- Email: [email protected]
Final Two-Year Outcomes for the Sentry Bioconvertible Inferior Vena Cava Filter in Patients Requiring Temporary Protection from Pulmonary Embolism
CLINICAL STUDY
Final Two-Year Outcomes for the Sentry Bioconvertible Inferior Vena Cava Filter in Patients Requiring Temporary Protection from Pulmonary Embolism €mer, MD, Michael D. Dake, MD, Timothy P. Murphy, MD, Albrecht H. Kra Michael D. Darcy, MD, Luke E. Sewall, MD, Michael A. Curi, MD, Matthew S. Johnson, MD, Frank Arena, MD, James L. Swischuk, MD, Gary M. Ansel, MD, Mitchell J. Silver, DO, Souheil Saddekni, MD, Jayson S. Brower, MD, and Robert Mendes, MD; for the SENTRY Trial Investigators
ABSTRACT Purpose: To report final 2-year outcomes with the Sentry bioconvertible inferior vena cava (IVC) filter in patients requiring temporary protection against pulmonary embolism (PE). Materials and Methods: In a prospective multicenter trial, the Sentry filter was implanted in 129 patients with documented deep vein thrombosis (DVT) and/or PE (67.5%) or who were at temporary risk of developing DVT/PE (32.6%). Patients were monitored and bioconversion status ascertained by radiography, computed tomography (CT), and CT venography through 2 years. Results: The composite primary 6-month endpoint of clinical success was achieved in 97.4% (111/114) of patients. The rate of new symptomatic PE was 0% (n ¼ 126) through 1 year and 2.4% (n ¼ 85) through the second year of follow-up, with 2 new nonfatal cases at 581 and 624 days that were adjudicated as not related to the procedure or device. Two patients (1.6%) developed symptomatic caval thrombosis during the first month and underwent successful interventions without recurrence. No other filter-related symptomatic complications occurred through 2 years. There was no filter tilting, migration, embolization, fracture, or caval perforation and no filterrelated deaths through 2 years. Filter bioconversion was successful for 95.7% (110/115) of patients at 6 months, 96.4% (106/110) of patients at 12 months, and 96.5% (82/85) of patients at 24 months. Through 24 months of follow-up, there was no evidence of late-stage IVC obstruction or thrombosis after filter bioconversion or of thrombogenicity associated with retracted filter arms. Conclusions: The Sentry IVC filter provided safe and effective protection against PE, with a high rate of intended bioconversion and a low rate of device-related complications, through 2 years of follow-up. From the Department of Cardiothoracic Surgery (M.D.Dak.), Stanford University School of Medicine, Falk Cardiovascular Research Center, 300 Pasteur Drive, Stanford, CA 94305; Department of Vascular & Interventional Radiology (T.P.M.), Rhode Island Hospital, Providence, Rhode Island; Department of lica de Vascular & Endovascular Surgery (A.H.K.), Pontificia Universidad Cato Chile, Santiago, Chile; Department of Vascular & Interventional Radiology (M.D.Dar.), Washington University, St. Louis, Missouri; Department of Vascular & Interventional Radiology (L.E.S.), Adventist Midwest Health, Hinsdale, Illinois; Department of Vascular Surgery (M.A.C.), Rutgers–New Jersey Medical School, Newark, New Jersey; Department of Vascular & Interventional Radiology (M.S.J.), Indiana University, Indianapolis, Indiana; Department of Cardiac & Vascular Disease (F.A.), Lakeview Regional Heart Center, Covington, Louisiana; Department of Vascular & Interventional Radiology (J.L.S.), OSF Saint Francis Medical Center, Peoria, Illinois; Department of Interventional Cardiology & Vascular Medicine (G.M.A.), Riverside Methodist Hospital, Columbus, Ohio; Department of Interventional Cardiology &Vascular Medicine (M.J.S.), OhioHealth Heart and Vascular Physicians, Columbus, Ohio; Department of Interventional Radiology & Oncology (S.S.), University of Alabama, Birmingham, Alabama; Department of Vascular & Interventional Radiology (J.S.B.), Providence Sacred Heart Medical Center, Spokane, Washington; Department of Vascular Surgery (R.M.), UNC Rex Hospital, NC Heart and Vascular Research, Raleigh, North Carolina. Received June 13, 2019; final revision received August 12, 2019; accepted August 20,
2019. Address correspondence to M.D.Dak.; E-mail: [email protected]. edu M.D.Dak., M.A.C., and J.S.B. are paid consultants for BTG Vascular (Bothell, Washington). A.H.K. is a paid consultant for BTG Vascular and PQ Bypass (Sunnyvale, California). M.S.J. is a paid consultant for BTG Vascular, Boston Scientific (Marlborough, Massachusetts), and Cook Medical (Bloomington, Indiana). J.L.S. is a shareholder in Brightwater Medical (Murrieta, California). G.M.A. is a paid consultant for BTG Vascular, CR Bard (Murray Hill, New Jersey), and Cook Medical. M.J.S. is a paid consultant for Boston Scientific, Cook Medical, W. L. Gore & Associates (Flagstaff, Arizona), and Inari Medical (Irvine, California), and is a shareholder in Contego Medical (Raleigh, North Carolina). Appendices A–C and Tables E1–E3 can be found by accessing the online version of this article on www.jvir.org and clicking on the Supplemental Material tab. © SIR, 2019. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/). J Vasc Interv Radiol 2019; ▪:1–12 https://doi.org/10.1016/j.jvir.2019.08.036
2 ▪ Final Report: Sentry Bioconvertible IVC Filter
Dake et al ▪ JVIR
ABBREVIATIONS CEC ¼ clinical events committee, CI ¼ confidence interval, DVT ¼ deep vein thrombosis, FDA ¼ US Food and Drug Administration, IDE ¼ investigational device exemption, IVC ¼ inferior vena cava, PE ¼ pulmonary embolism, SAE ¼ serious adverse event, VTE ¼ venous thromboembolic
In appropriately chosen patients, systematic reviews and meta-analyses have shown that inferior vena cava (IVC) filters are effective at providing mechanical protection against pulmonary embolism (PE) (1–3). In response to adverse events with permanent IVC filters, retrievable IVC filters were developed to provide protection during recognized periods of transient PE risk, with the intent of removal in separate procedures (4). In practice, an unanticipated proportion of retrievable devices have remained in place after the need for temporary PE prophylaxis has passed. Retrieval rates rarely exceed 50%, and in some studies rates are as low as 8.5% (1,5,6). Retrievable devices that remain indwelling have been associated with time-dependent complications such as tilting, fracture, embolization, migration, and IVC perforation (5–8). When the removal of retrievable IVC filters is actually attempted, the procedures can be complex (particularly for devices that have been indwelling for longer periods), with a failure rate of approximately 10% (5). The Sentry IVC filter (BTG Vascular, Bothell, Washington) was developed as an alternative to retrievable filters for patients for whom anticoagulation is contraindicated and in whom the risk of PE is clearly transient. On the basis of evidence, in part from large-population studies of the incidence and timing of PE, suggesting that the period of highest PE risk occurs within the first 30 days after an index event for a large proportion of trauma and hospitalized surgery patients (9,10), the Sentry device was designed to bioconvert from a filtering to a nonfiltering configuration after 60 days, leaving a patent IVC lumen (11). The bioconversion of the Sentry device obviates the need for a separate secondary device-removal procedure. In the prospective multicenter SENTRY clinical trial, interim 1-year outcomes with the Sentry device included clinical success for 97.4% of evaluable patients (95% confidence interval [CI], 92.5%–99.1%); technical success in 129 of 130 attempted filter deployments; no new cases of symptomatic PE through 1 year of follow-up; no filter-related imaging complications; a 1.6% rate of filter-related symptomatic complications; and successful filter bioconversion for 96.4% (106/110) of patients (12). Reported here are final results from the SENTRY clinical trial through completed 24month clinical follow-up.
MATERIALS AND METHODS The SENTRY clinical trial design was previously reported (12). This prospective, multicenter, nonrandomized, singlearm trial was conducted at 23 sites in the United States
(n ¼ 20), Belgium (n ¼ 2), and Chile (n ¼ 1). Patients eligible for inclusion had documented deep vein thrombosis (DVT) or PE or a high risk of developing DVT or PE and had a contraindication to or failure of anticoagulation. The trial allowed enrollment of up to 40% of patients with a prophylactic indication (no current PE or DVT but high risk of PE). The indications for enrollment were consistent with American College of Radiology and Society of Interventional Radiology practice and quality improvement guidelines (Table E1 [available online on the article’s Supplemental Material page at www.jvir.org]) (13,14). The protocol was approved by the appropriate institutional review boards or ethics committees, and all study procedures were performed in accordance with the guidelines of good clinical practice and applicable regulations. BTG Vascular was the sole sponsor of the study, which was conducted under an investigational device exemption (IDE G110111), in compliance with applicable provisions of 21 CFR Parts 812, 50, and 54, and in accordance with the ethical principles of the Declaration of Helsinki. Data were collected on case report forms (with supervision by Vitruvian Clinical Research; San Ramon, California) and reviewed and adjudicated by an independent data monitoring committee and a clinical events committee (CEC), which included independent clinicians (an interventional radiologist, a vascular surgeon, and an interventional cardiologist) experienced in the use of IVC filters and the conduct of clinical trials. The Global Institute for Research (Richmond, Virginia) served as the imaging core laboratory, providing independent measurement and analysis in accordance with the predefined study protocols. QST Consultations (Allendale, Michigan) developed the statistical hypothesis for the study, and Advanced Research Associates (Santa Clara, California) defined the statistical analysis plan and methods and provided statistical analysis support. The study was registered before the start of patient enrollment (ClinicalTrials.gov ID NCT01975090).
Study Device, Procedures, Patient Characteristics, and Disposition The study device, the filter implantation, and the targeted bioconversion are summarized in Appendix A (available online on the article’s Supplemental Material page at www. jvir.org). A total of 63 investigators performed the implantations of Sentry IVC filters in 129 patients at 23 sites between September 2014 and February 2016. Study patient disposition through 24 months is detailed in Figure 1. Baseline characteristics of the patient population,
Volume ▪ ▪ Number ▪ ▪ Month ▪ 2019
3
Visual Synopsis
previously described (12), are summarized in Table 1. The patient indications for filter placement, the reasons for inability to use anticoagulation therapy, and key procedural details are summarized in Table 2, along with the primary 6-month composite endpoint results. The overall anticoagulation status of the patients, which changed as their conditions progressed during the course of the study,
is charted in Table E2 (available online on the article’s Supplemental Material page at www.jvir.org).
Study Follow-up and Imaging Evaluation After device implantation, according to a US Food and Drug Administration (FDA)-approved schedule, patients were evaluated at 1 month (30–44 days), 2 months (60–67 days),
4 ▪ Final Report: Sentry Bioconvertible IVC Filter
Dake et al ▪ JVIR
Figure 1. Disposition of the 129 enrolled patients through 24-month follow-up in the SENTRY clinical trial. PET ¼ positron emission tomography.
Volume ▪ ▪ Number ▪ ▪ Month ▪ 2019
5
Table 1. Baseline Demographics and Medical History Characteristic Mean age (years) Range
Table 2. Filter Indications, Placement, and Primary 6-Month Efficacy Endpoint
n ¼ 129 62.6 ± 13.52 21.0–88.0
Gender
Variables Indications for filter placement Current PE only
n ¼ 129 11 (8.5)
Male
73 (56.6)
Current DVT only
Female
56 (43.4)
Current DVT and PE
18 (14.0)
Prophylaxis*
42 (32.6)
2
BMI (kg/m ) Range Race White (non-Hispanic) Black
30.5 ± 8.38 17.3–78.1 107 (82.9) 11 (8.5)
Hispanic
9 (7.0)
Unknown
2 (1.6)
Medical history/risk factors Hypertension
76 (58.9)
Recent surgery (30 days)
33 (25.6)
Diabetes
28 (21.7)
Note–Values are presented as mean ± standard deviation. Values in parentheses are percentages. BMI ¼ body mass index.
