Percutaneous Interventions in Adults with Congenital Heart Disease

Percutaneous Interventions in Adults with Congenital Heart Disease

Damien Kenny MB, MD

Qi-Ling Cao MD

Ziyad M. Hijazi MD, MPH, FSCAI

With improving survival rates following intervention for congenital heart disease (CHD) in infancy and childhood, the prevalence of adults with CHD is growing yearly. Indeed recent reports have suggested that the number of adults with CHD has surpassed pediatric numbers and that approximately 1 in 150 young adults will have some form of CHD (1, 2). Significant rates of intervention and reintervention are required in this cohort of patients, with one report demonstrating 20% of young adults with CHD requiring cardiovascular surgery (3). Increasingly, transcatheter alternatives to surgery exist in this population, particularly in the field of transcatheter pulmonary valve replacement (tPVR), and recommendations for catheter intervention in adults with CHD have been published in the context of overall care of these patients (2). Case complexity will vary according to the underlying diagnosis, ranging from more straightforward interventions such as atrial septal defect closure (covered in a separate chapter) to complex stenting interventions for interatrial baffle leaks in patients following atrial switch surgery for transposition of the great arteries. Table 44-1 outlines the majority of interventional procedures required in this group of patients. Many of the more complex patients may have had multiple previous interventions, and detailed review of previous data as well as sensible use of preprocedural imaging techniques is essential. A structured approach to diagnostic assessment with appreciation of the necessary data to accurately calculate the degree of intracardiac shunting is also vital and will be discussed in detail. Finally, procedural techniques and outcome data, wherever available, will be discussed in relation to the range of transcatheter interventions seen in this patient group.

TABLE 44-1 Congenital Cardiac Lesions Amenable to Transcatheter Interventions



Operation Replaced


Atrial septal defect

Device closure

Open heart repair

75% suitable for transcatheter closure (not sinus venosus or primum defects)

Ventricular septal defect

Device closure

Open heart repair

Muscular defects. Concerns regarding complete heart block with perimembranous device closure

Patent arterial duct

Coil or plug occlusion

Thoracotomy-closed repair

>90% suitable for transcatheter closure

Assess for pulmonary hypertension

Pulmonary stenosis

Pulmonary valvuloplasty

Open heart valvotomy

>85% will respond to valvuloplasty

Congenital aortic stenosis

Aortic valvuloplasty

Open heart valvotomy

Equivalent outcomes to surgica valvotomy in younger patients

Coarctation and recoarctation of aorta

Angioplasty or stent implantation

Thoracotomy-closed repair

Open surgery carries increased risk. Aortic aneurysm a risk in older patients

Venous and arterial collaterals in complex disease

Coil or plug occlusion

Surgical ligation

Recurrent intervention may be necessary

Pulmonary regurgitation

Percutaneous stented valve

Surgical homograft replacement

Evolving. Currently only suitable if adequate stenosis to anchor stent

Branch pulmonary stenosis

Pulmonary artery stent

Open surgical repair

Hybrid approach may be used if access to stenosis is technically challenging

Mustard/Senning baffle obstruction/leak

Baffle stent

Open surgical repair

Complex intervention. Covered stents necessary if baffle leak.

Post Fontan surgery-fenestration closure

Device closure

Open heart repair

40% spontaneous closure rate

Coronary artery fistula

Plug or coil closure

Open heart repair

>90% suitable for transcatheter closure. Close monitoring for myocardial infarction

Ruptured sinus of Valsalva

Plug or device closure

Open heart repair

>90% suitable for transcatheter closure


Flow and Shunt Calculations

A variety of techniques have been employed to calculate relative right and left heart cardiac outputs and intracardiac shunts; however, most laboratories now use simple oximetric methods. The foundation of these calculations is based upon Fick’s principle, which suggests that the total uptake of a substance by an organ is equal to the product of the blood flow to that organ and the arterial-venous (A-V) concentration difference (gradient) of the substance. When calculating cardiac output, the substance measured is oxygen. In estimating Qp or pulmonary blood flow, the A-V difference used is the difference between the oxygen content of the pulmonary artery and that of the pulmonary veins. In calculating Qs or systemic blood flow, the A-V difference is the difference between the oxygen content of the aorta and that of the systemic veins. Simply put, the equation reads:

As almost all oxygen in the blood is bound to the hemoglobin and each gram of hemoglobin is capable of carrying 1.36 mL of O2, the oxygen content of blood is calculated by:

Therefore, the arteriovenous oxygen content difference can be calculated by measuring the oxygen saturations in both the pulmonary arteries and the pulmonary veins. In order to calculate the cardiac output, oxygen consumption is required. This is influenced by a number of variables, including age, sex, height, and heart rate. Accurate measurement during cardiac catheterization is achievable using indirect calorimetry or mass spectrometry; however, it is commonplace to estimate VO2 values from published predictive equations or tables. It is noteworthy that studies have been published indicating poor levels of agreement between measured and estimated VO2 (from regression equations published separately by Lafarge and Miettinen, Lundell, Wessel, and Lindahl) in both sedated and mechanically ventilated patients with CHD during cardiac catheterization (4). Thus Qp can be calculated from the equation:

In the absence of any shunt, the pulmonary blood flow and the systemic blood flow should be the same. The systemic blood flow represents the flow from the aorta through the body to the right atrium (RA). Unfortunately, there is a normal variation in the saturation of blood arriving to the RA; the inferior vena caval (IVC) blood has higher oxygen content than the superior vena caval (SVC) blood because the kidneys remove far less oxygen relative to their degree of perfusion. The coronary sinus has very desaturated blood owing to the high oxygen extraction rate of the heart, but the amount of coronary sinus blood is minor, and its contribution is ignored in the equations. The IVC saturation is high enough normally that one needs at least a 11% step-up in the saturation from the SVC to the RA to be sure there is an atrial level shunt. As the blood mixes further downstream in the right ventricle (RV), a 7% step-up in the RV versus the SVC should be used to confirm a ventricular level shunt, and a 5% step-up is recommended in the PA to be confident of a pulmonary arterial left-to-right shunt. To normalize for the higher IVC oxygen content, a mixed venous (MV) saturation is derived from the formula:

Under normal circumstances with no intracardiac shunting, the aortic oxygen content should be the same as the PV oxygen content, and the MV oxygen content should be the same as the PA oxygen content. Therefore Qs is derived from:

Another concept used with calculation of shunt volume is that of effective blood flow (QEP). The QEP is the volume of desaturated systemic venous blood flowing through the lungs that is oxygenated in the lungs. In the absence of intracardiac shunting, the effective pulmonary blood flow (QEP) should be the same as the QP, which should be equal to Qs:

However, when there is a left-to-right shunt, the shunted blood has already passed through the lungs, and is therefore not part of QEP. Therefore, the volume of shunted blood is calculated by:

When there is a right-to-left shunt, QS will be formed by both the QEP and the blood that has shunted past the lungs. Therefore, the right-to-left shunt volume is calculated by:

A simpler method for assessing the systemic to pulmonary flow ratios can be calculated with the O2 saturations from the four main measurement areas used in the above calculations, namely the MV saturation, the pulmonary artery saturation, the pulmonary vein saturation, and the aortic saturation as in calculating relative QP:QS as all of the other variables will cancel each other out. Therefore, this can be calculated roughly by:

The above equation will work in providing an estimate of leftto-right shunt in most cases. However, if oxygen concentrations higher than room air are used, the dissolved oxygen content per milliliter of blood is required for accurate calculation flow. This is calculated from:

This is then incorporated into the calculations:

Calculation of Pulmonary Vascular Resistance

With invasive measurements, exact pressure recordings from the heart, pulmonary arteries, and the aorta can be obtained. It is also possible as outlined above to calculate cardiac output. Therefore, by applying Ohm’s law (pressure = flow × resistance), one can calculate accurate resistance measurements, particularly within the pulmonary arteries. This is extremely important as pressure may be elevated secondary to increased flow or increased resistance.
If flow is the predominant factor, such as in the case of a large ventricular septal defect (VSD) in a young infant, then surgical management is indicated. However, if distal irreversible pulmonary vessel remodeling occurs (which may be seen with chronic excessive pulmonary flow), then high resistance may preclude surgical intervention. This provides the cornerstone of diagnostic catheter assessment in CHD patients, where there are concerns regarding the contributing factors to increased pulmonary artery pressures. Pulmonary vascular resistance (PVR) is calculated by modifying Ohm’s law:

Similarly, the systemic vascular resistance (SVR) is derived from the mean aortic pressure, the mean RA pressure, and the systemic blood flow.

When the flow is expressed in terms of liters per minute, the units derived are referred to as Hybrid or Wood Units and measured in mm Hg/L/min, or simply referred to as “u”. It is important to note that these values when indexed to body surface area are expressed as u·m2. Smaller patients appear to have significantly greater resistance values when using nonindexed cardiac output values (because of their smaller body surface areas [BSA]), and therefore it is important to multiply, rather than divide, by the BSA to calculate the indexed resistance value. The Woods Unit number can be multiplied by 8 to convert to Pascal seconds per cubic meter (MPa·s/m3) and by 80 to derive dynes cm/s5. These values are outlined in Table 44-2.

