AVena Cava Filter Using Thermal Shape Memory Alloy
`
`Diagnostic
`Radiology
`
`Experimental Aspects 1
`
`Morris Simon, M.D., Roy Kaplow, Ph.D.,2 Edwin Salzman, M.D., and David
`Freiman, M.D.
`
`Surgical ligation of the vena cava in the treatment of pulmonary embolism is already being
`supersededby devices introduced via a peripheral vein. A new metal alloy (nitinol) with unique
`memory characteristics forms the basis of an experimental device which promises even
`greater safety, simplicity and speed of introduction. It is inserted as a straight thin wire via
`the small bore catheter used for angiographic diagnosis. Upon reaching the lumen of the
`inferior vena cava and sensing body temperature, it reverts to its preset complex filter shape
`and locks into place permanently. It will trap further thromboemboli from the pelvis or lower
`limbs.
`
`INDEX TERMS: Embolism, pulmonary 6[0].720 • (Inferior vena cava, mechanical implant
`9[82].456). Nitinol • Venae Cavae, filters
`
`Radiology 125:89-94, October 1977
`
`A CUTE PULMONARY THROMBOEMBOLISM must be treated
`"
`immediately to prevent further, possibly fatal, embolic
`episodes. Anticoagulant drugs such as heparin or warfarin
`can prevent further thrombus formation (1) and thrombo(cid:173)
`lytic agents such as streptokinase and urokinase have
`been used to promote rapid lysis of existing thrombi, in(cid:173)
`In some patients,
`cluding the embolus (TABLE I) (17).
`however, the danger of internal bleeding precludes the use
`of anticoagulants or thrombolytic drugs, e.g., in those who
`have sustained cerebral hemorrhage, recent surgery, or
`fractures or other injuries in automobile accidents.
`In
`others, anticoagulation treatment must be discontinued
`because of bleeding complications (7, 18). Occasionally,
`anticoagulant treatment may fail
`to prevent recurrent
`embolization. In these cases, mechanical venous inter(cid:173)
`ruption may be attempted (TABLE II). Traditionally, this has
`been accomplished by surgical
`interruption of the vena
`cava or other great veins by ligation, stapling or clipping
`(2, 8, 11). Venous blood is then forced to return to the right
`heart by multiple collateral venous pathways too small to
`carry large emboli. In recent years the same result has
`been achieved by obstructive devices such as miniature
`umbrellas or balloons (2, 6, 9, 10, 12-15) introduced into
`the vena cava indirectly after gaining entry to the venous
`system. None of these procedures, however, is simple or
`without risk.
`The ultimate challenge remains the development of a
`blood clot filter which can be introduced into the vena cava
`through a fine bore catheter such as that normally used for
`the angiographic diagnosis of pulmonary embolism.
`Placement of the device can then be done as part of the
`angiographic procedure.
`Recently, a new class of biomaterials with unique
`thermal-shaped memory properties has been developed.
`is an alloy that can exist in two predetermined
`Nitinol
`physical forms depending on its temperature (4,5, 16). By
`
`TABLE I:
`
`THERAPEUTIC OPTIONS
`
`A. Anticoagulant drugs (heparin, coumadin)
`Thrombolytic drugs (streptokinase, urokinase)
`B. Vena cava interruption
`Ligation (Homans 1944)
`Surgical
`Plication (Deweese, Hunter 1958, Spencer 1959)
`Staples
`Clips (Moretz 1959, Miles 1964, Adams and DeWeese
`1966)
`
`Transvenous
`Temporary Catheter Sieve, (Eichelter 1968, exp)
`Balloon (Moser 1971, exp)
`Silastic net, (Wholey 1976, exp)
`Permanent Wire Clip, (Pate 1968, exp)
`Balloon (Hunter 1970, clinical)
`Umbrella, (Mobin-Uddin 1969, clinical)
`Thermal memory wire filter, (Simon 1975. exp)
`
`C. Embolectomy
`
`TABLE II:
`
`INDICATIONS FOR VENOUS INTERRUPTION
`
`1. Anticoagulants Contraindicated
`Postoperative
`Post-traumatic
`Postpartum
`Internal bleeding Cerebral hemorrhage
`Peptic ulcer
`Varices
`2. Failure of Anticoagulants
`3. Major Pulmonary Embolism with Residual Venous Thrombi
`4. Prophylaxis Recurrent iliofemoral thrombosis ± P.E.
