`
`' LUME 81-3 1999
`
`N 0301-620X
`
`NUMBER FIVE (SEPTEMBER)
`
`”UL
`
`HEALTH SCIENCES LIMA“
`University oi Wisconsin
`
`592111999
`
`1305 Lindon Drive
`Madilon. WI 53708
`
`THEZ!OURNAL OF
`BONE AND JOINT
`
`SURGERY
`
`
`
`' UBLISHED JOINTLY IN LONDON ENGLAND AND BOSTON UNITED STATES OF AMERICA
`
`Page 1 of 1 1
`
`ZIMMER EXHIBIT 1011
`
`
`
`
`
`Page 1 of 11
`
`ZIMMER EXHIBIT 1011
`
`

`

`"In-l IIII III-II“. Il-‘I‘IHI'II
`
`Uninniiy a! win-nun
`
`SFP Pi 1993
`
`‘S'Si‘.éi”ii? its.
`*- THE JOURNAL OF
`s - e
`s BONE AND JOINT SURGERY
`o
`‘5
`’3
`u’
`'T
`6:,
`Q
`
`Volume 81-B. Number Five, September 1999
`
`
`Editorials
`
`The development of the Journal
`David Hambton and Frank Huron
`Infection in the operating room
`S. P. F. Hughes and F. M. Anderson
`
`Review article
`
`The dignosis and treatment of injury to the tarsometatarsal
`joint complex
`M. S. Myerson
`
`753
`
`‘154
`
`T56
`
`. ElflL__—___m———b
`764
`Pelvic obliquity after fusion of the spine in Duchenne
`821
`Detailed analysis of proprioception in normal and
`muscular dystrophy
`ACL-deficient knees
`G. Pap. A. Machner. W. Nebei'ong and F. Awiszns
`B. A. Alma and H. K. W. Kim
`
`Correction of genu recurvatum by the Ilizamv method
`i‘. H. Chef. C.
`if Chang, T-J. Cho and S. S. Park
`
`'
`
`Arthroscopic reconstruction of the anterior cruciate
`ligament with patellar-tendon autograft and interference
`screw fixation
`N. M. Joinha. L. A. Piiiczewski‘. A. Ctinget‘efer and
`D. D. Otto
`
`769
`
`775
`
`The placement of lumbar pedicle screws using
`computerised stereotactit: guidance
`E P. Girordi. F. P. Commie-a Jr. H. S. Soiidhu and
`L Aivm'ez
`
`The appearance on MRI of vertebrae in acute
`compression of the spinal cord due to metastases
`E M. Khan: 5'. A. Worthy. M. J. Gibson and A. Ghoikar
`
`825
`
`830
`
`780 Hi
`L_______—
`
`835
`
`843
`
`Poor eight-year survival of cemented
`zirconia-polyethylene total hip replacements
`.
`.
`.
`.
`JEA
`.
`.
`tD.
`dM.C.v
`Ham 5 Le Mme
`Gawain” a"
`0m"
`Nerve palsy after leg lengthening in total replacement
`arthroplasty for developmental dysplasia of the hip
`.
`’.
`.
`.
`.
`"it
`5 Egg“ 5 Ha”k"'”"’wa”dM E M" 8"
`
`,
`_
`Paedlamc
`The use or ultrasound in determining initiation
`of treatment in instability of the hip in neonates
`K. J. Helen. A. Tegiiander. S. H. Eik-Nes and TI Terjesen
`.
`.
`.
`.
`.
`.
`.
`Fmancual Justification for routine ultrasound screening
`ol’ the neonatal hip
`J. Ciegg, C. E. Boone and M if Ram
`
`846
`
`..,
`85-
`
`The treatment of congenital club foot by operation to
`correct deformity and achieve dynamic muscle balance
`{-12 Huang.
`iii Let. L. Zhao and J. Wang
`
`858
`
`Upper limb
`
`Shock-wave therapy is effective for chronic calcifying
`tendinitis of the shoulder
`
`363
`
`Osteoid osteoma
`
`314
`
`M. Componocc‘i'. P. Ruggieri. A. Gosbom'ni. A. Ferraro
`rind L Competing-ct
`
`M. Loew. W Doecke. D. Kosm‘emzaic M. Rohmonzadeh
`and it Ewerbeck
`
`Page 2 ofll
`
`Osteotomy of the tibia for correction of complex
`-
`deformity
`N 5- 51'5“!“
`.
`.
`.
`.
`Does arthritis progress in the retained compartments
`after ‘Oxford’ medial unicompartmental arthroplasty?
`A- E- Weaie- D. W “WM?- R- CWWOfd. V P-‘TChfli'iosx
`A. Bonomo. G. Hon-elf. J. O‘Connor and J. W. Goodfeiiow
`The length of the patellar tendon after
`790
`unicompartrnental and total knee replacement
`A. E. Weat'e. D. W. Murray J. H. Newman and
`C 5 mm)“
`
`“mun”
`Osteosarcoma of the pelvis
`R. J,- Grimei: S. R. can“: R. M. mm“. D. spoon“
`D. C. Maugham and K Kabukcoogin
`f
`tu
`Fr
`e distal
`osthetic reconstruction or moors o
`tibia and fibula
`.
`‘
`5.
`.
`.
`.
`.
`'
`,
`.—B. P ',
`.—‘K Rh'
`H Lee H -S Kim Y
`art T
`12 ondH K Lee
`Reconstruction and limb salvage after resection for
`'
`malignant bone tumour of the proxnmal humerus
`I. Wade, M. Usai'. K. Isa. 5. Yamowaki'i’ and S. (shit
`
`783
`
`796
`
`803
`
`808
`
`1‘ th
`
`Page 2 of 11
`
`

