Closure Devices for Patent Foramen Ovale and Atrial Septal Defects
Transcatheter closure of secundum atrial septal defects (ASD) may be considered medically necessary when using a device that has been FDA approved for that purpose and used according to the labeled indications.
At present, no PFO closure devices are FDA approved for patients with cryptogenic stroke. All uses of these PFO closure devices are currently off-label.
Closure of atrial septal defects (ASD)
There are 2 FDA-approved devices for ASD closure:
The labeled indications for these devices are similar and include:
Generally recognized indications for ASD closure include a pulmonary-to-systemic flow ratio of greater than 1.5, right atrial and right ventricular enlargement, and paradoxical embolism.
Patent foramen ovale (PFO) and atrial septal defects (ASDs) are relatively common congenital heart defects that can be associated with a range of symptoms. Depending on their size, ASDs may lead to left-to-right shunting and signs and symptoms of pulmonary overload. PFOs may be asymptomatic but have been associated with higher rates of cryptogenic stroke. “Closure” devices are intended as less invasive, catheter-based approaches of repairing PFO or ASDs. These devices are alternatives to open surgical repair for ASDs or treatment with antiplatelet and/or anticoagulant medications in patients with cryptogenic stroke and a PFO.
Patent Foramen Ovale
The foramen ovale, a component of fetal cardiovascular circulation, consists of a communication between the right and left atrium that functions as a vascular bypass of the uninflated lungs. The ductus arteriosus is another feature of the fetal cardiovascular circulation, consisting of a connection between the pulmonary artery and the distal aorta. Before birth, the foramen ovale is held open by the large flow of blood into the left atrium from the inferior vena cava. Over a course of months after birth, an increase in left atrial pressure and a decrease in right atrial pressure result in the permanent closure of the foramen ovale in most individuals. However, a PFO is a common finding in normal adults, detected in up to 25% of adults. (1) In some epidemiologic studies, PFO has been associated with cryptogenic stroke, a type of stroke defined as an ischemic stroke occurring in the absence of potential cardiac, pulmonary, vascular, or neurologic sources. Studies also show an association of PFO and migraine headache. There has been interest in either open surgery or transcatheter approaches to close the PFO in patients with a history of cryptogenic stroke to prevent recurrent stroke.
Atrial Septal Defects
In contrast to PFO, which represents the postnatal persistence of normal fetal cardiovascular physiology, ASDs represent an abnormality in the development of the heart that results in free communication between the atria. ASDs are categorized according to their anatomy. Ostium secundum describes defects that are located midseptally and are typically near the fossa ovalis. Ostium primum defects lie immediately adjacent to the atrioventricular valves and are within the spectrum of atrioventricular septal defects. Primum defects occur commonly in patients with Down syndrome. Sinus venous defects occur high in the atrial septum and are frequently associated with anomalies of the pulmonary veins. Ostium secundum ASDs are the third most common form of congenital heart disorder and one of the most common congenital cardiac malformations in adults, accounting for 30% to 40% of these patients older than age 40 years. The ASD often goes unnoticed for decades because the physical signs are subtle and the clinical sequelae are mild. However, virtually all patients who survive into their sixth decade are symptomatic; fewer than 50% of patients survive beyond age 40 to 50 years due to heart failure or pulmonary hypertension related to the left-to-right shunt. Symptoms related to ASD depend on the size of the defect and the relative diastolic filling properties of the left and right ventricles. Reduced left ventricular compliance and mitral stenosis will increase left-to-right shunting across the defect. Conditions that reduce right ventricular compliance and tricuspid stenosis will reduce left-to-right shunting or cause a right-to-left shunt. Symptoms of an ASD include exercise intolerance and dyspnea, atrial fibrillation, and, less commonly, signs of right heart failure. Patients with ASDs are also at risk for paradoxical emboli.
Repair of ASDs is recommended for those with a pulmonary to systemic flow ratio (Qp:Qs) exceeding 1.5:1.0. Despite the success of operative repair, there has been interest in developing a catheter-based approach to ASD repair to avoid the risks and morbidity of open heart surgery. A variety of devices have been researched over the past 20 years; technical challenges include minimizing the size of device so that smaller catheters can be used, developing techniques to properly center the device across the ASD, and ensuring that the device can be easily retrieved or repositioned, if necessary.
Several devices have been developed to treat PFO via a transcatheter approach, including the CardioSEAL® STARFlex™ Septal Occlusion System (NMT Medical) and the Amplatzer® PFO Occluder (Amplatzer Inc., now St. Jude Medical, St. Paul, MN). The STARFlex system is no longer manufactured. Transcathether PFO occluders consist of a single or paired wire mesh discs that are covered or filled with polyester or polymer fabric that are placed over the septal defect. Over time, the occlusion system is epithelialized.
ASD occluder devices consist of flexible mesh disks that are passed via catheter to cover the ASD.
PFO Closure Devices
In 2002, two transcatheter devices received approval for marketing from FDA as a treatment for patients with cryptogenic stroke and PFO: the CardioSEAL® Septal Occlusion System (no longer commercially available) and the Amplatzer® PFO Occluder. Both received approval by FDA through a Humanitarian Device Exemption (HDE), a category of FDA approval that is applicable to devices that are designed to treat a patient population of fewer than 4000 patients per year. This approval process requires the manufacturer to submit data on the safety and the probable clinical benefit. Clinical trials validating the device effectiveness are not required. The labeled indications of both limited the use of these devices to closure of PFO in patients with recurrent cryptogenic stroke due to presumed paradoxical embolism through a PFO and who have failed conventional drug therapy.
