A multi-center trial on efficacy and safety of the LifeTech CeraFlexTM ASD occluder for transcatheter closure in patients with secundum atrial septal defects.

Background: The last decades have brought remarkable improvements in treatment strategy and occluder modification of secundum atrial septal defect (ASD) closure. Approval, efficacy and safety of ASD closure devices have previously been demonstrated. This study investigated the clinical efficacy and safety of the LifeTech CeraFlexTM ASD occluder for interventional closure of secundum ASD with a 6-month follow-up (FU).
Methods: Procedure specific data was collected on patients considered for ASD closure with the CeraFlexTM occluder between April 2016 and December 2019 in three German centers. Efficacy and safety were assessed after device closure, at discharge, and at 6-month FU.
Results: The primary endpoint (successful ASD closure without severe complications) was reached by 102/103 patients (99%). Device embolization occurred in two patients (one early and one late embolization). After early snare-retrieval of an embolized device, this ASD was closed surgically and in the other patient with late device embolization the defect was closed with a larger CeraFlexTM occluder. The secondary endpoint (clincal efficacy after 6 months) was reached by 94/98 patients since new onset of arrhythmia occurred in four patients. Three patients had withdrawn their study-participation and one patient had moderate residual shunt, but not related to the occluder. Incomplete right bundle branch block (iRBBB) was seen in 31 patients. At last FU only 17 patients had remaining iRBBB documenting effective volume unloading of the right ventricle. Conclusions: Catheter interventional closure of secundum ASDs with the CeraFlexTM ASD occluder was feasible, safe and effective in this study. 2022 Cardiovascular Diagnosis and Therapy. All rights reserved.

Introduction

An atrial septal defect (ASD) is the most common congenital heart defect in adults with congenital heart disease (CHD) and the second most common cardiac defect in children (1,2). Previously, surgical closure was the only available treatment option (3,4). Since the first successful percutaneous closure of a secundum ASD by King et al. in 1976 (5), transcatheter closure has become the treatment of choice in many centers (6,7). Compared with surgical ASD closure, transcatheter treatment is associated with a shorter hospital stay, a reduced morbidity and complication rate, and even a reduced mortality. Additionally, a reduced occurrence of atrial fibrillation/flutter and a reduced rate of systemic thromboembolisms/ischemic strokes after catheter ASD closure is reported (7-10).

Consequently, numerous transcatheter ASD occluders have been developed and modified by different manufacturers (6). Due to their characteristics, most of the products currently available have outstanding efficacy and achieve excellent outcome in defect closure (7,11). These successes and excellent results are also attributable to significant improvements in procedural techniques (11). The CeraFlexTM ASD occluder (LifeTech Scientific Co., Shenzhen, China) is a self-expanding double-disk device made from a nitinol wire mesh that has been commercially available in Europe since its CE certification in 2013 (12-14). In contrast to the Amplatzer ASD occluder (Abbott, Lake Country, IL, USA) which has a stiff delivery cable, the CeraFlexTM occluder is implanted with a very flexible delivery cable, which allows device implantation in the final position, which does not change after device release.

We present results of a German prospective multi-center trial evaluating clinical efficacy and safety of the LifeTech CeraFlexTM ASD occluder for transcatheter closure of secundum ASDs in a large study population with a 6-month follow-up (FU) under controlled and structured study conditions. We present the following article in accordance with the TREND reporting checklist (available at https://cdt.amegroups.com/article/view/10.21037/cdt-21-798/rc).

Methods

Study design and formalities

This study was a German prospective, multi-center (German Heart Center Munich, University Heart Center Freiburg, and Heart Center, University of Leipzig) single-arm trial to evaluate the CeraFlexTM ASD occluder’s efficacy and safety.

It was conducted in accordance with the Declaration of Helsinki (as revised in 2013) and the Good Clinical Practice guidelines, DIN EN ISO 14155:2011 (German version) and the current national regulations including the Medical Devices Act. The study protocol was approved by the local ethical board of the Technical University Munich (project number 451/15S) and registered at the German Clinical Trial Register (https://www.drks.de/drks_web/; registration ID: DRKS00010233). Between April 2016 and December 2019 secundum ASDs were closed with the CeraFlexTM ASD occluder in 103 patients at the three participating study sites. All participants gave written informed consent for study participation after being provided with information about the study protocol.

