DETERIORATION OF BIOPROSTHETIC HEART VALVES
`
`M381
`
`16. Hughes H, Tipton LS, Feuchuk D, Prabhakar G, Aboul-Enen HY,
`Duran CMG: Glutaraldehyde 'Y carboxyglutamic acid and
`calcium in explanted bioprosthetic heart valves. J Heart Valve
`Di.d: 111 116, 1994.
`17. Lee M), Boughner DR: Tissue mechanics of canine pericardium
`
`in different test environment: evidence for time dependent
`accommodation, absence of plasticity and new roles for col(cid:173)
`lagen and elastin. Circ Res 49: 533-544, 1981.
`18. Scott M, Vesely I: Aortic valve cusp microstructure: the role of
`elastin. Ann Thorac Surg 60: S391-S394, 1995.
`
`Transluminal Aortic Valve Placement
`
`A Feasibility Study With a Newly Designed Collapsible Aortic Valve
`
`NADCR MOAZAMI, MARC BESSLER, MICHAEL ARGENZIANO, ASIM F. CitOUDHRI, SANTOS E. CABRCRIZA,
`JOHN D. F. ALLENDORF, ER1c A. RosE, AND MutMET C. Oz
`
`Percutaneous stents are used in vascular applications in con(cid:173)
`junction with angioplasty and in combination with graft ma(cid:173)
`terial for repair of abdominal aneurysms. The authors have
`designed a collapsible bioprosthetic aortic valve for place(cid:173)
`ment by a transluminal catheter technique. This trileaflet
`stent valve is composed of stainless steel and bovine peri(cid:173)
`cardium. Stent valves, 23 and 29 mm, were tested in a pulse
`duplicator system with rigid rings from 21 to 31 mm in 2
`mm increments. At a mean flow of 3.1 I/min (±0.7), normal
`systemic aortic pressure was generated with a transvalvular
`gradient of 14.9 ± 7 mmHg (mean± SD). Regurgitation frac(cid:173)
`tion ranged from 10 to 18% (mean 13.8 ± 3%) in the best
`ring size. Valves with the best hemodynamic profile were
`used for implantation in three 70 kg pigs in an open chest
`model. The valve was collapsed in a 24 Fr catheter designed
`to allow slow, controlled release. After resection of the na(cid:173)
`tive leaflets, the new valve was placed in the subcoronary
`position. No additional sutures were used for securing the
`valve. Two animals were successfully weaned from cardio(cid:173)
`pulmonary bypass and maintained systemic pressures of
`100/45 (±10) and 116/70 (±15) mmHg, respectively. Intra(cid:173)
`operative color echocardiography revealed minimal regur(cid:173)
`gitation, central flow, full apposition of all leaflets, and no
`interference with coronary blood flow. Both animals were
`sacrificed after being off bypass for 2 hr. Postmortem exam(cid:173)
`ination revealed the valves to be securely anchored. The
`third animal was weaned from cardiopulmonary bypass but
`developed refractory ventricular fibrillation because of
`valve dislodgment due to structural failure. Although long
`term survival data are needed, development of a hemody(cid:173)
`namically acceptable prosthetic aortic valve for translumi(cid:173)
`nal placement is feasible. ASA/0 Journal 1996;42:M381-
`M385.
`I ntravascular stents have gradually obtained acceptance for
`
`3 Com(cid:173)
`peripheral and coronary vascular occlusive disease. 1
`-
`binations of stents with grafts have allowed for intraluminal
`bypass of aneurysms and long segment atherosclerotic Je(cid:173)
`sions.4 Transfemoral percutaneous aortic valve replacement
`is an extended application of this new technology that offers
`several new challenges. First, the valve needs to be accu(cid:173)
`rately positioned without the benefit of direct open visual(cid:173)
`ization. Second, the valve must remain firmly in the annulus
`under physiologic conditions. Finally, in the absence of su(cid:173)
`tures, the valve must provide a sufficient seal to minimize
`perivalvular regurgitation.
