Long-Term Follow-Up After Transcatheter Closure of Atrial Septal Defect in Children

Over the past 20 years, closure of a secundum Atrial Septal Defect (ASD) has switched from a surgical approach to a transcatheter percutaneous approach. There are only few long-term studies on the efficacy and safety of percutaneous ASD closure in children. Data on all 220 children who admitted to the catheterization laboratory between 1999 and 2011 for attempted ASD closure were retrospectively reviewed in conjunction with a study performed at our institution to assess the results of surgical closure during the same study period. Of 220 patients, 205 underwent attempted ASD closure (mean age 7.96 years±4.15 years, mean weight 29.05Kg±17.0Kg), with effective closure in 203 patients (success rate: 92%). There were no deaths. There were no complications in 179 patients (88%). Major complications occurred in three patients (1.5%): two embolizations of the device (1%) requiring cardiac surgery to retrieve the device and close the ASD, and one patient (0.5%) with pericardial effusion requiring pericardial drainage. In hospital, minor transient complications were recorded in 22 patients (10.7%): 12 arrhythmias (5.8%); seven vascular complications (3.4%); four cases of nausea and vomiting (1.9%); three patients with headache (1.5%); two with skin rash (1%); one with chest pain (0.5%); one with epigastralgia (0.5%); and one skin infection at the puncture site requiring local treatment and antibiotics (0.5%).

treatment in 65% of children with isolated ASD2 at our institution.
In conjunction with a study performed to assess the results of ASD2 surgical closure, this review provides comparative data between the two approaches.

Patients and Methods
We retrospectively studied 220 children (≤18 years) who had undergone attempted closure of ASD2 by catheterization between 1999 and 2011. All the children had a significant ASD, diagnosed by Transthoracic Echocardiography (TTE), susceptible to be closed by catheterization according to physicians' opinions.
A significant ASD was defined by a pulmonary-to-systemic flow ratio of more than 1.5 (estimated by Doppler TTE). Patients underwent a physical examination, chest x-ray, EKG, and TTE echo the day before percutaneous closure. A blood test including the usual coagulation tests was performed to exclude thrombopenia or coagulation disorders. The procedures were performed under general anesthesia. An initial Transesophageal Echocardiography (TEE) specified the localization, size, and rims of the ASD.
Parenteral antibiotics were given (cefazolin 30mg/Kg) and vascular access was usually obtained via the femoral vein. Heparin (100IU/Kg) was administered prior to the procedure. During the procedure, the stretched diameter of the ASD was specified using a PTS® sizing balloon (B. Braun Interventional System, Bethlehem, Pennsylvania, USA) measured by angiography and TEE. An exchange guide was placed into the left superior pulmonary vein. Along this guide, a sheath was introduced into the left atrium. The device was deployed under fluoroscopic and echo graphic guidance. Before the release of the device, a Minnesota wiggle (push-pull maneuver) was performed to ensure the correct positioning and stability of the device, and TEE was then performed to rule out any interference from adjacent structures. After the effective release, a last TEE controlled the position of the device, a potential residual shunt, or any interferences with adjacent structures. Patients remained hospitalized for one night under EKG monitoring. Before discharge, every patient underwent an EKG, TTE, chest x-ray, and femoral vessel echography. A 6-month course of aspirin (3-
All patients had a single centrally located ASD2. ASD was an isolated

