Trans-catheter closure of large peri-membranous ventricular septal defect with multiple exits using two occluder devices

ABSTRACT

Aneurysmal peri-membranous ventricular septal defects (PM-VSD) may show multiple right ventricular exits separated by tricuspid chordae or trabeculations. Transcatheter PM-VSD closure is increasingly performed as a surgical alternative. An 8-year-old girl with a large PM-VSD, with moderate pulmonary hypertension, presented with dyspnea and recurrent respiratory infections. The defect measured 12 mm with two exits. Closure was successfully achieved using two devices through an antegrade approach with two arteriovenous loops. There was no residual shunt, aortic regurgitation, or conduction block on follow-up. Transcatheter closure of PM-VSDs with multiple exits using two devices is feasible and safe in selected patients.

INTRODUCTION

Peri-membranous ventricular septal defect (PM-VSD) accounts for approximately 70%–80% of all ventricular septal defects.[1] Membranous septal aneurysms or tricuspid chordal attachments around the defect may form multiple right ventricular (RV) exits.[2] Surgery is the standard of care for PM-VSD with multiple exits since 1954.[3] We report closure of a PM-VSD with multiple exits using two devices deployed antegradely from femoral venous access. This adds to the reports describing the use of multiple devices in PM-VSD closure.[4,5,9]

CASE REPORT

An 8-year-old girl (13 kg) with a history of recurrent respiratory tract infections in childhood and recent worsening breathlessness was diagnosed to have a large PM-VSD with moderate pulmonary arterial hypertension. Transthoracic and transesophageal echocardiography revealed a large 12 PM-VSD with two exits and left ventricular (LV) volume overload. There was no aortic regurgitation, and the estimated peak pulmonary artery systolic pressure was 55 mm-Hg. The heart team decided for a device closure after informed consent, as the parents refused open heart surgery. The procedure was done under conscious sedation. Following right femoral arterial (6F) and venous (7F) access, 100 units/kg heparin was administered. LV angiogram in left anterior oblique view with cranial tilt, showed a large 13.8 mm PM-VSD with two RV exits measuring 7 and 6 mm [Figure 1a, b and Video 1]. Oximetry data revealed a shunt ratio of 2.2:1. Invasive pulmonary artery pressure was 45/30 mmHg. As the distance between the two exit ports were <5 mm apart, closure was attempted with a single device. After retrograde VSD cannulation from the arterial side using a Judkin’s right coronary catheter, a 260 cm, 0.035” Glide-wire (Terumo Medical Corporation, Japan), was snared from the venous side, forming an arteriovenous (AV) loop. A 9F delivery sheath was advanced into the descending thoracic aorta over this AV loop. Deployment of a large 16/18 mm Cera duct occluder (Lifetech, Shenzhen, China) led to a significant residual shunt and development of Grade II aortic regurgitation, that prompted withdrawal of the device [Figure 2 and Video 2]. The plan was revised to close the two exits of the PM-VSD separately using two occluders.

Figure 1.

(a) Apical view on echocardiogram shows a large peri-membranous ventricular septal defect (PM-VSD) with left to right shunt, (b) Left ventriculogram shows a large PM-VSD with two exits measuring 7 and 6 mm

Figure 2.

Left ventriculogram after deployment of a 16/18 mm duct occluder shows significant residual shunt (red arrow)

A new AV loop was created to advance a 7Fr delivery sheath into the descending thoracic aorta. LV angiogram was repeated to identify the second exit, through which the second AV loop was formed. As advancing the second delivery sheath was challenging, the Glide-wire was exchanged for a super-stiff wire, which was snared from the arterial side to facilitate the passage of another 7 Fr delivery sheath [Figure 3]. A 10/12 mm Cera duct occluder was deployed across the inferior defect. A 10/12 mm Konar MFO (Lifetech, Shenzhen, PRC) was deployed across the superior defect that was close to the aortic valve. Final LV angiography showed negligible residual shunt without aortic regurgitation [Figure 4 and Videos 3, 4]. There was no electrocardiographic conduction disturbance at 6-month follow-up. There was no residual flow new aortic or tricuspid regurgitation [Figure 5 and Video 5].

Figure 3.

Left ventriculogram after formation of the arteriovenous loop through the first exit delineates the second exit

Figure 4.

Final left ventriculogram after deployment of both devices shows no residual shunt

Figure 5.

Colour Doppler echocardiography from apical view after 6 months shows complete closure of the shunt

DISCUSSION

Closure of PM-VSD with multiple exits with a single large device may succeed if the defects are within 5 mm from each other. However, closure of the far-spaced defects would need multiple devices.[4,5,6,7,8,9] A single device did not close the defect in our case, despite the exits being fluoroscopically <5 mm apart and in the absence of a septal aneurysm. While a critical assessment of complex PM VSD is essential for successful transcatheter closure, color Doppler turbulence across PM-VSD may fail to delineate the multiple exits in transthoracic echocardiography. Understanding the anatomy on LV angiography is equally essential for a successful intervention. Two principal concerns in PM-VSD device closure are conduction disturbances and aortic insufficiency that warrant a long-term electrocardiographic and echocardiographic surveillance. Closing PM VSD with multiple exits using a single device risks residual shunt and aortic valve interference due to the selection of a large device. While double-disc occluders offer an improved seal of multiple exits, duct occluder don’t crowd the RV inflow with large discs. In our case, a double-disc KONAR-MFO was chosen for the defect (close to the aortic valve) because of its low profile to avoid aortic valve interference. A duct occluder was chosen for the second exit to reduce metal burden in the RV inflow. Transcatheter closure of multiexit PM-VSD using two occluders needs planning in the selection of devices and execution of the procedure.

Declaration of patient consent

The authors certify that they have obtained all appropriate patient consent forms. In the form, the patient(s) has/have given his/her/their consent for his/her/their images and other clinical information to be reported in the journal. The patients understand that their names and initials will not be published and due efforts will be made to conceal their identity, but anonymity cannot be guaranteed.

Conflicts of interest

There are no conflicts of interest.

Funding Statement

Nil.