Early Clinical Experience with a Novel, Self-Centering, Bi-Directional Vena Cava Filter

Introduction

Pulmonary emboli (PE) are estimated to be the third leading cause of cardiovascular death in the United States. Over 600,000 cases are reported in the US annually, causing an estimated 120,000 to 150,000 deaths per year. Emboli most commonly arise from pre-existing deep venous thromboses (DVT). While anticoagulation remains the gold standard of care for both DVT and PE, many patients are contraindicated for anticoagulation¹ and up to one-third of PE occurs in patients already on anticoagulant therapy. Caval interruption by placement of an inferior vena cava filter (IVCF) is the common treatment for patients contraindicated for anticoagulation.^2–6 Currently, more than 140,000 percutaneous IVCF are implanted each year.⁸ While most filters were originally designed for permanent placement, concerns over long-term durability and clinical consequences have driven the increasing use of optional filters, which can be percutaneously removed or left in permanently. The latest generation of caval filters is low profile and designed to be retrievable,^7–14 yet consequences including recurrent PE, IVC thrombosis, filter migration, perforation, inability to retrieve, and device fracture have been reported, creating a need for a more optimal retrievable IVCF design. The purpose of our study was to evaluate the performance of a novel retrievable IVCF. The Crux IVCF (Crux Biomedical, Inc., Menlo Park, California) is a unique low-profile (6-Fr), self-centering, and retrievable (via both jugular and femoral access) IVCF. We have previously reported safe use and filter retrievability at 5 weeks in an ovine model.¹⁵ Herein, we present the use of this filter in the initial 10 patients to evaluate filter deployment accuracy, fluoroscopic visibility, resistance to migration, freedom from caval thrombosis/recurrent PE, and retrievability.

Methods

Device description. The Crux IVCF is constructed of two nitinol spiral elements crimped at the ends to form a symmetric double-looped helical structure. On one loop, a filter web is attached to the spiral element designed to capture clots. Attached to the opposite loop, three fixation anchors are crimped to the wireform, two of which are located at the loop midpoints and the third at the tail end. The filter is designed for either temporary or permanent use and each end has a retrieval element designed to facilitate capture by a commercial snare via both jugular or femoral access. A tantalum radiopaque marker allows for visualization under fluoroscopy. The filter is preloaded in a cartridge and delivered by a pushrod through a custom 6-Fr introducer sheath. During deployment, the sheath is retracted while the pushrod is held stable. As the sheath retracts, the filter self-expands, allowing the fixation anchors to engage the vessel wall. The filter is designed to self-center and oppose the IVC wall along the entire surface and sized to cover an IVC diameter up to 28 mm with two filter sizes available (small to treat cava 17–22 mm and large to treat 22–28 mm). Retrieval is accomplished with a double-sheath technique (6 and 10 Fr Cook Flexor Raabe Sheath [Bloomington, Indiana] that is 90 and 80 cm in length, respectively) via either femoral or jugular access. The smaller sheath is advanced through the larger sheath and advanced to the level of the device. The filter retrieval tail is captured by the smaller sheath with the snare, and the larger sheath is then advanced over the remainder of the device.

Study Design

This was a single-center pilot study conducted at the French Hospital in Asuncion, Paraguay to evaluate the performance of the retrievable Crux IVCF. The study objectives were to evaluate filter deployment accuracy, fluoroscopic visability, resistance to migration, freedom from caval thrombosis/recurrent PE, and retrievability. Ten filters were placed in the infrarenal IVC of consented patients meeting the inclusion criteria (5 in December 2006 and 5 in August 2007). Indications for filter placement included a medical condition with a high risk of venous thromboembolism, contraindication to anticoagulation, and inability to achieve/maintain therapeutic anticoagulation. Abdominal radiographs were obtained at 24 hours and 30 days post-implant to evaluate device position and integrity. Four patients were designated for filter retrieval at varying intervals: one device was retrieved at 39 days, two at 40 days, and one at 63 days. In those patients selected for device retrieval, intraoperative fluoroscopy was used to image the IVC post-explant. Retrieved filters were examined ex vivo for structural integrity and showed no evidence of fracture or loss of filter integrity.

