Abstract
Background: A multicenter retrospective study was conducted to determine the clinical and radiographic results of primary total hip arthroplasty with insertion of the S-ROM modular femoral stem without cement in a series of patients who had been followed for four to seven years. Four centers participated in the study, with one contributing surgeon at each center.Methods: Two hundred and eight consecutive patients who had a total hip arthroplasty with implantation of the S-ROM femoral prosthesis at one of the four centers during the study period were identified. Twenty-nine patients were lost to follow-up or had incomplete radiographic data, and twenty patients died from causes unrelated to the index arthroplasty. The remaining 159 patients formed the basis of this study. Sixteen of these patients had a bilateral procedure, resulting in 175 hips with complete clinical and radiographic data. The average age of the patients at the time of the index operation was fifty-nine years (range, twenty-two to ninety-three years). The duration of clinical follow-up averaged 5.3 years (range, four to 7.8 years), and the duration of radiographic follow-up averaged 4.9 years (range, four to 7.3 years).Results: One patient (0.6 percent) had a failed femoral component, which was evidenced by progressive subsidence and lack of bone ingrowth. In addition, two patients (1 percent) had a revision of the acetabular component.The average Harris hip score increased from 35 points (range, 10 to 76 points) preoperatively to 91 points (range, 52 to 100 points) at the most recent follow-up examination. The radiographic evaluation revealed that 172 hips (98 percent) had stable bone ingrowth, two hips (1 percent) had stable fibrous ingrowth, and one hip (0.6 percent) had unstable fibrous ingrowth.Periprosthetic osteolytic lesions were noted in twelve hips (7 percent). The lesions were observed in the femur in eight hips, in the acetabulum in two hips, and in both the femur and the acetabulum in two hips. All femoral osteolytic lesions were localized within the greater trochanter or the proximal-medial portion of the femoral neck. No osteolytic lesions were evident distal to the stem-sleeve junction.Conclusions: Use of the modular S-ROM femoral prosthesis yielded excellent intermediate-term outcomes with respect to standard radiographic and clinical criteria. The issue regarding the theoretical increase in the rate of osteolysis due to metal debris generated at the modular femoral stem-sleeve junction was specifically addressed. We found that the rate of osteolysis in this series was not notably higher than that in other series reported in the orthopaedic literature. Although many possible factors may influence the rate of osteolysis in total hip arthroplasty, this finding suggests that the potential increase in osteolysis theoretically associated with this modular femoral implant was not observed at intermediate-term follow-up. Although longer follow-up is warranted so that the potential for osteolysis can be evaluated fully, no osteolytic lesions were evident distal to the stem-sleeve interface at the time of intermediate-term follow-up. This finding suggests that there is a circumferential seal at the modular junction of the stem that prevents the distal egress of wear debris.
One of the primary goals of total hip arthroplasty without cement is to obtain intimate contact between the prosthesis and the host bone to provide both short and long-term stability of the implant. The wide range of sizes and shapes of the proximal part of the femur has made this goal difficult to achieve. Noble et al.16 recorded measurements from 200 cadaveric femora and found no constant relationship between the size and the shape of the proximal femoral metaphysis and those of the diaphysis. A variety of methods have been used in an attempt to match femoral implants with the bewildering array of shapes of the human femoral canal. Expanded ranges of implant sizes, in one-millimeter increments, with an assortment of proximal, or body, lengths and distal, or stem, lengths have been utilized. Custom implants, including those that can be customized intraoperatively, have been used to match the anatomy of an individual patient. During the past decade, however, the use of modular implants has become more widespread. In addition to the modular head-neck junction, which has been used for years, implants such as the Infinity (Dow Corning Wright, Memphis, Tennessee), the RMHS (Smith-Nephew Richards, Memphis, Tennessee), and the S-ROM (DePuy, a Johnson and Johnson Company, Warsaw, Indiana) allow the surgeon to independently match the anatomy of the metaphyseal and diaphyseal regions of the femoral canal of an individual patient. The use of modular systems, however, raises issues about the integrity of the modular junction as well as the potential generation of metallic wear debris due to fretting at the modular interface. Mechanical testing of these modular implants has demonstrated the production of several hundred thousand to several million metallic wear particles, ranging from one to three micrometers in size, after ten to twenty million load cycles2. It is unclear if this quantity of wear particles causes clinically important macrophage-mediated osteolysis. In addition, little clinical information with regard to the intermediate to long-term performance of modular implants is available to help to resolve the concerns about the potential deleterious effects of modularity.