Study patients

For inclusion, patients had to be diagnosed with a hemodynamically significant secundum ASD, which could be closed with an available CeraFlexTM ASD device. Exclusion criteria were myocardial infarction, unstable angina pectoris, decompensated heart failure, multiple defects hindering closure with only one occluder, and current active bacterial and/or viral infection or evidence of intra-cardiac thrombi on echocardiography and participation in another ongoing clinical trial.

In- and exclusion criteria were assessed using medical history in advance, and by transesophageal echocardiography (TEE), fluoroscopy, and balloon sizing directly before ASD occlusion (ASDO).

CeraFlexTM ASD occluder

The CeraFlexTM ASD occluder (see ) is a double-disk device for transcatheter secundum ASD closure. Polyethylene terephthalate (PET) membranes are sewn into each disc and the waist as foundation for growth of tissue over the occluder after placement and sealing of the hole (14). It has bioceramic titanium nitride (TiN) coating and no left atrial hub to reduce nickel ion elution and the risk of clot formation, and improve endothelialization and biocompatibility (13,14). The occluder has a 360° flexible rotation feature between the device and delivery cable, enabling accurate positioning without tension and a better adaptation to the interatrial septum (12). This device is available up to 42 mm waist diameters with 2 mm increments, and delivery sheath sizes from 8 to 14 Fr (14).

Figure 1

LifeTech CeraFlexTM ASD occluder. ASD, atrial septal defect.

Implantation procedure and FU

Interventional procedures were conducted according to instructions for use and the standard approach of the respective study site. Transcatheter treatment was performed using standard intraprocedural imaging techniques such as TEE with fluoroscopic guidance if necessary. Device size was determined mainly on the basis of the stretched defect diameter using the “stop-flow technique” or frequently using the “waist technique” (15). Successful implantation was defined as stable and correct placement of the occluder within the defect.

All patients were admitted to the hospital 1 day preceding the intervention for all necessary examinations and were discharged the day after ASD-closure to ensure a good procedural result. Efficacy and safety were assessed by physical examination, vital signs, electrocardiogram (ECG), transthoracic echocardiography (TTE), performed during admission and intervention, before discharge and at 6-month FU. Serious adverse events (SAEs) that occurred during the procedure, until discharge or FU were documented and treated as appropriate (see ).

Table 1

Schedule of assessments

Assessments	Admission	Procedure in cardiac catheter	Before discharge	Six-month FU (±1 month)
Informed consent	√
Inclusion/exclusion criteria	√
Medical history	√
Demographic data†	√			√
Examination	√		√	√
Vital signs‡	√		√	√
ECG	√		√	√
Transthoracic echocardiogram	√		√	√§
Transesophageal echocardiogram		√		√§
Data from cardiac catheter		√
Anticoagulation	√		√	√
SAE§§		√	√	√

†, sex, age (years), height (cm), weight (kg); ‡, vital signs: blood pressure, heart rate; §, patients <15 years: transthoracic echocardiogram, patients >15 years: in accordance with clinical routine transesophageal echocardiogram or transthoracic echocardiogram; §§, documentation of SAEs. ECG, electrocardiogram; SAE, serious adverse event; FU, follow-up.

Primary endpoint was defined as successful placement of the occluder and successful closure of the defect without severe complications [embolization of the occluder, surgical re-intervention, moderate or large residual shunt, new onset of atrioventricular (AV) block II° or III°]. Secondary endpoint was defined as clinical efficacy 6-month after the intervention, implicating a stable occlusion and a successful closure of the defect without severe complications (erosions, embolization of the occluder, surgical re-intervention, moderate or large residual shunt, new onset of arrhythmia or AV block II° or III°) in the FU period. The size of the residual shunt was classified into trivial (<1 mm), small (1–2 mm), moderate (>2–4 mm) and large (>4 mm).

Quality assurance was ensured by an external monitor at the three participating study sites. Two to eleven monitoring visits, depending on the number of patients enrolled, were performed per study site. All study specific activities and the regular inspections of the collected data by the certified monitor were performed in accordance with DIN EN ISO 14155:2011 (German version).

Statistics

Data of intention to treat analysis are presented as median and interquartile range (IQR 25; 75). Categorical variables are expressed as absolute numbers or percentage.

Statistical analyses were performed using SPSS (version 23.0, IBM Corporation, Armonk, NY, USA).

