Performance of Porous Tantalum vs. Titanium Cup in Total Hip Arthroplasty_ Randomized Trial with Minimum 10-Year Follow-Up.

Performance of Porous Tantalum vs. Titanium Cup in Total Hip Arthroplasty:Randomized Trial with Minimum 10-Year Follow-UpJulien Wegrzyn MD, PhD, Kenton R. Kaufman PhD, PE, Arlen D. HanssenMD, David G. Lewallen MDPII:DOI:Reference:To appear in:Received date:Revised date:Accepted date:S0883-5403(15)00023-6doi: 10.1016/j.arth.2015.01.013YARTH 54276Journal of Arthroplasty29 May 20145 January 201511 January 2015Please cite this article as: Wegrzyn Julien, Kaufman Kenton R., Hanssen Arlen D.,Lewallen David G., Performance of Porous Tantalum vs. Titanium Cup in Total HipArthroplasty: Randomized Trial with Minimum 10-Year Follow-Up, Journal of Arthro-plasty (2015), doi: 10.1016/j.arth.2015.01.013This is a PDF le of an unedited manuscript that has been accepted for publication.As a service to our customers we are providing this early version of the manuscript.The manuscript will undergo copyediting, typesetting, and review of the resulting proofbefore it is published in its nal form. Please note that during the production processerrors may be discovered which could aect the content, and all legal disclaimers thatapply to the journal pertain.ACCEPTEDMANUSCRIPTACCEPTED MANUSCRIPTPerformance of Porous Tantalum vs. Titanium Cup in Total Hip Arthroplasty:Randomized Trial with Minimum 10-Year Follow-UpRunning title: Minimum 10-Year Performance of Porous Tantalum CupJulien Wegrzyn, MD, PhD, julien.wegrzynchu-lyon.frKenton R. Kaufman, PhD, PE, Arlen D. Hanssen, MD, David G. Lewallen, MD, Department of Orthopedic Surgery, Mayo Clinic, 200 First Street SW, Rochester, MN 55905Corresponding author:David G. Lewallen, M.D.Department of Orthopedic SurgeryMayo Clinic200 First Street SWRochester, MN 55905Tel: (507) 284-4896Email: 1ACCEPTEDMANUSCRIPTACCEPTED MANUSCRIPTAbstractPorous tantalum monoblock cups have been proposed to improve survivorship of cementlessprimary THA. However, there are few direct comparative trials to established implants such asporous-coated titanium cups. 113 patients were randomized into two groups according to the cup:a porous tantalum monoblock cup (TM) or a porous-coated titanium monoblock cup (control).The follow-up averaged 1437 months in the TM group and 14510 months in the control group.At latest follow-up, no implants migrated. Two TM patients (4%) and 13 control patients (33%)presented with radiolucency around the cup (p0.001). In the control group, 1 cup (2%) wasrevised for aseptic loosening. At 12 years post-implantation, porous tantalum monoblock cupsdemonstrated 100% survivorship, and significantly less radiolucency as compared to porous-coated titanium monoblock cups.Level of evidence: Therapeutic Level I.Key words: total hip arthroplasty, porous tantalum monoblock cup, cementless fixation,osseointegration, aseptic loosening.2ACCEPTEDMANUSCRIPTACCEPTED MANUSCRIPTIntroductionDespite substantial improvements in cementless total hip arthroplasty (THA) implants, periprostheticosteolysis and subsequent aseptic loosening secondary to biologic reaction to polyethylene (PE) and metallicwear debris represents one of the leading causes of implant revision and accounts for 20% of the overallrevision THA procedures 1-3. In an attempt to reduce mechanical failures of implant fixation, monoblockacetabular components have been proposed as an alternative to cemented PE or conventional cementlessmodular implants 4, 5. The monoblock design combines advantages of no PE backside wear due to linermicromotions, no metallic debris generated by deficient locking rings, and no holes which decrease surface areafor bone ingrowth and allow pelvic entrance pathways for wear debris 4, 5. Accordingly, a porous tantalummonoblock cup (Hedrocel, Implex Corp., Allendale, NJ) and a porous-coated titanium monoblock cup(Elliptical, Implex Corp., Allendale, NJ) have been developed. The rationale of the Hedrocel cup is tocombine the advantages related to the monoblock design with the unique properties of porous tantalumimplants: a high friction coefficient that increases interface shear stress and primary frictional stability, a highvolumetric porosity that enables extensive tissue integration and bone ingrowth into the shell scaffold, and asubchondral bone-matched elastic modulus that results in a more physiological mode of load transfer to the hostbone with decreased periacetabular stress shielding 