Mullins-Sheath Facilitated Delivery of Gore Cardioform ASD Occluder Devices for Closure of Large or Challenging Secundum Atrial Septal Defects

Objectives. To describe a deployment technique of the Gore Cardioform atrial septal defect (ASD) occluder (W.L. Gore and Associates) for large secundum ASDs and ASDs with challenging anatomy. Background. The Gore Cardioform ASD occluder has recently been approved for closure of secundum ASDs; however, there are limitations to its delivery system. Methods. A retrospective study was conducted on the use of a Mullins sheath (Cook Medical) to facilitate Gore Cardioform ASD occluder delivery for secundum ASD closure in the cardiac catheterization laboratory from June, 2017 to December, 2019 at Texas Children’s Hospital/Baylor College of Medicine. Results. Out of 98 patients who underwent an attempt at ASD closure using the Gore Cardioform ASD occluder, a Mullins sheath was used in 52 patients (median age, 8 years [interquartile range, 4-13 years] and weight 27.2 kg [interquartile range, 17.9-51.2 kg]), with a successful implant in 46/52 patients (88%). The Mullins sheath was primarily used to deliver large devices (>32 mm) in 38/46 successful implants (83%). There were 2 major adverse events (atrial fibrillation requiring cardioversion). At a median follow-up of 43 days (interquartile range, 1-374 days), no patient had more than a mild residual shunt. The ASD size, maximum sheath size, and device size were larger in patients in whom the Mullins sheath was used as compared with those patients in whom a Mullins sheath was not used. Conclusions. The Mullins sheath-facilitated delivery of the Gore Cardioform ASD occluder device may be a useful adjunct technique for closure of large secundum ASDs and secundum ASDs with challenging anatomy.

J INVASIVE CARDIOL 2021;33(6):E425-E430. Epub 2021 April 23.

Key words: atrial septal defect, congenital heart disease, pediatrics, structural heart disease intervention

The Gore Cardioform atrial septal defect (ASD) occluder (W.L. Gore and Associates) was recently approved by the United States Food and Drug administration for device closure of secundum atrial septal defects (ASDs). Approval was received in June of 2019 based on clinical data from the Gore ASSURED Clinical Study (pivotal phase) from 125 patients, which revealed complete closure in all patients with a successful implant at evaluation at 6 months.¹ The device is available in sizes ranging from 27 to 48 mm and is approved to close secundum ASDs ranging from 8 to 35 mm in diameter. Similar to prior generations of Gore devices,^2,3 the device has a wire frame structure made of nitinol (although more reinforced than before) with the majority of the device made of expanded polytetrafluoroethylene. The new Gore Cardioform ASD occluder utilizes technology that enables the waist of the device to adapt to the size and shape of the ASD.

The larger sizes of the Gore Cardioform ASD occluder device make them appealing for closure of large secundum ASDs. The device is designed to be advanced through a short sheath and delivered with its own long delivery catheter that has a slight angulation near its tip to facilitate crossing the atrial septum. However, despite this angulation at the tip, the configuration in its current iteration may not be ideal to properly align the device with the atrial septum during device delivery when closure of large secundum ASDs with deficient septal rims is attempted. We describe a technique that may overcome some of these challenges and facilitate closure of large secundum ASDs or secundum ASDs with challenging anatomy using a long, preshaped sheath, ie, a Mullins sheath (Cook Medical).

Patients. This was a retrospective study conducted to investigate the use of a Mullins sheath to facilitate Gore Cardioform ASD occluder delivery for closure of secundum ASDs from June 2017 to December 2019 at Texas Children’s Hospital/Baylor College of Medicine in Houston, Texas. All secundum ASD closures, regardless of whether or not a Mullins sheath was used, were included in the study. Institutional review board approval was obtained for the study. Demographic data were collected, in addition to procedural and follow-up data and latest clinical status.

Gore Cardioform ASD occluder device. The Gore Cardioform ASD occluder device is available in sizes from 27 to 48 mm, requiring short delivery sheaths of 10 to 14 Fr. It is available with a preloaded system, consisting of a mandrel that helps conform the device, and its own delivery catheter that has an angulation at the tip to facilitate crossing the left atrium and aligning the device for closure of many secundum ASDs.

Procedure. Cardiac catheterization procedures were performed with general anesthesia in all patients. All of the procedures were guided using biplane fluoroscopy and transesophageal echocardiography (TEE) guidance. Patients were administered heparin to achieve activated clotting times >250 seconds. Right heart catheterizations were performed in all patients. A 0.035˝ Amplatz super-stiff wire (Boston Scientific) was positioned in the left upper pulmonary vein and the ASDs were sized utilizing the stop-flow technique.

