SEARCH

SEARCH BY CITATION

Keywords:

  • pmVSD;
  • transcatheter;
  • Amplatzer® membranous VSD occluder 2

Abstract

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. METHODS
  5. STATISTICAL ANALYSIS
  6. RESULTS
  7. DISCUSSION
  8. LIMITATIONS
  9. CONCLUSION
  10. ACKNOWLEDGEMENTS
  11. REFERENCES

Objectives

To describe the initial world experience and mid-term follow-up of perimembranous ventricular septal defect (pmVSD) closure with a newly designed occluder.

Background

Transcatheter closure of pmVSDs has been associated with a substantial risk of complete heart block, prompting many centers to abandon this intervention.

Methods

A prospective multicenter cohort study was conducted on patients with pmVSD undergoing catheter closure using the Amplatzer® Membranous VSD Occluder 2 in the initial 4 pilot centers.

Results

Nineteen patients, median age 6 years (range 1.4–62 years), were enrolled and followed for 14 ± 3 months (range 8–20 months). The median weight was 26 kg (range 9.3–96 kg) and the mean Qp/Qs ratio was 1.8 ± 0.7. The defect on left ventricular side measured 9.9 ± 3.5 mm and the orifice on right ventricular side was 8.1 ± 2.8 mm by echocardiography. Mean device size was 9.4 ± 2.4 mm (range 5–14 mm). An eccentric device was employed in 9 patients (47%) and a concentric device in 10 (53%). Overall, 18 patients (95%) had successful device implants. Procedural time was 122 ± 39 min. There were no procedural complications. Mild residual shunting was initially observed in 14 (78%) patients. At last follow-up, mild residual shunting persisted in only 3 (17%) patients. There was no significant increase in aortic or tricuspid regurgitation. No patient had any degree of AV block, although one developed a transient left anterior fascicular block. Holter evaluation, obtained in all patients, was unremarkable in all.

Conclusions

This early cohort experience using a novel adapted transcatheter closure device for pmVSD suggests that the procedure is feasible, safe, and effective. © 2013 Wiley Periodicals, Inc.


INTRODUCTION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. METHODS
  5. STATISTICAL ANALYSIS
  6. RESULTS
  7. DISCUSSION
  8. LIMITATIONS
  9. CONCLUSION
  10. ACKNOWLEDGEMENTS
  11. REFERENCES

Ventricular septal defect (VSD) is a common congenital heart abnormality [1]. Up to 80% of clinically significant VSDs, so-called perimembranous VSDs (pmVSDs) are located in the region between the tricuspid and aortic annulus [2]. Transcatheter VSD closure was introduced in 1988 as an alternative to open heart surgery that is less invasive, avoids extracorporeal circulatory support and surgical scars, and offers a faster recovery [3]. Since the early 2000's, the most widely used transcatheter VSD closure device was the Amplatzer Membranous VSD Occluder (AGA Medical Corporation, MN). Reported complete closure rates approached 100% at 6–12 months, with rare cases of progressive aortic or tricuspid insufficiency [4-7]. However, anatomic proximity to the cardiac conduction system led to an increased risk of complete atrio-ventricular block (cAVB), estimated to be ∼5% in most series [6-8], and up to 22% in one [8]. Many centers abandoned the technique due to this complication [9].

A new Amplatzer device was recently adapted for pmVSD closure (Amplatzer Membranous VSD Occluder 2, AGA Medical Corporation, St-Jude, MN). It was redesigned to prevent conduction abnormalities. Features of this device include a 75% reduction in radial force, 45% reduction in clamping force, and increased stability [10]. An excellent safety profile was demonstrated in preclinical studies [10]. Favorable results were obtained in the first human experience in one child and one adult [11]. Herein, we report results of the early international experience on the first 19 patients who underwent percutaneous device closure of a pmVSD using the Amplatzer Membranous VSD Occluder 2.

METHODS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. METHODS
  5. STATISTICAL ANALYSIS
  6. RESULTS
  7. DISCUSSION
  8. LIMITATIONS
  9. CONCLUSION
  10. ACKNOWLEDGEMENTS
  11. REFERENCES

Patient Selection

From June 2011 to February 2012, 19 patients with a pmVSD were prospectively enrolled in an early cohort study designed to assess the Amplatzer Membranous VSD Occluder 2, and were followed for 14 ± 3 months (range 8–20 months) postprocedure. Patients were recruited from the initial 4 centers performing the procedure worldwide, all of which had prior experience with percutaneous pmVSD closure. Written parental or patient informed consent was obtained in all cases.

