with excessivepulmonary blood flow who were treated withaortopulmonary shunt rather than PAB along withprophylactic DKS; they had one operative mortal-ity and 90% five year survival supporting theadvantage of this treatment strategy over thealternative PAB +/– aortic coarctation repair [26].Furthermore, a study by Serraf et al. from Pariscompared outcomes in SV patients who hadundergone various initial palliations and foundthat four year survival was better in those whohad undergone DKS or palliative arterial switch
Figure 5. View from a single ventricle patient with ventriculo-arterial discordance who underwent a palliative arterial switch operation as 1ststage palliation. Neo-aorta was reconstructed after coronary transfer, and a Lecompte maneuver was accomplished by placing the branchpulmonary arteries anterior to the reconstructed neo-aorta (A). The neo-pulmonary artery was reconstructed with a patch. Very often, theaddition of either an aortopulmonary shunt or a pulmonary artery band is necessary. In this example, an aortopulmonary shunt is shown (B).
operation than those who had undergone PAB andaortic coarctation repair (67% vs. 40%, P < 0.05)[22].
Other treatment strategies
Palliative arterial switch operationSeveral groups reported experience with pallia-
tive arterial switch operation for treatment of SVanomalies such as tricuspid atresia or DILV andventriculo-arterial discordance as an alternativeto neonatal DKS or Norwood operations [4,28–31] (Fig. 5). Palliative arterial switch operation re-sults in alignment of the systemic ventricle withthe systemic outflow tract and therefore elimi-nates SVOTO early on and decreases the chanceof development of ventricular hypertrophy anddiastolic dysfunction. Advantages of the palliativearterial switch operation include the ability to ad-dress SVOTO in the neonate, establishment oflaminar flow to the aorta potentially decreasingrecurrent arch obstruction, and finally low risk ofleft pulmonary artery or main bronchus compres-sion.[4] Nonetheless, potential disadvantages arerelated to the usual long-term concerns with thearterial switch operation such as late coronaryproblems, neo-aortic root dilatation, stretch andstenosis of the pulmonary arteries after theLecompte maneuver. Moreover, it was originallythought that the restricted bulboventricularforamen would help prevent excessive pulmonaryblood flow after the palliative arterial switchoperation. However, this effect is unpredictable,and often a PAB on the newly reconstructedpulmonary artery is needed to limit pulmonaryblood flow while an aortopulmonary shunt might
occasionally be necessary in patients with a veryrestricted bulboventricular foramen and dimin-ished pulmonary blood flow [28–31]. Furthermore,adjustments of PAB or a latter addition of aorto-pulmonary shunt can occasionally become neces-sary as bulboventricular foramen diminishes insize with subsequent decrease in pulmonaryblood flow after discharge [28–31]. Only a fewpatient series that underwent palliative arterialswitch operation exist in the literature. In a seriesfrom Ceresnak et al. from New York, nine patientsunderwent palliative arterial switch operation asinitial palliation with one operative and one latedeath while the seven remaining patients reachedor qualified for Fontan operation [30]. In a morerecent series by Heinle et al. from Houston, 14 pa-tients underwent palliative arterial switch opera-tion with 13/14 (93%) requiring concomitant archreconstruction. In their series, 6/14 (43%) patientsrequired concomitant PAB for excessive pulmon-ary blood flow while 1/14 (7%) required aorto-pulmonary shunt. There were no deaths and allpatients proceeded to 2nd stage BCPC. However,four patients required additional proceduresbefore BCPC that included addition of aortopul-monary shunt in three patients and PAB adjust-ment in one patient. Overall, 11/14 reachedFontan while the remaining 3/14 qualified forFontan surgery. While there was no incidence ofSVOTO, 64% of patients required pulmonaryartery augmentation at time of Fontan.[31]
Hybrid strategyIn the past decade, there has been an increased
experience in hybrid management of neonates
Figure 6. View from a single ventricle patient who underwent hybrid-type 1st stage palliation with branch pulmonary artery banding and ductalstenting (A). At time of 2nd stage surgery, the reconstruction was completed by performing a Damus-Kaye-Stansel anastomosis, archreconstruction and augmentation with patch, pulmonary artery de-banding and augmentation, in addition to Glenn bidirectional cavopulmonaryconnection (B).
