European Heart Journal (1992) 13, 704—708
`
`Transluminal implantation of artificial heart valves.
`Description of a new expandable aortic valve and initial results
`with implantation by catheter technique in closed chest pigs
`
`H. R. ANDl5RSEN"‘, L. L. KNUl)SF.N* AND J. M. HASBNKAMT
`
`Departments of “Cardiology, l'Thoracic and Cardiovascular Surgery. and the Institute of Experimental Clinical
`Research, Skejby University Hospital. Aarhus, Denmark
`
`KEY WORDS: Expandable stent—valve, transluminal implantation, prosthetic heart valve, pigs.
`
`A new artificial aortic valve prosthesis was developedfor implantation by the transluminal catheter technique wit/tout
`tlzoracotomy or extracorporal circulation. The new heart valve was prepared by mounting a porcine aortic valve into an
`expandable stent. Before implantation, the stent—valve was mounted on a balloon catheter and compressed around the
`deflated balloon. The stent——valve mounted balloon catheter was then advanced retrogradely to _the ascending aorta or the
`aortic root in anaesthetized pigs. Implantation was performed by balloon inflation which expanded the stent valve to a
`diameter exceeding the internal diameter of the vessel -— thus ensuring a stablefixation against the vessel wall. A total of
`nine implantations were performed in seven 70 kg closed chest pigs. Sub- and supracoronary implantation was performed in
`two and three pigs, respectively, while implantation in both positions was done in two. Angiographic and ltaemodynamic
`evaluation after implantation revealed no significant stenosis { <16 mmHg) in any of the nine valves and trivial regurgi~
`tation in only two. Complications were associated with restriction of the coronary blood flow in three animals. This
`preliminary study indicates that artificial aortic valves can be implanted in closed chest animals by transluminal catheter
`technique.
`
`Introduction
`
`in l952. Hufnagel and Harvey” performed the first
`implantation ofa prosthetic heart valve in a patient with
`severe aortic regurgitation. The artificial valve was
`implanted in the descending thoracic aorta and prevented
`severe regurgitation to the left vcntriclem. This technique
`was used in small series of operations for aortic regurgi-
`tation“). The development of extracorporal circulation
`made it possible for Harken et al.
`to perform the first
`subcoronary implantation in I960”), and since then,
`implantation of prosthetic heart valves has been an open
`heart surgical procedure. If, however, implantation could
`be accomplished without
`thoracotomy,
`it would be
`attractive.
`In 1989. we constructed a new artificial heart valve
`designed for implantation by transluminal catheter tech-
`nique without
`thoracotomy and heart surgeryl°1_ This
`paper describes the new prosthetic heart valve,
`the
`implantation technique, and the initial preliminary results
`with implantation in the sub~ and supracoronary position
`in pigs.
`
`Material and methods
`
`A NEW CO.\1CI:‘l’T FOR IMPLANT/\TION or ARTIFICIAL HI~:AItT
`vAI.vI:s
`
`The idea was conceived ofmounting a foldablc biologi-
`cal cardiac valve inside a balloon expandable metalic
`C'tIrre.I]70ttd?nt‘€.
`ilenning Rud Andersen. MD. Department of Cardiology.
`Skejby University Hospital. Brendstrupgaardsvej. DK—8200 Aurhus N.
`Duntnnrk.
`
`0l9S-()68X/‘)2.’050'l04 -l- 05 $03.00/'0
`
`stent. Implantation of such a device (stcnt+valvc=the
`stcnt~valve), would enable implantation of artificial heart
`valves by the transluminal catheter, technique.
`The stem was constructed of two 0-55 mm surgical
`stainless steel wires (monofilamcnt), each folded in IS
`loops (Fig. 1(a)). Three of the loops were 14 mm high.
