A Novel Approach for Transcatheter Management of Perimembranous Ventricular Septal Defect with a Subaortic Ridge

Introduction. Surgical closure of the perimembranous ventricular septal defect (PM VSD) and resection of the subaortic ridge are the standard methods of management, but there is no definitive agreement regarding the timing of surgery. Objectives. To evaluate the safety and efficacy of the management of patients with PM VSD and subaortic ridge with or without AR via transcatheter closure of the defect and compressing the ridge against the ventricular septum using Amplatzer ductal occluder type I (ADO-I). Patients and Methods. We introduced a new approach for transcatheter management of PM VSD and subaortic ridge by closing the VSD and capturing or compressing the ridge against the interventricular septum (IVS) using the ADO-I device. Thirty-eight (9.5%) of 398 patients with a PM VSD were found to have subaortic ridge and were enrolled in this study from August 1, 2014, to February 1, 2018, at the Ibn Albitar Center for Cardiac Surgery, Baghdad, Iraq. Results. The ages and weights of patients ranged from 1.5 to 25 years and 7 to 73 kg, respectively. The male-to-female ratio was 2.2 : 1. The VSD sizes ranged from 4 to 8 mm, and the median distance of the ridge from the proximal edge of the VSD was 2.5 mm. Prior to closure, 13 patients (34.2%) had mild and mild-to-moderate aortic regurgitation (AR), and nine patients (23.7%) had mild-to-moderate left ventricular outflow tract (LVOT) obstruction. The mean AR pressure half-time increased significantly after intervention (from 385 ± 38 ms to 535 ± 69 ms (significant 
 
 P
 
 value, 0.001)), and the mean of the peak pressure gradient across the LVOT decreased from 33 ± 7 mmHg to 15 ± 2.4 mmHg (significant 
 
 P
 
 value, 0.001). Successful procedures were achieved in 33 patients (86.8%). Conclusion. Transcatheter management of patients with PM VSD and subaortic ridges with or without AR is feasible and effective.


Introduction
Subvalvular aortic stenosis represents approximately 15% of left ventricular outflow tract (LVOT) stenosis [1,2]. Perimembranous ventricular septal defect (PMVSD) is found in about one-fifth of the cases of subvalvular aortic stenosis [1,3,4]. e subaortic ridge is defined as a ridge-like protrusion into the subaortic area from the crest of the ventricular septum that extended towards the mitral valve apparatus with or without Doppler evidence of obstruction [5,6]. Any fibrous or membranous echo-dense protrusion into the LVOT was considered by some authors as part of the spectrum of subaortic stenosis [6]. Subaortic ridge is best visualized in the parasternal long axis transthoracic echocardiographic view [1,4,6].
Subaortic ridge in patients with VSD is either caused by a fibrous, fibromuscular, or membranous ridge that is located at the lower border of the VSD or malalignment or deviation of the right ventricular outflow tract anteriorly or posteriorly [5,7]. Posterior deviation occurs when the outlet septum is displaced posteriorly into the LVOT resulting in narrow LVOT and obstruction above the VSD in the majority of patients. Alternatively, anterior deviation occurs when the outlet septum is displaced anteriorly into the right ventricle and is associated with aortic override and LVOT obstruction [7][8][9]. e progression of subaortic ridge in the presence of VSD is highly variable and unpredictable during childhood [1,[10][11][12]. Although progression is rapid and aggressive in some patients, others may exhibit slow and mild increase in LVOT gradient over the years, thereby mandating a prolonged follow-up of these patients [1,11,12].
Aortic regurgitation (AR) that occurs in cases of subaortic ridge and VSD may develop as a complication of VSD itself or due to progressive tethering and encroachment of the ridge onto the aortic valve leaflets. e frequency and progression of AR are variable depending on the age of the patient, anatomy of the ridge, and associated cardiac lesions [11,[13][14][15].
Surgical intervention is often required at some point in the clinical course of VSD and subaortic ridge. Surgery may be recommended in case of an LVOT pressure gradient >40 mmHg, LV volume overload, or aortic valve insufficiency. Some physicians prescribe surgery immediately after diagnosis, regardless of the LVOT pressure gradient, due to the progressive and nonbenign nature of this disease [10,15,16]. Early surgical intervention is negated by the high recurrence rate, reoperation risk, and development of aortic regurgitation [15,17,18].

