Cementless titanium tapered-wedge femoral stem

The Journal of Arthroplasty Vol. 19 No. 5 2004

Cementless Titanium Tapered-Wedge Femoral Stem 10- to 15-Year Follow-Up Amanda D. Marshall, MD,* Jeffrey G. Mokris, MD,† Richard D. Reitman, MD,‡ Alicia Dandar, MS,† and David R. Mauerhan, MD*

Abstract: This investigation is an ongoing clinical and radiographic analysis of a titanium tapered-wedge femoral component with a proximal plasma-spray porous coating. Integral femoral stems (Biomet, Warsaw, IN) were implanted in 200 hips in 186 patients. Nineteen patients died before 10-year follow-up, and 50 patients were lost to follow-up. The mean follow-up of the remaining 129 hips was 11.6 years. Harris Hip Scores improved from 58 to 93. Thigh pain was 2.3%. Radiographic analysis revealed adaptive distal remodeling in zones 2, 3, 5, 6, and 13, with no evidence of osteolysis below the level of the calcar and the greater trochanteric region. Only 2 femoral stems were revised: 1 with suspected fibrous fixation at 7 years postoperatively and another with a broken trunion at 10 years postoperatively. A tapered titanium femoral stem with circumferential plasma-spray porous coating provides excellent long-term fixation, durable clinical outcome, and protects against osteolysis below the level of the calcar and greater trochanter. Key words: total hip arthroplasty, cementless femoral fixation, integral femoral stem, radiographic analysis, arthritis. © 2004 Elsevier Inc. All rights reserved.

Cementless femoral fixation in total hip arthroplasty has evolved over the last 25 years, with early and intermediate results rivaling those of the best cemented outcomes at comparable follow-up periods [1–5]. Longer-term studies of cementless implants, however, have raised concern regarding the incidence of thigh pain, osteolysis around wellfixed components, and long-term fixation, with

mixed results among various stem designs [6 –12]. Cementless stems generally fall into 2 philosophical categories: 1) diaphyseal fixation, and 2) proximal fixation with a metaphyseal-filling component. Diaphyseal stems are usually cylindrical, with varying degrees of porous coating, whereas metaphyseal stems are either a tapered-wedge design or an anatomic (ie, curved) design. Diaphyseal-fitting stems have shown femoral stem survivorship up to 97% at long-term follow-up; however, concerns have included proximal stress shielding and thigh pain (with rates up to 25%) [13,14]. Proximal-fit stems also have demonstrated long-term durability, with lower incidences of thigh pain [15–21]. As reported in our previous study of 190 hips using the Integral stem (Biomet Inc., Warsaw, IN) at 5- to 8-year follow-up, 99% of stems demonstrated bony ingrowth, with a 4% incidence of thigh pain and no evidence of distal osteolysis [22]. This study reports the long-term clinical and radio-

From the *Carolinas Medical Center, Department of Orthopaedic Surgery, Charlotte, North Carolina; †Miller Orthopaedic Clinic, Charlotte, North Carolina; and ‡Texas Center for Joint Replacement, Plano, Texas. Submitted May 31, 2003; accepted January 26, 2004. Benefits or funds were received in partial or total support of the research material described in this article from Biomet, Warsaw, Indiana. Reprint requests: Amanda D. Marshall, MD, Carolinas Medical Center, Department of Orthopaedic Surgery, 1000 Blythe Blvd, Suite 503 MEB, Charlotte, NC 28203. © 2004 Elsevier Inc. All rights reserved. 0883-5403/04/1905-0003$30.00/0 doi:10.1016/j.arth.2004.01.006

546

Integral Femoral Stem • Marshall et al.

547

graphic results of this proximally porous-coated femoral component. In addition to radiographic analysis, implant stability, incidence of thigh pain, and femoral revision rates/survivorship with this implant are compared with reports of other cementless implants.