with hypoplastic left heart syndrome and otherrelated SV anomalies [27,32–35]. Hybrid strategyconsists of a 1st stage procedure that is done inthe neonatal period and involves bilateral PAbranch banding, ductal stenting with atrial septos-tomy or stenting as needed. The 2nd stageprocedure is usually done at four to six monthsof age and consists of stent removal, aortic archreconstruction, DKS and BCPC (Fig. 6). Advanta-ges of this hybrid strategy are related to avoidanceof cardiopulmonary bypass and cardioplegicarrest in the neonate with potential improvedlong-term myocardial function, deferment of
Figure 7. A proposed management algorithm for the treatment of singleoutflow tract obstruction.
reconstruction of the aortic arch and deep hypo-thermic circulatory arrest to an older age with po-tential improved long-term neurologicaloutcomes; and finally exit of the operation at 2ndstage with BCPC and a modified-in-series circula-tion that is more stable than a balanced circulationafter a Norwood operation with a shunt [27,32–35].Nonetheless, potential disadvantages of hybridstrategy include ductal and atrial stent migration,band migration, stent stenosis and thrombosis,all requiring close observation for potential needfor catheter-based re-interventions or early 2ndstage surgery [27,32–35]. Reported outcomes of
ventricle patients at potential risk of developing systemic ventricle
this hybrid strategy from experienced centers areencouraging, and it’s expected that additionalknowledge with this strategy and refinements oftechnique and devices would further improve out-comes and support the role of hybrid strategy inthe palliation of various SV anomalies. All theoret-ical long-term advantages of hybrid strategy areyet to be proven in order to replace other treat-ment strategies especially as Norwood outcomesin patients with SV, other than hypoplastic leftheart syndrome, are superior to those with hypo-plastic left heart syndrome [27].
Management of the old SV patients with SVOTOManagement of older SV patients with SVOTO
who have already completed palliation stages ischallenging [2]. It is very doubtful that the pul-monary valve in the transected pulmonary arterycould be utilized for a DKS operation especiallyas most surgeons over sew the pulmonary valvecusps at the time of BCPC or Fontan to preventthrombus formation [36]. Nonetheless, there arereports in the literature of late performance ofDKS in patients who previously had transectionof the main pulmonary artery at the time of BCPC[37]. At the King Faisal Specialist Hospital inRiyadh, one patient who had undergone transec-tion of the main pulmonary artery at time of BCPCfollowed by Fontan developed SVOTO nine yearsafter Fontan with 45 mm HG gradient across therestricted bulboventricular foramen. That patientunderwent re-operation with DKS that success-fully alleviated SVOTO with trivial neo-aorticregurgitation on late follow-up [17].
In most cases, the only available option is VSD/bulboventricular foramen enlargement that usu-ally requires a ventriculotomy through the rudi-mentary chamber underneath the aortic valve[2]. Late VSD/bulboventricular foramen enlarge-ment is challenging and long-term results are dis-turbing due to risk of heart block, inadequateSVOTO relief, recurrent obstruction, atrioventric-ular valve dysfunction, and hemodynamic effectsof hypertrophy on Fontan circulation. One inter-esting report from Meadows et al. from Bostondescribed eight patients who presented withsevere SVOTO following Fontan done in complexDORV. All had hypertrophied, hypertensive, su-pra-systemic ‘isolated LV chambers’, and threepatients had a false aneurysm of the left ventricle.Five patients underwent VSD creation and threeunderwent enlargement of existing VSD. Initialintervention resulted in a decreased gradient from76.9 mmHg to 20.3 mmHg (P = 0.004). There wasno procedural mortality or sustained heart block.