6-9.Short- to mid-term follow-up series evaluating porous tantalum monoblock cups have shownencouraging results with excellent initial fixation, osseointegration and stability; no evidence of progressiveradiolucent lines, osteolysis, migration or gross PE wear; and no reported acetabular component revision foraseptic loosening 10-17. However, to our knowledge, no prospective and randomized study to date has beenperformed to compare these promising early results obtained with such implants to more conventionalcementless monoblock acetabular components. Therefore, this randomized controlled trial (RCT) aimed tocompare the clinical and the radiological outcome of a porous tantalum monoblock cup to a porous-coatedtitanium monoblock cup in primary THA with a minimum of 10-year follow-up. We hypothesized that the useof porous tantalum for acetabular component in THA would improve implant osseointegration and survivorship,and reduce revisions associated with periprosthetic osteolysis and fixation failure at long-term follow-up.3ACCEPTEDMANUSCRIPTACCEPTED MANUSCRIPTPatients and MethodsA randomized controlled trial (RCT) on primary THAs with cementless monoblock acetabularcomponents made of either porous tantalum (Hedrocel) or porous-coated titanium-alloy (Elliptical) wasinitiated in January 1998 and completed in December 1999. The Hedrocel porous tantalum monoblock cupconsists of ultra-high molecular weight PE (UHMWPE, GUR 1020, Perplas Medical Ltd, Lancashire, UK)directly compression-molded into a backing made entirely of porous tantalum 6 (Figure 1). The poroustantalum is an open and fully interconnected tridimensional porous surface with porosity of 75-80% and anaverage pore size of 550 m 6, 9, 18. The elastic modulus is 3 GPa, which is between elastic modulus ofsubchondral (2 GPa) and cortical (15 GPa) bone 6, 9, 18. The UHMWPE is moderately crosslinked andsterilized in nitrogen with 30 kGy of -radiation from a 60Co source 6. The direct compression moldingresults in PE intrusion into the porous shell to a depth of 1 to 2 mm, leaving 2 to 3 mm of porous tantalum forbone ingrowth 6, 7 (Figure 1). The Elliptical porous-coated titanium monoblock cup consists of acompression-molded UHMWPE preassembled into a Ti6V4Al alloy shell 5, 19. The coating surface is madeof 3 layers of 200 to 300 m-diameter pure titanium beads 5, 19. The porosity of the coating surface is 30-50% and the elastic modulus of Ti6V4Al alloy shell is 110 GPa 18, 19. The characteristics of the UHMWPEare the same as for the Hedrocel cups. The geometry of both cups is a hemi-ellipsoid designed for peripheralinterference fit into an acetabular bone shaped with hemispherical reamers 5, 6. The equator diameter is 2 mmlarger than the polar diameter allowing maximal coaptation and press-fit at the time of implantation 5, 6.Inclusion in the current RCT has been proposed to 278 patients during their preoperative visit to one oftwo senior surgeons at our institution (Figure 2). The inclusion criterion was patient eligibility for primarycementless or hybrid THA with sufficient periacetabular bone stock for peripheral rim fixation. The exclusioncriteria were acetabular segmental or rim non-supportive defect, severe acetabular dysplasia, and severeacetabular deformity related to acetabular fracture or advanced osteoarthritis. We excluded 167 patients whomatched the inclusion criteria since they declined to participate in the study (Figure 2). The assigned treatmentwas generated by a computerized randomization program administered by our Department of Biostatistics. For4ACCEPTEDMANUSCRIPTACCEPTED MANUSCRIPTeach patient, an opaque and sealed envelope containing the assignment to a group was chosen at random by thefirst assistant surgeon in the operating room before cup implantation. One hundred and eleven patients (111hips) were randomized into the porous tantalum monoblock cup (TM, 56 patients) and porous-coated titaniummonoblock cup (control, 55 patients) groups. Seven patients in the TM group and 9 patients in the control groupwere lost to follow-up due to failure to return for the postoperative evaluation and no response to phone calls orletters. Three patients in the TM group and 5 patients in the control group died before reaching a minimum 10-year follow-up. Therefore, 45 patients in TM group and 41 patients in the control group with a minimum 10-year follow-up of their THA were analyzed in this study (Figure 2). No differences between groups were