A new vena caval filter for percutaneous placement and retrieval: experimental study

Accelerated migration and proliferation of smooth muscle cells cultured from neointima induced by a vena cava filter

Formation of a neointima is associated with grafted artery or vein, angioplasty, and stent and inferior vena cava filter (IVCF) implantation. Contributing to the neointima is a population of vascular smooth muscle cells (SMC) that migrates from media and subsequently proliferates within intima. The purpose of this present study was to culture SMC from normal vessel wall and from neointima and to compare migration and growth of these cells. Neointima was stimulated in the vena cava of pigs by placement of an IVCF for 30 days. Tissue was taken from the thickened wall between the struts and from a normal segment of the IVCF. After removal of the endothelium and adventitia, explants were placed in culture dishes and were observed for the migration of cells. Immunoassay for smooth muscle alpha-actin was used to identify cell origin. Proliferation was determined by cell counting. The cell cycle regulator cyclin D1 was detected by Western blot analysis. SMC phenotype was confirmed by positive immunostaining for smooth muscle alpha-actin. Cells migrated from the neointimal explants (NI-SMC) more rapidly than cells from explants of normal media (NM-SMC). Proliferation of NI-SMC was also more rapid than NM-SMC with or without exogenous mitogens. NI-SMC expressed more cyclin D1 than NM-SMC. Injury to the vena cava triggered neointima formation characterized by the expansion of a population of SMC with increased migration and replication compared with SMC from normal regions of the vessel.

A modified technique to minimize filter tilting during deployment of the Günther Tulip filter: in vitro study

With an in vitro model of the inferior vena cava, it was demonstrated that slight tension on the introducer during quick release of the jugular Günther Tulip filter led to less tilting of the filter compared with the technique recommended by the manufacturer. This technique may be useful to minimize significant filter tilting that may be associated with decreased filtration efficiency, and difficulty or impossibility of future filter retrieval.

A new optional vena cava filter: retrieval at 12 weeks in an animal model

PURPOSE

To test the feasibility and safety of percutaneous retrieval of a new inferior vena cava (IVC) filter, the Recovery Filter (RF), acutely and after 12-week implantation in sheep.

MATERIALS AND METHODS

The RF is a bilevel filter with stabilizing arms and elastic hooks that allow retrieval with a unique retrieval cone after incorporation into the wall of the IVC. Twenty-four filters were placed in the infrarenal IVCs of 18 sheep. In six sheep, two filters were placed and then removed immediately; three sheep were killed acutely and three were killed after a healing period of 3 weeks. In 12 sheep, a single filter was placed and then removed 12 weeks later; six were killed after retrieval and six were killed after an 8-week healing period.

RESULTS

The mean (+/-SD) transverse vena caval diameter was 15.3 mm +/- 2. All filters were deployed as intended and retrieved without difficulty. At sacrifice, there was no evidence of IVC perforation or retroperitoneal abnormality. The IVCs of the animals in the acute retrieval group showed minimal acute superficial injury that was largely reversed at 3 weeks. At 12 weeks, there was evidence of transmural incorporation of filter elements with narrowing of the IVCs. Solitary fibrotic abnormalities were present in the aorta adjacent to IVC lesions in nine of the 12 animals in the 12-week group. The IVC and aortic abnormalities were largely healed, with reversal of IVC narrowing after 8 weeks.

CONCLUSION

The recovery filter can be reliably and safely retrieved acutely and 12 weeks after implantation in sheep.

Clinical experience with temporary vena cava filters

An experience with temporary filter placement, which seems to be safe and effective for temporarily preventing pulmonary embolism, is reported. Since October 1997, six patients had temporary filters. There were two men and four women, with a mean age of 37 years. Three filters were placed at the infrarenal inferior vena cava, two at the suprarenal inferior vena cava, and one at the superior vena cava. All filters were placed before various surgical interventions. During filter placement, anticoagulation therapy was routinely performed. There were no complications at and during filter placement. No pulmonary emboli occurred during surgical intervention. All filters were successfully removed, two of which were exchanged for permanent filters. All patients are alive and well without recurrent deep vein thrombosis and/or pulmonary emboli during a follow-up period of 11 to 25 months. Although this experience is small, temporary filter placement is safe and effective for short-term prevention of pulmonary emboli even in older patients or those with malignant disease. Veins of the upper part of the body may be more favorable than the femoral vein for insertion of a temporary filter. Temporary filters can be safely placed not only at the infrarenal inferior vena cava, but also at the suprarenal inferior vena cava or superior vena cava.

