STRYKER EXHIBIT 1038, pg. 1
`
`STRYKER CORPORATION v. ORTHOPHOENIX, LLC
`
`IPR2014-01433
`
`

`
`Stephan Becker
`Michael Ogon (Eds.)
`
`Balloon Kyphoplasty
`
`Springer WienN ewYork
`
`STRYKER EXHIBIT 1038, pg. 2
`
`IPR2014-01433
`
`

`
`Dr. meeL Stephan Becker
`Univ.-Doz . Dr. med. M ichael Ogon
`W irbe lsau lenzentrum, 3. O rthopacl ische Abteilung, O"thop;idisches Spital Speising, Vienna, Austria
`
`This work is subject to copyright.
`All ri ghts are reserved, w hether the who le or part of the Illaterial is concerned, specifically those of translation, reprinting,
`re-use of illustrations, broadcasting, reproduction by photocopying machines or sim il ar Illeans, and storage in data banks.
`Product Liabil ity: The publisher can give no guarantee for all the information contai ned in this book. This does also refer to
`inrormation about drug dosage and application th ereof. In every individual case the respective user must check its accu(cid:173)
`racy by consulting other pharmaceutical literatu re, The use of registered n<lmes, trademarks, etc. in this publication does not
`imply, even in the absence of iI specific statement, that such names are exempt from the relevant protective laws and regu(cid:173)
`lation s and t.herefore free for general use.
`
`(\J 20 '10 Springer-Verlag/Wien
`Pri nted in Austria
`SpringerWienNewYork is a part or
`Springer Science + Business Media
`springe!'.at
`
`Printed on acid-free and chlorine-free bleached paper
`
`With 121 (partly coloured) Figures
`
`ISBN 978-3-211-99908-0
`
`e-ISBN 978-3-21 1-7422 1-1
`
`STRYKER EXHIBIT 1038, pg. 3
`
`IPR2014-01433
`
`

`
`Contents
`
`Contributors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
`Acknowledgement . . ...... . . ....... .. .. .... ...... ..... .... ... . ... . . .. . .... .. .... ... .. . .. . ....
`
`IX
`XI
`
`Chapter 1. Epidemiology of osteoporosis (S. Becker and M. Ogon) .... . .' . .. ... .. . ........ . ....... . .. . . .
`Incidence and prevalence of vertebral fractures. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . • . . . . . . . . . . .
`Sex differences . . . .... .. . . '. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . • . . . . . . . . . . . . • . . . . . . . . . . . . . . . . . . . . . .
`Socio-economic consequences ... . .. .... . .. . ... . .. . . .. .. . .. . . .. . . ... .. . •. .. : . ... . . . . .. . . ... . .. .
`
`Chapter 2. Drug therapy of osteo porosis (H. Resch an d C. Muschitz) .... . . . . . . . . .. ... .. ... . .....•.....•
`Pathogenic mechanism and pharmacologica l effects .................. . .. .. .... .... . . .. . ..... . ......•
`Therapeutic goals . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . • . . . . . . . . . . . • . . . . . . . . . . . . . . . . . . . . . . . . . .
`1) The treatment of pain .. ........ . . . .. . . .. . .. .. . . .. .... .. . .. . .. . . .. '.' . . . . . . . . . . . . . . . . . . . . . .
`2) Redu ction of the ri sk of fracture . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . • . . . . . . . .
`3) Increas e of bone density .. ... . . .. . ..................... . ...... ....... ..... . ...... . . •. . . . .
`4) Influencing biochemical markers of bone metabolism .... . . .... . . . ... . . . . . . . . . . . . . . . . . . . . . . . . . . .
`Indication for osteoporosis therapy ...... .. . .. ...... ...... ........ . . .. . ... .... . .•..... . .. . .......
`Treatment options .. ... . ..... ... ... .. ....... . . . .. . . ................•........ . . .. ... .. . . .. ....
`Basic medication w ith calcium-vitamin D ... ... . . . ...... .... .. .. .. .. . . .. .. .... . . . .. . . .. ... ... ..