Materials and Methods Between July 1986 and December 1989, 200 primary cementless total hip arthroplasties were performed in 186 patients. This was a consecutive series of patients selected for cementless fixation of both components. Since our initial publication, 10 hips in 7 patients (previously thought to be lost to follow-up) were located, evaluated, and subsequently included in this review. Nineteen patients (10%) expired before the minimum follow-up of 10 years. Of the remaining 167 living patients, 50 were lost to follow-up, leaving 129 hips in 117 patients available for analysis (70% follow-up rate). Multiple attempts, including phone calls, were made to contact patients and encourage follow-up. In many cases, the issues of cost, driving distance and time, loss of work, or change in health plan were cited as reasons for not participating. This study was approved by the Institutional Review Board. The cohort consisted of 63 men (54%) and 54 women (46%), with a mean age of 56 years (range, 31– 81 years). Right hips accounted for 45%, and left hips accounted for 55%. Preoperative diagnoses were primary osteoarthritis (64%), avascular necrosis (23%), posttraumatic arthritis (8%), developmental dysplasia (3%), and rheumatoid arthritis (2%). The Integral femoral stem is a tapered, collared, titanium component with a proximal circumferential titanium plasma-spray coating. The geometry consists of a proximal biplanar tapered wedge with a 3° proximal-to-distal taper. The substrate is gritblasted, with the overlying plasma-spray coating covering the proximal 30% to 50% of the stem, extending more distally at the lateral aspect of the prosthesis (Fig. 1). The acetabular component consisted of a threaded titanium plasma-sprayed hemispherical cup in 87 hips (67%) and a hemispherical titanium plasma-sprayed cup with screws in 42 hips (33%). Femoral head sizes were 28 mm or 32 mm, with the majority being titanium alloy. All surgeries were performed by the 2 senior surgeons (D.R.M. and J.G.M.) using a posterolateral approach. Sequential conical reaming and broaching were performed with broaches undersized by 0.5 mm to the implanted component to maximize

Fig. 1. Integral femoral stem (Biomet).

fit and fill of the femoral canal. The postoperative regimen consisted of 6 weeks of partial weight bearing, which then advanced to full weight bearing as tolerated. Outcome variables consisted of a clinical evaluation and radiographic analysis at each preoperative and postoperative visit. In addition to preoperative assessment, patients were followed at 6 weeks, 1 year, 2 years, 5 years, 7 years, and 10 years. The most recent evaluation ranged from 10 to 15 years postoperatively, with an average follow-up of 11.6 years. Radiographs obtained at all time points included an anteroposterior radiograph of the pelvis and hip and a frog-leg lateral of the hip. Clinical evaluation consisted of pain and function assessment using a Harris hip Score, as well as specific questioning regarding not only the presence, but also the severity of groin or thigh pain. Severity was graded as slight (occasional pain not requiring alteration in pattern of activity), mild (occurring only after unusual activity or requiring anti-inflammatory agent), moderate (requiring alteration of activities or requiring narcotic pain medication), or severe (resulting in serious limitations to any activity). Harris Hip Scores of 90 to 100 were considered excellent, 80 to 89 good, 70 to 79 fair, and ⬍70 considered poor. Radiographic analysis was provided by an independent observer (R.R.) at an outside institution without knowledge of the patients’ clinical scores. Preoperative radiographs were evaluated for Dorr bone subtype [23]. The immediate postoperative and most recent follow-up radiographs were re-