The purpose of the present study was to report the clinical and radiographic results of primary total hip arthroplasty with insertion of the modular S-ROM femoral stem without cement in a series of patients who were followed for four to seven years.
*One or more of the authors has received or will receive benefits for personal or professional use from a commercial party related directly or indirectly to the subject of this article. In addition, benefits have been or will be directed to a research fund, foundation, educational institution, or other nonprofit organization with which one or more of the authors is associated. Funds were received in total or partial support of the research or clinical study presented in this article. The funding source was Joint Medical Products, Stamford, Connecticut (now DePuy, a Johnson and Johnson Company, Warsaw, Indiana).
†Southern Joint Replacement Institute, Baptist Medical Plaza, 2021 Church Street, Suite 104, Nashville, Tennessee 37203. E-mail address: research@srji.com.
‡414 Navarro, Suite 1128, San Antonio, Texas 78205.
§2400 Patterson Street, 300, Nashville, Tennessee 37203.
#5920 Forest Park Road, 600, Dallas, Texas 75235.
**766 Lakeland Drive, Suite B, Jackson, Mississippi 39216.
††Department of Orthopaedics, West Virginia University, P.O. Box 9196, Morgantown, West Virginia 26506.
A multicenter retrospective study was conducted to determine the clinical and radiographic results after primary total hip arthroplasty with use of the S-ROM femoral prosthesis in a series of patients who had been followed for a minimum of four years. Four centers participated in the study, with one contributing surgeon (L. W. T., J. C. B., R. E. J., or G. T. V.) at each center. The indications for implantation of the S-ROM prosthesis varied among the participating centers. One center used the S-ROM prosthesis for most patients who had a primary total hip arthroplasty, whereas other centers reserved this implant for use in younger patients or in patients who had type-A bone stock according to the classification system of Dorr et al.8. Two centers were primarily community-based practices, and two centers were located at academic institutions. This variation resulted in a diverse patient cohort for the study population.
The study was designed to have the clinical data evaluated independently by two surgeons (M. J. C. and D. K. DeB.) and the radiographic analysis performed independently by a reviewer (T. A. G.) who had expertise in the evaluation of radiographic outcomes after total hip arthroplasty. All of the patients in the study had had implantation of an S-ROM three-piece femoral component made of titanium alloy (except for the head ball, which is cobalt-chromium) (DePuy, a Johnson and Johnson Company) without cement as part of a primary total hip arthroplasty (Fig. 1). Although the sizes of the components varied according to the anatomy of the individual patients, no changes in the design of the implant or in the manufacturing process were made during the time-period of the study. The modular femoral component consisted of a head ball, a fluted stem for torsional stability with a coronal clothespin-like slot to reduce bending stiffness, and a circumferentially porous-coated sleeve.
A variety of acetabular implants were used in the present study; however, all of them consisted of a hemispherical cup made of titanium-alloy and all were fixed without cement. Although any failure of an acetabular component was reported in this review, a detailed analysis of the acetabular results was not the objective of the study.
Between January 1988 and September 1991, 225 S-ROM stems were implanted in 208 patients. All of the procedures were performed by the contributing surgeons at their respective institutions, and all of the hips in which an S-ROM prosthesis was implanted at each center during this time-period were included in the study.
Twenty-nine patients (thirty hips) who did not return for follow-up examinations for a minimum of four years after the operation were considered to be lost to follow-up. A review of the office notes for these patients did not reveal any evidence of overt or pending clinical or radiographic failure at the latest follow-up examination. Nine of these patients (ten hips) were contacted by telephone for an interview, and each reported that the hip was functioning well.
