Author + information
- Received March 22, 2014
- Revision received May 24, 2014
- Accepted June 8, 2014
- Published online December 1, 2014.
- John W. Moore, MD, MPH∗∗ (, )
- Jessica Greene, BS†,
- Salvadore Palomares†,
- Alexander Javois, MD‡,
- Carl Y. Owada, MD§,
- John P. Cheatham, MD‖,
- Mark H. Hoyer, MD¶,
- Thomas K. Jones, MD#,
- Daniel S. Levi, MD∗∗,
- Pivotal and Continuing Access Studies of the Nit Occlud PDA Investigators
- ∗Division of Cardiology, Rady Children's Hospital, University of California, San Diego School of Medicine, San Diego, California
- †PFM Medical, Inc., Carlsbad, California
- ‡Division of Cardiology, Advocate Children’s Hospital, University of Illinois School of Medicine, Park Ridge, Illinois
- §Division of Cardiology, Children’s Hospital Central California, Fresno, California
- ‖Division of Cardiology, Nationwide Children’s Hospital, Ohio State University School of Medicine, Columbus, Ohio
- ¶Division of Cardiology, Riley Hospital for Children, Indiana University School of Medicine, Indianapolis, Indiana
- #Division of Cardiology, Seattle Children’s Hospital, University of Washington School of Medicine, Seattle, Washington
- ∗∗Division of Cardiology, Mattel Children’s Hospital UCLA, David Geffen School of Medicine, University of California, Los Angeles, California
- ↵∗Reprint requests and correspondence:
Dr. John W. Moore, Rady Children’s Hospital, University of California, San Diego School of Medicine, 3020 Children’s Way, MC 5004, San Diego, California 92123.
Objectives This study aimed to compare the efficacy and safety of the Nit-Occlud PDA device (PFM Medical, Cologne, Germany) to benchmarks designed as objective performance criteria (OPC).
Background The Nit-Occlud PDA is a nitinol coil-type patent ductus arteriosus (PDA) occluder with a reverse cone configuration, which is implanted using a controlled delivery system.
Methods Patients with <4-mm minimum diameter PDA were prospectively enrolled in the Pivotal and the Continuing Access Studies from 15 sites in the United States and were followed up for 12 months post-procedure. Investigator-reported outcomes were compared to OPC including a composite success criterion, efficacy criteria of successful closure (clinical and echocardiographic), and safety criteria incidence of adverse events (serious and of total).
Results The Pivotal Study enrolled patients between November 1, 2002 and October 31, 2005, and the Continuing Access Study enrolled additional patients between September 1, 2006 and October 31, 2007. A total of 357 patients were enrolled, and 347 had successful device implantations. After 12 months, 96.8% had complete echocardiographic closure (OPC = 85%) and 98.1% had clinical closure (OPC = 95%). There were no deaths or serious adverse events (OPC = 1%). The total adverse event rate was 4.7% (OPC = 6%). Composite success was 95.1% in the study patients (OPC = 80%).
Conclusions Closure of small- and medium-sized PDA with the Nit-Occlud PDA is effective and safe when compared with OPC.
Persistent patency of the ductus arteriosus is a common defect accounting for up to 6% of patients seen in busy congenital heart programs (1). Patent ductus arteriosus (PDA) may be associated with endarteritis, left heart volume overload, pulmonary congestion, and pulmonary hypertension. Closure of moderate and large PDA is indicated at the time of discovery, and closure of smaller audible PDA is also recommended (2).
Surgical PDA ligation was first performed by Gross and Hubbard in 1939 (3). Transcatheter closure of PDA was first accomplished by Porstmann et al. (4) in 1967 using an Ivalon plug. Rashkind et al. (5) designed an umbrella-type PDA occluder that was tested in the United States in the 1980s but did not perform well enough to achieve U.S. Food and Drug Administration (FDA) clearance. In 1992, the first report of off-label use of Gianturco coils for closure of PDA (6) rapidly lead the way to widespread use of coils to close small to moderate PDA. Off-label use of Gianturco coils continues to the present time (7,8). The Amplatzer Duct Occluder (St. Jude Medical, St. Paul, Minnesota), a nitinol wire mesh plug, designed for moderate to large PDA, achieved FDA clearance in 2004 (9). It continues to be used widely for closure of larger PDA (7,8).