Retrievable IVC square stent filter: experimental study

PURPOSE

In vitro and in vivo evaluation of a new retrievable, home-made, inferior vena cava (IVC) Square stent filter (SSF) with two trapping levels.

METHODS

In vitro, the SSF was compared in a flow model with the stainless steel Greenfield filter (SGF) for emboli-trapping efficiency by serially passing 300 emboli of 3 and 6 mm in diameter and 15-30 mm in length in each type of filter. Nine swine were used for the in vivo testing of the SSF for deployment and retrievability, emboli-trapping efficiency, stability, and self-centering ability and two were used (total of 11 swine) for testing repositioning and retrievability of the SSF at 2 weeks and for gross and histologic IVC changes at 2 months.

RESULTS

In vitro, the SSF and SGF had similar efficiency in trapping large emboli but the SSF had significantly better efficiency than the SGF for trapping all sizes of emboli (91.7% vs 81%), medium size emboli (93% vs 80%), and small emboli (86% vs 69%). Efficiency decreased in both filters from the first to the fifth embolus in each series but was still significantly better for the SSF. With the SSF, 89% of emboli were caught at the primary and 11% at the secondary filtration level. In the nine animals used for acute studies, the SSF was easily placed in all 27 attempts, assumed a central position 26 times, and was easily retrieved in 21 of 22 attempts. One tilted filter needed additional manipulation for retrieval. During emboli injection in five swine, the SSF had 97.2% emboli-trapping efficiency and demonstrated good stability. In the two animals used for longer-term evaluation, the filters were easily retrieved 2 weeks after implantation. Histologic evaluation at 2 months showed neointimal proliferation around the SSF wires in contact with the IVC wall, which was otherwise normal.

CONCLUSION

The SSF is a promising filter. It is easy to place and retrieve, is stable after placement, and has high efficiency for trapping emboli. Promising results justify further experimental and eventual clinical studies with a commercially manufactured SSF.

[Morphometric study of the wall of the caudal vena cava after the implantation of a filter in the sheep]

The long stabilizers of the VT-LGM filter rest on large areas of the vascular wall. The morphometric study of the layers of the vascular wall, after implantation of filter in 15 ewes, divided into 3 groups of follow-up (2, 4 or 8 weeks) of 5 animals, was made at 3 levels opposite the filter and 1 level outside of it. Changes are seen for all the layers. Filter produces intimal hyperplasia. Stabilizers are quickly isolated from the blood flow. The thickness of the intimal hyperplasia grows for 4 weeks. It is more important at the base of the filter than at its head. There is also hyperplasia of the media with no change according to the follow-up or the level opposite the filter. The adventitia becomes thinner without variation of time or level of the filter. Smooth muscle cells of the adventitia become less frequent and smaller. Their density in front of the stabilizers, is the smallest by 4 week follow-up and remains the same along the stabilizer. The full thickness of the wall is bigger opposite the stabilizers than between them. The filter produces changes that are limited in time and in space. The most important changes are seen at 4 weeks after insertion and opposite the stabilizers.

In vitro model for the evaluation of inferior vena cava filters: effect of experimental parameters on thrombus-capturing efficacy of the Vena Tech-LGM filter

PURPOSE

To determine the experimental parameters in an in vitro model that influence the thrombus-capturing efficacy of the Vena Tech-LGM filter.

MATERIALS AND METHODS

The Vena Tech-LGM filter was evaluated in an in vitro model of the vena cava with a computer-controlled flow system with a total of 5,200 thrombi. The influences of the following experimental parameters on the capture rate were analyzed with a multiple logistic regression model: type of testing (single, double, and multiple shot testing), thrombus diameter and length, IVC diameter and orientation, flow quality and quantity, flow velocity, and the length of the prepositioned thrombus.

RESULTS

A significant influence on the capture rate could be demonstrated for the type of testing, the thrombus diameter and length, the IVC diameter, and with double shot testing for the length of the prepositioned thrombus and the IVC orientation. The flow quality and the peak velocity were not significant. Based on these results, a protocol for in vitro testing of IVC filters was designed.

CONCLUSIONS

Experimental parameters influence the thrombus-capturing efficacy of the Vena Tech-LGM filter and should be taken into account when in vitro testing is performed.