`We have developed a novel sutureless bovine pericardia!
`bioprosthetic aortic valve mounted on a collapsible stent.
`This prosthetic valve can be collapsed in a 24 Fr catheter for
`delivery. This report represents our initial characterization of
`this valve in a pulse duplicator system and in an open chest
`porcine model.
`
`Methods
`
`Stent Valve Construction
`
`Stainless steel wire, size 0.020", was used to construct all
`stents, and size 0.022" was soldered on the ring to serve as
`the anchoring fixation point in the aortic annulus. For con(cid:173)
`structing the stent valves at different sizes, because of the
`collapsible nature of the design, all valves were placed in a
`ring of predetermined size. Subsequently, bovine pericar(cid:173)
`dium stabilized by a photooxidation process 5 was sewn on
`the stent from a single sheet to construct trileaflet valves con(cid:173)
`strained to different sizes.
`
`In Vitro Measurements
`
`From the Department of Surgery, Columbia-Presbyterian Medical
`Center, New York, New York.
`Reprint requests: Nader Moazami, MD, Columbia Presbyterian
`Medical Center, PO Box 3fl3, 630 W 168 St., New York, NY 10032.
`
`A total of 16 stent valves were constructed. The four that
`were acceptable in terms of overall symmetry were sub(cid:173)
`jected to further studies. A pulse duplicator similar to a pre(cid:173)
`viously described mock circulatory system 6 was constructed
`
`Edwards Lifesciences Corporation, et al. Exhibit 1008, p. 1 of 5
`
`

`

`M382
`
`MOAZAMI fT AL.
`
`ments, Milford, MA) recorded with the MacLab system and
`stored on the hard drive for subsequent analysis. Regurgitant
`fraction was calculated by the ratio of the integral of back(cid:173)
`ward flow to the integral of total flow (forward and backward)
`using customized wave analysis software (IGOR; Wave(cid:173)
`metrics, Inc., Lake Oswego, OR).
`
`Animal Care
`
`Aorlk Pn:s::;urc
`Transducer
`
`Left Ventricular
`Pressure Transducer
`
`Animals received humane care in compliance with the
`Principles of Laboratory Animal Care formulated by the Insti(cid:173)
`tute of Laboratory Animal Resources and the Guide for the
`Care and Use of Laboratory Animals published by the Na(cid:173)
`tional Institutes of Health (NIH Publication No. 86-23, re(cid:173)
`vised 1985).
`
`Dual Chamber
`Pulsatilc Pump
`
`Figure 1. Diagram of the mock circulatory loop used for in vitro
`testing of valves. The chamber allows for adjustment of compliance
`and resistance.
`
`(Figure 1). This system uses an Abiomed BVS 5000 dual
`chamber (Abiomed, Inc., Danvers, MA) to generate pulsatile
`flow. Each stroke ejects 80-85 ml of saline into the circuit.
`The system possesses an open reservoir to simulate the ve(cid:173)
`nous and atrial systems. In addition, a piston-cylinder-spring
`arrangement functions as an adjustable compliance cham(cid:173)
`ber. Peripheral vascular resistance could be adjusted by ap(cid:173)
`plication of a clamp to the circuit partially to impede forward
`flow. The valve was placed on a ring in a rigid chamber made
`of Plexiglas. The ring size in the chamber could be changed
`to a diameter ranging from 21 to 31 mm in 2 mm increments.