548 The Journal of Arthroplasty Vol. 19 No. 5 August 2004 viewed for signs of femoral and acetabular failure. The focus of this study was the performance of the femoral component, and as such, no detailed radiographic analysis of the acetabular components is included in this report. Femoral component orientation was deemed neutral if the center lines of the component and the femur were within 3°; otherwise, the component was designated as either varus or valgus. The fill of the medullary bone by the component was measured at the junction of the porous-coated and non–porous-coated area. This value was expressed as a percentage. Using a fixed point of reference on the prosthesis and the lesser trochanter, component position was assessed. Subsidence was present if the component settled 5 mm or more. Ectopic bone was graded per the Brooker et al [24] classification. Using a zonal analysis described by Gruen [25], all changes around the femoral component were documented. Calcar changes (atrophy or hypertrophy), radiolucent lines, endosteal new bone formation near the prosthesis (spot welds), presence of endosteal bone at the distal tip of the prosthesis (an intramedullary shelf of new bone), and cancellous bone changes were evaluated according to definitions described by Engh et al [8]. The locations of osteolysis also were recorded as such. Indicators of stable component fixation or osseointegration included the absence of radiolucent lines around the proximal porous surface, the presence of new bone bridging the gap between the endosteal surface and implant (spot welds), and no subsidence. Signs of implant instability included the presence of progressive implant migration or subsidence, calcar hypertrophy, and radiolucent lines about the area of the porous surface. A radiolucency encompassing ⬎50% of the interface in the area of the porous coating was indicative of fibrous fixation or failure of osseointegration.

Results Clinical Outcome Improvement in the hip scores was notable at the 10- to 15-year final follow-up. The average preoperative pain score was 21, increasing to 41 at longer than 10 years postoperatively. The average preoperative function score was 58, increasing to 93 postoperatively. Clinical outcome was rated excellent in 75%, good in 13%, fair in 7%, and poor in 5%. With the exception of 2 patients with multiple medical comorbidities and radiographic evidence of acetabular osteolysis, all patients with poor and fair results were either awaiting or within 1 year post-

operatively from an acetabular revision. The prevalence of thigh pain was 2.3% (3 of 129 hips). Thigh pain was documented as mild in 2 patients and moderate in 1 patient. No patients had severe or disabling thigh pain. One patient underwent revision secondary to fibrous fixation, and the other 2 patients are awaiting acetabular revision. Both have evidence of acetabular osteolysis, with no femoral osteolysis or other signs of instability of the femoral component. Two stems were revised (2%), 1 because of suspected fibrous fixation and another secondary to a broken trunion. The 1 case of fibrous fixation occurred in a 61-year-old woman with posttraumatic arthritis. The acetabular component was revised at 5 years postoperatively, and despite radiographic suspicion for fibrous fixation, the femoral component was stable intraoperatively. Postoperatively, she became chronically unstable, and 1 year from her revision surgery, an attempt at trochanteric advancement was unsuccessful. At 7 years from her index arthroplasty, she was revised at an outside institution, which included revision of her femoral component to gain stability. The mechanical failure occurred in a 47-year-old man with a diagnosis of posttraumatic arthritis. At 9.8 years postoperatively, the patient experienced acute onset of pain, and radiographs revealed a broken trunion at the junction of the 28 mm long-neck head with the trunion. He subsequently underwent revision to a diaphyseal-fit, long-stem prosthesis. Three patients sustained periprosthetic fractures. The components were all well fixed and intact (Vancouver B1) and were successfully managed by open reduction and internal fixation. Seventy-four hips (57%) underwent revision of the acetabular component. Of the revisions, 74% were threaded acetabular components and 26% had a porous hemispherical cup with adjuvant screws. At the time of acetabular revision, no femoral component showed radiographic evidence of loosening, and intraoperative evaluation of the femoral component found all to be solidly fixed and stable. The use of titanium femoral heads as well as thin polyethylene liners with first-generation locking mechanisms contributed considerably to these failures. Radiographic Analysis Forty of the hips (31%) were Dorr type A, 83 hips (64.3%) were type B, and 6 hips (4.7%) were type C. One hundred three stems (80%) were in neutral position, 17 were in varus, and 9 were in valgus. Femoral subsidence ⬍2 mm was seen in 6 hips. All

Integral Femoral Stem • Marshall et al.