Twenty patients (twenty hips) died for reasons unrelated to the arthroplasty. A review of the office notes for these patients confirmed that all of them had had good function of the hips at the latest clinical examination. These patients had ranged in age from forty-eight to 100 years old (average, seventy-four years old) at the time of death. None of the deaths occurred in the perioperative period.
The remaining 175 hips (78 percent of the original 225 hips) in 159 patients had been followed clinically and radiographically for a minimum of four years, and they form the basis of this study.
The average age of the patients at the time of the index operation was fifty-nine years (range, twenty-two to ninety-three years); thirty-five (20 percent) of the procedures were performed in patients who were less than forty-five years old (Table I). There were ninety women (57 percent) and sixty-nine men (43 percent). There were ninety-one reconstructions (52 percent) of the right hip and eighty-four (48 percent) of the left hip. The average weight of the patients at the time of the index arthroplasty was seventy-eight kilograms (range, forty-eight to 137 kilograms).
The underlying diagnoses leading to the total hip arthroplasty were osteoarthritis (120 hips), avascular necrosis (twenty), rheumatoid arthritis (seventeen), congenital dysplasia (eight), posttraumatic arthritis (three), Legg-Calvé-Perthes disease (two), ankylosing spondylitis (two), hemophilia (one), scleroderma (one), and diastrophic dwarfism (one). Eight of the patients who had rheumatoid arthritis had notable involvement of other joints as well.
The operative approach was anterolateral in 143 hips, posterior in thirty-one hips, and transtrochanteric in one hip. Nine trochanteric osteotomies were performed. Eight of them were so-called trochanteric slides in patients in whom an anterolateral approach was used, and one was a formal trochanteric osteotomy during a transtrochanteric approach.
Thirteen patients (8 percent) had radiographic evidence of osteoporosis or a type-C femur according to the system of Dorr et al.8, which was determined by a cortical index of less than 0.40 preoperatively. The femora in the remaining 146 patients (92 percent) were either type A or type B, according to the system of Dorr et al.
The size of the S-ROM femoral component reflects the diameter of the metaphyseal and the diaphyseal region of the stem in millimeters. In this study cohort, three stems were fourteen by nine millimeters, thirty were sixteen by eleven millimeters, forty-eight were eighteen by thirteen millimeters, sixty-five were twenty by fifteen millimeters, twenty-eight were twenty-two by seventeen millimeters, and one was twenty-four by nineteen millimeters. A cobalt-chromium femoral head was used in all hips; the diameter of the head was twenty-six millimeters in two hips, twenty-eight millimeters in forty-two hips, and thirty-two millimeters in 131 hips.
The clinical evaluations were performed by the contributing surgeons at their respective institutions according to a modification of the method described by Harris13 and with use of the Total Hip Arthroplasty Outcome Evaluation Form of The American Academy of Orthopaedic Surgeons15. In addition, forms concerning the operative procedure, details about the implants, medical history, and intraoperative and postoperative complications were completed and a self-administered questionnaire was filled out by the patient. All patient evaluations were based on the initial preoperative visit and the most recent postoperative examination, with documentation of pain, activity level, ability to perform activities of daily living, use of walking aids, gait, and range of motion of the hip. This information was forwarded to the independent observer's institution, where it was entered into a computerized relational database for tabulation.
