Percutaneous Device Closure of Patent Ductus Arteriosus in Adult Patients with 10-Year Follow-up.

OBJECTIVES: We report our 10-year experience with transcatheter closure of patent ductus arteriosus (PDA) in adult using different closure devices.
BACKGROUND: Transcatheter closure of PDA in adults can be challenging because of frequently associated comorbidities. Reports on immediate and intermediate-term results of PDA closure are excellent. This study aimed to provide the outcomes of PDA closure using different devices on long and very long term follow-up in adults.
MATERIALS AND METHODS: Between September 2009 and December 2018, data were retrospectively reviewed from 27 patients who underwent transcatheter closure of PDA. Outcome parameters were procedural success, procedure-related complications, evidence of residual shunt, and improvement in the signs/symptoms for which the procedure was performed. The mean follow-up interval was 72 months.
RESULTS: A device was successfully implanted in 27 of 27 patients (15 females). Median age and weight were 24 years (range: 18-57 years) and 69 kg (range: 53-102 kg), respectively. The mean PDA diameter was 4.1 ± 2.1 mm. Devices used were Amplatzer Duct Occluder (19/27), Occlutech Duct Occluder (6/27), and PFM Nit-Occlud (2/27). Doppler transthoracic echocardiography (TTE) demonstrated 92.6% of full occlusion at day 1, rising to 96.3% at 1 month. Three procedure-related complications occurred with no death. Among symptomatic 26 patients (96.3%), there was marked improvement in symptoms. Among 22 patients (81.5%) for whom the procedure was performed to address left ventricular (LV) enlargement, there was reduction or stabilization in LV size on serial TTEs.
CONCLUSIONS: Transcatheter closure of PDA in the adult patient appears to be safe and effective. Copyright:

INTRODUCTION

Ductus arteriosus is a vascular structure that connects the left pulmonary artery (LPA) near its origin to the descending aorta just after the left subclavian artery; it is an essential fetal structure that closes spontaneously in about 90% of full-term infants during the first 48 h of life. Patent ductus arteriosus (PDA) is considered a form of congenital heart disease (CHD), defined as a persistent patency beyond the 3rd month of life in term infants.

PDA accounts for 5%–10% of all CHDs. It can be associated with various other CHD.[1] The prevalence in adulthood is 0.05% and usually an isolated lesion.[2]

The presence of a hemodynamically significant PDA with left-to-right shunt may result in left ventricular (LV) volume overload with signs and symptoms of heart failure. In patients with evidence of LV volume overload, closure is needed to prevent complications such as LV dysfunction, arrhythmias, and pulmonary arterial hypertension (PAH). An additional reason for PDA closure is infective endocarditis to prevent infection.[34]

Surgical closure of PDAs was first performed by Gross and Hubbard in 1939 and has long been considered the gold standard treatment.[5] In most reports, surgical PDA closure allows a complete closure rate of 94%–100% with a 0–2% mortality rate.[6]

The most common complications of surgical PDA intervention include pneumothorax, bleeding, and recurrent laryngeal nerve injury. However, surgical closure in adults can be complicated by the presence of calcified ductus, aortic fragility due to atheromatous lesions, LV dysfunction, and PAH. These complications would make the operation more hazardous than when undertaken in the young patients.[7] Surgical closure remains the treatment of choice in the rare patients with a PDA too large for device closure or with unsuitable anatomy, such as aneurysmal PDA.

Transcatheter closure of PDA was started by Porstmann et al. in 1967 and became more widespread in 1976 after Rashkind and Cuaso developed a closure device.[89] Since then, multiple devices have been developed. The most common devices we use in our practice are the Amplatzer® Ductal Occluder (ADO, St. Jude Medical Inc., St. Paul, Minnesota, USA), the Occlutech® Duct Occluder (ODO, Occlutech, Helsingborg, Sweden), and the Nit-Occlud® PDA (PFM Medical, Cologne, Germany).

Transcatheter closure of PDA in adults can be challenging because of anatomical variations, associated findings, and complications such as PAH, LV dysfunction, infective endocarditis, calcification, and aneurysm formation. Most of the experience with transcatheter PDA closure has been in the pediatric population. Although some of the pediatric reports included small numbers of adults, contemporary data focusing on adults are limited. Much of the published data of PDA device closure in adults is focused on ADO.[10] However, there is a paucity of data on the use of other devices in adults.

In this study, we report our 10-year experience with transcatheter closure of PDA in 27 adult patients using various closure devices with focusing on safety and efficacy (long term) of device closure in adults with PDA.