Following this limited FDA approval, the use of PFO closure devices increased by more than 50-fold, well in excess of the 4000 per year threshold intended under the HDE. (2) As a result, in 2006, FDA withdrew the HDE approval for these devices. At this time, FDA also reiterated the importance of RCTs of PFO closure devices versus medical therapy but noted that ongoing trials were hampered by slow enrollment. Withdrawal of the HDE approval was, in part, intended to spur greater enrollment in ongoing RCTs of these devices.2 Currently, all uses of closure devices to treat PFO are off-label uses.
ASD Closure Devices
At present, 2 devices are FDA-approved for ASD closure: the AMPLATZER™ Septal Occluder (St. Jude Medical, Minneapolis, MN), and the GORE HELEX™ Septal Occluder (W.L. Gore & Associates Inc., Newark, DE). The Amplatzer Septal Occluder was approved through FDA’s premarket approval process in June 2002. The Amplatzer device is intended for the occlusion of ASDs in the secundum position or in patients who have undergone a fenestrated Fontan procedure who require closure of the fenestration. Patients indicated for ASD closure have echocardiographic evidence of ostium secundum ASD and clinical evidence of right ventricular volume overload.
The Gore Helix Septal Occluder was approved through the premarket approval process in August 2006. It is indicated for the percutaneous, transcatheter closure of ostium secundum ASDs.
FDA product code: MLV.
Medical policies are systematically developed guidelines that serve as a resource for Company staff when determining coverage for specific medical procedures, drugs or devices. Coverage for medical services is subject to the limits and conditions of the member benefit plan. Members and their providers should consult the member benefit booklet or contact a customer service representative to determine whether there are any benefit limitations applicable to this service or supply. This medical policy does not apply to Medicare Advantage.
In 1996, the FDA created a new category of approval for humanitarian use devices (HDE). A HDE is a device that is intended to benefit patients by treating or diagnosing a disease or condition that affects fewer than 4,000 individuals in the United States per year. An approved HDE in2002 Authorized marketing of the CardioSeal® device and the Amplatzer® Patent Foramen Ovale Occluder for treatment of patient foramen ovale. However, following a large increase in use of these devices that exceeded the HDE threshold of 4,000 per year, the HDE approval was withdrawn in 2006. Currently, all use of PFO closure devices is off-label use.
FDA approval continues for 2 closure devices for ASD which include the AMPLATZER® Septal Occluder, and the GORE HELEX® Septal Occluder.
Transcathether Device Closure of Patent Foramen Ovale
Patent Foramen Ovale Closure for Thromboembolism Prevention
Conventional therapy for cryptogenic stroke consists of either antiplatelet therapy (aspirin, clopidogrel, or dipyridamole given alone or in combination) or oral anticoagulation with warfarin. In general, patients with a known clotting disorder or evidence of pre-existing thromboembolism are treated with warfarin, and patients without these risk factors are treated with antiplatelet agents. Closure devices are non-pharmacologic alternatives to medical therapy for cryptogenic stroke in patients with a patent foramen ovale (PFO).
Evidence on the efficacy of PFO closure devices consists of three randomized controlled trials (RCTs), a few nonrandomized, comparative studies, and numerous case-series. Meta-analyses of the published studies have also been performed.
Randomized Controlled Trials
CLOSURE I trial
The Evaluation of the STARflex Septal Closure System in Patients with a Stroke and/or Transient Ischemic Attack due to Presumed Paradoxical Embolism through a Patent Foramen Ovale (CLOSURE I) study was a multicenter, randomized, open-label trial of percutaneous closure versus medical therapy. (3) A total of 909 patients between the ages of 18 and 60 years, with cryptogenic stroke or transient ischemic attack (TIA) and a PFO were enrolled. Patients in the closure group received treatment with the STARflex device and also received anti-platelet therapy. Patients in the medical therapy group were treated with aspirin, warfarin, or both at the discretion of the treating physician. The primary endpoint was a composite of stroke/TIA at 2 years, death from any cause during the first 30 days after treatment, and death from neurologic causes at 2 years.
Of 405 patients in the closure group, 362 (89.4%) had successful implantation without procedural complications. At 6 months, echocardiography revealed effective closure in 315 of 366 patients (86.1%). The composite primary outcome was reached by 5.5% of patients in the closure group and 6.8% of patients in the medical therapy group (adjusted hazard ratio [HR], 0.78; 95% confidence interval [CI], 0.45 to 1.35; p=0.37). Kaplan-Meier estimates of the 2-year rate of stroke were 2.9% in the closure group and 3.1% in the medical therapy group (adjusted HR=0.90; 95% CI: 0.41 to 1.98). Serious adverse events were reported by16.9% of patients in the closure group versus 16.6% in the medical group. Adverse events that were increased in the closure group included vascular procedural complications (3.2% vs. 0, p<0.001) and atrial fibrillation (5.7% vs. 0.7%, p<0.001).
The RESPECT trial was a multicenter RCT comparing PFO closure with medical therapy in 980 patients between the ages of 18 and 60 years with a previous cryptogenic stroke and documented PFO. (4) The RESPECT trial was a multicenter RCT comparing PFO closure with medical therapy in 980 patients between ages 18 and 60 years with a previous cryptogenic stroke and documented PFO. Patients were randomly assigned to PFO closure with the Amplatzer Occluder, or to medical therapy. Medical therapy consisted of 1 of 4 regimens prescribed at the discretion of the treating physician: aspirin, aspirin plus dipyridamole, clopidogrel, or warfarin. The primary endpoint was a composite of fatal ischemic stroke, nonfatal ischemic stroke, or early death within 30 days of randomization. Mean follow-up for the entire group was 2.6±2.0 years.