All patients underwent standard trans-thoracic echocardiography (TTE) screening. Indications for pmVSD closure were left chamber volume overload, history of bacterial endocarditis, symptoms of heart failure, significant left-to-right shunt (Qp/Qs ≥ 2), and/or pulmonary hypertension. Patients were required to have a documented left-to-right shunt, calculated pulmonary vascular resistance <2/3 systemic vascular resistance, and to weigh >9 kg. Patients were excluded if they had any of the following criteria: (i) predominant right-to-left shunt, (ii) more than mild aortic regurgitation (AR), (iii) more than moderate tricuspid regurgitation (TR), (iv) severe aortic valve prolapse, (v) associated congenital defects requiring surgery, or (vi) calculated pulmonary vascular resistance ≥2/3 calculated systemic vascular resistance. All patients had a detailed physical examination, chest X-ray, and standard 12-lead electrocardiogram (ECG). Selected patients underwent pre- and postprocedural electrophysiology studies (EP), according to the availability of resources.

Procedure and Device

The procedure was performed under general anesthesia, with trans-esophageal echocardiographic (TEE) guidance. After obtaining arterial and venous access, intravenous antibiotics and heparin (75–100 IU/kg) were administered. The diameter of the VSD [left ventricular (LV) side and right ventricular (RV) orifice] was assessed by echocardiography (two-dimensional and Doppler color flow) in a short axis aortic view (35°–45° angulation) and long axis LV outflow 3-chamber view (110°–120° angulation). When more than one orifice was present through a so-called “aneurysm” (accessory tricuspid tissue in most instances), the summation of diameters was used. Angiography of the LV [60° left anterior oblique (LAO) − 20° cranial view] was performed using a 5 or 6F marker pigtail catheter. Diameters of the VSD defect were measured on the LV side and RV orifice (though the “aneurysm,” if present). The device size was selected based on TEE and angiography and, in the absence of an “aneurysm” of the LV side, it was oversized by 1–2 mm. In case that an “aneurysm” of the LV side was present, the device was undersized in relation to the LV defect, while remaining larger than the RV orifice.

The device design and technique of deployment have been previously described [11]. In brief, the device's left disc has an elliptical and concave shape (Fig. 1) that adapts to the LV outflow tract (LVOT) and provides improved retention and stability. It is available in two configurations: eccentric, with a 1 mm superior rim and a 3 mm inferior rim and concentric, with 3 mm superior and inferior rims. An eccentric device is used when the distance between the defect and the aortic valve is <3 mm. The nitinol wire is considerably thinner, which decreases the rigidity of the device, and is in a dual layer configuration. The external layer is especially thin, imparting minimal radial pressure, while the inner portion provides stability. To reduce clamp force against the ventricular septum, the waist length is also increased from 1.5 to 3 mm. Polyester patches are sewn into the disks, to ensure rapid occlusion. Both versions of the device (eccentric and concentric) are available in 9 waist diameters (from 4 to 10 in 1 mm increments plus 12 and 14 mm). Finally, the new TorqVue 4 delivery system was redesigned to facilitate better positioning in the LVOT. The deployment technique is similar to the technique reported with the original device. The defect is crossed from the LV, usually using a JR4 catheter. After obtaining a femoral arterio–venous guidewire loop, the delivery sheath is advanced to the ascending aorta and pushed back to the LV. The device is loaded into the pushing catheter and delivery sheath. It is then advanced to the tip of the sheath, which is positioned no more than 1–2 cm inside the LV. The left disk is deployed by withdrawing the sheath. Appropriate position of the device's platinum marker is verified, ensuring that it pointed towards the LV apex in a 30° RAO view and in a 60° LAO −20° cranial view. In such a position, the elliptical shape of the LV disc should be seated perpendicular to the axis of the LVOT. Importantly, if the platinum marker is not in optimal position, the device can be completely exteriorized in the LV while remaining attached to the pusher catheter and wire. The correct orientation may then be achieved by rotating the pusher catheter clockwise or counter clockwise, followed by recapturing the RV disk into the sheath. After confirmation of optimal positioning, the left disk is pulled against the septum and the right disk is deployed in the RV. Depending on the exact anatomy of the pmVSD and operator preferences, the device can be centred on the LV defect (Fig. 2) or on the orifice of the “aneurysm,” when present (Fig. 3). After verification of adequate positioning, the device is released.