MATERIALS AND METHODS

The study was conducted as a retrospective, nonrandomized review of 27 procedures performed on a total of 27 patients by a single operator. The procedures were performed from September 2009 to December 2018 at four medical centers. The patient population consisted of individuals who were referred by their primary cardiologists to a center specialized in the care of patients with CHD.

Data were gathered by means of retrospective chart review. All patients provided informed consent regarding the use of their patient data for retrospective analysis and anonymous reporting. The institutions' IRBs approved the study protocol.

The data collected included the following:

Demographics: procedure date, age, gender, height, weight, and body surface area (BSA)

Clinical data: symptoms, signs, medical history, and indication(s) for PDA closure

Electrocardiography: right ventricular hypertrophy/left ventricular hypertrophy/biventricular hypertrophy, atrioventricular block, right or left bundle branch block, other intraventricular conduction delay, as well as any other abnormal findings

PDA size and type: measured by transthoracic echocardiography (TTE), cardiac computed tomography, cardiac magnetic resonance imaging, and angiography

TTE: LV end-diastolic dimension (LVEDD), left atrial dimension, pulmonary artery size, LV systolic function, and presence and quantification of residual shunt(s)

Hemodynamics: Qp Qs ratio, systolic and mean pulmonary artery pressure, and pulmonary vascular resistance, Rp:Rs ratio

Procedural details: fluoroscopy time, device size, device used, and sheath sizes

Adverse events: details and outcome of any adverse event during the procedure or follow-up period.

Residual shunts were evaluated using color Doppler TTE. Grading was as follows: trivial (<1 mm); mild (1–2 mm); moderate (2–4 mm), or severe (>4 mm) based on the width of the color jet upon exit from the PDA. The classification is similar to that used to describe atrial level residual shunts after device closure of atrial septal defects (ASD) as described by Boutin et al.[11]

We used the American Society of Echocardiography guidelines for chamber size and valvular function assessments.[12] Furthermore, PAH was defined in accordance with the standards set forth by the World Health Organization fifth World Symposium on Pulmonary Hypertension.[13]

Patient characteristics

The patients ranged in age from 18 to 57 years with a median age of 24 years. Fifteen (56%) of the patients were female and the remaining 12 (44%) patients were male. BSA ranged from 1.5 to 2.7 m2 with a median of 1.8 m2. Most of PDAs were isolated not associated with other CHD.

Most patients were in normal sinus rhythm (96.3%) except one patient (3.7%) had atrial fibrillation. This patient was elderly and known to have longstanding systemic hypertension. Most of the patients were symptomatic (96.3%). One (3.7%) patient had a history of PDA-related infective endocarditis.

The majority of the patients had TTE evidence of LV volume overload (81.5%), but only a small percentage showed evidence of depressed LV function (15%). The median pulmonary artery systolic pressure was 35 mmHg (range: 24–65 mmHg). Table 1 summarizes the baseline patient characteristics as well as preprocedural TTE findings.

Table 1

Baseline patient characteristics and findings

Characteristic	Patient
Age, median (years)	24 (18-57)
Gender (%)
Female	15 (55.6)
Male	12 (44.4)
Height, median (cm)	164 (152-190)
Weight, median (kg)	69 (53-102)
BSA, median (m²)	1.8 (1.5-2.7)
Associated abnormalities (%)
Congenital
ASD	2 (3.7)
Persistent left SVC to CS	1 (3.7)
Bicuspid aortic valve	1 (3.7)
Coronary artery fistula (small)	1 (3.7)
Acquired
Coronary artery disease	3 (11)
Atrial fibrillation	1 (3.7)
Moderate aortic valve stenosis	1 (3.7)
Moderate mitral regurgitation	1 (3.7)
Systemic illness (%)
Systemic hypertension	2 (7.4)
Diabetes	2 (7.4)
Hypothyroidism	1 (3.7)
Bronchial asthma	1 (3.7)
ECG (%)
Normal sinus rhythm	26 (96.3)
Atrial fibrillation	1 (3.7)
LVH/dilatation	6 (22)
LBBB/RBBB/IVCD	3 (11)
Nonspecific ST-T changes	2 (7.4)
TTE (%)
Normal LV function	23 (85)
Depressed LV function	4 (15)
LV dilation	22 (81.5)
Right ventricular systolic pressure, median (mmHg)	35 (24-65)

BSA: Body surface area, ASD: Atrial septal defect, SVC: Superior vena cava, CS: Coronary sinus, ECG: Electrocardiography, LVH: Left ventricular hypertrophy, LBBB: Left bundle branch block, RBBB: Right bundle branch block, IVCD: Intraventricular conduction delay, TTE: Transthoracic echocardiography, LV: Left ventricular

Patent ductus arteriosus

According to the PDA classification adopted by Krichenko et al.,[14] 81.5% had Type A, 11.1% had Type E, and 7.4% had Type C. The mean PDA diameter on angiography was 4.1 ± 2.1 mm, and the mean length was 8.3 ± 3.1 mm. All PDAs were restrictive. The median shunt fraction (Qp/Qs) was 1.8:1 with a range from 1.3 to 2.4:1. Table 2 summarizes the characteristics of the PDAs.