Discussion
Since ASD transcatheter closure became an alternative to surgical procedure using cardiopulmonary bypass, closure of the secundum ASD has increasingly switched from a surgical approach to a percutaneous transcatheter approach in most instances.
There are only few long-term studies on the efficacy and safety of percutaneous ASD closure in children. The data on all 220 children who were admitted to our catheterization laboratory between 1999 and 2011 for attempted ASD closure were retrospectively reviewed, in conjunction with a study performed at our institution to assess the results of surgical closure during the same study period [4].
The current study shows that percutaneous ASD closure was feasible in 205/220 children referred to the catheterization lab for ASD closure (93%). The procedure was successful in 203/205 children (99%), with secondary embolization of the device in two cases; very effective with no residual shunt in 191/203 patients (94%); and with a non-significant shunt (QP/QS ratio <1.5) on echocardiography in the 12 patients with a residual shunt. A similar 94% success rate, with no residual shunt at late follow-up, was reported in the literature [5]. Major complications occurred in three patients (1.5%) during the first 24 hours after ASD closure. Surgery was required in two patients, secondary to device embolization (1%), an incidence similar to that reported by another study [6], and one patient required the drainage of a pericardial effusion. We did not experience any fatal issues relating to the procedure nor delayed wall erosion, the latter being a rare complication with a reported incidence of 0.1-0.3% [7][8][9][10][11][12]. Cardiac perforation and erosion are rare but potentially fatal. Most cases reported in the literature were clustered within the first 6 months after device closure (76%), though erosions were still reported as late as 3 years after deployment [12], with a mortality rate of 0.05%. This observed mortality is still lower than the overall surgical mortality of 0.13% determined from the Society of Thoracic Surgeons database [13].
Other major complications described in the literature were not observed in our series: retroperitoneal hematoma, air embolism [14], deep vein thrombosis, arteriovenous fistula [15], transient ischemic attack [9], late device thrombus formation [6], and late device embolization [16]. These complications, though rare and usually observed during the first days after percutaneous closure, emphasize the need for careful monitoring and follow-up after device closure, in addition to providing appropriate information for the patients and their families on alerting symptoms. Identification of high-risk cases, early recognition, and prompt intervention can minimize the risk of undesirable events [17,8]. Following ASD closure, the incidence of symptoms reported by the patients decreased from 46.7% to 13.3% at late follow-up. Palpitations (12/203) and headaches (4/203), the most common complains, were transient, with reassuring work-up in all cases. Palpitations following ASD closure, as previously reported in the literature [5], were rarely associated with arrhythmia. In comparison with surgical ASD closure, conducted during the same period of time at our institution [4], the results after device closure compare favorably in terms of mortality and morbidity (Table 1) for the surgical group in the paper published by Du et al. [18], along with an undesirable effect rate of 5.9% for the device group using the Helex septal occluder versus 10.9% for the surgical group in the paper published by Jones et al. [19]. There were no deaths in the device group, whereas there was one death from pericardial tamponade in the surgical group [19]. Due to the ease of implantation and low rate of major complications, percutaneous closure of ASD when technically possible, is currently the preferred therapeutic option. Pre-intervention TTE is essential to delineate the septal defect, evaluate the rims, and detect additional defects or other cardiac abnormalities. Pre-intervention TEE is useful to further specify the septal defect, confirm the size of the rims, specify the relationship between the device and cardiac structures, and control for any complication, both during and at the end of the procedure. As it remains difficult to correctly identify ASD and its rims by TTE, TEE is of particular relevance. Selection of the correct device size, aseptic technique, correct heparinization, stability maneuver, antibioprophylaxis, as well as antiplatelet therapy are likely to decrease the complication rate. In this study, although some historical ASD occluders were employed, the majority of ASD closures were performed using Amplatzer Septal Occluders and, more recently, by means of Occlutech® devices. Actual ASD closure devices are very safe and able to close the vast majority of secundum ASDs. Device closure of very large defects, ASD with small rims, and multifenestrated ASD is still challenging, but can be facilitated by the greater device flexibility following deployment achieved through a bioptome-like delivery system [20]. To decrease exposure to radiation, increased use of TEE is being proposed, as it may be associated with better ASD management, especially in children, who are very susceptible to radiation.

Study Limitations
During the study period (11 years), ASD closure devices have improved to some extent. The great variety of devices used could be seen a bias for the correct interpretation of our results, given that the numbers for each device were too small to obtain relevant statistics. Retrospective studies inevitably involve selection bias and the lack of some informative parameters.

Conclusion
Percutaneous ASD closure is a safe procedure in children, with regards to medium-and long-term follow-up. This study confirms that percutaneous ASD closure is an appropriate alternative to surgical closure. Despite major advances made in the field of ASD closure devices, along with the increasing experience in implanting such devices, complications tend to persist and cannot be eliminated. Careful patient selection, good understanding of potential complications, appropriate information provided to the patient's family, and close monitoring after percutaneous ASD closure are all mandatory measures to decrease complications [17][18][19][20].