The radiographic analysis was performed by an independent observer (T. A. G.), who was blinded with respect to the clinical data. Preoperative and postoperative sequential anteroposterior radiographs of the pelvis and anteroposterior and lateral radiographs of the hip were reviewed at each participating institution, with the final analysis and interpretation of data conducted by the independent observer. The radiographs were evaluated with respect to the zones around the femoral component described by Gruen et al.12, and the cortical index, the amount of filling of the distal part of the canal by the stem, the orientation of the femoral component, the stability of the implant, the degree of subsidence, and the extent of bone-implant demarcation were documented. The cortical index is defined as the ratio of the width of the femoral cortex to the total width of the femur measured ten centimeters distal to the lesser trochanter on an anteroposterior radiograph of the hip. The orientation of the component was classified as valgus, slight valgus, neutral, slight varus, or varus. Slight varus or slight valgus alignment was used to describe a femoral stem with less than 5 degrees of malalignment with respect to the neutral axis of the femoral canal. Varus or valgus alignment was used to describe a femoral stem that was oriented 5 degrees or more beyond neutral.
Evidence of adaptive bone-remodeling, cancellous metaphyseal hypertrophy, so-called spot-welding along the sleeve, and osteolytic lesions were included in the analysis. Periprosthetic osteolysis was defined as the radiographic appearance of a focal area of bone resorption evidenced as a cystic lesion that was not linear and that was at least five millimeters wide. These lesions were classified according to the seven zones around the femoral implant-bone interface described by Gruen et al.12, and they were identified as areas of either endosteal, intracortical, or cancellous destruction of bone.
The radiolucency associated with an osteolytic lesion was measured for maximum length and width, and the presence of a lesion was recorded if either the length or the width was five millimeters or more. All measurements were made with use of a transparent ruler positioned directly on the radiograph. The area of the radiolucency was calculated on the basis of a semi-ellipse, with a correction factor for magnification.
Clinical Results
The duration of clinical follow-up for the study cohort averaged 5.3 years (range, four to 7.8 years). The average preoperative score for pain was 12.7 points (range, 0 to 30 points). Preoperatively, pain was reported as mild in five hips, moderate in fifty-nine hips, severe in ninety hips, and disabling in three hips (Fig. 2). No data were available for eighteen hips. The average postoperative score for pain improved to 41.5 points (range, 20 to 44 points). Postoperatively, 164 hips (94 percent) were slightly or not painful. Eleven patients (6 percent; eleven hips) had postoperative pain in the thigh. Ten of the patients rated the pain in the thigh as mild, with no effect on their usual activity level, whereas one patient had moderate pain that affected his activity level.
Preoperatively, three hips were in patients who had no limp, twelve hips were in patients who had a slight limp, seventy-five hips were in patients who had a moderate limp, and sixty-six hips were in patients who had a severe limp. No data were available for nineteen hips. Postoperatively, 117 hips were in patients who had no limp, twenty-six hips were in patients who had a slight limp, twenty-six hips were in patients who had a moderate limp, and six hips were in patients who had a severe limp. Preoperatively, two hips were in patients who were bedridden, 124 hips were in patients who reported a sedentary or semi-sedentary activity level, eighteen hips were in patients who were able to perform light labor, and four hips were in patients who were able to perform moderate labor. No data on the preoperative activity level were available for twenty-seven hips. Postoperatively, none of the patients were bedridden. Fifty-nine hips were in patients who were either sedentary or semi-sedentary, seventy hips were in patients who were able to perform light labor, twenty-seven hips were in patients who were able to perform moderate labor, and nineteen hips were in patients who were able to perform strenuous labor.
Preoperatively, 101 (64 percent) of the 157 hips for which data were available were in patients who needed external support for walking (ninety-six hips) or who were unable to walk (five hips). Twelve hips were in patients who needed a walker, six hips were in patients who needed two crutches, five hips were in patients who needed one crutch, sixty-nine hips were in patients who used one cane full-time, and four hips were in patients who needed one cane for long walks. Fifty-six hips were in patients who needed no external support, and the data for eighteen hips were incomplete. Postoperatively, 135 hips (77 percent) were in patients who needed no external support for walking. Fourteen hips were in patients who needed one cane for long walks, eighteen hips were in patients who used one cane full-time, two hips were in patients who needed one crutch, and one hip was in a patient who needed two canes. Four hips were in patients who needed a walker because of advanced rheumatoid arthritis with involvement of multiple joints. The data were incomplete for one patient.