Although initial assessment of vascular resistance may suggest significantly raised PVR, it is possible that this increased resistance may be reversible and therefore react favorably to certain pulmonary vasodilators. Such agents include oxygen and pulmonary vasoactive drugs such as adenosine, calcium-channel blockers, prostacyclin, and inhaled nitric oxide. A “positive” response is generally one in which there has been both a reduction in mean PA pressure of at least 10 mm Hg to an absolute mean PA pressure of less than 40 mm Hg without a decrease in cardiac output (5). A positive response has been associated with an improved longterm outcome, although the response may not predict clinical improvement with the various pulmonary hypertensive agents now available. Eisenmenger’s physiology is defined by the presence of pulmonary hypertension and evidence for shunt reversal (cyanosis), which occurs because of increased distal PVR secondary to chronic excessive blood flow. Initial assessment and management protocols for these patients have been defined (2). Significant symptom improvement has been shown following treatment with newer oral pulmonary hypertensive agents (6).

TABLE 44-2 Definitions in Pulmonary Hypertension

Measurement of PVR

Wood units = mm Hg/L/min (to index multiply by BSA rather than divide). To convert to Pascals multiply by 8, and to convert to dynes multiply by 80

Normal pulmonary vascular resistance (Nonindexed):

0.25-1.6 Wood units

Normal systemic vascular resistance (Nonindexed):

9-20 Wood units

Definition of pulmonary arteria hypertension:

Mean pulmonary artery pressure >25 mm Hg with PCWP/LAP/LVEDP<15 mm Hg and PVR > 3 Wood units

Acute response with testing:

Decrease in mPAP of at least 10 mm Hg to an absolute value <40 mm Hg without a decrease in cardiac output

Contraindication to closure of L-R shunt lesion

PVR > 7 Wood units (although lesions demonstrating reversibility with higher initial PVR have been closed)


Transcatheter VSD Closure

The majority of congenital VSDs are located within the region of the membranous septum (>75%) with the remainder located within the muscular septum. Rarely, these defects are doubly committed lying directly beneath the septum separating the aortic and pulmonary valves, and classically these defects have been closed surgically. Most clinically relevant VSDs are closed in childhood; however, hemodynamically significant shunts (defined as a Qp:Qs >1.5:1 in adults) (2) may become clinically relevant only in later years. Reports outlining transcatheter closure exclusively in adults have been published (7, 8). Indication for closure was symptoms and/or ventricular dilatation in 40% to 60% of patients. Other more common indications included pulmonary hypertension and endocarditis. Aortic valve prolapse leading to aortic regurgitation may be seen with small subaortic VSDs secondary to the Venturi effect in essence “sucking,” most commonly, the right or non-coronary cusps inferiorly toward the VSD, and surgery is usually preferred in these instances as device placement may impede upon the aortic leaflets, worsening the aortic regurgitation. Transcatheter closure of perimembranous defects has been halted in the United States owing to significant concerns regarding complete heart block (˜5% in the European registry series [9] and 3% in the US trial [10]) following device placement. Recovery of atrioventricular block has been reported with early treatment with steroid (11) or early surgical removal of the membranous occluder (12). Comparable rates of complete heart block following surgical closure in younger patients have been reported at less than 1% (13). Newer devices are in development with design modifications to reduce radial force on the conduction tissue and potential for heart block. Overall, reported complication rates have ranged from 6.5% in the European registry to 10.7% in the US perimembranous registry, including device embolization, cardiac perforation, stroke, and two deaths (10). Exclusion criteria have also been cited, and include inadequate distance for device placement (usually 4 mm) between the VSD and either the atrioventricular or the semilunar valves, active sepsis or PVR > 7 indexed Wood units (14). The approach to transcatheter closure is outlined in Figure 44-1.

Transcatheter closure of acquired VSDs secondary to myocardial infarction is associated with significant morbidity and mortality, mirroring surgical mortality of 50%. In a recent study looking at transcatheter closure of acute postinfarct VSDs, although procedural success was achieved in 86%, 41% of patients had a procedurerelated complication, including significant residual shunt, with
overall 30-day survival of 35% (15). In a previous study looking solely at postinfarct VSD closure, 30-day mortality figures of 28% were reported; however, a significant proportion of these patients underwent closure in the subacute or chronic phase of their acute myocardial infarction (16). Patient selection may be key, and alternative forms of imaging such as cardiac computer tomography (CT) or magnetic resonance imaging (MRI) may assist in delineating the morphology of the defect and surrounding tissue and thus whether the patient may be suitable for transcatheter closure.

Premium Wordpress Themes by UFO Themes