`Femoral fractures in elderly
`Other high risk situations
`
`delicately varying the composition of the alloy it is possible
`to control precisely the temperatures at which the trans(cid:173)
`formation takes place. For this particular application the
`alloy is designed to exist at room temperature in its first
`form as a fine pliable wire that can easily pass through the
`standard angiographic catheter used for venography or
`pulmonary angiography. Upon reaching the vena cava the
`wire is extruded into the venous blood stream, encounters
`
`1 From the Departments of Radiology (M.S.), Surgery (E.S.), and Pathology (D.F.), Beth Israel Hospital/Harvard Medical School, Boston, Mass.
`Presented at the Sixty-second Scientific Assembly and Annual Meeting of the Radiological Society of North America, Chicago, 111., Nov. 14-19.
`1976.
`Supported by USPHS Grant HL 11414.
`2 Department of Metallurgy, Massachusetts Institute of Technology, Cambridge, Mass.
`
`shan
`
`89
`
`Not for External Distribution
`
`Edwards Exhibit 1029, p. 1
`
`

`

`90
`
`MORRIS SIMON AND OTHERS
`
`October 1977
`
`1000°
`FURNACE
`
`ANNEAL
`
`98°
`BODY
`
`70°
`ROOM
`
`PHASE
`CHANGE
`
`FULL SHAPE
`RECOVERY
`
`~~"'iETEMP @rfi))
`~
`+-
`~~ ~~E~! ~CO%R~
`
`PARTIAL
`
`"
`,.,.
`
`TIME
`Idealized diagram of thermal shape memory effect. The alloy is limp at room temperature, permitting loading and
`Fig. 1.
`unloading of the jig which restrains the wire during high temperature annealing. The imprinted filter shape is recovered at body
`temperature and the wire is now rigid.
`
`Discoverer
`Alloy Composition
`Physical Properties
`
`Bioacceptability
`
`TABLE III: NITINOL
`
`William Beuhler, U.S. Naval Ordnance Labs, 1962
`Nickel 53, Titanium 45, Cobalt 2, (iron?)
`Thermal shape memory (two crystal phases)
`Resists corrosion, abrasion, oxidation
`Nonmagnetic
`Forms wire, tubing, plates
`? Nonreactive in dogs
`? Nontoxic in tissue culture, dogs, sheep
`? Nonthrombogenic in dogs
`? Comparable to stainless steel, vitallium
`
`body temperature and is immediately transformed into its
`second form, a complicated rigid filter shape previously
`imprinted in its "memory." This permits the passage of
`normal blood but will arrest any significant thrombi on their
`way to the right side of the heart and lungs. The filter is also
`designed to lock itself firmly into the wall of the vena cava
`at the chosen level.
`
`THE THERMAL SHAPE MEMORY PHENOMENON
`
`The archetype of this phenomenon is seen in Exodus
`3, Chapter 4, Verse 2:
`
`"And the LORD said unto Moses, 'What is that in your hand?' And
`he said, 'A staff.'
`Then He said,
`'Throw it on the ground.' So he threw it on the
`ground, and it became a serpent; and Moses fled from it.
`But the LORDsaid to Moses, 'Stretch out your hand and grasp it by
`its taW-so he stretched out his hand and caught it, and it became
`a staff in his hand-"
`
`The contemporary metallurgic breakthrough which
`makes such transformations possible and which promises
`
`to usher in a new generation of medical prostheses derives
`from the work of Dr. William Buehler on the nickel/titanium
`alloys (nitinol) in the U.S. Naval Ordnance Laboratory in the
`early 1960s (TABLE III)(3, 19). Initial efforts were aimed at
`the delivery of a huge radio antenna into space by
`launching a small ball of "programmed" nitinol wire. Re(cid:173)
`cently the alloys have been made available for civilian use
`and have been adapted to a variety of industrial applica(cid:173)
`tions. In addition to the unique shape-memory Character(cid:173)
`istics they are non-magnetic and resistant to corrosion,
`oxidation and abrasion, all distinct advantages for potential
`biological usage.
`The way the thermal shape memory effect works is il(cid:173)
`lustrated diagrammatically in Figure 1. In its low temper(cid:173)
`ature phase, for example at room temperature, the ma(cid:173)
`terial can be very pliable and readily straightened or de(cid:173)
`formed to any desired shape. In this limp state the wire is
`easily wound into an appropriately designed jig to assume
`the desired shape of the vena caval filter. To imprint this
`shape on the wire's "memory" it is now annealed in a
`furnace at a temperature of about 1000 0F for about 25
`minutes while being restrained in the filter shape by the
`jig. The wire and jig are then allowed to cool to the low
`temperature phase permitting a reverse phase change in
`the wire's crystal structure so that it again becomes limp
`and plastic.