`

`
`
`Work practices and histopathological changes in the
`twosynovium and Horror retinaculurn in carpal tunnel
`syndrome in women
`J, Che”, A. Stevens and 7: R. C. Davis
`
`The relationship between the site of nonunion of the
`.mphoid and scaphoid nonunion advanced collapse
`{sure}
`H, Moritomo. K. Tada. I Yosnidn and T. Masaromr'
`
`CONTENTS
`
`868
`
`The regeneration of sensory neurones in reconstruction
`of the anterior cruciate ligament
`M. Ochf. J. Mara. Y. Uchio, N. Adacht’ and if Suntan
`
`Characteristics of bone ingmwth and interface
`871 mechanics of a new porous tantalum hiornaterial
`J. D. Robyn, G. J. Stackpool. S. A. Hacking. M. antzer
`and J. J. Krygr‘er
`
`per-cutaneous repair of the ruptured tendo Achillis
`J. M. Webb and G. C. Bannister
`
`377
`
`83]
`
`
`General
`
`'Inn-aoperative bacterial contamination in operations
`for joint replacement
`N. Davis. A. Curry, A. K. Garnbht’t: H. Panr‘grahi.
`C. R. C. Walker; E. G. L. Wilkins, M. A. Wot-slay and
`P. R. Kay
`
`886
`
`The connective-tissue envelope in revasculariaation
`of pateliar tendon grafts
`A. Scker'l, M. Leum'g. C. R. Fraitzt'. R. Ganz and
`F. I Bailmer
`
`c-Myc protein in the rabbit growth plate
`T. Aizawa, 5'. Kokuburt. TI Kuwnmatn.
`l’. Tannka and
`H. I. Roach
`
`Extensor carpi radiaiis brevis
`B. Greenbatrm. J. Marmara, C. T. Vnngsnesr. J. Tibone
`
`and R. Atkinson
`
`Obituary
`Gumec. du Toit 1909 -— 1999
`
`902
`
`907
`
`915
`
`921
`
`926
`
`930
`
`932
`934
`934
`
`935
`
`936
`
`938
`
`939
`
`
`
`
`
`
`foot and ankle
`
`
`'Total dislocations of the navicular: are they ever
`isolated injuries?
`
`M. S. Dht‘llon and 0. N. Nagt'
`
`
`
`
`
`Pyodenna gangrenosnm
`
`P. M. Annstmng. I. Hyas. R. Pandefi A. R. Bemndt,
`
`C. F! Carrion and A. H. R.
`lit Simpson
`
`
`"Muscle recovery after irnrnohilisation by external
`fixation
`N. Zarzhevsky. R. Coleman. G. Voipr'n. D. Fuchs.
`
`H. Stein and A. Z Reznt'ck
`
`Vascular complications of osteotomiee in
`ilinib reconstruction
`M. Rickntnn, M. Saieh. P. A. Gaines and K. Eyres
`
`arch
`
`890 Correspondence
`The tourniquet in total knee arthroplasty
`Acute fractures of the scaphoid
`Salvage of the head of the radius after
`fracture-dislocation of the elbow
`
`893
`
`896
`
`The pathology of bone allograit
`
`Book reviews
`
`Correction
`
`Forthcoming meetings
`
`h'untents, appliances, drugs and preparation
`Allhougl: it
`is intended than all advertising material accepted conforms to ethical medical standards. acceptance that not imply endorsement by the Journal. The Journal
`es. the light In refuse advertising material.
`
`
`
`'l’ilflriodicals postage paid at Railway, New Jersey. Postmaster Send address emotions to the Joumal of Bone and Joint Surgery. cto Mercury Airfreighr international
`ited. 365 Blair Road. Avenel. NJ 0700] (US mailing agent).
`Journal of Bone and Joint Surgery. ISSN MOI-620x British Volume. is produced at 21 Buckingham Sheet. London WC2N SET. It is published six times a year by
`British Editorial Society at Bone and Joint Surgery in January. March. May. July, September and November. Subscription rate $67 (£42). Printed in England.
`selling. scanning, printing and binding by Unwin Brothers Ltd. Old Wetting. a member of Martins Printing Group.
`
`ii
`
`Page 3 ofll
`
`Page 3 of 11
`
`