`designed to the commisures ofa porcine aortic valve. The
`remaining loops in the first wire and all the second wire
`loops were 8 mm high. Each folded wire was bent into a
`circle (diameter, 22 mm) which was closed end-to—cnd by
`soldering. The two circles were then stacked upon each
`other and intcrfixcd by Mcrsclinc 2-0 sutures (Fig. 1(a)).
`The foldable valve was a porcine aortic valve taken
`from a 90 kg slaughtered pig (mixed Danish Landrace
`and Yorkshire). The aortic valve was carefully dissected
`and cleaned manually to remove unwanted material. The
`diameter,
`thickness, and height of the cleaned valve
`annulus was 27 mm, l mm, and 2mm. respectively. The
`height 0!‘ the three cornmisural sites were 8 mm.
`The stent—valve was prepared by mounting the cleaned
`aortic valve inside the stent (Fig. l(b) and (c)). The aortic
`annulus, which included the three commissural sites, were
`fixed to the metalic stent by 45-50 Prolene 5-0 sutures.
`The external diameter of the stent—-valve was approxi-
`mately I2 mm when collapsed (Fig.
`l(d) and (c)), and
`32 mm when entirely expanded (Fig. l(l) and (g)). After
`the stent--—valve had been manually compressed on the
`carrier balloon catheter. the stiffness of the metal pre-
`vented it from uncoiling spontaneously (Fig. l(d) and (c)).
`After expansion.
`the stillness ol‘ the metal minimized
`spontaneous recoil, when the balloon was deflated (Fig.
`l(i) and (g)). However, a small recoil (< l0"/n diameter
`
`© 1992 The European Society of Cardiology
`
`Edwards Lifesciences Corporation, et al. Exhibit 1017,—
`
`Edwards Lifesciences Corporation, et al. Exhibit 1017, p. 1 of 5
`
`

`
`Implantation 0/'(1rIift'c'z'rI/heart valves
`
`705
`
`Figure I The stem was constructed with two 0-55 mm stainless steel wires Folded in IS loops (at). A three-leallet
`porcine aortic valve was mounted inside the stent and fixed to the metal by sutures to form the stcnt- -valve (h) and (c).
`Before implantation the stcnt—valvc was mounted on a deflated three-foiled balloon dilatalinn catheter (d). The
`diameter of the collapsed stent~—valve was 12 mm (d) and (e). Balloon inflation expanded the stent —valve to an
`external diameter of 32 mm (1) and (g). Each of the three balloons were mounted with two elastic blocks (indicated by
`arrows). to prevent migration ofthc stent—valvc from the middlcof the hallnnns.
`
`reduction) was often seen during balloon deflation. After
`preparation, the stcnt~valves were kept frozen (~20 °C)
`until
`implantation some days later. Since only acute
`studies were performed in this initial phase, the valves
`were neither sterilized. heparinized nor treated by drugs
`or chemical agents.
`The carrier balloon catheter used For implantation
`was a conventional No. 12 F three-foiled aortic valvulo-
`plasty balloon dilatation catheter (Schneider, Zurich,
`Switzerland). Each of the three balloons was 70 mm long
`and had a diameter of‘ IS mm. The total diameter of the
`three balloons were 31 mm when inflated. Two soft rubber
`blocks (3 mm high) were mounted on each of the three
`
`balloons, separated by a distance of18 mm (Fig. l(d) and
`(D). The blocks ensured the stent --valve's stable position in
`the middle of the balloons, and avoided migration during
`catheter advancement and balloon inflation. The carrier
`balloon catheter was mounted in a self-constructed
`No. 41 F flexible introducer sheath (external diameter
`l3-6 mm, internal diameter 125 mm, length 75 cm). The
`stentvvalve loaded carrier balloon catheter was retracted
`into the introducer sheath during intravascular introduc-
`tion and advancement to minimize friction against the
`vessel wall. A standard guidewire, 300 cm long and 0-9 mm
`in diameter, was used for conventional cathcter~over-
`guidewire advancement of the carrier balloon catheter.