image

Figure 1. Picture and fluoroscopic image of the first in human implanted device. Note the elliptical and concave shape of left disk and dual nitinol layer, with thinner external layer (arrow). [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]

Download figure to PowerPoint

image

Figure 2. Perimembranous VSD with small aneurysmal pouch in an 18 kg patient. (a) TEE preimplantation showing a large defect (11 mm), loosely covered by accessory tricuspid tissue. Note the absence of rim to the AoV. (b) LV angiography preimplantation, same findings as in (a). (c) LV angiography postimplantation of an eccentric device (12 mm) with a mild residual shunt (arrow). (d) TEE postimplantation. Note the close relationship of the device with the AoV, with trivial aortic regurgitation (arrow). (e) TTE one-week postimplantation demonstrating a mild residual shunt (arrow). (f) TTE 6 months postimplantation with disappearance of shunt and unchanged minimal aortic regurgitation (arrow). AoV, aortic valve; LV, left ventricle; TEE, transesophageal echocardiography; TTE, transthoracic echocardiography. [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]

Download figure to PowerPoint

image

Figure 3. Perimembranous VSD with multifenestrated aneurysmal pouch in a 9.4 kg patient. (a) LV angiogram pre: Large defect on LV side (10.5 mm), partially covered by accessory tissue, with residual orifices on RV side (arrows). (b) LV angiogram postdeployment of an eccentric device (10 mm), undersized in relation to LV defect. Note the LV disk deployed inside aneurysm and a residual shunt superior to it (arrow). (c) TTE (short axis) 3 months postprocedure showing a small residual shunt (arrow). LV, left ventricle; RV, right ventricle; TTE, transthoracic echocardiography. [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary. com.]

Download figure to PowerPoint

Procedural success was determined by device delivery in an appropriate and stable position, without persistent conduction abnormalities or an acute increase in AR or TR. In 4 patients, EP studies with programmed atrial and ventricular stimulation were performed to assess properties of the AV node and His-Purkinje conduction system (i.e., baseline intervals, Wenckebach cycle length, antegrade, and retrograde AV nodal effective refractory periods) prior to and following device implantation.

Patients were discharged 1–3 days after the intervention and were prescribed aspirin 2–5 mg/kg/day (children) or 80 mg/day (adults). Therapy was discontinued after 6 months if no significant residual shunt was present.

Follow-Up

All patients were systemically followed with a physical examination, chest X-ray, standard TTE, and 12-lead ECG one day, and serially during the first year after the procedure. Exact timing of the follow-ups varied from center to center and according to families' availability, but included at least a visit around the first month, from 3 to 6 months at around 1 year postimplantation. The TTE included a detailed assessment of changes in AR, TR, and residual VSD shunting. Grading of severity of residual shunts was based on color-flow mapping and was defined as mild (0–2 mm color Doppler jet or foaming through device), moderate (2–4 mm), or severe (>4 mm). ECG parameters included the PR interval, QRS duration, and QRS axis. Holter monitoring (24 hr) was performed at least 30 days after the procedure, or if deemed clinically indicated.

STATISTICAL ANALYSIS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. METHODS
  5. STATISTICAL ANALYSIS
  6. RESULTS
  7. DISCUSSION
  8. LIMITATIONS
  9. CONCLUSION
  10. ACKNOWLEDGEMENTS
  11. REFERENCES

Continuous variables are presented as mean ± standard deviation or median (range), depending on whether or not they were normally distributed. Categorical variables are presented as frequencies and percentages. For intrapatient comparisons of continuous or ordinal variables (e.g., shunt grade, TR, and AR) between two time points, a paired sample t test or Wilcoxon signed rank test for related samples was used for normally distributed or skewed data, respectively. A constant difference between values was assumed for ordinal variables. A one-way ANOVA test was used for comparisons between continuous variables at three time points. A two-sided P-value <0.05, with Bonferroni adjusted alpha levels, was considered statistically significant. All statistical analyses were performed using the Statistical Package for Social Sciences, version 17.0 software (SPSS, Chicago, IL).