`

`Characteristics of bone ingrowth and
`interface mechanics of a new porous
`tantalum biomaterial
`
`J. D. Bobyn, G. J. Stackpool, S. A. Hackino. M. Tanzer, J. J. Krygicr
`From the Montreal General Hospital and McGt'll University. Montreal. Canada
`
`
`
`We have studied the characteristics of bone
`ingth of a new porous tantalum biomaterial
`in a simple transcortical canine model using
`cylindrical implants 5x10 mm in size. The material
`was 75% to 80% porous by volume and had a
`repeating arrangement of slender interconnecting
`struts which formed a regular array of
`dodecahedron-shaped pores. We performed
`histological studies on two types of material, one with
`a smaller pore size averaging 430 pm at 4, 16 and 52
`weeks and the other with a larger pore size averaging
`650 pm at 2, 3. 4, 16 and 52 weeks. Mechanical
`push-out tests at 4 and 16 weeks were used to assess
`the shear strength of the bone-implant interface on
`implants of the smaller pore size.
`The extent of filling of the pores of the tantalum
`material with new bone increased from 13% at two
`weeks to between 42% and 53% at four weeks. By 16
`and 52 weeks the average extent of bone ingrowth
`ranged from 63% to 80%. The tissue response to the
`small and large pore sizes was similar, with regions of
`contact between bone and implant increasing with
`time and with evidence of Haversian remodelling
`within the prom at later periods. Mechanical tests at
`four weeks indicated a minimum shear fixation
`strength of 18.5 MPa. substantially higher than has
`been obtained with other porous materials with less
`volumetric porosity.
`
`
`
`J. D. Bobyn. PhD. Associate Professor
`Departments of Surgery and Biomedical Engineering. Montreal General
`Hospital. Rm All—156. 1650 Cedar avenue. Montreal. Quebec. Canada
`H30 lArt.
`
`G. .t. Stackpool. MD
`S. A. Hacking, MEng
`.l. J. Krygier. CE’T
`Jo Miller Orthopaedic Research Laboratory. Montreal General Hospital
`Research institute, RMLs 141-09. lost} Cedar Avenue. Montréal. Quebec.
`Canada H3G M4.
`
`M. Tanzer. MD. FRCS C. Clinical Associate Professor
`Division of Orthopaedic Surgery. Montreal General Hospital, Rm A2444,
`[650 Cedar lavenue. Montreal. Quebec. Canada H36 lA-i.
`
`Conespondence should be sent to Dr J. D. Bobyn.
`@999 British Editorial Society of Bone and Joint Surgery
`0301—620X399359233 $3.00
`
`VOL. SIvB. No. 5. SEPTEMBER 1999
`
`This porous tantalum biomaterial has desirable
`characteristics for bone ingmwth; further studies are
`warranted to ascertain its potential for clinical
`reconstructive orthopaedics.
`J Brine Joint 5mg {31'} |W9;Rl 43190114.
`Received 9 July .3998; Accepted after revision 5 Marti! £999
`
`In the last 20 years a variety of porous surfaces and
`materials has been used to obtain fixation of bone ingrown-t
`in total hip and knee prosdteses. The most common include
`titanium and cobalt-chrome-alloy sitttered beads. diffusion-
`bonded titanium. fibre metal, and titanium plasma spray.t'3
`Based on the clinical outcome and histological evidence
`from retrieved implants.
`it
`is clear that porous surfaces
`support
`tissue ingrowth or ongrowtli and are generally
`effective for supplementing the stability of the implant by
`biological fisation?‘15
`Nevertheless. conventional porous materials each have
`certain deficiencies or weaknesses. For instance. sintered
`beaded and fibre metal coatings have a porosity which is
`limited to 30% to 50% by volume. a factor which directly
`limits the maximum interfacial strength that can develop by
`bone ingrowtb. Conventional metallic porous materials are
`best suited for use as coatings on implants since they do not
`readily have the required mechanical and processing char—
`acteristics which would allow them to be used as bulk
`structural materials for implants. bone augmentation, or
`substitutes for bone graft.
`A new porous biomatcrial made of tantalum has recently
`been developed for potential application in reconstructive
`orthopaedics and other surgical disciplinesm‘ '8 The materi-
`al has an unusually high and interconnecting porosity with
`a very regular pore shape and size. It can be made into
`complex shapes and used either as a bulk implant or as a
`surface coating. Our aim in this study was to characterise
`this porous tantalum material in terms of the extent and rate
`of bone ingrown: as well as the strength of fixation at the
`interface.
`
`Materials and Methods
`
`Implants. The manufacture of the porous tantalum begins
`with the pyrolysis of a thermosening polymer foam pre-
`90'?
`
`Page 4 ofll
`
`Page 4 of 11
`
`