The Nit-Occlud PDA (PFM Medical, Cologne, Germany) is a coil-type, controlled release device specifically designed for occlusion of small to moderate PDA. The device consists of reinforced spiral nitinol coils containing multiple loops of varying diameters and graduated stiffness. These devices accommodate a variety of ductal sizes and configurations. The device is implanted through a 4- or 5-F catheter, and it has a controlled delivery system, which allows for repositioning or retrieval if required.
Although the Nit-Occlud PDA has been available outside the United States for several years (10), the device only recently received FDA clearance for the indication of PDA occlusion. The aim of this report is to describe the clinical study data reviewed by the FDA in granting device clearance, focusing on efficacy and safety data (11).
The Nit-Occlud PDA device is a nitinol coil designed and shaped specifically for PDA closure. The Nit-Occlud coils are available as “flex” and “medium” coils in a range of sizes. The flex coils are made from nitinol wire windings with a 0.85-mm outer diameter whereas the medium devices are thicker and stiffer because they have a 0.95-mm outer diameter. Each device is named by identifying the diameter of the most distal coil diameter (always the largest) and the most proximal coil loop diameter. For example, in an 11 × 6 device, the first coil loop deployed (the distal loop) has a diameter of 11 mm and the final loop (most proximal) has a diameter of 6 mm.
As the device is deployed, the coil’s diameter initially decreases distally to proximally, which allows the device to form a cone designed to fit into the ductal ampulla to occlude the PDA and prevent the device from pulling through to the pulmonary artery. As the additional device coils are deployed, they form a second “reverse cone” at the device’s proximal end by progressively increasing in diameter as shown in Figure 1A. As shown in Figure 1B, the proximal windings act to both reinforce the distal windings and allow for anchoring to the pulmonary side of the PDA. There have been no additional modifications of this device since the study described in this report.
Although the Nit-Occlud device is designed primarily for delivery from the venous/antegrade approach, the 4 × 4 and 5 × 4 devices are symmetric enough that they can also be delivered retrograde from the aortic side. Coil occlusion is most effective if the majority of the windings are deployed on the aortic side. Often only 1 loop is positioned on the pulmonary side. The 4 × 4, 5 × 4, and 6 × 5 devices are designated “flex” devices and are implanted through a 4-F delivery catheter. The larger device sizes (7 × 6, 9 × 6, and 11 × 6) are the “medium” devices, which require a 5-F delivery catheter. The device has a controlled release system that allows it to be retrieved and repositioned as needed. If embolized, the Nit-Occlud coil can be snared and retrieved into a 6-F sheath.
The Nit-Occlud PDA pivotal study and continuing access study had identical study designs. The studies were controlled nonrandomized, prospective multicenter trials conducted in 15 medical centers in the United States. The studies evaluated the effectiveness and the safety of the Nit-Occlud PDA for closure of small and moderate PDA. The studies’ outcomes were defined to determine whether the Nit-Occlud PDA met or exceeded objective performance criteria (OPC) derived from historical results of device and surgical PDA closure (12). Study centers with experienced pediatric interventional cardiologists were selected. The cardiologists received training and proctoring in the use of the device and agreed to adhere to the study designs. Each participating center obtained local institutional review board approval of the studies, and informed consent was obtained for each study participant.
Inclusion criteria were as follow: presence of a PDA with minimum diameter of 4 mm or smaller as measured by the width of the color Doppler stream and secondarily by angiogram (Figure 2A); weight at least 5 kg; and age 6 months to 21 years. Exclusion criteria included the following: presence of associated cardiac anomalies requiring surgery; bleeding or clotting disorders; pulmonary hypertension (with pulmonary vascular resistance of 5 Wood units or greater); angiographic contrast allergy; pregnancy; and acute illnesses.
Patients were evaluated prior to catheterization, during catheterization, at hospital discharge, and 6 months and 12 months after device implant. Follow-up evaluations included a patient history, physical examination, and transthoracic echocardiogram. Data report forms were maintained in the patient study books at the sites and were transmitted to the study sponsor (PFM Medical) where an electronic study database was maintained. Periodic site audits, by the sponsor, confirmed consistency of reports with source patient data.
