`
`Aortic and venous valve for
`percutaneous insertion
`
`D. Pavcnik, B.T. Uchida, H. Timmermans, C.L. Corless, ES. Keller and J. Rosch
`
`Dotter Interventions! Institute, Oregon Health Sciences University, Portland, OR, USA
`
`Summary
`
`The purpose of this paper is to present in vitro and in vivo experimental evaluation of a new, artificial,
`bicuspid, aortic and venous valve. Valves were constructed from square stents with barbs covered by
`porcine small intestine submucosa (SIS). A valve 15 mm in diameter was tested in a flow model (2.5
`l/min) with pressure measurement. A 100-ml rubber bag attached to a side arm of the flow model
`simulated heart ejection fraction. In acute (n26) and short—term (n=3) experiments conducted in four swine and
`four dogs, valves ranging from 16 - 28mm in diameter were placed into the ascending aorta through 10 F
`sheaths; three were placed subcoronary and six in a supracoronary position. Function and stability of the valves
`were studied with pressure measurements and aortograms. Three short-term animals were sacrificed for gross
`and histologic evaluation at one, two and four weeks respectively. In an acute experiment, venous valves with
`four barbs were placed into the IVC through an 8 F guiding catheter in three dogs. For longer-term testing,
`valves were placed into the NOS and iliac veins of three young swine. The animals were followed up after two
`weeks with venograms, then were sacrificed for gross and histologic evaluation.
`
`Keywords
`
`aortic valve, venous valve, stents and prostheses, interventional procedures, experimental, biomaterial
`
`Introduction
`
`diameter. Selection of the wire diameter depends on
`Expandable stents have been widely used for more
`the desired size and degree of expansile force of the
`then 10 years in the treatment of obstructions in
`square stent. The selected wire was hand—bent, 0” a
`vascular and nonvascular systems. Expandable stents
`also have a great potential as carriers of per- wooden template With fixed metal pegs enabling
`cutaneouslyplaced intravasculardevices. They have
`bending the wire into an exact square. Stent sizes
`been explored as carriers for an interiorvena cava filter
`ranging from 5 mm TO 50 mm can be made. The
`[1—6], a vascu|ar occlude-r [7], a prosthetic venous va|Ve
`corners Of the square stent were coil—bent, t0 I‘GdUCG
`[8], a Monodisk for closure of cardiac septal defects [9]
`stress and fatigue 01‘ the stent. When barbs are
`and a prosthetic aortic valve [10-12]_ Selfeexpandable
`needed for better fixation of the square stent, one or
`Gianturco Z« stents served as carriers for most of these
`bOTh wire ends were extended 1 2 mm ever the stent
`devices. We present a report on a new stent, a selfa
`frame 10 form a barb(s) on one or bO’th sides 0f the
`expandable square stent, and its potential asacarrier
`Steh’f- One,
`tWO or more anchoring barbs can be
`foravenous and aortic valve.
`attached to the other side of the square stent with
`metal cannullae. Square stents can be made as a
`single stent (Figure 1 a. b, c), or connected by an
`elongated barb into combination of two or more
`stents. These combinations can be made of stents of
`
`Square Stent
`The square stent was constructed in our research
`laboratory from stainless steel wire 0.006—0.02"
`
`Correspondence: D. Pavcnik MD, Dotter inten/entiona/ Institute, Oregon Health Sciences University. L342, 37 87 SW Sam Jackson Park
`Road, Portland, 0/? 97201,USA.
`'
`
`© 2000 Isis Medical Media Ltd
`
`287
`
`Edwards Lifesciences Corporation, et al. Exhibit 1032, p. 1 of 6
`
`
`
`D. Pavcnik er a/.
`
`various sizes and lengths, and with different degrees
`of expansile force.