`It can then be removed from the jig and
`straightened. If the wire is now rewarmed to a temperature
`just above the phase change but well below the annealing
`temperature, i.e., to body temperature, it is immediately
`transformed to the filter shape. In this "high temperature
`phase" the wire is rigid.
`The transformation occurs with great rapidity and con(cid:173)
`slderable.torce if the metal
`is simply transferred from a
`
`Not for External Distribution
`
`Edwards Exhibit 1029, p. 2
`
`

`

`Vol. 125
`
`VENA CAVA FILTER USING THERMAL SHAPE MEMORY ALLOY
`
`91
`
`Diagnostic
`Radiology
`
`medium at one temperature to a medium at the other.
`However, by introducing the wire into the blood stream
`through a catheter, only that segment extruded beyond the
`tip of the catheter becomes transformed so that the op(cid:173)
`erator can control the rate of transformation by the speed
`of extrusion.
`At a microscopic level it is probable that the high tem(cid:173)
`perature phase is characterized by a highly symmetrical
`crystal structure, for example a cubic unit cell with one
`atom of the alloy at each corner of the cube and a second
`atom in the center. The low temperature phase is probably
`characterized by a lower level of symmetry of the crystal
`cell.
`In this phase the material allows considerable re(cid:173)
`arrangement of the crystals during low temperature de(cid:173)
`formation without normal plastic flow. This allows the re(cid:173)
`covery of the high temperature filter shape upon reheating
`(Fig. 1). The details of the phenomenon involve a diffu(cid:173)
`sionless, or "Martensitic" transformation but are still very
`controversial. However, the gross physical manifestations
`are apparent (3).
`The implications for designing a new class of medical
`prostheses are clear.
`In the low temperature phase a
`prosthesis can be given a form which will
`facilitate its
`passage through channels such as blood vessels or other
`anatomic conduits to reach the target site. At body tem(cid:173)
`perature or other appropriate high temperature the device
`is transformed into a shape designed to carry out an in(cid:173)
`tended therapeutic function. It is not essential to understand
`the basic physics of the transformation in order to exploit
`the phenomenon for medical applications. It is essential,
`however, that the alloy be manufactured to necessarily
`precise medical specifications. Small variations in com(cid:173)
`position of the alloy, not only of its major elements but also
`of trace contaminants, may radically alter the physical
`behavior of the material. The physical properties are also
`affected by the methods of manufacture and handling of
`the alloy. As a result, present commercial sources are
`unreliable and it is not possible to predict with accuracy
`the thermal memory characteristics of a particular nitinol
`sample. Samples must be tested until a suitable alloy with
`the degree of accuracy of composition, reliability and
`predictability of physical behavior essential
`for human
`medical applications is obtained.
`
`MEDICAL BACKGROUND
`
`The medical background originated with the suggestion
`by Homans in 1944 (8) that vena caval
`interruption is a
`rational method to prevent recurrent pulmonary embolism.
`However, the initial complication rate of direct surgical
`ligation exceeded the risks of recurrent embolism so that
`a number of safer surgical methods were developed using
`a variety of special sutures, staples or external vena caval
`clips (2, 11). However, all necessitated significant addi(cid:173)
`tional surgery, even though the patients were frequently
`already seriously ill as a result of their pulmonary embolism
`or underlying phlebothrombosis. To bypass this problem
`a number of investigators attempted to develop devices
`
`which could be introduced with less trauma by gaining
`access to the vena cava via a remote vein in the groin or
`neck (TABLE I). The best known and most successful of
`these is the Mobin-Uddin umbrella (12). This requires a
`cut-down procedure on the internal jugular vein in the neck
`under local anesthesia since the metal capsule which
`contains the folded umbrella is still fairly large. Further(cid:173)
`more, the umbrella plugs the catheter so that preliminary
`venographic control
`is not possible unless a separate
`angiographic catheter is introduced into the inferior vena
`cava for this purpose. The device is passed downward
`through the right atrium of the heart to reach the inferior
`vena cava where it is released and locked into position to
`serve as a barrier to future emboli. The thermal shape
`memory filter is designed to accomplish the same goal
`more simply, more efficiently and with less risk.