Aim of the Study
e aim of the study was to evaluate the safety and efficacy of the management of patients with PM VSD and subaortic ridge with or without AR using transcatheter closure of the defect and compressing the subaortic ridge against the ventricular septum using Amplatzer ductal occluder type I (ADO-I). e goal of our technique is to protect the aortic valve and prevent progression of preexisting aortic regurgitation by ameliorating the subaortic ridge.

Patients and Methods
is prospective study was performed at the Ibn Albitar Center for Cardiac Surgery, Baghdad, Iraq, from August 1, 2014, to February 1, 2018. irty-eight (9.5%) of 398 patients with PM VSD were found to have subaortic ridge and were enrolled in this study. e inclusion criteria allowed patients with PM VSD and significant subaortic ridge causing LVOT flow disturbance with mild-to-moderate LVOT obstruction and mild and mild-moderate AR or LV volume overload. e exclusion criteria included patients with tunnel type and ring-like subaortic obstruction, severe prolapse of the right coronary cusp (RCC), moderate and severe AR, VSD with septal aneurysm, or VSD with malalignment of more than 25%, patients with a distance between the ridge and the proximal border of the VSD more than 5 mm, and patients who had other cardiac defects that required surgical intervention.
Informed consent was obtained from all patients or their guardians. Procedures were performed under deep sedation and fluoroscopic and transthoracic echocardiographic guidance. Patient data such as age, sex, weight, and TTE measurements were recorded. Echocardiographic assessment included defect size, location, distances of the ridge from the aortic valve and the VSD borders, presence of LV volume overload, peak and mean pulmonary arterial pressures, aortic cusp prolapse, and AR severity. In all included patients, the anatomical characteristics of the lesions as clarified by TTE evaluation were PM VSD with a significant subaortic membranous, fibrous, or fibromuscular ridge below the VSD that protrudes into the LVOT, with or without mild-to-moderate LVOT obstruction and mild-tomoderate AR.
Right and left heart catheterization was performed in all patients to evaluate the pulmonary and systemic pressures, magnitude of the left-to-right shunt, and measurement of the pressure gradient across the LVOT, which were obtained by catheter pullback from the left ventricle to the aorta. e patients were heparinized to achieve an activated clotting time of more than 200 seconds at the time of device implantation. A modified pigtail or Judkins right catheter was used to cross the VSD from the left ventricle with a 0.035 inch hydrophilic guidewire to either the pulmonary artery or the superior vena cava, which was snared and exteriorized from the femoral vein to create an arteriovenous loop. An appropriately sized delivery sheath was advanced from the femoral vein to the ascending aorta. e appropriate device size was chosen to be 1-4 mm larger than the VSD size as measured by the left ventricular angiogram at end-diastole, taking into consideration the anatomical characteristics of the subaortic stenosis as evaluated by TTE (the distance between the subaortic ridge and VSD border should not be more than 5 mm for effective capturing or compressing the ridge by the aortic disc of the ADO-I) (Figure 1(a)). e aortic disk of the ADO-I was deployed completely just under the aortic valve and pulled toward the defect, capturing and compressing the subaortic ridge against the ventricular septum, which was confirmed by continuous TTE guidance. e waist and the remainder of the device were deployed into the VSD and subsequently into the right ventricle (RV) side of the septum (Figure 1 All patients were discharged a day after the procedure and received aspirin (3-5 mg/kg) daily for six months. Most patients were followed up at one, three, six, and twelve months after the closure and then annually with a TTE study evaluating the device position, direction of the subaortic ridge, residual shunt, right ventricular systolic pressure, or any degree of AR. e mean follow-up interval was 27 ± 7 months (range: 17-41 months). e data are expressed as means ± standard deviations, medians, percentages, and ranges obtained using the IBM/SPSS statistical program, version 24 (IBM Corporation, Armonk, NY).