6 months (150–210 days), 12 months (335–395 days), and 24 months (670–790 days). Each follow-up visit included clinical assessment for symptoms of PE and DVT, monitoring of adverse events, and assessment of venous thromboembolic (VTE) risk factors. Imaging for filter configuration and complications was performed by ultrasound of the lower extremities and computed tomography (CT) venography at 1 month, and by anterior-posterior and lateral x-ray (or by CT venography if thrombus was observed at 1 month) at 2 months. Imaging for filter bioconversion status and complications was performed by CT at 6 months, by anterior-posterior and lateral x-ray at 12 months, and by CT venography at 24 months.
Study Endpoints Study endpoints are specified and defined in Appendix B (available online on the article’s Supplemental Material page at www.jvir.org). Study endpoints were formulated in accordance with Society of Interventional Radiology reporting standards (15) and American College of Radiology guidelines (13), with reference to recent IDE studies (16–19) and the PRESERVE trial (NCT02381509), which was also designed in conjunction with the FDA, following FDA safety communications in 2010 and 2014 responding to the reported high rates of complications associated with prolonged indwelling of retrievable IVC filters and emphasizing the need for vigilant follow-up (20,21). Clinical assessment for potential PE-related symptoms was performed at each follow-up timepoint, and suspicion of PE was supported by appropriate imaging. For adjudication of the procedure- and device-relatedness of potential serious adverse events (SAEs), cases were presented to the CEC as
58 (45.0)
Primary factor for filter placement Surgery
77 (59.7)
Medical condition
28 (21.7)
Trauma Other
8 (6.2) 16 (12.4)
Reasons for inability to use anticoagulation therapy Inability to use anticoagulation therapy Risk of bleeding and/or injury from anticoagulation Contraindication
129 (100.0) 79 (61.2) 49 (38.0)
Access site for filter placement Right internal jugular vein
64 (49.6)
Right femoral vein
54 (41.9)
Left femoral vein
11 (8.5)
Pre-implant IVC diameter (mm) Mean ± SD Range Primary composite endpoint of clinical success at 6 months
19.3 ± 2.23 14.0–26.0 111/114 (97.4)
Technical success of filter deployment
129/130 (99.2)
Freedom from symptomatic PE during the 60-day protection period
129/129 (100)
Freedom from IVC filter-related complications at 6 months
112/114 (98.2)
Note–Values in parentheses are percentages. DVT ¼ deep vein thrombosis; IVC ¼ inferior vena cava; PE ¼ pulmonary embolism; SD ¼ standard deviation. *Prophylactic indication: no current PE or DVT but high risk of PE.
clinical summaries, including drug history and available autopsy reports, with imaging.
Statistical Analysis Endpoints were analyzed per patient on an intent-to-treat basis. Continuous variables were described as mean ± standard deviation, and categorical variables were described as counts and percentages. For the statistical analysis on the primary endpoint of clinical success at 6 months, the observed rate was tested against the acceptance criterion using the 2-sided 95% Wilson CI for the binomial proportion. If the lower confidence limit for the true proportion equaled or exceeded 80%, the endpoint was deemed to be successfully achieved. For the secondary endpoints and other clinical outcomes data, the number and percentage of observed patients for each endpoint were determined, and,
6 ▪ Final Report: Sentry Bioconvertible IVC Filter
Dake et al ▪ JVIR
Figure 2. Filtering configuration and bioconversion status through 24 months in the SENTRY clinical trial. Filtering configuration ¼ all 6 pairs of arms held in the central portion of the IVC lumen. Bioconverted ¼ 1 or more pairs of filter arms separated from the central portion of the IVC lumen.
where appropriate, the 2-sided 95% CI using the Wilson score interval around the proportion was calculated.
RESULTS Outcomes through 1 year of follow-up in the SENTRY clinical trial, which have been previously presented (12), are summarized in Appendix C (available online on the article’s Supplemental Material page at www.jvir.org).
Filter Bioconversion Figure 2 shows the bioconversion status for all patients with imaging assessment at 6, 12, and 24 months. Figure 3 shows representative CT imaging for a single patient through 24-month follow-up. Filter bioconversion was successful for 95.7% (110/115) of patients at 6 months, 96.4% (106/110) of patients at 12 months, and 96.5% (82/85) of patients at 24 months. Through the 24-month follow-up, no new DVT, PE, or IVC filter-related complications were reported in any of the 5 patients whose devices were not bioconverted at 6 months. Through 24 months of follow-up, there was no evidence of late-stage IVC obstruction or thrombosis after filter bioconversion or of thrombogenicity associated with retracted filter arms.
VTE Outcomes through 24 Months All patients available for follow-up were reported to be free from new symptomatic PE through 12 months (n ¼ 111) (Table 3). Between 12 and 24 months, new symptomatic nonfatal cases of PE were identified in 2 of 85 evaluable
patients (2.4%). In 1 of these patients, with new emboli found at 581 days in both the right and left pulmonary arteries, the study device was determined to have been bioconverted at 6 months, and the filter arms were fully retracted to the IVC wall. The new PE was successfully treated and resolved with anticoagulation and thrombolysis. The other case of new PE in the second year of follow-up occurred at day 624 after uncomplicated implantation of a Sentry device in a patient with DVT who had suffered severe gastrointestinal bleeding while on anticoagulation. In this patient, a small clot was identified in the filter at 1 month after implantation, for which no intervention was undertaken, and 6-month imaging showed that the filter had bioconverted, with the filter arms fully retracted to the IVC wall, and no incident of PE. At 624 days, while receiving low-molecular-weight heparin, the patient developed new DVT in the right femoral and popliteal veins. After the patient developed dyspnea, CT pulmonary angiography showed a saddle PE in both lungs extending into the segmental pulmonary arteries, with no cardiac abnormalities. The DVT and the PE were successfully treated with anticoagulation. In both of the cases of new symptomatic nonfatal PE during the second year of follow-up, the core laboratory and CEC review at 24 months adjudicated the events as not related to the study device or procedure. Through the 24-month follow-up, as adjudicated by the CEC, in 14 patients, there were 12 cases of new DVT (8 symptomatic and 4 asymptomatic) and 5 cases of worsening DVT (3 symptomatic and 2 asymptomatic) (Table 3). No DVTs were considered to be device related, and 1 (previously described) new symptomatic DVT at 8 days was
Volume ▪ ▪ Number ▪ ▪ Month ▪ 2019
7
Figure 3. Representative CT imaging for a single patient. Left: Coronal image acquired as part of a helical dataset 1 month after filter placement, showing the Sentry device in filtering configuration. Center: Coronal image at 24-month follow-up, showing the device in bioconverted configuration and endothelialized. Top right: Axial view of the bioconverted device at 6 months at the level of the filter tips. Bottom right: Axial view of the bioconverted device at 24 months at the level of the filter tips.
Table 3. IVC Filter-Related Complications and VTE Outcomes through 24 Months Filter-related imaging complications
1 mo n ¼ 122
2 mo n ¼ 119
6 mo n ¼ 114
12 mo n ¼ 111
24 mo n ¼ 85
Tilting
0 (0)
0 (0)
0 (0)
0 (0)
0 (0)
Migration
0 (0)
0 (0)
0 (0)
0 (0)
0 (0)
Perforation
0 (0)
0 (0)
0 (0)
0 (0)
0 (0)
Embolization
0 (0)
0 (0)
0 (0)
0 (0)
0 (0)
Fracture
0 (0)
0 (0)
0 (0)
0 (0)
0 (0)
Any imaging complication
0 (0)
0 (0)
0 (0)
0 (0)
0 (0)
Filter-related symptomatic complications
0–1 mo n ¼ 129
Filter-related death Symptomatic caval thrombosis
0 (0) 2 (1.6)*
0 (0) 0 (0)
0 (0) 0 (0)
0 (0) 0 (0)
0 (0) 0 (0)
Other symptomatic complications requiring invasive intervention
0 (0)
0 (0)
0 (0)
0 (0)
0 (0)
CEC Adjudicated VTE Outcomes
0–1 mo n ¼ 129
New symptomatic DVT
3 (2.3)
New asymptomatic DVT
4 (3.1)
Symptomatic worsening DVT Asymptomatic worsening DVT New symptomatic PE
1–2 mo n ¼ 127
1–2 mo n ¼ 127
2–6 mo n ¼ 126
6–12 mo n ¼ 111
12–24 mo n ¼ 85
2–6 mo n ¼ 126
6–12 mo n ¼ 111
12–24 mo n ¼ 85
0 (0)
1 (0.8)
1 (0.9)
3 (3.5)
0 (0)
0 (0)
0 (0)
0 (0)
1 (0.8) 1 (0.8)
0 (0) 1 (0.8)
0 (0) 0 (0)
1 (0.9) 0 (0)
1 (1.2) 0 (0)
0 (0)†
0 (0)
0 (0)
0 (0)
2 (2.4)†
Note–Values in parentheses are percentages. DVT ¼ deep vein thrombosis; IVC ¼ inferior vena cava; PE ¼ pulmonary embolism; VTE ¼ venous thromboembolism. *Two cases of symptomatic caval thrombosis developed and were successfully treated during the first month of follow-up. † Two cases of new symptomatic PE developed at 581 and 624 days after implantation and after IVC filter bioconversion. The PE events between 12 and 24 months were adjudicated by the clinical events committee as not device related and not procedure related.
8 ▪ Final Report: Sentry Bioconvertible IVC Filter
adjudicated as being related to the procedure (12). Of the cases of new symptomatic DVT, 3 were recorded during the second year of follow-up—at days 395, 411, and 624 (discussed above)—in patients who had been symptomatic at baseline with known DVT or PE. In the case recorded at day 395, the patient (whose device remained non-converted through 24month follow-up) was not receiving anticoagulation, and compression stockings were prescribed. The case recorded at day 411 involved bilateral DVT in the popliteal and calf veins in a patient who had also experienced a new symptomatic DVT at 224 days and a worsening symptomatic DVT at 351 days; the patient had received an additional IVC filter at 147 days above the bioconverted Sentry device. The bilateral DVTs resolved with physical therapy and anticoagulation. Of the cases with symptomatic worsening DVT, 1 occurred during the second year, at day 580 in the same patient who experienced a new symptomatic nonfatal PE at 581 days. The DVT, in the left femoral and popliteal veins, was treated with anticoagulation and physical therapy. Through 24 months, 3 (2.3%) patients developed new transient VTE risks requiring the placement of an additional IVC filter chosen at the discretion of the investigator (the study device was not available for secondary implantation). Each of these cases occurred beyond the 60-day targeted study-filter protection period, and none was placed to extend the initial risk period. Two of the cases, previously described (12), occurred during the first 6 months of follow-up, with the new risk due to severe epistaxis in 1 case and to hemorrhagic brain metastasis in the other. In each of these 2 cases, the additional filter was placed above the bioconverted study device. The third case occurred at 613 days in a patient originally enrolled due to a history of DVT after total knee replacement who returned to undergo gastric bypass revision. In this case, the additional filter (Cook Celect) was placed inside the bioconverted study device. In all 3 cases, CEC adjudication confirmed that the additional device placement was required due to new VTE risk and not to the study device or procedure or to extend the index protection period.