Table 2

Patent ductus arteriosus characteristics (angiography)

Angiography	Patients
PDA type (%)
Type A	22 (81.5)
Type E	3 (11.1)
Type C	2 (7.4)
PDA size, mean±SD (mm)	4.1±2.1
PDA length, mean±SD (mm)	8.3±3.1
Hemodynamic data (median)
Qp: Qs	1.8 (1.3-2.4:1)
PVR (wood unit)	1.7 (1.3-4.9)
PASP (mmHg)	35 (24- 64)
Mean PA pressure (mmHg)	22 (16-36)

PDA: Patent ductus arteriosus, SD: Standard deviation, PVR: Pulmonary vascular resistance, PASP: Pulmonary artery systolic pressure

Indications

The indications for PDA closure included signs and symptoms attributable to the defect such as dyspnea on exertion and fatigue. Other indications included the presence of a hemodynamically significant shunt as evidenced by otherwise unexplained LV enlargement or a shunt fraction (Qp/Qs) ≥1.5 by cardiac catheterization. Otherwise unexplained deterioration in LV function, recurrent endocarditis, and PAH also constituted indications for PDA closure.[1516]

In the majority of cases, symptoms of dyspnea on exertion and/or unexplained LV enlargement constituted the indications for closure. Table 3 summarizes the indications for PDA closure. It should be noted that many patients had multiple indications for PDA closure.

Table 3

Indications for patent ductus arteriosus closure

Indications	Patients
Attributable symptoms or signs (%)
Dyspnea on exertion, fatigue	26 (96.3)
Significant left-to-right shunt (%)
Unexplained LV enlargement	22 (81.5)
Unexplained deterioration of LV function (%)	4 (15)
Recurrent endocarditis (%)	1 (3.7)
Pulmonary hypertension with (%)
PASP <50% of systemic and/or PVR <1/3 systemic	9 (33.3)
Net L-R shunt and PASP 50% or greater systemic, and/or PVR >1/3 systemic

PVR: Pulmonary vascular resistance, PASP: Pulmonary artery systolic pressure, LV: Left ventricular

Procedure

Most of the procedures were performed under general anesthesia (74%) while the remaining procedures were done under conscious sedation (26%). Most of the patient requested general anesthesia. A single dose of intravenous (IV) antibiotic was administered 30 min before the procedure. Aspirin administered at least 24 h before the procedure. Femoral arterial and venous access were obtained for all patients following which they received IV heparin.

After hemodynamic measurements, aortic angiogram was performed in lateral and anteroposterior views to visualize the ductus. Antegrade approach was used to cross the PDA in most patients while snare-assisted retrograde approach was utilized in two patients. The choice of the device was depending on the PDA diameter and length.

Devices were deployed through venous approach in all the patients. In most cases, the device was successfully deployed in a single attempt. In two instances, two deployment attempts were needed and in one instance, three deployment attempts were necessary; in this particular case, the device type was changed from ADO to ODO (long type) in the third attempt.[17] In one case, the tip of the dilator passed the PDA but the rest of the delivery catheter stuck within the PDA due to protruded calcium near the aortic side. In this particular case, we pulled the assembly to main pulmonary artery, and we predilated the PDA with 3 mm × 8 mm coronary noncomplaint balloon through another femoral artery access using coronary wire and right Judkins catheter, then the delivery catheter advanced gently.

Additional interventional procedures were performed during the same cardiac catheterization in two patients (ASD closure device in one patient and coronary artery stenting in one patient). Table 4 summarizes the procedural data. Postprocedure, patients were maintained on aspirin or equivalent antiplatelet therapy for 6 months.

