Clinical Outcome of a Fibermetal Taper Stem

Minimum 5-year Followup

Abstract

Tapered femoral components were designed to provide a cementless alternative in patients with narrow femoral canals. Major concerns with this stem include perioperative fracture and thigh pain. To ascertain the frequency of these concerns, we retrospectively reviewed 100 consecutive hips in 97 patients who underwent total hip arthroplasty with a tapered fibermetal femoral component. Patients were evaluated postoperatively using Harris hip scores and sequential radiographs. The average age of the patients was 56.5 years (range, 33.7-73.5 years). Minimum followup was 5 years (average, 6.2 years; range, 5-9 years). One patient died at 1.3 years and was excluded from the study. Six patients had incomplete perioperative fractures of the calcar without extension below the lesser trochanter. All fractures were recognized perioperatively and fixed with cerclage wires without further problems. Four patients underwent revision surgery, two for recurrent dislocation secondary to malpo-sitioning of the acetabular cup, one for a greater trochanteric fracture after a fall, and one for a fractured zirconium femoral head. At last followup, the mean Harris hip score was 98 (range, 86-100); two patients had mild anterior thigh pain. All femoral stems appeared to have bony integration radio-graphically. Tapered femoral stems provide a cementless alternative in patients with a narrow femoral canal without undue risk of perioperative fracture or thigh pain.

Level of Evidence: Level IV, therapeutic study. See the Guidelines for Authors for a complete description of levels of evidence.

The need for initial stability in uncemented total hip arthroplasty has led to development of larger implants with both metaphyseal and diaphyseal fit.^11,31 This design approach, however, may contribute to an increased perioperative fracture rate and may be one of the major causes of anterior thigh pain, longer term stress shielding and femoral bone loss.²²

The incidence of proximal femur fracture during stem insertion has also been higher with uncemented stems when compared with cemented femoral components, ranging from 3% to as high as 27.8%.^1,4,5,9,32 This is much higher than the 0.1% to 3.2% for cemented stems.^4,5,32 Among the reasons stated for these fractures are improper sizing of the canal and improper insertion of the stem,^4,5,21,27,32 female gender,^5,32 and an anterolateral approach.⁵ Many different tapered designs have been developed to address these issues.^3,6,9,10,30 Although many of these stems have had excellent clinical success, the puzzling issue of thigh pain along with the concern for intraoperative proximal femur fractures still remains.

Although the etiology of thigh pain remains uncertain in uncemented THA, the incidence ranges from 0.5% to 40%. ^{2,3,7,9,10,12-14,25,26,29} Several factors have been implicated, including instability of the stem, excessive stress transfer secondary to various stem properties such as stem size, material, and design, the amount of porous coating, and the quality of the host bone. Although the cause is most likely multifactorial, implant designs over the years have been altered to address these concerns.

We therefore defined the incidence of perioperative fracture rate and thigh pain with the use of one tapered femoral stem design.

MATERIALS AND METHODS

We retrospectively reviewed the occurrence of perioperative fractures and thigh pain in 97 selected patients with primary and secondary osteoarthritis who underwent implantation of the first 100 fibermetal taper stems used by the senior author (VMG) between 1997 and 2001. During the same time period, there were 428 total hip arthroplasties performed by the same surgeon, of which 164 had uncemented stems. These patients were selected for this stem based on the anatomy of their proximal femurs. We included only femurs classified as type A or B according to the criteria by Dorr et al.¹⁶ For this stem we excluded patients with inflammatory arthropathy or type C femurs.

One patient died at 1.3 years from causes unrelated to the implant and was excluded from the study. Eighty-three patients (86 hips) returned for clinical and radiographic followup at a minimum of 5 years (mean, 6.3 years; range, 5-9 years). Thirteen patients declined to return for last clinical followup but were contacted by phone at an average of 6.2 years from the time of surgery (range, 5.2-8.3 years). These latter patients were last seen for clinical and radiographic followup at an average of 4.1 years (range, 3.9-4.4 years).