`Calcium ... ... . . .... . ... ... ...... .. .. . . .. ..... . .. .... . . . . ..... . . . .... .. .. .. .. .. ..... .
`. ... ... .. ... ........ . . . . ..... . ... . .. ........ .. ......
`The combination of calcium w ith vitamin D
`Hormone repl acement therapy - a fundamental change in evalu ation ... . .. .. ..... . ........ .. ..... ... .
`Substances th at inhibit bone abso rption . ...... . ... ... . .... ..... . ...... ..... .. .. . . ... . . . . . . . . . . .
`Bisphosphonates ..... ..... ....... ..... ... .. .. .. ........ . . . ... .. . ........ . .... . ..........
`Selective estrogen -receptor modulators .... . . .... . ... .. ... . .... . . . . . . . ... ... ... . . . . . .. .. ..... . .
`Calcitonins . . .. .. ..... ... ...... .. ............... . ... . ........ ..... ... . . ..... . . ... . . . ... ...
`Tibolone ......... . .. . ...... ... . . . . .. ..... '.' . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
`Substances that in crease bone formation .. . .............. ... . . . .. .. ... .. . . . . .... ... . ... .. . . . . ..
`Parathormone . .. .. . . ...... .. ... .. .. .................. ....... ..... . . ... . .. .. .. . . . . . . . . .
`Combination therapies
`. ... . ......... .... ............. . . ........ ... . ... ... ...... . . ..... .....
`Parathormone and antiresorptives . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
`Fluorides . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
`Substances w ith a synchronous effect on bone formation and absorption ..... ... .... . ........ ... ... . ..
`Substances with a biological effect. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
`The RANKL antibody - new approaches in the treatment of osteoporosis ... .... . . ..... .. . ...... .. . .
`Osteoporosis in men
`. . ..... . . . ..... . .. .. ............. . . ... . .. ........ ... .........• .. •.. . .... .
`Prevention and therapy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
`
`Chapter 3. Clinical aspects and mortality risk of the osteoporotic spine fracture (S. Becker and M. Ogo n) . . . . . .
`. .. . ... .. .. .. .. .. .. ....... . . . ... . . . . .. .. .. ....... . ... . ...... ..
`Cli nical diagnosis of spinal fracture
`The clinical co nseq uences of a spinal fracture .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
`Mortality after fractures of the spine
`. .. . . . .. ..... . . .................... . ... . ..... ...... .. ... .... .
`
`. ........ .. . . . ..... . .. ... ............... ............• .. .... .. .. .... ..
`Chapter 4. Biom echanics
`Biomechanics of cement injection in vertebropl asty (G. Baroud and A. Schleyer) .. . . ... .... •. . . . .. .. . . . . . . .
`Summa ry ... . .. . .. . .... ... . .. .......... . .... .. ............ . . .......... .... . . . . . . . . . .. . . . .
`Introduction .... .. .. .. . . ... . . .. .... . ...... . .... . ..... . ......... . .. . . .. .. . ...... . . .. ..... .
`In vivo measurements of inj ection pressure/volume versus tim e for three represen tative cases of vertebropl asty
`. .
`. .. .. . ..... . . . . ....... .. .. . .... . . ........... ..
`Analysis of inj ecti on pressures during vertebropl asty
`Experim ental determination of the different pressure components during a cement injection ... ... . . . ... . . .