Two patients had moderate-to-severe atrioven-tricular valve regurgitation, and one required sur-gical repair. At last follow-up, all VSDs remainedpatent but with recurrent obstruction in themajority of cases caused by muscular hypertrophybeyond the stent margins [38].
Management algorithm of SV patients withSVOTO
Management strategy of patients at potentialSVOTO risk should be carefully planned begin-ning at the time of 1st palliative procedure andindividualized to patient’s anatomy (Fig. 7). Deci-sion-making in neonates depends on the presenceof aortic coarctation/hypoplastic aortic arch, sizeof VSD/bulboventricular foramen and presenceof subaortic obstruction. In neonates with smallbulboventricular foramen diameter less than aor-tic annulus or bulboventricular foramen area in-dex less than 2 cm2/m2, or with evidence of aorticor subaortic obstruction, neonatal DKS + aorto-pulmonary shunt is indicated. In those patientswith associated aortic coarctation/hypoplastic aor-tic arch, the Norwood operation is performed. Inpatients in whom bulboventricular foramen islarge and who have no evidence of subaorticobstruction or aortic coarctation/hypoplastic aorticarch, it is justified to perform a PAB followed byclose echocardiographic follow-up and a plannedDKS at time of BCPC. In patients in whom bulbo-ventricular foramen is large and who have no evi-dence of aortic or subaortic obstruction but withaortic coarctation/hypoplastic aortic arch, initialNorwood palliation is indicated, although somesurgeons report acceptable results of PAB andaortic coarctation repair with close echocardio-graphic follow-up and a planned DKS at the timeof BCPC. In our experience, aortic coarctation inthis subset of patients is often associated with se-vere hypoplastic aortic arch, and adequate archrepair via a thoracotomy is often not possible thusnecessitating a Norwood operation in the majorityof patients. VSD/bulboventricular foramenenlargement should be reserved for patients withsevere pulmonary regurgitation following PAB,DORV with non-committed VSD, those with pul-monary stenosis or atresia, and, of course, olderSV patients with SVOTO. Finally, the role of hy-brid 1st stage palliation with ductal stenting andbranch pulmonary artery banding is evolvingbut the advantage in those ‘favorable’ non-hypoplastic left heart syndrome patients is yet tobe proven considering their improved outcomesin the current era with the Norwood operation.
Author has nothing to disclose with regard tocommercial support.
R
References
[1] Freedom RM, Benson LN, Smallhorn JF, Williams WG,Trusler GA, Rowe RD. Subaortic stenosis, theuniventricular heart, and banding of the pulmonaryartery: an analysis of the courses of 43 patients withuniventricular heart palliated by pulmonary arterybanding. Circulation 1986;73(4):758–64.
[2] Lacour-Gayet F. Management of older single functioningventricles with outlet obstruction due to a restricted‘‘VSD’’ in double inlet left ventricle and in complexdouble outlet right ventricle. Semin Thorac CardiovascSurg Pediatr Card Surg Annu 2009;1:130–2.
[3] Matitiau A, Geva T, Colan SD, et al. Bulboventricularforamen size in infants with double-inlet left ventricle ortricuspid atresia with transposed great arteries: influenceon initial palliative operation and rate of growth. J Am CollCardiol 1992;19(1):142–8.
[4] Fraser Jr CD. Management of systemic outlet obstructionin patients undergoing single ventricle palliation. SeminThorac Cardiovasc Surg Pediatr Card Surg Annu2009;1:70–5.
[5] Cheung HC, Lincoln C, Anderson RH, et al. Options forsurgical repair in hearts with univentricularatrioventricular connection and subaortic stenosis. JThorac Cardiovasc Surg 1990;100(5):672–81.
