Glenn Circulation: Physiologic Principles, Cavopulmonary Anastomosis, and Determinants of Post-Glenn SpO₂

1. Overview

The bidirectional Glenn procedure is the standard second-stage palliation for most patients with functionally univentricular circulation. It connects the superior vena cava (SVC) directly to the pulmonary arteries, directing upper-body systemic venous return to the lungs without passage through the systemic ventricle. This creates two simultaneous benefits: pulmonary blood flow is partially separated from ventricular ejection, and the systemic ventricle is relieved of a substantial component of its volume load.

The Glenn circulation is not merely an anatomic shunt. It is a pressure-dependent, passive pulmonary circulation in which flow is governed by the amount of SVC return and the resistance between the SVC and the pulmonary venous atrium. Conversion from a systemic-to-pulmonary arterial shunt circulation to a Glenn circulation has also been associated with improved cerebral oxygenation, despite an acute increase in central venous pressure [1]. Postoperative assessment must therefore integrate systemic oxygen saturation, Glenn pressure, pulmonary vascular resistance, ventricular filling pressure, cardiac output, hemoglobin concentration, respiratory mechanics, and systemic oxygen consumption.

2. Pre-Glenn Single-Ventricle Physiology

Before the Glenn procedure, the single functional ventricle usually supplies both the systemic and pulmonary circulations. Pulmonary blood flow may arise through a systemic-to-pulmonary arterial shunt, a right ventricle-to-pulmonary artery conduit, a patent native pulmonary outflow tract, or another surgically controlled source.

The systemic ventricular output can be expressed as:

COₛᵥ = Qp + Qs

where COₛᵥ is systemic ventricular output, Qp is pulmonary blood flow, and Qs is systemic blood flow. Every unit of pulmonary blood flow must be ejected by the systemic ventricle before returning through the pulmonary veins. Pulmonary flow is therefore “stolen” from effective systemic output, although it remains essential for oxygenation.

Excessive Qp produces ventricular volume overload, atrioventricular valve regurgitation, elevated atrial pressure, increased myocardial oxygen demand, and reduced systemic perfusion. Inadequate Qp causes severe cyanosis. The pre-Glenn circulation is consequently vulnerable to changes in pulmonary vascular resistance, systemic vascular resistance, shunt resistance, and ventricular function.

3. Physiologic Transformation Produced by the Glenn Procedure

The bidirectional Glenn diverts SVC blood directly into the pulmonary arteries. The pulmonary circulation is placed in series between the SVC and the heart:

SVC → pulmonary arteries → pulmonary capillaries → pulmonary veins → atrium → systemic ventricle

The systemic ventricle no longer directly pumps the SVC-derived component of pulmonary blood flow. Its output therefore approaches systemic flow:

COₛᵥ ≈ Qs

This reduces ventricular preload and volume work. The reduction is particularly valuable in patients with ventricular dilation, systolic dysfunction, atrioventricular valve regurgitation, or excessive pulmonary blood flow after first-stage palliation.

The physiologic cost is that pulmonary blood flow becomes passive and usually non-pulsatile unless antegrade pulmonary blood flow is intentionally retained. In the absence of additional pulmonary flow or major collateral flow:

Qp ≈ Qₛᵥ꜀

The driving pressure for Glenn flow is approximately:

SVC pressure − pulmonary venous atrial pressure

Resistance is distributed across the cavopulmonary anastomosis, branch pulmonary arteries, pulmonary arterioles and capillaries, pulmonary veins, atrium, and ventricular diastolic chamber. Obstruction or elevated pressure anywhere along this pathway can reduce pulmonary flow and increase SVC pressure.

4. Cavopulmonary Anastomosis

The bidirectional Glenn is generally constructed as an end-to-side anastomosis between the SVC and right pulmonary artery. When cardiopulmonary bypass is used, the SVC cannula is positioned sufficiently cephalad to permit division near the cavoatrial junction. Inferior venous drainage is obtained through the right atrium or IVC. The operative plan must account for bilateral SVCs, a persistent left SVC, innominate vein anatomy, previous shunts, branch pulmonary artery distortion, and systemic venous collaterals.

The SVC is divided near the right atrial junction while protecting the sinoatrial node and sinus node artery. Excessive dissection, traction, or extension into the cavoatrial junction may increase the risk of sinus node dysfunction. The cardiac end is securely oversewn, and the cephalad SVC is mobilized only enough to permit a tension-free connection without torsion.

