Feasibility and Success of Muscular Ventricular Septal Defect Occluders and Mushroom-Shaped Occluders in Transcatheter Patent Ductus Arteriosus Closure in Low-Weight Children: A Propensity Score-Matched Retrospective Analysis from a Chinese National Regional Health Center

Patent ductus arteriosus (PDA) is a common congenital heart defect that often leads to hemodynamic challenges in affected children [1]. Since the pioneering report of transcatheter closure by Porstmann in 1966, transcatheter PDA closure (TCPC) has gained widespread acceptance and has become the preferred treatment option due to its high success rates, less trauma, and expedited recovery compared to surgical closure [2, 3]. Currently, the mushroom-shaped occluder (MSO) and coil occluders are commonly selected in TCPC, but they are mostly approved for use in patients with a weight of ≥6 kg.

For some special types of PDA (such as C type PDAs), and/or low-weight implantation of those occluders, the potential for post-implantation complications should be concerned such as thrombocytopenia, device-related left pulmonary artery (PA)/descending aorta (AO) obstruction, occluder displacement, detachment, or even unplanned surgery [4]. Therefore, some institutions have attempted to explore alternative occluders such as muscular ventricular septal defect occluder (MVSDO) to reduce the aforementioned complications, achieving promising results with high success rates and favorable short-term outcomes [5‒11]. Nonetheless, most of the studies reported small numbers of cases, lacked comparisons with MSO, and provided inadequate descriptions of follow-up.

Consequently, this study reported the preliminary results of TCPC using MVSDO at our center, accompanied by an analysis of postoperative outcomes in two groups of children using MSO and MVSDO. And propensity socore matching was employed to address potential selection bias.

This study conducted a retrospective analysis of children who underwent TCPC procedures at a Chinese national regional health center (Children’s Hospital of Chongqing Medical University) from 2015 to 2021. Figure 1 provides an overview of the study design. We encompassed the medical records of children at our center. Adhering to stringent criteria, a total of 417 children were finally included. Children were stratified into two groups according to the occluder employed: the MSO group and the MVSDO group. Notably, children initially assigned to the MSO were intraoperatively transitioned to the MVSDO following further evaluation – these subjects were subsequently classified within the MVSDO group. Employing a 1:1 propensity score matching (PSM) methodology, matched pairs for a comprehensive assessment were identified.

Inclusion criteria were as follows: (1) weight less than 10 kg; (2) presence of clinical symptoms and signs of cardiac overload; (3) no concurrent surgical indications for other cardiac anomalies; (4) meeting the criteria for TCPC. Children with missing data, those who received types of occluders other than MVSDO and MSO such as coil were excluded. The patients were categorized into MSO group (using MSOs) and the MVSDO group (using MVSDOs) based on the types of occluder used.

This study was conducted in compliance with the Helsinki Declaration [12] and received approval from the Ethics Committee of Children’s Hospital of Chongqing Medical University (202383). Children’s parents, legal guardians, or next of kin to children in this study were thoroughly informed about the potential benefits and risks of TCPC before procedure, and a written informed consent was obtained from them.

All procedures strictly adhered to the guidelines outlined in the Chinese Expert Consensus on Interventional Treatment of Common Congenital Heart Diseases [13, 14]. Prior to the procedure, all children underwent comprehensive assessments, including echocardiography, complete blood count, coagulation function tests, electrocardiography, and chest radiography. Under general anesthesia, femoral vein and femoral artery punctures were performed. After successful puncture, heparin was administered at a dose of 100 U/kg. Pulmonary artery (PA) pressure and descending aortic pressure were measured. A pigtail catheter was introduced for aortic arch angiography to determine the size, morphology, and location of the PDA to select an appropriate occluder. Subsequently, the catheter was advanced to the descending AO, and an exchange wire was placed below it. The selected occluder was then delivered along the delivery sheath to the target location and released. Routine anticoagulation and antiplatelet therapy were not administered after PDA closure.