Results

Between April 2016 and December 2019, a total of 148 patients were identified as potential study patients in the three participating study sites. Forty-five patients had to be excluded due to multiple defects or too large defect size. Course of Study, study characteristics and procedure related data of the patients are displayed in detail in , .

Figure 2

Course of study.

Table 2

Anthropometric data of patients with a successfully closed ASD using CeraFlexTM occluder

Anthropometric data	Successfully closed ASDs (n=103)
	Median [IQR 25-75]	Min/max
Sex, female, n (%)	75 (72%)	–
Age at procedure (years)	9 [6-47]	3/80
Weight (kg)	29 [21-64]	13/120
Height (cm)	135 [117-167]	97/192
Defect size (mm)	12 [10-16]	4/30

ASD, atrial septal defect; IQR, interquartile range.

Table 3

Procedural data of patients with a successfully closed ASD using CeraFlexTM occluder

Procedural data	Successfully closed ASDs (n=103)
	Median [IQR 25-75]	Min/max
Defect size (mm)	12 [10-16]	4/30
Device size (mm)	12 [10-16]	6/32
Procedure duration (min)	42 [28-65]	15/221
Fluoroscopy time (min)	0.9 [0-5]	0/34

ASD, atrial septal defect; IQR, interquartile range.

Due to the flexible delivery system placement of the occluders was simple and the occluder did not change its position in the heart during and after detachment from the delivery cable. The primary endpoint (successful ASD closure without severe complications) was reached by 102/103 patients (99%; see ). One device embolized early and was removed with a snare catheter on the first post interventional day. Surgical ASD closure was performed in this patient, and he did not reach the primary endpoint.

Another patient was found to have late device embolization into the aortic arch 4 weeks after device closure. The occluder was removed with a snare via the femoral artery from the aorta and the ASD was then closed with a larger CeraFlexTM device. This patient did not reach the secondary endpoint. Four patients developed clinically non-relevant supraventricular tachycardias (atrial flutter/fibrillation) during the FU period and were treated with antiarrhythmic medication. Atrial flutter, paroxysmal atrial fibrillation, supraventricular extra systoles and palpitations, as well as ECG abnormalities in 24-hour ECG were diagnosed for the first time in a total of four patients (4%) at 6-month FU. Accordingly, these four patients did not reach the secondary endpoint.

In the 6-month FU period three patients withdrew their consent and were excluded from the study. Hence, 98 patients completed the 6-month FU and 94 patients reached the secondary endpoint. One patient showed moderate residual shunt at 6-month FU. Since this shunt was newly detected and not related to the occluder a previously not diagnosed second ASD was postulated and the patient was not excluded. In 92/98 patients (94%) the defect was completely closed, without residual shunt at TTE (6-month FU). In 5/98 patients (5%) a trivial residual shunt was visible adjacent to the occluder.

Preceding secundum ASD closure incomplete right bundle branch block (iRBBB) was diagnosed in 31/103 patients (30%). At 6-month FU it was diagnosed in 17/98 patients (17%) no new ECG changes were noted.

Known minor side effects (adverse events) such as groin hematoma, hemorrhage, vertigo, nausea, vomiting, headache and migraine occurred only temporarily without permanent sequelae after intervention. SAEs like embolizations (n=2), vessel complications (n=1), syncope (n=1), vasovagal reaction (n=1), and arrhythmia (n=1) were gathered after transcatheter treatment and during FU period. In all SAEs the original state of health could be recovered.

No erosion, perforation or death occurred in any of the study patients during study period.

Device embolization

First patient

A 19-year-old woman presented an 11 mm ASD II in echocardiography (TTE and TEE). During balloon sizing a balloon occlusion diameter (BOD with “stop flow” technique) of 12 mm was assessed. A 12 mm CeraFlexTM ASD device was positioned. It was not stable and, hence, a 16 mm CeraFlexTM ASD occluder was implanted. On the next day this occluder had migrated to the left ventricle, from where it was removed by gentle retrieval into the descending aorta. It was withdrawn through a 12 F sheath from the right femoral artery with a Maslanka biotome. Six months later, during a second catheter interventional procedure, the defect now had a diameter of 16 mm in TEE and a BOD of 19 mm. A 21 mm Occlutech ASD device was positioned, but during stability testing, preceeding device release, the device fell easily into the left atrium. It was retrieved and surgical closure was scheduled. During surgery an oval shaped ASD II 16×25 mm was seen, which was closed with a GoreTex patch.

Due to device embolization before discharge the primary endpoint was not reached by this patient.