`Analysis of risk of ceme nt 'extravasation out of the vertebral body .... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
`
`1
`2
`2
`
`5
`5
`6
`6
`6
`6
`6
`7
`7
`7
`7
`8
`8
`9
`9
`10
`10
`10
`10
`10
`10
`11
`11
`11
`11
`11
`12
`12
`
`17
`17
`18
`19
`
`23
`23
`23
`23
`24
`25
`26
`27
`
`STRYKER EXHIBIT 1038, pg. 4
`
`IPR2014-01433
`
`

`
`Chapter 4
`
`Biomechanics
`
`Biomechanics of cement injection in
`vertebroplasty
`
`G . Baroud and A. Schleyer
`
`Summary
`
`Vertebroplasty is being increasingly used for con(cid:173)
`solidation of osteoporotic vertebrae or other patho(cid:173)
`logical findings; for example, in bone cancer. In this
`chapter we present a combination of theoretical
`considerations and in vivo and ex vivo studies on
`cement injection. The unexpected results reflect the
`fact th at approximately 95% of th e overall injection
`pressure is necessary for cement delivery through
`the cannula, and only approximately 5% for the
`dispersion of cement in the spongiosa. On e of our
`most important findings is that the process of ce(cid:173)
`ment injection makes conflicting demands on bon e
`cements, which are req uired to be more viscous
`and less viscous at the same time. A low viscosity
`
`eases cement delivery through the injection can(cid:173)
`nula, whereas a high viscosity reduces the risk of
`cement leakage out of th e vertebra. Th e challenge
`therefore is to develop biomaterials, techniques and/
`or devices that ca n overcome or manage the con(cid:173)
`flicting demands concerning cement viscosity.
`
`Introduction
`
`Vertebropl asty is a relatively new technique for th e
`treatment of vertebral fractures originating in osteo(cid:173)
`porosis or resulting from other pathological fin dings
`[Cotton et al. 1996; Deramond et al. 1998; Heini et
`al. 2000; Jensen et al. 1997; Mathis et al. 2001 J. In
`this procedure, bone cement is injected under pres(cid:173)
`sure through a cannula into the porous structure of
`the cancellous bone. Th e bone marrow is thereby
`displaced o ut of the cavities of the vertebra (Fig. 2).
`Th e in situ curing of th e cement in th ese cavities
`strengthens the weakened vertebra [H eini et al.
`2000, 2001; Jensen et al. 1997; Krause et al. 1982;
`
`Fig. 2. Pictures of a three-dimensional reconstru ction of trabeculae of the spongiosa of healthy (A) and osteoporotic bone
`(B). The bone is depicted in turquoise, the bone marrow in violet
`
`STRYKER EXHIBIT 1038, pg. 5
`
`IPR2014-01433
`
`

`
`24
`
`G. Baroud and A. Schleyer
`
`;'i
`
`Mathis et al. 2001; San Millan Ruiz et al. 1999; Wil(cid:173)
`son et al. 2000] .
`Until now it has not been possible to develop
`uniform standards for the cement-injection proce(cid:173)
`dure. Furthermore, there are no clear guidelines
`from which to choose the parameters necessary for
`en suring reproducibl e and safe injection with a pre(cid:173)
`dictable outcome. Because of this situation, the out(cid:173)
`come of cement injection is often unpredictable.
`In the following, we present various studies that
`have contributed to a clearer understanding and
`therefore to potential enhancements of the injection
`process.
`Initially, we analyzed the process of th e injection
`pressure and injection volume for successful in vivo
`cement injection, an
`insufficient cement injection
`(aborted injection because of too high an injection
`pressure) and a risky in vivo injection (aborted be(cid:173)
`cause of potential cement leakage out of the verte(cid:173)
`bra). Because the injection pressure seemed to play
`an important role in the outcome of an injection, it
`was further analyzed in a theoretical study. The in(cid:173)
`jection pressure was divided into an extravertebral
`component (delivery of cement through the cannula)
`and an intravertebral component (spreading of ce(cid:173)
`ment throughout the vertebral cavities). Following
`the theoretical analysis, the different pressu re com(cid:173)
`ponents were measured and evaluated in an ex vivo
`experiment. W e discovered that the major part of the
`injection pressure is needed for cement delivery.
`In addition to examining the injection pressure,
`we addressed the risk of cement leakage. Specifi(cid:173)
`cally, the role of cement rh eological properties in
`cement leakage was examined in both a theoretical
`and experimental manner.
`In the last section of this chapter, we present a
`newly developed injection cannula, which is ex(cid:173)
`pected to significantly reduce the injection pressure.
`In an additional study, the injection pressures neces(cid:173)
`sary for the new cannula were compared with those
`for a conventional cannula. For this, ex vivo experi(cid:173)
`ments were performed under simulated clinical
`conditions.