Table 4

Procedural data

Variable	Patients
General anesthesia (%)	20 (74)
Conscious sedation (%)	7 (26)
Fluoroscopy time, median (min)	14 (11-45)
Procedure time, median (min)	52 (31-90)
Approach for crossing (%)
Antegrade	25 (92.5)
Retrograde	2 (7.5)
Delivery through femoral vein (%)	27 (100)
Deployment attempts (%)
One	24 (88.9)
Two	2 (7.4)
Three	1 (3.7)
Device (%)
ADO	19 (70.4)
ODO	6 (22.2)
Nit-Occlud	2 (7.4)
Combined procedures addressing associated abnormalities (%)
ASD closure	1 (3.7)
Coronary PCI	1 (3.7)

ADO: Amplatzer® ductal occlude, ODO: Occlutech® duct occlude, ASD: Atrial septal defect, PCI: Percutaneous coronary intervention

RESULTS

Measured outcome parameters were procedural success, procedure/device-related complications, evidence of residual shunt by TTE, and improvement in the signs/symptoms for which the procedure was performed. Procedural success was defined by device release in appropriate position without embolization. Procedure/device-related complications were determined by chart review.

The presence of residual shunt was assessed by color Doppler TTE. Furthermore, improvement in signs/symptoms for which the intervention was performed was determined by history and serial TTE during follow-up visits. Follow-up occurred at 1, 30, 60, 180, and 360 days with yearly follow-up thereafter with the longest follow-up being 10 years for the first few patients. The mean follow-up was 72 months.

In all the 27 patients, the device was successfully deployed with appropriate position confirmed both by angiography. Device success was achieved with the first chosen device in all except one patient who needed a longer device (the ODO long version). No device embolization occurred. Furthermore, there was no LPA or aorta stenosis-one patient developed moderate size hematoma at calcified access site that was treated conservatively. In one patient with significant aortic calcification, there was distal cholesterol embolization. It was complicated by acute renal failure requiring no dialysis; there was also distal limbs ischemia and skin discoloration. This was treated conservatively by IV heparin infusion. In 1 week, the renal function and limb ischemia improved to the baseline statistics.

A third complication was recurrent symptomatic nonsustained ventricular arrhythmia during the procedure in a patient with severe LV dysfunction. This patient had ICD implantation before discharge. Ultimately, no long-term complications were noted. Table 5 summarizes the procedure-related complications. Immediate complete occlusion was achieved in 60% on angiography. Doppler TTE demonstrated 92.6% of full occlusion on day 1, rising to 96.3% at 1 month and 100% at 6 months. Figure 1 summarizes the data with respect to the presence of a residual shunt.

Table 5

Adverse events during and after device closure of patent ductus arteriosus

Events	Patients
Death (%)	0 (0)
Device embolization (%)	0 (0)
LPA or Dao stenosis (%)	0 (0)
Arrhythmias (%)	1 (3.7)
Distal cholesterol embolization (%)	1 (3.7)
Moderate access site hematoma not requiring blood transfusion (%)	1 (3.7)
Overall (%)	11

LPA: Left pulmonary artery

Figure 1

Echocardiographic shunt status (percentage of patients) pre and postpatent ductus arteriosus closure. At day 1, there were two residual shunts (1 mild and 1 trivial). At 30 days, there was one trivial shunt. At 6 months, there was no residual shunt

For those 26 (96.3%) patients for whom the procedure was performed to address dyspnea on exertion, all but two noted marked improvement in their symptoms from the New York Heart Association (NYHA) class 2–4 to NYHA class 1. The remaining two noted mild-to-moderate improvement in their shortness of breath from NYHA class 3–4 to NYHA class 2. Of note, one of the two patients suffered from ischemic cardiomyopathy. The other one suffered from bronchial asthma.

For those patients for whom the procedure was performed to address unexplained LV enlargement, all experienced reduction or stabilization in LV size on serial TTE. A total of 22 (81.5%) patients had LV enlargement. Eleven female patients had mild-to-moderate LV enlargement before the procedure (seven patients had mild enlargement with LVEDD of 5.4–5.6 cm with a median of 5.5 cm and four patients had moderate LV dilation with LVEDD of 5.8–6.1 cm with a median of 5.9 cm). On follow-up, the LVEDD normalized in ten patients (3.9–5.2 cm) with a median of 4.9 cm.

Most of the improvement occurred in the first 3–6 months, and it was sustained during the follow-up period. Only one patient's LVEDD did not improve, instead remaining stable over 4-year follow-up. This patient had hypertensive cardiomyopathy in addition to her PDA at baseline. Among male patients, 11 had evidence of LV enlargement before PDA closure. Eight of the 11 men had mild LV dilation with LVEDD ranging from 6.0 to 6.2 cm (median 6.1 cm). The remaining three men displayed moderate LV dilation with LVEDD ranging from 6.6 to 6.7 cm.