The mean age of the patients was 56.5 years (range, 33.7-73.5 years). Eighty-three patients had a diagnosis of primary osteoarthritis. The remainder had secondary osteoarthritis from avascular necrosis or Crowe I developmental dysplasia of the hip (Table 1).

TABLE 1: Patient Demographics

The VERSYS fibermetal taper (FMT) stem (Zimmer, Warsaw, IA) is composed of a Tivanium (Ti-6Al-4V) core with a fibermetal layer proximally and a roughened, grit-blasted surface in the midportion of the stem. Distally, the stem is polished to discourage distal fixation and tapered to increase flexibility and avoid contact with cortical bone in narrow femurs. The fibermetal surface is intended to promote ingrowth, whereas the roughened surface promotes ongrowth.

The procedure was performed in a laminar airflow operating room, and the surgeons wore body-exhaust isolation suits. All procedures were performed through a standard posterolateral approach to the hip. Femoral stem sizes ranged from 10 mm to 19 mm in diameter. Canal preparation was performed with broaches of increasing size with the last broach selected based on the visual fill of the metaphysis. No power reamers were used during canal preparation. All stems were evaluated with an operative torque wrench to 120 Nm to assess for rotational stability. Femoral head sizes varied from 28 mm to 36 mm in diameter in 4-mm increments. Thirty-three of the femoral heads were zirconium ceramic and 67 were cobalt-chrome alloy. The material selected for the femoral head was based on the age of the patients, with patients younger than 55 years receiving a ceramic femoral head. All patients had a porous cup (Trilogy; Zimmer) with an ultra-high-molecular-weight polyethylene liner sterilized by gamma irradiation in nitrogen ranging in size from 48 mm to 66 mm. The fixation was augmented in 69 hips with one, two, or three screws. The ultra-high-molecular-weight polyethylene line was a minimum of 9.3 mm in thickness.

A prophylactic antibiotic (cefazolin) was administered intravenously (one dose preoperatively and for 48 hours postoperatively). All patients received prophylactic anticoagulation with warfarin to maintain an international normalized ratio around 2.0; wore thromboembolic hose; and were managed with sequential compression devices on both legs. All patients were monitored for deep venous thrombosis with a duplex ultrasound study on the third postoperative day. After the procedure, patients were kept 50% weightbearing with two crutches for 6 weeks after which they proceeded with progressive weightbearing. All patients were fully weightbearing by 8 weeks after surgery with no assistive devices.

A third party (PC) not involved in the care of the patients preoperatively evaluated all patients at 3 months and 6 months postoperatively, then at 2-year intervals thereafter. During each visit, a Harris hip score²⁴ (HSS) was determined and radiographs were obtained. As part of the HSS, patients were asked whether they experienced any thigh pain. If a positive response was obtained, the location and degree of thigh pain was then graded. Pain was graded as mild when it was occasional, did not interfere with daily activities, and did not necessitate medications; as moderate when it led to a modification of daily activities and the occasional use of nonsteroidal antiinflammatory medications; and as severe when it was constant, necessitated the use of narcotics, and interfered with daily activities.

All patients with femoral stems had immediate postoperative radiographs. Anteroposterior, lateral, and oblique radiographs of the hip and anteroposterior radiographs of the pelvis were obtained in a standard fashion preoperatively, postoperatively at 3 months and 6 months, and at 2-year intervals thereafter. All radiographs were evaluated using Hip Society guidelines.²⁸ Calcar remodeling, pedestal formation, femoral stem fixation, radiolucent lines, osteolysis, stem subsidence, and heterotopic ossification were evaluated according to published methods.^{8,15,18-20,23,33} All measurements were made by the senior author (VMG). We did not specifically address polyethylene wear in our report because the focus of our study was on the femoral stem.