IVC filter tilt and asymmetry: comparison of the over-the-wire stainless-steel and titanium Greenfield IVC filters

PURPOSE

A comparison of tilting, caval coverage, asymmetry, and insertion problems with the over-the-wire stainless-steel and titanium versions of the Greenfield filter.

MATERIALS AND METHODS

The study compared 104 stainless-steel and 141 titanium Greenfield inferior vena cava (IVC) filter insertions. The angle the sheath and deployed filter made relative to the cava, as well as filter strut distribution, were determined from spot films. The proportionate caval coverage was computed from the cavogram (anteroposterior projection). Mean filter tilts, subgrouped by insertion site, and caval coverage were compared with the Student t test, whereas strut patterns were analyzed with a contingency table.

RESULTS

The filter caval and sheath caval angles correlated. The filter caval angles varied with insertion site, but were lowest with a right jugular approach. Caval coverage was identical with both designs. The stainless-steel version resulted in a more uniform distribution of struts in comparison with the titanium version. The incidence of insertion problems was not significantly different between the filter types.

CONCLUSIONS

While IVC filter tilting was not improved with the newer design, the pattern of struts was more uniformly symmetric with the stainless-steel device. The right jugular insertion site was associated with the lowest filter caval angles and the most symmetric pattern of struts.

A physiologic in vitro model of the inferior vena cava with a computer-controlled flow system for testing of inferior vena cava filters

RATIONALE AND OBJECTIVES

The authors develop a physiologic in vitro model of the inferior vena cava (IVC) for testing of filters.

METHODS

The model is driven by a centrifugal pump. A computer-controlled electromagnetic valve is used for generation of different flow patterns. Limitation of the pressure increase in case of IVC occlusion is achieved by a bypass circuit. A glycerin solution is used for perfusion. Artificial clots are made from polyacrylamide gel. Data acquisition includes continuous monitoring of flow and difference pressure over the filter and video recording of the testing events.

RESULTS

The model can generate constant and pulsatile flows. The pressure increase can be limited to 70 mm Hg in case of occlusion. Calculation of the flow velocities in the IVC is possible. A classification of thrombus capturing is presented. The testing of most of the results are reproducible.

CONCLUSIONS

The in vitro model simulates the physiologic conditions in the IVC. It can be used for comparative testing of different filters and the evaluation of new filter designs.

Caval incorporation of the LGM Vena Tech filter: an experimental study

PURPOSE

To analyze LGM Vena Tech filter incorporation and the rapidity of the process.

MATERIALS AND METHODS

A filter was inserted into the infrarenal inferior vena cava (IVC) of 15 ewes assigned to one of three groups depending on the length of follow-up (2, 4, or 8 weeks). Radiologic data concerning IVC diameter and filter patency, stability, and incorporation were obtained before and after insertion and before euthanasia. Histopathologic analysis concerned wall thickness and smooth muscle cell area (SMCA) at three levels of the filter and at one point outside the filter.

RESULTS

All filters remained patent during follow-up. Incorporation of struts was dependent on time (P = .006), level of the filter (P = .0001), and strut surface (P < .0001). Neointimal thickness increased during follow-up (P = .0002), being more marked in the midportion of the filter (P = .0037). Adventitial thinning was observed (P = .0001), corresponding to a significant decrease in SMCA (P < .0001) above the struts as a function of the length of the follow-up period (P = .0021).

CONCLUSIONS

The LGM Vena Tech filter was well tolerated and is suitable for incorporation into the IVC wall of normal animals without risk of any deleterious reactions due to biological incompatibility.

Subcutaneously tethered temporary filter: pathologic effects in swine

PURPOSE

To evaluate the histopathologic effects of the Tempo-filter, a temporary caval filter, on the caval wall and determine the feasibility of deployment and removal of the device in swine.

MATERIALS AND METHODS

Filters were placed in the infrarenal inferior vena cava of 11 swine. The tethering catheter was sutured in a subcutaneous pocket near the puncture site. The original tethering catheter used in humans and a stiffer catheter designed to prevent migration in swine were evaluated. Postplacement, mid-study, and preexplant vena cavography procedures were performed. Four swine underwent in situ dissection at 3-10 weeks. Filters were removed from seven animals just before they were killed at 1-6 weeks.