`
`Edwards Lifesciences Corporation, et al. Exhibit 1017, p
`
`
`
`Edwards Lifesciences Corporation, et al. Exhibit 1017, p. 2 of 5
`
`

`
`706 H. R. Andersen et al.
`
`
`ANIMAL PREPARATION
`
`Seven pigs weighing 70 kg (mixed Danish Landracc and
`Yorkshire) were used for implantation. The animals
`were anaesthetized, endotrachcally intubated, and venti-
`lated artificially. Surface ECG and blood pressures
`were recorded on a Sirecust 961 Monitor (Siemens: AG,
`Erlangen. Germany) and on paper by a Mingograph-62
`ink-jet galvanometer
`recorder
`(Siemens, Stockholm,
`Sweden). A No. 9 F introducer sheath was placed in the
`right carotid artery after surgical exposure of the vessel.
`A No. 8 F pigtail catheter was advanced to the ascending
`aorta through the sheath, and used for pressure monitor-
`ing and for angiography. After stent—valve implantation,
`the pigtail catheter was exchanged with a No. 8 F multi-
`purpose coronary arteriography catheter with an open
`end—hole which was advanced retrogradely through the
`stcnt—valve, and used for pressure measurements. A No.
`12 F Foley balloon catheter (Riisch, Kernen, Germany)
`with a balloon diameter of 40 mm was introduced into a
`neck vein, and advanced to the pulmonary trunk guided
`by fluoroscopy. The balloon could be inflated to decrease
`blood flow through the lungs and consequently through
`the left ventricle. This was used to minimize the high
`blood velocities past the valve prosthesis, causing pulsa-
`tilc movements of the carrier balloon catheter during
`stent-valve implantation in the heart.
`Because the femoral arteries of 70-kg pigs are only
`3-4 mm in diameter, rctroperitoneal access to the abdomi-
`nal aorta was made through a midline laparotomy. The
`aorta was exposed over a distance of 6- 7 cm cranial to the
`renal arteries and cross-clamped proximally and distttlly.
`During temporary cross-clamping, a 4crn long incision
`enabled a 8—- I0 cm long vascular prosthesis (diameter
`20mm),
`to be sutured end-to-side to the aorta at an
`angle of 45°. The prosthesis was used to gain intra-
`vasculur access for the No. 41 F introducer sheath. The
`animals were given no antiplatelet or anticoagulant
`drugs, and after implantation, angiography and pressure
`measurements,
`the pigs were exsanguinated ‘under
`continuous anaesthesia.
`
`lMl’l.ANTA’I‘l()N or "run sTaNT—vAr.vn
`
`The guidcwirc was advanced retrogradely into the
`left ventricle under continuous fluoroscopy. and subse-
`quently,
`the introducer sheath was advanced over the
`guidewire into the descending thoracic aorta. The carrier
`balloon catheter was then pushed out from the sheath and
`advanced further around the aortic arch. For supracor-
`onary implantation, the stent~valve was positioned just
`beneath where the right brachiocephalic artery started.
`For subcoronary implantation the stent—valve was pos-
`itioncd in the aortic root/left ventricular outflow tract
`beneath the coronary arteries at the level of the native
`aortic valve. The position ofthe stent---valve was guided by
`transthoracic echocardiography and fluoroseopy in the
`initial four pigs. Due to inaccuracy of echocardiography
`(heavy echoes from the stem), the irnplantations .in the
`subsequent three pigs were guided by fluoroseopy and
`angiography. In these three experiments. ventriculogra-
`phy and aortography were recorded on videotape for
`
`immediate playback to guide the stent»-valve implan-
`tation. When the stcnt-—valve was placed in the right
`position, implantation (stent-valve expansion) was per-
`formed by balloon inflation (4 atmospheres in 15 s) which
`overdilated (overstretched) the vessel. The elastic recoil of
`the vessel secured fixation and minimized periprosthetic
`leakage. Subsequently, the deflated balloon catheter, the
`guidewire, and the sheath were withdrawn. Two pigs were
`exposed to double stent-—valve implantation with the first
`one implanted in the supracoronary position; the second
`stent—valve was advanced retrogradely through the first
`valve and implanted in the subcoronary position.
