Why, When and How Should Atrial Septal Defects Be Closed in Adults

In this review, evidence is presented to indicate that hemodynamically significant (right ventricular volume overload) atrial septal defects (ASDs) in adults should be transcatheter occluded, irrespective of symptomatology. While surgical closure is safe and effective, device closure carries less morbidity. Several devices have been investigated over the last few decades, but at the present time, only two devices, namely, the Amplatzer and the Helex, have received FDA approval; the former is useful in most defects, while the latter is useful in small- and medium-sized defects. A detailed description of Amplatzer device implantation is presented. Finally, approaches to occlude ASDs with complex anatomy are reviewed.


Introduction
The most common defects in the atrial septum are ostium secundum, ostium primum and sinus venosus atrial septal defects (ASDs) and patent foramen ovale. The management of ostium primum and sinus venosus defects is by surgery because of associated abnormalities, namely, cleft in the mitral valve causing mitral regurgitation in ostium primum defects and partial anomalous pulmonary venous connection in sinus venosus defects and is addressed in Chapter 1. Patent foramen ovale (PFO) in relation to presumed paradoxical embolism, platypnea-orthodeoxia syndrome, migraine, decompression illness and others may also require closure and the considerations for closure of such PFOs are different than those of closure of ostium secundum ASDs and some of these are discussed in other chapters in this book and will not be addressed in this chapter. In this chapter only ostium secundum ASDs in adult subjects will be discussed; I will address issues related to why, when and how should atrial septal defects be closed in these subjects. The methods of transcatheter closure in adults will also be reviewed as are the approaches to occlude complex forms of ASD.

Why should atrial septal defects be closed in adults?
In the past it was generally thought that closure ASDs in adult subjects is not necessary if they are not symptomatic. Some early studies (Ward 1994, Gatzoulis et al 1996, Webb 2001 suggested that there is no major benefit if surgical closure is performed in adulthood. Based on more recent analysis however, it would appear that the ASDs should be closed as and when they are identified. The purpose of this section of this chapter is to present evidence that the ASDs in adults should be closed.

Evidence in favor of closing ASDs in adults
In this section I will review some of the published evidence supporting closure of ASDs in all adults

Complications in unrepaired ASD patients
In a follow-up study (Rosas et al 2004) of 200 patients older than 40 years (49 ± 9 years) with unrepaired ASD for 2 to 22 years, it was found that 37 (18.5%) had major events, namely heart failure in seven, sudden death in five, severe pulmonary infection in 13, embolism in five, stroke in four and miscellaneous complications in three. In addition, more than half of the patients had dyspnea at follow-up evaluation. Predictors of complications were analyzed and age at presentation, elevated pulmonary artery pressures and O2 saturation less than 80% were found to be associated with complications. These data suggest that major cardiovascular events are likely to occur in older adult patients with unrepaired ASD.

Safety and efficacy of surgical closure
Horvath et al (1991) examined safety and efficacy of surgical closure of ASDs. In this study, surgical closure was performed in 166 patients with a mean age of 44 years who had an average pulmonary to systemic flow ratio (Qp:Qs) of 3.0:1.0. The operative mortality was 1.2% (two deaths). The remaining patients were followed for a mean of 7.5 years. The overall survival and event-free survival rates were 98% and 97% at five years, respectively. Similarly, ten-year overall (94%) and event-free (92%) survival rates were high. Their (Horvath et al 1991) conclusion was that surgical closure is safe and effective with high event-free survival rates. Konstantinides et al (1995) made a comparison of surgical closure with medical follow-up without surgery. One hundred-seventy-nine patients older than 40 years were examined; 84 of these had surgical closure while 95 had no surgery. The follow-up duration for both groups was 10 years. The actuarial 10-Year survival rate was 95% for the surgery group and 84% for no surgery group. In addition surgery also appears to have prevented deterioration of NYHA functional class. Based on these data the authors (Konstantinides et al 1995) conclude that surgical repair of ASD in adult subjects increases long-term survival and decreases functional deterioration when compared to medical therapy (no surgery).