The MSO and MVSDO used in our center are manufactured by LifeTech Scientific (Shenzhen, China), Shanghai Shape Memory Alloy (Shanghai, China), and Starway Medical (Beijing, China). The MSO is characterized by its mushroom-shaped design, and when choosing it, the diameter typically exceeds the narrowest PDA diameter by 3–6 mm. For infants and young children with elastic PDA, the occluder diameter is recommended to be twice the PDA diameter. The MVSDO is an Amplatzer-sized modified domestic double-disc occluder for muscular ventricular septal defects (MVSD), as shown in our previous work [15]. It features symmetrically positioned left and right discs, with the disc’s diameter being 4 millimeters larger than the waist and a waist length ranging from 5 to 7 millimeters. Given the limited experience with MVSDO and the absence of guidelines, the selection of MVSDO primarily relies on the operator’s expertise and insights from previous literature reports.

Clinical data were extracted from the hospital’s electronic medical record system, encompassing the following categories: (1) demographic information: gender, height, weight, and age; (2) preoperative data: the narrowest diameter of the PDA as determined by ultrasound, PDA’s PA end diameter, PDA’s aortic end diameter, PDA length, PDA type, degree of PA hypertension, and preoperative blood routine parameters (red blood cell count, hemoglobin, and platelet count); (3) intraoperative data: PDA angiographic diameter, PDA type, right ventricular pressure, PA systolic pressure, occluder size, occluder type, and delivery sheath size; (4) postoperative follow-up: thrombocytopenia, hemolysis, occluder displacement or migration, residual shunting, valve regurgitation, femoral arteriovenous fistula or aneurysm, thrombosis, puncture bleeding, as well as measurements of descending aortic flow velocity (DAF) and stenosis, and PA flow velocity and stenosis.

Thrombocytopenia is defined as a post-TCPC platelet count below the lower physiological limit of 100 × 10⁹/L. Depending on the degree of decrease in platelet count, it was further classified as mild (postoperative platelet counts between 51–100 × 10⁹/L), moderate (postoperative platelet counts between 20–50 × 10⁹/L), and severe (postoperative platelet counts <20 × 10⁹/L) [16].

During hospitalization, patients undergo assessments at least 3 times (prior to closure, on the first post-closure day, and 1 day before discharge). Subsequently, they receive a minimum of four follow-up evaluations (at 1 month, 3 months, 6 months, and 12 months after discharge), including complete blood routine tests, electrocardiograms, and echocardiograms.

All statistical analyses were conducted using IBM SPSS 22.0 (SPSS Inc., Chicago, IL, USA), R (http://www.R-project.org), and EmpowerStats statistical software (www.empowerstates.com, X&Y solutions, Inc., Boston, MA, USA). To mitigate the impact of confounding variables on post-procedural outcomes across various occluder types and ensure comparability between the two groups, we performed PSM. Gender, height, weight, BSA, and BSA-corrected PDA diameter were matched at a 1:1 ratio to address baseline differences. Propensity scores were calculated using logistic regression, and matched pairs were formed utilizing the nearest-neighbor matching algorithm.

Categorical variables were expressed as frequencies (percentages), normally distributed continuous variables as means (standard deviations), and skewed continuous variables as medians (interquartile range). Between-group comparisons were performed using Student’s t test, Mann-Whitney U test, Pearson χ² test, and Fisher’s exact test, with statistical significance defined as a p value <0.05.

A total of 1,135 cases were performed TCPC in our center between 2015 and 2021. Then, we finally enrolled 417 children with a body weight less than 10.0 kg in our study, including 372 children in MSO group and 45 children in MVSDO group. Among those 45 cases with a mean weight 5.92 ± 1.32 kg, 29 was type A PDA, 11 was type C PDA, and 5 was type B PDA. The mean PDA diameter was 6.2 ± 0.2 mm (ranging from 2.5 to 9.7 mm) by transthoracic echocardiography. Furthermore, the angiography diameters at the PA and aortic ends were 4.5 ± 0.2 mm (ranging from 1.5 to 8.5 mm) and 7.9 ± 0.4 mm (ranging from 0.8 to 12.0 mm). The average diameter of MVSDO was 9.3 ± 0.3 mm (ranging from 4.0 to 14.0 mm). The mean procedure duration was 45.36 ± 15.92 min and the mean hospital stay was 12.11 ± 7.16 days.