`

`908
`
`I. D. BDBYN. G. J. STACKPOOL‘ S. A. HACKING. M. TANZER. J. .I. KRYGI'ER
`
`
`Figure la — Scanning electron micrograph of porous tantalum showing the cellular
`structure fonned by the tantalum struts. There is the occasional smaller opening or portal
`which interconnects with the larger pores or cells. Figure lb — Higher power scanning
`electron micrograph of a single pore illustrating the surface microtexture on the struts
`caused by crystal growth during the
`process of tantalum deposition. Figure 1c -
`Photographs showing nanscortical implants with small and large pore sizes.
`
`
`
`cursor to obtain a low—density vitreous carbon skeleton
`which has a repeating dodecahedron array of pores inter-
`connected by smaller openings or portals. Commercially
`pure tantalum is deposited into and about the carbon skele-
`ton using chemical vapour depositionlinfiltration (CVD:f
`CVI) to create a porous meta] construct. Because of the
`crystallographic growth and orientation of the tantalum
`during deposition. the process results in a surface with a
`distinct microtexture as shown in Figure 1. Thin CVDICVI
`films of 10 to 100 am can impart very high mechanical
`properties because the deposit
`is typically [00% dense.
`with grain sizes of less than 1 to 5 pm and impurities of
`<0.05%. The typical thickness of the tantalum coating is
`approximately 50 pm. An increase in thickness of the
`tantalum deposition can affect the pore size and mechanical
`properties.
`implants 5mm in diameter and 10mm in
`Cylindrical
`length were made from porous tantalum [Implex Corpora-
`
`tion, Allendale. New Jersey} for use in a canine model as a
`transcortical implant (Fig. 1). They had two different pore
`sizes. The mean pore size was determined using two tech-
`niques in each of which the data were expressed as a mean
`with 95% confidence intervals (CI). The first
`involved
`measuring the diameter of 100 dodecahedron-shaped pores,
`excluding the smaller connecting portals, based on second-
`ary electron SEM images of the two pore sizes. This
`method gave a mean pore diameter of 547 pm (95% CI 537
`to 557) for implants of smaller pore size and 710 pm (95%
`CI 696 to 724} for those of the larger. The second technique
`used the line intercept method based on the digital meas—
`urement of [000
`re openings on SEM images of ten
`different
`implants. 9 It
`included all pore lengths inter-
`cepting the grid lines, not just the full diameters of the
`pores, and therefore gave smaller mean pore sizes with
`larger (315. By this method the mean pore size was found to
`be 430 pm {95% Cl 413 to 447} in implants with the small
`THE JOURNAL OF BONE AND JOINT SURGERY
`
`Page 5 ofll
`
`Page 5 of 11
`
`