`
`Venous and aortic valve
`
`The square stent becomes a valve when the stent
`with barbs on all four corners is covered with low
`
`porous material, such as small intestine submucosa
`(SIS) or polytetrafluoroethylene (PTFE) (Figure 2 a, b,
`c). SIS provides an acellular framework that becomes
`remodeled by host tissue, while being degraded and
`reabsorbed over time [18]. This makes SIS a unique
`covering for intravascular devices.
`same
`the
`have
`Venous
`and
`aortic
`valve
`construction and differ only by the sizes of the square
`stent and diameter of the wire from which they are
`
`made. The square stent has four barbs and its
`diagonal axis is constrained to the length of m,
`forming a diamond (rhombus or two equal triangles) in
`order to fit in the vein or aortic circumference of 2 nr.
`For veins 15 mm in diameter, a diagonal axis of 20
`mm square stent is constrained to 22 mm and for a
`80 mm aorta diagonal axis of 40 mm square stent is
`constrained to 46 mm.
`
`Two separate triangular pieces of SIS were sutured
`to the square frame with 7.0 Prolene monofilament,
`running sutures allowing for the gap between the
`diagonal axes. The valve was front-loaded into a
`guiding catheter; the 15 mm venous valve was loaded
`into an 8 F guiding catheter, the 40 mm aortic valve
`into a 10 F guiding catheter. For deployment a special
`
`
`
`
`Figure 1. The square stent. (a) Single square stent 28 mm in length with four barbs for self—attachment to the vessel wall
`(arrows). (b) Square stent retained by wire pusher connected to one barb. (c) Square stent deployed into a tube 20 mm in
`diameter.
`
`Figure 2. Valve design. (a) Non-restricted valve 20 mm in length with four barbs, (b) Deployed valve in a plastic tube, open
`position. (c) Deployed valve in a plastic tube, closed position.
`
`288
`
`Edwards Lifesciences Corporation, et al. Exhibit 1032, p. 2 of 6
`
`
`
`Aortic and venous valve for percutaneous insertion
`
`it
`pusher, with a small hook at its end was used.
`assured valve placement
`in proper position and
`prevented its dislodgement by blood flow before its
`barbs engage into the vessel wall. The valve was self-
`expanding so that the valve automatically assumed its
`operational form upon insertion. When the valve was
`deployed, two valvular sinuses were created between
`the venous or aortic wall and SIS or PTFE-mounted
`leaflets on the square stent. The valve was open in
`systole to permit fluid flow. In diastole, the valve was
`closed as its two triangular leaflets sealed against
`each other to prevent fluid flow.
`
`In vitro testing
`SIS-covered 20 mm square—stent valves, with four
`barbs, were repeatedly tested in a flow model,
`15 mm in diameter, for competency. Pressure was
`measured proximal and distal to the valve during
`prograde flow. The valves were exposed to retro—
`grade hydrostatic pressure of 60 mmHg, provided by
`a water column in a plastic tube. For venous testing,
`continuous ton/vard flow was approximately 250 mL
`min“. For aortic testing, continuous forward flow
`was 2.5 L min“. Valves were also tested with
`additional pulsatile flow. A 100 mL rubber bag,
`attached by a side arm at the lower end of the flow
`model, provided pulsatile flow. Manual compression
`of the bag was used to simulate the calf muscle
`pump and heart ejection fraction. To simulate the calf
`pump, a mild~force bag compression was applied for
`2~3 s; to simulate the heart ejection fraction, a fast,
`< 1 3 bag compression was used. The flow model
`was in a vertical position during testing.
`At rest, without flow, the valve was closed, with a
`hydrostatic pressure of 61 mmHg below and 60 mm
`above the valve. With initiation of continuous non-
`
`pulsatile forward flow, the valve opened practically
`immediately with low venous or high arterial flow. The
`valve stayed in an open position during the whole
`duration of continuous flow. With pulsating flow, and
`either mild or strong force compression of the bag,
`the valve stayed open but closed immediately after
`the injection.
