Subaortic Stenosis in Single Ventricle: Surgical Management

1. Overview

Subaortic stenosis (SAS), more broadly described as systemic ventricular outflow tract obstruction (SVOTO), is an important and potentially progressive problem in selected forms of functionally univentricular congenital heart disease with ventriculoarterial discordance. It is particularly relevant in double-inlet left ventricle (DILV) with transposition of the great arteries (TGA) and tricuspid atresia (TA) with TGA.

In these anatomies, the dominant systemic ventricle is usually a morphologic left ventricle, whereas the aorta arises from a rudimentary or hypoplastic morphologic right ventricular outlet chamber. Systemic blood must therefore pass from the dominant ventricle through the bulboventricular foramen (BVF), or VSD, into the rudimentary RV before reaching the aorta:

Dominant LV → BVF/VSD → rudimentary RV → aorta

The BVF is therefore not simply an interventricular communication; it constitutes a critical component of the systemic ventricular outflow pathway.

The principal surgical challenge is that this pathway may be adequate in early infancy but progressively narrow during staged single-ventricle palliation. In a contemporary series of 30 patients with DILV/TGA or TA/TGA, 20 ultimately required a systemic outflow relief operation. Five-year freedom from systemic outflow relief was only 34.5%. Associated aortic arch obstruction was a particularly strong predictor of earlier intervention, with a hazard ratio of 20.6, while a smaller indexed end-systolic systemic outflow area independently predicted subsequent obstruction [1].

Accordingly, management must address not only the severity of obstruction at presentation but also the likelihood that the systemic outflow pathway will remain adequate as ventricular loading conditions change.

2. Mechanisms of Systemic Outflow Obstruction

SVOTO in DILV/TGA and TA/TGA most commonly develops at or around the BVF. Several mechanisms may coexist:

• A congenitally small BVF.
• Progressive muscular narrowing around the BVF.
• Hypertrophy of the ventricular septum or rudimentary outlet chamber.
• Adverse remodeling of ventricular geometry.
• Increased systemic ventricular pressure after pulmonary artery banding (PAB).
• Relative failure of BVF growth as cardiac output requirements increase.
• Associated aortic arch obstruction, which adds another level of systemic impedance.

The obstruction is therefore frequently dynamic and progressive, rather than a fixed congenital lesion.

This differs conceptually from the discrete subaortic membrane encountered in a biventricular circulation. In a single ventricle with ventriculoarterial discordance, the relevant obstructive pathway may include the BVF, muscular outlet chamber, subaortic region, and proximal aorta. The term SVOTO therefore often describes the physiology more accurately than isolated “subaortic stenosis.”

3. Why Pulmonary Artery Banding Can Promote SAS

PAB remains useful when pulmonary blood flow is excessive and the systemic outflow pathway is initially adequate. However, its effect on ventricular loading is particularly important in DILV/TGA and related anatomies.

Before banding, the dominant ventricle may eject a substantial proportion of its output relatively freely toward the pulmonary artery. After PAB, resistance to pulmonary ejection rises, increasing ventricular pressure and redirecting a greater proportion of output through the BVF-dependent systemic pathway. Progressive ventricular hypertrophy may then further reduce the effective size of the BVF.

Clinical data support this phenomenon. In a series of patients with univentricular hearts at risk for systemic outflow obstruction, the indexed ventricular outflow diameter decreased from a median of 103% of normal before PAB to 75% after banding [2]. The investigators suggested that a DKS should be considered before Fontan when the post-PAB indexed outflow diameter was below approximately 70%, and at the time of Fontan when it was below approximately 120% [2]. These values are useful historical indicators of risk rather than absolute contemporary thresholds.

Long-term experience also demonstrates that the absence of obstruction at the time of PAB does not guarantee durable systemic outflow. Among 18 infants with DILV/TGA and aortic arch obstruction treated with an arch-repair/PAB strategy, all but one ultimately developed SAS; 15 required subsequent relief of the obstruction [3].

Thus, PAB protects the pulmonary vascular bed but may expose or accelerate vulnerability of the systemic outflow pathway.