of these demonstrated spot welds. One hundred twenty-three hips demonstrated no subsidence. No hip subsided ⬎2 mm. Calcar atrophy was noted in 93 hips (72%): above the level of the lesser trochanter in 82 hips and to the level of the lesser trochanter in 11 hips. Spot welds were present in 99.3% of the hips. This was most common adjacent to the porous coating. Endosteal bone formation at the tip of the stem was identified in 66 hips (51%). This phenomenon was associated with direct bone apposition to the tip of the prosthesis. Cortical hypertrophy was evident in 34 hips (26%). Most common zones of hypertrophy were 2, 3, 5, 6, and 13. There were no cases of distal osteolysis. Six stems demonstrated radiolucencies ⬎2 mm. In only 1 of these did the radiolucency constitute ⬎50% of the porous-coated zone. This correlated with the clinical suspicion of the study’s 1 case of fibrous fixation. One hundred twenty-eight hips were stably fixed (99%). An example of long-term radiographic follow-up is seen in Fig. 2. There was 1 case of fibrous fixation (1%). Survivorship at 10 years was 98%.

Discussion Since the introduction of uncemented hip implants in the 1970s, femoral and acetabular designs have undergone substantial change. On the femoral side, various stem geometries combined with different surface coatings have been used with variable results in the literature. Because of the historically accepted heuristic approach to implant validation, this era of evolution has produced designs with unacceptably high rates of clinical failure [26 –28]. We now know that initial implant stability is critical to long-term fixation and that circumferential porous coatings contribute to long-term bone ingrowth, with a very low incidence of distal migration of polyethylene wear debris and subsequent osteolysis [17–20]. In comparison with the Integral’s tapered-wedge design, the Anatomic Medullary Locking (AML) prosthesis (DePuy, Warsaw, IN) relies on a diaphyseal fit for stability. Engh and Culpepper [13] reported an average of 11-year follow-up data on 174 hips. Thigh pain occurred in 13% of patients, with a 50% incidence of proximal stress shielding. No distal lysis was noted. Femoral stem survivorship was 97%. Kim et al [14] also investigated long-term results of the AML prosthesis, reporting a 25% incidence of thigh pain, 55% of stems demonstrating lysis, and a 6% rate of femoral component loosening.

549

Clinical results of the titanium tapered-wedge stems have been relatively consistent. McLaughlin et al [17] followed 114 hips using the TaperLoc stem (Biomet) over a period of 8 to 12.5 years. With a 100% follow-up rate, only 2% of patients experience moderate thigh pain, a 6% incidence of proximal osteolysis, and a 15% revision rate for aseptic loosening. Bourne et al [18] reported 10- to 13-year follow-up on 307 tapered plasma-spray stems (Mallory-Head, Biomet). No femoral revisions were noted despite a 10% acetabular revision rate and 50% mild proximal stress shielding. Three percent of patients reported thigh pain. Head et al [19] also followed 188 Mallory Head stems with 11-year follow-up. Survivorship of the femoral component was 99.5%, with no reports of osteolysis. Thigh pain was present in 2% of patients. Meding et al [20] presented data on 105 tapered titanium stems (Bimetric, Biomet), followed for an average of 10.4 years. No revisions and no osteolysis were reported. Thigh pain was noted to be 2%. Using the Corail stem (DePuy) with an extensive hydroxyapatite coating, Froimson et al [21] did not report any femoral stem revisions either. One hundred fifty-three stems with minimum 10-year data displayed 2% stress shielding and 21% osteolysis. Other tapered-wedge designs include the TriLock (Depuy) and the Omnifit (Howmedica, Allendale, NJ) stems. Both are cobalt-chromium implants with proximal sintered beads. Sakalkale et al [15] reported on the TriLock (Depuy) in 71 hips at an average of 11.5 years postoperatively. No stem was revised for loosening despite 5% evidence of radiographic loosening, and only 1.4% of patients experienced thigh pain. Hellman et al [16] reported on a group of 67 hips with 5- to 13-year follow-up (average, 10 years) implanted with the Omnifit stem (Howmedica). Proximal femoral osteolysis was common (54%), and 13 hips (20%) required a “reoperation for large osteolytic lesions independent of component loosening.” Limiting thigh pain was present in 4%. The incidence of thigh pain in this review was 2.3%, which has decreased from the 4% reported in our 5- to 8-year review. In the current study, we noted that 51% of our hips had endosteal bone formation at the tip of the stem. In addition, cortical hypertrophy was noted in zones 2, 3, 5, and 6. We believe that this endosteal new bone formation at the tip of the implant in this titanium taperedwedge design is a positive adaptive remodeling response. It is our impression that as time passes, the femur remodels around the implant, resulting in less thigh pain as compared with the early postoperative period when thigh pain can be quite both-