RESULTS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. METHODS
  5. STATISTICAL ANALYSIS
  6. RESULTS
  7. DISCUSSION
  8. LIMITATIONS
  9. CONCLUSION
  10. ACKNOWLEDGEMENTS
  11. REFERENCES

Baseline Characteristics

A total of 19 patients were enrolled in the study between June 2011 and February 2012, and followed serially for 1 year. Baseline characteristics are summarized in Table 1. One patient had atrial fibrillation and another had a permanent pacemaker for a postsurgical cAVB. Eight patients (42%) had a concomitant nonsignificant congenital cardiac abnormality. Three patients had trisomy 21. Two patients had residual VSDs after surgery and one patient after previous device implantation (residual jet from left ventricle to right atrium, Fig. 4). In four cases, the perimembranous defect was extending to the inlet portion of the septum. In one patient, a moderate right coronary cusp prolapse was present, with trace aortic insufficiency (Fig. 5). The indication for closure was left chamber overload in 16 patients (84%) and prior endocarditis in the remaining three.

Table 1. Baseline Characteristics
N= 19 
  1. Variables are expressed as mean ± SD (range) or n (%).

Male gender8 (42)
Age (yr)17.3 ± 20.6 (1.4–62)
Weight (kg)40 ± 27 (9.3–96)
Height (cm)128 ± 30 (80–176)
Atrial fibrillation1 (5)
Permanent pacemaker1 (5)
Concomitant congenital cardiac abnormalities8 (42)
Indication for closure 
Left chamber dilatation16 (84)
Endocarditis3 (16)
image

Figure 4. Perimembranous VSD with aneurysmal pouch and a previous generation Amplatzer pmVSD occluder in place and residual LV to RA shunt. (a) TTE and (b) TEE preimplantation, showing the previous device (arrow) and residual LV to RA shunt. (c) Angiography preimplantation with residual shunt superior to the previous device (arrow). (d) Angiography postrelease of the second device (concentric 8 mm, arrow) with no residual shunt. TTE, transthoracic echocardiography; TEE, transesophageal echocardiography; LV, left ventricle; RA, right atrium. [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]

Download figure to PowerPoint

image

Figure 5. Perimembranous VSD with conal extension and mild prolapse of right aortic cusp in an 80 kg patient. (a) TEE preimplantation, LVOT long axis. Note the close proximity of defect (measuring 8 mm) to the aortic valve and the prolapse of right cusp (arrow). (b) TEE preimplantation, LVOT short axis. Note the extension of shunt to the conal septum (arrow). (c) TEE postimplantation, LVOT long axis and 4 chambers. Note the device (eccentric, 10 mm) straddling the aortic prolapse (arrows). (d) Aortogram postimplantation (same patient). Straddling of the aortic cusp (arrow) does not cause aortic regurgitation. (e) TEE postimplantation. Note the foaming through the device and trivial aortic regurgitation, unchanged from preimplantation. TEE, transesophageal echocardiography; LVOT, left ventricular outflow tract. [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]

Download figure to PowerPoint

Procedural Characteristics

Procedure related variables are shown in Table 2. The defect size on the LV side measured 9.3 ± 3.3 mm by angiography and 9.9 ± 3.5 mm by echocardiography. The orifice on the RV side (total of all orifices if multiple) was 8.1 ± 2.8 mm by echocardiography. An “aneurysm” (accessory tricuspid tissue in most cases) of the interventricular septum was present in 10 patients (53%). The Qp/Qs ratio was 1.8 ± 0.7 overall and 2.0 ± 0.6 for patients with left chamber overload as the main indication for closure. The average procedural time was 122 ± 39 min, with a mean device size of 9.4 ± 2.4 mm (range 5–14 mm). An eccentric device was used in 9 patients (47%) and a concentric device in 10 (53%).

Table 2. Procedure Related Variables
N = 19 
  1. LV, left ventricle; RV, right ventricle.

  2. Variables are expressed as mean ± SD (range) or n (%).

  3. a

    Transient nodal rhythm in one patient (full recovery in minutes).