e VSD sizes ranged from 4 to 8 mm ( Figure 5(a) and Figures 4(a) and 4(d)), and the median distance of the subaortic ridge from the proximal edge of the VSD was 2. 5 mm (range: 1-5 mm) as measured by TTE in parasternal long axis view (Figure 1(a)). e Qp/Qs values ranged from 1.6 to 3.1 (mean � 2.2 ± 0.4), and the left ventricular diastolic dimensions which have been measured by TTE ranged from 24 mm to 59 mm (mean � 37.6 mm, median � 35 mm), which were significant in all patients. e mean of peak systolic pulmonary artery pressure was 30 ± 6 mmHg (range: 20-45 mmHg), which declined to 19 ± 2.6 mmHg after transcatheter closure of the VSD. None of the patients exhibited evidence of pulmonary hypertension on follow-up.
Eight (21%) of the patients were found to have mild aortic override (VSD with malalignment of 15-25%) due to anterior deviation of the outlet septum and or right ventriculo-infundibular fold into the right ventricle without compromising the right ventricular outflow tract ( Figure  1 Twenty-four (63%) patients were found to have mild and moderate prolapsed noncoronary aortic cusp (NCC) and/or right coronary cusp (RCC) (Figure 1  Journal of Interventional Cardiology pressure half-time before intervention was 385 ± 38 ms, which increased significantly after intervention to 535 ± 69 ms (vena contracta from 0.3-0.6 to ˂0.3) with a significant P value (0.001). Significant peak systolic pressure gradient across left ventricular outflow tract (LVOT) was found in nine patients, 23.7% (two patients less than 6 years, 3 patients 6-12 years, 2 patients 12-18 years, and 2 patients >18 years). e mean of the LVOT peak and mean systolic pressure gradient before intervention were 33 ± 7 mmHg and 15 ± 3 mmHg, which decreased after the intervention to 15 ± 2.4 mmHg and 11 ± 2 mmHg, respectively, with a significant P value (0.001). Patients with peak systolic pressure gradient of ≤20 mmHg were not included. e hemodynamic characteristics of patients with LVOT obstruction and AR are demonstrated in Table 1.
Five patients with mild AR showed improvement immediately after the procedure, while patients with mild-tomoderate AR (no. � 4) showed regression of the AR to a mild degree in two patients and to a trivial degree in the other two patients during the follow-up period (Figures 2(d) and 2(h) and Figure 5(d)).
e Amplatzer ductal occluder type one (ADO-I) (sizes: 6/4 (6 patients), 8/6 (19 patients), 10/8 (11 patients), 12/10(1 patient), and 16/14 (1 patient)) was used to close the VSD and capture and compress the subaortic ridge in this study. We have chosen ADO I device because we have managed a lot of cases of subaortic PM VSD using this device (in more than 400 cases) with satisfying results without major complications like complete heart block and aortic regurgitation through long-term follow-up period. We had used ADO II device in one patient with unfortunately an unsuccessful result as the device is too soft and flexible with a waist between two equal discs that make the compression of the ridge with closure of the VSD not applicable, as the soft left ventricular disc cannot compress the ridge and will not be well aligned with the IVS, making the residual shunt more significant. is is in contrary to the harder LV disc of ADO I device, which is capable of compressing the ridge effectively and would be well aligned with the IVS, making the residual shunt negligible.
Successful procedures were achieved in 33 patients (86.8%), in whom the VSD had been closed without residual shunt and the subaortic ridge had been captured and compressed against the interventricular septum toward the apex by the aortic disc of the ADO-I. e procedures failed in 5 patients due to a deficient aortic rim in three patients and increased severity of the AR in the other two patients.
At the mean of the follow-up period (27 ± 7 months), the highest LVOT systolic pressure gradient as detected by TTE was 22 mmHg, and there was no worsening in the degree of aortic regurgitation. All patients completed the follow-up period of 12 months, four patients missed the follow-up during the 2nd year, and three patients missed the follow-up during the 3rd year. All patients with AR and/or LVOT obstruction completed the follow-up period because we contacted them at the scheduled follow-up time. We encountered no significant complications acutely or on followup, and no patients developed heart block.