Postoperatively, the average range of motion of the hip improved in all planes. The average preoperative range of motion was 92 degrees of flexion, 20 degrees of abduction, 17 degrees of adduction, 20 degrees of external rotation, and 9 degrees of internal rotation. The average postoperative range of motion increased to 109 degrees of flexion, 35 degrees of abduction, 28 degrees of adduction, 34 degrees of external rotation, and 21 degrees of internal rotation.
The preoperative Harris hip score13 for this study cohort averaged 35 points (range, 10 to 76 points) (Fig. 3). At the time of the most recent follow-up examination, the average Harris hip score was 91 points (range, 52 to 100 points). Four patients had a postoperative Harris hip score of less than 60 points. Two of them had rheumatoid arthritis with involvement of multiple joints, and one had had a lumbar arthrodesis and had residual pain in the lower extremities and functional limitations related to low-back pain. The functional deficits in these three patients was not directly attributable to the index arthroplasty. The hip arthroplasty in the remaining patient, who had diastrophic dwarfism, was considered both a clinical failure, with symptoms related to the procedure, and a radiographic failure, with documentation of progressive subsidence of the femoral implant and lack of bone ingrowth.
Radiographic Results
The duration of radiographic follow-up averaged 4.9 years (range, four to 7.3 years). One hundred and seventy-two hips (98 percent) demonstrated stable bone ingrowth according to the criteria of Engh et al.10. Two stems were found to have stable fibrous ingrowth radiographically, and one stem had unstable fibrous ingrowth. Nonprogressive subsidence (three millimeters or less) of two stems was observed. Only one stem demonstrated progressive subsidence, and it was considered a clinical and radiographic failure. Evidence of radiolucency at the bone-component interface was found in 101 hips (58 percent). The radiolucency primarily involved either one zone of Gruen et al.12 (seventy-five hips; 43 percent) or two zones (seventeen hips; 10 percent). Of the 101 hips, ninety had radolucency in zone 4 (the tip of the stem), where the stem is fluted, slotted, and polished. Seventy-four stems had no radiolucent areas.
Eighty-seven of the femoral stems were in a neutral position, fifty-eight were in a slightly varus position, eight were in a varus position, and ten were in a slightly valgus position. No stems were in a valgus position. (No data were available for twelve hips.) The stem filled less than 80 percent of the distal part of the canal in two hips, between 80 and 89 percent in fifteen hips, between 90 and 94 percent in forty-nine hips, between 95 and 99 percent in nine hips, and 100 percent in eighty-seven hips. No data were available for thirteen hips. So-called corticocancellization of the femoral cortex or decreased radiodensity, indicative of adaptive bone-remodeling, was present in 138 hips (79 percent) and involved only a single zone in 122 hips. The proximal-medial part of the neck (zone 7) was involved in most (136) of the 138 hips. Cortical thinning was evident in sixty-eight hips (39 percent) (Fig. 4), and the greater trochanter (zone 1) was involved in sixty of them. New endosteal bone formation was evident in fifty-seven hips (33 percent). Cancellous (metaphyseal) hypertrophy was noted in fifty-four hips (31 percent). Evidence of stable bone ingrowth as demonstrated by spot welds adjacent to the sleeve was noted in eighty-seven hips (50 percent). Spot welds were found at the lateral-distal corner of the sleeve in thirty-four hips (39 percent), at the medial-distal corner of the sleeve in fifteen hips (17 percent), and at both distal corners in thirty-eight hips (44 percent) (Fig. 5).
Periprosthetic osteolytic lesions were evident radiographically in twelve hips (ten patients). The lesions were seen in the femur in eight hips, in the acetabulum in two hips, and in both the femur and the acetabulum in two hips. All femoral osteolytic lesions were localized in the proximal part of the femur (zone 1 [proximal-lateral] in seven hips and zone 7 [proximal-medial] in three hips). In six of the ten hips that had femoral osteolysis, the lesions did not actually occur at the bone-prosthesis interface. Rather, they occurred in the greater trochanter superior to the proximal-lateral aspect of the S-ROM sleeve. In all ten hips, the femoral osteolytic lesions were in only one zone of Gruen et al.12; however, one hip had a large lesion that extended from the bone-prosthesis interface of zone 1 into the proximal aspect of the greater trochanter. The average size of the femoral lesions was 102.6 square millimeters (range, 19.6 to 298.3 square millimeters). No osteolytic lesions were evident distal to the sleeve-stem modular junction.