`With the imitation of venous testing, pressures
`below the valve fell to 6—1 6 mm Hg (median 1 1 mmHg)
`and returned to the original 60 mmHg after 24—32 3
`continuous flow. When pressure increased to 45 mm
`Hg, the valve opened partially and stayed open to the
`next pulsatile injection. The valve pressure was
`unchanged.
`tested at hydrostatic
`The aortic valve was
`pressures of 100 mmHg. At rest without flow, the
`valve was Closed by a hydrostatic pressure of 101
`mmHg below and 100 mmHg above the valve.
`In
`
`bag
`fast
`a
`after
`immediately
`flow,
`pulsatile
`compression pressure above, it increased to 110 ( :t
`5) mm Hg and below it decreased to 79 ( 1 7)
`mmHg.
`
`Pilot animal study
`Aortic valve
`
`In acute (n =: 6) and short-term experiments (n = 3)
`conducted in four swine and four dogs, valves ranging
`from 16—28 mm in diameter were placed into the
`ascending aorta through 10 F sheaths; three were
`placed in the subcoronary and six in the supra—
`coronary position (Figure 3 a, b). Function and stability
`of valves was studied with pressure measurements
`and aortograms. Three animals were sacrificed for
`short-term gross and histologic evaluation at 1 , 2 and
`4 weeks, respectively.
`All the animals survived the initial post-implant
`period. The animals tolerated the procedure well and
`no arrhythmias or aortic pressure changes were
`observed. Valve movements were regular and there
`was no gradient across the valve. In two short—term
`animals with valves placed in the sub coronary
`position, one of the cusps of the native valve was
`trapped between the square stent and the aortic
`wall. This created considerable regurgitation and an
`appropriate model for evaluating SIS—valve efficacy.
`Both animals maintained systemic pressures of
`92/74 (
`:i: 9) mmHg and 110/80 ( :t 11) mmHg. None
`of the square stent valves caused stenoses and
`only small contrast regurgitation was seen in two
`animals. Left ventricular end-diastolic pressures were
`unchanged after
`stent~valve
`implantation in
`all
`animals.
`
`All nine prosthetic valves were undamaged by the
`implantation procedure. Aortic rupture was seen in
`one of the nine animals after 1 week. Aortograms
`revealed minimal regurgitation and no interference
`with coronary blood flow in all animals. Postmortem
`examination revealed the valves to be securely
`anchored. Histologic evaluation at 2 and 4 weeks
`revealed early remodeling of SIS with fibrocytes,
`fibroblasts and endothelial cells.
`
`Venous valve
`
`Venous valves were tested in an acute experiment
`on three dogs using a preloaded 20 mm square
`stent valve. The valve was placed into the 15 mm
`inferior vena cava (IVC)
`through an 8 F guiding
`catheter from the transjugular approach. Function
`of the valves and their stability was studied in the
`supine and upright positions, with injection of
`contrast medium and
`pressure measurement
`below and above the valve (Figure 4 a, b). For
`
`289
`
`Edwards Lifesciences Corporation, et al. Exhibit 1032, p. 3 of 6
`
`
`
`D. Pavcnik et a/.
`
`longer—term testing, valves were placed into the weeks by venograms and pressure measurement,
`IVC and iliac veins of three young swine via the
`and were then sacrificed for gross and histologic
`transjugular approach. Animals were followed for 6
`evaluation.
`
`Figure 3. Aortic valve aortogram showing competency of the supra coronary placed prosthetic aortic valve. (a) Diastole.
`(b) Systole.
`
`
`
`
`Figure 4. Venous valve cavogram in upright position with injection of contrast agent from the right femoral vein.
`(a) Early phase of injection shows valve patency with flow to upper IVC. (b) Late phase of contrast injection
`shows valve closure.
`
`290
`
`Edwards Lifesciences Corporation, et al. Exhibit 1032, p. 4 of 6
`
`
`
`Aortic and venous valve for percutaneous insertion
`
`Good valve function was observed in all animals.
