Ebstein Anomaly — #5 Starnes Procedure

Rationale and pathophysiology

In the most severe neonatal Ebstein phenotype, failed delamination of the septal and inferior tricuspid leaflets produces a massively atrialized RV and a diminutive functional RV. Torrential TR elevates RA pressure and drives right-to-left shunting at the atrial level. When an incompetent pulmonary valve and a patent ductus are present, a circular shunt can form—ductal inflow to the PA washes retrograde across the pulmonary valve and returns via the pulmonary veins to the LA, diverting output away from the systemic circulation while the RV provides little or no antegrade PA flow. Clinically, neonates exhibit profound cyanosis, acidosis, cardiogenic shock, and low RV systolic pressure (~20–25 mmHg), signaling RV incapacity for a biventricular plan. In this context, fenestrated RV exclusion with controlled pulmonary blood flow is a purpose-built rescue that breaks the shunt and stabilizes physiology [1].

Indications

The Starnes procedure is selected to (1) abolish the circular shunt, (2) stabilize systemic perfusion, and (3) establish a single-ventricle pathway—while preserving options for later RV rehabilitation in select patients [4]. Typical triggers include:

• Carpentier C/D morphology with a tiny functional RV, severe TR, and markedly low RVSP versus systemic pressure.
• Functional pulmonary atresia or negligible RV–PA forward flow, often with pulmonary insufficiency.
• Refractory hypoxemia/systemic steal despite optimized ventilation (to reduce PVR), careful ductal strategy, and inotropes.

Institutional algorithms commonly assess the feasibility of primary tricuspid repair first; when not achievable, RV exclusion with atrial septectomy and systemic-to-PA shunt is undertaken [1].

Operative strategy (canonical steps)

1. Patch exclusion of the tricuspid valve with fenestration to decompress Thebesian/coronary venous return and avoid RV compartment syndrome. Fenestration diminishes RV dilation and facilitates septal realignment and LV mechanics [1,2].
2. Definitive interruption of the RV–PA circuit via proximal MPA ligation/division to eliminate pulmonary valve insufficiency and break the circular shunt.
3. Generous atrial septectomy for unobstructed atrial mixing and venous decompression in single-ventricle physiology.
4. Systemic-to-pulmonary shunt (e.g., mBTS) sized to balance oxygen delivery against over-circulation risk.
5. Right-atrial reduction (frequent) to improve geometry, lung mechanics, and reduce arrhythmic substrate.

In unstable presentations or extreme cardiomegaly, a rapid two-stage approach (early extra-CPB steps—MPA ligation, RA/RV plication, mBTS—followed by formal RV exclusion days later) is reported as effective rescue [6,7].

Immediate postoperative expectations

• Oxygenation depends on shunt flow and PVR; ventilatory management targets mild hyperventilation and adequate oxygenation to temper PVR without excessive systemic desaturation.
• Inotropes are tailored to LV (systemic ventricle) performance and SVR.
• Vigilance for arrhythmias and pleural effusions is essential after extensive RA manipulation.
• Echocardiography should confirm an intact TV patch with patent fenestration, complete RV–PA interruption, shunt patency, and a widely patent atrial septectomy [1,2].

Outcomes and longitudinal pathway

Modern series demonstrate excellent late survival among neonatal survivors of the modified (fenestrated) Starnes: in one cohort, survival at 1, 5, and 10 years was 87%, 87%, and 81%, respectively; most survivors proceeded successfully to Glenn and Fontan with preserved LV function and favorable rhythm profiles [3]. Importantly, contemporary perspectives emphasize that Starnes “first” does not necessarily mean Fontan “forever.” With growth and falling PVR, selected patients can undergo cone repair to rehabilitate the RV and convert to 1.5- or 2-ventricle physiology, with mid-term results showing trivial/mild TR and satisfactory RV function in carefully chosen cases [4,5].

Technical pearls and pitfalls

• Fenestration size matters: too small risks RV hypertension; too large re-loads the right heart. Aim for a controlled, low-resistance pop-off [1,2].
• MPA division must be definitive—residual PI defeats the purpose.
• Shunt sizing is hemodynamics: undersizing → hypoxemia; oversizing → overcirculation, systemic hypotension, shunt-dependent failure [1].
• RA reduction: improves mechanics but avoid conduction injury near the triangle of Koch/coronary sinus.
• Consider a two-stage strategy when rapid bedside stabilization is needed or when prohibitive cardiomegaly/edema makes immediate intracardiac work unsafe [6,7].

Decision boundaries and alternatives

• If RV and RVSP are adequate and durable antegrade RV–PA flow is achievable as PVR falls, a primary or staged cone repair with biventricular goals may be preferred [1].
• Pharmacologic ductal manipulation is insufficient in the face of PI-driven circular shunt; surgical elimination of PI plus a controlled shunt is more reliable [1].
• In an era of evolving strategies, teams should reassess RV potential longitudinally; staged conversion (e.g., Glenn → cone ± 1.5V) may improve lifetime physiology in select children [4,5].

References

[1] Reemtsen BL, Polimenakos AC, Fagan BT, Wells WJ, Starnes VA. Current surgical therapy for Ebstein anomaly in neonates. J Thorac Cardiovasc Surg. 2006;132(5):1285-1290.

[2] Reemtsen BL, Fagan BT, Wells WJ, Starnes VA. Fate of the right ventricle after fenestrated right ventricular exclusion (Starnes procedure) for severe neonatal Ebstein anomaly. J Thorac Cardiovasc Surg. 2007;134(6):1406-1411.

[3] Kumar SR, Walters HL 3rd, Turek JW, et al. Single-Ventricle Outcomes After Neonatal Palliation of Ebstein Anomaly or Tricuspid Valve Dysplasia. Circulation. 2016;134(17):1245-1253.

[4] Stephens EH, Dearani JA. Commentary: Neonatal Ebstein—Starnes procedure first… Fontan not necessarily the last. J Thorac Cardiovasc Surg. 2023;166(6):1744-1745.

[5] da Silva JP, Seese L, Olivieri L, et al. The Da Silva cone operation after the Starnes procedure for neonatal Ebstein anomaly: Conversion to biventricular physiology. JTCVS Tech. 2020;3:176-179.

[6] Kajihara N, Asou T, Takeda Y, et al. Rapid two-stage Starnes procedure for a symptomatic neonate with Ebstein anomaly. Ann Thorac Surg. 2010;90(6):2073-2075.

[7] Asou T, Osaka K, Kado H, et al. Rapid two-stage Starnes approach in high-risk neonates with Ebstein anomaly. Ann Thorac Cardiovasc Surg. 2020;26(3):170-177.