Four hips had acetabular osteolytic lesions: three hips had an isolated lesion, and one hip had two lesions. The femoral head was thirty-two millimeters in diameter in three hips and twenty-eight millimeters in the fourth. The five acetabular lesions in the four hips ranged in size from 35.3 to 274.8 square millimeters.
Two acetabular revisions were performed because of acetabular osteolysis. Both patients had a thirty-two-millimeter femoral head and evidence of polyethylene wear on plain radiographs. In addition, one hip had a trochanteric avulsion associated with an osteolytic lesion that had expanded into the greater trochanter. A twenty-eight-millimeter femoral head had been used in this hip.
Associations Between Clinical and Radiographic Outcomes and Demographic and Technical Factors
The clinical and radiographic outcomes were examined in relation to the age of the patient, the cortical index, the filling of the distal part of the canal by the stem, and the orientation of the femoral component in order to identify any association between the patient outcomes and the demographic data and the technical aspects of the operation (Table II).
Thirty-five hips were in patients who were less than forty-five years old at the time of the index procedure, fifty-eight hips were in patients who were between forty-five and sixty-four years old, fifty-eight hips were in patients who were between sixty-five and seventy-five years old, and twenty-four hips were in patients who were more than seventy-five years old. The average postoperative Harris hip score13 for each age-group was 89.1, 92.2, 91.5, and 88.4 points, respectively. Spot welds were observed in eighteen, twenty-nine, twenty-three, and twelve hips, respectively. Corticocancellization was noted in nineteen, forty-six, thirty-six, and seventeen hips, respectively (Table II).
The cortical index was less than 0.40 for thirteen hips, between 0.40 and 0.55 for ninety hips, and greater than 0.55 for fifty-eight hips; the average postoperative Harris hip score for each group was 89.5, 90.0, and 93.4 points, respectively. (The cortical index could not be calculated for fourteen hips because the data were incomplete.) Spot welds were observed in four, forty-nine, and twenty-nine hips, respectively, and corticocancellization was noted in ten, seventy-four, and forty-five hips, respectively (Table II).
The canal fill was less than 80 percent in two hips, between 80 and 89 percent in fifteen hips, between 90 and 94 percent in forty-nine hips, and between 95 and 100 percent in ninety-six hips; the average postoperative Harris hip score for each group was 96.5, 89.4, 89.1, and 92.3 points, respectively. (No data were available for thirteen hips.) Spot welds were observed in one hip, nine hips, twenty-two hips, and fifty-one hips, respectively. Corticocancellization was noted in one hip, eleven hips, forty-four hips, and seventy-nine hips, respectively (Table II).
Eighty-seven of the femoral components were in neutral position, fifty-eight were in slight varus, eight were in varus, and ten were in slight valgus; the average postoperative Harris hip score was 92.2, 90.1, 98.8, and 85.1 points, respectively. (No components were implanted in a valgus orientation, and twelve hips had incomplete data with regard to orientation.) Spot welds were observed in forty-two, twenty-nine, seven, and nine hips, respectively. Corticocancellization was noted in seventy, forty-eight, seven, and ten hips, respectively (Table II).
Complications
Intraoperative complications occurred in less than 2 percent (three) of the hips and included a crack in the calcar, a sciatic nerve injury, and a pelvic vascular injury in one hip each. Postoperative complications included dislocation in three hips (2 percent), one of which had recurrent dislocation that necessitated revision to a constrained acetabular insert. One hip had a periprosthetic femoral fracture, which necessitated revision to a prosthesis with a longer stem. A trochanteric fracture related to an osteolytic lesion was documented in one hip. Two hips had revision of the acetabular component because of acetabular osteolysis. The femoral component was noted to be well fixed and was not revised in either of these hips. No postoperative infections were reported for this series of patients.
