Short and Mid-Term Outcomes of Percutaneous Ostium Secundum Atrial Septal Defect Closure in a Single-Centre Adult Population within the National Program of Percutaneous Procedures in Adults with Congenital Heart Disease

Patients with ostium secundum ASD referred to our institution from November 1^st 2017 to September 30^th 2021 were assessed for percutaneous closure within the National Program for percutaneous procedures in adults with congenital heart disease. A total of 66 patients were assessed for the percutaneous procedure. Of the 66 patients, 2 had too small defects with no indication for closure, 4 had too large defects (> 36 mm) and were referred for surgery, 2 had associated PAH with PVR >3 Wood units and the closure was deferred, and PAH-specific treatment was initiated, 10 underwent percutaneous closure in a different centre, 4 chose surgery, 5 are still waiting for the procedure and 39 underwent percutaneous closure. The procedure was successful in 38 patients; in 1 patient the procedure was aborted due to the instability of the device (the defect's rims were too soft) and 1 patient lacked complete pre-procedure transthoracic echocardiography and was excluded from this study. Written informed consent of the patients was obtained as part of our institutional filing system, and consent regarding the use of collected information for research purposes was signed by all patients.

Indications for ASD closure were the presence of shunt related symptoms and/or echocardiographic signs of right ventricle (RV) volume overload.⁶ Patients with significant LV systolic and/or diastolic dysfunction underwent a complete hemodynamic assessment with invasive measurement of PAP and measurement of LV end-diastolic pressure and LA pressure. No patient had a LA pressure more than 10 mmHg.

All patients underwent a clinical and echocardiographic evaluation; demographics, biological data (e.g. BNP levels, haemoglobin level and platelets number) were collected. A transoesophageal echocardiography (TOE) was performed in all patients in order to assess the feasibility of percutaneous closure: ostium secundum ASD, no anomalous pulmonary venous drainage, ASD size <36 mm, all atrial septal rims >5 mm except for the aortic rim.

A transthoracic echocardiography was performed a day prior to the percutaneous closure. Two dimensional echocardiography-derived parameters of RV size (RV basal diameter from RV-focused apical 4C view, RV end-diastolic and end-systolic area) and function (TAPSE, TDI-derived lateral tricuspid annulus S wave velocity, fractional area change), RA size (RA mid-lateral diameter and end-systolic area), PA size (PA transversal diameter at end-diastole) and estimated pressure (systolic PA pressure using the peak velocity of the tricuspid regurgitation jet, mean PA pressure using the peak velocity of the pulmonary regurgitation jet and the size and respiratory variations of the inferior vena cava), left chambers size (LV end-diastolic and end-systolic diameter, LA antero-posterior diameter and volume, LV systolic (TDI-derived lateral and septal mitral annulus S wave velocity) and diastolic function (E and A waves of the mitral flow, TDE, E/e’ ratio) were obtained as recommended by current guidelines.^7,8 Whenever the acoustic window was appropriate, 2D speckle-tracking echocardiography was used to assess RV longitudinal strain as an average value of 6-segments from the RV-focused apical 4C view.

A similar echocardiography study was performed at follow-up (at 1, 3 months, 1 year after the procedure and yearly afterwards). Due to the Covid 19 pandemic during the last 2 years this follow-up schedule was not possible. For this reason, the last follow-up data were used in the present analysis.

A control group of 33 age- and gender-matched healthy volunteers was analysed in order to assess the right and left heart remodelling and function in ASD patients.

Percutaneous closure of the ASD was performed under general anaesthesia using fluoroscopic and 3DTOE guidance. Prior to the ASD closure, re-measurement of defect size and rims was performed from all the dedicated views in 2D and 3D-derived 2D images (Fig. 3) using the VIVID E95 ultrasound machine (GE, Horten, Norway). The largest ASD diameter was considered and device size was established by adding 4–8 mm to the largest ASD diameter. Balloon-sizing was performed only in 2 cases with floppy rims; this technique, which could have overestimated the ASD size, was not necessary in the other cases due to a good assessment of the ASD diameter by TOE. The deployment of the device was both fluoroscopic and TEE-guided. Percutaneous ASD closure was performed using Cocoon and Lifetech Cera devices. Post-interventional treatment included 100 mg/day of aspirin and 75 mg/day of clopidogrel for 3 months, followed by aspirin monotherapy for up to 5 years.

Continuous variables are displayed as mean and standard deviation. The paired-samples t test was used to compare paired continous data. Categorical data are presented as n (%) and differences were compared using Fisher's exact test. Two-sided p value <0.05 was considered statistically significant. Data were analyzed using IBM SPSS Statistics version 23 (IBM Corp., Armonk, NY, USA).

The study population included 37 adult patients of whom 29 (78.4%) were women. At the time of percutaneous ASD closure, the patients had a mean age of 43.5 ± 14.4 years (range 19 – 68 years). The most frequent symptom was fatigue (28 patients, 75.7%) with 27 patients (73%) in NYHA functional class II and 1 patient (2.7%) in NYHA class III. Nine patients (24.3%) were asymptomatic or exhibited fatigue only at intense exertion (Table 1). Eight patients had a history of supraventricular arrhythmias (7 patients had atrial fibrillation and 1 patient atrial flutter) and 1 patient had persistent atrial fibrillation at the time of ASD closure. Among other comorbidities, 3 patients (8.1%) had coronary artery disease, 1 of the 3 had also moderate aortic stenosis and 1 patient (2.7%) had a history of stroke.