Effect of ASD closure on cardiac function
Myocardial performance index (MPI), a Doppler-derived non-geometric measure of ventricular function, has been used to for quantitative assessment of ventricular function in patients with congenital heart disease both in adults and children; this measure appears to be relatively independent of changes in preload and afterload (Eidem 2000). Right ventricular (RV) MPI did not improve following surgical closure of ASD despite relief of RV volume overload (Eidem 2000). This was attributed to adverse effect of cardiopulmonary bypass on ventricular function. Salehian et al (2005) evaluated twenty-five patients at a mean age 46 years prior to and 3 months (mean) after device closure of ASD. Right ventricular MPI improved from 0.35 ± 0.14 to 0.28 ± 0.09 (p = 0.004) while left ventricular MPI enhanced from 0.37 ± 0.12 to 0.31 ± 0.11 (p = 0.04) (Figure 1). These authors (Salehian et al 2005) conclude that ASD closure improves cardiac function.

Effect of ASD closure on functional capacity
Improvement in functional capacity following ASD closure was studied by Brochu et al (2002). Thirty-seven patients with a mean age of 49 years whose mean Qp:Qs was 2:1 were evaluated. The VO2 max was measured and NYHA classification assessed prior to and 6 months after ASD closure. They found that VO2 max improved from 23 ± 6 to 27 ± 7 (p < 0.0001) following ASD closure. Fifteen out of 37 patients were in NYHA Class I prior to www.intechopen.com

Summary of why should atrial septal defects should be closed in adults
Based on review of the above and other reports, I conclude that untreated ASD patients tend to have decreased event-free survival rates when compared to normal population and surgical closure is safe and effective with high event-free survival rates. ASD closure also prevents functional deterioration, improves cardiac function and increases functional capacity. Consequently all adult patients with ASD should undergo closure of ASD.

When should atrial septal defects be closed in adults?
Murphy et al (1990) examined the effect of age at surgical closure of ASD. Patients who had surgical closure of ASD, performed between 1956 and 1960 at Mayo Clinic, were studied; they followed 123 patients and compared their actuarial survival rates with those of normal population. In the groups of patients who had surgery after 24 years of age, the actuarial survival rates are lower ( Figure 2). When surgery is performed prior to 24 years of age, there was no significant difference in survival rates. The earlier the surgery was performed the better were the 27-year survival rates ( Figure 2).
Based on these data Murphy concludes that early intervention may be beneficial; earlier the closure, the better is the long-term outlook. Consequently, it is prudent to close hemodynamically significant ASDs in all adults. Since there is no advantage in waiting beyond 24

How should atrial septal defects be closed in adults
Following the introduction of cardiopulmonary bypass techniques for open heart surgery and the description of surgical closure of atrial septal defect (ASD) by Gibbon, Lillehei and Kirklin in 1950s, it rapidly became a standard form of treatment for atrial defects. The conventional treatment of choice of moderate and large defects until recently is surgical correction. Although surgical closure of ASDs is safe and effective with low mortality (Galal et al 1994, Pastorek et al 1994, the morbidity associated with sternotomy/thoracotomy is unavoidable. Consequently, substantial efforts have been made by the cardiology community to develop a non-surgical, catheter-based method of ASD occlusion. Since the initial description in mid 1970s by King (Rao 2003a) is now an established practice in most centers providing state of the art care to patients with heart disease.

Surgical closure
When surgical closure is contemplated, a median sternotomy or a right sub mammary incision is made under general anesthesia. The aorta and vena cavae are cannulated and the patient is placed on cardiopulmonary bypass and right atriatomy is performed. The defect is exposed and closed either by approximating the defect margins with suture material or by using a pericardial patch, depending upon the size of the defect.
However, at the present time, surgical repair is largely reserved for ASDs with poor septal rims in which the interventional cardiologist opines that defect is difficult to close with trans-catheter methodology or was unsuccessful in closing the defect. If intra-cardiac repair of other defects is contemplated, surgical closure of ASD could also be performed at the same time.