Second patient

A 63-year-old male patient, who was treated previously for severe stenosis of the right coronary artery with a drug eluting stent, presented an 18 mm ASD II with thin interatrial septum combined with an atrial septum aneurysm. Balloon sizing revealed a BOD (“stop flow” technique) of 20–22 mm. To exclude severe diastolic dysfunction, left ventricular end-diastolic pressure (LVEDP) was assessed during test occlusion of the ASD. LVEDP remained unchanged at 17 mmHg. A 20 mm CeraFlexTM ASD occluder slipped through at the first attempt, but was positioned correctly at the second try.

During an examination 4 weeks after ASDO, not considering our study, the patient had no cardiac symptoms, however, the occluder had migrated to the proximal aortic arch. It was removed with a 6 F Multisnare 20 mm through a 14 F sheath introduced into the right femoral artery. At the same session the ASD was closed with a 28 mm CeraFlexTM ASD occluder, which remained stable until the last routine examination.

Due to device embolization during 6-month FU the secondary endpoint was not reached by this patient.

Discussion

This is a German prospective multi-center trial evaluating the efficacy and safety of the LifeTech CeraFlexTM ASD occluder for transcatheter closure in patients with secundum ASD. Unique strengths of this trial compared to published studies (12,13,16) are a prospective multi-center study design and a sample size of more than 100 patients that underwent ASD closure with the LifeTech CeraFlexTM ASD occluder. The results of this intention to treat analysis demonstrate efficient and safe ASDO and extend these positive results to the first visit 6 months after ASDO. The primary endpoint (successful placement of the occluder and successful closure of the defect without severe complications) was reached by 102/103 patients and the secondary endpoint (clinical efficacy 6 months after the intervention) was reached by 94 patients.

Advantages/features

The advent of the Amplatzer Septal occluder (Abbott) in 1997 enabled catheter interventional closure of a secundum ASD in most patients. A recently published systematic review and meta-analysis, including 2,972 patients, reported safe use with adverse event rates for 1 in 20 patients (17). Studies documented significantly lower morbidity and shorter hospital stay after interventional ASD-closure with the Amplatzer ASD occluder compared to surgical ASD-closure (8,18). The Amplatzer ASD occluder is self-centering, easy to handle and fully retrievable until release. Even after device release the right atrial hub can be snared and the occluder can be removed safely through an adequately sized longsheath. However, the stiff connection of the Amplatzer ASD occluder to the delivery cable may cause difficult device placement, especially in patients with larger ASDs and deficient aortic septal rims. Just recently Abbott introduced a more flexible delivery cable (Treviso) (19), which improves the flexibility at device delivery. However, the effectivity of this modification still needs to be documented in larger patient series. The CeraFlexTM ASD occluder is comparable to the Amplatzer ASD occluder, as it is a nitinol double-disk device with PET membranes sewn into each disk. It is comparable to the Amplatzer ASD occluder with good outcomes and low complication rates (20). However, the 360° flexible rotation features between the device and delivery cable facilitate safe device implantation. Accurate positioning, even in difficult defects, is reported by Apostolopoulou et al. (12) and was also achieved in our study. Deployment without tension of the delivery catheter was confirmed by our study (13).

Correct device placement (primary endpoint of the study) on the first attempt was reached in 98 of 103 patients, the results being comparable with the use of the Occlutech Figulla Flex II Occluder (21). Both occluders have enhanced flexibility and a softer left atrial disc, due to the lack of a left atrial hub, resulting in a reduced risk of prolapse of the left atrial disc into the right atrium (22,23). Additionally, the missing left atrial hub may lower the risk for left atrial thrombus formation on the left disc of the device in the long-term.

A second implantation trial during the same procedure was necessary in five patients. One had an obvious residual shunt, one device could not be positioned properly. Repositioning of the same occluder was performed in one patient and one occluder was removed because of contact with the mitral valve. No additional maneuvers, such as deployment of the left disk in a pulmonary vein and additional balloon during deployment, were needed. Another potential advantage of the CeraFlexTM ASD occluder is the bioceramic TiN coating of the double disk, done to reduce nickel ion elution (14). Although sensitization to nickel is not a contra indication to ASDO with a nitinol device, increased levels of nickel were assessed in the blood of patients after ASDO with the Amplatzer device (24). Few case reports exist on the necessity to remove a nitinol device for suspected overreaction to nickel (25,26). Since at present there are no reports on nickel levels after ASD-closure with the CeraFlexTM ASD occluder, the advantage in this respect is speculative.