Defect size (mm) 
Angiography–LV side9.3 ± 3.3 (4.0–14.8)
Echocardiography–LV side9.6 ± 3.3 (4.5–16.0)
Echocardiography–RV side8.1 ± 2.8 (3.9–14.0)
Aneurysm of inter-ventricular septum10 (53)
Qp/Qs1.8 ± 0.7
Indication: LV volume overload2.0 ± 0.6
Indication: Endocarditis1.1 ± 0.1
Device size (mm)9.4 ± 2.4 (5.0–14.0)
Device type 
Eccentric9 (47)
Concentric10 (53)
Procedure time, sheath in-out (min)122 ± 39 (60–207)
Fluoroscopy time (min)35 ± 16 (16–69)
Procedural success18 (95)
On first attempt17 (94)
On second attempt1 (6)
Use of >1 device0 (0)
Procedural complicationsa1 (6)

The device was successfully implanted in 18 patients (95%) and no procedural complications occurred. Device implantation was unsuccessful in a two-year-old with a 14.5 mm defect with no aneurysm or aortic rim. Despite attempts using the largest eccentric device (14 mm), a stable position was not achievable, as the device repeatedly prolapsed through the defect to the RV. A transient nodal rhythm was observed when crossing the defect with the delivery system during the first attempt, with full recovery within minutes. The procedure was eventually aborted and the patient underwent surgical pmVSD closure weeks later, with an uneventful postoperative course.

Changes in TR and AR

Immediately postimplantation, the degree of TR increased slightly (Fig. 6, P < 0.05). However, over a follow-up of 14 ± 3 months, the degree of TR improved, with no patient having more than mild TR. Similarly, immediately following device implantation, AR increased nonsignificantly (from none or trivial to mild in 3 patients). At last follow-up, the degree of AR was similar to baseline.

image

Figure 6. Changes in residual shunt, AR, and TR based on echocardiography. The duration of follow-up was 14 ± 3 months. AR, aortic regurgitation; TR, tricuspid regurgitation. [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]

Download figure to PowerPoint

Residual Shunt

Immediately postimplantation, mild, and moderate residual shunts were observed in 11 (61%) and 3 patients (17%), respectively (Fig. 6). At last follow-up, there was a significant improvement in the degree of residual shunting (P < 0.001). The only 3 patients (17%) with residual shunting, all of whom had mild shunts, had multiple orifices through an aneurysm, with the device positioned inside it (Fig. 3) rather than on the true LV defect (Fig. 2).

Conduction Abnormalities

No significant changes in PR and QRS duration were observed at 1 day and at last follow-up (Table 3). All 18 patients had at least one Holter monitor (last Holter 9 ± 6 months postprocedure, range 1–17 months). No Holter showed any degree of AVB at any time, except for the patient with postsurgical AVB. In one patient, 2 consecutive Holter monitors (1 month and 16 months postprocedure) showed numerous monomorphic premature ventricular beats, which were not noted preprocedure, and for which the patient was asymptomatic. EP studies, performed in 4 pediatric patients pre- and postdeployment, revealed no acute changes in the properties of the AV node and His-Purkinje system. There were no new left or right bundle branch blocks. One patient, however, developed a left anterior fascicular block, noted one day after the procedure, which persisted for 6 months, before returning to normal at last follow-up (20 months postprocedure). This patient had a normal PR interval, a normal EP study (immediately postdeployment), and a normal 30-day and 6 month Holter test. Another patient (who had device implantation after a previous surgery) had junctional ectopic beats with pauses that prompted Holter monitoring and a complete EP study, including a procainamide challenge test. The AV node and His-Purkinje conduction system were entirely normal. Holter monitoring, 5 months after implantation, showed disappearance of junctional ectopic beats and normal AV conduction.

Table 3. ECG Changes at 1-Day and Last Follow-Up
 Baseline1-day follow-upLast follow-upP-value
  1. Variables are expressed as mean ± SD.

  2. a

    N = 16 (one patient with AF and one with a permanent pacemaker were excluded).

  3. b

    N = 17 (one patient with a permanent pacemaker was excluded).

PR interval (msec)a147 ± 25140 ± 28146 ± 23NS
QRS interval (msec)b87 ± 1785 ± 3093 ± 21NS
QRS axis (°)b19 ± 4817 ± 4112 ± 33NS

DISCUSSION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. METHODS
  5. STATISTICAL ANALYSIS
  6. RESULTS
  7. DISCUSSION
  8. LIMITATIONS
  9. CONCLUSION
  10. ACKNOWLEDGEMENTS
  11. REFERENCES

Our initial multicenter experience suggests that transcatheter pmVSD closure using the Amplatzer pmVSD Occluder 2 is feasible, safe, and effective. A wide range of anatomical variants (pmVSD with and without aneurysm, inlet extension, conal extension with aortic cusp prolapse, residual shunt after surgical or device closure, LV to RA shunt) was attempted, with generally positive results (Figs. 2-5). Notwithstanding the relatively small number of patients, no major conduction abnormality or other complication was observed on short and mid-term follow-up. The procedural success rate was comparable to previous reports using other devices [5-7, 12, 13]. The one case of failed implantation was secondary to the insufficient size of the largest available device (14 mm). This may have been overcome with a larger device (e.g., 16 mm, eccentric). With currently available device sizes, the maximal size of a VSD that may be attempted for closure appears to be around 13 mm (on LV size or orifice through aneurysm, if present). It may be reasonable; however, to defer development of larger devices until larger studies confirm the safety of current sizes, particularly in small patients.