RESULTS

All filters were successfully placed. All seven filters were successfully removed at up to 6 weeks after placement. Cephalic migration of more than 1 cm was observed in 10 of 11 swine (100% of original catheters, 83% of stiff catheters). Other complications were more common with stiffer tethering catheters, including caval stenosis in 40% of original catheters and 100% of stiff catheters, filter cone thrombus in 0% and 67%, tethering catheter thrombus in 20% and 83%, pulmonary embolism in 0% and 50%, and death in 0% and 17%, respectively. There was mild vessel wall damage in the vena cava.

CONCLUSION

Placement of the Tempofilter and removal at up to 6 weeks after placement is feasible.

Temporary inferior vena cava filters: in vitro comparison with permanent IVC filters

PURPOSE

An in vitro comparison of clot-trapping abilities of permanent and temporary inferior vena cava (IVC) filters.

MATERIALS AND METHODS

A flow model was used to simulate the IVC. Two permanent IVC filters, the titanium Greenfield and LG-Medical (LGM), were compared with two temporary filters, the Filcard International and Gunther. Clot sizes used were 2.5 x 2.5 mm, 2.5 x 5 mm, 5 x 5 mm, 5 x 10 mm, and 5 x 20 mm. Individual clots were presented to the filters with the simulated IVC in a horizontal or vertical orientation. Clot-trapping dynamics and pressure gradient changes during the injection of multiple, sequential clots were also examined.

RESULTS

As clot size diminished, all filters trapped fewer clots; however, the temporary filters trapped more small clots than the permanent filters. Very little difference was observed in clot-trapping abilities among the filters for clots of 5 x 10 mm or greater. In the horizontal orientation, the permanent filters trapped 38% of all clots delivered, while the temporary filters trapped 73%, chi 2 = 24.8 (P < .001). In the vertical orientation, the overall clot-trapping abilities of the filters improved, with the permanent filters trapping 73% of all clots delivered, while the temporary filters trapped 95%, chi 2 = 18 (P < .001). During trapping of multiple clots, the temporary filters allowed fewer clots to pass.

CONCLUSION

The temporary filters performed better than the permanent filters in both individual clot-trapping orientations. During multiple clot-trapping experiments, fewer clots were allowed to pass by the temporary filters. The temporary filters demonstrated the ability to capture clots both inside and outside the wire struts.

Retrievable IVC filter: preliminary in vitro and in vivo evaluation

PURPOSE

To develop an inferior vena cava (IVC) filter that is retrievable even after neointimal formation and incorporation into the caval wall.

MATERIALS AND METHODS

Eight filters were placed percutaneously in the infrarenal IVC of eight dogs. Four weeks after placement, percutaneous retrieval of the filters was attempted. Vena cavograms were obtained before and immediately after retrieval. Three dogs were killed immediately after the retrieval procedure was completed. In the other five dogs, follow-up vena cavography was performed 4-14 weeks after retrieval, and autopsy was performed.

RESULTS

The filters were placed and retrieved successfully in all eight dogs. No migration, caval penetration, or tilting occurred. The IVCs were completely patent both before and immediately after retrieval in all eight dogs. In the three dogs killed immediately after retrieval, neointimal hyperplasia was seen around the struts, but there was no detachment. Delayed stenosis was not seen in any of the five dogs that were followed up, and the inner surface of the caval wall was smooth.

CONCLUSION

This IVC filter can be easily placed and safely retrieved percutaneously, even after neointimal formation.

In vitro evaluation of vena cava filters

PURPOSE

The authors tested in vitro nine caval filters to evaluate clot-trapping capacity and transfilter pressure gradients under varying experimental conditions.

MATERIAL AND METHODS

The filtering efficiency of the stainless-steel Greenfield filter was evaluated in a modified flow model. A total of 2,100 clots were injected. In addition, controlled parameters simulating in vivo conditions were selected to appreciate clot-trapping capacity and pressure gradients for nine filters in horizontal and vertical positions, by using small (3 x 30 mm) and medium-sized (5 x 30 mm) clots. Statistical evaluation was performed for 3,600 clot deliveries.

RESULTS

Depending on experimental conditions, the clot-trapping capacity of the Greenfield filter varied greatly (0%-78%) by using small and medium clots. Bird's Nest and Simon nitinol filters demonstrated the highest filtering efficiency (94%-100%) for small clots. Medium clots were effectively captured by Bird's Nest, Simon nitinol, Antheor, and Günther Tulip filters (79%-100%).