Considering the associated congenital heart defects, 2 patients (5.4%) had an associated valvular pulmonary stenosis (one moderate and one severe) and 1 patient (2.7%) had an aneurysm of the atrial septum. One patient had two ASDs (22/14 mm and 9/7 mm); we chose to close the largest defect with a 26 mm device, the residual shunt through the smallest defect being haemodynamically nonsignificant.

The mean value of the BNP levels was mildly increased: 55.7 ± 48.4 pg/ml (range 10–189.5 pg/ml).

Table 2 displays the main echocardiographic parameters of the 37 patients included in the present study compared with the healthy controls, showing the right and left heart remodelling in the presence of ASD.

The ASD patients had a significant dilation of the RV (higher basal diameter, end-systolic and end-diastolic area) and RA (higher mid-lateral diameter, endsystolic area) compared to controls. The parameters of RV longitudinal and radial function had higher values in ASD patients, but statistical significance was only reached for TAPSE (25.9 ± 3.8 mm vs 24.0 ± 2.8 mm, p = 0.023) and RV fractional area change (51 ± 5% vs 47 ± 6%, p = 0.009). The RA was also significantly dilated in ASD patients. Moreover, the pulmonary artery (PA) trunk transversal diameter was increased in ASD patients compared to healthy volunteers (27.6 ± 5.1 mm vs 18.6 ± 2.7 mm, p <0.001) and the estimated pulmonary artery pressures (PAP) were also significantly higher in the ASD group (systolic PAP 37 ± 7 mmHg vs 23 ± 4 mmHg, p <0.001; mean PAP 23 ± 7 vs 16 ± 2 mmHg, p = 0.007).

The LV diastolic diameter was significantly smaller in ASD patients compared to healthy volunteers (42.7 ± 3.7 mm vs 46.7 ± 4.5 mm, p <0.001) while the endsystolic diameter was similar in both groups. The LV systolic function was normal in ASD patients, while parameters of LV diastolic function (septal and lateral e’, E/e’ ratio) were mildly impaired in ASD patients (Table 2). The LA was significantly dilated in ASD patients (Table 2).

The maximum ASD diameter assessed using TOE was 9 to 30 mm with a mean of 19.2 ± 4.6 mm. Device size ranged from 12 to 34 mm, with a mean of 24.3 ± 4.8 mm. No intraprocedural complications occurred.

The ASD size significantly correlated with RV dimensions (RV basal diameter, RV systolic and diastolic area), RA size (RA mid-lateral diameter and RA endsystolic area) and with PA trunk transversal diameter (Table 3) but not with LV or LA size. Also, ASD size did not impact RV function parameters or estimated systolic PAP (Table 3).

At least 1 follow-up visit data was available in 23 patients and the mean follow-up was 12 ± 8 months (range 1–27 months). The functional class significantly improved after ASD closure, but BNP levels remained similar between baseline and the last follow-up visit (Table 4).

Four patients presented with supraventricular arrythmias after ASD closure, all of them with a history of atrial fibrillation/atrial flutter prior to the percutaneous closure: 1 patient with preprocedural persistent atrial fibrillation underwent successful electrical cardioversion to sinus rhythm after ASD closure; 1 patient with a history of paroxysmal atrial fibrillation presented in the first month after the procedure with an episode of atrial flutter that required electrical cardioversion; 1 patient with a history of paroxysmal atrial flutter developed atrial fibrillation one year after ASD closure that was chemically converted to sinus rhythm. No other early or mid-term complications were noticed during follow-up.

The echocardiographic data showed a significant decrease in RV and RA size after the ASD closure (Table 4). The RV longitudinal function parameters (TAPSE, RV free wall S-wave) decreased after ASD closure. Also, at the follow-up visit, the PA trunk diameter (24.4 ± 5.0 mm vs 27.7 ± 5.8 mm) and the systolic PAP (29 ± 5 mmHg vs 39 ± 8 mmHg) were significantly lower (p <0.001) compared with the pre-procedural values. The LV end-diastolic diameter increased after ASD closure (45.6 ± 2.8 mm vs 42.8 ± 2.9 mm, p <0.001), while the increase in LV end-systolic diameter did not reach statistical significance (Table 4).

The ASD patients were divided into two groups according to their age: less than 60 and more or equal to 60 years. The frequency of supraventricular arrhythmias was similar in both groups, but patients older than 60 years were more functionally impaired (Table 5).There were no differences in baseline (pre-procedural) biological or echocardiographic parameters between the two subgroups of ASD patients, except for the baseline estimated systolic PAP which was higher in patients over 60 years (44.4 ± 5.9 mmHg vs 35.7 ± 6.6 mmHg, p = 0.01) and for the LA dimensions (LA antero-posterior diameter and LA volume) which were higher in older patients (Table 5). Despite these differences, reverse remodelling of the right chambers was similar in both groups: the decrease in RV, RA and PA size was similar in patients younger and older than 60 years (Table 5) but the decrease in estimated systolic PAP was higher in older patients (16.3 ± 4.3 mmHg decrease in patients over 60 years vs 7.5 ± 5.6 mmHg in patients below 60 years, p = 0.01). The BNP value at the last follow-up visit was higher in ASD patients older than 60 years (40.2 ± 8.9 pg/ml vs 23.8 ± 11.2 pg/ml) even if the change in LA volume was not significantly different in the two groups (−1.5 ± 5.5 ml/m² in patients older than 60 years vs 2.5 ± 7.4 ml/m², p = 0.39).

The first limitation of our study is the small number of patients with an even smaller proportion of patients in the two groups of age – less than 60 and more than 60 years. Furthermore, the duration of follow-up was relatively reduced and so late complications, such as arrhythmia, may have been underestimated in the study. Also, novel parameters of RV function such as RV global longitudinal strain were not available in all patients during follow-up.