`There was no pressure gradient through the valve in the
`supine position. A pressure gradient of 12—15 mmHg
`developed immediately in upright position, with less
`pressure below the valves. All veins remained patent
`at 6 weeks and smooth incorporation of the SIS
`valves into the vein wall was observed. Valves
`mimicked natural vein valves. Histologic evaluation
`demonstrated host tissue replacement and collagen
`tissue stroma remodeling, with variable fibrocytes,
`fibroblasts and some inflammatory cells. Vascular
`endothelial cells covered valve leaflets.
`
`Discussion
`
`The catheter balloon-valve principle was suggested in
`1980. The concept was balloon periodical closure of
`the insufficient aortic valve orifice, using a system
`attached to an aortic balloon-pump control unit [14].
`A completely revolutionary concept, a catheter-based
`aortic valve, was
`introduced in 1992 with a
`percutaneously introduced ball-valve [10]. Whereas in
`the previous systems the inserted valve remained
`connected to the introducing catheter, the deployed
`ball valve stayed in the aorta without any support of
`the catheter. The ball valve consisted of a modified Z-
`
`stent, serving as a valve cage, and a detachable
`balloon as valve. The cage was deployed first,
`followed by introduction of the detachable balloon.
`Animal experiments showed good potential for this
`ball—valve [10]. The ‘stent valve‘ bioprosthesis was
`introduced in 1992 [11].
`It consisted of an explanted
`porcine valve, fixed on a wire stent—skeleton. The
`whole system was compressed and mounted on a
`modified balloon catheter for vaivuloplasty, it was then
`placed into the aortic position using a sheath with an
`outer diameter of 13.6 mm. Balloon inflation was used
`to press the wire skeleton against the aortic wall.
`Testing was performed both for the supracoronary
`and infracoronary positions [1 1]. A similar model, with
`a valve made from the porcine pericardium mounted
`on a stent base, was tested successfully in animals.
`This model required a 24 F catheter for introduction
`[12]. The size of the delivery catheter for the disc valve
`(10 F sheath) is similar to that for delivery of the ball-
`valve and much smaller then the size of the stent-
`valve bioprostheses delivery catheters (24—41 F). A
`percutaneoust-introduced disc valve was described
`in 1999 [15]. As with the ball-valve and stent-valve
`bioprostheses, the disc valve was delivered with a
`catheter, but stayed in place on its own.
`The square stent was designed to be an
`intravascular implant—device carrier [16]. in order to
`squeeze a device through a delivery catheter small
`enough for percutaneous delivery, the stent structure
`
`must have a low profile and the covering material must
`be thin. Devices placed within the aorta must also have
`adequate strength and durability to withstand the
`aortic pressure of the blood flow. The main effort of the
`engineering involved was to construct the optimal stent
`framework with the lowest profile and enough
`expandable force to distend the covering material and
`secure hemostatic apposition of the occluder or valve
`to the artery, aorta or vein. We have shown that, with
`the square stent as a carrier, it is possible to introduce
`an aortic or venous valve covered with SIS per—
`cutaneously through a 6—1 0 F guiding catheter.
`In a pilot study, only square stents covered with
`SIS were tested as venous or aortic valves. Square
`stents can also be covered with other materials, such
`as PTFE or Dacron.
`
`Square stents and square—stentrbased devices are
`radiopaque and easy to place. Once the device is
`anchored against the vessel wall,
`it is released and
`the pusher catheter with the retention wire
`is
`removed. After deployment. the square stent self-
`centres, self-adapts and self-attaches with four barbs
`to the wall of tubular structure.
`
`The square stent is a new device with the potential
`to improve minimally-invasive treatment as a venous
`and aortic valve. The valve design is bicuspid and
`mimics natural valve anatomy. Initial studies showed
`that percutaneoust-placed SIS square—stent valves
`are promising one-way valves, capable of sustaining
`aortic and venous back—pressure, while allowing
`forward-flow with minimal resistance.