CONCLUSION

Experimental parameters influence a filter's clot-trapping capacity. Bird's Nest filters demonstrated the highest clot-trapping capacity.

In vitro and in vivo experimental evaluation of a new vena caval filter

PURPOSE

A new stainless steel (MP35N alloy) vena cava filter without a central stasis point was evaluated in vitro and in vivo.

MATERIALS AND METHODS

The clot-trapping efficiency and hemodynamic flow pattern of the filter were assessed in a flow model and were compared with those of currently available commercial filters including the Vena Tech-LGM, Simon nitinol, Greenfield, and Bird's Nest filters. The new filter was placed in the inferior vena cava (IVC) of 31 dogs; 21 of the 31 dogs were followed up with cavography for up to 3 months. At the termination of the study, the filters and IVCs were examined grossly and histologically. An in vivo clot-trapping test was carried out in five dogs.

RESULTS

The least turbulence was noted with the new filter and the titanium Greenfield filter. The stainless steel Greenfield and Simon nitinol filters caused major flow disturbances. Migration within 5 cm of initial placement occurred in two animals (9.5%). There were no IVC thromboses, perforations, or filter embolizations. An in vivo clot-trapping study showed an 80% efficiency for small thrombi (3 x 20 mm) and 100% efficiency for large thrombi (6 x 20 mm) with the new filter. The Simon and the new filter had the best clot-trapping capabilities. The Vena Tech-LGM and Bird's Nest filters were slightly inferior and the Greenfield filter demonstrated by far the lowest trapping capacity.

CONCLUSION

The new vena cava filter is easily introduced percutaneously through a 12-F sheath and appears to be very promising due to its high filtering capability, low turbulence, nonmagnetic properties, good mechanical stability, and hypothrombogenicity. Clinical trials are warranted.

Effect of a 1.5 Tesla magnetic field on Greenfield filters in vitro and in dogs

The authors have tested four Greenfield filters for deflection in a 1.5 Tesla magnetic field and found large variations in the amount of deflection among filters. They also placed two filters in dogs and checked the filter for migration by taking radiographs before and after magnetic resonance imaging (MRI) scans. They found no evidence of migration. They conclude that, while most Greenfield filters respond to a magnetic field, the chance of migration of a filter because of an MRI scan is small. Therefore, MRI scanning of patients with Greenfield filters has little risk.

In vitro evaluation of caval filters

This experiment demonstrated the clottrapping ability of two commercially available filters, the Mobin-Uddin and Greenfield, and three experimental filters developed by Amplatz, Günther, and Gianturco. Each filter was tested in a polyethylene tube simulating the inferior vena cava. Separate series of 10 clots, each 3 cm long and 6 mm or 9.2 mm in diameter, were exposed to the test filter. The Mobin-Uddin and Amplatz filters failed by overload: acutely elevated pressures forced clots outside the skirt of the former, and between the limbs of the latter. The Günther filter trapped all incident clots, but migrated downstream when occluded by clot. The Greenfield filter passed clots at normal pressures between its widely spaced legs. The Gianturco (bird's nest) passed clots at normal pressures as well. While filters performed suboptimally, strengthening the anchoring struts of the Günther filter would result in a secure, effective filter.

First clinical results with a new caval filter

A new percutaneous inferior venacaval filter was used in 15 patients. The filter is constructed of stainless steel wires formed into a helical basket and anchoring legs. It can be inserted through a 10F catheter in an antegrade or retrograde fashion and can also be retrieved percutaneously. In a follow-up period of up to 13 months, the filter was well tolerated. One patient died of recurrent embolism despite filter protection. Local thrombogenicity was noted in 2 patients, but no other side effects were observed. Short-term heparinization following filter insertion is advisable. Heparin coating of the filter is under evaluation.

Misplacement of a vena cava filter in the retroperitoneum

An inferior vena cava filter was misplaced into the pericaval retroperitoneum. This was probably secondary to placement into and avulsion of the right spermatic vein. The filter delivery system was modified, and such misplacement should now be impossible.

Retrieving the Amplatz retrievable vena cava filter

The new Amplatz retrievable filter was placed 15 times into the inferior vena cava (IVC) of 7 dogs. Retrieval of the filter was attempted in 11 cases after 1 week and in 3 cases after 2 weeks. The retrieval was successful and without complication in all 14 cases. The 15th placement resulted in thrombotic occlusion of the IVC, and no retrieval was attempted.