Filter-Related Complications and Safety Outcomes through 24 Months Through 24-month follow-up, there were no further imagingidentified or symptomatic IVC filter-related complications besides the 2 cases of symptomatic caval thrombosis that were reported and successfully treated during the first month after implantation (Table 3) (12). There were no cases of surgical removal of the filter and/or vascular repair of the IVC, and there were no deaths adjudicated as filter related. On all CT imaging performed at any timepoint, no instances of filter perforation of the IVC wall were observed. In the entire safety analysis population of 129 patients, 1 or more adverse events were experienced by a total of 104 (80.6%) during the 24 months of follow-up, and 1 or more SAEs were experienced by 63 (48.8%) patients (Table E3 [available online on the article’s Supplemental Material page at www.jvir.org]). The most common SAEs were infections (17.8%). None of the SAEs was confirmed by the
Dake et al ▪ JVIR
CEC to be related to the study device. None of the 12 deaths in patients eligible for follow-up through 24 months (due to cancer [n ¼ 3], cardiopulmonary arrest [n ¼ 2], respiratory failure [n ¼ 3], multi-organ failure, hypertensive cardiovascular disease, liver failure, or myasthenia gravis) was related to the Sentry filter or to the index procedure, as reported by the sites and confirmed by CEC adjudication.
DISCUSSION For the patients in the SENTRY trial—all of whom had documented DVT or PE or were at high risk of developing DVT or PE and had a contraindication to or failure of anticoagulation—freedom from new symptomatic PE was 100% through the first year of follow-up, an outcome comparing favorably with PE rates in recent trials of retrievable IVC filters (16–19,22). Between 12 and 24 months, new symptomatic nonfatal PE was identified in 2 of 85 evaluable patients (2.4%), in each of whom the 6-month imaging had shown the study device to be bioconverted with the filter arms fully retracted to the IVC wall. In each of these cases, the new PE was successfully resolved. Both of these events occurred well past the index period of early transient risk, and neither of the patients had an indication for a permanent IVC filter. Both cases were adjudicated by the CEC as not related to the study device or procedure. The situation in these 2 patients was similar to one in which a retrievable filter would have previously and appropriately been removed. The timing of these 2 new cases of PE in patients, in whom the Sentry device had already bioconverted at 6 months, would strongly indicate that they were not due to release of entrapped thrombus. Such a low rate of late PE would suggest that physicians were capable of identifying patients who had a defined period of risk from PE and who did not require placement of a permanent IVC filter. Through the 24-month follow-up, there were a total of 17 cases of CEC-confirmed new or worsening DVT in 14 patients (10.9%), which also compared favorably with recent findings for retrievable filters (16–19,22). Only the 1 new symptomatic access-site DVT that occurred at 8 days was considered to be procedure related, whereas none was confirmed as being device related. The bioconversion of the Sentry IVC filter—targeted to occur after 60 days after implantation, after the period of transient PE risk—and subsequent endothelialization obviate the need for, and avoid any potential for complications that would be associated with, a second separate device-removal procedure. Also potentially avoided were the risks of time-dependent complications such as have been reported in association with retrievable IVC filters that remain indwelling (5–7). The confirmed execution of the mandated 1-month, 2-month, 6month, 12-month, and 24-month CT, CT venography, and xray examinations documented the success of device bioconversion and the absence of device-related imaging-identified complications or issues with the filter-frame integrity through 24 months—no cases of tilting, migration, perforation,
Volume ▪ ▪ Number ▪ ▪ Month ▪ 2019
embolization, or fracture. There were no filter-related deaths through 24 months of follow-up. Through 24 months of followup, there was no evidence of late-stage IVC obstruction or thrombosis after filter bioconversion or of thrombogenicity associated with retracted filter arms, which was a finding that confirmed the results of preclinical studies with the device in an experimental ovine model (23). There was no PE experienced by any patient who had thrombus in the filter in the protocolmandated 1-month CT venography (see Appendix C [available online on the article's Supplemental Material page at www.jvir. org]). The Sentry IVC filter was designed not to tilt, and any trapped thrombus is thus retained in the center of the vessel where the body’s own lysis capabilities may be maximized. It is unlikely that the trapped clot is “free floating,” as it tends to become adherent to the filter elements. Through 2 years, 3 patients required the placement of an additional IVC filter beyond the 60-day protection period and after the index device had bioconverted. In each of these cases, the CEC confirmed that the new filter was required due to a new distinct VTE risk and not to the study device or to extend the index protection period. The secondary filter placements in response to new VTE risk did not take place until well after the targeted 60-day protection period, suggesting (as does the low rate of late PE) that all of the Sentry trial implanting physicians were successful in identifying patients whom they felt would benefit from temporary protection due to identification of transient risk based on individual case presentations. As is the case with other IVC filters, data are lacking regarding the safety and efficacy of the Sentry bioconvertible filter in hypercoagulable patients or those requiring multiple surgeries over short intervals. However, along with retrievable filters, which are still often used in these cases, the Sentry IVC filter can be considered in such temporary filtration scenarios. Limitations of the present study include the nonrandomized single-arm design and the inherent potential for bias in a manufacturer-funded regulatory device trial. It is possible that instances of asymptomatic PE may have gone undetected (consistent with other IVC filter studies), as imaging was performed only in patients with suggestive clinical symptoms. Likewise, although 6 of the 17 detected cases of new and worsening DVT were asymptomatic, after the 1-month ultrasonography there was no mandated lower extremity imaging, and DVT status was assessed based on symptoms and any site-performed imaging that was part of follow-up of high-risk patients. Although the 32.6% proportion of patients with prophylactic (no current PE or DVT but high risk of PE) indications for study device implantation was consistent with the enrollment in previous IDE filter studies and established guidelines, the ability of the filter to capture already present emboli of lower extremity origin could not be observed in those patients. This study offers no evidence to suggest that a bioconverted Sentry device would pose a technical challenge to placement of an additional filter in a situation of new VTE risk, but the concern with such new filter placement could be greater in the rare case of a Sentry device remaining nonconverted.
9
Although short-term outcomes are promising, data are needed regarding any effect of the long-term presence of the bioconverted and endothelialized Sentry device. The SENTRY clinical trial has demonstrated that for patients requiring temporary protection against PE and unable to use anticoagulation, the Sentry IVC filter provides high rates of technical and clinical success with minimal complications, within the efficacy and safety thresholds suggested by the Society of Interventional Radiology, and outcomes through 2 years that compare favorably with those reported for retrievable IVC filters. The results of this trial confirm that the Sentry bioconvertible device can provide an effective alternative to existing retrievable IVC filters that often remain indwelling long after the period of transient PE risk has passed and that are associated with high complication rates.
ACKNOWLEDGMENTS ClinicalTrials.gov ID NCT01975090. The study was funded by BTG Vascular (Bothell, Washington). The authors thank Diane Gargus for clinical trial management. Writing support services were provided by Galen Press, Inc (Austerlitz, New York) and paid for by the study sponsor. The following investigators and institutions participated in the SENTRY Clinical Trial (Michael D. Dake, MD, principal investigator) in addition to the authors: Robert Feezor, MD (University of Florida, Gainesville, FL); Sanjeeva Kalva, MD (UT Southwestern Medical Center, Dallas, TX); Darren Kies, MD (Emory University, Atlanta, GA); Marc Bosiers, MD (AZ Sint-Blasius, Dendermonde, Belgium); Werner Ziegler, MD (Memorial Hospital of Colorado Springs, Colorado Springs, CO); Mark Farber, MD (University of North Carolina, Chapel Hill, NC); David Paolini, MD (Jobst Vascular Institute, Toledo, OH); Robert Spillane, MD (Hartford Hospital, Hartford, CT); Steven Jones, MD (Cardiovascular Associates of the Southeast, LLC, Birmingham, AL); and Patrick Peeters, MD (Imelda Hospital, Bonheiden, Belgium).
REFERENCES 1. Angel LF, Tapson V, Galgon RE, Restrepo MI, Kaufman J. Systematic review of the use of retrievable inferior vena cava filters. J Vasc Interv Radiol 2011; 22:1522–1530 e1523. 2. Bikdeli B, Chatterjee S, Desai NR, et al. Inferior vena cava filters to prevent pulmonary embolism: systematic review and meta-analysis. J Am Coll Cardiol 2017; 70:1587–1597. 3. Haut ER, Garcia LJ, Shihab HM, et al. The effectiveness of prophylactic inferior vena cava filters in trauma patients: a systematic review and meta-analysis. JAMA Surg 2014; 149:194–202. 4. Kinney TB. Update on inferior vena cava filters. J Vasc Interv Radiol 2003; 14:425–440. 5. Jia Z, Fuller TA, McKinney JM, et al. Utility of retrievable inferior vena cava filters: a systematic literature review and analysis of the reasons for nonretrieval of filters with temporary indications. Cardiovasc Intervent Radiol 2018; 41:675–682. 6. Sarosiek S, Crowther M, Sloan JM. Indications, complications, and management of inferior vena cava filters: the experience in 952 patients at an academic hospital with a level I trauma center. JAMA Intern Med 2013; 173:513–517.
10 ▪ Final Report: Sentry Bioconvertible IVC Filter
7. Deso SE, Idakoji IA, Kuo WT. Evidence-based evaluation of inferior vena cava filter complications based on filter type. Semin Intervent Radiol 2016; 33:93–100. 8. US Food and Drug Administration. Removing retrievable inferior vena cava filters: FDA safety communication. Issue date: May 6, 2014. Available at: https://wayback.archive-it.org/7993/20161022044053/http:// www.fda.gov/MedicalDevices/Safety/AlertsandNotices/ucm396377.htm. Accessed March 1, 2019. 9. Coleman JJ, Zarzaur BL, Katona CW, et al. Factors associated with pulmonary embolism within 72 hours of admission after trauma: a multicenter study. J Am Coll Surg 2015; 220:731–736. 10. Hope WW, Demeter BL, Newcomb WL, et al. Postoperative pulmonary embolism: timing, diagnosis, treatment, and outcomes. Am J Surg 2007; 194:814–818; discussion 818–819. 11. Dake MD, Ansel GM, Johnson MS, Mendes R, Smouse HB. The clinical rationale for the Sentry Bioconvertible Inferior Vena Cava Filter for the prevention of pulmonary embolism. Int J Vasc Med 2019; 2019:16. 12. Dake MD, Murphy TP, Kr€amer AH, et al. One-year analysis of the prospective multicenter SENTRY clinical trial: safety and effectiveness of the Novate Sentry Bioconvertible Inferior Vena Cava Filter. J Vasc Interv Radiol 2018; 10:1350–1361. 13. ACR-SIR-SPR Practice Parameter for the Performance of Inferior Vena Cava (IVC) Filter Placement for the Prevention of Pulmonary Embolism. Revised; 2016. Available at: https://www.acr.org/-/media/ACR/Files/PracticeParameters/ivc-fliterplacement.pdf?la¼en. Accessed March 1. 2019. 14. Caplin DM, Nikolic B, Kalva SP, et al. Quality improvement guidelines for the performance of inferior vena cava filter placement for the prevention of pulmonary embolism. J Vasc Interv Radiol 2011; 22: 1499–1506. 15. Millward SF, Grassi CJ, Kinney TB, et al. Reporting standards for inferior vena caval filter placement and patient follow-up: supplement for
Dake et al ▪ JVIR
16.
17.
18.
19.
20.
21.
22.