`
`MEASUREMENTS AFTER IMPLANTATION
`
`Pressure measurements were performed immediately
`after implantation. Measurements were obtained during
`slow withdrawal ofthe catheter from the left ventricle to
`the aortic arch distal to the stent—-valve. Afterwards, ven-
`triculography and aortography were obtained by contrast
`injection through the plgtail catheter (Fig. 2). Following
`exsanguination, the heart and the aorta were excised and
`gross pathological examination was performed.
`
`Results
`
`"rue IMPLANTATION YROCEDURB
`
`The introducer sheath was easily inserted through the
`vascular prosthesis and advanced to the thoracic aorta
`guided by fluoroscopy. Before balloon inflation (stent-—
`valve expansion) the carrier balloon was kept easily in a
`stable position in the blood-stream. When the carrier bal-
`loon was inflated, the blood-flow carried it approximately
`another 3 -4 mm distally (downstream). The pulmonary
`artery balloon was inflated in two pigs. both with sub-
`coronary implantation, because of pulsatile movements
`of the carrier balloon. The blood-flow could be totally
`obstructed for a short period (10-15 s,
`if needed), or
`decreased for a longer time (min). The latter method was
`used in two animals (Nos 5 and 6). Seven pigs had nine
`stcnt--valves implanted (four in the subcoronary and five
`in the supracoronary position; Table l).
`
`HAEMODYNAMICS AND ANGIOGRAPHY
`
`All the animals survived the initial post-implant period;
`and pressure measurement and angiography was ac-
`complished (Table l). None of the stent-valves caused
`severe stenosis and only trivial contrast regurgitation
`was seen in two pigs (Nos 1 and 3). Left ventricular end-
`diastolic pressures were unchanged after stent~—valve
`implantation in live out ofseven pigs, but increased in two
`(Nos 3 and 4) due to left ventricular failure caused by
`restriction ofthe coronary blood-flow.
`
`ANATOMICAL FINDINGS
`
`All nine prosthetic valves were undamaged by the
`implantation procedure, and the sutures kept the biologi-
`cal valves inside the stents stable. No haematoma, bleed-
`ing or aortic dissection was seen in any of the seven pigs.
`Four pigs, in which no mechanical complications were
`
`Edwards Lifesciences Corporation, et al. Exhibit 1017, p._
`
`Edwards Lifesciences Corporation, et al. Exhibit 1017, p. 3 of 5
`
`

`
`Implantation o_far11/icia/ heart valves
`
`707
`
`Figure 2 Angiography showing slcnt-valves (indicated by arrows) implanted in suhcoronary position (a) and
`supracoronary position (b). an: aorta. LV: left ventricle.
`
`Table I Hloadpremrure measurementsfollowing stem valve imp/anlalion
`
`
`
`.
`Egg
`'
`
`.
`Suagval
`
`Site ofimplantation
`
`Blood pressure (mmHg)
`
`Subcoronary
`
`M
`Supracoronary
`
`Left ventricle
`
`Above
`subcoronary
`stent~~valve
`
`Below
`supracoronary
`stentvvalvc
`
`Above
`supracoronary
`stent- valve
`
`100/60
`100/60
`100/0
`2-5
`1
`105/75
`—
`120/7
`2-5
`2
`—-
`60/35
`64/30
`0-25
`3
`40/ 17
`~
`—
`43/18
`0-25
`4
`-
`~
`134/97
`150/—5
`+
`1-5
`5
`103/58
`102/58
`100/60
`I10/0
`+
`l~
`2-5
`6
`100/57
`100/57
`-—
`115/6
`+
`2-0
`7
`
`
`+
`
`+
`
`+
`+
`
`+
`
`100/60
`105/75
`
`seen. fulfilled the study's protocol. In three animals, the
`coronary flow was restricted.