`A 5 Fr Millar catheter (Millar Instruments, Houston, TX) was
`placed on the ventricular side and the aortic side of the valve
`to measure transvalvular pressures gradients. An ultrasonic
`flow probe (Transonics, Ithaca, NY) was placed on the out(cid:173)
`flow tubing to monitor forward and backward flow. All data
`were digitized at 200 Hz using a 12 bit A-D board (AD lnstru-
`
`Experimental Procedure
`
`Pigs weighing 60-70 kg were anesthetized with ketamine
`(20 mg/kg) and thiamylal sodium (4.5 mg/kg). After endotra(cid:173)
`cheal intubation, anesthesia was maintained with isoflurane
`(1.5-2.5%) mixed with 100% oxygen. Lidocaine 2 mg/kg
`and bretyllium 5 mg/kg were given as a loading dose, and
`a continuous drip of 1 mg/kg/hr for each was maintained
`throughout the experiment. Standard concentrations of
`phenylephrine (Neo-Synephrine; Winthrop Pharmaceuti(cid:173)
`cals, New York, NY), norepinephrine bitartrate (Levophed;
`Winthrop Pharmaceuticals), and epinephrine drips were
`used as needed to maintain a mean blood pressure of 60
`while on cardiopulmonary bypass (CPB). Internal paddles
`charged up to 50 V were used for defibrillation as needed.
`
`Surgical Procedure
`
`Sternotomy was performed, and the pericardium was
`opened. Heparin (300 µg/kg) was administered and ade(cid:173)
`quate anticoagulation was verified by an activated clotting
`time of greater than 400 sec. A 14 Fr cannula was inserted in
`the aorta and a 36 Fr dual stage venous cannula was inserted
`in the right atrium for CPB. The aorta was dissected proxi-
`
`Table 1. Hemodynamic Characterization of the Stent Valves in the Mock Circulatory System
`
`Valve Size (mm)
`
`Ring Size (mm)
`
`VP(mmHg)
`
`AoP(mmHg)
`
`Regurgitant Fx (%)
`
`Flow (L/min)
`
`Empty chamber (no valve)
`
`23
`
`23
`
`29
`
`29
`
`25
`27
`29
`31
`21
`23
`25
`27
`21
`23
`25
`27
`29
`31
`27
`29
`31
`
`177
`158
`165
`157
`173/20
`188/23
`160/25
`193/35
`130/15
`174/9
`153/25
`161/10
`163/12
`169/18
`137/9
`145/18
`144/25
`
`167
`149
`156
`152
`163/70
`178/70
`150/33
`184/36
`120/60
`169/52
`145/37
`146/63
`149/52
`147/46
`122/60
`120/42
`114/30
`
`100
`100
`100
`100
`10
`14
`26
`50
`14
`18
`27
`13
`15
`20
`10
`17
`27
`
`VP, ventricular pressure; AoP, aortic pressure; Fx, fraction.
`
`3.3
`3.3
`3.3
`3.3
`2.7
`3.2
`2.2
`2.1
`2
`2.1
`2.1
`3.71
`3.67
`4.5
`3.24
`3.46
`3.28
`
`Edwards Lifesciences Corporation, et al. Exhibit 1008, p. 2 of 5
`
`

`

`PERCUTANEOUS AORTIC VALVE
`
`M383
`
`160
`
`120
`
`80
`
`40
`
`0
`
`0:33
`
`0:34
`
`0:35
`
`Time (secs)
`
`Figure 2. Representative in vitro aortic and ventricular pressure
`tracing for a 23 mm stent valve in a 21 mm ring.
`
`mally until the origin of the left main coronary artery could
`be visualized. Subsequently, the animal was placed on CPB
`and cooled to 32 ° C. After placement of a left ventricular vent
`through the apex of the heart, the proximal aorta was cross(cid:173)
`clamped. One liter of 4 °C standard crystalloid cardioplegia
`solution was delivered to the aortic root to arrest the heart. A
`partial transverse aortotomy was performed as close to the
`coronary artery as technically feasible. The deep aortic root
`in this animal model necessitated the use of three traction
`sutures placed at each commissure to pull the native leaflets
`into view. After sizing the annulus with standard sizers, the
`appropriate stent valve, collapsed in a 24 Fr catheter, was
`delivered under direct vision and secured in placed. No su(cid:173)
`tures were used for fixation of the stent valve. All traction
`sutures were removed and the aortotomy was closed with
`a double layer of running 4-0 polypropylene suture. After
`removing the cross-clamp, deairing the left ventricle, and re(cid:173)
`warming, the animal was slowly weaned from CPB.