A total of 9 events occurred in 499 patients assigned to closure, and 16 events occurred in 464 patients assigned to medical therapy. All of the events were non-fatal strokes. The HR for this outcome was 0.49, but this result did not reach statistical significance in the intention-to-treat (ITT) analysis (95% CI: 0.22 to 1.11; p=0.08). On per-protocol analysis, there was a statistically significant effect, with a HR of 0.37 (95% CI: 0.14 to 0.96; p=0.03). On subgroup analyses, there were no statistically significant differences in outcomes, although there were trends for better outcomes in the closure group for patients with a substantial right -to -left shunt (p=0.07) and for patients with an atrial septal aneurysm (p=0.10). The rate of serious adverse events did not differ between the closure and medical therapy groups (23.0% vs. 21.6%, p=0.65). Major bleeding (n=2) and cardiac tamponade (n=2) were the most frequent procedure-related adverse events.
The PC trial was a multicenter RCT comparing PFO closure with medical therapy in 414 patients younger than 60 years of age with a prior cryptogenic stroke or peripheral embolization and a documented PFO. (5) Patients were recruited from 29 centers worldwide and randomly assigned to PFO closure with the Amplatzer device or medical therapy. Recommended antiplatelet therapy in the closure group was aspirin plus ticlopidine, or clopidogrel alone. Medical therapy in the control group was at the discretion of the treating physician, with the requirement that patients receive at least one appropriate medication. The primary endpoint was a composite of death, nonfatal stroke, TIA, or peripheral embolism. The median duration of follow-up was 4.1 years in the closure group and 4.0 years in the medical therapy group.
The primary outcome, after independent adjudication, occurred in 9 of 204 patients (3.4%) in the closure group compared with 11 of 210 patients (5.7%) in the medical group. The HR for this outcome was 0.63 (95% CI: 0.24 to 1.62; p=0.34) on ITT analysis. On per-protocol analysis, results were similar with an HR of 0.70 (95% CI: 0.27 to 1.85; p=0.48). There were no significant differences in the rate of the individual components of the primary outcome, and there were no significant differences in outcome on subgroup analyses. The adverse event rate was 34.8% in the closure group compared with 29.5% in the medical therapy group.
A large number of systematic reviews with meta-analysis of the 3 available RCTs have been published; several representative studies are summarized here. Rengifo-Moreno et al. (6) performed a combined analysis of the 3 RCTs previously discussed. The analysis included a total of 1,150 patients randomized to PFO closure and 1,153 patients randomized to medical therapy followed for a mean of 3.5 years. Two endpoints were included, recurrent vascular events and a combined endpoint of death plus recurrent vascular events. On combined analysis, there was a statistically significant reduction in recurrent vascular events with a pooled HR of 0.59 (95% CI: 0.36 to 0.97; p=0.04). For the composite outcome of death plus recurrent vascular events, combined analysis revealed a reduction for the closure group of borderline statistical significance (HR=0.67; 95% CI: 0.12 to 1.03; p=0.05). On subgroup analysis, there was a trend for greater benefit in patients with a substantial right -to -left shunt, although this result did not reach statistical significance (HR=0.35; 95% CI: 0.12 to 1.03; p=0.06).
Another meta-analysis of the same 3 RCTs was reported by Kitsios et al. (7) This study used recurrent stroke as the primary outcome. The authors noted that the rates of recurrent stroke varied widely across the studies, thereby raising the possibility of ascertainment bias for this outcome. On combined analysis, the difference between groups did not reach statistical significance, with an HR of 0.55 (95% CI: 0.26 to 1.18). Combined analysis was also performed for the composite outcomes reported in the trials, even though the composite outcomes were not defined in the same way. The combined result for the composite outcome was of borderline statistical significance, with an HR of 0.67 (95% CI: 0.44 to 1.00). There were no significant differences found on combined analysis of the subgroup analyses from the trials.
Meta-analyses of the same 3 RCTs were reported by Chen et al., (8) Hakeem et al., (9) Khan et al., (10) Kwong et al., (11) Nagaraja et al., (12) Ntaios et al., (13) Pandit et al., (14) and Pineda et al. (15) Results from these meta-analyses generally supported findings from previous meta-analyses. For the primary outcome of recurrent stroke or TIA, Chen et al found a pooled risk ratio with PFO device closure for recurrent stroke or TIA of 0.70 (95% CI: 0.47 to 1.04; p=0.08). Hakeem et al reported a pooled risk ratio for a composite outcome of death or recurrent stroke or TIA of 0.71 (95% CI: 0.48 to 1.06; p=0.09). Neither the Chen et al nor the Hekeem et al meta-analyses found significant differences between PFO device closure and medical management for the risk of death or adverse events. Khan et al reported pooled analyses for the primary outcome of recurrent stroke, with a pooled effect-estimated HR for the primary outcome of recurrent stroke in patients treated with PFO device closure compared with medical management of 0.67 (95% CI, 0.44 to 1.00). In analysis of only the RESPECT and PC trials, which used the Amplatzer PFO occluder device, the HR for recurrent stroke in patients treated with PFO device closure was 0.54 (95% CI: 0.29 to 1.01). Similarly, Pandit et al reported sensitivity analyses including only the RESPECT and PC trials, and found that patients who received the Amplatzer PFO occluder device had a lower risk of recurrent strokes compared with medical therapy (HR=0.44; 95% CI: 0.21 to 0.94; p=0.03). In addition to pooled estimates for the risk of the primary outcomes of recurrent stroke, TIA, or death, Kwong et al reported pooled outcomes for risk of new-onset atrial fibrillation and found that PFO closure was associated with a significantly higher incidence of new-onset atrial fibrillation compared with medical therapy (OR=3.77; 95% CI: 1.44 to 9.87; p=0.007). In the analysis by Ntaios et al, the risk of new-onset atrial fibrillation with device closure compared with medical therapy was not higher in patients who received the Amplatzer PFO Occluder device (1.28% vs. 0.72%; OR=1.81; 95% CI: 0.60 to 5.42), but was higher in patients who received the STARFlex device (5.14% vs. 0.64%; OR=8.30; 95% CI: 1.44 to 9.87).