23.
temporary and retrievable/optional filters. J Vasc Interv Radiol 2009; 20: S374–S376. Stavropoulos SW, Sing RF, Elmasri F, et al. The DENALI Trial: an interim analysis of a prospective, multicenter study of the Denali retrievable inferior vena cava filter. J Vasc Interv Radiol 2014; 25:1497–1505, 1505,e1. Stavropoulos SW, Chen JX, Sing RF, et al. Analysis of the final DENALI trial data: a prospective, multicenter study of the Denali inferior vena cava filter. J Vasc Interv Radiol 2016; 27:1531–1538.e1. Smouse HB, Mendes R, Bosiers M, et al. The RETRIEVE trial: safety and effectiveness of the retrievable crux vena cava filter. J Vasc Interv Radiol 2013; 24:609–621. Johnson MS, Nemcek AA Jr, Benenati JF, et al. The safety and effectiveness of the retrievable Option inferior vena cava filter: a United States prospective multicenter clinical study. J Vasc Interv Radiol 2010; 21: 1173–1184. US Food and Drug Administration. Removing retrievable inferior vena cava filters: initial communation. Issue date: August 9, 2010. Available at: https://wayback.archive-it.org/7993/20161022180008/http:/www.fda.gov/ MedicalDevices/Safety/AlertsandNotices/ucm221676.htm. Accessed March 1, 2019. US Food and Drug Administration. Removing retrievable inferior vena cava filters: FDA Safety Communication. Updated May 6, 2014, Available at: https://www.fda.gov/MedicalDevices/Safety/AlertsandNotices/ucm39 6377.htm. Accessed March 1, 2019. Lyon SM, Riojas GE, Uberoi R, et al. Short- and long-term retrievability of the Celect vena cava filter: results from a multi-institutional registry. J Vasc Interv Radiol 2009; 20:1441–1448. Gaines PA, Kolodgie FD, Crowley G, et al. Sentry Bioconvertible Inferior Vena Cava Filter: study of stages of incorporation in an experimental ovine model. Int J Vasc Med 2018; 2018:10.
Volume ▪ ▪ Number ▪ ▪ Month ▪ 2019
APPENDIX A. DEVICE DESIGN, IMPLANTATION, AND BIOCONVERSION The Sentry bioconvertible IVC filter (BTG Vascular, Bothell, Washington) is designed to provide temporary protection against PE during transient high-risk periods and then to bioconvert after 60 days after implantation, thus obviating the need for device retrieval in a secondary procedure. The Sentry IVC filter is made from a single piece of laser-cut nitinol, which is formed into a cylindrical frame with an integral filter cone consisting of 6 pairs of arms held together in the center of the IVC by means of a bioabsorbable filament composed of poly-p-dioxanone, a biodegradable synthetic polymer. The Sentry is designed to provide filtering protection for a minimum duration of 60 days and then to bioconvert to a nonfiltering configuration. Upon deployment, the cylindrical frame expands to appose the IVC wall, inciting its incorporation into the caval wall by means of neointimal healing. During bioconversion, the bioabsorbable filament hydrolyzes, releasing the filtering arms from the filtering cone. The filtering arms then retract to the IVC wall into a nonfiltering configuration, where they also are endothelialized, and IVC lumen patency is restored. The Sentry is indicated for use in IVCs with diameters between 16 and 28 mm and has a maximum deployed length of 57.7 mm. Contemporary data from an analysis of 17 largepopulation studies of the incidence and timing of PE in post-injury trauma patients and major surgery patients support the Sentry design premise that the period of highest risk for PE in such patients with temporary contraindications to anticoagulants occurs early, with most PEs diagnosed within 30 days of an index event (hospitalization, traumatic injury, or surgery) (11).
APPENDIX B. STUDY ENDPOINTS The predefined primary endpoint of the Sentry clinical trial was clinical success at 6 months, which was a composite of technical success (filter deployment as intended without acute events), freedom from symptomatic PE through 60 days, and 6-month freedom from filter-related complications. Secondary efficacy endpoints included the technical success rate at day 0; filter status at months 1 and 2 (the percentage of patients with devices in filtering configuration, based on all 6 pairs of arms being held together in the central portion of the IVC lumen, and the percentage in nonfiltering configuration with arms separated from the central portion of the lumen); bioconversion status at months 6, 12, and 24 (the percentage bioconverted and the percentage not converted); and new symptomatic PE through 6, 12, and 24 months. Clinical assessment for potential PE-related symptoms was performed at each followup timepoint, and suspicion of PE was supported by appropriate imaging.
10.e1
Secondary safety endpoints included freedom from filterrelated complications on day 0 and at months 1, 2, 6, 12, and 24; invasive filter interventions such as thrombolysis, thrombectomy, surgical removal of the filter, placement of a second filter, and/or vascular repair of the IVC; and placement of an additional IVC filter (categorized as filter related or for extension of protection from PE beyond 60 days). For adjudication of the procedure- and device-relatedness of potential SAEs, cases were presented to the CEC as clinical summaries, including drug history and available autopsy reports, with imaging. Other clinical outcomes included DVT of the lower extremities; the assessment of VTE risk factors; and anticoagulation status on day 0 and at months 1, 2, 6, 12, and 24.
APPENDIX C. SENTRY CLINICAL TRIAL OUTCOMES THROUGH THE FIRST YEAR OF FOLLOW-UP A Sentry device was successfully implanted in all 129 patients enrolled in the trial; and after deployment in all patients, the protocol-mandated venography verified that all implanted filters were in the intended location and in the filtering configuration with the IVC patent. At the 1-month follow-up, 100% (119/119) of the devices were in filtering configuration; at the 2-month follow-up, 95.3% (101/106) of the devices remained in filtering configuration. There was no evidence that any device had bioconverted before 60 days after implantation. Subsequent imaging confirmed that successful device bioconversion had occurred for 95.7% of patients by 6 months, for 96.4% of patients by 12 months, and for 96.5% of patients by 24 months. All 3 of the component criteria were met by 111 (97.4%) of 114 patients evaluable for the composite primary clinical success endpoint (95% CI, 92.5%–99.1%), exceeding the predefined acceptance criterion of 80%. Technical success of deployment was achieved in 99.2% (129/130) of deployment attempts (95% CI, 95.8%–99.9%). The rate of freedom from new symptomatic PE through 60 days was 100% (n ¼ 129; 95% CI, 97.1%–100.0%), and the rate of freedom from IVC filter-related complications through 6 months was 98.2% (112/114; 95% CI, 93.8%–99.5%), based on findings of symptomatic caval thrombosis in 2 patients during the first month of follow-up (adjudicated by the CEC as SAEs having an unknown relationship to the device and procedure). After successful treatment with thrombectomy and thrombolysis in both cases of symptomatic caval thrombosis, the 2-month follow-up by x-ray and/or CT venography confirmed that the filter was in correct filtering configuration, and there was no recurrence of the caval thrombosis in either patient. The protocol-mandated 1-month CT venography revealed the presence of thrombus in the filters of 18 (15.8%) of 114 patients with core laboratory review. The thrombus was symptomatic only in the 2 successfully treated cases of caval
10.e2 ▪ Final Report: Sentry Bioconvertible IVC Filter
thrombosis. No patient with thrombus in the filter at 1 month went on to experience PE during the next 30 days. There were no reports of symptomatic IVC stenosis caused by the device. Filter bioconversion occurred for 95.7% of patients by 6 months, for 96.4% of patients by 12 months, and for 96.5% of patients by 24 months. Through 24 months of follow-up, there was no evidence of late-stage IVC obstruction or
Table E1. SIR Standards of Practice Committee Classification of Indications for IVC Filter Placement* Therapeutic indications (documented thromboembolic disease) Evidence of PE or IVC/iliac/femoropopliteal DVT and 1 or more of the following: Absolute or relative contraindication to anticoagulation Complications of anticoagulation Failure of anticoagulation Recurrent PE despite adequate therapy Inability to achieve/maintain adequate anticoagulation Propagation/progression of DVT during therapeutic anticoagulation Massive PE with residual DVT in a patient at risk for further PE Free-floating ileofemoral or IVC thrombus Severe cardiopulmonary disease and DVT (eg, corpulmonale with pulmonary hypertension) Prophylactic indications (no current thromboembolic disease) Temporary risk of PE in 1 of the following settings: Severe trauma without documented PE or DVT Closed head injury Spinal cord injury Multiple long-bone or pelvic fractures High-risk situations (eg, patient immobilized or in ICU) Note–Values in parentheses are percentages. DVT ¼ deep vein thrombosis; ICU ¼ intensive care unit; IVC ¼ inferior vena cava; PE ¼ pulmonary embolism. *Caplin DM, Nikolic B, Kalva SP, et al. Quality improvement guidelines for the performance of inferior vena cava filter placement for the prevention of pulmonary embolism. J Vasc Interv Radiol 2011; 22:1499–1506.
Dake et al ▪ JVIR
thrombosis after filter bioconversion or of thrombogenicity associated with retracted filter arms. In the unusual event of non-bioconversion of the Sentry filter, a subsequent interventional procedure to convert the Sentry into a nonfiltering configuration is not recommended. The Sentry device was designed not to tilt, with the intention that any trapped thrombus would be central within the vessel where the body’s own lysis capabilities can be maximized.
Volume ▪ ▪ Number ▪ ▪ Month ▪ 2019
10.e3
Table E2. Patient Anticoagulation Status through 24 Months Days –7 to –1 N in interval
Device Deployment Day 0
129
Days 1–7
Month 1 Month 2 Months 2–6 Months 6–12 (Days 8–44) (Days 45–67) (Days 68–210) (Days 211–395)
Months 12–24 (Days 396–760)
129
129
128
127
126
117
85
95 (74%) n (%) with no anticoagulation use for all or part of the interval
72 (56%)
80 (62%)
65 (51%)
42 (33%)
72 (57%)
69 (59%)
55 (65%)
n (%) with 34 (26%) continuous anticoagulation for the interval
57 (44%)
49 (38%)
63 (49%)
85 (66%)
54 (42%)
48 (41%)
30 (35%)
Note–Values in parentheses are percentages. Anticoagulants included heparin, factor Xa inhibitors, direct thrombin inhibitor, factor II inhibitor, and Coumadin derivatives. A patient could use more than 1 anticoagulant medication within the same time interval. At study baseline, all 129 enrolled patients were reported by the sites to have permanent, temporary, or predicted contraindications to anticoagulation. The overall anticoagulation status of the patients changed as their conditions progressed during the course of the study. The proportion of patients who had interruption of anticoagulant use was greatest during the first 7 days after implantation, which was likely to be the period of highest risk for many. More patients were treated continuously with anticoagulation between day 8 and day 67, as they no longer had contraindications. Between days 68 and 210, the proportion of patients on continuous anticoagulation was lower, as was expected due to the passing of the period of transient risk of pulmonary embolism.
Table E3. SAEs by System Organ Class through 24-Month Follow-up Category
Patients with ≥1 SAE
All patients, all SAEs
63 (48.8)
Blood and lymphatic system disorders Cardiac disorders
5 (3.9) 12 (9.3)
Congenital, familial, and genetic disorders
2 (1.6)
Gastrointestinal disorders
8 (6.2)
General disorders and administration site conditions
6 (4.7)
Hepatobiliary disorders Infections and infestations Injury, poisoning, and procedural complications
1 (0.8) 23 (17.8) 9 (7.0)
Metabolism and nutrition disorders
4 (3.1)
Musculoskeletal and connective tissue disorders
6 (4.7)
Neoplasms benign, malignant, and unspecified (including cysts and polyps)
8 (6.2)
Nervous system disorders
6 (4.7)
Psychiatric disorders
1 (0.8)
Renal and urinary disorders
7 (5.4)
Respiratory, thoracic, and mediastinal disorders Skin and subcutaneous tissue disorders
12 (9.3) 1 (0.8)
Surgical and medical procedures
4 (3.1)
Vascular disorders
7 (5.4)
Note–Values in parentheses are percentages. SAE ¼ serious adverse event.