`With supracoronary implantation, all five stem-«valves
`were fixed in the ascending aorta; the aortic diameter was
`overstretchcd by 3-4 mm. Four of the five stcnt—valves
`were positioned more than 1-5 cm above the genuine
`valves; none of the valves obstructed the brachiocephalic
`artery. in one animal (No. 4), in which the stent—valvc was
`implanted 3 mm above the native aortic valve, the stent—
`valve was competent, thus leaving only a small volume
`of blood between the native valve and the stent--valve
`available for coronary flow in diastole. This caused ST
`elevation and pump failure. All
`the stent—valves were
`competent at inspection and without coagulated blood on
`the cusps. H owever, small thrombi were seen on the metal
`and sutures.
`With subcoronary implantation, all stcnt—valves were
`implanted at the level of the genuine valves which were
`completely compressed between the metal stent and the
`vessel/heart wall. Two of the four stcnt-—valves were
`
`implanted beneath the origin ofthe coronary arteries. The
`other two (Nos 3 and 5) restricted coronary flow. In pig
`No. 3. both coronary arteries were obstructed. and the pig
`died due to pump failure. In pig No. 5 the left coronary
`ostium was free of the stcrtt-valve, but the right coronary
`artery was partially obstructed. This animal was haemo~
`dynamically stable, but died suddenly from ventricular
`fibrillation 1-Sh after implantation. Mild regurgitation
`at aortography was found in two stent valves. This was
`caused by tightness of one of thc stent-valve cusps in one
`case (probably caused by the preparation); in the other
`there was a small paraprosthetic leak.
`
`Discussion
`
`IMPLANTATION or PROSTIIETIC HEART VALVES WITHOUT
`'I'll0RACOTOMY
`
`This paper presents the first description and prelimi~
`nary results of a new expandable artificial heart valve
`designed for pennanent implantation by transluminal
`
`Edwards Lifesciences Corporation, et al. Exhibit 1017,fl
`
`Edwards Lifesciences Corporation, et al. Exhibit 1017, p. 4 of 5
`
`

`
`708
`
`II. R. Anderson et al.
`
`
`
`thoracotomy. Catheter-
`technique without
`catheter
`mounted valves have previously been constructed for
`short-term treatment of acute aortic insulficiencylml.
`These devices were mounted on long catheter wires which
`extended out through the vessel wall. Their position in
`the bloodstream was secured by external fixation of the
`extending catheter wires to the skin. Consequently, such
`valves were not suitable for permanent implantation. In
`contrast, the new stent—valvc is fixed intravascularly at
`the site ofimplantation without stent material projecting
`into the bloodstream or penetrating the vessel wall, thus
`making it more suitable for permanent implantation.
`
`THE STENT—VALVE
`
`The present devices were self~constructed from avail-
`able materials. The ()-55 mm wire fulfilled the criteria
`of minimal spontaneous ‘uncoil’ and ‘recoil’ after com-
`pression and dilatation. If the metal loops were longer
`than 8 mm, it was much easier to dilate the stent, but this
`resulted in larger deformation by the opposing elastic
`recoil from the vessel and/or the heart. If the loops were
`smaller, the stent was too still‘ to be fully expanded by
`the balloon. We used biological valves, because they
`were easy to obtain and mount inside the stent. Other
`types of foldable valves may also prove suitable, c.g. the
`tricuspid polyurethane heart valve""""‘.
`
`'l'Hl-2 IMPLANTATION PROCEDURE
`
`The extrathoracic approach was mandatory. As femoral
`arteries are very small in pigs we chose the abdominal
`aortic route for catheterization. Obviously, the femoral
`route should be used in humans, preferentially by a percu-
`taneous approach, alternatively by artcriotomy. Implan-
`tation was easy in both the ascending and descending
`aortas“) where small movements of the carrier balloon
`catheter were not critical. However, with subcoronary
`implantation such movements proved to be a problem. A
`catheter which does not obstruct the blood—llow during
`stent—vvalve expansion could be the solution.