A longitudinal arteriotomy is created on the superior aspect of the right pulmonary artery. The opening should match the full SVC caliber. A restrictive arteriotomy or narrowed posterior suture line may produce elevated Glenn pressure, reduced pulmonary blood flow, facial and upper-extremity venous congestion, impaired cerebral venous drainage, and desaturation. Conversely, poor orientation may distort the right pulmonary artery or produce preferential flow to one lung.

The completed anastomosis should be tension-free, non-restrictive, non-twisted, and oriented for balanced bilateral pulmonary arterial distribution. After de-airing and reperfusion, the surgeon should confirm unobstructed flow, acceptable SVC–pulmonary artery pressure, stable systemic hemodynamics, and adequate oxygenation.

5. The Primary Determinant of Post-Glenn Oxygen Saturation

Systemic arterial saturation after the Glenn procedure is primarily determined by the pulmonary-to-systemic flow ratio and by the oxygen saturations of the blood streams that mix within the heart. Clinical measurements after cavopulmonary anastomosis have shown a mean Qp/Qs of approximately 0.58, with collateral flow capable of increasing both Qp/Qs and systemic saturation [2].

Without additional pulmonary blood flow:

Qp/Qs ≈ Qₛᵥ꜀ / (Qₛᵥ꜀ + Qᵢᵥ꜀)

Thus, the relative distribution of SVC and IVC flow is fundamental. A larger proportion of upper-body/SVC return produces a greater fraction of systemic blood that traverses the lungs before systemic ejection. As the child grows and IVC flow becomes proportionally larger, systemic saturation may decrease even when the Glenn pathway remains anatomically unobstructed. Theoretical modeling similarly demonstrates that a higher SVC-to-IVC flow ratio increases arterial saturation and total systemic oxygen delivery [3].

A simplified mixing relationship is:

SaO₂ ≈ [(Qp × SpvO₂) + (Qᵢᵥ꜀ × Sᵢᵥ꜀O₂)] / Qs

where SpvO₂ is pulmonary venous saturation and Sᵢᵥ꜀O₂ represents the oxygen saturation of venous blood returning directly to the heart. This equation identifies two clinically important “levers”: pulmonary venous oxygenation and systemic venous oxygenation.

6. Pulmonary Venous Saturation and Pulmonary Vascular Resistance

Pulmonary venous saturation is usually near 100% when ventilation, diffusion, and pulmonary venous anatomy are normal. Therefore, increasing an already normal SpvO₂ produces only a modest improvement in systemic saturation. Nevertheless, pulmonary venous desaturation should be suspected with atelectasis, pulmonary edema, pneumonia, airway obstruction, pulmonary hemorrhage, ventilation-perfusion mismatch, or pulmonary venous obstruction.

Management includes lung recruitment, secretion clearance, correction of hypoxemia, hypercarbia, and acidosis, and treatment of the underlying pulmonary process. Inhaled nitric oxide may be useful when elevated or reactive pulmonary vascular resistance is suspected, but it cannot correct an anastomotic stenosis, branch pulmonary artery obstruction, pulmonary venous obstruction, or markedly elevated atrial pressure.

Because no subpulmonary ventricle drives Glenn flow, even modest increases in pulmonary vascular resistance may significantly reduce Qp. In a clinical series, pulmonary vascular resistance was inversely related to arterial saturation, and larger patients with body surface area greater than 0.65 m² were at increased risk of excessive postoperative cyanosis [4]. Positive-pressure ventilation should therefore provide adequate recruitment while avoiding excessive mean airway pressure and alveolar overdistension.

7. Systemic Venous Saturation: The More Modifiable Lever

Systemic venous saturation reflects the balance between oxygen delivery and oxygen consumption:

VO₂ = CO × (CaO₂ − CvO₂)

For a given arterial oxygen content, venous saturation falls when cardiac output decreases or oxygen consumption rises. Severely desaturated IVC blood can substantially lower systemic arterial saturation after mixing, even when pulmonary venous oxygenation is normal.

Cardiac output may be limited by ventricular systolic or diastolic dysfunction, atrioventricular valve regurgitation, inadequate preload, arrhythmia, excessive afterload, myocardial ischemia, or residual anatomic lesions. Improving ventricular performance and effective systemic flow may raise SpO₂ without a major change in lung function.