Cappodano et al. published an updated systematic review and meta-analysis of studies that compared outcomes associated with medical management or PFO closure among patients with cryptogenic stroke. (16) This analysis included the 3 RCTs previously described, along with 11 observational studies. In the randomized trials, PFO closure was not associated with significantly lower rates of stroke than medical therapy (HR=0.62; 95% CI, 0.34 to 1.11; p=0.10) or with lower rates of TIA (HR=0.77; 95% CI: 0.46 to 1.32; p=0.34). When the analysis was restricted to the RESPECT and PC trials, which used the Amplatzer PFO occluder device, PFO closure was significantly associated with lower recurrent stroke risk (HR=0.44; 95% CI, 0.20 to 0.95; p=0.04). In the observational studies, which included 2231 patients, PFO closure was significantly associated with lower rates of stroke than medical therapy (HR=0.23; 95% CI: 0.11 to 0.49; p<0.01).
Similarly, Wolfrum et al. conducted a systematic review and meta-analysis of controlled trials that compared outcomes for PFO closure with medical management among patients with cryptogenic stroke, including 3 RCTs and 11 nonrandomized studies. (17) Again, among the RCTs, there was no significant improvement in stroke risk with PFO closure compared with medical management. However, among the non-RCT studies, PFO closure was associated with a reduced risk of stroke (RR=0.37; 95% CI: 0.20 to 0.67; p<0.001). In a time-to-stroke analysis that included 3 RCTs and 2 non-RCTs that had multivariable adjustments, PFO closure was associated with a borderline significant stroke risk reduction compared with medical therapy (HR=0.58; 95% CI: 0.33 to 0.99; p=0.047.)
A number of systematic reviews of the observational studies have also been published, comparing outcomes of PFO closure with medical therapy. (18-20) Similar to the findings reported by Cappodano et al., These reviews are consistent in reporting that the combined rate of recurrent stroke is lower for patients treated with a closure device compared with medical therapy.
Kitsios et al. published a systematic review of observational studies and the single RCT in 2012. (19) This review included 52 single-arm studies, 7 non-randomized comparative studies, and 1 RCT. The combined incident rate for recurrent stroke was lower for patients treated with PFO (0.36 events/100 patient-years; 95% CI: 0.24 to 0.56) compared with patients treated medically (2.53 events/100 patient-years; 95% CI: 1.91 to 3.35). The incident rate ratio was 0.19 (95% CI: 0.18 to 0.98), which indicated an approximately 80% reduction in the rate of strokes for the closure group. This systematic review noted that the incident rate for recurrent strokes in patients treated with closure devices was much lower in the RCT compared with the observational studies, while the incident rate for recurrent stroke in patients treated medically was only slightly lower in the RCT compared with observational studies. This finding raises the possibility that ascertainment bias in the observational studies may have led to a spuriously low rate of recurrent stroke reported for patients treated with PFO closure.
Wohrle (20) compared the results of 12 series of PFO closure (n=2,016) with 8 series (n=998 patients) of medical therapy. At 2-year follow-up, the range of recurrent stroke was 0% to 1.6% for PFO closure and 1.8% to 9.0% for medical therapy. The combined annual incidence of stroke or TIA was 1.3% (95% CI: 1.0% to 1.8%) following PFO closure compared with 5.2% (95% CI: 4.4–6.2) for medical therapy. In an earlier review, Khairy et al. (18) analyzed 6 series of medical therapy (n=895 patients) and 10 series of PFO closure (n=1,355 patients). These authors noted differences in key clinical characteristics among patients in the 2 treatment groups. Patients treated with medical therapy were older, had a greater proportion of men, and higher rates of smoking and diabetes. Patients treated with PFO closure were more likely to have had more than one cerebrovascular event. The recurrence rate at 1 year ranged from 0% to 4.9% with PFO closure, compared with 3.8% to 12.0% with medical therapy. There was an estimated major complication rate (death, hemorrhage requiring transfusion, tamponade, need for surgical intervention, pulmonary embolus) for PFO closure of 1.5%, and a minor complication rate of 7.9%.
Abaci et al conducted a systematic review and meta-analysis of studies of both PFO and ASD device closure procedures. (21) The authors reviewed 203 articles, 111 of which reported ASD closure, 61 of which reported PFO closure, and 31 of which reported both ASD and PFO closures. Among patients undergoing PFO closure, the pooled rate of major complications was 1.1% (95% CI: 0.9% to 1.3%), most commonly device embolization requiring surgery.
Nonrandomized Comparative Studies
A number of nonrandomized comparative studies of closure devices versus medical therapy have been published. Wahl et al. performed a non-randomized comparative study using propensity matching in 308 consecutive patients with stroke or TIA that was presumed due to a PFO. (22) A total of 103 pairs of matched patients were compared on the primary composite outcome of stroke, TIA, or peripheral embolism. After a mean of 9 years of follow-up, the primary endpoint was reached by 11% of patients in the closure group compared with 21% in the medical therapy group (HR=0.43; 95% CI: 0.20 to 0.94; p=0.039). The main difference in the outcome measure seemed to be driven by differences in TIA, which occurred in 5% of closure patients compared with 14% of medical therapy patients.
Windecker et al. (23) compared 150 patients who underwent PFO closure between 1994 and 2000 with 158 medically treated patients over the same time period. The choice of therapy was based on clinician and/or patient preference. The patients who received closure differed from the medically treated patients on key clinical variables, including the percentage with more than one cerebrovascular event and the size of the PFO. At 4 -year follow-up, there was a trend toward lower recurrence of stroke or TIA in the PFO group that did not reach statistical significance (7.8% vs. 22.2%, p=0.08).