`
`In vivo .measurements of injection pressure!
`volume versus time for three representative
`cases of vertebroplasty
`
`In this section, the in vivo injection data (pressure(cid:173)
`versus-time and volume-versus-time) for the three
`possible outcomes (successful, insufficient, unsafe
`injection) are described. Details on technique and
`
`guidelines for patient selection are described in Hei(cid:173)
`ni et al. [2001].
`The biomaterial used in all three cases presented
`here was a low-viscosity acrylic cement (Palacos
`E-f1ow, Essex Chemie, Lucerne, Switzerland). Ten
`milliliters of cement was divided into two 5-cc stan(cid:173)
`dard syringes and then injected through a biopsy
`cannula (8 G, Somatex, Berlin, Germany). After the
`liquid was added to th e powder, there was a waiting
`period (elapsed time) of approximately 2 minutes,
`durin'g which time th e cement attain ed the appro(cid:173)
`priate consistency
`for safe
`injection. After this
`elapsed time, the cement had a handling time of
`about 3 minutes, during which th e procedure had
`to be completed. The injection data described be(cid:173)
`low (for successful, for insufficient and for risky de(cid:173)
`livery) are three cases chosen to represent in vivo
`pressure-injection data.
`injection device
`A custom-made sterilizable
`(Fig. 3) instrumented with force and displacement
`transducers was used to monitor the injection data
`and was calibrated using a universal material testing
`machine (Mini Bionix 856, MTS, Eden Prairie, Min(cid:173)
`nesota, USA). A 5-cc syringe was filled with cement
`and placed in the device. Th e pressure-versus-time
`and injection volume-versus-time curves of the in(cid:173)
`j ection were collected using a Palm Pilot.
`
`Fig. 3. Device for measuring injecti on pressure and injec(cid:173)
`tion volume. The device consists of a delivery tool, in w hich
`you can insert a 5-cc syringe, and a Palm Pilot
`
`STRYKER EXHIBIT 1038, pg. 6
`
`IPR2014-01433
`
`

`
`Biomechanics of cement inj ec tion in vertebropl asty
`
`25
`
`On the basis of the cement expansion and th e
`resista nce encountered during the first 20 seconds
`of the injection, the cl inician is often able to predict
`whether or not th e injection wi ll be successful. The
`injection in this case was considered to be success(cid:173)
`ful because the injection forces required were mod(cid:173)
`erate and the cement expansion (i nfil tration) was
`uniform.
`An injection pressure and volume-versus-time
`graph is shown in Fig. 4. In this case, two syringes
`were injected consecutively, with a time delay whe n
`the first syringe was replaced by the second. The
`first syringe was injected over a longer time peri od
`than the second syringe because the cement was
`initially too liquid and thus had to be delivered cau(cid:173)
`tiously. A total of approximately 8.4 cc was injected
`in strokes of approxi mately 0.4-1.0 cc. The pressure
`in response to th ese strokes varied substantially be(cid:173)
`cause of th e ongoing cement polymerization. Th e
`maximum pressure of the first syringe was approxi(cid:173)
`mately 0. 5 MPa, for the second one it increased to
`about 1.7 MPa.
`By tactile and visual feedback, the cl inician is
`often able to predict the outcome of th e treatment
`within th e first 20 seconds of an injection. Th e in(cid:173)
`jection is considered successful when th e required
`pressures are moderate and the cement filling in the
`spongiosa is uniform. An injection is considered
`insufficient whe n it is aborted at an early stage be(cid:173)
`cause of too high an injection pressure. Th e injec(cid:173)
`tion process in an unsafe injection is aborted at an
`ea rl y stage because the cement leaks; for example,
`
`through blood vessels and out of th e vertebra, thu s
`endangering the life of the pati ent.
`For th e purpose of clarity, on ly the pressure and
`volume progression of the successfu l injection are
`shown in Fig. 4. Afterwards, they are compared
`w ith the results of th e insufficient and the unsafe
`injection. Further resu lts are published in Baroud et
`al. [2004] .