The new Amplatzer® TorqVue4 delivery system functioned much like the previous version and was, in the majority of cases, easily placed in the LV apex. However, in the smallest patients, occasional difficulties were encountered when attempting to advance the sheath tip from the LVOT to LV apex, because of its stiffness and radius of curvature. The pusher catheter's easy maneuverability, with the ability to rotate the device inside the LV, facilitated proper alignment, and correct positioning on the first attempt in the vast majority of patients.

Although the device was modified to reduce radial and clamping forces, proximity of the perimembranous defect to the AV conduction system nevertheless carries risk of AV block. Mechanical injury to conduction fibers may occur at any step of the procedure (defect crossing, loop creation, delivery catheter insertion, device deployment, or recapture etc.). Therefore, careful manipulation of catheters and sheaths is required throughout. It may be speculated that device size selection is critical to preventing such a complication and that device oversizing may have been responsible for a substantial proportion of AV blocks reported with the previous version of the Amplatzer membranous VSD occluder. Nitinol is a superelastic alloy that expands with time to reach the original diameter of the device. While the new version is less traumatic, with less radial and clamping forces and a softer external layer of nitinol, sizing must nevertheless still be performed with great precision. In the current study, when no significant “aneurysm” was present (Fig. 2), the device size (which represents the diameter of its central waist) was selected to be 1–2 mm larger than the LV defect. When an “aneurysm” was present, the device was generally undersized in relation to the LV defect, while remaining larger than the RV orifice (or total of orifices if multiple holes, Fig. 3). This frequently resulted in the left disk of the device “flipping” partially or totally inside the aneurismal pouch, while its central waist aligned with the RV orifice. Although this position is probably less traumatic for the conduction system (smaller device and less direct contact with the conduction system), it may be associated with a higher incidence of residual shunts, especially in multifenestrated aneurysms. Further studies are required to define the optimal approach to deploying devices in the presence of an aneurysm. When a sufficient rim (>3 mm) was present between the upper edge of the pmVSD and aortic valve, a concentric device was selected with the rationale that it would provide greater stability while decreasing the probability of residual shunting. Introduction of 11 and 13 mm devices would, in our opinion, allow for a more precise sizing of device, in larger defects, thus decreasing oversizing.

The impact of device closure on aortic and tricuspid valve function is multifactorial. Considerations include the presence and length of an aortic rim and the relation between the defect and septal tricuspid leaflet. In a substantial proportion of patients, some degree of AR or TR was present before the procedure, secondary to the pmVSD jet. Elimination of this jet can lessen valve regurgitation, particularly TR. In our series, new onset AR and TR was trivial or mild in all cases, and tended to improve during follow-up. It appears, therefore, that the larger wings and oval concave shape of the left disc [10] do not significantly interfere with the aortic valve if the device is properly sized and optimally rotated during its deployment. In cases with no aortic rim, the eccentric device, with its 1 mm superior rim, did not interfere with aortic valve function, even in the one case where moderate prolapse of the right cusp was present (Fig. 5). Reasons as to why a small clinically insignificant increase in grade of TR grade was observed immediately after the procedure remain unknown. It may have reflected the effect of anesthesia, or patient volume status or, more probably, transient interference by the newly deployed device. In any event, at later follow-up, the grade of TR was globally unchanged compared to preprocedural levels and had improved in some patients.