`

`L'HARAC'TERESTICS OF BONE TNGRDWTH AND INTER FAt'l-i MECHANICS OF A NEW POROUS TANTALUM BIUMATERIAL
`
`“09
`
`pore size and 650 pm {95% CI 629 to 679i in those with
`the larger size. The volume porosity of all implants ranged
`from T59? to 30%.
`
`Experimental study. We used skeletally mature mongrel
`dogs weighing betwean 25 and 35 kg. Under standard
`aseptic surgical techniques a lateral approach was made to
`the femoral diaphysis. A specially designed C~shaped
`adjustable jig was clamped to the femur to ensure accurate
`and reproducible alignment and sizing of perpendicular
`drill holes.‘nBushings which fitted into theJig were used to
`guide pilot (2.5 mm) and final (4.95 mm] drills All
`the
`implant sites were prepared using slow drill speeds and
`copious saline irrigation to minimise mechanical and ther—
`mal
`trauma to cortical bone. The final drill hole gave a
`slight press—fit of the transcortical
`implants which were
`inserted into the holes with 2 to 3 mm left proud of the
`periost'eal surface. Four implants were inserted into each
`femur of each animal. After surgery the animals were
`allowed to recover without limitation of diet or activity.
`Implants with a large pore size were the first
`to be
`manufactured and supplied for evaluation. This precluded a
`direct comparison of the two pore sizes in the same ani—
`mals. Three animals with implants of large pore size [24
`implants) were studied at each period of 2. 3. 4. 16 and 53
`weeks and three (24 implants} with implants of small pore
`size at 4. 16 and 52 weeks. At the end of each study. the
`femora were removed and prepared for either undecalcil'ied
`thin section histology or mechanical testing.
`Histological examination. Histological preparation includ-
`ed dehydrating. defatting. and embedding in acrylic'2"
`The implants were sectioned either longitudinally through
`the centre or transversely. All sections were polished and
`sputter—coated for analysis by backseattered SEM. The
`longitudinal sections were created as close to the centre of
`the implant as possible because only central sections gave a
`true impression of the extent of penetration of the ingrowth
`of new bone across the full diameter of the implant. We
`used computerised image analysis on the longitudinal sec—
`tions to determine the extent of bone ingrowth. which was
`
`defined as the percentage of available porosity filled with
`new bone. Some sections were subsequently mounted on
`slides. ground. polished. and stained with paragon for
`qualitative examination by transmitted light microscopy.
`Mechanical testing. We performed mechanical tasting only
`on implants of small pore size. Twelve specimens from the
`four-week and six from the [6-week periods had a push-out
`test. Femora from each of the three animals in each group
`were randomly selected for mechanical testing. They were
`stripped of soft tissue and the proximal and distal femoral
`metaphyses excised. thus leaving the diaphyseal segment
`containing the implants. This was divided longitudinally
`into two halves with a band saw to give access to the
`
`implant inside the intramedullary canal. Each implant was
`pushed out of the femur with a cylindrical plunger of
`4.8 mm diameter attached to the cross-head of a servo-
`
`hydraulic Instron test apparatus (Canton. Massachusetts).
`VOL. RIAB. No. 3. BEP'TEMBER "399
`
`Page 6 ofll
`
`'
`Conical Bone fl
`
`Instron Grosshead
`I
`,—I
`mt? 3"“
`
`Fig. 2
`
`Diagram of the test arrangement used for the mechanical
`push-out tests.
`
`into contact and
`The end of the plunger was brought
`aligned with the intramedullary end of the implant. The
`bone was supported in a metal cradle with the implant
`centred over a hole 5.5 mm in diameter to provide room for
`push—out (Fig. 2]. Gaps in the curvature between the femora
`and the metal cradle were filled with cold-curing dental
`acrylic cement to give uniform support during mechanical
`testing. The cradle Was mounted to a support on a universal
`joint which allowed alignment of the plunger with the
`implant. Load was applied to the implant at a cross——head
`speed of U05 cmfmin. a rate which was selected based on
`previous studies. Load was applied until the bone-implant
`interface. ruptured or there was compressive collapse of the
`implant as defined from the peak on the load~deformation
`curve.
`
`Following the design of a previous study.I after comple-
`tion of each push-out test. the remaining femur was divided
`into halves through the implant site to allow measurement
`of the cortical thickness. This was done at four equidistant
`locations to obtain a mean thickness which was used for
`calculation of the bone interfacial area. The shear strength
`of fixation was calculated by dividing the peak load by this
`area. In instances when the implant underwent compressive
`failure before rupture of the bonc~implant interface. a value
`for shear strength could not be calculated but
`it was
`assumed to be at
`least as great as those obtained from
`implants which were successfully tested.
`Statistical analysis. Each implant in each femur was trea-
`ted independently for statistical analysis. Sample independ-
`ence was not verified in separate experiments since in
`previous studies this has alwas1 been presumed for the
`transcortical
`implant model.
`The data
`from all
`implants of a given pore size at a given time were tested for
`normality using the Kolmogorov—Smirnov and the Marti—
`nez—Iglewicz tests. For each data set the mean and 95% C!
`were calculated. When data were normally distributed,
`difierences in the extent of bone ingrowth between the two
`types of implant at various times were compared using
`
`Page 6 of 11
`
`