`The pressure and volume progression of the in (cid:173)
`su fficient injection showed that, despite a cl earl y
`lower injection rate, the injection pressure increased
`over 2 MPa and therefore the injection had to be
`aborted. In the case of an unsafe injection, the pro(cid:173)
`gression curves were similar to those of the success(cid:173)
`ful
`injection, therefore presumably facto rs other
`than pressure are responsible for cement leakage
`out of the vertebra.
`In summary, it can be concluded that in most
`cases the injection pressure plays an important role
`in the outcome of vertebroplasty.
`
`Analysis of injection pressures during
`vertebroplasty
`
`For a clearer understanding of the pressure mecha(cid:173)
`nisms in vertebroplasty, we divided the overall pres(cid:173)
`sure, which is in equi libration with the injection
`pressure, into two components: (1) the overall extra(cid:173)
`vertebral injection pressure, which is necessary to
`overcome the friction between the cement and the
`cannula wall w hile delivering cement, and (2) the
`intravertebral pressu re. Equation (I) represents th e
`
`13:55:40
`
`13:56:20
`
`13:57:00
`Time
`
`13:57:40
`
`13:58:20
`
`Fig. 4. Injection pressure and -injection vo lume versus time for successful injection
`
`STRYKER EXHIBIT 1038, pg. 7
`
`IPR2014-01433
`
`

`
`26
`
`pressure required for the infiltration of cement into
`the cavities of the spongiosa and for the displace(cid:173)
`ment of bone marrow:
`P inj = Pextra + Pintra
`(I)
`where P inj = injection pressure, Pextra = extravertebral
`pressure, and P intra = intravertebral pressure. The
`intravertebral pressure can be further subdivided
`into (a) the pressure required to infiltrate the tra(cid:173)
`becular cavities with cement, and (b) the shell pres(cid:173)
`sure as hydrostatic resistance caused by the dis(cid:173)
`placement of the bone marrow out of the vertebra
`into the adjacent structure. The hydrostatic resis(cid:173)
`tance strongly depends on the trabecu lar structure
`and on the porosity of the vertebral shell. The re(cid:173)
`vised mathematical representation is as follows:
`P inj = Pextra + Pin! + Pkomp
`where Pin! == infiltration pressure and P komp = shell
`pressure. To analyze the different components of
`the injection pressure, we built a theoretical model.
`The compacta pressure is neglected in this model
`because it is very complex and there is no way to
`describe it mathematically. By means of this model,
`we were able to point out the relationship between
`the extravertebral pressure and the infiltration pres(cid:173)
`sure, as well as the significance of the physical
`properties of the cement and the other injection
`parameters for the injection pressure.
`The approach for this model was the equilibrium
`of the injection force, based on Pinj, and the forces
`evoked by the infiltration pressure Pin! and the extra(cid:173)
`vertebral pressure Pextra [Baroud et al. 2003]. To de(cid:173)
`scribe the infiltration pressure, we used th e Law of
`Darcy (infiltration of a fluid into a porous medium),
`and for the extravertebral pressure, the Law of Ha(cid:173)
`gen-Poisseuille (flow of a Newtonian fluid through
`a cylindrical tube):
`
`(II)
`
`(III)
`
`where rs = radius of the syringe, rk = radius of the
`cannula, y == shear rate, f1 = cement viscosity, and I
`= length of th e cannula. Because the cement infil(cid:173)
`trates the spongiosa uniformly in a successful filling,
`Darcy's Law can be integrated in spherical coordi(cid:173)
`nates. Assuming that cement flow in the cannula is
`laminar, the Law of Hagen-Poiseuille can be inte(cid:173)
`grated over the length of the cannula. Accordingly,
`we can write Eq. (III) for the injection pressure in the
`following way:
`.
`
`.. !
`!