The 77% incidence of early residual shunting is higher than reported with other series, which is probably because of our inclusion of patients with any degree of foaming. However, at last follow-up, the prevalence of residual shunts had decreased to 17% and was mild (color jet of 2 mm) in all. Part of this reduction is likely due to initial clotting of the polyester patches of the device, which eliminates foaming through it. It may, therefore, be reasonable to accept a small degree of initial residual shunting rather than upgrade to a larger device that may increase the risk of damage to the conduction system. Later reductions in residual shunting may be due to device reconfiguration and periprosthetic fibrosis. The 3 patients with residual shunts at last follow-up had multifenestrated aneurysmal pouches. In these cases, a concentric device smaller than the pmVSD diameter on the LV side was deliberately selected and positioned partly or totally inside the aneurysm (Fig. 3). In this series, 55% of devices were deployed partly or completely in this fashion, to minimize contact with the conduction system. In most cases, the observed initial shunt decreased progressively and eventually disappeared, which might also be observed in the 3 patients with residual shunt, with longer follow-up. If, at longer follow-up, such shunts tend to persist and, on the other hand, the more “anatomical” positioning of the device on the LV defects shows no interference with AV conduction, the latter might become the preferred deployment technique.

No AV block was observed in our series, after a 14 ± 3 month's follow-up. Two thirds of previously reported cAVB cases with the original device occurred within the first week [5-7, 12-17], but could develop as late as 1 year after the procedure [18]. The fact that all our patients have now reached this length of follow-up without any degree of AVB is obviously reassuring but longer follow-up and a much greater number of subjects is needed before claiming that this device is as safe as open heart surgery in this regard. Therefore, these initial favorable results should be interpreted with cautious optimism as experience with the device grows and follow-up duration is extended. In particular, two of our patients showed mild and self-improving alterations of their AV conduction physiology (left anterior fascicular block and junctional ectopic beats). In these two patients, the device was centered on the LV defect. This tends to demonstrate that direct contact of the device with the AV conduction system does elicit changes; however subtle, and that only a long period of follow-up can rule-out progressive inflammation or fibrosis of the latter. Another patient has experienced new-onset monomorphic ventricular ectopic beats, (noted at Holter 1 month and 16 months postprocedure), which were asymptomatic and not associated to sustained ventricular tachycardia on Holter. Although the patient rhythm was not assessed with a Holter monitor before procedure, complete review of the chart (physical examinations and nurses notes, multiple ECGs, records of transthoracic and transoesophageal echocardiograms, and Magnetic Resonance reports) did not show ectopic beats, making it very possible that the arrhythmia appeared after the procedure, and is possibly related to the device. In this adult patient, the device was placed inside a very large aneurismal pouch. Movements of the device inside the right ventricle in systole might explain this benign arrhythmia, which obviously will need to be followed.

In our opinion, extrapolation of this technique to very small patients (<9 kg) seems premature [11]. The delivery system might still be too stiff and the radius of curvature of the left disk may not be suitable to the LVOT anatomy of an infant, resulting in subaortic stenosis or valve interference. Smaller patients also appeared to be at higher risk of AV block with the first version of the device [8]. It would seem prudent to first extend safety data with the second generation of the device to a larger population before extending indications to infants. Minor modifications of the current device, such as increasing the concavity of the left disk, may be required for small patients.

As with every invasive procedure, the learning curve effect may influence procedural results and patient outcomes. The operators in this study have extensive experience with catheter interventions in congenital heart diseases and, more specifically, with implanting the first generation of this device. To obtain favorable outcomes, it would seem sensible to initially restrict use of this new version of the device to experienced manipulators committed to systematically gathering data within the context of a formal study protocol.

LIMITATIONS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. METHODS
  5. STATISTICAL ANALYSIS
  6. RESULTS
  7. DISCUSSION
  8. LIMITATIONS
  9. CONCLUSION
  10. ACKNOWLEDGEMENTS
  11. REFERENCES

The study is observational in nature and limited to a small number of patients and brief duration of follow-up.

CONCLUSION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. METHODS
  5. STATISTICAL ANALYSIS
  6. RESULTS
  7. DISCUSSION
  8. LIMITATIONS
  9. CONCLUSION
  10. ACKNOWLEDGEMENTS
  11. REFERENCES

We report the first series of patients to undergo transcatheter pmVSD closure using a second-generation occluder device specifically adapted to reduce complications, including damage to the adjacent conduction system. In this multicenter experience, device implantation was feasible, safe, and effective, with no more than mild residual shunting on follow-up. No patient experienced any degree of AV block or new onset complete bundle branch block at one-year follow-up. These initial favorable findings suggest that further studies should be pursued. Larger multicenter studies with even longer follow-up are required to validate and extend these initial results before widespread use.

ACKNOWLEDGEMENTS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. METHODS
  5. STATISTICAL ANALYSIS
  6. RESULTS
  7. DISCUSSION
  8. LIMITATIONS
  9. CONCLUSION
  10. ACKNOWLEDGEMENTS
  11. REFERENCES

The investigators wish to thank Professor Kurt Amplatz, Dr Xiaoping Gu, and Mr John Oslund for their precious advice and technical support during most of the interventions.