550 The Journal of Arthroplasty Vol. 19 No. 5 August 2004

Fig. 2. (A) Six-week follow-up radiograph (anteroposterior). (B) Six-week follow-up radiograph (lateral). (C) Ten-year follow-up radiograph (anteroposterior). (D) Ten-year follow-up radiograph (lateral).

Integral Femoral Stem • Marshall et al.

ersome depending on which stem design is used. The distal-fit prosthesis thigh pain rates range from 13% to 25% [13,14]. Tapered-wedge designs have a relatively lower range of reported thigh pain (1.4%– 4%) [15–22]. The use of a titanium tapered-wedge design with plasma-spray porous coating as reported in this study has excellent 10- to 15-year clinical and radiographic results. We have demonstrated endosteal spot welds in 99.3% of cases, with adaptive bone remodeling indicative of bony fixation. There have been no cases of distal osteolysis observed, validating the concept that circumferential plasma spray does indeed protect against distal migration of particulate debris at long-term follow-up. Osteolysis has been confined to the upper calcar area and the upper greater trochanter, without evidence of distal progression in spite of continuing wear-debris generation. This has been observed in light of the fact that the failure rate of the acetabular components continues to rise: from 25% in the intermediate follow-up period to 57% in this study. These early patients were implanted with first-generation acetabular components, in particular threaded designs, and poor polyethylene locking mechanisms. Additionally, 28-mm and 32-mm titanium heads were used as the articulating bearing surface. Despite these inferior bearing surfaces by today’s standards and the subsequent wear debris generated, the femoral component and its bone-component interface has demonstrated excellent durability, both clinically and radiographically. Inherent to the majority of longer-term follow-up studies is the possibility of bias introduced with increasing number of patients lost to followup, from which this report is not exempt. The clinical and radiographic follow-up rate for the current study is 70%. As previously stated, multiple attempts were made to contact patients and encourage follow-up.

Conclusion The study’s results demonstrate that the Integral femoral stem provides 99% stable fixation and 98% survivorship at long-term follow-up. Despite the high rates of acetabular failures necessitating acetabular revision, the femoral stems remained well fixed. This investigation further supports the continued use of titanium tapered-wedge circumferentially porous-coated cementless femoral stems, and the concept that once stable bone ingrowth fixation is achieved, the bone-prosthesis interface remains durable at longer-term follow-up.