`
`I
`.1 '
`
`==> P inj = fJ -.SL. . (2- - ~) + fJ 8Q I
`
`4rrK
`
`r k
`
`r
`
`Summand 1
`
`Summand 2
`
`nr~
`
`(IV)
`
`where f1 = cement viscosity, Q = flow rate, K = bone
`permeability, and r = radiu s of th e spreading cement
`cloud. The first term displ ays the infiltration pressure
`Pin!, the second term displays the extravertebral pres(cid:173)
`sure Pextra' Equation IV shows that the injection pres(cid:173)
`sure depends on a combination of geometrical (e.g.,
`length 'and radius of a cannula) and physical (e.g.,
`viscosity and flow rate) parameters,
`Using values taken from the References [Baroud
`et al. 2003a, 2004a, b; Krause et al. 1982; Nau(cid:173)
`mann et al. 1999J for fJ, Q and K, as well as the
`geometrical dimensions of an 8-gauge cannula (I =
`200 mm, rk = 2 mm) in Eq. IV, a very interesting
`and surprising result emerges: th e infiltration pres(cid:173)
`sure of an injection contributes 0% at the beginning
`and only 5.6% at the end to the overall pressure.
`Consequently, the extravertebral pressure, which is
`necessary to overcome the friction in the cannula,
`has to be considered as the limiting factor for ce(cid:173)
`ment injection in vertebroplasty, because it is ap(cid:173)
`proximately 95% of the required injection pressure.
`It is therefore clear that cement delivery through a
`cannula represents the bottleneck of the cement
`injection.
`
`Experimental determination of the different
`pressure components during a cement
`injection
`
`Using the results of the theoretical consideration of
`the
`injection forces, we measured the injection
`pressure and intravertebral pressure on the lateral
`shell of the vertebra during an injection process in
`an ex vivo experiment with cadaveric vertebrae.
`Our assumption, derived from the theoretical mod(cid:173)
`el, was that the injection pressure would be much
`higher than the intravertebral pressure.
`For the experiments, 15 lumbar vertebrae were
`harvested from three osteoporotic spines. The bone
`mineral density ranged from 0.136 to 0.620 g/cm2.
`The injection cannula with the connected syringe
`was placed in the vertebra so that it entered the
`right vertebral pedicle, and its end was placed as
`accurately as possible one third of the overall width
`from the right lateral side and one third of the over(cid:173)
`all length from the frontal side. To measure the in(cid:173)
`travertebral pressure at the left lateral shell, a pres(cid:173)
`sure sensor was connected to a vent in the com-
`
`STRYKER EXHIBIT 1038, pg. 8
`
`IPR2014-01433
`
`

`
`Biomechanics of cement injec tion in vertebroplasty
`
`27
`
`pacta of the vertebral body. Th e overall test arrange(cid:173)
`ment was in stalled in a servohydraulic-engine test
`bench, where th e integrated load cell measured the
`injection forces. The schematic test arrangement is
`shown in Fig. 5. Instead of cement, silicon oil with
`a comparable viscosity (100 Pa·s) was used, giving
`reproducible results because of the constant viscos(cid:173)
`ity. More exact information abou t test preparation
`and test arrangeme nts ca n be found in Baroud et al.
`[2005].
`Injection of the silicon oil was carri ed out under
`controlled kinematic conditions. Th e pressure ac(cid:173)
`quired by th e load cell that was needed to inject the
`silicon oil corresponds to the pressure that a physi(cid:173)
`cian has to apply manually to a syringe during a
`vertebropl asty procedure. The intravertebra l pres(cid:173)
`sure produced from the cement dispersion in th e
`vertebral body on the lateral shell was acq uired
`with the pressure transducer.
`After initiation of the injection process, the injec(cid:173)
`tion pressure quickly reached a relatively constant
`level of (344 ± 62) kPa and did not change signifi(cid:173)
`cantly during the remainder of the injection process.
`In contrast to the injection pressure, the intraverte-
`
`'.
`
`L0ad
`cell
`
`1
`Pressure '
`sensor; I
`
`•• ~
`
`.; ••••••• _.1
`
`Fig. 5. Schematic representation of the experimental setup
`for measuring intra- and extravertebral pressure
`
`bral pressure on the cortical shell increased sign ifi(cid:173)
`cantly, though the increase was very slight, w ith
`maximal values of (3 .54 ± 2.92) kPa . The hypothesis
`proposed at th e beginning of this study was af(cid:173)
`firmed very clearly, because the measured injection
`pressure was approximately 97 times higher than
`the intravertebral pressure.