Overall, one femoral component was revised because of mechanical loosening. Radiographically, this instability was manifested by progressive subsidence of the femoral stem. Two hips had reactive lines between the sleeve and the host bone without subsidence, which represented stable fibrous ingrowth, but these two hips were not considered clinical failures. Overall, three hips (2 percent) in this series failed when a revision of the acetabular or femoral component was used as the end point. Two acetabular components were revised, and one femoral component was revised. Additionally, there was one reoperation for exchange to a constrained acetabular insert. Thus, the rate of failure of the femoral components was 0.6 percent.
The primary focus of the present study was to report the intermediate-term clinical and radiographic results after primary total hip arthroplasty with use of a modular S-ROM femoral component. We specifically addressed the issue of whether fretting and generation of metallic wear debris from the modular sleeve-stem junction would result in a substantial rate of osteolysis compared with that seen when less modular implants were used.
Osteolytic lesions were observed in twelve hips (7 percent) in our series. This finding compares favorably with those in other series of press-fit femoral implants. Xenos et al.19 reported a rate of osteolysis of 15 percent in ninety-two patients (100 hips) who had been followed for a minimum of seven years after a primary total hip arthroplasty with insertion of a femoral stem with a proximal, circumferential porous coating. Engh et al.9, in a study of 393 hips, reported a notably higher rate of osteolysis of 41 percent after an average duration of follow-up of seven years in a subset of seventy-four hips that had been treated with an extensively porous-coated femoral stem. Likewise, the rate of osteolysis in the series of D'Antonio et al.7 was 43 percent (ninety-six of 224 hips) at an average of six years after total hip arthroplasty performed with insertion of a proximally hydroxyapatite-coated femoral stem. Furthermore, D'Antonio et al. reported that significantly higher rates of osteolysis were found to be associated with an age of less than fifty years (p = 0.009), use of a thirty-two-millimeter femoral head (p = 0.0008), and use of a polyethylene liner with a thickness of less than eight millimeters (p = 0.003). The prevalence of osteolysis in our series is particularly important, as the patients were relatively young and a thirty-two-millimeter femoral head had been used in most hips (131; 75 percent). Many factors influence the rate of osteolysis, and it is impossible to directly compare series of patients that are not matched for age, gender, weight, or diagnosis. Although a longer duration of follow-up is still needed, the low rate of osteolysis in our series suggests that the theoretical concern that metallic wear debris from this modular femoral stem greatly increases the rate of osteolysis is not valid.
There has been speculation that polyethylene wear debris traverses the sleeve-stem interface and causes endosteal osteolysis along the femoral stem. This phenomenon was not observed in the present series. None of the hips had distal osteolysis. The absence of any osteolytic lesions or endosteal scalloping distal to the proximal sleeve suggests that an intact implant-bone interface between the femoral sleeve and the surrounding host bone had been attained. The circumferential extent of the porous coating on the modular sleeve appears to act as a proximal seal, preventing the distal egress of polyethylene wear particles generated from the polyethylene liner.
The clinical results of the present series are similar to those of other reported studies of primary total hip arthroplasty5-8,14. The average Harris hip score in our series was 91 points, with a good or excellent result in 91 percent (159) of 175 hips. The functional results were similar to those of other series with intermediate-term and long-term follow-up after insertion of a variety of femoral components. Heekin et al.14, in a study of the results of 100 consecutive primary hip arthroplasties with use of a PCA (Porous-Coated Anatomic) stem (Howmedica, East Rutherford, New Jersey), reported an average Harris hip score of 92 points after an average duration of follow-up of five years. In a study of 201 patients (224 hips) who were followed for an average of six years after insertion of a proximally hydroxyapatite-coated stem (Osteonics, Allendale, New Jersey), D'Antonio et al.7 reported an average Harris hip score of 95 points. Our results are also comparable with those of Schmalzried and Harris17, who reported an average Harris hip score of 93 points for ninety-seven hips that had been followed for a minimum of five years after fixation of the femoral implant with a so-called third-generation cementing technique. Again, it is impossible to directly compare functional results of patient cohorts that were not matched; however, the functional results in our series, in which a modular femoral implant had been used, do not appear to be appreciably different from those of series in which a variety of femoral implants were used.