All children using MVSDO achieved immediate successful closure. However, in one child, successful closure was initially achieved, but occluder displacement occurred one day after the procedure. This child, aged 12 months and weighed 7.0 kg, presented with a type C PDA characterized by an angiographic diameter of 4.9 mm at the aortic end and a narrowest diameter of 5.2 mm at the descending aortic arch. Ultrasound measurements indicated a PDA diameter of 6.1 mm. And an 8 mm MVSDO was selected for the occlusion. Immediate postoperative cardiac echocardiography confirmed the favorable positioning of the occluder, with no evidence of residual shunting or secondary vascular stenosis. However, routine echocardiography performed on the first day following the procedure revealed occluder displacement in the left PA. This necessitated the subsequent removal of the occluder and arterial duct ligation, which was subsequently performed by the cardiothoracic surgery team.

In our examination of 45 pediatric cases ultimately assigned to MVSDO group, 18 had initially undergone MSO before transitioning to MVSDO. A review of surgical records revealed that a significant constriction of the descending AO or the left PA occurred following the implantation of a large-diameter MSO, as depicted by intraoperative angiography and cardiac echocardiography. This significantly altered hemodynamics, despite preoperative discussions anticipating such an outcome. When such scenarios occurred, comprehensive consent was secured from the families of the patients: if they assented to retry interventional occlusion, MSO would be retracted and an appropriately sized MVSDO reintroduced; conversely, if they opted against further interventional occlusion attempts or if the interventional specialist panel deemed the patient’s circumstances unfit for interventional occlusion, MSO would be removed and the patient scheduled for elective surgical intervention at a later time.

Table 1 summarizes the characteristics of the two groups of children before and after PSM. Before PSM, the MVSDO group had younger average age, lower weight, shorter height/length, special types of PDA, higher rate of PA hypertension, smaller delivery sheaths, larger occluder diameters, longer procedure duration, less impact on platelet count, and longer hospital stay (p < 0.05). PSM was performed to balance the baseline characteristics between the two groups. Finally, 41 well-matched patient pairs were included for analysis (standardized mean difference [SMD] <0.2). After PSM, the MVSDO group had younger average age, more special types of PDA, higher pulmonary hypertension, smaller delivery sheaths, and larger occluder diameters (all p < 0.05).

Table 2 shows the incidence of in-hospital complications in the two groups of children before and after PSM. Before PSM, the MVSDO group had a higher incidence of residual shunts, postoperative PAH, a less pronounced impact on platelet count, and a longer hospital stay than the MSO group (all p < 0.05). After PSM, the MVSDO group had a higher incidence of postoperative residual shunt and postoperative PAH, a lower impact on platelet count, reduced incidence of thrombocytopenia (all p < 0.05). Postoperative PAH degrees decreased significantly in both groups, but there was no significant difference in the improvement between the two groups.

Table 3 presents a comparison of the follow-up results of the propensity score-matched children in the two groups, including left ventricular ejection fraction, left pulmonary artery flow velocity (LPAF), and descending aorta flow velocity (DAF). LPAF of both groups increased on postoperative day 1 compared with the preoperative period and then gradually showed a gradual decline. DAF reduced after the device placement compared with preoperative velocities.

This study presents the experience of TCPC using the MVSDO at a Chinese national regional health center as well as the leading pediatric tertiary hospital in Western China. We treated 45 patients with a mean weight of 5.92 ± 1.32 kg, achieving a 100% immediate procedural success rate. However, one patient required unplanned surgery for occluder displacement on the first postoperative day. Subsequently, we performed a comparative analysis of outcomes and follow-up between children using MVSDO and MSO. Our findings suggest that choosing MVSDO is a safe and technically feasible option, especially for patients with low-weight, special PDA types, and concurrent PAH. After PSM, compared to the MSO group, MVSDO group had a lower incidence of thrombocytopenia and a higher incidence of residual shunts. At 1-year follow-up, the descending AO and left PA velocities were acceptably altered, and no secondary vascular stenosis was observed.