foundin the following pre-operative parameters: age at surgery, sex ratio, BMI and preoperative Harris hip score(Table 1). Institutional review board approval and written informed consent were obtained from each patientbefore their enrollment in the study. A Hardinges lateral transgluteal approach was performed in all the patientsin both groups by one of two senior surgeons at our institution. Reaming was performed to the size of theacetabular component to be used. Positioning of the cup was in relation to the neighboring anatomic landmarks,seeking anteversion of 15 to 20 degrees and cup abduction of 40 to 45 degrees. All the procedures wereperformed without cement, screws or autologous bone graft or other osteogenic factors placed onto the reamedacetabular bony surface or onto the acetabular component. A 28-mm femoral head was used in all cases andmade of either cobalt-chromium alloy (CoCr) or zirconia ceramic (Table 2). For the femoral component, acemented wrought CoCr alloy stem or a press-fit femoral fixation with a porous-coated CoCr alloy orhydroxyapatite-coated titanium alloy femoral stem was implanted depending on bone quality (Table 2).Data were collected from our institutional Total Joint Registry. Clinical evaluation was performed usingthe Harris Hip Score (HHS) 20. In addition, intra- and post-operative operative complications were recorded.Standard anteroposterior (AP) and cross-table lateral hip radiographs were obtained for all the patientspreoperatively, six weeks postoperatively (for baseline postoperative radiographs) and then at regular intervalsthereafter until latest follow-up. All the radiographs were evaluated by a single independent surgeon notinvolved in any of the surgical procedures or patient management. The baseline AP radiograph was used toassess the initial angle of abduction of the cup with respect to the landmarks of the teardrop line and the long5ACCEPTEDMANUSCRIPTACCEPTED MANUSCRIPTaxis of the projected ellipse of the face of the cup. Special emphasis was given to the initial postoperative fit ofthe cup as manifested by radiographic evidence of interface “gaps.” Gaps were defined as areas in which theporous surface of the cup was not initially in contact with acetabular bone, to differentiate the areas of apparentdecreased density from radiolucencies that appear on subsequent films in areas where no gaps had existedinitially 21, 22. Such gaps were recorded as present when it involved 50% or more of the zone in which it wassituated. The width of the gaps was measured to the nearest millimeter. The cup-bone interface was evaluated inthe baseline postoperative and all subsequent AP radiographs using the three contiguous acetabular zonesdelineated by DeLee and Charnley 23.Cup fixation was evaluated by measures of migration, radiolucency, osteolytic lesions, and gap filling. Acomponent was considered to have migrated if change in position was greater than 3 mm and/or 5 with respectto the baseline radiograph according to the criteria of Massin et al. 24-26. A cup with progressivecircumferential radiolucent lines greater than 2 mm or evidence of migration was considered as loosened 24-26. Evaluation of cup fixation also included evidence of periacetabular osteolytic lesions with a minimumdimension of 3 mm either along and/or perpendicular to the interface within each zone, and initial postoperativegap filling 24-26. Finally, gross PE wear was qualitatively evaluated by the occurrence of obvious femoralhead migration into the PE on latest follow-up radiographs.Descriptive statistics are presented as meanSD. Qualitative variables were compared using a chi-squaretest. Quantitative and continuous variables were analyzed using the following parametric tests: (1) Studentspaired t-tests for intragroup comparison of two variables, (2) two-sample t-tests for intergroup comparison oftwo variables, and (3) repeated measures ANOVA to determine the presence of significant differences forcontinuous variables between the pre-operative and the two consecutive post-operative periods. Statisticalanalyses were performed using SPSS 16.0 (IBM Corp., Armonk, NY) with a level of significance set atp0.05.Results6ACCEPTEDMANUSCRIPTACCEPTED MANUSCRIPTThe follow-up averaged 1437 months in the TM group and 14510 months in the control group(p=0.603). The HHS improved significantly at two months and at latest follow-up in both groups (Table 3). Nosignificant difference was detected in HHS improvement between