Fever, agitation, pain, shivering, respiratory distress, and increased work of breathing raise oxygen consumption and reduce venous saturation. Analgesia, temperature control, appropriate sedation, and respiratory support can therefore improve systemic oxygenation.

Hemoglobin concentration is another determinant of oxygen delivery:

CaO₂ ≈ 1.34 × hemoglobin × SaO₂

Clinically significant anemia increases tissue oxygen extraction and may lower systemic venous saturation. However, excessive hematocrit increases blood viscosity and may impair passive cavopulmonary flow. Transfusion decisions should balance oxygen-carrying capacity against viscosity, SVC pressure, and overall perfusion rather than targeting saturation alone.

8. Predictors of Early Desaturation and Adverse Outcomes

Age influences Glenn oxygenation through several mechanisms and should not be interpreted as a simple linear variable. In patients undergoing Glenn without additional pulmonary blood flow, age younger than 8 months and severe ventricular volume overload independently predicted lower early arterial saturation. Lower early saturation was also associated with mortality or exclusion from further univentricular palliation within 24 months, and higher postoperative SVC pressure correlated inversely with saturation [5].

In contrast, another cohort found that older or larger patients were more likely to develop excessive cyanosis, particularly when body surface area exceeded 0.65 m² [4]. This apparent difference is physiologically plausible. Very young infants may have unfavorable postoperative hemodynamics, pulmonary vascular reactivity, or persistent ventricular volume loading, whereas older and larger patients may have a lower SVC-to-IVC flow ratio and a greater IVC contribution to systemic venous return. Age must therefore be interpreted together with body size, Qₛᵥ꜀/Qᵢᵥ꜀, pulmonary vascular resistance, ventricular loading, and filling pressure.

9. Collaterals, Additional Pulmonary Blood Flow, and Late Glenn Physiology

Systemic venous collaterals may decompress the high-pressure SVC pathway into the IVC, atrium, or other lower-pressure venous channels. They reduce effective Glenn flow and deliver desaturated blood directly to the heart. Venous collaterals developed in approximately one-third of patients in one series and were associated with higher early transpulmonary gradients and lower follow-up arterial saturation. Coil embolization produced a substantial saturation increase in selected patients [6].

Aortopulmonary collaterals may increase pulmonary blood flow and saturation, but they also increase pulmonary venous return and systemic ventricular volume load. They were common after Glenn and Fontan procedures, particularly in patients with a previous Blalock–Taussig shunt [7]. Their significance should be judged by the balance between improved oxygenation and the adverse effects of ventricular volume loading, elevated filling pressure, and inefficient recirculation.

The decision to preserve antegrade pulmonary blood flow is similarly individualized. Additional flow may improve saturation, promote pulsatility, and support pulmonary artery development, but it may also elevate SVC pressure, increase ventricular volume load, and obscure the assessment of true Glenn performance. A 160-patient series evaluating additional pulmonary blood flow reported an early failure rate of 7.5%, indicating that accessory flow is not universally protective [8]. Conversely, a multicenter series of 246 patients managed with Glenn plus antegrade pulmonary blood flow reported no in-hospital deaths; 29.7% subsequently underwent Fontan completion, while the remainder maintained satisfactory palliation with resting saturation near 90% during intermediate follow-up [9].

Long-term surveillance must also consider pulmonary artery development. After Glenn, central pulmonary artery dimensions may decrease, particularly in the branch contralateral to the anastomosis, and prolonged Glenn duration has been associated with reduction in the Nakata index [10]. Balanced branch pulmonary artery flow, timely relief of stenosis, and appropriate timing of Fontan completion are therefore important.

10. Structured Evaluation of Post-Glenn Desaturation

Post-Glenn desaturation should be evaluated in a defined sequence:

1. Confirm pulmonary venous oxygenation: assess lung expansion, airway patency, ventilation-perfusion matching, pulmonary edema, infection, and pulmonary venous obstruction.
2. Assess Glenn pathway resistance: review SVC pressure, transpulmonary gradient, anastomotic caliber, branch pulmonary arteries, pulmonary vascular resistance, and mean airway pressure.
3. Assess systemic oxygen delivery: evaluate ventricular function, atrioventricular valve regurgitation, rhythm, preload, hemoglobin, temperature, pain, agitation, and metabolic demand.
4. Identify abnormal vascular pathways: evaluate venovenous collaterals, aortopulmonary collaterals, residual antegrade pulmonary blood flow, and systemic venous anomalies.
5. Interpret saturation in context: consider age, body size, SVC-to-IVC flow distribution, cerebral and somatic near-infrared spectroscopy, lactate, urine output, and end-organ perfusion.