TCPC, as a well-established technique, is the preferred treatment of choice for PDA closure in patients weighing ≥6 kg [17]. MSOs are commonly used worldwide for closing small- to medium-sized PDAs (PDA diameter ≤4.0 mm), but in children of low weight and large PDAs, they are at risk of protrusion into the aortic lumen, aortic deformation, luminal narrowing, and embolization into the descending AO [17]. Limited literature reports transcatheter closure for such special types of PDAs. Facing the challenge head-on, the MVSDO, characterized by a symmetrical double-disc structure with each disc 4 mm larger than the waist tending to anchor the device, theoretically reduces the impact on the PA and AO while ensuring effective closure, thus lowering the risk of occluder displacement and subsequent embolization.

Therefore, several institutions have explored the use of MVSDO for transcatheter closure in low-weight, large, and special type PDAs and achieved favorable outcomes [5‒11, 18]. Vijayalakshmi et al. [5] shared their preliminary clinical experience in closing 10 large PDAs with a modified and tilted MVSDO. Thanopoulos et al. [6] reported the treatment of 7 children with large PDA and severe pulmonary arterial hypertension using the Amplatzer MVSDO. Salam et al. [7] presented the experience of using Occlutech and Amplatzer MVSDOs in 8 children with a body weight of less than 10 kg and type C PDA accompanied by heart failure is presented. Cubeddu et al. [11] reported a case about a 28-year-old female with 16 mm PDA. Mei Jin et al. [18] also shared their experience at Beijing Anzhen Hospital in treating 12 cases of large PDA with the Amplatzer MVSDO. Some Chinese domestic institutions also reported their outcomes [8‒10]. It is evident that the technique of TCPC using MVSDO has proven to be viable, especially in special cases such as low weight, large PDAs, and cases associated with PAH, with acceptable postoperative complications.

The occlusion success rate at our center is consistent with previous studies [19‒21]. However, occluder displacement occurred on the first postoperative day in one child after immediate occlusion. This may have been related to the child’s underlying condition, disease state, or an inappropriate choice of occluder. This underscores the importance of postoperative ultrasound follow-up and highlights the need to be cautious when selecting MVSDO, despite its high success rate. García-Montes et al. [22] recommended prioritizing MVSDO for patients with larger type C PDAs (≥8 mm) or partial type A PDAs (aortic potbelly diameter exceeding 1.5 times the PA end). Given the limited research and lack of established guidelines for MVSDO selection, exercising caution is advisable. Factors such as the child’s age, weight, PDA characteristics, dimensions, and occluder material should be carefully considered to reduce the likelihood of necessitating unplanned surgery.

It merits emphasis that subsequent to MSO occlusion, the transition to MVSDO in those 18 children due to pronounced effects on the descending AO or left PA culminated in successful occlusion for all. Theoretically, assuming the technical success of the intervention, the MVSDO, with its symmetric double-disk structure, is more suitable for anchoring the occluder without causing vascular obstruction. Therefore, if intraoperative findings indicate narrowing of the descending AO or a significant change in blood flow velocity, we proceed to remove the MSO and opt for MVSDO for re-occlusion. This approach  has, to a certain extent, reduced the occurrence of unplanned secondary surgeries in children and alleviated the economic and psychological burden on families. This is a real-world scenario encountered in our clinical practice.

The findings underscore the technical viability of utilizing MVSDO for occlusion, with no similar adverse effects on the descending AO/left PA following MVSDO implantation. Such outcomes logically account for the protracted interventional procedural time observed in the MVSDO group. Importantly, the implementation of MVSDO markedly curtailed the frequency of unplanned surgery in PDA management; as bereft of MVSDO, these patients would be relegated to awaiting surgical closure. These observations, in turn, obliquely indicate a requisite for further refinement in our comprehension of implantation indications for MVSDO.