`
`PRESENT STATUS
`
`The feasibility of using the nitinol alloy to form a pros(cid:173)
`thetic device has been demonstrated in a series of in vitro
`and in vivo experiments performed under an initiation grant
`in the Atherosclerosis and Thrombosis Project at Beth
`Israel Hospital and Harvard Medical School. The research
`project involves collaborative medical and metallurgic
`efforts. The medical design and testing of inferior vena
`caval filters is being conducted in the Radiologic Research
`Laboratories at Beth Israel Hospital. Further studies of the
`nature, behavior and limitations of the alloys will be un(cid:173)
`dertaken in the metallurgic laboratories of the Massa(cid:173)
`chusetts Institute of Technology.
`The samples of nitinol wire that we have been able to
`obtain thus far have not yet met the precise medical
`specifications though some have come close. In general,
`wires that complete the phase transformation below nor(cid:173)
`mal body temperature, i.e., 90-9SoF, commence their
`shape change between 4S-S00F, i.e., below normal room
`temperature. For this reason we are presently required to
`work under water cooled below 4So in order to wind the
`wire into the jig for annealing or subsequentlyto straighten
`it for passage through the catheter into the canine vena
`cava or in vitro system. For the 20-30 second transit time
`through the catheter a cold solution is infused through the
`catheter to maintain the low temperature phase of the wire
`until
`the moment of extrusion into the blood stream.
`Working at low temperatures complicates the sterile in(cid:173)
`troduction procedure considerably but these problems will
`be corrected as soon as material of optimal transition
`temperature range is developed.
`
`FILTER DESIGN
`
`There are two functional elements in the filter, the filter
`mesh, and the locking system (Fig. 2). Many designs are
`being investigated for these elements, each requiring
`construction of an appropriate jig. The jig is made of
`stainless steel to withstand repeated annealing tempera(cid:173)
`tures in the furnace. The designsof the mesh have included
`
`Not for External Distribution
`
`Edwards Exhibit 1029, p. 3
`
`

`

`92
`
`MORRIS SIMON AND OTHERS
`
`October 1977
`
`Fig. 3. Experimental designs for the filter mesh-cloverleaf, grid
`and overlapping circles. The third design will accommodate to a range
`of sizes of the vena cava.
`
`Outer Catheter
`
`Inner Catheter
`
`Guide-Wire with
`Notched Adapter
`
`Fig. 4. Diagram of delivery system for filter.
`
`Fig. 2. A. Side view of filter implanted in dog vena cava.
`B. Radiograph of filter in excised vena cava with opacified embolus
`trapped in its mesh.
`
`three-leaf, four-leaf and five-leaf clover patterns, spiral and
`overlapping ring patterns (Fig. 3). In these designs the
`mesh has resulted in polygonal openings ranging from
`1 mm to 5 mm in width. More recently we have used a
`crisscross grid pattern with uniform square openings no
`greater than 2 mm in width. This will prevent passage of
`all but the tiniest emboli. The locking design comprises
`sharp-tipped leading and trailing points, both directed
`forward, which penetrate the endotheliumto a depth of 1-2
`mm. Further penetration is prevented by either a small
`metal stud near each tip or tight loops of the wire itself. The
`trailing end behaves like a fishhook or anchor.
`
`TECHNIQUE
`
`Preparation of the Filters
`
`a temperature of about 40°F and it is disassembled to
`allow removal of the new limp wire. The wire is then
`straightened in the cold water by drawing its ends apart.
`A small stud of metal tubing is attached to each end of the
`filter by a crimping tool.
`
`The Wire Delivery System
`
`The delivery system (Fig. 4) consists of a radiopaque
`outer catheter (3.0 mm outer diameter, 60 cm long), a
`matched Teflon inner catheter (1.85 mm outer diameter)
`and a guide wire (1 mm diameter) with a notched adapter
`at its tip to hold the rear stud of the filter wire during its
`transit through the catheters. The filter wire studs are at(cid:173)
`tached 1-2 mm from each end. They serve dual purposes:
`they limit the depth of penetration of the wire into the wall
`of the vena cava, and the rear stud is held firmly in the
`notch at the end of the guide wire while being drawn into
`or extruded from the inner catheter.