Harrer et al. (24) reported on 124 patients with cryptogenic stroke and PFO treated over a 10-year period. Eighty-three patients were treated with medical therapy, 34 were treated with percutaneous PFO closure, and 7 were treated with surgical closure. After a mean follow-up of 52±32 months, annual recurrence rates of stroke were not different between medical therapy and PFO closure (2.1% vs. 2.9%, respectively, p=NS).
Paciaroni et al. (25) performed a prospective observational study on 238 consecutive patients with cryptogenic stroke and PFO treated at 13 Italian centers. A total of 117 patients were treated with anti-thrombotic therapy, and 121 patients were treated with a closure device, with the treatment decision made according to patient and physician preference. Procedure-related adverse events were reported in 8 of 121 (6.8%) patients treated with a closure device (4 patients with tachycardia, 2 patients with allergic reaction, 1 patient with atrial fibrillation, 1 patient with sepsis). After a 2-year follow-up, 10 of 117 patients (8.5%) in the medical therapy group had a recurrent neurologic event (stroke or TIA), compared with 7/121 patients (5.8%, p=0.28) in the closure device group. For recurrent stroke, the difference between the groups was statistically significant, with 8/117 (6.8%) in the medical therapy group compared with 1/121 (0.8%, p=0.018) in the closure device group. On multivariate analysis, treatment with a closure device was a significant predictor of a reduced stroke rate (OR=0.1; 95% CI: 0.0 to 1.0; p=0.05) but was not a significant predictor of the combined outcome of stroke or TIA (OR=0.1; 95% CI: 0.02 to 1.5; p=0.10).
Alushi et al. reported results from a prospective, single-center study comparing outcomes after PFO device closure or medical management in 418 patients presenting with PFO and cryptogenic stroke or TIA. (26) Two hundred sixty-two patients underwent percutaneous PFO closure, while 156 were treated medically. The choice of medical intervention versus device closure was determined by the treating physician and patient. Percutaneous device closure was preferably advised for patients younger than age 55 years, with recurrent cerebrovascular events, large interatrial right-to-left shunt, and nonlacunar ischemic events on neuroimaging. Patients undergoing percutaneous closure were younger and more frequently presented with a larger interatrial right-to-left shunt, previous venous thromboembolism, and hypercoagulability state. Patients treated medically presented more frequently with multiple cerebrovascular accident risk factors. In a multivariable model to predict the composite outcome of TIA, stroke, or all-cause mortality, treatment strategy (percutaneous closure vs medical management) was not significantly associated with the outcome (adjusted OR=1.1; 95% CI: 0.44 to 2.74; p=0.81), after controlling for age, multiple prior cerebrovascular events, and the presence of aspirin.
Single-Arm Case Series
Many case series report on outcomes of PFO closure in an uncontrolled fashion; some examples of these series are as follows. Cifarelli et al. (27) reported on 202 consecutive patients treated with a closure device for secondary prevention of thromboembolism. They reported no periprocedural deaths or strokes, and one case of device migration 24 hours after placement. Recurrence-free survival was reported in 99% of patients 55 years of age or younger, and 84% in patients older than age 55 years. Recurrence of thromboembolism was associated with a septal aneurysm, with all patients who experienced recurrence of thromboembolism having a septal aneurysm. Onorato et al. (28) reported on 256 patients with paradoxical embolism who received transcatheter closure of PFO. The authors reported a 98.1% full closure rate of the PFO and no neurologic events at a mean follow-up of 19 months. Martin et al. (29) also reported on a study of 110 patients with paradoxical embolism who received transcatheter closure of PFO. While the full closure rate of PFO was 71% at 2 years, only 2 patients had experienced a recurrent neurologic event. Windecker et al. (30) reported on a case series of 80 patients with a history of at least 1 paradoxical embolic event and who underwent closure of a PFO with a variety of transcatheter devices. Patients were followed up for a mean of 1.6 years. During 5 years of follow-up, the risk of an embolic event (TIA, stroke, peripheral embolism) was 3.4%, considered comparable with either medical therapy with anticoagulation or open surgical approaches. The presence of a post-procedural shunt was a predictor of recurrent thromboembolic events, emphasizing the importance of complete closure. Butera et al. reported results from a registry that included 122 consecutive patients who underwent PFO closure with the Gore Septal Occluder device, 110 of whom underwent closure for previous stroke or TIA, and 12 of whom underwent closure for a history of migraines. (31) During a median follow-up of 9 months (range, 1-18 months), 7 patients experienced atrial arrhythmias, 4 of whom required medical treatment. On chest radiograph, 2 patients were found to have evidence of wire fractures in the device; the devices were not removed and the patients had no evidence of further problems from the wire fractures at 12 months of follow up. Three patients experienced neurologic problems, 1 of which was recurrent migraines. None of these patients were found to have residual shunt or intracardiac or device thrombi.
Other single-arm studies of transcathether PFO closure in patients presenting with stroke or TIA and PFO generally report high rates of freedom from embolic events. (32-37)
No clinical trials focus specifically on patients who have failed medical therapy, as defined by recurrent stroke or TIA while on therapy. Many of the published studies include both patients with first cryptogenic stroke, as well as patients with recurrent stroke or TIA, and generally do not analyze these patient populations separately. As a result, it is not possible to determine from the evidence whether PFO closure in patients who have failed medical therapy reduces the risk of subsequent recurrences.
