PICS~AICS


Author: John L Bass MD

How I do it: Tips, Tricks, and Techniques

A PICS Society education series

Perventricular/hybrid device closure of muscular VSD using the Amplatzer® muscVSOD

Introduction:

Muscular VSDs (muscVSD) are the most common type of congenital heart disease1. The majority are small and without hemodynamic significance, and frequently close spontaneously. Defects large enough to require closure must often be addressed in the first few months of life because of volume overload with symptoms of congestive heart failure. Surgical closure can be challenging because of difficulty identifying defect margins. Surgically placed patches may be sutured to trabeculations leaving residual shunts. When the defect is approached from the left ventricle where margins are easier to identify, reduced left ventricular function may result 2. The Amplatzer muscular ventricular septal occlude device (muscVSOD) (Figure 1) was developed to avoid these limitations. The muscVSOD has a woven nitinol frame allowing deployment through a sheath like a self-expanding stent. The device has a waist with a length of 7mm to extend the thickness of the ventricular septum, and a diameter to fill the defect. Retention discs on each end extend 4mm beyond the waist. Fabric baffles are sewn in the retention discs, and "stuffing" of the same material fills the waist to produce thrombosis and occlusion.

Anticipated challenges of the procedure:

  1. Determining the appropriate device size
  2. Location of entrance through right ventricular free wall
  3. Crossing the muscVSD from the RV
  4. Deploying the muscVSOD without complications

TIP 1. Planning and Preparation

  • Access-The spot to enter the right ventricular free wall is located opposite the muscVSD. Using transesophageal echocardiography, the indentation of the RV free wall by the surgeon can be adjusted with communication and visualization by the surgeon and echocardiographer (Figure 2) Alternatively, the location can be determined by directly imaging the muscVSD with epicardial echo on the RV free wall.

  • "Hybrid deployment of the muscVSO device requires the skill to manipulate the delivery equipment and device using echocardiographic imaging. Communication between the surgeon/cardiologist team and the echocardiographer at every step of the procedure and prior to deployment of the device is critical."—John Bass MD"

  • Defect Size-There must be at least 4mm between the muscVSD and cardiac valves. Choosing a device waist that fills the defect has the greatest chance of leaving no residual shunt. Measurements are made at end-diastole as the size of muscVSDs decreases in systole. Most muscular VSD's are oval3 and this geometry must be considered in choosing the device size. This is best done from 2-dimensional imaging of the VSD. Either separate images of muscVSD maximum and minimum diameters, or en face imaging can be done (Figure 3). The simplest way to choose the device size is to average the two diameters, although a diameter can be derived from the cross-sectional area or circumference. The device diameter is chosen 1-2mm larger than the end diastolic diameter of the muscVSD.
  • Tip 2. Tools needed

  • Long 18 Ga needle
  • Device: chosen 1-2mm larger than the calculated end-diastolic muscVSD diameter
  • A standard vascular sheath that accepts the device and can be flushed, with the packaged 0.035 in. J-wire for    vascular access
  • Delivery cable with microscrew from an Amplatz 45° TorqueVue delivery sheath.
  • Excellent echocardiographic imaging

  • Tip 3. How I do it

  • Hybrid/"Perventricular" implant of the device with an open chest is performed on the beating heart. The location for entering the RV is located opposite the muscVSD using echocardiography, observing the indentation of the RV free wall by the surgeon on TEE (Figure 2A), or locating it by directly imaging the muscVSD with epicardial echo. A purse string suture is placed at the predetermined site on the RV free wall, and a needle used to enter the RV cavity under echocardiographic observation. A J-wire is advanced through the needle across the VSD on echocardiography (Figure 2B). A J-tipped guidewire is easiest to maneuver across the muscVSD avoiding RV trabeculations and is less likely to perforate or injure the heart. Radiographic imaging is not needed. The short delivery sheath is positioned in the LV by advancing over the dilator and guidewire to the tip of the J-wire with careful TEE guidance (and pre-determination of the distance between the VSD and the LV free wall) to make sure the sheath and dilator does not get pushed too far as it can perforate the LV free wall (location confirmed by the change in the J tip shape, Figure 2C). The guidewire and dilator are removed leaving the sheath tip free in the LV, and the sheath flushed. The muscVSOD is attached to the microscrew of the delivery cable, withdrawn into a second sheath and flushed, and the sheath advanced through the hemostatic valve of the LV sheath. The device is advanced to the end of the delivery sheath in the LV on echocardiography.