In 2/103 patients (2%) the device embolized. One device was seen in the left ventricle the day after the initial closure and an additional device was noticed in the aortic arch during an examination, not related to our study. Both patients were described in detail above. Device embolization occurs at a rate of 0.6–2% and usually, as in our cases, device retrieval is successful (22,27-29). Obviously in both patients the actual defect size was wrongly underestimated, although in both patients balloon sizing with the “stop flow” technique was employed. After successful retrieval it was tried in both patients to close the defect with a larger occluder. A 5 mm larger Occlutech device was not stable in the first patient and the defect was closed surgically, where a large defect (16×25 mm) was seen and closed with a patch. The other defect was successfully closed with a CeraFlexTM ASD occluder 5 mm larger than the first one.

Removing large CeraFlexTM occluders with a snare may be difficult since it may slip of the right atrial hub. Georgiev et al. reported on a viable method of interventional occluder removal using the Maslanka biopsy forceps assisting the snare (28). Left ventricle, abdominal aorta and femoral vessels are familiar embolization locations if the device embolizes to the systemic circulation (30,31). An overall embolization rate of 2% is rather high although in concordance with other study reports, accounting an embolization rate of 0.2–2.2% (21,22,28,29,32-35). Undersized device and large defects with either deficient or floppy septal tissues are also associated with these complications (15,36,37). It is assumed that embolizations are more related to anatomy than to specific devices (38). Nevertheless, operator-related technical issues like excessive tension on delivery cable during device deployment and malposition during the “push-pull” maneuver have to be taken into account (37).

Residual shunts

The presence of residual shunts after device implantation maybe caused by a second defect which previously was not diagnosed (22). Kaya et al. reported that residual shunting before discharge was present in 8 of 205 patients (4%) receiving Cera septal occluder, in 3 patients (1%) at 6-month FU and in 1 patient (1%) at 12-month FU (38). Decreasing shunts was confirmed by Haas et al. reporting that in 1,291 patients trivial residual shunting was present in 17% before discharge and in 4% at 6-month FU after ASD closure with the Occlutech occluder device. At 5-year FU a residual shunt was diagnosed solely in 2% (22). In 1 patient (1%) a moderate residual shunt was diagnosed not related to the occluder at the 6-month FU examination, while no residual shunt was apparent before discharge. This may be explained by a second ASD which had not been detected previously.

Arrhythmia and RBBB

Atrial arrhythmia is a frequent complication following transcatheter ASD closure (39). Occurrence of atrial tachyarrhythmia after ASD closure is associated with a prevalence of 2–4.8% (15,34,40). This data was confirmed by our study, since new-onset cardiac arrhythmia was detected in 4% (4 patients) with an average patient age of 64 years. Affected patients were diagnosed with new onset atrial flutter, paroxysmal atrial fibrillation, supraventricular extra systoles and palpitations, as well as ECG abnormalities in 24-hour ECG. Arrhythmia resolved after medical treatment in all patients. Particularly the occurrence of atrial fibrillation and flutter as well as ventricular extra systoles in this cohort is common. A relationship with the device is possible in all affected patients. In one patient hemodynamically relevant bradycardia occurred. This vasovagal reaction was handled with emergency treatment. A relationship with the procedure was considered possible. However, the symptoms completely resolved and the patient remained clinically stable.

In our study, iRBBB was diagnosed in 31 patients before secundum ASD closure. When comparing the QRS-time before the transcatheter treatment and at 6-month FU, the QRS-time was normalized in 14 patients. Moreover, no AV block occurred, confirming previously reported prevalence of advanced heart block as being below 1% (35,41,42). Careful monitoring of the development of conduction disturbance is recommended (43).

Study limitation

All three participating centers have 20 years and more experience with catheter interventional closure of secundum ASDs with self-centering double disc devices. Hence, no detailed implantation protocol was applied for this very standardized procedure. The implantations were conducted according to the CeraFlexTM ASD occluder instructions for use.

Conclusions

Interventional secundum ASD closure with the CeraFlexTM ASD occluder proved to be feasible, effective and safe. Prospective, randomized long-term studies with larger patient cohorts are needed to fully evaluate the advantages of this occluder comparing its results with other devices and capture possible long-term benefits.

The article’s supplementary files as