REFERENCES

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. METHODS
  5. STATISTICAL ANALYSIS
  6. RESULTS
  7. DISCUSSION
  8. LIMITATIONS
  9. CONCLUSION
  10. ACKNOWLEDGEMENTS
  11. REFERENCES
  • 1
    Hoffman JI, Kaplan S. The incidence of congenital heart disease. J Am Coll Cardiol 2002;39:18901900.
  • 2
    Jacobs JP, Burke RP, Quintessenza JA, Mavroudis C. Congenital heart surgery nomenclature and database project: Ventricular septal defect. Ann Thorac Surg 2000;69:S25S35.
  • 3
    Lock JE, Block PC, McKay RG, Baim DS, Keane JF. Transcatheter closure of ventricular septal defects. Circulation 1988;78:361368.
  • 4
    Hijazi ZM, Hakim F, Haweleh AA, et al. Catheter closure of perimembranous ventricular septal defects using the new Amplatzer membranous VSD occluder: Initial clinical experience. Catheter Cardiovasc Interv 2002;56:508315.
  • 5
    Fu YC, Bass J, Amin Z, et al. Transcatheter closure of perimembranous ventricular septal defects using the new Amplatzer membranous VSD occluder: results of the U.S. phase I trial. J Am Coll Cardiol 2006;47:319325.
  • 6
    Holzer R, de Giovanni J, Walsh KP, et al. Transcatheter closure of perimembranous ventricular septal defects using the amplatzer membranous VSD occluder: Immediate and midterm results of an international registry. Catheter Cardiovasc Interv 2006;68:620628.
  • 7
    Butera G, Carminati M, Chessa M, et al. Transcatheter closure of perimembranous ventricular septal defects: Early and long-term results. J Am Coll Cardiol 2007;50:11891195.
  • 8
    Predescu D, Chaturvedi RR, Friedberg MK, Benson LN, Ozawa A, Lee KJ. Complete heart block associated with device closure of perimembranous ventricular septal defects. J Thorac Cardiovasc Surg 2008;136:12231228.
  • 9
    Gatzoulis MA, Webb GD, Daubeneyet P. Diagnosis and Management of Adult Congenital Heart Disease, 2nd ed. Philadelphia: Elsevier; 2011. p 171.
  • 10
    Bass JL, Gruenstein D. Transcatheter closure of the perimembranous ventricular septal defect—Preclinical trial of a new Amplatzer device. Catheter Cardiovasc Interv 2012;79:11531160.
  • 11
    Velasco-Sanchez D, Tzikas A, Ibrahim R, Miró J. Transcatheter closure of perimembranous ventricular septal defects: Initial human experience with the Amplatzer® membranous VSD occluder 2. Catheter Cardiovasc Interv 2013;82:474–479.
  • 12
    Bass JL, Kalra GS, Arora R, et al. Initial human experience with the Amplatzer perimembranous ventricular septal occluder device. Catheter Cardiovasc Interv 2003;58:238245.
  • 13
    Zuo J, Xie J, Yi W, et al. Results of transcatheter closure of perimembranous ventricular septal defect. Am J Cardiol 2010;106:10341037.
  • 14
    Xunmin C, Shisen J, Jianbin G, Haidong W, Lijun W. Comparison of results and complications of surgical and Amplatzer device closure of perimembranous ventricular septal defects. Int J Cardiol 2007;120:2831.
  • 15
    Pedra CA, Pontes SC Jr, Pedra SR, et al. Percuta- neous closure of postoperative and past-traumatic ventricular septal defects. J Invasive Cardiol 2007;19:491495.
  • 16
    Zhou T, Shen XQ, Zhou SH, et al. Atrioventricular block: A serious complication in and after transcatheter closure of perimem- branous ventricular septal defects. Clin Cardiol 2008;31:368371.
  • 17
    Oses P, Hugues N, Dahdah N, et al. Treatment of isolated ventricular septal defects in children: Amplatzer versus surgical closure. Ann Thorac Surg 2010;90:15931598.
  • 18
    Carminati M, Butera G, Chessa M, Drago M, Negura D, Piazza L. Transcatheter closure of congenital ventricular septal defect with Amplatzer septal occluders. Am J Cardiol 2005;96:52L58L.