Final Two-Year Outcomes for the Sentry Bioconvertible Inferior Vena Cava Filter in Patients Requiring Temporary Protection from Pulmonary Embolism €mer, MD, Michael D. Dake, MD, Timothy P. Murphy, MD, Albrecht H. Kra Michael D. Darcy, MD, Luke E. Sewall, MD, Michael A. Curi, MD, Matthew S. Johnson, MD, Frank Arena, MD, James L. Swischuk, MD, Gary M. Ansel, MD, Mitchell J. Silver, DO, Souheil Saddekni, MD, Jayson S. Brower, MD, and Robert Mendes, MD; for the SENTRY Trial Investigators
ABSTRACT Purpose: To report final 2-year outcomes with the Sentry bioconvertible inferior vena cava (IVC) filter in patients requiring temporary protection against pulmonary embolism (PE). Materials and Methods: In a prospective multicenter trial, the Sentry filter was implanted in 129 patients with documented deep vein thrombosis (DVT) and/or PE (67.5%) or who were at temporary risk of developing DVT/PE (32.6%). Patients were monitored and bioconversion status ascertained by radiography, computed tomography (CT), and CT venography through 2 years. Results: The composite primary 6-month endpoint of clinical success was achieved in 97.4% (111/114) of patients. The rate of new symptomatic PE was 0% (n ¼ 126) through 1 year and 2.4% (n ¼ 85) through the second year of follow-up, with 2 new nonfatal cases at 581 and 624 days that were adjudicated as not related to the procedure or device. Two patients (1.6%) developed symptomatic caval thrombosis during the first month and underwent successful interventions without recurrence. No other filter-related symptomatic complications occurred through 2 years. There was no filter tilting, migration, embolization, fracture, or caval perforation and no filterrelated deaths through 2 years. Filter bioconversion was successful for 95.7% (110/115) of patients at 6 months, 96.4% (106/110) of patients at 12 months, and 96.5% (82/85) of patients at 24 months. Through 24 months of follow-up, there was no evidence of late-stage IVC obstruction or thrombosis after filter bioconversion or of thrombogenicity associated with retracted filter arms. Conclusions: The Sentry IVC filter provided safe and effective protection against PE, with a high rate of intended bioconversion and a low rate of device-related complications, through 2 years of follow-up. From the Department of Cardiothoracic Surgery (M.D.Dak.), Stanford University School of Medicine, Falk Cardiovascular Research Center, 300 Pasteur Drive, Stanford, CA 94305; Department of Vascular & Interventional Radiology (T.P.M.), Rhode Island Hospital, Providence, Rhode Island; Department of lica de Vascular & Endovascular Surgery (A.H.K.), Pontificia Universidad Cato Chile, Santiago, Chile; Department of Vascular & Interventional Radiology (M.D.Dar.), Washington University, St. Louis, Missouri; Department of Vascular & Interventional Radiology (L.E.S.), Adventist Midwest Health, Hinsdale, Illinois; Department of Vascular Surgery (M.A.C.), Rutgers–New Jersey Medical School, Newark, New Jersey; Department of Vascular & Interventional Radiology (M.S.J.), Indiana University, Indianapolis, Indiana; Department of Cardiac & Vascular Disease (F.A.), Lakeview Regional Heart Center, Covington, Louisiana; Department of Vascular & Interventional Radiology (J.L.S.), OSF Saint Francis Medical Center, Peoria, Illinois; Department of Interventional Cardiology & Vascular Medicine (G.M.A.), Riverside Methodist Hospital, Columbus, Ohio; Department of Interventional Cardiology &Vascular Medicine (M.J.S.), OhioHealth Heart and Vascular Physicians, Columbus, Ohio; Department of Interventional Radiology & Oncology (S.S.), University of Alabama, Birmingham, Alabama; Department of Vascular & Interventional Radiology (J.S.B.), Providence Sacred Heart Medical Center, Spokane, Washington; Department of Vascular Surgery (R.M.), UNC Rex Hospital, NC Heart and Vascular Research, Raleigh, North Carolina. Received June 13, 2019; final revision received August 12, 2019; accepted August 20,
2019. Address correspondence to M.D.Dak.; E-mail: [email protected]. edu M.D.Dak., M.A.C., and J.S.B. are paid consultants for BTG Vascular (Bothell, Washington). A.H.K. is a paid consultant for BTG Vascular and PQ Bypass (Sunnyvale, California). M.S.J. is a paid consultant for BTG Vascular, Boston Scientific (Marlborough, Massachusetts), and Cook Medical (Bloomington, Indiana). J.L.S. is a shareholder in Brightwater Medical (Murrieta, California). G.M.A. is a paid consultant for BTG Vascular, CR Bard (Murray Hill, New Jersey), and Cook Medical. M.J.S. is a paid consultant for Boston Scientific, Cook Medical, W. L. Gore & Associates (Flagstaff, Arizona), and Inari Medical (Irvine, California), and is a shareholder in Contego Medical (Raleigh, North Carolina). Appendices A–C and Tables E1–E3 can be found by accessing the online version of this article on www.jvir.org and clicking on the Supplemental Material tab. © SIR, 2019. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/). J Vasc Interv Radiol 2019; ▪:1–12 https://doi.org/10.1016/j.jvir.2019.08.036
2 ▪ Final Report: Sentry Bioconvertible IVC Filter
Dake et al ▪ JVIR
ABBREVIATIONS CEC ¼ clinical events committee, CI ¼ confidence interval, DVT ¼ deep vein thrombosis, FDA ¼ US Food and Drug Administration, IDE ¼ investigational device exemption, IVC ¼ inferior vena cava, PE ¼ pulmonary embolism, SAE ¼ serious adverse event, VTE ¼ venous thromboembolic
In appropriately chosen patients, systematic reviews and meta-analyses have shown that inferior vena cava (IVC) filters are effective at providing mechanical protection against pulmonary embolism (PE) (1–3). In response to adverse events with permanent IVC filters, retrievable IVC filters were developed to provide protection during recognized periods of transient PE risk, with the intent of removal in separate procedures (4). In practice, an unanticipated proportion of retrievable devices have remained in place after the need for temporary PE prophylaxis has passed. Retrieval rates rarely exceed 50%, and in some studies rates are as low as 8.5% (1,5,6). Retrievable devices that remain indwelling have been associated with time-dependent complications such as tilting, fracture, embolization, migration, and IVC perforation (5–8). When the removal of retrievable IVC filters is actually attempted, the procedures can be complex (particularly for devices that have been indwelling for longer periods), with a failure rate of approximately 10% (5). The Sentry IVC filter (BTG Vascular, Bothell, Washington) was developed as an alternative to retrievable filters for patients for whom anticoagulation is contraindicated and in whom the risk of PE is clearly transient. On the basis of evidence, in part from large-population studies of the incidence and timing of PE, suggesting that the period of highest PE risk occurs within the first 30 days after an index event for a large proportion of trauma and hospitalized surgery patients (9,10), the Sentry device was designed to bioconvert from a filtering to a nonfiltering configuration after 60 days, leaving a patent IVC lumen (11). The bioconversion of the Sentry device obviates the need for a separate secondary device-removal procedure. In the prospective multicenter SENTRY clinical trial, interim 1-year outcomes with the Sentry device included clinical success for 97.4% of evaluable patients (95% confidence interval [CI], 92.5%–99.1%); technical success in 129 of 130 attempted filter deployments; no new cases of symptomatic PE through 1 year of follow-up; no filter-related imaging complications; a 1.6% rate of filter-related symptomatic complications; and successful filter bioconversion for 96.4% (106/110) of patients (12). Reported here are final results from the SENTRY clinical trial through completed 24month clinical follow-up.
MATERIALS AND METHODS The SENTRY clinical trial design was previously reported (12). This prospective, multicenter, nonrandomized, singlearm trial was conducted at 23 sites in the United States
(n ¼ 20), Belgium (n ¼ 2), and Chile (n ¼ 1). Patients eligible for inclusion had documented deep vein thrombosis (DVT) or PE or a high risk of developing DVT or PE and had a contraindication to or failure of anticoagulation. The trial allowed enrollment of up to 40% of patients with a prophylactic indication (no current PE or DVT but high risk of PE). The indications for enrollment were consistent with American College of Radiology and Society of Interventional Radiology practice and quality improvement guidelines (Table E1 [available online on the article’s Supplemental Material page at www.jvir.org]) (13,14). The protocol was approved by the appropriate institutional review boards or ethics committees, and all study procedures were performed in accordance with the guidelines of good clinical practice and applicable regulations. BTG Vascular was the sole sponsor of the study, which was conducted under an investigational device exemption (IDE G110111), in compliance with applicable provisions of 21 CFR Parts 812, 50, and 54, and in accordance with the ethical principles of the Declaration of Helsinki. Data were collected on case report forms (with supervision by Vitruvian Clinical Research; San Ramon, California) and reviewed and adjudicated by an independent data monitoring committee and a clinical events committee (CEC), which included independent clinicians (an interventional radiologist, a vascular surgeon, and an interventional cardiologist) experienced in the use of IVC filters and the conduct of clinical trials. The Global Institute for Research (Richmond, Virginia) served as the imaging core laboratory, providing independent measurement and analysis in accordance with the predefined study protocols. QST Consultations (Allendale, Michigan) developed the statistical hypothesis for the study, and Advanced Research Associates (Santa Clara, California) defined the statistical analysis plan and methods and provided statistical analysis support. The study was registered before the start of patient enrollment (ClinicalTrials.gov ID NCT01975090).
Study Device, Procedures, Patient Characteristics, and Disposition The study device, the filter implantation, and the targeted bioconversion are summarized in Appendix A (available online on the article’s Supplemental Material page at www. jvir.org). A total of 63 investigators performed the implantations of Sentry IVC filters in 129 patients at 23 sites between September 2014 and February 2016. Study patient disposition through 24 months is detailed in Figure 1. Baseline characteristics of the patient population,
Volume ▪ ▪ Number ▪ ▪ Month ▪ 2019
3
Visual Synopsis
previously described (12), are summarized in Table 1. The patient indications for filter placement, the reasons for inability to use anticoagulation therapy, and key procedural details are summarized in Table 2, along with the primary 6-month composite endpoint results. The overall anticoagulation status of the patients, which changed as their conditions progressed during the course of the study,
is charted in Table E2 (available online on the article’s Supplemental Material page at www.jvir.org).
Study Follow-up and Imaging Evaluation After device implantation, according to a US Food and Drug Administration (FDA)-approved schedule, patients were evaluated at 1 month (30–44 days), 2 months (60–67 days),
4 ▪ Final Report: Sentry Bioconvertible IVC Filter
Dake et al ▪ JVIR
Figure 1. Disposition of the 129 enrolled patients through 24-month follow-up in the SENTRY clinical trial. PET ¼ positron emission tomography.
Volume ▪ ▪ Number ▪ ▪ Month ▪ 2019
5
Table 1. Baseline Demographics and Medical History Characteristic Mean age (years) Range
Table 2. Filter Indications, Placement, and Primary 6-Month Efficacy Endpoint
n ¼ 129 62.6 ± 13.52 21.0–88.0
Gender
Variables Indications for filter placement Current PE only
n ¼ 129 11 (8.5)
Male
73 (56.6)
Current DVT only
Female
56 (43.4)
Current DVT and PE
18 (14.0)
Prophylaxis*
42 (32.6)
2
BMI (kg/m ) Range Race White (non-Hispanic) Black
30.5 ± 8.38 17.3–78.1 107 (82.9) 11 (8.5)
Hispanic
9 (7.0)
Unknown
2 (1.6)
Medical history/risk factors Hypertension
76 (58.9)
Recent surgery (30 days)
33 (25.6)
Diabetes
28 (21.7)
Note–Values are presented as mean ± standard deviation. Values in parentheses are percentages. BMI ¼ body mass index.