The results of 3 RCTs do not support the conclusion that closure devices improve outcomes for patients with cryptogenic stroke and PFO. These trials, which included a total of 1,108 patients who underwent PFO closure and 1,178 patients who received medical management, did not report significant improvement in outcomes with PFO closure. These results contrast with the results of nonrandomized, comparative studies and systematic reviews of observational studies, which report lower rates of recurrent events following closure of PFO. The discrepancy in these results may arise from selection bias, because selection for either closure devices or medical therapy may vary, resulting in populations that may have unequal distribution of confounders. Also, the rate of recurrent stroke for patients treated with closure devices in the RCT was much higher than combined estimates from observational studies. This raises the possibility that ascertainment bias in the observational studies may have resulted in a spuriously low stroke rate for patients treated with a closure device. Multiple meta-analyses of the 3 RCTs, with or without the addition of nonrandomized studies, have come to varied conclusions, with some reporting a statistically significant reduction in recurrent events on pooled analysis and others reporting a trend for benefit that does not reach statistical significance. While these results suggest that a benefit might be present, the evidence is not definitive and the risk-benefit ratio is not well-defined.
PFO Closure for Migraine
Noncontrolled observational studies have reported improvement in migraine headaches after PFO closure.
A sham-controlled randomized trial of PFO closure for the indication of refractory migraine headache was published in 2008. (38) Migraine headache is another condition that has been associated with PFO in epidemiologic studies. In this study, there was no significant difference observed in the primary end point of migraine headache cessation (3/74 in the implant group, 3 /73 in the sham group, p=0.51). The results of this study cast some doubt on the causal relationship between PFO and migraine.
In 2014, Lip et al. published a primarily descriptive systematic review of studies that reported either the prevalence of PFO and migraine or the effects of PFO-related interventions on migraine attacks. (39) The authors included 20 studies that evaluated the prevalence of PFO in patients with migraines and 21 studies that presented data on the effects of PFO closure. In case series and cohort studies of patients with migraines, the prevalence of PFO in patients with migraines ranged from 14.6% to 66.5%. In case control studies, the prevalence of PFO in control patients ranged from 16.0% to 25.7%, while the prevalence of PFO in patients with migraine-with-aura or aura migraine-without-aura ranged from 26.8 to 96.0% and 22.6% to 72.4%, respectively. In the 18 case series that reported migraine outcomes after PFO closure, rates of resolution for migraine with aura and migraine without aura ranged from 28.6% to 92.3% and 13.6% to 82.9%, respectively. In 2 case-control studies that compared PFO closure and medication for migraines with intervention, improvement in migraine symptoms occurred in 83% to 87%, compared with 0% to 21% of those managed medically. The single RCT identified (Dowson et al. (38)) did not identify significant improvements in migraine symptoms in the PFO closure group.
In a study not included in the Lip et al. systematic review, Biasco et al. retrospectively compared transcatheter PFO closure with medical therapy in terms of impact on daily activities. (40) The study included 217 patients with migraine and echocardiographic evidence of PFO, 89 were managed with percutaneous PFO closure and 128, medically managed. PFO device closure was recommended for patients with migraine associated with previous suspected paradoxical embolic events, or for those without a history of suspected embolic events only in the case of severely disabling symptoms not controlled by multiple therapies. At a mean follow up of 1299 days, both groups demonstrated significant improvements in scores on the Migraine Disability Assessment Questionnaire (MIDAS). However, there were no significant differences in MIDAS score between groups (p=0.204). The degree of residual right-to-left shunt was not associated with symptom perception.
Although observational studies have shown a possible association between PFO closure and improvement in migraine symptoms, 1 sham-controlled RCT did not demonstrate significant improvements in migraine symptoms after PFO closure. Nonrandomized studies show highly variable rates of migraine improvement after PFO closure.
PFO Closure for Other Indications
Several other medical conditions have been reported to occur more frequently in patients with PFOs, including platypnea-orthodeoxia syndrome, myocardial infarction with normal coronary arteries and decompression illness in response to change in environmental pressure, high altitude pulmonary edema, and obstructive sleep apnea. (41) Evidence about clinical outcomes related to these conditions after PFO closure is limited to case reports and case series.
Transcatheter Device Closure for ASDs
At present there are 2 FDA-approved devices for ASD closure: the AMPLATZER™ Septal Occluder, and the GORE HELEX™ Septal Occluder.
Overview of the Evidence
Evidence supporting the efficacy of devices for closure of ASD consists of nonrandomized comparative studies and case series. However, in contrast to the situation of PFO and cryptogenic stroke, the relationship of closure of the ASD and improved clinical outcomes is direct and convincing, because the alternative treatment is open surgery. Results generally show a high success rate in achieving closure and low complication rates. FDA approval of the AMPLATZER Septal Occluder was based on the results of a multicenter, nonrandomized study comparing the device with surgical closure of ASDs; 423 patients received 433 devices. (42) This study was subsequently published with slightly different numbers but similar quantitative findings. (43) All patients had an ostium secundum atrial septal defect and clinical evidence of right ventricular volume overload. The results for the septal occluder group, showed comparably high success rates with surgery; the 24-month closure success rate was 96.7% in the septal occluder group compared with 100% in the surgical group. While the pattern of adverse events was different in the 2 groups, overall, those receiving a septal occluder had a significantly lower incidence of major adverse events (p=0.03). Similarly, there was a significantly lower incidence of minor adverse events in the septal occluder group (p<0.001). It should be noted that the mean age of patients of the 2 groups was significantly different; in the septal occluder group, the mean age was 18 years, compared with 6 years in the surgically treated group.