Mullins-sheath facilitated Gore Cardioform ASD occluder delivery technique. The mechanism of favorable exaggeration of the angle of the Gore Cardioform ASD occluder delivery system when used with a Mullins sheath is seen in Figure 1. In large defects or in defects with deficient rims, the angulation of the Gore delivery catheter may not allow for alignment of the device parallel to the atrial septum, with resultant device prolapse superiorly through the defect into the right atrium (Figure 2). In such situations, the short sheath was exchanged for an appropriately sized 63 cm-long or 75 cm-long Mullins sheath in the left atrium and the dilator and wire were removed (same Fr size as required short sheath Fr size). When used in combination with a Mullins sheath, the angulation is exaggerated enough to facilitate better alignment parallel to the atrial septum (Figure 3). The Mullins sheath was rotated clockwise to assume a posterior orientation to align the device parallel to the atrial septum as it was being delivered. The Gore Cardioform ASD occluder delivery catheter was then advanced through the Mullins sheath, which was positioned with continuous clockwise rotation, often by the assistant, as the primary operator deployed the device. In some cases, despite this maneuver, the device was not aligned parallel to the atrial septum. In such cases, once the left atrial side of the device was deployed and aligned in the best possible way, the Mullins sheath was withdrawn (if 63 cm-long sheath was used) to the inferior vena cava or right atrium to serve as a “fixed point,” whereby further clockwise rotation of the Gore delivery catheter could be performed if needed to facilitate further device alignment as parallel to the atrial septum as possible (Figure 4). In some cases, the Gore delivery catheter was rotated 180° clockwise despite the use of the Mullins sheath to facilitate capture of the aortic and superior vena cava rims. Use of the Mullins sheath also facilitated delivery of the device when inferior disc prolapse into the left atrium was encountered (Figure 5). Finally, in some cases, the tip of the Gore Cardioform ASD occluder delivery catheter was also hand-shaped to further exaggerate its angle. The use of the Mullins sheath was at the discretion of the implanting physician.

After the right atrial side of the device was deployed, the device was locked with subsequent removal of the retrieval cord once in satisfactory position. Hemostasis was achieved with manual pressure. All patients were observed overnight with telemetry and underwent chest radiographs, electrocardiograms, and transthoracic echocardiograms the next day. Follow-up was arranged at 1 month, 6 months, 1 year, and yearly after the procedure.

Statistical analysis. Statistical analyses were performed using SPSS, version 26 (IBM) and statistical significance was assessed at the .05 level. Normality of continuous variables was assessed using histograms. Descriptive statistics are presented as aggregates and percentages for categorical variables and medians with interquartile range (IQR) for continuous variables. Continuous data were compared where a Mullins sheath was used and where a Mullins sheath was not, using Wilcoxon rank-sum tests. Categorical variables were compared using Chi-square tests or Fisher’s exact test when expected cell counts were <5.

A total of 98 patients underwent attempted ASD closure during the study period (70 patients from the Gore ASSURED Clinical Study-Pivotal and Continued Access studies) and 28 patients after the study period. Of these patients, 52 underwent attempted secundum ASD closure using a long Mullins sheath (Supplemental Figures S1 and S2), while 46 patients underwent attempted secundum ASD closure with the Gore Cardioform delivery catheter through a short sheath in standard fashion.

Baseline demographic data and procedural characteristics of both groups can be seen in Table 1. The device size and maximum sheath size are summarized in Supplemental Table S1. In a univariate analysis, ASD size determined by stop-flow technique and correspondingly the device size and the maximum sheath size were significantly larger in the Mullins sheath group. The procedure/fluoroscopy times were also significantly longer in cases where a Mullins sheath was used.

There were 5 complications in the Mullins sheath group, 2 major (atrial fibrillation requiring cardioversion) and 3 minor (transient atrial tachycardia in 2 patients and groin bleeding the next day in 1 patient). There was 1 major complication (device embolization) and 1 minor complication (transient atrial tachycardia) when a Mullins sheath was not used. At a median follow-up of 43 days (IQR, 1-374 days) in the Mullins sheath group and 185 days (IQR, 31-364 days) in patients in whom a Mullins sheath was not used, all patients are alive with no more than a mild residual shunt in any patient. Of the 6 patients who did not have successful closure with the Mullins-sheath assisted technique, 3 were referred for surgery, 2 underwent successful closure of their ASD with the Gore Cardioform ASD occluder using a Hausdorf sheath (Cook Medical), and 1 defect was closed with an Amplatzer septal occluder device (Abbott). The 2 patients who had unsuccessful closure when a Mullins sheath was not used underwent successful surgical closure.