`
`A 20-cm segment of the nitinol wire (.5 mm diameter)
`is wound into the jig with aid of a transparent Plexiglas
`frame, with use of insulated gloves for working under
`cooled water (40°F). The furnace is preheated to 10200F
`and the wire which is now restrained in the filter shape by
`the jig is placed in the furnace for 20 minutes. The jig and
`wire are then removed and allowed to air-cool to room
`temperature. The jig is then immersed in the ice water at
`
`Preparation of Emboli
`
`Radiopaque emboli of the standardized size are used
`to test the filters, either in vitro or in vivo. The emboli are
`3 mm in diameter and 20 cm in length and are coated with
`tantalum dust for ready visibility by x-ray fluoroscopy as
`well as on subsequent plain films of the abdomen and
`chest. The emboli are prepared in a length of polyethylene
`
`Not for External Distribution
`
`Edwards Exhibit 1029, p. 4
`
`

`

`Vol. 125
`
`VENA CAVA FILTER USING THERMAL SHAPE MEMORY ALLOY
`
`Diagnostic
`Radiology
`
`93
`
`tubing after moistening its lumen by drawing steam through
`it and then lining it with tantalum dust (5 micron particle
`size) which adheres to the moistened wall. Blood is then
`drawn into the tubinq and allowed to clot. The tantalum is
`incorporated into the surface of the clot which then ap(cid:173)
`pears as a worm-like object coated with contrast agent in
`the bloodstream. The diameter of the embolus is readily
`measurable and its change in size with progressive lysis
`can be studied. The trapped emboli are visible for many
`months as they undergoorganization and incorporation into
`the vessel wall.
`
`The In Vitro System
`
`A section of thin-walled latex rubber tubing 30 cm in
`length and 15 mm in diameter is used to simulate the dog
`vena cava. It is connected to a hot/cold water mixing-valve
`which is adjusted to a temperature of 100°F and a rate of
`flow of 1 litre per minute, approximating the normal state
`of the dog vena cava. A Y-connector permits introduction
`of the catheter delivery system for the nitinol wire through
`a side arm into the lumen of the latex rubber tubing. This
`is done under fluoroscopic control. The catheter is per(cid:173)
`fused by cold solution while it is being introduced. The
`process of extrusion of the nitinol wire from the cool
`catheter into the warm tubing is recorded on cine film (35
`mm, 36 frames/second). A second Y-connector is used
`for delivery of the previously prepared radiopaque embolus
`and the arrival of the embolus at the filter mesh is again
`documented by cine filming. Any particles of radiopaque
`thrombus that penetrate the mesh are collected in a res(cid:173)
`ervoir at the end of the return tubing. The size and number
`of particles are noted as a measure of effectiveness of the
`particular filter device.
`
`The In Vivo System
`
`To test the effectiveness of the filter, its tolerance by
`the local tissues and the fate of the emboli we have used
`large dogs (18-20 kg), with adequate sized inferior vena
`cavas, to perform preliminary biological studies. After
`anesthetizing the animal with 30 mg/kilo of veterinary
`Diabutal (sodium pentobarbital), intubation of the airway
`and control of breathing with a Harvard pump respirator,
`the two femoral veins are exposed by cut-down on each
`side. On the left side the outer catheter of the wire intro(cid:173)
`duction system is inserted by venotomy and a slow drip
`infusion of saline started. On the right side the larger
`catheter housing the opacified embolus is inserted by
`venotomy. The animal
`is positioned for fluoroscopy with
`a C-arm intensifier system with 35mm filming. A prelimi(cid:173)
`nary vena cavography is performed to check the caliber
`of the vena cava and to determine the optimal position of
`the outer catheter tip just below the level of the renal
`veins.
`
`Insertion Procedure
`
`The nitinol wires thus far available have had to be cooled
`
`below room temperature to about 40-50oF for the low
`temperature phase manipulations. In ice-cooled water, the
`rear stud of the straightened filter wire is engaged in the
`notched end of the guide wire and drawn into the fine-bore
`inner Teflon catheter. An infusion of cooled normal saline
`is now started with a double syringe pump connected both
`to the inner catheter housing the straightened filter and the
`outer catheter in the femoral vein of the dog. The infusion
`rate is .75 ml per second on each side, providing cooled
`solution for 40 seconds from two 30-ml syringes. Under
`fluoroscopic control and cine recording, the inner catheter
`is now rapidly advanced into the outer catheter through a
`Touhy-Borst seal su the catheter tips line up. The guide
`wire is then extruded into the blood stream of the vena
`cava by first withdrawing the catheters over it for about 5
`ern, then advancing the guide wire about 10 cm to form the
`grid of the filter and then again withdrawing the catheters
`till the rear hook of the filter is released. As the extrusion
`proceeds the wire is warmed to body temperature and
`instantly restored to its filter shape. This also locks it into
`place. The inner catheter and guide wire are now removed
`and the outer catheter is withdrawn into the iliac vein for
`follow-up venography to confirm proper seating and pa(cid:173)
`tency of the device.