The primary outcome metric of the studies was composite success, defined as technical success of device implantation (Figure 2B), clinical and echocardiographic closure of the PDA 1 year after device implantation, and absence of device- or procedure-related death or serious adverse events (AE) after 1-year follow-up. Secondary outcomes included individual efficacy and safety metrics after 1-year follow-up. Efficacy measures were clinical closure (defined as absence of a typical PDA heart murmur) and echocardiographic closure (defined as absence of color Doppler evidence of residual PDA flow). For control of data quality, an independent core echocardiographic laboratory reviewed one-third of the 1-year follow-up echocardiograms reported by the study sites. Safety measures included device- and procedure-related deaths, serious AE, and total AE through the 1-year follow-up evaluation. Serious AE were defined as events, which were life-threatening, required surgery to correct, resulted in hospitalization or prolonged hospital stay, caused long-term disability, or resulted in genetic damage or birth defect. Serious AE included, but were not limited to, cerebral or pulmonary embolism, bacterial endocarditis, device embolization requiring surgery, and persistent cardiac arrhythmia requiring a pacemaker. Total AE included serious AE as well as non-life-threatening events, which were resolved either by nonsurgical intervention or with no intervention. Adverse events were periodically reviewed and adjudicated as needed by an independent data safety monitoring board convened by the study sponsor.
For binary outcomes per person, such as technical success at implantation, clinical closure, echocardiographic closure, and composite success at 12 months, binomial proportions expressed in terms of percentages were calculated. Standard errors and exact 95% confidence intervals based on the Clopper-Pearson method were also computed. For AE outcomes where there was the possibility of more than 1 event per study patient, standard errors and exact 95% confidence intervals using exact Poisson methods were calculated. Statistics were computed with the help of the SAS statistical software (SAS Institute Inc., Cary, North Carolina) and STATA statistical software (version 10, StataCorp, College Station, Texas) and StatXact software (Cytel Software, Cambridge, Massachusetts).
Outcome measures were compared with OPC, which were either recommended by a multiorganization advisory panel to the FDA for pediatric cardiovascular devices (11) or were specified by the FDA at the time of granting the investigational device exemption for the pivotal study. A tipping point analysis is performed for each outcome for which an OPC is specified.
A total of 357 patients from 15 sites in the United States were enrolled in the Pivotal Study (between November 2002 and October 2005) and the Continuing Access Study (between September 2006 and October 2007). These patients are presented here together. All patients met study inclusion and exclusion criteria.
Patient demographics included median age of 2.96 years (range 6 months to 21.9 years). Patient weight ranged from 4.7 kg to 109 kg (median: 13.9 kg, mean: 18.1 kg). Female patients constituted 68.1% of the patients (243 of 357). Additional congenital heart defects were present in 13% of the patients (45 of 357), most commonly atrial communications or small ventricular septal defects. Baseline clinical assessment showed that 96.8% or 336 of 357 patients had an audible PDA murmur. Eleven patients had no audible murmurs (silent PDA).
Angiographic evaluation of the 357 PDA in the catheterization laboratory demonstrated that 74.8% were classified as Krichenko (13) type A (conical), 14.0% as type E (elongated), 4.8% s type D (complex), 4.7% as type B (short), and 1.2% as type C (tubular). Mean minimum PDA diameter was 1.9 mm, and the median PDA diameter was also 1.9 mm (range 0.5 mm to 3.9 mm). Based on measured oxygen saturations in 355 patients, average ratio of pulmonary blood flow (Qp) to systemic blood flow (Qs) was 1.46. Median Qp/Qs was 1.3.
Successful implantation of a Nit-Occlud PDA device, or procedure technical success, was achieved in 347 of 357 patients (97.2%). There were 10 technical failures, which were attributed to ductal configuration (type B or C) in 4, ductal minimum diameter (small in 1, large in 2), or operator learning curve in 3 patients. Median procedure times at sites implanting more than 50 devices ranged from 78.0 to 109.8 min. At those same sites, median fluoroscopy times ranged from 16.5 to 21.0 min. Several sites implanting fewer than 15 devices had median fluoroscopy times below this range.