`

`‘JIU
`
`J.
`
`l). BOBYN. G.
`
`.I. S‘I'ALTKPOOL. S. A. HACKING. M. TANZER. J. J. KRYGIER
`
`-—
`
`95% C]
`
`for difference
`—
`
`p value
`—
`
`Table l. Extent of mean bone ingrowth (9%.) as a function of pore size and time
`Bone ingrowlh
`Time — '
`tweelesl
`Small pare
`Large pore
`2*
`-
`13.3 tn 2 ‘4)
`(95% CI 10.8 to I38}
`23.0 (n = 24]
`{95% C] 20.0 to 36.0)
`52.9 {n = 23%
`41.5 tn = ll}
`(95% CI 50.4 to 55.4!
`{95‘}: CI 37.3 to 45.3)
`69.201: 24}
`63.| {n = IE)
`(959?- CI 61010 TLS}
`{95% CI 58.2 to (18.0)
`CLIKHKIIIB
`—12.9 EU -S.4
`rat- in = 2311'
`T9]? (:1 = 24]
`52
`
`(95% CI T6.9tt182.51
`(95% CI 68.3 to 1'10)
`"‘ implants of small pore size were not studied at two and three weeks
`l l
`implant at four and 52 weeks was not included because of a sectioning error
`
`3*
`
`-
`
`4
`
`I6
`
`-
`
`—
`
`6.9 to 151-,l
`
`0.00003
`
`LS to 10.8
`
`llfll
`
`to assess sig—
`the 95% CI
`unpaired Student‘s r—tests at
`nificance. Data for shear strength from the successful
`mechanical tests at four weeks were expressed as a mean
`with 95% CI. An unpaired Student's r—test was used to
`compare these with data from an earlier four-week mechan—
`ical study in which the dimensions of the implant.
`the
`animal model. surgical technique, mechanical test protocol.
`and sample size were virtually identical]
`
`Results
`
`the groups of data from the
`Histological analysis. All
`histological analyses were found to have a normal distribu-
`tion and were evaluated by Student’s r-tests. Table I gives
`details of bone ingrowth and the results of statistical analy-
`sis on differences between means. In implants of large pore
`size. the mean extent of ingrowth at two and three weeks
`was 13.3% and 23.0%. respectively. At four weeks,
`the
`
`‘i-{r for
`extent of bone ingrowth increased to a mean of 52.
`the large pore size compared with 41.5% for the small pore
`size. a difference which was statistically significant. At 16
`weeks, the difference in mean ingrowth between the large
`pore size (69.2%) and the small pore size (63.1%) was also
`significant. By 52 weeks. the mean extent of ingrowth of
`implants with a small pore size (79.7%.) was significantly
`greater than that of those with the large size (70.6%).
`Backseattercd SEM showed a clear pattern of formation
`of new bone within the porous tantalum material. At two
`weeks. the primary source of the initial formation of bone
`was the intramedullary canal and the edges of the drill hole
`(Fig. 3).
`Ingrowth was generally scant and only small
`regions of close apposition of bone to the struts of the
`porous tantalum were observed. At three weeks it appeared
`that more new bone arose from. or was continuous with. the
`drilled edges of the cortex (Fig. 4). in some sections there
`was a slight periosteal reaction and new bone seemed to he
`
`
`
`Fig. 3
`
`Buckscattered scanning electron micrograph of a two-week implant with a
`large pore size showing bone growth of 1393, New bone appears to arise
`from the intramedullary canal and the edges of the drill hole. with little
`bone on the periosteal surface. The bran},I
`trabeculac are immune and
`slender.
`
`Page 7 ofll
`
`
`
`i“:
`
`l.
`
`i"
`
`1mm N033
`
`Backs-tattered scanning: cleclron micrograph of a three-week implant with
`:1 large pore size showing bone ingrowtlt of 21%. Compared with th:
`ilnplnnl at two weeks lFig. 3) there is increased continuity of new burn:
`with Ihe drilled cortical edges. additional new bone arising from the
`perimleai surface and a general thickening of the bony trtlbeculac.
`
`THE JOURNAL OF BONE AND JOINI' SURGERY
`
`Page 7 of 11
`
`