  • The LV disc is uncovered by withdrawing the sheath and allowing the disc to expand (Video 1), being careful to avoid opening it within mitral chordae. It is easy to control the position of the disc relative to mitral chordae, the LV free wall and the ventricular septum by observing on echocardiography. The muscVSOD shortens as the LV disc expands. After the LV disc is expanded the waist of the device is uncovered by withdrawing the sheath allowing more control over device position. The device and sheath are withdrawn together until the LV disc reaches the IVS. The RV disc should then be deployed by slowly advancing the device out of the sheath, (Video 2) rather than withdrawing the sheath over the device as is usual with the Amplatzer atrial septal defect device. This gives the disc a chance to expand normally without entanglement in RV trabeculations or tricuspid valve chordae. The so called "cobra" deformity of the muscVSOD usually occurs when there is insufficient room for full expansion of the device, or when the disc is entangled in RV trabeculations or tricuspid valve apparatus. This deformity prevents complete expansion of the device with a greater chance of a residual shunt. With the RV disc incompletely expanded, embolization is also a possibility.

    1. Pitfalls to avoid

  • As mentioned above, target vessel diameter is important. Device instability has been found to be associated with larger target vessel diameters in patients with CHD (12.8 ± 1.0 mm vs. 8.6 ± 1.4 mm, p = .01)3.

  • While a long sheath is often helpful in navigating the tortuous course of an atrial switch anatomy, this can sometimes complicate the procedure if there are stents and pacing leads in the way. It is often feasible to deliver the sensor directly over the wire and through a short sheath using landmarks for the target vessel.

  • Tip 4. What complications to expect and how to deal with them

  • Perforation of the LV free wall can occur during hybrid placement of the device if the needle entering the heart is advanced too far without controlled observed on echocardiography (Figure 4).

  • Perforation of the LV free wall can also occur if the LV disc is advanced out of the sheath rather than being uncovered. The device extends 8-10 mm beyond the sheath before expanding (Figure 5), and forms a sharp spear that could puncture the LV free wall. If perforation of the LV is suspected, the posterior surface of the LV should be examined by the surgeon. TEE can also help identify the site of perforation. If perforation is suspected, it is best to inspect the LV free wall and interrogate the LV wall on TEE before withdrawing the dilator/sheath. Bleeding will also be noticed in the field after the catheter is withdrawn, and if significant, hemodynamic compromise may result.

  • Laceration of mitral chordae can occur when the LV disc is entangled in the chordae and pulled through them towards the IVS. This was experienced in feasibility animal trials, and has been reported in humans leading to mitral regurgitation. The LV disc should be opened freely in the LV cavity. If the LV disc does not move freely when withdrawn towards the IVS, entanglement in mitral chordae should be considered, and the device recaptured and the delivery sheath tip repositioned.

  • If a "cobra" deformity of the RV disc occurs (Figure 6), the RV disc should be recaptured and redeployed, even if partially within the muscVSD. The device can also be completely deployed in the LV cavity, the RV disc partially recaptured, and withdrawn with gentle tension on the delivery apparatus until it slides into the VSD, and then the RV disc is completely redeployed. Some surgeons have chosen to suture a deformed RV disc to the RV to prevent embolization.

  • Elimination of the volume load on the left ventricle by occlusion of a muscVSD is accompanied by an increase in afterload to the LV. Both this increase in afterload and a possible stenting of IVS contraction by a large device can produce a transient decrease in LV function. Short term support of the infant may be necessary after the procedure, but LV function should recover.

  • Summary:

    The Amplatzer® muscVSOD allows simple occlusion of muscVSDs off of cardiopulmonary bypass4,5. Placement using echocardiographic imaging is somewhat different from standard manipulation on fluoroscopy, but this technique has been learned with balloon atrial septostomy with transposition. A small amount of peri-device shunting may disappear over time. Careful attention to device selection and deployment of the device should allow safe and simple device closure of muscVSDs.