`
`Whether square-stent advantages in design, as a
`carrier for aortic and venous valves, will translate into
`long—term clinically—useful
`intravascular devices
`remains to be determined. More experimental studies
`are necessary to evaluate their long—term potential for
`possible future clinical use.
`
`References
`
`Inferior vena
`1 Wallace MJ, Ogawa K, Wright K et al.
`caval stent
`filters. Am J Roentgenol 1986;147(6):
`1247—50.
`2 Teitelbaurn GP, McGurrin M, Davies BL et a/. Insertion
`and recovery of a new retrievable vena caval filter. Work
`in progress. lnvest Rad/cl 1988;23(7):527—33.
`3 Ricco JB, Crochet D. Sebilotte P et al. Percutaneous
`transvenous caval interruption with the ‘LGM‘ filter: early
`results
`of
`a multicenter
`trial. Ann Vasc Surg
`1988;2(3):242—7.
`4 Pavcnik D, Wallace 8, Wright KC. Double nest vena
`cava filter
`for percutaneous
`insertion. Radiology
`1991;181(P):126~6.
`5 Sochman J, Peregrin J. A new inferior vena cava filter: in
`vitro
`test:
`and
`animal
`experiments. Cor Vasa
`1995;37(6):327—33.
`
`291
`
`Edwards Lifesciences Corporation, et al. Exhibit 1032, p. 5 of 6
`
`
`
`D. Pavcnik et al.
`
`Pavcnik D, Uchida BT, Keller FS, et a/. Retrievable IVC
`square stent
`filter: experimental study. Cardiovasc
`Intervent Rad/OI 1999;22:239—45.
`Kato N, Semba CP, Dake MD. Use of a self—expanding
`vascular occluder for embolization during endovascular
`aortic aneurysm repair. J Vasc
`Intervent Radiol
`1997;8s27—33.
`Uflacker H. Percutaneously Introduced artificial venous
`valve: Experimental use in pigs. The 1993 Annual
`Meeting of the Western Angiographic and Interventional
`Society, Portland, OR Abstract book 1993; 80.
`Pavcnik D, Wright KC, Wallace S. Monodisk: device for
`percutaneous transcatheter closure of cardiac septal
`defects. Cardiovasc Intervent Radio! 1993;16:308—12.
`Pavcnik D, Wright KC, Wallace 8. Development and initial
`experimental evaluation of a prosthetic aortic valve for
`transcatheter placement. Radiology 1992;183:151—4.
`Andersen HR, Knudsen LL, Hasenkam JM. Transluminal
`implantation of artificial heart valves: description of a
`new expandable aortic valve and initial results with
`
`1O
`
`11
`
`12
`
`13
`
`14
`
`15
`
`16
`
`implantation by catheter technique in closed chest pigs.
`Euro Heart J 1992;13:704—8.
`Moazami N, Basiler M, Argencia M et al. Transluminal
`aortic valve replacement — a feasibility study with a
`newly designed collapsible aortic valve. ASA/O J
`1996;42(5): M 381—5.
`l antz G0 at a/. Mechanical
`Hiles MC, Badylak SF,
`properties of xenogenic small—intestinal submucosa
`when used as an aortic graft in the dog. J Biomed Mater
`Res 1995;29:883—95.
`Moulopoulos SD, Anthopoulos LP, Antonatos PG et al.
`Intraraortic balloon pump for relief of aortic regurgitation.
`Experimental
`study.
`J Thorac Cardiovasc Surg
`1980;80(1):38—44.
`Sochman J, Peregrin J, Pavcnik D, Resch J. Prosthetic
`aortic disk valve for percutaneous insertion. Cardiovasc
`lntervent Rad/oi In press.
`Pavcnik D, Uchida B, Tirnn'iermans H at al. Square stent
`based large vessel occluder: an experimental pilot study.
`J Vasc Intervent Radiol In press.
`
`292
`
`Edwards Lifesciences Corporation, et al. Exhibit 1032, p. 6 of 6
`
`