551

References 1. Alexiades MM, Clain MR, Bronson MJ: Prospective study of porous coated anatomic total hip arthroplasty. Clin Orthop 269:205, 1991 2. Bands R, Pelker RR, Shine J, et al: The noncemented porous-coated hip prosthesis. Clin Orthop 269:209, 1991 3. Callaghan JJ, Dysart SH, Savory CG: The uncemented porous-coated anatomic total hip prosthesis. J Bone Joint Surg Am 70:337, 1988 4. Lachiewicz PF, Anspach WE, DeMasi R: A prospective study of 100 consecutive Harris-Galante porous total hip arthroplasties. 2-5 year results. J Arthroplasty 7:519, 1992 5. Rosenberg A: Cementless total hip arthroplasty: femoral remodeling and clinical experience. J Orthop 12:1223, 1989 6. Barrack RL, Jasty M, Bragdon C, Haire T, et al: Thigh pain despite bone ingrowth into uncemented femoral stems. J Bone Joint Surg Br 74:507, 1992 7. Campbell ACL, Rorabeck CH, Bourne RB, Chess D, et al: Thigh pain after cementless hip arthroplasty: annoyance or ill omen. J Bone Joint Surg Br 74:63, 1992 8. Engh CA, Massin P, Suthers KE: Roentgenographic assessment of the biologic fixation of poroussurfaced femoral components. Clin Orthop 257: 107, 1990 9. Haddad RJ Jr, Skalley TC, Cook SD, et al: Clinical and roentgenographic evaluation of noncemented porous-coated anatomic medullary locking (AML) and porous-coated anatomic (PCA) total hip arthroplasties. Clin Orthop 258:176, 1990 10. Harris WH: The problem is osteolysis. Clin Orthop 311:46, 1995 11. Smith RN, Harris WH: Increasing prevalence of femoral lysis in cementless total hip arthroplasty. J Arthroplasty 10:407, 1995 12. Xenos JS, Hopkinson WJ, Callaghan JJ, Heekin RD, et al: Osteolysis around an uncemented cobaltchrome total hip arthroplasty. Clin Orthop 317: 29, 1995 13. Engh CA Jr, Culpepper WJ II, Engh CA: Long-term results of use of the anatomic medullary locking prosthesis in total hip arthroplasty. J Bone Joint Surg Am 79:177, 1997 14. Kim YH, Kim JS, Cho SH: Primary total hip arthroplasty with the AML total hip prosthesis. Clin Orthop 360:147, 1999 15. Sakalkale DP, Eng K, Hozack WJ, Rothman RH: Minimum 10-year results of a tapered cementless hip replacement. Clin Orthop 362:138, 1999 16. Hellman EJ, Capello WN, Feinberg JR: Omnifit cementless total hip arthroplasty: a 10 year average F/U. Clin Orthop 364:164, 1999 17. McLaughlin JR, Lee KR: Total hip arthroplasty with an uncemented femoral component. J Bone Joint Surg Br 79:900, 1997

552 The Journal of Arthroplasty Vol. 19 No. 5 August 2004 18. Bourne RB, Rorabeck CH, Patterson JJ, Guerin JG: Tapered titanium cementless total hip replacements: a 10- to 13-year followup study. Clin Orthop 393: 112, 2001 19. Head WC, Emerson RH, Higgins LL: Twelve years with a primary press-fit titanium stem. Poster 400. Presented at the American Academy of Orthopaedic Surgeons, San Francisco, CA, March 2001 20. Meding JB, Keating EM, Ritter MA, Faris PM, et al: Minimum 10-year F/U of a straight stemmed, plasma-sprayed, titanium alloy, uncemented femoral component. Poster 396. Presented at the American Academy of Orthopaedic Surgeons, San Francisco, CA, March 2001 21. Froimson MI, Garino JP, Machenaud A, Vidalain JP: Minimum 10-year results of a tapered, titanium, hydroxyapatite coated hip system: an independent review. Poster 374. Presented at the American Academy of Orthopaedic Surgeons, San Francisco, CA, March 2001 22. Mauerhan DR, Mesa J, Gregory AM, Mokris JG: Integral porous femoral stem: 5 to 8 year follow-up study. J Arthroplasty 12:250, 1997

23. Dorr LD: Anatomic porous replacement hip arthroplasty: first 100 consecutive cases. Semin Arthroplasty 1:77, 1990 24. Brooker AF, Bowerman JW, Robinson RA, Riley LH: Ectopic ossification following total hip replacement: incidence and a method of classification. J Bone Joint Surg Am 55:1629, 1973 25. Gruen TA, McNeice GM, Amstutz HC: “Modes of failure” of cemented stem-type femoral components: a radiographic analysis of loosening. Clin Orthop 141:17, 1979 26. Dorr LD, Lewonowski K, Lucero M, Harris M, et al: Failure mechanisms of anatomic porous replacement I cementless total hip replacement. Clin Orthop 334: 157, 1997 27. Martell JH, Galante JO, Pierson RH, Jacobs JJ, et al: Clinical experience with primary cementless total hip arthroplasty. Chirurgia Degli Organi di Movimento 77:383, 1992 28. Mathur SK, Mont MA, McCutchen JW: Intraoperative custom press-fit and standard press-fit femoral components in total hip arthroplasty. A comparison of surgery, charges, and early complications. Am J Orthopedics 25:486, 1996