RESULTS

We had six incomplete intraoperative femoral neck fractures. None extended below the lesser trochanter. Two fractures occurred in patients with a diagnosis of DDH (two of nine, 22.2%) and the remaining four occurred in patients with a diagnosis of osteoarthritis (four of 83, 4.8%). In three cases, the fracture occurred during the preparation of the canal as it was being broached. In the remaining three cases, the fracture was recognized during final impaction of the stem. The fracture occurred in two female patients and four male patients. In four patients, the stem diameter was 16 mm or larger. Four fractures occurred during the first 50 procedures and two occurred in the last 50. In each case, these fractures were treated with one to two cerclage wires placed below the level of the fractures after the broach or stem was removed and the stems were then reinserted.⁴ The patients were placed on partial weightbearing for a total of 6 weeks. There was no migration of these stems. No patients had any complications as a result of the fractures, and the mean HSS for these patients was 97 (range, 87-100).

No patient seen for followup had any evidence of functionally important anterior thigh pain; however, two patients reported mild anterior thigh pain and two patients had lateral trochanteric thigh pain. The mean preoperative HHS for the 83 patients (86 hips) who returned for followup was 58 (range, 30-85). It improved to 97 (range, 86-100) by 1 year postoperatively and 98 (range, 86-100) by the last followup. Only three patients had scores less than 90, and no patient had a score less than 80. At last followup, all stems had evidence of bony ingrowth, and none had subsided more than 3mm (Fig 1). Thirty-four patients had evidence of cortical hypertrophy and medullary sclerosis without a solid bone block at the tip of the stem (Fig 2). The 13 patients with 13 hips who declined to return for their last followup had a mean preoperative HSS of 56 (range, 43-71). This improved to 98 (range, 94-100) by 1 year postoperatively and 98 (range, 96-100) at their last clinical exam. When last contacted by telephone, no patient reported problems with their hip. No thigh pain or the use of external aids was experienced. The one patient (one hip) who died at 1.3 years had an HSS of 87 at the time of death.

Fig 1A: B. (A) Anteroposterior and (B) lateral radiographs 8 years after the initial operation. Trabecular contact is seen along the entire length of the implant.

Fig 2A: C. (A) Anteroposterior, (B) lateral, and (C) closeup view of the reactive bone formation seen at the tip of the stem.

One patient sustained a greater trochanter fracture after a fall 6 months after the operation. This patient was treated with a hook plate and cerclage wiring. At last followup, she was walking without walking aids and had a HSS of 86. There were three additional revision operations. Two patients underwent acetabular cup revisions for recurrent hip dislocation at 6 months and 5 years postoperatively, respectively. The last patient had a fracture of a zirconium femoral head 4 years after the index operation and had the femoral head replaced with a cobalt-chrome femoral head. At the time of revision, all stems were stable and well-fixed. The minimum HSS for these four patients at last followup was 96 (mean, 99; range, 96-100) and none had any further complications related to the hip replacement.

DISCUSSION

The purpose of the current study was to evaluate the incidence of perioperative femoral fracture and thigh pain with the use of an uncemented fibermetal stem. Among 100 consecutive patients having this stem implanted, six patients had an incomplete femoral neck fracture recognized and treated intraoperatively with no further complication, and two patients had nonlimiting mild anterior thigh pain.

Ours was a retrospective study and, as a result, is susceptible to recall bias. Patients contacted by telephone were asked to recall whether any problems occurred with their hips and were not clinically examined. We were unable to obtain complete radiographic evaluation for all patients, although all were contacted. The patients who did not have radiographic assessment denied being in pain and functioned without any external aids. The incidence of thigh pain was determined and graded as part of the HSS score. It was not asked separately or graded with a separate validated scale. This was a single-surgeon, selected patient population with osteoarthritis and type A or B femurs as described by Dorr et al.¹⁶ Finally, while this implant has performed well in this patient cohort, further study is necessary to evaluate how it would function in patients outside of our selection criteria.