A systematic review of percutaneous closure versus surgical closure was published by Butera et al. in 2011. (44) Thirteen non-randomized comparative studies that enrolled at least 20 patients were included, with a total of 3,082 patients. The rate of procedural complications was higher in the surgical group (31%; 95% CI: 21% to 41%) compared with the percutaneous group (6.6%; 95% CI: 3.9% to 9.2%), with an odds ratio for total procedural complications of 5.4 (95% CI: 2.96 to 9.84; p<0.0001). There was also an increased rate of major complications for the surgical group (6.8%; 95% CI: 4% to 9.5%) compared with the percutaneous group (1.9%; 95% CI: 0.9% to 2.9%), for an odds ratio of 3.81 (95% CI: 2.7 to 5.36; p=0.006).
In the Abaci et al systematic review and meta-analysis of periprocedural complications after ASD/PFO device closures referenced earlier, for ASD closure, the pooled rate of major complications after ASD closures was 1.6% (95% CI: 1.4% to 1.8%). (21)
Nonrandomized Comparative Studies
Other nonrandomized studies comparing transcatheter closure with surgery show similar success rates. Suchon et al., in a study of 100 patients, had a 94% success rate in the transcatheter closure group compared with a 100% success rate in the surgical group. (45) A study by Berger et al. showed identical 98% success rate in both treatment groups. (46) A non-randomized comparative analysis by Kotowycz et al. (47) Reported that mortality rates at 5-year follow-up did not differ between transcatheter and surgical closure (5.3% vs. 6.35% respectively, p=1.00), but that reintervention rates were higher for patients undergoing transcatheter closure (7.9% vs. 0.3% respectively, p<0.004). Xu et al. reported a retrospective analysis of transcathether (n=35) and surgical (n=43) closure of ASD in patients with ASD and pulmonary stenosis.48 Complication rates were not significantly different between groups, and all patients in both groups were reported to have complete correction of their ASD.
Single-arm studies show high success rates of ASD closure. The FDA study discussed previously was the largest series, with an enrollment of 423 patients. Fischer et al. reported on use of the AMPLATZER device in 236 patients with secundum ASD. (49) In this evaluation study, closure was achieved in 84.7% of patients, and intermediate results were reported as excellent. Other smaller studies have reported favorable results for transcatheter closure of ASD. In Du et al., transcatheter closure of ASD in 23 patients with deficient ASD rims was compared with transcatheter closure of 48 patients with sufficient ASD rims. (50) The authors reported no significant differences in closure rates between the groups (91% for deficient rims, 94% for sufficient rims) along with no major complications at 24 hours and 6-month follow-up. Oho et al. also reported a successful closure rate of 97% at 1-year follow-up in 35 patients receiving transcatheter closure of ASD, while only 1 patient complication of second-degree atrioventricular block was noted. (51) Brochu et al. evaluated 37 New York Heart Association (NYHA) class I or II patients who underwent transcatheter closure of ASD. (52) At 6-month follow-up, maximal oxygen uptake improved significantly, and the dimensions of the right ventricle decreased significantly while 20 patients moved from NYHA class II to class I and improved exercise capacity. Numerous other small, single-arm studies report similar results, with procedural success approaching 100% and successful closure on follow-up reported in the 90% to 100% range. (3, 5)
Single-Arm Studies in Pediatrics
Several single-arm studies have reported outcomes from transcathether ASD closure in children and adolescents. Grohmann et al reported outcome from a single-center series of children aged 3 to 17 years (median, 6) who were treated with the Gore Septal Occluder, with technical success in 41 of 45 patients in whom closure was attempted (91%).(53 Nyboe et al reported outcomes from 22 patients with secundum ASD who underwent ASD closure with the Gore Septal Occluder, 10 of whom were children younger than age 15, with technical success in all patients. (54) Yilmazer et al reported improvements in echocardiographic parameters in a series of 25 pediatric patients (mean age, 9.02) who underwent successful transcathether closure of secundum ASD. (55)
For patients with an ASD, nonrandomized comparative studies and single-arm case series show high success rates of closure using closure devices approaching the high success rates of surgery. The percutaneous approach has a low complication rate and avoids the morbidity and complications of open surgery. If the percutaneous approach is unsuccessful, ASD closure can be achieved using surgery. Because of the advantages of percutaneous closure over open surgery, the use of percutaneous ASD closure devices can be considered medically necessary.
Ongoing and Unpublished Clinical Trials
An online search of ClinicalTrials.gov on August 6, 2014, identified the following interventional trials evaluating device closure for atrial septal defects and patent foramen ovale that are currently ongoing.
Summary of Evidence
The evidence on the efficacy of closure devices for patients with patent foramen ovale (PFO) and cryptogenic stroke does not permit conclusions on whether health outcomes are improved. Three randomized controlled trials have been completed that compare closure devices with medical therapy in patients with cryptogenic stroke and PFO. None of the 3 trials reported statistically significant improvements on their main outcome using intention-to-treat analysis. In all 3 trials, low numbers of outcome events in both groups limited the power to detect differences between groups. One trial showed a significant benefit for the closure group on per-protocol analysis and another showed significant benefit on secondary outcomes. Meta-analyses of these trials have also come to different conclusions, with some reporting a statistically significant reduction in recurrent events on pooled analysis and others reporting a trend for benefit that does not reach statistical significance. While these results suggest that a benefit might be present, the evidence is not definitive and the risk-benefit ratio is not well-defined.
PFO closure devices have been investigated as a treatment for refractory migraines and for other medical conditions. The evidence on the use of PFO device closure for the treatment of migraines is limited to nonrandomized studies and 1 RCT that showed no benefit to migraine symptoms from PFO device closure. The evidence for the use of PFO closure for conditions other than stroke prevention and refractory migraines is limited to case reports and case series.
In addition to the limitations in the evidence base, there are no PFO closure devices that are currently FDA-approved for use in the United States. Therefore closure devices are considered investigational for prevention of paradoxical emboli in patients with PFO and cryptogenic stroke.