On follow-up, nine patients experienced normalization of LVEDD (5.0–5.5 cm) with a median of 5.3 cm. Again, most of the improvement occurred in the first 3–6 months following PDA closure and was sustained during follow-up. Only two male patients failed to have normalization of the LVEDD although one LVEDD improved from moderate to mild (6.6–6.3 cm) following PDA closure. Of note, the two patients suffered from a concomitant ischemic cardiomyopathy. Three of the four patients with deterioration of LV function saw marked improvement in LV function following PDA closure.

Among those patients with LV dysfunction, the preprocedure LV ejection fraction (LVEF) ranged from 30% to 45%. Following PDA closure, three of the four experienced normalization of the LVEF ≥55%, which was sustained during follow-up. The fourth patient with LV dysfunction showed only mild improvement in LV function with the EF improving from 30% to 40%. However, it should be noted that those patients suffered from concomitant ischemic cardiomyopathy.

For those patients for whom the procedure was done to address recurrent endocarditis, no further episodes of endocarditis occurred. Regarding the pulmonary artery pressure, the overall right ventricular systolic pressure by TTE improved from a mean of 44 mmHg (40 mmHg invasively) before the PDA closure to 32 mmHg after the PDA closure. Of note, the pulmonary pressure was improved uniformly in all patients including the one with severe pulmonary hypertension. Figure 2 summarizes the follow-up data regarding signs and symptoms.

Figure 2

Percentage of patients on follow-up who experienced improvement in dyspnea on exertion, left ventricular size, left ventricular function, freedom from endocarditis

DISCUSSION

PDAs in adult represent special cardiac condition. The impact of long-standing PDA-related shunt on adults with probably acquired cardiomyopathies or systemic illness is different from the pediatric population. The LV is less compliant (stiffer) in adult with/or without cardiomyopathy compared to pediatric.[4] This probably explains why most of the patients were symptomatic throughout the range of shunt degree. This series reports cohort of adult patients with PDAs who underwent percutaneous closure with various devices and very long follow-up.

The results indicate that transcatheter PDA closure effectively reduces left-to-right shunting. The success in eliminating hemodynamically significant shunts appears to have produced stabilization or improvement in the signs and symptoms for which PDA closure was undertaken. All patients with dyspnea experienced improvement in their shortness of breath. The vast majority noted marked improvement in their shortness of breath while two patients with preexisting unrelated cardiomyopathies experienced mild-to-moderate improvement in their shortness of breath. Furthermore, all patients experienced stabilization in or reduction in their LVEDD on serial TTE following PDA closure. In fact, among those patients with impaired LV function, three of four experienced improvements in the LVEF. Transcatheter PDA closure, therefore, appears to ameliorate the pathologic changes that follow from hemodynamically significant left-to-right shunting.

The clinical benefits from PDA closure were maintained through the follow-up. The complication associated with percutaneous PDA closure in adult clearly related to the presence of atherosclerotic changes at the aorta and the access site. Furthermore, the LV dysfunction put them at risk of arrhythmias during the stress of the procedure. The complication rate was 11%.

Several reports on the safety and efficacy of percutaneous PDA closures have been published.[18192021] Most of the reports focused on the pediatric population with limited number of adults included, and the follow-up duration was short or medium term at most.

Different reports focused on different devices.[2223] Recently, a group published a retrospective review of percutaneous PDA closure in 70 patients; of this group 37 were adults.[24] Different closure devices were used including coils, ADO, and other devices.

Devices were successfully deployed in all the patients. At 24-h postprocedure, the success rate of the transcatheter intervention was 95.7%. At 6-month follow-up, no residual shunt was observed in all the patients. The mean follow-up duration was 531 days (range: 11–2059 days). No major procedure-related complications reported.

The present study does have a number of limitations. It is a retrospective review of a nonrandomized population. Furthermore, the data are somewhat incomplete given that the duration of follow-up varied widely among patients. The small sample size also precludes drawing larger conclusions from the data obtained.

CONCLUSION

Based on this series of 27 adult patients who underwent percutaneous PDA closure for indications ranging from dyspnea to LV dilatation, percutaneous PDA closure appears to be safe and effective. No long-term procedural complications were encountered. Furthermore, no significant residual shunt remained with respect to all closed defects.

Patients also noted marked symptomatic improvement with TTE confirmation of reduction in or stabilization of LV size and improvement in LV function. The percutaneous approach showed excellent immediate and long-term results and fewer complications.

Percutaneous closure should, therefore, be the preferred method of closure for most anatomically suitable PDAs.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.