Transcatheter closure
A number of devices are available to the interventional cardiologist for closure of ASD, but selection of an appropriate device is difficult because of lack of randomized clinical trials.  (Rao 1998a, Rao 1998b, Rao 2000, Rao 2003c) the implantation feasibility (ratio of implantations vs. patients taken to the catheterization laboratory with the intent to occlude), percentage of device dislodgements/miss-placements/embolizations, percent of patients with effective occlusion and re-intervention-free rates during follow-up; these results were tabulated elsewhere (Rao 2000, Rao 2003c). These comparisons revealed that these parameters are similar and comparable for most, if not all devices that I had the opportunity to evaluate. While the feasibility, safety and effectiveness are most important, availability, cost, size of the delivery sheath and other factors should also be considered in the process of device selection.
Of the devices tabulated in the prior publications (Rao 2000, Rao 2003c) and others that entered clinical trials since those reviews (see chapter 1), some devices were discontinued, shelved or withdrawn because of different reasons and some others continue to be in clinical trials either within or outside the US. Amplatzer (AGA Medical Corp., Golden Valley, MN) and HELEX (W.L. Gore, Flagstaff, AZ) devices are the only devices approved by the FDA at the present time, for general clinical use for closure of the ASDs.
The Amplatzer septal occluder is the most commonly used ASD closure device worldwide at the present time. The feasibility, safety and efficacy of device occlusion are based on self-expandable, retrievable and re-positionable design of the device (Hamdan et al 2003). Even very large defects can be closed successfully with Amplatzer device using a variety of techniques (Nagm andRao 2004, Rao 2007).

Diagnosis and indications
After a clinical and echocardiographic diagnosis of moderate to large ostium secundum ASD is made, consideration for transcatheter closure should be given. Because of poor echo windows, most adult subjects require transesophageal echocardiography (TEE) to confirm the diagnosis, to quantify its size and define the septal rims. The indications for closure in adults are similar to those used in children (see Chapter 1) and are echocardiographic finding of right ventricular volume overloading and/or catheterization findings of Qp:Qs greater than 1.5:1.0. The reasons for closure of ASDs in children are prevention of pulmonary vascular obstructive disease in adulthood, to prevent arrhythmias and to prevent symptoms later in life. Additional reasons in adult subjects are to prevent heart failure, prevent functional deterioration, improve myocardial function and prevent paradoxical embolism.

Consent, catheterization and transesophageal or intracardiac echocardiography
Informed consent is obtained and cardiac catheterization is performed preparatory to transcatheter occlusion, at the same sitting. Right heart catheterization is undertaken percutaneously to confirm the clinical and echocardiographic diagnosis with particular attention to exclude partial anomalous pulmonary venous return.

Device description and implantation
Detailed descriptions of Amplatzer Septal Occluder and HELEX devices and their implantation were included in Chapter 1 of this book. Device placement protocol in adults is similar to that described for children and will not be detailed here except to state that the procedure is performed more often under ICE guidance in adults than in children and Clopidogrel 75 mg/day for the first 2 to 3 months after device implantation in addition to Aspirin 185 or 325 mg/Kg/day by mouth for six months is given in adults. However, discussion of some issues germane to adult subjects and device closure of complex ASDs will be included hereunder.