Our fracture rate was 6% (six of 100). Fracture occurred in four men and two women. These fractures occurred during either preparation of the canal or during final impaction of the stem. Indeed, in four hips (67%), the stem diameter was 16 mm or larger. This may be the result of the tendency to attempt to achieve complete fill of the canal for initial stability, especially in patients with a very large femoral canal. We also experienced a learning curve similar to other authors²¹ with four fractures occurring in the first 50 patients and only two fractures occurring in the last 50. A recent assessment of the last 100 tapered stems performed by the senior author (VMG) between 2005 to 2006 found the incidence of fracture had decreased to 2% (unpublished data, VMG). Clearly, insertion of these stems requires an accurate understanding of the amount of force necessary for proper femoral canal preparation and insertion of the stem.

Cementless femoral implants have traditionally been associated with higher intraoperative fracture rates than cemented implants.^4,5,21,32 Berend et al⁵ reported a 6.4% incidence of femoral neck fracture in uncemented stems versus only 0.9% in cemented stems. They discussed several risk factors for fractures, including female gender and the use of an anterolateral approach. Experience with the anatomic medullary locking stem (DePuy, Warsaw, IA) has demonstrated an intraoperative fracture rate of 3%, including both proximal femoral neck fractures as well as fractures at the tip of the stem. In all cases, the fractures occurred either during preparation of the canal or during final impaction of the stem.³² Fitzgerald et al²¹ observed a fracture rate of 6.3% with the use of various cementless systems in both primary and revision surgeries. They attributed fractures to improper sizing of the implant in almost all cases. They reported a substantial learning curve with an 8% fracture rate during the first year of the study being reduced to 4% in the last year of the study. The fractures in this study were recognized intraoperatively and were fixed with one to two cerclage wires placed below the level of the fracture. All stems healed without any complications and without any migration of the stems. These findings are similar to that of other authors^4,5,32 and stress the importance of proper recognition of these fractures intraoperatively and their prompt treatment.

Our patients denied clinically important thigh pain although assessed by an individual not involved in the care of the patients. The incidence of thigh pain, however, has been higher in patients with cementless stems. There have been several potential causes for this thigh pain. Several authors^6,14 report a high rate of thigh pain in patients with radiographic evidence of unstable stems. One study¹⁴ found an incidence of 27% at 2 years with a porous-coated anatomic stem. The incidence was more than twice as high in patients with more than 2 mm of subsidence on postoperative radiographs.¹⁴ The incidence of thigh pain has been as high as 34% in stems with stable fibrous ingrowth compared with 8% to 10% in similarly designed stems with radiographic evidence of bony ingrowth.^17,18 All of the stems in our study had radiographic evidence of bony ingrowth.

Thirty-four patients in our study had evidence of cortical hypertrophy and medullary sclerosis without a solid bone block at the tip of the stem. The importance of this finding and its relationship to thigh pain is unknown. It may indicate reactive bone formation at the tip of the stem secondary to stress transfer or distal stem motion. Another study²⁶ correlated this finding to increased uptake on technetium bone scans and a possible correlation with thigh pain.

Stem design reportedly contributes to thigh pain. In theory, an implant made of a less stiff material such as titanium would result in less thigh pain than one made of cobalt-chromium alloy.^9,22 The material of the stem (cobalt-chromium alloy versus titanium) contributes to the stiffness of the implant and the amount of stress seen at the tip of the stem. The femoral stem we evaluated was composed of a titanium alloy core, which has a lower stiffness than stems with a cobalt-chromium alloy core. In addition, the tapered design has two advantages. First, the stem size is reduced secondary to the distal taper. As a result, the overall stiffness is lower than a nontapered stem of the same size. Second, a tapered design allows the stem to wedge itself into a stable position. There were four stems with evidence of subsidence measuring less than 2 mm at 1-year followup. None of these stems, however, demonstrated any further migration on subsequent examinations. The tapered design theoretically could prevent distal migration of the stem and may have provided an added layer of stability for early bony ingrowth.

Our minimum of 5 years of followup demonstrates a high rate of clinical success of the VERSYS fibermetal taper femoral component with no evidence of thigh pain but with a small but important intraoperative fracture rate. Longer followup is necessary to evaluate the long-term performance of this implant.

Acknowledgments

We thank Patricia Conroy, RN, and Rebecca Thomas for their help in gathering data and preparing the manuscript.