`The filter is now tested by flushing the embolus from the
`plastic tubing into the left femoral vein with a bolus of
`normal saline. The passage and trapping of the embolus
`are recorded on cine film for slow motion analysis.
`If
`particles of the embolus penetrate the filter they are fol(cid:173)
`lowed into the lungs of the dog. The effect of the trapped
`embolus on venous return is recorded by a further vena
`cavogram.
`
`Follow-up Studies
`
`Since the experimental thrombus is opacified by tan(cid:173)
`talum powder incorporated into its surface layer the fate
`of the trapped embolus and of transmitted particles in the
`lungs can be documented by follow-up plain films of the
`abdomen and chest. These are taken periodically at in(cid:173)
`tervals of hours, days and months. Follow-up venography
`is also performed at intervals to check changes in patency
`of the vena cava. Finally the animals are sacrificed at
`varying stages to determine the effect of the device on the
`vein wall and local tissues, the degree of lysis, disinte(cid:173)
`gration or organization of the embolus or the absorption
`of the embolus into the vessel wall. The nitinol wire is also
`examined for evidence of formation of new thrombus. The
`fate of embolic fragments in the lung is studiedat the same
`time.
`
`RESULTS
`
`Approximately 50 filters have been tested in vitro and
`16 have been successfully placed in dogs. The initial de(cid:173)
`signs with relatively large openings permitted passage of
`significant sized fragments of the experimental embolus
`which is only 3 mm in diameter. Later designs have proved
`much more effective and permitted passage of only oc-
`
`Not for External Distribution
`
`Edwards Exhibit 1029, p. 5
`
`

`

`94
`
`MORRIS SIMON AND OTHERS
`
`October 1977
`
`casional minute fragments of opacified clot, a few milli(cid:173)
`meters in length and almost certainly too small to be of
`clinical significance. The most effective design is a fine
`mesh grid forming 2mm square openings (Fig. 3). Other
`problems encountered in dogs of varying size and shape
`included vena cavas which were too large for the filter and
`allowed small sections of emboli to pass around them and
`vena cavas which were too small and prevented the
`transformation to a perfect filter shape. In one small animal
`the rear hook was unable to form properly so that the filter
`was not locked into position and was pushed upwards
`about 1.5 cm by the embolus. In no other animal has the
`device failed to lock into position or been shown to be(cid:173)
`come dislodged. We have now designed an expandable
`filter which will accommodate itself to a large range of
`vena caval cross sections.
`Autopsy examination of the vena cava has been per(cid:173)
`formed in 12 dogs, 9 of them on the day of the experiment,
`3 after delays of one week, three weeks and three months,
`respectively. In all animals the points of the device were
`found to have penetrated the wall of
`the vena cava
`obliquely to the level of the stud, i.e., 1 or 2 mm. None
`showed perivenous hemorrhage. One leading point of an
`early device with a small stud was shown to have perfo(cid:173)
`rated the vena cava and was embedded in the wall of the
`aorta without hemorrhage. Larger studs have been used
`subsequently. The delayed autopsies showed development
`of a cowl of tissue on the outer surface of the vena cava
`over the penetrating points of the filter.
`In the three chronic experiments the opacified embolus
`trapped by the filter had retracted towards the wall of the
`vena cava and had become incorporated into it. In the dog
`sacrificed after three months the outer portions of the filter
`loops were buried in the vessel wall. This suggeststhat the
`device may become incorporated into the wall of the vena
`cava as if fixed in place by surgical sutures. No fresh
`thrombus was observed on the exposed surface of the filter
`wire in these cases.
`While these initial experiments have proved very en(cid:173)
`couraging it is evident that systematic efforts must now be
`undertaken to develop an optimum filter and delivery
`system suitable for human applications and to permit more
`detailed biological and pathological evaluation. Particular
`efforts must be made to optimize the composition, the
`thermal treatment and the mechanical preparation of the
`alloy so that the behavior of the device will more closely
`conform to the idealized medical specifications in a reliable
`and reproducible manner.