For patients with atrial septal defect (ASD) that require closure, nonrandomized comparative studies and single-arm case series show high success rates of closure using closure devices, approaching the high success rates of surgery. The percutaneous approach has a low complication rate and avoids the morbidity and complications of open surgery. Because the main alternative to percutaneous closure is open surgery, this evidence is sufficient to conclude that percutaneous closure achieves similar outcomes with less risk compared with the alternative. If the percutaneous approach is unsuccessful, ASD closure can be achieved using surgery. Because of the advantages of percutaneous closure over open surgery, the use of percutaneous ASD closure devices can be considered medically necessary for this purpose.
Practice Guidelines and Position Statements
The American College of Chest Physicians (ACCP)
The ACCP published guidelines on antiplatelet and antithrombotic therapy in 2012, (56) which were an update to previous guidelines published in 2008. (57) These guidelines contained the following statements about the treatment of patients with a PFO:
Closure devices are not discussed.
Previously published Guidelines from ACCP (2008) compared outcomes from medical management and percutaneous closure in patients with PFO and cryptogenic stroke and concluded that there was no difference in risk of death or between major adverse clinical events between patients with cryptogenic stroke who underwent medical management and those who underwent percutaneous closure procedures. (58)
The American Heart Association (AHA)/American Stroke Association (ASA)
The AHA/ASA published guidelines on the prevention of stroke in patients with ischemic stroke or transient ischemic attack (TIA) in 2006 that makes the following recommendations about the management of patients with a PFO (59):
AHA/ASA guidelines (59) published in 2006 offer somewhat more specific recommendations. These guidelines do not recommend PFO closure as initial therapy for patients with a first ischemic stroke and PFO, stating that, “Insufficient data exist to make a recommendation about PFO closure in patients with a first stroke and a PFO.” They also state that “…aspirin (50-325mg/d), aspirin and extended-release dipyridamole in combination, and clopidrogrel are all acceptable options for initial therapy (class IIa, level of evidence A),” and that “Warfarin is reasonable for high-risk patients who have other indications for oral anticoagulation, such as underlying hypercoagulable state or evidence of venous thrombosis (class IIa, level of evidence C).” For patients with stroke or TIA while on medical therapy, they state that, “PFO closure may be considered for patients with recurrent cryptogenic stroke despite optimal medical therapy (class IIB, level of evidence C).”
The American Academy of Neurology (AAN)
The AAN (60) published in 2004, states that the evidence is inconclusive regarding the comparative efficacy of PFO closure devices and medical therapy. These guidelines do not offer specific recommendations as to when PFO closure devices should be used.
American College of Cardiology
Guidelines issued by the American College of Cardiology and AHA in 2008 on the management of congenital heart disease recommend closure of an ASD by either percutaneous or surgical methods for several indications. (61) For sinus venosus, coronary sinus, or primum ASD, however, surgical rather than percutaneous closure is recommended.
U.S. Preventive Services Task Force Recommendations
Use of closure devices for PFO or ASDs is not a preventive service.
Medicare National Coverage
There is no national coverage determination (NCD). In the absence of an NCD, coverage decisions are left to the discretion of local Medicare carriers.
Percutaneous transcatheter closure of congenital interatrial communication (i.e., fontan fenestration, atrial septal defect) with implant
Repair of atrial septal defect with prosthesis, closed techniques
Other and unspecific repair of atrial septal defect (i.e., patent foramen ovale)
Atrial septum repair, percutaneous
Atrial septum repair, percutaneous endoscopic
Atrial septum repair, percutaneous, with device
Atrial septum repair, percutaneous endoscopic, with device
Type of Service
Place of Service
Add to Medicine Section - New Policy
Replace policy - New information on patent foramen ovale; rest unchanged.
Replace policy - Revised; added requirements to policy statement patients with PFO must fail trial of oral anticoagulants. Noted FDA approval of Amplatzer device. Policy replaces 2.02.09.
Replace policy - Policy statement revised to indicate transcatheter treatment of ASD may be considered medically necessary. Replaces P2.02.100.
Replace policy - Policy reviewed; no change to policy statement; CPT codes updated.
Update Scope and Disclaimer - No other changes.
Cross Reference Update - No other changes.
Cross Reference Update - No other changes.
Replace policy - Policy updated with literature search; policy rationale extensively revised. Policy statement for PFO changed to investigational due to the FDA's withdrawal of the humanitarian device exemption approval. References added.
Replace policy - Policy updated with literature search; no change to the policy statement. References added.
Replace policy – Policy updated with literature search. Policy statements unchanged. References 5, 8, 15 and 25 added. ICD-10 codes added to policy.
Replace policy - Policy updated with literature search. References 3, 6, 7, and 30 added. No change to policy statement.
Replace Policy. Policy guidelines reformatted for usability. Updated Regulatory Status with information about the FDA medical device alert issued 10/17/13 for the Amplatzer ASO. Rationale updated with literature search through July 31, 2013. References 4-7, 25 added; others renumbered/removed. Policy statement unchanged.
Annual Review. Policy updated with literature review through August 1, 2014. References 8-17, 21, 26, 33-37, 40, 48, 53-55, 58-59 added. Policy statement unchanged. ICD-9 and ICD-10 diagnosis codes removed; these are not utilized in adjudication of the policy.
Disclaimer: This medical policy is a guide in evaluating the medical necessity of a particular service or treatment. The Company adopts policies after careful review of published peer-reviewed scientific literature, national guidelines and local standards of practice. Since medical technology is constantly changing, the Company reserves the right to review and update policies as appropriate. Member contracts differ in their benefits. Always consult the member benefit booklet or contact a member service representative to determine coverage for a specific medical service or supply. CPT codes, descriptions and materials are copyrighted by the American Medical Association (AMA).