`For a further control, silicon oil was injected into
`the air, and at the same time, the injection pressure
`was measured under the same conditions as for the
`ex vivo vertebral body experim ents. Compariso n of
`th e pressures in both test sequences resulted in non (cid:173)
`significant differences. This affirms the accuracy/va(cid:173)
`lidity of the assumption made at the beginning of
`the study.
`The conclusions drawn from this study are pre(cid:173)
`sented in th e following. The experiment clearly con(cid:173)
`firm s that th e largest amount of the injection pres(cid:173)
`sure is required for cement delivery through th e
`cannula; th e intravertebral pressure seems to be
`minimal. The problem of insufficient filling of a ver(cid:173)
`tebral body resulting from an injection pressure that
`is too high is not associated with an increase of the
`intraverteb ral pressure and is th erefore only explain(cid:173)
`able by a change in the required extravertebral pres(cid:173)
`su re, which has to be rega rded as the key factor of
`th e injection process. M ethods such as the opening
`of the shell for release of pressure or making a cav(cid:173)
`ity in th e spongiosa do not contribute significantly
`to reducti on in the risk of insufficient filling.
`One implication of th e findin gs of our study is
`that because th e shell pressure contributes very little
`to the overall injection pressure, th e shell pressure
`cannot be important for insufficient cement deliv(cid:173)
`ery. However, shell pressure appears to be a signifi(cid:173)
`cant component of intravertebral pressure, and
`therefore it is hypothesized that it may be importa nt
`in how the cement spreads in the vertebral body.
`
`Analysis of risk of cement extravasation out
`of the vertebral body
`
`In earlier experiments [Baroud et al. 2005; Heini et
`al. 2000; Jensen et al. 1997; M athi s et al. 2001] it
`became clear that cement leakage out of the verte(cid:173)
`bral body (for example, through blood vessels or a
`fracture line) is a freq uently occurring and serious
`problem in vertebroplasty, and can culminate in
`nerve damage, pulmonary embolism or even the
`death of the patient. In thi s paragraph we consider
`the influence of various factors on extravasation ri sk
`at theoretical and experimental levels.
`
`STRYKER EXHIBIT 1038, pg. 9
`
`IPR2014-01433
`
`

`
`28
`
`G. Baroud and A. Schleyer
`
`Because the cement generally chooses the path
`of least resistance, an "extravasation factor" was
`analytically calculated in the theoretical model. This
`factor describes the relation of the pressure neces(cid:173)
`sary for uniform dispersion of the cement and the
`pressure required for injecting the cement into the
`leakage path [Bohner et al. 2003], In addition to the
`geometrical factors, such as the diameter of the
`extravasation path or the porosity, which cannot be
`influenced by the physician and are not further ex(cid:173)
`plained here for this reason, the relation of cement
`viscosity f-lc to the viscosity of bone marrow f-lb plays
`an important role. From qualitative calculations, we
`conclude that a low ratio fJc/fJb significantly increas(cid:173)
`es the risk of extravasation, whereas increasing the
`ratio diminishes the risk. For a physician, this means
`using cement that is as highly viscous as possible.
`On examining the dependence of cement viscosity
`on time after mixing the powder with the monomer
`(f-lc increases with time) [Baroud et al. 2004aL it is
`possible to increase the cement viscosity while post(cid:173)
`poning the injection to a later point in time. Be(cid:173)
`cause of the shear-thinning properties of the ce(cid:173)
`ment, it would also be advantageous if one could
`inject the cement at a low flow rate, which means
`with a low injection pressure.
`Although cement with higher viscosity would re(cid:173)
`duce the risk of extravasation, there are certain con(cid:173)
`straints; for example, the trabeculae could break
`under the too high charge during the injection pro(cid:173)
`cess, or the delivery system could fail because of
`the high forces (failing of the syringe), so that there
`is no longer an optimal connection between bone
`and cement.