In our study, the MVSDO group was younger, after PSM, had a higher prevalence of special PDA types, and had greater comorbidity with PAH compared to the MSO group. This suggests that the use of MVSDO for PDA occlusion in these children is a safe and viable option. And we primarily focused on postoperative complications including residual shunts, improvement of PAH, and impact on the PA and AO. Nineteen of 45 children (42.2%) developed residual shunts after MVSDO occlusion, more frequently than MSO occlusion. However, most residual shunts were minor and resolved spontaneously within 6 months, consistent with previous studies [23]. The occurrence of microscopic hematuria was more frequent in the MVSDO group, potentially linked to a higher prevalence of residual shunting. MVSDO-treated patients had a higher comorbidity with PAH. However, postoperative PAH significantly improved, with most patients returning to a mild or normal state at 1-year follow-up, consistent with previous studies [7‒9, 24]. These outcomes are considered acceptable for postoperative residual shunts, microscopic hematuria, and PAH improvement.

Thrombocytopenia after TCPC may be related to residual shunts and platelet aggregation around the occluder [18‒20]. In our study, two out of 45 children in the MVSDO group experienced thrombocytopenia associated with mild residual shunts. Their platelet counts returned to normal after hormone therapy, consistent with findings by Wang et al. [25]. After PSM, the MVSDO group exhibited reduced postoperative impact on platelet count and a lower likelihood of thrombocytopenia compared to the MSO group. Our findings indicate that postoperative thrombocytopenia associated with MVSDO should not be overlooked and present a favorable alternative to MSO, despite its lower incidence.

Similar to previous study [26], neither group had significant left PA or descending AO stenosis or obstruction at the 1-year follow-up, although blood flow velocity was affected. After device implantation, increased LPAF after device implantation is common as previous studies [27, 28]. However, DAF decreased after device implantation compared with preoperative velocities, likely attributable to decreased blood volume traversing the aortic isthmus after elimination of the left-to-right ductal shunt [29]. The MVSDO’s aortic end disk surface fits into the aortic vascular surface of the PDA to form a microcurve, which stabilizes the device and eliminates the left-to-right shunt while minimizing obstruction of blood flow to the aortic end of the AO [8‒10]. However, no intergroup differences were observed at 1-year follow-up. Our short-term follow-up suggested that the effects of MVSDO on the left PA and descending AO were benign and anticipated to improve spontaneously without leading to secondary stenosis. However, further research and long-term follow-up are needed to determine if the MVSDO group has a reduced impact on descending aortic stenosis.

This study acknowledges some limitations. First, as a single-center retrospective study with a limited sample size, it is subject to potential biases. Larger prospective studies with longer-term follow-up are needed to evaluate the postoperative outcomes and long-term prognosis of children using MVSDO. Second, missing values for left PA and descending AO diameters limit the assessment of vessel size and changes, which can only be indirectly evaluated by blood flow velocity measured by ultrasound. Finally, MVSDO occluder selection in this study was based on empirical criteria, and the outcome of the procedure may be related to the operator’s technique and experience. Therefore, the results of this study may not be generalizable to all interventionalists.

MVSDO used in TCPC is technically feasible and effective in pediatric patients with low body weight (<10.0 kg) who have large PDA with a high immediate occlusion rate and a low incidence of thrombocytopenia. Postoperative complications were satisfactory and no artery secondary stenosis was observed in 1-year follow-up. Moreover, occluder displacement and thrombocytopenia, although occurring in small number of children, should not be underestimated. Larger sample sizes and longer follow-up periods studies are needed to evaluate long-term efficacy and to identify potential complications.

The authors would like to thank all the patients who enrolled in this study, along with colleagues and medical staff who took part in patient care.

This study was approved by the Ethics Committee of Children’s Hospital of Chongqing Medical University (Approval No. 202383). A written informed consent was obtained from children’s parents/legal guardians/next of kin to children in this study.

All authors declare that there are no conflicts of interest regarding the publication of this paper.

This work was supported by Key Topic of Chongqing Municipal Health and Family Planning Commission (2016ZDxm018).

Kaijun Zhang, Le Yang, Ping Xiang, Mi Li, and Xue Zhou: concept and design. Kaijung Zhang, Le Yang, Rensen Zhang, Penghui Yang, and Jingdong Ma: data collection, analysis, and interpretation. Kaijung Zhang and Min Cheng: drafting of the manuscript. Ping Xiang, Mi Li, and Xue Zhou: administrative support, supervision, and critical revision of the manuscript.

The original contributions presented in this study are included in the article; further inquiries can be directed to the corresponding author.