The objective is not to normalize pulse oximetry at any cost. The clinical goal is adequate tissue oxygen delivery with unobstructed passive pulmonary flow, acceptable SVC pressure, stable ventricular performance, and a sustainable balance between pulmonary and systemic circulation.

11. Conclusion

The Glenn procedure unloads the systemic ventricle by placing the pulmonary circulation between the SVC and the heart. Post-Glenn oxygen saturation is primarily determined by Qp/Qs, which largely reflects SVC flow relative to total systemic venous return, and by the oxygen content of pulmonary venous and systemic venous blood. Pulmonary venous saturation is usually near maximal; consequently, systemic venous saturation is often the more modifiable determinant through optimization of cardiac output, oxygen consumption, and hemoglobin.

Persistent desaturation should prompt systematic evaluation of pulmonary vascular resistance, ventricular loading and filling pressure, anastomotic or branch pulmonary artery obstruction, collaterals, and the relative contribution of SVC and IVC flow. Successful Glenn management requires integration of anatomy, operative geometry, respiratory physiology, and systemic oxygen-delivery physiology rather than reliance on a single saturation or pressure value.

References

[1] Bertolizio G, DiNardo J, Laussen P, Polito A, Pigula F, Zurakowski D, Kussman B. Evaluation of cerebral oxygenation and perfusion with conversion from an arterial-to-systemic shunt circulation to the bidirectional Glenn circulation in patients with univentricular cardiac abnormalities. J Cardiothorac Vasc Anesth. 2015. doi:10.1053/j.jvca.2014.06.001.

[2] Salim M, Case CL, Sade RM, Watson DC, Alpert BS, DiSessa TG. Pulmonary/systemic flow ratio in children after cavopulmonary anastomosis. J Am Coll Cardiol. 1995. doi:10.1016/0735-1097(94)00441-R.

[3] Santamore W, Barnea O, Riordan C, Ross MP, Austin E. Theoretical optimization of pulmonary-to-systemic flow ratio after a bidirectional cavopulmonary anastomosis. Am J Physiol Heart Circ Physiol. 1998. doi:10.1152/ajpheart.1998.274.2.H694.

[4] Gross GJ, Jonas R, Castaneda AR, Hanley F, Mayer J, Bridges ND. Maturational and hemodynamic factors predictive of increased cyanosis after bidirectional cavopulmonary anastomosis. Am J Cardiol. 1994. doi:10.1016/0002-9149(94)90314-X.

[5] Aeba R, Katogi T, Kashima I, Omoto T, Kawada S, Omae K. Factors influencing arterial oxygenation early after bidirectional cavopulmonary shunt without additional sources of pulmonary blood flow. J Thorac Cardiovasc Surg. 2000. doi:10.1067/MTC.2000.108727.

[6] McElhinney D, Reddy V, Hanley F, Moore P. Systemic venous collateral channels causing desaturation after bidirectional cavopulmonary anastomosis: evaluation and management. J Am Coll Cardiol. 1997. doi:10.1016/S0735-1097(97)00223-4.

[7] Triedman J, Bridges N, Mayer J, Lock J. Prevalence and risk factors for aortopulmonary collateral vessels after Fontan and bidirectional Glenn procedures. J Am Coll Cardiol. 1993. doi:10.1016/0735-1097(93)90836-P.

[8] McElhinney D, Marianeschi S, Reddy V. Additional pulmonary blood flow with the bidirectional Glenn anastomosis: does it make a difference? Ann Thorac Surg. 1998. doi:10.1016/S0003-4975(98)00581-5.

[9] Calvaruso D, Rubino A, Ocello S, Salviato N, Guardì D, Petruccelli DF, Cipriani A, Fattouch K, Agati S, Mignosa C, Zannini L, Marcelletti C. Bidirectional Glenn and antegrade pulmonary blood flow: temporary or definitive palliation? Ann Thorac Surg. 2008. doi:10.1016/j.athoracsur.2008.01.013.

[10] Mendelsohn A, Bove E, Lupinetti F, Crowley D, Lloyd T, Beekman R. Central pulmonary artery growth patterns after the bidirectional Glenn procedure. J Thorac Cardiovasc Surg. 1994. doi:10.1016/S0022-5223(94)70049-4.