`In experiments with a leakage model, we have
`shown that a higher ratio of cement viscosity to
`bone marrow viscosity and a delayed injection point
`
`reduce the risk of cement leakage. The leakage
`model consisted of a porous ceramic filter or alu(cid:173)
`minium foam with porosity similar to that of the
`spongiosa. The leakage path was simulated by a
`cylindrical drilling in the test specimen. With the aid
`of a materials-testing machine, cement was injected
`through a cannula into the probe. The behavior of
`dispersion at different moments of the injection af(cid:173)
`ter mixing the ingredients became apparent after
`the injection by means of x-ray images (Fig. 6) of the
`models and affirmed the theoretical perceptions.
`The following problems arose from the results for
`the injection process. The time scope in which the
`polymerization process allows an injection is rela(cid:173)
`tively small when injecting cement with a high vis(cid:173)
`cosity. This requires a high flow rate to fill the ver(cid:173)
`tebral body sufficiently and with this, a high injec(cid:173)
`tion pressure (Eq. IV). In addition, the injection pres(cid:173)
`sure increases because of the increased cement
`viscosity. If the injection exceeds the forces a hu(cid:173)
`man being can apply, the injection will have to be
`the preferred
`aborted prematurely. Furthermore,
`goal is to have low injection pressures to reduce the
`extravasation risk as a result of the shear-thinning
`properties of the cement. The requirements of the
`injection process for uniform infiltration (high ce(cid:173)
`ment viscosity, low pressure) and for sufficient filling
`of the vertebral bodies are thus exactly contrary.
`
`Development of a new injection cannula
`
`On the basis of the results of the preceding studies
`(the extravertebral pressure represents 95% of the
`overall injection pressure and is the key factor in the
`injection process; high viscosity and low pressure
`are necessary for a uniform dispersion of cement in
`the spongiosa; low viscosity and high pressure are
`
`DlarneterH
`
`A
`
`B
`
`c
`
`Fig. 6. X-ray pictures of the cement-filling pattern of strong cement leakage (A, low-viscous cement), moderate cement leak(cid:173)
`age (B, mid-viscous cement), and no cement leakage (e, high-viscous cement). The graph on the right side depicts a digita(cid:173)
`lization that was made to keep the numeric values for the amount of cement leakage
`
`STRYKER EXHIBIT 1038, pg. 10
`
`IPR2014-01433
`
`

`
`Biomechanics of cement injection in vertebroplasty
`
`29
`
`req uired for a sufficient filling of the vertebral bodYL
`we developed a new injection cannula. Th e goa l of
`this development was to achieve significa nt diminu(cid:173)
`ti on of th e extravertebral pressure.
`Th e new injection can nu la co nsists of two parts
`w ith different inner and outer diameters (Fig. 7). The
`distal third of the cannula has th e same dimensions
`as a conventiona l 8-gauge cannula (inner diameter,
`3.38 mm), because this part is introduced through
`the pediculus arcus vertebrae
`into the vertebral
`body and thus has to be adapted to th ese anatomi(cid:173)
`ca l conditions. The inner diameter of the proximal
`part, w hich partially penetrates the soft tissue, is
`6.92 mm, which is nearly double the size of the
`dista l third. Th e overa ll length of the new ca nnula
`is 135 mm.
`A theoretical model, which is based on th e law
`of H agen-Poiseuille and in which th e anthropome(cid:173)
`try of the human body was incorpo rated, was estab(cid:173)
`lished to find the ideal dimensions for th e cannula.
`Th e result, proven in an earlier experimental study,
`is that by doubling th e inner diameter of th e proxi(cid:173)
`mal part of the cannula the extravertebral pressure
`is decreased by 63% compared w ith a conventional
`8-gauge cannula [Baroud and Steffe n 2005].
`In a furth er study, th e newly developed can nula
`was tested under simulated clini cal conditions. In(cid:173)
`jectio n pressures were measured w hil e a surgeon in
`
`spina l orthopedics injected bone cement into ca(cid:173)
`daveric vertebrae through a conventional ca nnula
`and through the newly developed cannu la under
`the same injection cond itions (e.g., flow and pres~
`su re) as