`

`CHARACTERISTICS OF BONE [NGROWTH AND lNTERFACE MECHANICS OF A NEW POROUS 'I'AN'I'AIUM BIOMATERIAL
`
`‘Jll
`
`Figure 5 — Transmitted light photolnicrugraplt of an interface region of a four—week implant with a small pore size. The reversai Jim: from
`the original drill hole is evident and Ill!) niierotcxlure of the tanlalum struts is apparent (paragon X 15). Figure 6 — Backseattered scanning
`electron micrograph of a 10—week implant of small pore size showing ingrowth of T654». There are many regions of contact between new
`hone and the tantalum struts. A pronounced pcriosteal reaction has resulted in additional incorporation of the implant by new bone.
`Haversian remodelling is evident within the porous nelworlrt.
`
`
`
`
`there was very dense ingrowth and extensive bone-implant
`apposition although there were regions of porosity that
`were not
`filled with bone and tantalum struts which
`
`remained incompletely surrounded by new bone. The
`ingrowth often extended abundantly above the periosteal
`aspect of the femur and showed cortical remodelling within
`the pores into Haversian systems. In some sections. partic-
`ularly at four and lo weeks. there was noticeable intra—
`cortical porosity adjacent to the implants. suggestive of a
`remodelling phenomenort associated with increased blood
`flow and bone turnover during the repair Process (Figs 5
`and 6).
`Mechanical testing. Six of the four-week implants with
`small pore sizes were successfully tested mechanically in
`that the interface ruptured before mechanical collapse of
`the implant. The mechanical test data were found to have a
`normal distribution. The mean shear strength was calcu-
`lated to be 18.5 MPa (95% Ci 1?.4 to 19.6). The remaining
`six implants at four weeks and all six implants tested at 16
`weeks had compressive failure before rupture of the bone—
`implant interface. For both time periods, this suggested that
`the values of shear strength were higher than those calcu—
`lated from the successful tests at four weeks.
`
`Fig. ?
`
`Backscaltered scanning electron micrograph of a transverse section
`through a 52—week implant with a small pore size with ittgmwtli of 84%.
`There is some residual porosity around the tantalum struts but relatively
`complete incorporation of the implant.
`
`partly generated from the periosteal surface. Overall. there
`was more ingrowth than at
`two weeks and hence more
`regions of apparent contact of bone with tantalum struts.
`The four—week implants showed more complete ingrowtli
`and many regions of apparent contact between new bone
`and the porous material in all histological sections (Fig. 5).
`[ngrowth of bone across the full diameter of the implant
`was common. At 16 (Fig. 6} and 52 weeks [Figs 7 and 8)
`
`VOL. til-l3. No. 5. SEPTEMBER “£99
`
`Page 8 ofll
`
`
`Discussion
`
`Our study has given an initial characterisation of the
`response of bone to a new porous tantalum biomaterial in a
`canine transcot‘tical model. Substantial filling of the pores
`with new bone to 40% to 50% occurred by four weeks with
`
`Page 8 of 11
`
`