the two groups at two months and at latestfollow-up (p=0.454 and 0.104, respectively). No perioperative complication related to the cup was reported.However, 1 (2%) patient in the TM group presented with a posterior dislocation related to a fall at 117 months.Closed reduction was performed without recurrence at latest follow-up.No significant difference was detected in cup abduction angle at 2 months and at latest follow-upindicating that no implant migrated (Table 3). Baseline radiograph analysis showed 5 (11%) and 8 (19%) initialpostoperative gaps in the TM group and in the control group without significant difference between the twogroups (p=0.206) (Table 4). In addition, the initial postoperative gaps demonstrated evidence of filling for bothgroups at latest follow-up (Figures 3A the molded PE was reamed away using hemisphericalreamers and a highly-crosslinked PE liner (Longevity Standard, Zimmer, Warsaw, IN) was cemented into thecup 27 (Figure 4B). In addition, the cemented femoral component was revised since there was evidence ofdebonding of the stem from the cement mantle. At the latest follow-up of 157 months from the index procedureand 40 months from the revision procedure, there were no progressive radiolucencies or radiographic evidenceof osteolysis around the implants, migration, or PE wear (Figure 4B). However, porous tantalum debris,produced by the reaming, was visible on the postoperative radiographs (Figure 4B). This debris couldpotentially cause third-body wear.7ACCEPTEDMANUSCRIPTACCEPTED MANUSCRIPTIn the control group, one THA was revised for acetabular aseptic loosening at 144 months (Figure 5).The index procedure was performed on a 37-year-old man for avascular necrosis of the femoral head andconsisted of a cementless THA with a 58-mm Elliptical cup articulated against a 28-mm zirconia femoralhead. The radiographs showed clear evidence of fibrous rather than bone ingrowth around the cup and minimalPE wear (Figure 5). Once the cup was explanted during the revision procedure, the back-side showed noevidence of bone attachment on the porous-coated surface.DiscussionOsteolysis resulting from biologic reaction to PE and metallic wear debris has been recognized as amajor limitation in long-term survival of modular cementless acetabular components, particularly whenassociated with central dome and multiple screw holes 1-4, 24. Combining the advantages of a monoblockdesign with the dramatic properties of primary stability and bone ingrowth obtained with porous tantalum, aporous tantalum monoblock cup has been proposed to improve survivorship of cementless primary THA 10-17. However, no RCT in literature has been designed to evaluate the potential advantages of porous tantalum ina monoblock design for acetabular component with direct comparison to a more conventional porous-coatedtitanium monoblock cup.The current randomized study demonstrated excellent overall survivorship in both groups. At a mean of12 years after THA, 100% of the porous tantalum monoblock cups were found to be osseointegrated with nocup revision for osteolysis, progressive radiolucencies or aseptic loosening. In contrast, one porous-coatedtitanium monoblock cup (2%) was revised for aseptic loosening at 12 years after implantation (Figure 5). Inaddition, the number of patients with radiolucent lines was significantly higher in the porous-coated titaniumgroup and might raise concerns about the potential for future cup loosening and revision (Table 5). This studyalso confirmed that the initial postoperative gaps at the bone-implant interface, although high with themonoblock design, did not influence long-term stability regardless of the shell cast. In addition, extensive PEwear did not appear to be of concern in our series at a mean of 12-year follow-up. In the porous tantalum group,one THA (1%) was revised for recurrent instability associated with mild PE wear without radiolucent line or8ACCEPTEDMANUSCRIPTACCEPTED MANUSCRIPTosteolysis around the cup at 10 years after implantation (Figure 4A). In this case, the molded PE was reamingaway and the well-fixed porous tantalum shell was not removed (Figure 4A&B). Revision consisted of adouble-socket technique with cementation of a PE liner into the remaining porous tantalum shell with excellentstability of the cemented fixation of the PE liner at 3 years postoperatively (Figure 4B) 27. Therefore, thisrandomized study confirms and strengthens the results of the previous series