Experimental Procedure

All procedures were performed in routine sterile fashion under local anesthesia. Vascular access was achieved through the femoral vein (n = 6) or jugular vein (n = 4). A cavogram was performed routinely with a 5-Fr pigtail catheter (Boston Scientific, Natick, Massachusetts) to determine caval size and corollary vessel location. Markers spaced 2 cm apart on the Crux delivery sheath were used for caval diameter measurements and the filter size selected accordingly. Crux filters were then implanted through the custom 6-Fr delivery sheath and accompanying pushrod. All 10 filters were deployed in the infrarenal position. Final positioning and evaluation of vessel integrity was evaluated post-implant with fluoroscopic imaging. Abdominal radiographs were used to assess filter placement in all patients 24 hours and 30 days after implantation. In 4 patients selected at the physicians discretion for filter retrieval, filters were retrieved from the subjects at varying time intervals; one device was retrieved at 39 days, two at 40 days, and one at 63 days. Filters were retrieved from femoral vein access using a two-sheath technique. A 10-Fr Cook Flexor Raabe 80-cm sheath was advanced to the level of the filter. Then a 6-Fr, 90-cm Cook Flexor Raabe sheath was advanced to snare the device using an EnSnare. The device was then pulled under tension to capture it within the smaller sheath. The larger sheath was then advanced over the device allowing complete device capture. Post-retrieval fluoroscopy was used to image vessel integrity.

Results Implantation. All 10 devices were successfully implanted in their intended location within the infrarenal IVC via jugular or femoral venous access. Mean deployment time was 11 minutes once the vessel was accessed. There were no complications related to filter placement. All devices were positioned accurately and were normally oriented according to post-placement fluoroscopy and 24-hour post-placement kidney, ureter, bladder (KUB). The vena cava filters (VCF) remained fixed in orientation and location over the 30-day period as by KUB.

Retrieval. Prior to retrieval in the four subjects selected for filter removal, intra-operative fluoroscopy demonstrated patent venous system with preserved integrity of the vena cava and implanted filter. There was no evidence of vessel damage, narrowing, thrombosis, filter fracture, or migration. All four filters were subsequently removed successfully, with minimal force required for capture. All devices were recovered without complications. All filters were retrieved within 5 minutes of sheath placement. Completion venograms revealed no evidence of vessel damage.

Follow up. All 10 patients have been followed since filter placement. There have been no complications related to implanted filters reported with implantation times up to 16 months at the time of this report. Specifically, there have been no reports of filter fracture, migration, recurrent PE, caval thrombosis, or penetration. In patients post-retrieval, no complications have been reported since filter removal.

Discussion

The Crux IVCF is a novel low-profile (6-Fr) retrievable filter with the option to function as a permanent filter, if warranted. The device is available in two sizes based on internal diameters small (17–22 mm) and large (22–28 mm); and can be sized to fit the majority of IVC. The advantage of having two devices is to allow for appropriate placement of the correct size device rather than placing too large or small a device. Due to its spiral shape, appropriate sizing is necessary, especially when oversizing as the device can elongate. We have previously reported the safe and efficacious use of the Crux IVCF in an ovine model.¹⁵ Twelve filters were implanted in the infrarenal IVC of 12 sheep with no adverse events occurring during deployment or the subsequent implantation period of 12 weeks. There were no instances of filter fracture, migration, IVC thrombosis, or PE. Filters were retrieved at 5-weeks post-implantation without complication or evidence of vessel damage. This study demonstrated both ease of deployment via jugular and femoral access and retrievablity in 10 patients. All filters were deployed in their intended location in the infrarenal IVC without complications. Four patients were selected for filter retrieval. All filters were removed within 5 minutes of sheath placement at up to 9 weeks post-implantation. Vessel integrity was maintained after filter recovery, as demonstrated by post-retrieval venograms. In the remaining 6 patients, filters were intended to function as permanent filters. There have been no reports of caval thrombosis, IVC wall penetration, filter fracture, or migration during the implantation period out to 16 months post-deployment.

The Crux IVCF has several advantages over currently available IVCF. Currently available VCF share a similar conical design. This design has several drawbacks. First, the conical shape of the filter provides a natural tendency for the filter to tilt and the hooks to fail to engage or pull away from the caval wall, potentially contributing to filter migration. Additionally, retrievability of existing conical filters may be limited by filter struts perforating the cava wall and the tilted retrieval hook apex becoming incorporated into the cava wall, making snare engagement difficult or impossible. In contrast to the conical design of existing filters, the Crux IVCF has a symmetrical double-looped helical structure. This novel filter is designed to be self-centering in the cava and retrievable via both jugular and femoral access. It is also designed to be resistant to migration by maintaining complete apposition to the cava wall with both radial force and active tissue anchor fixation with no free struts to perforate the cava wall while providing consistent clot trapping over the entire IVC circumference. Furthermore, there are fewer hooks than other conical filters to allow for a greater window of retrievability. This report outlines the first experience of the Crux IVCF in man. The results are encouraging, with no complications resulting from device deployment, indwelling filters, or device retrieval. Specifically, there were no instances of caval thrombosis, recurrent PE, filter fracture, or migration. The major advantages to this filter include its ease of insertion, low profile, self-centering capability thereby deterring filter tilting, bi-directional design, and ease of retrievablity at up to 9 weeks post-implantation. However, while early animal and human trials have demonstrated great success, further clinical study with a larger number of patients is indicated and currently underway.