`
`Hemodynamic Measurements
`
`Pressure was measured with a 5 Fr Millar micromanometer
`placed in the left ventricle through the apex, and a second
`catheter guided to the proximal aorta through the left carotid
`artery. All hemodynamic data were recorded off CPB on the
`Maclab system. All data are reported as mean ± standard
`deviation.
`
`1:chocardiography
`
`Two dimensional and m-mode echocardiograms were ob(cid:173)
`tained using a hand held 5 MHz ultrasound transducer
`(Vingmed CFM 750; Vingmed Sound, Inc., Salt Lake City,
`UT). Color flow Doppler was obtained to assess valvular re(cid:173)
`gurgitation.
`
`Table 2. Size of the Aortic Annulus After Cardiac Arrest
`Relative to the Size of the Stent Valve Implanted
`
`Animal
`
`1
`2
`3
`
`Annulus Size
`(mm)
`
`Valve Size
`(mm)
`
`18
`25
`20
`
`23
`29
`23
`
`Figure 3. (A) Short axis echocardiogram demonstrates the three
`leaflets to be fully apposed during diastole. One of the leaflets is
`slightly redundant. (B) Long axis view demonstrates opening of the
`leaflets. The lower, redundant leaflet does not open completely.
`
`Results
`
`In Vitro Studies
`
`Table 1 shows the results obtained from studying valves of
`different sizes in our pulse duplicator system. Flow gener(cid:173)
`ated ranged from 2.0 to 4.5 L/min with a mean of 3.1 L/min
`(±0.7). The lowest regurgitation fractions were obtained by
`placing the valves in rings 2 mm smaller than the stent valve.
`Placing the valve in smaller chambers caused too much
`leaflet redundancy and interference with proper valve open(cid:173)
`ing. The valves were considered adequate for animal implan(cid:173)
`tation if the regurgitant fraction was less than 15% at a given
`annulus size and a diastolic pressure of 60 mm Hg could be
`maintained. Calibration of the pulse duplicator in absence of
`any valve shows that the system has a transvalvular gradient
`of 5-10 mm Hg. The stent valves had a mean transvalvular
`gradient of 14.9 (±7.4) mmHg. In Figure 2, representative
`pressure and flow curves generated in our mock circulatory
`loop are shown.
`
`Edwards Lifesciences Corporation, et al. Exhibit 1008, p. 3 of 5
`
`

`

`M384
`
`MOAZAMI ET AL.
`
`Animal 2 was successfully weaned from bypass, but within
`5 min refractory ventricular fibrillation developed. Postmor(cid:173)
`tem examination revealed dislodgment of the valve second(cid:173)
`ary to structural failure.
`Animal 3 had an annulus sized at 20 mm and received a
`23 mm stent valve. This animal maintained a systemic blood
`pressure of 116/70 (±15) mm Hg. Pressure recordings at the
`aortic root and left ventricle revealed an asystolic gradient of
`2 (±1) mmHg. The mean diastolic gradient was 66 (±3)
`mmHg with the left ventricular diastolic pressure at 4.9 (±1)
`mmHg, suggesting absence of significant regurgitation (Fig(cid:173)
`ure 4). Two dimensional echocardiography revealed no tur(cid:173)
`bulence during systole, with maintenance of central flow.
`Color Doppler flow showed trace central regurgitation. No
`perivalvular leak could be identified. m-Mode echocardiog(cid:173)
`raphy showed full opening of the leaflets with complete ap(cid:173)
`position during diastole (Figure 5).