`ACKNOWLEDGMENTS: We are indebted to Dr. William Buehler for
`initial assistanceand guidanceand to Mr. DavidGoldstein, his successor
`in the Materials Division of the U.S. Naval Surface Weapons Center,
`for many helpful suggestions and for continuingefforts to develop better
`
`nitinol materials. We would also like to express appreciation for the
`valuableassistanceof SandyPeterson, StevenKeeganandJerry Garlitz,
`research radiologicaltechnologists.We are particularlygratefulto Itzhak
`Bentov, Engineering Consultant, Chad Sutton, research fellow for two
`summers, and Norma Elkind, for assisting with many essential de(cid:173)
`tails.
`
`Morris Simon, M.D.
`Department of Radiology
`Beth Israel Hospital
`330 Brookline Avenue
`Boston, Mass. 02215
`
`REFERENCES
`
`1. Barritt OW, Jordan SC: Anticoagulant drugs in the treatment
`of pulmonary embolism: a controlled trial. Lancet 1:1309 -1312, 18
`Jun 1960
`2. BernsteinEF: The place of venousinterruptionin the treatment
`of pulmonary embolism.
`[In] Moser KM, Stein M, eds: Pulmonary
`Thromboembolism. Chicago, Year Book Medical Publishers, 1973,
`pp 312-323
`3. Buehler WJ, Cross WB: 55 Nitinol-unique wire alloy with a
`memory. Wire Journal 1:41-49, Jun 1969
`4. Buehler WJ, Wang FE: A summary of recent research on the
`nitinol alloys and their potential application in ocean engineering.
`Ocean Eng 1:105-120, Jul1968
`[In] Transactions Quarterly of the
`5. Buehler WJ, Wiley RC:
`American Society for Metals 55:269, 1962
`6. DeWeese MS, Hunter DC Jr: A vena caval filter for the pre(cid:173)
`vention of pulmonary emboli. Bull Soc Int Chir 17:17-25, Mar 1958
`7. Hodin E, Dass T: Spontaneous retroperitoneal hemorrhage
`complicating anticoagulant therapy. Ann Surg 170:848-851, 1969
`8. Homans J: Deep quiet venous thrombosis in the lower limb.
`Surg Gynecol Obstet 79:70-82, Jul 1944
`9. Hunter JA, Sessions R, Buenger R: Experimental balloon
`obstruction of the inferior vena cava. Ann Surg 171:315-320, Feb
`1970
`10. Hunter JA, Sessions R, Petasnick J: Therapeutic balloon
`occlusion of the inferior vena cava. JAMA 234:1034-1037,8 Dec
`1975
`11. Lindenauer SM: Prophylactic staple plication of the inferior
`vena cava. Arch Surg 107:669-675, Nov 1973
`12. Mobin-lJcldin K, McLeanR, Jucle JR: A new catheter technique
`of interruption of inferior vena cava for prevention of pulmonary em(cid:173)
`bolism. Am Surg 35:889-894, Dec 1969
`13. Mobin-Uddin K, Callard GM, Bolooki H, et al: Transvenous
`caval interruption with umbrella filter. New EnglJ Mad 286:55-58, 13
`Jan 1972
`14. Moser KM, Harsanyi PG, Harvey-Smith W, et al: Reversible
`interruption of inferior vena cava by means of a balloon catheter.
`Preliminary report.
`J Thorac Cardiovasc Surg 62:205-212, Aug
`1971
`15. Pate JW, Melvin 0, Cheek RC: A new form of vena caval in(cid:173)
`terruption. Ann Surg 169:873-880, Jun 1969
`16. Robinson AL: Metallurgy: extraordinaryalloys that remember
`their past. Science 191:934-936, 1976
`17. Urokinase Pulmonary Embolism Trial Study Group: Urokinase
`pulmonary embolism trial. Circulation 47 (Suppl 2), 1973
`18. Vieweg WV, Piscatelli RL, Houser JJ, et al: Complications
`of intravenous administration of heparin in elderly women.
`JAMA
`213:1303-1306,24 Aug 1970
`19. Wang FE, Buehler WJ, Pickart SJ: Crystal structure and a
`unique "Martensitic" transition of TiNi. J Appl Phys 36:3232-3239,
`1965
`
`~§~~
`
`Not for External Distribution
`
`Edwards Exhibit 1029, p. 6
`
`