[6] Shimada M, Hoashi T, Kagisaki K, Shiraishi I, Yagihara T,Ichikawa H. Clinical outcomes of prophylactic Damus-Kaye-Stansel anastomosis concomitant with bidirectionalGlenn procedure. J Thorac Cardiovasc Surg2012;143(1):137–43.
[7] Uemura H, Ho SY, Adachi I, Yagihara T. Morphologicspectrum of ventriculoarterial connection in hearts withdouble inlet left ventricle: implications for surgicalprocedures. Ann Thorac Surg 2008;86(4):1321–7.
[8] Daenen W, Eyskens B, Meyns B, Gewillig M. Neonatalpulmonary artery banding does not compromise theshort-term function of a Damus-Kaye-Stanselconnection. Eur J Cardiothorac Surg 2000;17(6):655–7.
[9] Odim JN, Laks H, Drinkwater Jr DC, et al. Staged surgicalapproach to neonates with aortic obstruction and singleventricle physiology. Ann Thorac Surg 1999;68(3):962–8.
[10] Clark AJ, Kasahara S, Andrews DR, et al. Mid-term resultsfor double inlet left ventricle and similar morphologies:timing of Damus-Kaye-Stansel. Ann Thorac Surg2004;78(2):650–7.
[11] Fujii Y, Kasahara S, Kotani Y, et al. Double-barrel Damus-Kaye-Stansel operation is better than end-to-side Damus-Kaye-Stansel operation for preserving the pulmonaryvalve function: the importance of preserving the shape ofthe pulmonary sinus. J Thorac Cardiovasc Surg2011;141(1):193–9.
[12] Fiore AC, Rodefeld M, Vijay P, et al. Subaortic obstructionin univentricular heart: results using the double barrelDamus-Kaye-Stansel operation. Eur J Cardiothorac Surg2009;35(1):141–6.
[13] Huddleston CB, Canter CE, Spray TL. Damus-Kaye-Stansel with cavopulmonary connection for singleventricle and subaortic obstruction. Ann Thorac Surg1993;55(2):339–46.
[14] Miura T, Kishimoto H, Kawata H, Hata M, Hoashi T,Nakajima T. Management of univentricular heart withsystemic ventricular outflow obstruction by pulmonaryartery banding and Damus-Kaye-Stansel operation. AnnThorac Surg 2004;77(1):23–8.
[15] Donofrio MT, Jacobs ML, Norwood WI, Rychik J. Earlychanges in ventricular septal defect size and ventricular
geometry in the single left ventricle aftervolume-unloading surgery. J Am Coll Cardiol 1995;26(4):1008–15.
[16] Carter TL, Mainwaring RD, Lamberti JJ.Damus-Kaye-Stansel procedure: midterm follow-upand technical considerations. Ann Thorac Surg 1994;58(6):1603–8.
[17] B. Alsoufi, A. Al-Wadai, M Khan, Outcomes of Damus-Kaye-Stansel anastomosis at time of cavopulmonaryconnection in single ventricle patients at risk ofdeveloping systemic ventricular outflow tract obstruction,Eur J Cardiothorac Surg. (in press).
[18] Pass RH, Solowiejczyk DE, Quaegebeur JM, et al.Bulboventricular foramen resection: hemodynamic andelectrophysiologic results. Ann Thorac Surg2001;71(4):1251–4.
[19] Hiramatsu T, Imai Y, Kurosawa H, et al. Midterm results ofsurgical treatment of systemic ventricular outflowobstruction in Fontan patients. Ann Thorac Surg2002;73(3):855–60.
[20] Cerillo AG, Murzi B, Giusti S, Crucean A, Redaelli S,Vanini V. Pulmonary artery banding and ventricularseptal defect enlargement in patients with univentricularatrioventricular connection and the aorta originating froman incomplete ventricle. Eur J Cardiothorac Surg2002;22(2):192–9.
[21] Tchervenkov CI, Shum-Tim D, Béland MJ, Jutras L, PlattR. Single ventricle with systemic obstruction in early life:comparison of initial pulmonary artery banding versus theNorwood operation. Eur J Cardiothorac Surg2001;19(5):671–7.
