Patient population

From September 2009 to December 2012, 131 patients underwent ASD closure with either an ASO ( n = 100) or an FSO ( n = 31; Flex I device, n = 16; Flex II device n = 15). A retrospective comparative analysis of the two groups was performed.

Devices

The FSO device is a double disc system, similar to the ASO, with different structural modifications that make it quite attractive. This device is made of a nitinol wire mesh, to create a smooth and flexible outer layer using a unique braiding technique. The two retention discs are connected to a central 4 mm waist, and the size of the device is determined by the diameter of the waist, as usual. The left atrial disc is usually 12–16 mm larger than the waist and the right atrial disc is 8–11 mm larger than the waist . Polyester patches are sewn within both discs and the waist to facilitate thrombogenicity and to increase the occlusion rate. Compared with the ASO ( Fig. 1 ), the FSO has a reduced amount of material, with no hub on the left disc to reduce trauma risk and clot formation ( Fig. 2 ). The connecting system from the right disc to the distal tip of the delivery cable has evolved from a microscrew initially – as in the ASO – to a hub attached to the loader by two lateral hooks. With this latter connection, the double disc can be angled some 50° without tension on the system ( Fig. 3 ). All these modifications have increased the flexibility of the device. The FSO is available in different sizes, ranging from 4 mm to 40 mm .

Amplatzer ^® Septal Occluder: notice the hub on the left disc. The connecting system on the right disc uses a screwing mechanism, and the angle between the delivery cable and the device is 90 degrees.

The Figulla ^® ASD Occluder is a double disc system with a reduced amount of material and no hub on the left disc compared with the Amplatzer ^® Septal Occluder.

Figulla ^® ASD Occluder: the delivery cable allows a tilt of up to 50° without any tension, facilitating placement before release.

Implantation technique

For both groups, indications for ASD closure were pulmonary-to-systemic flow ratio > 1.5:1 with right ventricular volume overload. Exclusion criteria were similar for both devices, including:

• •
  

  small ASD with (mainly) a non-significant shunt;

  
  
• •
  

  no sign of right ventricular dilatation;

  
  
• •
  

  no reactive elevated pulmonary vascular resistance; associated cardiac lesions requiring surgical repair (mainly associated partial anomalous pulmonary venous return);

  
  
• •
  

  deficient rim (the minimal rim accepted was 4–5 mm, except for the anterior rim, which could be completely absent);

  
  
• •
  

  bleeding condition, such as untreated ulcer; and contraindication to aspirin .

Informed written consent to the procedure was obtained from all patients or their parents before closure.

The implantation technique was similar for both devices. Usually, the procedure was carried out under general anaesthesia with transoesophageal echocardiography guidance for device implantation. Patients received intravenous heparin (100 IU/kg) at the beginning of the procedure. Activated clotting time measurements were not usually taken during the procedure. Device choice was determined by measurement of the stretched diameter, using a compliant balloon catheter (Equalizer™ Occlusion Balloon Catheter; Boston Scientific, Natick, MA, USA) placed across the defect and controlled by colour Doppler transoesophageal echocardiography, using the “stop-flow” technique. The size of the occluder implanted was usually the same as the stretched diameter ± 2 mm . The appropriate Mullins delivery sheath was then advanced into the left atrium over a guidewire previously placed in the left upper pulmonary vein. The introduction of the occluder within the delivery sheath was achieved by a loader for the FSO flushed with saline serum. The connection to the delivery sheath was performed after de-airing the delivery sheath under water to avoid any risk of air embolism. Device positioning techniques were similar for both devices and correct positioning was confirmed by means of transoesophageal echocardiography and fluoroscopy. When placement was judged to be appropriate, the occluder was released by unscrewing the delivery cable or by advancing the two hooks outside the loader for the FSO. The patient left the hospital 1 or 2 days after implantation, and received aspirin 75–160 mg daily for a 6-month period.

During follow-up, control transthoracic echocardiography was carried out 1 month, 3–6 months and 12 months after implantation. Residual shunting was defined on colour Doppler as trivial (< 1 mm colour width), small (1–2 mm colour width), moderate (2–4 mm colour width) and large (> 4 mm colour width).

Statistical analysis

Analysis included patient criteria (i.e. age, sex, weight, size of the defect established from the stretched diameter), success of implantation, procedure duration, fluoroscopy time, radiation dose and rate of full occlusion on colour Doppler transthoracic echocardiography at implantation and during follow-up. For continuous variables, results are expressed as means and standard deviations (SDs) in case of normal distribution and as medians and SDs otherwise. The normality of distribution was tested by the Shapiro–Walk test. Categorical variables are expressed as frequencies and percentages. For continuous variables, comparisons between the two groups (patients with the FSO device or the ASO device) were performed using Student’s t -test or the Mann–Whitney test, according to the distribution of the variable. For categorical variables, comparisons were performed using the chi ² test or Fisher’s exact test. Odds ratios (ORs) with 95% confidence intervals (CIs) were computed. Because our study was not randomized, we used a propensity score method to adjust the analysis for potential differences between the two groups (FSO–ASO). To compute the propensity score, we used multivariable logistic regression, with the group considered as dependent variable and the following as independent variables: stretched diameter, weight, age, fluoroscopy time, radiation dose and procedure duration. The comparisons between the two groups were then adjusted for the propensity score.

We used an analysis of covariance for continuous variables. For continuous variables, adjusted means with 95% CIs were computed. The adjusted effect of the device on each outcome was assessed by using multivariable logistic regression, with the device and the propensity score as dependent variables. Adjusted ORs with 95% CIs were computed. Statistical significance was defined as P < 0.05. Statistical analyses were performed using SAS software (SAS Institute, Inc., Cary, NC, USA).