4. Surveillance and Timing of Intervention

Because SVOTO may evolve rapidly, surveillance should focus on the trajectory of the obstruction rather than a single Doppler gradient.

Assessment should include:

• BVF/VSD dimensions and indexed dimensions when possible.
• Cross-sectional area of the systemic outflow pathway.
• Doppler velocity and gradient.
• Ventricular hypertrophy.
• Systemic ventricular systolic and diastolic function.
• Atrioventricular valve regurgitation.
• Aortic and arch obstruction.
• Clinical evidence of impaired systemic perfusion.
• Timing relative to the planned Glenn or Fontan stage.

Aortic arch obstruction deserves particular attention. In the 30-patient DILV/TGA and TA/TGA series, arch obstruction and smaller end-systolic systemic outflow area were the strongest predictors of subsequent systemic outflow relief [1].

The practical implication is that patients with both a potentially restrictive BVF and arch obstruction should be considered high risk for later SVOTO, even when neonatal Doppler findings appear acceptable.

5. Surgical Strategy When Significant SAS Is Present Early

When significant systemic outflow obstruction is already present in the neonatal or early infant period, simply performing PAB does not address the principal problem.

The preferred strategy is generally to establish a systemic pathway that does not depend on passage through the restrictive BVF.

Two major approaches accomplish this:

1. Palliative arterial switch operation (pASO), usually with PAB and arch reconstruction when required.
2. Norwood-type reconstruction incorporating a Damus–Kaye–Stansel connection, with a separate source of pulmonary blood flow.

Both strategies convert the pulmonary root into part of an unobstructed systemic outflow pathway but differ substantially in great-artery reconstruction and pulmonary blood-flow physiology.

6. Palliative Arterial Switch Operation

Physiologic principle

The palliative ASO changes the ventriculoarterial pathway so that systemic output from the dominant LV no longer needs to traverse the BVF and rudimentary RV.

Before pASO:

LV → BVF → rudimentary RV → aorta

After arterial switch:

LV → native pulmonary root → neo-aorta → systemic circulation

The operation therefore eliminates the restrictive BVF/RV/RVOT pathway from the principal systemic circuit.

When aortic arch obstruction is present, arch reconstruction can be performed concurrently. Pulmonary blood flow can be regulated by PAB, producing a controlled band physiology until subsequent cavopulmonary connection.

Outcomes

A single-center series of 14 neonates undergoing pASO reported no deaths. All patients subsequently underwent bidirectional cavopulmonary shunt, and 11 had completed Fontan palliation at the time of reporting. At median 5-year echocardiographic follow-up, ventricular function remained normal, neoaortic insufficiency was trivial to mild, and no LV outflow obstruction was identified [4].

The major tradeoff was the pulmonary arterial reconstruction. Pulmonary artery interventions were required during staged palliation, and 7 of the 11 patients undergoing Fontan required pulmonary artery augmentation [4].

Thus, pASO provides a particularly attractive physiologic solution to BVF-dependent obstruction, but branch pulmonary artery geometry and later pulmonary artery reintervention are important considerations.

7. Norwood-Type Reconstruction and DKS

The alternative approach is a Norwood-type systemic outflow reconstruction, in which the pulmonary root is connected to the native systemic arterial pathway through a DKS-type reconstruction.

Systemic flow then follows:

Dominant LV → pulmonary root → DKS/neo-aorta → systemic circulation

The obstructed BVF–rudimentary RV pathway is effectively bypassed.

Because the pulmonary root is incorporated into systemic outflow, pulmonary blood flow must be supplied separately, traditionally by a modified Blalock–Thomas–Taussig shunt or another surgically constructed pulmonary blood-flow source. The resulting circulation therefore has shunt physiology, rather than the band physiology commonly used after pASO.

A Norwood-type approach is particularly attractive when severe systemic outflow obstruction coexists with significant arch hypoplasia or interruption and when a durable BVF-independent systemic pathway is desired from the initial palliation.

However, neonatal Norwood reconstruction introduces its own physiologic and technical burden, including shunt-dependent pulmonary circulation and extensive great-artery reconstruction.