`
`LlMl'l‘ATl()NS or THE srunv AND FUTURE TECHNICAL
`DEVELOPMENTS
`
`This is a very preliminary technical study, and many
`important questions remains to be answered about the
`stent—valvc. Since only acute studies were performed the
`long-term durability of the valve is unknown. Questions
`
`regarding neointimalization. calcification, thrombogeni-
`city, and dislodgement during long-term follow-up
`should be addressed. There may be a risk ofdislodgement
`and distal migration due to long-term gradual dilation or
`even necrosis ofthe portion of the aorta where the valve is
`implanted. Furthermore, valvular or aortic pathology,
`such as calcium, vegetative debris, fibrosis, and abscess
`formation could prevent accurate fixation or become a
`source of embolization. Thus, many more complex and
`long~term animal studies must be performed before even
`speculation concerning clinical use is begun.
`
`POSSIBLE CLINICAL IMPLICATIONS
`
`This preliminary feasibility study cannot clarify the
`clinical applicability of the stent~valve. However, it might
`be a treatment for patients with aortic regurgitation who
`are not candidates for open heart surgery. Implantation
`of the stent—valvc in a supracoronary position may
`protect the left ventricle from severe regurgitation""l.
`
`This study was supported by a grant from the Danish Heart
`Foundation, Copenhagen, Denmark.
`
`References
`
`[2]
`
`[ll Hufnagel CA. Harvey WI’. The surgical correction of aortic
`regurgitation. Preliminary Report, Bull. Georgetown Univ."
`Med. Center 1953; 6: 34.
`l-lufnugcl CA. Basic concepts in the development of cardio-
`vascular prosthcses. Am J Surg I979; I37: 285-300.
`[3] Hufnngel CA. Harvey WP. Rabil Pl, McDermott TF. Surgical
`correction ofaortic insutllcicncy. Surgery l954; 35: 673-83.
`[4] Conklin WS, Grismer JT. Aalpocl JA. Hufnagel valve surgery
`for aortic insufiiciency. J Thoracic Surg l958; 36: 238-46.
`[5] Harken DIE, Sorofi HS, Taylor WJ, Lefemine AA. Gupta SK.
`Lunzcr 3. Partial and complete prosthesis in aorta insullicicncy.
`J Thorac Cardiovasc Surg I960; 40: 74442,
`[6] Andersen HR, Knudsen LL, Hasenkam JM. Trtmsluminal
`catheter implantation of a new expandable artificial cardiac
`valve (the stent—valvc) in the aorta and the beating heart of
`closed chest pigs (Abstract). Eur Heartl 1990; l
`l (Suppl): 224a.
`[7] Borctos JW, Polrer RA. Aortic heart valve catheter. United
`States Patent (No. 4.056.854), l977.
`[8] Moulopoulos S. Catheter mounted artificial heart valve for
`implanting in close proximity to a defective natural heart valve.
`United States Patent (No. 3.671.979), I972.
`[9] Lo HB, Herold M, Reul H er al. A tricuspid polyurethane heart
`valve as an alternative to mechanical prostheses or hiopros-
`theses. Trans Am Soc Artiflntem Organs 1988; 34: 839-44.
`[l0] Hilbert SL, Ferrans VJ, Tomila Y, Eidbo BE. Jones M.
`Evaluation of explanted polyurethane trileallet cardiac valve
`prosthesis. J Thorac Cardiovasc Surg I987; 94: 4 l9~29.
`
`Edwards Lifesciences Corporation, et al. Exhibit 1017, p—
`
`Edwards Lifesciences Corporation, et al. Exhibit 1017, p. 5 of 5