Precautions in subjects with pulmonary hypertension
In some adult ASD patients pulmonary hypertension may be present. In these patients, a particular attention should be paid to calculate pulmonary vascular resistance. Pulmonary vascular resistance (PVR) may be calculated:

PVR = (Mean PA presence -Mean LA pressure)/Pulmonary blood flow index
Where, PA and LA are pulmonary artery and left atrium respectively.
The calculated resistance is normally between 1 and 2 units and a resistance higher than 3.0 units is considered elevated. Marked elevation of the resistance (>8.0 units) contraindicates closure of the ASD. When the resistance is elevated, oxygen and other vasodilating agents, particularly Nitric oxide (NO) should be administered to demonstrate the reversibility. In addition, pulmonary arterial wedge angiography and sometimes, even lung biopsy may be necessary to determine the suitability for closure. Patients with calculated pulmonary vascular resistance less than 8 wood units with a Qp:Qs >1.5 are generally considered suitable candidates for ASD occlusion. In patients with increased pulmonary vascular resistance, if the calculated resistance drops to levels below 8 units after administering oxygen or other vasodilator agents (NO), the patient becomes a candidate for closure of ASD.
If the results of the testing of pulmonary vascular reactivity are marginal or the pulmonary vascular resistance remains elevated (>8.0 units) following vasodilator testing, a fenestrated Amplatzer device may be implanted across the ASD (Lammerset al 2007, Kretschmar et al 2010). The device will reduce the left-to-right shunt, thus removing the effect of continued increase in pulmonary blood flow and may result in improvement. Should the pulmonary vascular resistance continue to increase despite the fenestrated device closure, the fenestrations in the device will serve as a pop-off escape mechanism and maintain near normal cardiac index, though at the expense of arterial oxygen desaturation.

Approaches for closure of complex ASDs
Secundum ASDs located centrally in the atrial septum are found in only 24% of cases (Podnar et al 2001). These authors reviewed the characteristic of ASDs in 190 patients who had transcatheter or surgical repair and found deficient superior anterior rim in 42%, deficient inferior posterior rim in 10%, perforated aneurysm of the atrial septum in 8%, multiple defects in 7%, deficient inferior anterior and superior anterior rims in 3%, deficient inferior posterior and posterior rims in 2% and deficient inferior anterior, superior posterior and coronary sinus rims in 1 % each. In another study complex ASDs were present in 40 (28%) of 143 patients (Pedra et al 2004). These authors arbitrarily defined complex anatomy as ASDs with stretched diameters larger than 26 mm with a deficient (<4 mm) rim in 23 (16%), two separate defects with a distance greater than 7 mm in 8 (5.6%), fenestrated atrial septum in 5 (3.5%) or redundant and hyper mobile (>10 mm) atrial septum in 4 (2.8%). In the ensuing paragraphs I will address how the complex ASDs can be closed by transcatheter methodology.

Amplatzer device
While deploying the Amplatzer device in ASDs with deficient anterior superior rim, the left atrial disk tends to become perpendicular to the atrial septum leading to prolapse of the left disk into the right atrium. Several techniques have been proposed to overcome such difficulties; these were reviewed elsewhere ( delivery sheath. The cardiologists should consider all options reviewed above and select the method that is most likely to result in successful implantation of the device in their patient.

HELEX device
Since the HELEX device is not suitable for large defects, it should not be used to occlude large ASDs.  (2004) defined absent PI rim as a rim < 3 mm. As the difference in radius length between right and left atrial disks of the Amplatzer device is 2-3 mm, a rim < 3 mm will not allow both disks to hang on both sides of the rim. They found that defects with absent PI rim tend to be larger in diameter. They concluded that, although a stable Amplatzer device deployment is possible, these defects are more liable for complications such as pulmonary vein or inferior vena caval obstruction, encroachment onto the anterior mitral leaflet or frank embolization (Mathewson et al 2004). The number of cases reported is too small to make a generalized conclusion as to what is the best approach to address these ASDs.