Comparative evidence remains limited. Published experience comparing initial PAB-based palliation with Norwood-type reconstruction is retrospective, spans different surgical eras, and includes heterogeneous anatomy [6]. Consequently, neither strategy can be selected solely from historical outcome comparisons; anatomy, institutional experience, arch morphology, BVF dimensions, semilunar valve anatomy, and anticipated subsequent reconstruction remain central.

8. Arch Repair + PAB When SAS Is Not Initially Significant

When there is significant arch obstruction but the BVF and systemic outflow pathway remain adequate, a less extensive initial operation may be appropriate:

Aortic arch repair + PAB

This strategy restores arch continuity while controlling pulmonary overcirculation without committing the neonate to arterial switch or Norwood-type reconstruction.

The important limitation is that it does not eliminate the potentially restrictive BVF from the systemic pathway.

Results from staged approaches demonstrate that this can nevertheless be successful when surveillance is rigorous and obstruction is treated promptly. In a 25-patient series treated initially with PAB and repair of associated arch obstruction, 20 of 22 survivors ultimately achieved Fontan completion [5]. DKS performed during later staging provided durable systemic outflow, with no reported DKS obstruction and no more than mild pulmonary regurgitation [5].

Similarly, the earlier 18-patient DILV/TGA series demonstrated 72% long-term survival, with 12 patients reaching Fontan or bidirectional cavopulmonary palliation after recognition and relief of progressive SAS [3].

The strategy should therefore be understood as staged management of a potentially evolving lesion, rather than reassurance that the systemic outflow is permanently protected.

9. What to Do if SAS Progresses After Arch Repair + PAB

At the time of follow-up or planned bidirectional Glenn, the systemic outflow pathway should be reassessed.

No significant SAS

If the BVF remains adequately sized and there is no clinically important systemic outflow gradient, the patient may proceed to bidirectional Glenn without direct systemic outflow intervention.

Continued surveillance remains necessary because narrowing may occur later, particularly as ventricular geometry and loading conditions change.

Progressive or significant SAS

When obstruction develops, the principal options are:

• Subaortic muscular resection
• BVF/VSD enlargement
• DKS anastomosis

These procedures can frequently be combined with BDG when the patient has reached the appropriate stage of cavopulmonary palliation.

10. Direct Muscle Resection and BVF Enlargement

Direct enlargement treats the native obstructed pathway rather than bypassing it. Muscular tissue around the BVF can be resected, and the BVF/VSD can be enlarged when necessary.

This approach may be attractive when the obstruction is localized and an adequate systemic pathway can be created without excessive ventricular septal intervention.

However, important technical concerns include:

• Injury to the atrioventricular conduction axis and complete heart block.
• Residual obstruction.
• Recurrent narrowing as the child grows.
• Excessive ventricular septal muscle resection.
• Potential injury to adjacent atrioventricular valve structures.

The durability of direct BVF enlargement is particularly important. In the staged series reported by Clarke and colleagues, some patients who underwent BVF resection subsequently developed recurrent obstruction and ultimately required DKS [5].

Thus, direct resection should not be viewed simply as a less invasive alternative to DKS. The key question is whether it can produce a durable systemic outflow pathway without unacceptable conduction or ventricular risk.

11. DKS at the Time of Bidirectional Glenn

When significant SAS has developed after initial arch repair and PAB, DKS combined with BDG provides an effective staged solution.

The two components address complementary physiologic objectives:

• BDG unloads the single ventricle by directing superior caval return directly to the pulmonary arteries.
• DKS establishes a BVF-independent systemic outflow pathway.

This strategy avoids repeated attempts to enlarge an anatomically vulnerable BVF and can provide durable relief of systemic obstruction during the later Fontan pathway.

The decision between direct BVF enlargement and DKS depends on the mechanism and severity of obstruction, ventricular geometry, semilunar valve function, great-artery relationship, prior pulmonary artery band placement, conduction-system risk, and anticipated Fontan anatomy.