The Gore Cardioform ASD occluder has become commercially available in the United States and recently received a CE Mark in Europe in October, 2019. With these approvals, the device has been introduced to multiple institutions in the United States and Europe. The device has been received with enthusiasm, as it offers a new option for closure of secundum ASDs, particularly in large defects and defects with deficient rims. However, many implanters have faced significant challenges with aligning the device parallel to the atrial septum using the delivery system in which it is preloaded. In this report, we describe a technique that may help overcome some of these alignment challenges in select patients. We found that when the Mullins sheath was used, successful device deployment and proper alignment of the device was achieved in 88% of patients with a good clinical result in all patients who underwent successful ASD closure. As expected, the ASD size, sheath size, and device size were larger in the Mullins group. This reflects the selection bias as the Mullins sheath was used by the primary operator in cases with larger ASDs and with deficient rims. The longer procedure and fluoroscopy times in the Mullins group are likely secondary to operator learning curve of a new technique as well as the tendency to close larger and more complex ASDs with the Mullins-sheath assisted procedure. In our initial experience, we attempted closure of large and challenging secundum ASDs with the Gore Cardioform ASD occluder delivery catheter first; however, this resulted in a number of attempts and devices that had to be used before ultimately using a Mullins sheath for delivery of the device. Based on our experience thus far, we now utilize the Mullins-sheath assisted technique as a first-line method of device deployment for large or challenging secundum ASDs, particularly those with deficient rims.

Use of the Gore Cardioform ASD occluder delivery catheter may be adequate for small-to-moderate secundum ASDs or large secundum ASDs with good rims. In these instances, the relatively perpendicular orientation of the device before locking is acceptable as long as all rims have been captured, since the device will shift significantly after locking, in a favorable manner and in a parallel orientation to the defect. This is in contrast to when the Gore Cardioform ASD occluder delivery catheter is used on its own for large defects or defects with deficient rims. In such cases, the Gore Cardioform ASD occluder delivery catheter may align remote from the atrial septum due to its relatively straight trajectory, rendering the device perpendicular to the ASD.

The Mullins sheath provides an exaggerated curve extending to the tip of the sheath. This curve straightens out with the introduction of many implements, eg, balloons and stents, particularly if confined within a vessel or relatively straight trajectory. In our experience, when positioned in a large chamber, eg, the left atrium, the Mullins sheath maintains enough of a curve at its distal tip, even with the introduction of the Gore Cardioform ASD occluder delivery catheter to accentuate the curve of Gore Cardioform ASD occluder delivery catheter to align the device parallel to the atrial septum.

Although we did use a 75 cm-long Mullins sheath in a number of cases, we generally prefer to use a 63 cm-long sheath Mullins sheath. The 63 cm-long Mullins sheath can be withdrawn to the inferior vena cava or right atrium to allow further rotation/manipulation of the Gore Cardioform ASD occluder delivery system if needed, particularly if the defect is large and there is a deficient aortic or superior vena cava rim. If needed, the tip of the Gore Cardioform ASD occluder delivery system can be hand shaped further to exaggerate its curve as well; however, care should be taken to do this gently, as the mandrel and delivery system can be damaged if a kink develops in the system.

A number of techniques have been described as modifications for delivery of large ASD devices with challenging anatomy and deficient rims, mainly when using the ASO device. These include deployment via the right upper pulmonary vein,⁴ balloon-assisted delivery,⁵ use of a Hausdorf sheath, and use of a straight side-hole sheath.⁶ The use of a straight side-hole sheath (Mullins sheath tip cut on a bias) is a useful technique for delivery of the ASO device parallel to the atrial septum. However, this is facilitated by advancement of the device with a flexible cable, as opposed to a relatively stiff delivery catheter, such as used for delivery of the Gore Cardioform ASD occluder device. For this reason, in our experience, use of a Mullins sheath that has not been modified may be a better option for aligning the Gore Cardioform ASD occluder.

Study limitations. This study is limited by its retrospective description of a technique and lack of a true control group. It is possible that some of the secundum ASDs closed with Gore Cardioform ASD occluder utilizing the Mullins-sheath facilitated delivery technique could have been closed without this technique as well, using standard or other techniques.

Closure of large secundum ASDs or secundum ASDs with challenging anatomy with Gore Cardioform ASD occluder may be facilitated utilizing the Mullins sheath delivery technique we describe. This technique may be useful until a specific delivery sheath is designed and available for delivery of the Gore Cardioform ASD occluder in such circumstances.