Multiple or fenestrated defects
Multiple or fenestrated ASDs may be successfully occluded by different techniques or devices. One method used for addressing fenestrated defects was to perform balloon atrial septostomy to create a single large defect which was then closed with a single large Amplatzer device (Carano et al 2001). While these authors were successful in occluding the defect, I am not in favor of using such a technique. A large single Amplatzer device may be deployed in the larger defect to occlude two or more smaller defects as used by  and others (Roman et al 2002). While examining this technique, Szkutnik et al (2004) found that a smaller defect less than 7 mm distance from the larger defect had a 100% closure rate at 1 month follow-up. The device in the larger defect decreases the distance between the two defects or even compress the smaller defect . However, if the distance between the two defects is >7 mm, a residual left to right shunt persisted

Anuerysmal atrial septum
Single or multiple defects with septal aneurysm represent a different type of complex ASD. Such defects may be better dealt with devices that don't rely on stenting the defect (for example Amplatzer device) in order to achieve stabilization in the septum. Devices with two discs such as the hybrid buttoned device (Rao 2003d), the HELEX, the CardioSeal or the Cribriform Amplatzer (Musto et al 2009) are more appropriate choices to close such defects. Closing two small but distant defects within an aneurismal atrial septum may effectively close both defects but carry a higher risk of later development of thrombus formation (Krumsdorf et al 2004). We have successfully closed defects associated with atrial septal aneurysms with hybrid buttoned devices ( Figure 3) by compressing the aneurysm between the occluder and the square shaped counter-occluder (Rao 2003d).

Conclusions -Approaches for closure of complex ASDs
It may be concluded that most complex ASDs can be closed successfully either by using traditional or special devices or techniques. Defects with deficient or absent PI rim continue to be challenging for most interventionalists and may require surgical closure.

Long-term complications of transcatheter closure
Reports of erosion of the aortic wall by the Amplatzer device with development of aorta-toright atrium (Chun 2003) or aorta-to-left atrium (Aggoun et al 2002) fistulae led to the suggestion of over-sizing the device (4 mm larger than the measured stretched diameter) in large defects with deficient anterior superior rim. This is meant to ensure that the device disks straddle and remain flared around the ascending aorta to prevent discrete areas of pressure where erosion may occur. Of course, when over-sizing the device, care must be taken not to interfere with atrio-ventricular valve function and venous return. Device migration and erosion of aorta (Amin et al 2004, AGA 2006) was observed during follow-up in 37 out of 35,000 ( 0.11% or 1 in 1,000) Amplatzer device implants world-wide. This complication was also similar in the US study population: 18 out of 15,900 implants ( 0.12% or 1 in 1,000). Review of data by the Review Board and AGA Medical (AGA 2006) suggested that the device erosion is related to over-sizing of the device ( Figure 4) and recommended that device size >1.5 times the TEE/ICE diameter of ASD should not be used.  Fig. 4. Bar diagram demonstrating relationship of atrial septal defect (ASD) and device size data between the group of patients without perforation (Trial) and those that had perforation (Erosion). The diameter (Dia) of the ASD was similar in both groups whereas the stretched diameter (Stretch), device size (Dev size) and ratio of device to ASD (Ratio) were larger in the patients who had perforation than those who did not. Based on these data a recommendation made not to use devices larger 1.5 X ASD size (Constructed from the data of AGA 2006).

Summary and conclusions
ASDs in adult subjects should be closed at presentation, electively, irrespective of their age. Evidence was presented to indicate that hemodynamically significant (right ventricular volume overload) ASDs in adults should be transcatheter occluded irrespective of symptomatology. While surgical closure is safe and effective, device closure carries less morbidity. Multiple devices have been investigated over the last few decades, but only Amplatzer and HELEX devices received FDA approval as of this time. The Amplatzer is useful in most ASDs while the HELEX device is useful in small and medium-sized defects. Some procedural details were mentioned with particular emphasis on the need for test occlusion of ASD in the elderly. Approaches taken to occlude ASDs with complex anatomy were also reviewed. Amplatzer device appears to be best available option at the present time. Careful attention to the details of the technique are mandatory to achieve a successful outcome