`

`‘Jll
`
`I. D. BOBYN. G. J. STACKPOOL. S. A. HACKING. M. 'I‘ANZER. J. J. KRYGIER
`
`
`
`Fig. 8a
`
`Fig. 8b
`
`Transmitted light micmgraphs of a transverse section through a 52-week implant With small pore size. There are multiple regions of intimate brme-strur
`contact and evidence of a vascular supply throughout the ingrown bone (paragon: a] K II. bi X 27}.
`
`implants of both pore sizes. This resulted in a mean shear
`fixation strength of at least l8.5 MPa (HZTOO PSI). Com—
`pared with previous studies using porouswcoated trans-
`cortical
`implants, high fixation strength occurred much
`earlier with porous tantalum. The data from this investiga-
`tion are best compared with those of an earlier study in
`which
`sintered.
`beaded
`cobalt—chrome
`transcortical
`implants spanning similar pore sizes were used because
`both had equivalent protocols] The maximum mean shear
`strength obtained at four weeks with the beaded cobalt-
`chrome implants was 9.3 MPa (95% CI 8.3 to 10.3). a
`difference of 9.2 MPa (95% CI 7.8 to 10.6) which was
`significant (Student's r-test: p20.004}. Studies using other
`porous metals and different implants and testing techniques
`have reported fixation strengths to cortical bone at
`four
`weeks ranging from 1.2 MPa to 13.1 Ml:‘a.3‘2iL
`The increased rate of development of the interfacial
`shear strength with porous tantalum can best be attributed
`to the higher volume fraction available for ingrowth. While
`fibre metal coatings have a porosity of 40% to 50% and
`sintered—beaded coatings of the type studied by Robyn et
`alI of only 30% to 35%.
`the porous tantalum material
`examined in this study had a substantially higher porosity
`of 75% to 80%. This meant that for any given percentage
`filling. a greater volume of boae was present within the
`porous tantalum,
`thus giving a proportionate increase in
`interface strength. The end result was a faster rate of
`development of strength. The higher porosity of the porous
`tantalum also theoretically allowed a higher ultimate
`strength to develop. Clinically. however. the rate of fixation
`may be more important since the. time required for second-
`ary stabilisation of an implant by bone ingrowth can influ—
`ence the load-bearing after operation.
`The measured strength of the interface probably repre—
`sents a lower bond since half of the implants at four weeks
`and all at 16 weeks failed in compression before rupture of
`the interface. The compressive failure indicated a certain
`reduction in mechanical properties associated with the high
`
`porosity ‘of porous tantalum. The latter. however. has a high
`strengthiweight
`ratio and OVerall mechanical properties
`which would be adequate for most clinical applications.”
`The bulk properties of the porous tantalum can be con
`trolled by the manufacturing process: additional strength
`can be obtained simply by additional deposition of tantalum
`an to the struts. This indicates the need for supplementary
`characterisation of the mechanical properties of the materi-
`al with different processing parameters.
`The pore sizes of both types of implant which we
`evaluated were at the upper limit of the characteristics of
`pore size of conventional porous materials. Nevertheless.
`the rates of bone ingrowth and the overall extent of filling
`of the porosities with new bone were very high. There were
`statistical differences in the mean extent of bone ingrowth
`between the two pore sizes at the different time periods.
`Implants with the large pore size initially had greater
`ingrowth. but by 52 weeks those with the small pore size
`were more completely ingrown. These differences may
`have resulted from the slight inherent variations in pore
`characteristics between the two types of implant or from
`interanimal variability in the healing response. In any case.
`the relative differences were not large and unlikely to result
`in any practical or meaningful clinical consequence. The
`material of small pore size. by virtue of its higher density of
`struts per unit volume. inherently has stronger mechanical
`properties and hence would probably be preferred for most
`clinical applications.
`It is possible that the surface microtexture of the struts
`forming the porous material contributed to the overall
`osteogenic response. Cell~culture studies and studies in
`animals with both unloaded and dynamically loaded
`implants have shown that microtextured surfaces such as
`those produced by grit blasting or acid etching are highly
`osteophilic.22'25 The histological studies clearly showed
`that the porous tantalum served as an effective scaffold f0?
`relatively complete incorporation with new bone by 1‘5
`weeks, with little change after 52 weeks of implantation.
`
`THE JOURNAL OF BONE AND JOINT SURGERY
`
`Page 9 ofll
`
`Page 9 of 11
`
`

`

`CIHRAC'TERIS'I'ICS {JF BONE INt'rRl'JW'l'l-l AND lN'l'EHFAC'E MECHANICS OF A NEW I’ORDUS TANTALLTM BIDMATERIAI.
`
`913
`
`load-
`Although not as realistic as a fully-functional
`bearing model. the transconical model is very useful for the
`initial characterisation of new porous hiomaterials.
`It
`allows control of several parameters which influence the
`hone-healing response. Using the appropriate instruments.
`holes can be drilled with accuracy and reproducibility. thus
`ensuring a uniform initial fit of the implant. The density of
`the peri-implant bone is consistent and thus allows equita-
`ble comparison of data within and between animals.
`Although implants for joint replacement are primarily sur-
`rounded by cancellous bone. retrieval analyses have shown
`that
`the most consistent and abundant source of bone
`
`ingrowth in hip prostheses is from cortical regions. The
`model has also been used in many previous studies of
`porous materials and therefore allows comparison with
`historical controls. 15433“
`tantalum is partic-
`From a manufacturing standpoint.
`ularly well suited to the complex CVDlCVl process used
`for deposition on to lhe vitreous carbon substrate.
`It is a
`strong. ductile metal with excellent corrosion resistance.
`More than 50 years ago its potential in regard to human
`implanttswtvas
`studied in animal
`and human experié
`ments.‘
`‘
`It has been used for a wide variety of implants“
`including pacemaker electrodes.“ cranioplasty plates.32
`ligation clips.33 femoral endoprostheses.34 as wire. foil and
`mesh for nerve repair?5 contrast media for atrwave radio-
`graphic studiesfifi and as a radiopaque marker for following
`bone growth and implant migration.” Osseointegration has
`previously been demonstrated using non—porous tantalum
`implants in dental and orthopaedic applications for periods
`of up to 8 to 12 years:Ml Its superb biocompatibility and
`suitable mechanical properties have led to its standardisa—
`tion as a surgical implant material'm'u
`The tantalum construct which we have evaluated repre—
`sents a departure from conventional porous materials in
`many respects. Because of its high por