6 months (150–210 days), 12 months (335–395 days), and 24 months (670–790 days). Each follow-up visit included clinical assessment for symptoms of PE and DVT, monitoring of adverse events, and assessment of venous thromboembolic (VTE) risk factors. Imaging for filter configuration and complications was performed by ultrasound of the lower extremities and computed tomography (CT) venography at 1 month, and by anterior-posterior and lateral x-ray (or by CT venography if thrombus was observed at 1 month) at 2 months. Imaging for filter bioconversion status and complications was performed by CT at 6 months, by anterior-posterior and lateral x-ray at 12 months, and by CT venography at 24 months.
Study Endpoints Study endpoints are specified and defined in Appendix B (available online on the article’s Supplemental Material page at www.jvir.org). Study endpoints were formulated in accordance with Society of Interventional Radiology reporting standards (15) and American College of Radiology guidelines (13), with reference to recent IDE studies (16–19) and the PRESERVE trial (NCT02381509), which was also designed in conjunction with the FDA, following FDA safety communications in 2010 and 2014 responding to the reported high rates of complications associated with prolonged indwelling of retrievable IVC filters and emphasizing the need for vigilant follow-up (20,21). Clinical assessment for potential PE-related symptoms was performed at each follow-up timepoint, and suspicion of PE was supported by appropriate imaging. For adjudication of the procedure- and device-relatedness of potential serious adverse events (SAEs), cases were presented to the CEC as
58 (45.0)
Primary factor for filter placement Surgery
77 (59.7)
Medical condition
28 (21.7)
Trauma Other
8 (6.2) 16 (12.4)
Reasons for inability to use anticoagulation therapy Inability to use anticoagulation therapy Risk of bleeding and/or injury from anticoagulation Contraindication
129 (100.0) 79 (61.2) 49 (38.0)
Access site for filter placement Right internal jugular vein
64 (49.6)
Right femoral vein
54 (41.9)
Left femoral vein
11 (8.5)
Pre-implant IVC diameter (mm) Mean ± SD Range Primary composite endpoint of clinical success at 6 months
19.3 ± 2.23 14.0–26.0 111/114 (97.4)
Technical success of filter deployment
129/130 (99.2)
Freedom from symptomatic PE during the 60-day protection period
129/129 (100)
Freedom from IVC filter-related complications at 6 months
112/114 (98.2)
Note–Values in parentheses are percentages. DVT ¼ deep vein thrombosis; IVC ¼ inferior vena cava; PE ¼ pulmonary embolism; SD ¼ standard deviation. *Prophylactic indication: no current PE or DVT but high risk of PE.
clinical summaries, including drug history and available autopsy reports, with imaging.
Statistical Analysis Endpoints were analyzed per patient on an intent-to-treat basis. Continuous variables were described as mean ± standard deviation, and categorical variables were described as counts and percentages. For the statistical analysis on the primary endpoint of clinical success at 6 months, the observed rate was tested against the acceptance criterion using the 2-sided 95% Wilson CI for the binomial proportion. If the lower confidence limit for the true proportion equaled or exceeded 80%, the endpoint was deemed to be successfully achieved. For the secondary endpoints and other clinical outcomes data, the number and percentage of observed patients for each endpoint were determined, and,
6 ▪ Final Report: Sentry Bioconvertible IVC Filter
Dake et al ▪ JVIR
Figure 2. Filtering configuration and bioconversion status through 24 months in the SENTRY clinical trial. Filtering configuration ¼ all 6 pairs of arms held in the central portion of the IVC lumen. Bioconverted ¼ 1 or more pairs of filter arms separated from the central portion of the IVC lumen.
where appropriate, the 2-sided 95% CI using the Wilson score interval around the proportion was calculated.
RESULTS Outcomes through 1 year of follow-up in the SENTRY clinical trial, which have been previously presented (12), are summarized in Appendix C (available online on the article’s Supplemental Material page at www.jvir.org).
Filter Bioconversion Figure 2 shows the bioconversion status for all patients with imaging assessment at 6, 12, and 24 months. Figure 3 shows representative CT imaging for a single patient through 24-month follow-up. Filter bioconversion was successful for 95.7% (110/115) of patients at 6 months, 96.4% (106/110) of patients at 12 months, and 96.5% (82/85) of patients at 24 months. Through the 24-month follow-up, no new DVT, PE, or IVC filter-related complications were reported in any of the 5 patients whose devices were not bioconverted at 6 months. Through 24 months of follow-up, there was no evidence of late-stage IVC obstruction or thrombosis after filter bioconversion or of thrombogenicity associated with retracted filter arms.
VTE Outcomes through 24 Months All patients available for follow-up were reported to be free from new symptomatic PE through 12 months (n ¼ 111) (Table 3). Between 12 and 24 months, new symptomatic nonfatal cases of PE were identified in 2 of 85 evaluable
patients (2.4%). In 1 of these patients, with new emboli found at 581 days in both the right and left pulmonary arteries, the study device was determined to have been bioconverted at 6 months, and the filter arms were fully retracted to the IVC wall. The new PE was successfully treated and resolved with anticoagulation and thrombolysis. The other case of new PE in the second year of follow-up occurred at day 624 after uncomplicated implantation of a Sentry device in a patient with DVT who had suffered severe gastrointestinal bleeding while on anticoagulation. In this patient, a small clot was identified in the filter at 1 month after implantation, for which no intervention was undertaken, and 6-month imaging showed that the filter had bioconverted, with the filter arms fully retracted to the IVC wall, and no incident of PE. At 624 days, while receiving low-molecular-weight heparin, the patient developed new DVT in the right femoral and popliteal veins. After the patient developed dyspnea, CT pulmonary angiography showed a saddle PE in both lungs extending into the segmental pulmonary arteries, with no cardiac abnormalities. The DVT and the PE were successfully treated with anticoagulation. In both of the cases of new symptomatic nonfatal PE during the second year of follow-up, the core laboratory and CEC review at 24 months adjudicated the events as not related to the study device or procedure. Through the 24-month follow-up, as adjudicated by the CEC, in 14 patients, there were 12 cases of new DVT (8 symptomatic and 4 asymptomatic) and 5 cases of worsening DVT (3 symptomatic and 2 asymptomatic) (Table 3). No DVTs were considered to be device related, and 1 (previously described) new symptomatic DVT at 8 days was
Volume ▪ ▪ Number ▪ ▪ Month ▪ 2019
7
Figure 3. Representative CT imaging for a single patient. Left: Coronal image acquired as part of a helical dataset 1 month after filter placement, showing the Sentry device in filtering configuration. Center: Coronal image at 24-month follow-up, showing the device in bioconverted configuration and endothelialized. Top right: Axial view of the bioconverted device at 6 months at the level of the filter tips. Bottom right: Axial view of the bioconverted device at 24 months at the level of the filter tips.
Table 3. IVC Filter-Related Complications and VTE Outcomes through 24 Months Filter-related imaging complications
1 mo n ¼ 122
2 mo n ¼ 119
6 mo n ¼ 114
12 mo n ¼ 111
24 mo n ¼ 85
Tilting
0 (0)
0 (0)
0 (0)
0 (0)
0 (0)
Migration
0 (0)
0 (0)
0 (0)
0 (0)
0 (0)
Perforation
0 (0)
0 (0)
0 (0)
0 (0)
0 (0)
Embolization
0 (0)
0 (0)
0 (0)
0 (0)
0 (0)
Fracture
0 (0)
0 (0)
0 (0)
0 (0)
0 (0)
Any imaging complication
0 (0)
0 (0)
0 (0)
0 (0)
0 (0)
Filter-related symptomatic complications
0–1 mo n ¼ 129
Filter-related death Symptomatic caval thrombosis
0 (0) 2 (1.6)*
0 (0) 0 (0)
0 (0) 0 (0)
0 (0) 0 (0)
0 (0) 0 (0)
Other symptomatic complications requiring invasive intervention
0 (0)
0 (0)
0 (0)
0 (0)
0 (0)
CEC Adjudicated VTE Outcomes
0–1 mo n ¼ 129
New symptomatic DVT
3 (2.3)
New asymptomatic DVT
4 (3.1)
Symptomatic worsening DVT Asymptomatic worsening DVT New symptomatic PE
1–2 mo n ¼ 127
1–2 mo n ¼ 127
2–6 mo n ¼ 126
6–12 mo n ¼ 111
12–24 mo n ¼ 85
2–6 mo n ¼ 126
6–12 mo n ¼ 111
12–24 mo n ¼ 85
0 (0)
1 (0.8)
1 (0.9)
3 (3.5)
0 (0)
0 (0)
0 (0)
0 (0)
1 (0.8) 1 (0.8)
0 (0) 1 (0.8)
0 (0) 0 (0)
1 (0.9) 0 (0)
1 (1.2) 0 (0)
0 (0)†
0 (0)
0 (0)
0 (0)
2 (2.4)†
Note–Values in parentheses are percentages. DVT ¼ deep vein thrombosis; IVC ¼ inferior vena cava; PE ¼ pulmonary embolism; VTE ¼ venous thromboembolism. *Two cases of symptomatic caval thrombosis developed and were successfully treated during the first month of follow-up. † Two cases of new symptomatic PE developed at 581 and 624 days after implantation and after IVC filter bioconversion. The PE events between 12 and 24 months were adjudicated by the clinical events committee as not device related and not procedure related.
8 ▪ Final Report: Sentry Bioconvertible IVC Filter
adjudicated as being related to the procedure (12). Of the cases of new symptomatic DVT, 3 were recorded during the second year of follow-up—at days 395, 411, and 624 (discussed above)—in patients who had been symptomatic at baseline with known DVT or PE. In the case recorded at day 395, the patient (whose device remained non-converted through 24month follow-up) was not receiving anticoagulation, and compression stockings were prescribed. The case recorded at day 411 involved bilateral DVT in the popliteal and calf veins in a patient who had also experienced a new symptomatic DVT at 224 days and a worsening symptomatic DVT at 351 days; the patient had received an additional IVC filter at 147 days above the bioconverted Sentry device. The bilateral DVTs resolved with physical therapy and anticoagulation. Of the cases with symptomatic worsening DVT, 1 occurred during the second year, at day 580 in the same patient who experienced a new symptomatic nonfatal PE at 581 days. The DVT, in the left femoral and popliteal veins, was treated with anticoagulation and physical therapy. Through 24 months, 3 (2.3%) patients developed new transient VTE risks requiring the placement of an additional IVC filter chosen at the discretion of the investigator (the study device was not available for secondary implantation). Each of these cases occurred beyond the 60-day targeted study-filter protection period, and none was placed to extend the initial risk period. Two of the cases, previously described (12), occurred during the first 6 months of follow-up, with the new risk due to severe epistaxis in 1 case and to hemorrhagic brain metastasis in the other. In each of these 2 cases, the additional filter was placed above the bioconverted study device. The third case occurred at 613 days in a patient originally enrolled due to a history of DVT after total knee replacement who returned to undergo gastric bypass revision. In this case, the additional filter (Cook Celect) was placed inside the bioconverted study device. In all 3 cases, CEC adjudication confirmed that the additional device placement was required due to new VTE risk and not to the study device or procedure or to extend the index protection period.