Efficacy endpoints included clinical and echocardiographic closure after 1 year; however, these efficacy endpoints were also assessed at hospital discharge, after 6 months, and after 12 months. The 1- year follow-up evaluation occurred at an average of 14.3 months after implantation (SE 3.78 months). At this point, 314 of 347 patients presented for clinical evaluations; however, only 309 patients had an echocardiogram performed. Among the 314 patients who presented, 308 (98.1%) had no detectable PDA murmur on examination. In the 309 patients having echocardiograms, 299 (96.8%) had no residual PDA shunting by color Doppler. Among the 1-year follow-up patients having echocardiograms, 226 of 234 (96.6%) with type A, 11 of 13 (84.6%) with type B, 4 of 4 with type C, 16 of 16 with type D, and 42 of 42 with type E had no residual shunting.
Safety endpoints included device- and procedure-related deaths, serious AE, and total AE assessed cumulatively up to the 1-year follow-up evaluation. There were no deaths and no serious adverse events reported. There were 6 device-related and 9 procedure-related AE reported (15 of 316 or 4.7%). Device-related events included 3 device embolizations (1 to right pulmonary artery and 2 to the left femoral artery). In these cases, percutaneous retrieval was performed (after 14 days from the pulmonary artery, and during the implantation procedure and after 215 days post-implantation from the femoral arteries). Additional device-related AE included the following: device protrusion into the descending thoracic aorta, causing a 10-mm Doppler gradient (n = 1); device tip causing an apparent very small false aneurysm of the aorta (n = 1); and presence of an apparent small thrombus on the pulmonary end of the device by echocardiogram. Procedure-related adverse events included the following: temporary loss of pulse (n = 2); need to reintubate because of excessive post-procedure hypoventilation (n = 1); post-anesthesia nausea and vomiting (n = 5); and allergic skin reaction to tape (n = 1).
Composite success was designated as the principle study outcome because it took into account all aspects of device use including procedure technical success, 1- year efficacy, and death and serious AE. This metric was calculated to be 95.1% (294 of 309).
In the absence of a surgical or other control group, OPC were selected to function as benchmarks for comparison with the study outcomes. Table 1 presents the study outcomes with the relevant OPC. It is apparent from a simple review that all study outcomes satisfied or improved the levels specified by individual OPC. However, to more rigorously compare the study outcomes to the OPC, a tipping point analysis was performed for each outcome having an OPC. A tipping point value was defined to be the observed value closest to but worse than the OPC value that corresponds to a 1-sided p value of ≤0.025. That is, if the OPC is the null hypothesis value in a hypothesis test, the tipping point value was defined as the value closest to the OPC in the unfavorable direction that would cause one to reject the null hypothesis with p < 0.025 1-sided, corresponding to p < 0.050 2-sided—the usual statistical significance criterion. This is also equivalent to the 95% confidence bound no longer including the OPC. The tipping point values computed are based on the actual sample size (n) reported, which is not the same for all outcomes. For example, regarding technical success of device implantation, a value of 97.2% success was observed, or 347 successes of 357 attempts. For this outcome, the sample size was n = 357. The OPC for this outcome was 95%. The tipping point value for this outcome was 92.4%. Therefore, given a sample of n = 357, if only 92.4% success or less (330 of 357) was observed, the p value for testing true OPC = 95% would be <0.025 1-sided equivalent to p < 0.05 2-sided. That is, had 92.4% instead of 97.2% been observed, the null hypothesis of OPC = 95% would have to be rejected, and one would have had to conclude that the true technical success in general (i.e., in all patients in this population) was under 95%. Table 2 provides tipping point values calculated for all of the study outcomes, which have assigned OPC. It is clear from this analysis that the outcomes observed are comfortably consistent with the OPC.
The Nit-Occlud PDA is the only coil-type device that has achieved FDA clearance for the indication of PDA occlusion. As is well known, Gianturco coils (Cook Medical, Bloomingdale, Indiana) have been used off-label for closure of small to moderate PDA for decades. Unfortunately, there has been no large controlled study demonstrating safety and efficacy of coils for PDA occlusion in spite of widespread coil use. This study, therefore, is unique and long overdue. In a large pediatric study population, it demonstrates acceptably high levels of efficacy and safety in the use of the Nit-Occlud coil for PDA closure. Although echocardiographic closure at 1 year was not universal, 96.8% definitive (echocardiographic) closure was attained. This closure rate comfortably exceeds the OPC benchmark designated by the FDA advisory panel. Furthermore, this level of efficacy was associated with a level of adverse events, which is low and consistent with the contemporary practice environment. With no mortality and no serious AE noted in this study, all observed AE were treated without surgery and resulted in no sustained disability or injury. The total rate of 4.7% met the OPC benchmark. The most important AE was device embolization (n = 3), which in all cases was handled with percutaneous retrieval of the devices.