    References:
        1. Hoffman JIE, Kaplan S. The incidence of congenital heart disease. J Am Coll Cardiol 2002;39:1890-1900.
        2. Hanna B, Colan SD, Bridges SD, Mayer JE, Castaneda AR. Clinical and myocardial status after left ventriculotomy for ventricular septal defect. J Am Coll Cardiol. 1991;17(suppl):110A. Abstract
        3. Berry JM, Krabill KA, Pyles LA, Lohr J, Steinberger J, Bass JL. Muscular ventricular septal defect geometry and Amplatzer device closure: a new en face view. Circulation 1999 100:18(I):I-30. Abstract.
        4. Amin Z, Gu X, Berry JM, Bass JL, Titus JL,Urness M, Han Y-M, Amplatz K. New device for closure of muscular ventricular septal defects in a canine model. Circulation 1999;100:320-328.
        5. Gray RG, Menon SC, Johnson JT, Armstrong AK, Bingler MA, Breinholt KP, Kenny D, Lozier J, Murphy JJ, Sathanandam SK, Taggert N, Trucco SM, Goldstein. BH, Gordon BM. Acute and midterm results following perventricular device closure of ventricular septal defects: a multicenter PICES investigation. Cath Cardiovasc Interv 2017 90:281-289.

     

    Illustrations

     

    Figure 1

    Figure 1 Legend: The Amplatzer® muscular Ventricular Septal Occluder Device is a radially and longitudinally symmetrical device of woven Nitinol with fabric baffles in the discs and "stuffing" in the waist. The device is attached to a microscrew (A) allowing easy retrieval for repositioning before detachment. In B the device has been implanted in a midmuscular defect.

     

    Figure 2

    Figure 2 Legend: Transesophageal images during placement of the delivery sheath through the right ventricular free wall, arrow indicating the surgeon's finger indenting the RV wall (A), advancing a J-wire across the muscular VSD into the left ventricle, arrow indicating the tip of the J-wire (B) and advancing a delivery sheath to the end of the guidewire indicated by the arrow (C).

    Figure 3

    Figure 3 Legend: Enface imaging of a mid-muscular VSD. The defect has a vertical diameter of 6mm with a horizontal diameter of 13mm. An average of the two diameters is 9.5mm, the calculated diameter from the circumference assuming a circle is 11mm, and from the cross sectional area 9.5mm. A 10 or 12mm device would be appropriate.

    Video 1:

     

    Video 1: The LV disc of the device is deployed uncovering by withdrawing the sheath. Note that the device shortens pulling away from the LV free wall and mitral apparatus as it expands.

     

    Video 2:

    Video 2: Deployment of the right ventricular disc of the device achieved by pushing the disc out of the delivery sheath rather than uncovering it by withdrawing the sheath. This allows the disc to expand against the septum rather that catching RV trabeculations.

     

    Figure 4

    Figure 4 Legend: After advancing the needle through the RV free wall, a J-tipped guidewire was advanced but could not be seen in the RV or LV cavity. Arrows in A and B point to the guidewire tip outside the LV free wall in the pericardial space. After completion of device deployment, a small perforation was found in the LV free wall and sutured closed.

     

    Figure 5

    Figure 5 Legend: In A, a left ventricular angiogram is performed 3 months after device deployment. There was no inferior rim and a residual left to right shunt is present below the device. Opposite the device there is a pseudoaneurysm that resulted from perforation of the LV free wall during deployment. In B a muscular VSD device is pushed out of a sheath. The device extends 8-10mm beyond the sheath as a sharp "spear" before expanding. This could perforate the LV free wall if pushed out of the sheath. The device shortens as it expands so that uncovering the disc by withdrawing the sheath pulls it away from the LV wall and mitral apparatus.

    Figure 6

    Figure 6 Legend: In A, a muscVSO device has been implanted with a "cobra" deformity of the RV disc, probably caught in RV trabeculations in the apex. Within 24 hours, the RV disc had reformed within the beating heart. Comparing the device in A with images from a heart catheterization 2 months later (B), it is clear that the waist does not completely expand when the RV disc is constrained, and residual shunts are more likely.