Filter-Related Complications and Safety Outcomes through 24 Months Through 24-month follow-up, there were no further imagingidentified or symptomatic IVC filter-related complications besides the 2 cases of symptomatic caval thrombosis that were reported and successfully treated during the first month after implantation (Table 3) (12). There were no cases of surgical removal of the filter and/or vascular repair of the IVC, and there were no deaths adjudicated as filter related. On all CT imaging performed at any timepoint, no instances of filter perforation of the IVC wall were observed. In the entire safety analysis population of 129 patients, 1 or more adverse events were experienced by a total of 104 (80.6%) during the 24 months of follow-up, and 1 or more SAEs were experienced by 63 (48.8%) patients (Table E3 [available online on the article’s Supplemental Material page at www.jvir.org]). The most common SAEs were infections (17.8%). None of the SAEs was confirmed by the
Dake et al ▪ JVIR
CEC to be related to the study device. None of the 12 deaths in patients eligible for follow-up through 24 months (due to cancer [n ¼ 3], cardiopulmonary arrest [n ¼ 2], respiratory failure [n ¼ 3], multi-organ failure, hypertensive cardiovascular disease, liver failure, or myasthenia gravis) was related to the Sentry filter or to the index procedure, as reported by the sites and confirmed by CEC adjudication.
DISCUSSION For the patients in the SENTRY trial—all of whom had documented DVT or PE or were at high risk of developing DVT or PE and had a contraindication to or failure of anticoagulation—freedom from new symptomatic PE was 100% through the first year of follow-up, an outcome comparing favorably with PE rates in recent trials of retrievable IVC filters (16–19,22). Between 12 and 24 months, new symptomatic nonfatal PE was identified in 2 of 85 evaluable patients (2.4%), in each of whom the 6-month imaging had shown the study device to be bioconverted with the filter arms fully retracted to the IVC wall. In each of these cases, the new PE was successfully resolved. Both of these events occurred well past the index period of early transient risk, and neither of the patients had an indication for a permanent IVC filter. Both cases were adjudicated by the CEC as not related to the study device or procedure. The situation in these 2 patients was similar to one in which a retrievable filter would have previously and appropriately been removed. The timing of these 2 new cases of PE in patients, in whom the Sentry device had already bioconverted at 6 months, would strongly indicate that they were not due to release of entrapped thrombus. Such a low rate of late PE would suggest that physicians were capable of identifying patients who had a defined period of risk from PE and who did not require placement of a permanent IVC filter. Through the 24-month follow-up, there were a total of 17 cases of CEC-confirmed new or worsening DVT in 14 patients (10.9%), which also compared favorably with recent findings for retrievable filters (16–19,22). Only the 1 new symptomatic access-site DVT that occurred at 8 days was considered to be procedure related, whereas none was confirmed as being device related. The bioconversion of the Sentry IVC filter—targeted to occur after 60 days after implantation, after the period of transient PE risk—and subsequent endothelialization obviate the need for, and avoid any potential for complications that would be associated with, a second separate device-removal procedure. Also potentially avoided were the risks of time-dependent complications such as have been reported in association with retrievable IVC filters that remain indwelling (5–7). The confirmed execution of the mandated 1-month, 2-month, 6month, 12-month, and 24-month CT, CT venography, and xray examinations documented the success of device bioconversion and the absence of device-related imaging-identified complications or issues with the filter-frame integrity through 24 months—no cases of tilting, migration, perforation,
Volume ▪ ▪ Number ▪ ▪ Month ▪ 2019
embolization, or fracture. There were no filter-related deaths through 24 months of follow-up. Through 24 months of followup, there was no evidence of late-stage IVC obstruction or thrombosis after filter bioconversion or of thrombogenicity associated with retracted filter arms, which was a finding that confirmed the results of preclinical studies with the device in an experimental ovine model (23). There was no PE experienced by any patient who had thrombus in the filter in the protocolmandated 1-month CT venography (see Appendix C [available online on the article's Supplemental Material page at www.jvir. org]). The Sentry IVC filter was designed not to tilt, and any trapped thrombus is thus retained in the center of the vessel where the body’s own lysis capabilities may be maximized. It is unlikely that the trapped clot is “free floating,” as it tends to become adherent to the filter elements. Through 2 years, 3 patients required the placement of an additional IVC filter beyond the 60-day protection period and after the index device had bioconverted. In each of these cases, the CEC confirmed that the new filter was required due to a new distinct VTE risk and not to the study device or to extend the index protection period. The secondary filter placements in response to new VTE risk did not take place until well after the targeted 60-day protection period, suggesting (as does the low rate of late PE) that all of the Sentry trial implanting physicians were successful in identifying patients whom they felt would benefit from temporary protection due to identification of transient risk based on individual case presentations. As is the case with other IVC filters, data are lacking regarding the safety and efficacy of the Sentry bioconvertible filter in hypercoagulable patients or those requiring multiple surgeries over short intervals. However, along with retrievable filters, which are still often used in these cases, the Sentry IVC filter can be considered in such temporary filtration scenarios. Limitations of the present study include the nonrandomized single-arm design and the inherent potential for bias in a manufacturer-funded regulatory device trial. It is possible that instances of asymptomatic PE may have gone undetected (consistent with other IVC filter studies), as imaging was performed only in patients with suggestive clinical symptoms. Likewise, although 6 of the 17 detected cases of new and worsening DVT were asymptomatic, after the 1-month ultrasonography there was no mandated lower extremity imaging, and DVT status was assessed based on symptoms and any site-performed imaging that was part of follow-up of high-risk patients. Although the 32.6% proportion of patients with prophylactic (no current PE or DVT but high risk of PE) indications for study device implantation was consistent with the enrollment in previous IDE filter studies and established guidelines, the ability of the filter to capture already present emboli of lower extremity origin could not be observed in those patients. This study offers no evidence to suggest that a bioconverted Sentry device would pose a technical challenge to placement of an additional filter in a situation of new VTE risk, but the concern with such new filter placement could be greater in the rare case of a Sentry device remaining nonconverted.
9
Although short-term outcomes are promising, data are needed regarding any effect of the long-term presence of the bioconverted and endothelialized Sentry device. The SENTRY clinical trial has demonstrated that for patients requiring temporary protection against PE and unable to use anticoagulation, the Sentry IVC filter provides high rates of technical and clinical success with minimal complications, within the efficacy and safety thresholds suggested by the Society of Interventional Radiology, and outcomes through 2 years that compare favorably with those reported for retrievable IVC filters. The results of this trial confirm that the Sentry bioconvertible device can provide an effective alternative to existing retrievable IVC filters that often remain indwelling long after the period of transient PE risk has passed and that are associated with high complication rates.
ACKNOWLEDGMENTS ClinicalTrials.gov ID NCT01975090. The study was funded by BTG Vascular (Bothell, Washington). The authors thank Diane Gargus for clinical trial management. Writing support services were provided by Galen Press, Inc (Austerlitz, New York) and paid for by the study sponsor. The following investigators and institutions participated in the SENTRY Clinical Trial (Michael D. Dake, MD, principal investigator) in addition to the authors: Robert Feezor, MD (University of Florida, Gainesville, FL); Sanjeeva Kalva, MD (UT Southwestern Medical Center, Dallas, TX); Darren Kies, MD (Emory University, Atlanta, GA); Marc Bosiers, MD (AZ Sint-Blasius, Dendermonde, Belgium); Werner Ziegler, MD (Memorial Hospital of Colorado Springs, Colorado Springs, CO); Mark Farber, MD (University of North Carolina, Chapel Hill, NC); David Paolini, MD (Jobst Vascular Institute, Toledo, OH); Robert Spillane, MD (Hartford Hospital, Hartford, CT); Steven Jones, MD (Cardiovascular Associates of the Southeast, LLC, Birmingham, AL); and Patrick Peeters, MD (Imelda Hospital, Bonheiden, Belgium).
REFERENCES 1. Angel LF, Tapson V, Galgon RE, Restrepo MI, Kaufman J. Systematic review of the use of retrievable inferior vena cava filters. J Vasc Interv Radiol 2011; 22:1522–1530 e1523. 2. Bikdeli B, Chatterjee S, Desai NR, et al. Inferior vena cava filters to prevent pulmonary embolism: systematic review and meta-analysis. J Am Coll Cardiol 2017; 70:1587–1597. 3. Haut ER, Garcia LJ, Shihab HM, et al. The effectiveness of prophylactic inferior vena cava filters in trauma patients: a systematic review and meta-analysis. JAMA Surg 2014; 149:194–202. 4. Kinney TB. Update on inferior vena cava filters. J Vasc Interv Radiol 2003; 14:425–440. 5. Jia Z, Fuller TA, McKinney JM, et al. Utility of retrievable inferior vena cava filters: a systematic literature review and analysis of the reasons for nonretrieval of filters with temporary indications. Cardiovasc Intervent Radiol 2018; 41:675–682. 6. Sarosiek S, Crowther M, Sloan JM. Indications, complications, and management of inferior vena cava filters: the experience in 952 patients at an academic hospital with a level I trauma center. JAMA Intern Med 2013; 173:513–517.
10 ▪ Final Report: Sentry Bioconvertible IVC Filter
7. Deso SE, Idakoji IA, Kuo WT. Evidence-based evaluation of inferior vena cava filter complications based on filter type. Semin Intervent Radiol 2016; 33:93–100. 8. US Food and Drug Administration. Removing retrievable inferior vena cava filters: FDA safety communication. Issue date: May 6, 2014. Available at: https://wayback.archive-it.org/7993/20161022044053/http:// www.fda.gov/MedicalDevices/Safety/AlertsandNotices/ucm396377.htm. Accessed March 1, 2019. 9. Coleman JJ, Zarzaur BL, Katona CW, et al. Factors associated with pulmonary embolism within 72 hours of admission after trauma: a multicenter study. J Am Coll Surg 2015; 220:731–736. 10. Hope WW, Demeter BL, Newcomb WL, et al. Postoperative pulmonary embolism: timing, diagnosis, treatment, and outcomes. Am J Surg 2007; 194:814–818; discussion 818–819. 11. Dake MD, Ansel GM, Johnson MS, Mendes R, Smouse HB. The clinical rationale for the Sentry Bioconvertible Inferior Vena Cava Filter for the prevention of pulmonary embolism. Int J Vasc Med 2019; 2019:16. 12. Dake MD, Murphy TP, Kr€amer AH, et al. One-year analysis of the prospective multicenter SENTRY clinical trial: safety and effectiveness of the Novate Sentry Bioconvertible Inferior Vena Cava Filter. J Vasc Interv Radiol 2018; 10:1350–1361. 13. ACR-SIR-SPR Practice Parameter for the Performance of Inferior Vena Cava (IVC) Filter Placement for the Prevention of Pulmonary Embolism. Revised; 2016. Available at: https://www.acr.org/-/media/ACR/Files/PracticeParameters/ivc-fliterplacement.pdf?la¼en. Accessed March 1. 2019. 14. Caplin DM, Nikolic B, Kalva SP, et al. Quality improvement guidelines for the performance of inferior vena cava filter placement for the prevention of pulmonary embolism. J Vasc Interv Radiol 2011; 22: 1499–1506. 15. Millward SF, Grassi CJ, Kinney TB, et al. Reporting standards for inferior vena caval filter placement and patient follow-up: supplement for
Dake et al ▪ JVIR
16.
17.
18.
19.
20.
21.
22.