Discussion
We have introduced a new approach for the management of patients with PM VSD and subaortic ridge before and after  [20] projection demonstrating no AR despite mild prolapse of the RCC. e peak systolic pressure gradient across the LVOT decreased to 10 mmHg. development of AR via transcatheter closure of the VSD and concomitant capturing or compressing the subaortic ridge against the ventricular septum by the aortic disk of the ADO-I. e incidence of subaortic ridge in our study was 9.5% (38/398), which is comparable to those reported by Zielinsky et al. [5] (10.5%, 32/295) and Cassidy et al. [6] (7.2%, 16/223). Gabriels et al. [19] reported a higher incidence (14%), which may be due to the older age of their patients (median age � 29 years), including postoperative cases. e progressive nature of the subaortic ridge was confirmed by many investigators, such as Vogel et al. [12] and Horta et al. [10], who reported progression of LVOT pressure gradient in 63.89% of cases during the follow-up period. e preoperative development and progression of AR has also been mentioned by authors like Rohlicek et al. [15] (the incidence of AR increased from 18% (12/68) to 53% (36/68) during the follow-up period of 3.6 years), Babaoglu et al. [14] (91% of patients), and others [10,12].
Many authors documented the development and progression of AR postsurgical resection of the subaortic ridge, even in patients without preexisting detectable AR, such as Babaoglu et al. [14], Rohlicek et al. [15], Tefera et al. [20], and Donald et al. [18], who documented progression of aortic  Surgical resection of the subaortic ridge with or without extensive myectomy carries the risk of development of complete heart block, which has been mentioned in many studies like that of Parry et al. [21] (5 patients, one initially and 4 after reoperation), Serraf et al. [22], and Drolet et al. [23] (6%, 3/49), while, with our technique, no patient developed heart block throughout the follow-up period.
Patients with VSD and subaortic ridge without AR were treated successfully using this technique with amelioration of the LVOT obstruction without interference with the aortic valve function (Figures 1, 3, and 4). e regression of AR in some patients after device closure of the VSD and compression of the subaortic ridge is not completely understood and may be related to multiple factors (Figures 2(b), 2(d), 2(f ), and 2(h); Figures 5(b) and 5(d)). We thought that closure of the VSD with abolishment of the left-to-right shunt effect on the aortic valve leaflets and the restoration of the LVOT anatomy (after closure of the VSD and compressing the subaortic ridge against the ventricular septum) greatly decreased the blood flow disturbance through the LVOT, a finding that was also observed after closure of the VSD with AR and without an associated subaortic ridge. If device implantation induces a new onset or worsening of preexisting AR, the device was retrieved and the patient was referred for surgery. In this study, most patients had a prolapse of the noncoronary cusp (NCC), which did not interfere with the device implantation or resulted in induction or worsening of AR (Figures 2(a) and 2(h) and Figure 3(i)). Patients with severe prolapse of the right coronary cusp (RCC) were excluded from this study as transcatheter closure of the VSD was unsuitable because the RCC was prolapsing into the defect (Figure 6(b)). All patients with prolapsing cusps (24/ 38, 63%) who underwent successful procedures without induction of AR continued to have no AR during the followup period of 12 months. ree patients missed follow-up during the next two years.
In patients with LVOTobstruction (no. � 9), it was found that the severity of obstruction increased with age of the patients (7 patients were more than 6 years), a finding that was reported by other authors, such as Shem-Tov et al. [11], and Rohlicek et al. [15].
Perimembranous subaortic VSD with malalignment associated with subaortic ridge has been described by many authors. Zielinsky et al. [5] reported malalignment VSD in all patients (100%), and Kitchiner et al. [7] reported nearly equal percentages between subaortic ridge with malalignment VSD (33/65, 50.8%) and those with short-segment fibromuscular ridge without malalignment (32/65, 49.2%). In our study, only eight patients (21%) had VSD with mild malalignment due to anterior deviation of the outlet septum that resulted in aortic override of 15-25% according to our global assessment, which did not interfere with our technique to close the VSD and compress the subaortic ridge that was abuting the lower border of the defect (Figures 2(a)-2(c), 2(e) and 2(g), Figures 4(a), 4(d), 4(e) and 4(f ), and Figure 1(a)). Patients with malalignment VSD ≥ 30% were excluded from this study because we found that our technique cannot be applied in such cases (Figures 6(c) and 6(d)). Patients with associated septal aneurysm were excluded because our technique cannot be applied to close the defect and compress the subaortic ridge against the septal aneurysm ( Figure 6(a)).

Conclusion
Transcatheter management of patients with PM VSD and subaortic ridge is feasible and effective in protecting the aortic valve by ameliorating the blood flow disturbance across the LVOT created by the left-to-right shunt across the VSD and the presence of subaortic ridge with regression of the preexisting AR. In addition, this technique was found to be effective in relieving LVOT obstruction by compressing the subaortic ridge against the interventricular septum without creating a new AR.
Data Availability e data used in this study are available within the article.

Conflicts of Interest
e authors declare that there are no conflicts of interest.