`
`Discussion
`
`Development of a percutaneously implantable aortic
`valve poses many challenging considerations. Two investiga(cid:173)
`tors have recently described aortic stent valves fur percuta(cid:173)
`neous aortic valve replacement: a ball-cage and a balloon
`expandable porcine valve. 7
`8 This study presents a new de(cid:173)
`•
`sign for a collapsible, sutureless bioprosthetic valve for trans(cid:173)
`luminal application. The acute study was undertaken to eval(cid:173)
`uate several features of this design.
`First, the stability of a sutureless stent in the aortic annulus
`needed to be determined. In all in vitro studies, the valve
`remained firmly anchored in the ring with no displacement.
`This was confirmed in the animal studies, with the exception
`of animal 2. In this animal, valve dislodgment occurred sec(cid:173)
`ondary to structural failure of the stent at the weld point. This
`was probably caused by excessive manipulation and wear of
`the stent during the multiple in vitro studies that were per(cid:173)
`formed. Previous prosthetic valves have also been plagued
`with this problem, and design changes have included making
`the strut from a single sheet of metal to eliminate weld
`10
`sites. 9
`•
`
`O,H
`
`0,4S
`
`0,46
`
`Time (secs)
`
`Figure 4. Representative in vivo aortic and ventricular pressure
`tracings. Mean aortic root pressure was 116/70 mmHg, with mainte(cid:173)
`nance of a diastolic gradient of 66 mmHg between the aorta and left
`ventricle.
`
`In Vivo Studies
`
`Table 2 demonstrates the size of the porcine annulus and
`the most appropriately sized valve that was available for im(cid:173)
`plantation. Engagement of the valve required approximately
`5 min. Cross-clamp time was approximately 20 min in ani(cid:173)
`mals 1 and 3 and 40 min in animal 2 because of inadvertent
`dislodgment of the aortic cannula. All animals were success(cid:173)
`fully weaned from CPB.
`Animal 1 maintained a systemic pressure of 100/45 (±10)
`mm Hg over a pPriod of 2 hr. lntraoperative echocardiogra(cid:173)
`phy revealed complete opening of two leaflets, with the third
`leaflet redundant and relatively immobile. All leaflets were
`fully apposed during diastole (Figure 3). The slent valve was
`in the subcoronary position without any obstruction to cor(cid:173)
`onary blood flow. Color Doppler studies showed mild regur(cid:173)
`gitation at the level of the non mobile posterior leaflet. Post(cid:173)
`mortem examination revealed complete fixation of the valve
`at the annulus without any displacement. Because of the
`small size of the annulus, one of the leaflets appeared redun(cid:173)
`dant, as had been observed during echocardiography.
`
`Figure 5. (A) m-Mode echocardiogram demonstrates full opening and closing of the leaflets. (B) Long axis view with trace regurgitation
`during diastole represented by the mosaic pattern at the center of the valve. No perivalvular leak is observed.
`
`Edwards Lifesciences Corporation, et al. Exhibit 1008, p. 4 of 5
`
`

`

`PERCUTANEOUS AORTIC VALVE
`
`M385
`
`Second, an effective watertight seal to prevent perivalvular
`leakage is of paramount importance. Our studies have dem(cid:173)
`onstrated that the stent valve mechanism effectively pre(cid:173)
`vents regurgitation when the valve is anchored properly. The
`dynamic nature of the aortic root during the cardiac cycle
`that causes the aorta to decrease in size 11 is probably another
`feature in favor of effective sEialing. We postulate that this
`dynamic interaction between the stent valve and aorta is the
`reason that minimal to trace regurgitation was observed in
`the animal studies, as opposed to the in vitro data, which
`I ti% re(cid:173)
`were collected in a rigid system and resulted in IO
`gurgitation.
`Finally, we have shown that when an appropriately sized
`valve is implanted in the subcoronary position, flow is unidi(cid:173)
`rectional, with no turbulence and minimal transvalvular gra(cid:173)
`dient and central flow. These factors are important to the
`design of a prosthetic: valve with good hernodynamic perfor(cid:173)
`mance.1?