12. Practical Surgical Decision Framework

A useful approach is to begin with two questions:

Is systemic outflow obstruction significant?

Is aortic arch obstruction present?

Significant SAS/SVOTO present

A BVF-independent systemic pathway should generally be considered.

Options include:

Palliative ASO ± arch repair + PAB

• Directs dominant LV output to the neo-aorta.
• Avoids BVF-dependent systemic flow.
• Maintains controlled pulmonary flow with band physiology.
• Allows subsequent BDG and Fontan.

or

Norwood-type reconstruction with DKS + pulmonary blood-flow shunt

• Bypasses the obstructed BVF.
• Provides reliable systemic output through the pulmonary root/neo-aorta.
• Uses shunt-dependent pulmonary blood flow.
• Progresses subsequently to BDG and Fontan.

No significant SAS, but arch obstruction requires repair

Consider:

Arch repair + PAB

Then reassess the systemic outflow pathway carefully during follow-up.

At the next stage:

• No significant SAS → BDG
• Localized, safely resectable SAS → muscle resection ± BVF enlargement + BDG
• Significant, recurrent, or anatomically unfavorable obstruction → DKS + BDG

This is a framework rather than a fixed algorithm. The available evidence is predominantly retrospective, derived from relatively small cohorts, and strongly influenced by anatomy, surgical era, and institutional strategy [1–6].

13. Key Surgical Principle

The fundamental lesson is that systemic outflow obstruction in DILV/TGA and TA/TGA should be anticipated rather than treated only after a large gradient develops.

The combination of a small or narrowing BVF and associated arch obstruction identifies a particularly high-risk substrate [1]. PAB can successfully control pulmonary blood flow, but it may accelerate narrowing of the systemic outflow pathway as ventricular pressure and hypertrophy increase [2,3].

When significant obstruction is already present, pASO and Norwood/DKS strategies solve the problem by bypassing the vulnerable BVF pathway. When the systemic outflow is initially adequate, arch repair with PAB can preserve a simpler neonatal strategy, provided that surveillance is rigorous and later DKS or BVF intervention is performed before obstruction imposes sustained pressure load on the systemic ventricle [3,5].

The optimal operation is therefore not determined by a single gradient. It is the strategy that provides durable systemic outflow, appropriately controlled pulmonary blood flow, preserved ventricular and semilunar valve function, and a favorable pathway toward Glenn and Fontan completion.

References

[1] Park WK, Baek JS, Kwon BS, Im YM, Lee JH, Choi ES, Park CS, Yun TJ. Revisitation of double-inlet left ventricle or tricuspid atresia with transposed great arteries. Ann Thorac Surg. 2019;107(4):1212-1217.

[2] Miura T, Kishimoto H, Kawata H, Hata M, Hoashi T, Nakajima T. Management of univentricular heart with systemic ventricular outflow obstruction by pulmonary artery banding and Damus-Kaye-Stansel operation. Ann Thorac Surg. 2004;77(1):23-28.

[3] Webber SA, LeBlanc JG, Keeton BR, Salmon AP, Sandor GG, Lamb RK, Monro JL. Pulmonary artery banding is not contraindicated in double inlet left ventricle with transposition and aortic arch obstruction. Eur J Cardiothorac Surg. 1995;9(9):515-520.

[4] Heinle JS, Carberry KE, McKenzie ED, Liou A, Katigbak PA, Fraser CD Jr. Outcomes after the palliative arterial switch operation in neonates with single-ventricle anatomy. Ann Thorac Surg. 2013;95(1):212-219.

[5] Clarke AJ, Kasahara S, Andrews DR, Cooper SG, Nicholson IA, Chard RB, Nunn GR, Winlaw DS. Mid-term results for double inlet left ventricle and similar morphologies: timing of Damus-Kaye-Stansel. Ann Thorac Surg. 2004;78(2):650-657.

[6] Ruzmetov M, Geiss DM, Fortuna RS. Outcomes of double inlet left ventricle and similar morphologies: a single center comparison of initial pulmonary artery banding versus a Norwood-type reconstruction. J Card Surg. 2013;28(5):569-575.