[22] Serraf A, Conte S, Lacour-Gayet F, et al. Systemicobstruction in univentricular hearts: surgical options forneonates. Ann Thorac Surg 1995;60(4):970–7.
[23] Franklin RC, Spiegelhalter DJ, Anderson RH, et al.Double-inlet ventricle presenting in infancy. II. Resultsof palliative operations. J Thorac Cardiovasc Surg1991;101(5):917–23.
[24] Franklin RC, Spiegelhalter DJ, Rossi Filho RI, et al.Double-inlet ventricle presenting in infancy. III.Outcome and potential for definitive repair. J ThoracCardiovasc Surg 1991;101(5):924–34.
[25] Mosca RS, Hennein HA, Kulik TJ, et al. Modified Norwoodoperation for single left ventricle and ventriculoarterialdiscordance: an improved surgical technique. Ann ThoracSurg 1997;64(4):1126–32.
[26] Bradley SM, Simsic JM, Atz AM, Dorman BH. The infantwith single ventricle and excessive pulmonary blood flow:results of a strategy of pulmonary artery division andshunt. Ann Thorac Surg 2002;74(3):805–10.
[27] Alsoufi B, Bennetts J, Verma S, Caldarone CA. Newdevelopments in the treatment of hypoplastic left heartsyndrome. Pediatrics 2007;119(1):109–17.
[28] Karl TR, Watterson KG, Sano S, Mee RB. Operations forsubaortic stenosis in univentricular hearts. Ann ThoracSurg 1991;52(3):420–7.
[29] Lacour-Gayet F, Serraf A, Fermont L, et al. Early palliationof univentricular hearts with subaortic stenosis andventriculoarterial discordance. The arterial switch option.J Thorac Cardiovasc Surg 1992;104(5):1238–45.
[30] Ceresnak SR, Quaegebeur JM, Pass RH, Hordof AJ,Liberman L. The palliative arterial switch procedure forsingle ventricles: are these patients suitable Fontancandidates? Ann Thorac Surg 2008;86(2):583–7.
[31] Heinle JS, Carberry KE, McKenzie ED, Liou A, KatigbakPA, Fraser Jr CD. Outcomes after the palliative arterialswitch operation in neonates with single-ventricleanatomy. Ann Thorac Surg 2013;95(1):212–9.
[32] Akintuerk H, Michel-Behnke I, Valeske K, et al. Stenting ofthe arterial duct and banding of the pulmonary arteries:basis for combined Norwood stage I and II repair inhypoplastic left heart. Circulation 2002;105(9):1099–103.
[33] Pizarro C, Murdison KA. Off pump palliation forhypoplastic left heart syndrome: surgical approach.
[34] Bacha EA, Hijazi ZM. Hybrid procedures in pediatriccardiac surgery. Semin Thorac Cardiovasc Surg PediatrCard Surg Annu 2005:78–85.
[35] Galantowicz M, Cheatham JP, Phillips A, et al. Hybridapproach for hypoplastic left heart syndrome:intermediate results after the learning curve. Ann ThoracSurg 2008;85(6):2063–71.
[36] Alsoufi B, Manlhiot C, Awan A, et al. Current outcomes ofthe Glenn bidirectional cavopulmonary connection for
single ventricle palliation. Eur J Cardiothorac Surg2012;42(1):42–8 [discussion 48–9].
[37] Broekhuis E, Brizard CP, Mee RB, Cochrane AD, Karl TR.Damus-Kaye-Stansel connections in children withpreviously transected pulmonary arteries. Ann ThoracSurg 1999;67(2):519–21.
[38] Meadows J, Pigula F, Lock J, Marshall A. Transcathetercreation and enlargement of ventricular septal defects forrelief of ventricular hypertension. J Thorac CardiovascSurg 2007;133(4):912–8.