23.
temporary and retrievable/optional filters. J Vasc Interv Radiol 2009; 20: S374–S376. Stavropoulos SW, Sing RF, Elmasri F, et al. The DENALI Trial: an interim analysis of a prospective, multicenter study of the Denali retrievable inferior vena cava filter. J Vasc Interv Radiol 2014; 25:1497–1505, 1505,e1. Stavropoulos SW, Chen JX, Sing RF, et al. Analysis of the final DENALI trial data: a prospective, multicenter study of the Denali inferior vena cava filter. J Vasc Interv Radiol 2016; 27:1531–1538.e1. Smouse HB, Mendes R, Bosiers M, et al. The RETRIEVE trial: safety and effectiveness of the retrievable crux vena cava filter. J Vasc Interv Radiol 2013; 24:609–621. Johnson MS, Nemcek AA Jr, Benenati JF, et al. The safety and effectiveness of the retrievable Option inferior vena cava filter: a United States prospective multicenter clinical study. J Vasc Interv Radiol 2010; 21: 1173–1184. US Food and Drug Administration. Removing retrievable inferior vena cava filters: initial communation. Issue date: August 9, 2010. Available at: https://wayback.archive-it.org/7993/20161022180008/http:/www.fda.gov/ MedicalDevices/Safety/AlertsandNotices/ucm221676.htm. Accessed March 1, 2019. US Food and Drug Administration. Removing retrievable inferior vena cava filters: FDA Safety Communication. Updated May 6, 2014, Available at: https://www.fda.gov/MedicalDevices/Safety/AlertsandNotices/ucm39 6377.htm. Accessed March 1, 2019. Lyon SM, Riojas GE, Uberoi R, et al. Short- and long-term retrievability of the Celect vena cava filter: results from a multi-institutional registry. J Vasc Interv Radiol 2009; 20:1441–1448. Gaines PA, Kolodgie FD, Crowley G, et al. Sentry Bioconvertible Inferior Vena Cava Filter: study of stages of incorporation in an experimental ovine model. Int J Vasc Med 2018; 2018:10.
Volume ▪ ▪ Number ▪ ▪ Month ▪ 2019
APPENDIX A. DEVICE DESIGN, IMPLANTATION, AND BIOCONVERSION The Sentry bioconvertible IVC filter (BTG Vascular, Bothell, Washington) is designed to provide temporary protection against PE during transient high-risk periods and then to bioconvert after 60 days after implantation, thus obviating the need for device retrieval in a secondary procedure. The Sentry IVC filter is made from a single piece of laser-cut nitinol, which is formed into a cylindrical frame with an integral filter cone consisting of 6 pairs of arms held together in the center of the IVC by means of a bioabsorbable filament composed of poly-p-dioxanone, a biodegradable synthetic polymer. The Sentry is designed to provide filtering protection for a minimum duration of 60 days and then to bioconvert to a nonfiltering configuration. Upon deployment, the cylindrical frame expands to appose the IVC wall, inciting its incorporation into the caval wall by means of neointimal healing. During bioconversion, the bioabsorbable filament hydrolyzes, releasing the filtering arms from the filtering cone. The filtering arms then retract to the IVC wall into a nonfiltering configuration, where they also are endothelialized, and IVC lumen patency is restored. The Sentry is indicated for use in IVCs with diameters between 16 and 28 mm and has a maximum deployed length of 57.7 mm. Contemporary data from an analysis of 17 largepopulation studies of the incidence and timing of PE in post-injury trauma patients and major surgery patients support the Sentry design premise that the period of highest risk for PE in such patients with temporary contraindications to anticoagulants occurs early, with most PEs diagnosed within 30 days of an index event (hospitalization, traumatic injury, or surgery) (11).
APPENDIX B. STUDY ENDPOINTS The predefined primary endpoint of the Sentry clinical trial was clinical success at 6 months, which was a composite of technical success (filter deployment as intended without acute events), freedom from symptomatic PE through 60 days, and 6-month freedom from filter-related complications. Secondary efficacy endpoints included the technical success rate at day 0; filter status at months 1 and 2 (the percentage of patients with devices in filtering configuration, based on all 6 pairs of arms being held together in the central portion of the IVC lumen, and the percentage in nonfiltering configuration with arms separated from the central portion of the lumen); bioconversion status at months 6, 12, and 24 (the percentage bioconverted and the percentage not converted); and new symptomatic PE through 6, 12, and 24 months. Clinical assessment for potential PE-related symptoms was performed at each followup timepoint, and suspicion of PE was supported by appropriate imaging.
10.e1
Secondary safety endpoints included freedom from filterrelated complications on day 0 and at months 1, 2, 6, 12, and 24; invasive filter interventions such as thrombolysis, thrombectomy, surgical removal of the filter, placement of a second filter, and/or vascular repair of the IVC; and placement of an additional IVC filter (categorized as filter related or for extension of protection from PE beyond 60 days). For adjudication of the procedure- and device-relatedness of potential SAEs, cases were presented to the CEC as clinical summaries, including drug history and available autopsy reports, with imaging. Other clinical outcomes included DVT of the lower extremities; the assessment of VTE risk factors; and anticoagulation status on day 0 and at months 1, 2, 6, 12, and 24.
APPENDIX C. SENTRY CLINICAL TRIAL OUTCOMES THROUGH THE FIRST YEAR OF FOLLOW-UP A Sentry device was successfully implanted in all 129 patients enrolled in the trial; and after deployment in all patients, the protocol-mandated venography verified that all implanted filters were in the intended location and in the filtering configuration with the IVC patent. At the 1-month follow-up, 100% (119/119) of the devices were in filtering configuration; at the 2-month follow-up, 95.3% (101/106) of the devices remained in filtering configuration. There was no evidence that any device had bioconverted before 60 days after implantation. Subsequent imaging confirmed that successful device bioconversion had occurred for 95.7% of patients by 6 months, for 96.4% of patients by 12 months, and for 96.5% of patients by 24 months. All 3 of the component criteria were met by 111 (97.4%) of 114 patients evaluable for the composite primary clinical success endpoint (95% CI, 92.5%–99.1%), exceeding the predefined acceptance criterion of 80%. Technical success of deployment was achieved in 99.2% (129/130) of deployment attempts (95% CI, 95.8%–99.9%). The rate of freedom from new symptomatic PE through 60 days was 100% (n ¼ 129; 95% CI, 97.1%–100.0%), and the rate of freedom from IVC filter-related complications through 6 months was 98.2% (112/114; 95% CI, 93.8%–99.5%), based on findings of symptomatic caval thrombosis in 2 patients during the first month of follow-up (adjudicated by the CEC as SAEs having an unknown relationship to the device and procedure). After successful treatment with thrombectomy and thrombolysis in both cases of symptomatic caval thrombosis, the 2-month follow-up by x-ray and/or CT venography confirmed that the filter was in correct filtering configuration, and there was no recurrence of the caval thrombosis in either patient. The protocol-mandated 1-month CT venography revealed the presence of thrombus in the filters of 18 (15.8%) of 114 patients with core laboratory review. The thrombus was symptomatic only in the 2 successfully treated cases of caval
10.e2 ▪ Final Report: Sentry Bioconvertible IVC Filter
thrombosis. No patient with thrombus in the filter at 1 month went on to experience PE during the next 30 days. There were no reports of symptomatic IVC stenosis caused by the device. Filter bioconversion occurred for 95.7% of patients by 6 months, for 96.4% of patients by 12 months, and for 96.5% of patients by 24 months. Through 24 months of follow-up, there was no evidence of late-stage IVC obstruction or
Table E1. SIR Standards of Practice Committee Classification of Indications for IVC Filter Placement* Therapeutic indications (documented thromboembolic disease) Evidence of PE or IVC/iliac/femoropopliteal DVT and 1 or more of the following: Absolute or relative contraindication to anticoagulation Complications of anticoagulation Failure of anticoagulation Recurrent PE despite adequate therapy Inability to achieve/maintain adequate anticoagulation Propagation/progression of DVT during therapeutic anticoagulation Massive PE with residual DVT in a patient at risk for further PE Free-floating ileofemoral or IVC thrombus Severe cardiopulmonary disease and DVT (eg, corpulmonale with pulmonary hypertension) Prophylactic indications (no current thromboembolic disease) Temporary risk of PE in 1 of the following settings: Severe trauma without documented PE or DVT Closed head injury Spinal cord injury Multiple long-bone or pelvic fractures High-risk situations (eg, patient immobilized or in ICU) Note–Values in parentheses are percentages. DVT ¼ deep vein thrombosis; ICU ¼ intensive care unit; IVC ¼ inferior vena cava; PE ¼ pulmonary embolism. *Caplin DM, Nikolic B, Kalva SP, et al. Quality improvement guidelines for the performance of inferior vena cava filter placement for the prevention of pulmonary embolism. J Vasc Interv Radiol 2011; 22:1499–1506.
Dake et al ▪ JVIR
thrombosis after filter bioconversion or of thrombogenicity associated with retracted filter arms. In the unusual event of non-bioconversion of the Sentry filter, a subsequent interventional procedure to convert the Sentry into a nonfiltering configuration is not recommended. The Sentry device was designed not to tilt, with the intention that any trapped thrombus would be central within the vessel where the body’s own lysis capabilities can be maximized.
Volume ▪ ▪ Number ▪ ▪ Month ▪ 2019
10.e3
Table E2. Patient Anticoagulation Status through 24 Months Days –7 to –1 N in interval
Device Deployment Day 0
129
Days 1–7
Month 1 Month 2 Months 2–6 Months 6–12 (Days 8–44) (Days 45–67) (Days 68–210) (Days 211–395)
Months 12–24 (Days 396–760)
129
129
128
127
126
117
85
95 (74%) n (%) with no anticoagulation use for all or part of the interval
72 (56%)
80 (62%)
65 (51%)
42 (33%)
72 (57%)
69 (59%)
55 (65%)
n (%) with 34 (26%) continuous anticoagulation for the interval
57 (44%)
49 (38%)
63 (49%)
85 (66%)
54 (42%)
48 (41%)
30 (35%)
Note–Values in parentheses are percentages. Anticoagulants included heparin, factor Xa inhibitors, direct thrombin inhibitor, factor II inhibitor, and Coumadin derivatives. A patient could use more than 1 anticoagulant medication within the same time interval. At study baseline, all 129 enrolled patients were reported by the sites to have permanent, temporary, or predicted contraindications to anticoagulation. The overall anticoagulation status of the patients changed as their conditions progressed during the course of the study. The proportion of patients who had interruption of anticoagulant use was greatest during the first 7 days after implantation, which was likely to be the period of highest risk for many. More patients were treated continuously with anticoagulation between day 8 and day 67, as they no longer had contraindications. Between days 68 and 210, the proportion of patients on continuous anticoagulation was lower, as was expected due to the passing of the period of transient risk of pulmonary embolism.
Table E3. SAEs by System Organ Class through 24-Month Follow-up Category
Patients with ≥1 SAE
All patients, all SAEs
63 (48.8)
Blood and lymphatic system disorders Cardiac disorders
5 (3.9) 12 (9.3)
Congenital, familial, and genetic disorders
2 (1.6)
Gastrointestinal disorders
8 (6.2)
General disorders and administration site conditions
6 (4.7)
Hepatobiliary disorders Infections and infestations Injury, poisoning, and procedural complications
1 (0.8) 23 (17.8) 9 (7.0)
Metabolism and nutrition disorders
4 (3.1)
Musculoskeletal and connective tissue disorders
6 (4.7)
Neoplasms benign, malignant, and unspecified (including cysts and polyps)
8 (6.2)
Nervous system disorders
6 (4.7)
Psychiatric disorders
1 (0.8)
Renal and urinary disorders
7 (5.4)
Respiratory, thoracic, and mediastinal disorders Skin and subcutaneous tissue disorders
12 (9.3) 1 (0.8)
Surgical and medical procedures
4 (3.1)
Vascular disorders
7 (5.4)
Note–Values in parentheses are percentages. SAE ¼ serious adverse event.