`
`Conclusion
`
`We have developed a collapsible bioprosthetic aortic
`valve for implantation in the subcoronary position at the
`level of the aortic annulus. This report summarizes the pre·(cid:173)
`liminary evaluation of this stent valve in terms of secure an(cid:173)
`choring, prevention of regurgitation, and adequate flow
`properties. The study is limited in that long term analyses
`were not performed, and many questions regarding durabil(cid:173)
`ity of the stent and leaflets n~main open. In addition, chronic
`studies are needed to assess valve thrombogenicity, throm(cid:173)
`boernbolism, and absence of stent migration. This prelimi(cid:173)
`nary study establislws that c1 sut ureless prosthetic stent valve
`is functional in the aortic position, and is a step toward de(cid:173)
`velopment of a percutaneous approach in the foreseeable
`future.
`
`Acknowledgments
`The authors thank Carbomedics, Inc. (Austin, TX) for preparation
`of bovine pericardium and assistance with construction of the valves
`on the stents.
`
`References
`1. Parodi JC: Endovascular repair of abdominal aortic aneurysms
`and other arterial lesions./ Vase Surg 21: 549-5'>7, 1995.
`2. Fischman DL, Leon MB, Baim DS, et al: A randomized compari··
`son of coronary stent placement and balloon angioplasty in
`the treatment of coronary artery disease. N Engl J Med .B 1:
`496-501, 1994.
`3. Serruys PW, Jaegere P, Kiemeneij F, et al: A comparison of bal-·
`loon-expandable stent implantation with balloon angioplasty
`in patients with coronary arleiy disease. N fngl J Med 331:
`489-495, 1994.
`4. Chuter TAM, Green RM, Ouriel K, er al: Transfemoral cndovas·
`cular aortic graft placement. J Vase Surg 1 B: 18'5 ··· 197, 1 ')93.
`·,. Moore MA, !:lohachevsky IK, Cheung DT, et al: Stabiliz;ition of
`pericardia! tissue by dye-mediated photooxidation. / /iiomcc/
`Mater Res 28: 611 618, 1994.
`6. Rosenberg C, Phillips WM, Landis DL, Pierce WS: Dt>sign and
`evaluation of the Pennsylvania State University Mock Circu(cid:173)
`latory System. ASA/0 J 4(2): 41 -49, 1981.
`7. Pavcnik D, Wright KC, Wallace S: Development and initial PX(cid:173)
`perimental evaluation of a prosthetic aortic valvP for trans ..
`catheter placerncnt. l,adiology 183: 1.51-1.54, 1992.
`B. Andersen Hi<, Knudsen LL Hasenkam JM: Transluminal implan
`tation of artificial heart valves: description of a new cxpand(cid:173)
`;1ble aortic valve and initial results with implantation by cath
`eter technique in closed chest pigs. Eur I /earl/ 13: 704 .. 70B,
`1992.
`9. Bjork VO: The Bjork-Shiley tilting disc valve: past, pn•sent and
`future. Cardiac Surgery: Stare of the 1\rt Reviews 19137, I IB
`000.
`·10. Starek l"JK, Beiludet RL, I fail KV: The Medtronic-I !ail valve: de(cid:173)
`velopment and clinical Pxperience. Cardiac '>urgery.· Stale of
`the Art Reviews 19137, 223 000.
`11. Tindale WB, Black MM, Martin TR: In vitro ev,iluation of pros(cid:173)
`thetic heart valves: anomalies and limitations. Clin f'hys l'hys(cid:173)
`iol Meas 3: 115-130, 19132.
`12. Arorn KV: Choice of mechanical aortic prosthetic valve, in Em
`cry RW, Arom KV (eds), The Aortic Valve, Philadelphia, Han
`ley & Belfus, 1991, pp. 1 B7 201.
`
`Edwards Lifesciences Corporation, et al. Exhibit 1008, p. 5 of 5
`
`