Although there have been no other prospective multicenter studies of PDA coils. There have been registry studies presenting retrospective data about the use of coils for PDA occlusion. The Michigan PDA Coil Registry was an early attempt to accumulate a large multicenter data set. Unfortunately, results were not published beyond abstract form. In 2001, the European PDA Registry retrospectively reported on 1,291 intended PDA coil occlusion procedures (14). In this registry, procedure success was 94.3% in patients having PDA minimum diameter up to 6 mm and the rate of suboptimal outcomes (including failed procedures and adverse events) was 10%. Contemporary registries including MAGIC (Mid-Atlantic Group of Interventional Cardiology) (7), C3PO (Congenital Cardiac Catheterizations Outcomes Project) (8), and IMPACT (Improving Pediatric and Adult Congenital Treatment) (15) have or are reporting fairly substantial PDA datasets (n = 359 [MAGIC], n = 496 [C3PO], and n = 1,375 [IMPACT]) with use of a mix of approved devices and unapproved coils for PDA closure. These registries report only acute data such as procedure technical success and procedural or short-term device-related AE. Technical success in these studies ranges from 96.8% to 99.4%. Major AE ranged from 0.2% to 2.0%, and total AE ranged from 4.6% to 9.0%.
This study lacked a surgical and an approved device control group. Given the likely barriers to enrollment in a surgical study arm and the absence of an approved alternative device for small to moderate PDA, the study design using OPC was a practical compromise agreed to by the FDA. The development and the use of OPC as benchmarks for this study provided an opportunity to consolidate historical data and contemporary expert opinion into specific criteria that functioned as surrogates for control group data. Outcomes of the study compare favorably to these benchmarks.
This report presented the study data provided to the FDA in its review of the Nit-Occlud PDA. High levels of composite success and efficacy and low levels of adverse events were presented, suggesting that the Nit-Occlud PDA is safe and effective for closure of small to moderate PDA.
For the list of additional site investigators, please see the online version of this paper.
All the physician authors participated as site investigators at their respective institutions. Dr. Moore was the principal investigator for both studies and a site investigator at UCLA. Dr. Moore had a consulting relationship with PFM Medical during the conduct of the studies and the submission of the PMA application; he has no current relationship with PFM Medical Inc. Ms. Greene and Mr. Palomares acted as study monitors and data coordinators for both studies, and are PFM Medical, Inc. employees. Dr. Javois is a consultant to PFM Medical. Dr. Owada has received consulting and proctoring fees from BBraun. Dr. Hoyer has received consulting and/or proctoring fees from Gore Medical and St. Jude Medical. Dr. Levi has received consulting fees from PFM Medical, but does not do so currently. All other authors have reported that they have no relationships relevant to the contents of this paper to disclose. See the Online Appendix for investigators not credited as authors.
- Abbreviations and Acronyms
- adverse event(s)
- U.S. Food and Drug Administration
- objective performance criteria
- patent ductus arteriosus (arteriosi)
- Received March 22, 2014.
- Revision received May 24, 2014.
- Accepted June 8, 2014.
- American College of Cardiology Foundation
- Fyler D.C.
- Feltes T.F.,
- Bacha E.,
- Beekman R.H. 3rd.,
- et al.
- Porstmann W.,
- Wierny L.,
- Warnke H.
- Rashkind W.J.,
- Mullins C.E.,
- Hellenbrand W.E.,
- Tait M.A.
- El-Said H.G.,
- Bratincsak A.,
- Foerster S.R.,
- et al.
- Pass R.H.,
- Hijazi Z.,
- Hsu D.T.,
- Lewis V.,
- Hellenbrand W.E.
- ↵U.S. FDA. Summary of safety and effectiveness data (SSED) for the Nit-Occlud PDA. August 26, 2013. Available at: http://www.accessdata.fda.gov/scripts/cdrh/cfdocs/cfTopic/pma/pma.cfm?num=P120009. Accessed May 5, 2014.
- Magee A.G.,
- Huggon J.C.,
- Seed P.T.,
- et al.,
- for the Association for European Cardiology
- Moore J.W.,
- Vincent R.N.,
- Beekman R.H. III.,
- et al.