Pain in the thigh after total hip arthroplasty with insertion of the femoral stem without cement has been a major concern, and the prevalence has been reported to be as high as 30 percent in series ranging from twenty-eight to 393 hips1-4,9,18. In a report of the results after total hip arthroplasty with use of an S-ROM stem, Cameron4 observed that postoperative pain in the thigh was reduced from two of five patients to 5 percent (two of forty-three patients) after the design was changed to include a distal coronal slot. Although the orientation of the slot, which is parallel to the neck, varies with the degree of anteversion of the component, this slot helps to reduce the bending stiffness of the femoral stem. The prevalence of pain in the thigh in our cohort was 7 percent (twelve hips), which is consistent with Cameron's findings4 and is substantially less than the rate of 14 percent (of eighty-nine patients) after use of a fully porous-coated anatomic (PCA) medullary locking prosthesis reported by Engh et al.9 and the rate of 16 percent (of twenty-seven hips after seven years of follow-up) to 18 percent (of forty-six hips after six years of follow-up) reported by Heekin et al.14 in a study of a PCA stem. The rate in the present series is comparable with the rate of 2 percent (of 201 patients) reported by D'Antonio et al.7 in a study of a proximally hydroxyapatite-coated stem.
The modularity of the prosthesis at the stem-sleeve junction allows the surgeon to select the anteversion of the femoral component independent of the sleeve position that best fills the proximal part of the femoral canal. The surgeon therefore may elect to increase or decrease the anteversion of the femoral component, to match the native femoral anteversion, independent of the position dictated by the best fit and fill in the femoral canal. This control of the anteversion of the stem theoretically should decrease the rate of dislocation by allowing the surgeon to adjust the version for the greatest stable arc of motion of the hip. Rates of dislocation after primary total hip arthroplasty have ranged from 0 to 9 percent in series of forty-eight to 330 hips in recent reports5,7,14,18. The rate of dislocation in our series was 2 percent (three hips). All of the dislocations occurred in patients who had had an anterolateral approach to the hip.
One of the interesting consistent radiographic findings in the present series was the evidence of adaptive bone-remodeling. Femoral corticocancellization or decreased radiodensity was evident in 79 percent (138) of the hips, and it was noted in a single proximal-medial zone (zone 7 according to the system of Gruen et al.12) in 136 of those hips. Cortical thinning was also noted in sixty-eight hips (39 percent). Fifty-six (82 percent) of the sixty-eight hips had involvement of only one zone, four hips (6 percent) had involvement of two zones, six hips (9 percent) had involvement of three zones, and two hips (3 percent) demonstrated cortical thinning in four zones. Cortical thinning involving the greater trochanter (zone 1) was present in sixty (88 percent) of the sixty-eight hips. Ten hips demonstrated cortical thinning in zone 2; one hip, in zone 3; no hip, in zone 4; two hips, in zone 5; nine hips, in zone 6; and eight hips, in zone 7. These radiographic data suggest that some degree of stress-shielding occurs in the proximal part of the femur in association with the S-ROM prosthesis. These findings are similar to the radiographic changes observed after the use of proximally hydroxyapatite-coated femoral stems7.
The major radiographic signs of osseointegration according to the criteria of Engh et al.11 include the absence of complete reactive lines adjacent to the ingrowth or ongrowth surface of an implant and the presence of spot welds of endosteal new bone in contact with the ongrowth surface. These spot welds were observed in eighty-seven hips (50 percent) in our series, suggesting that load transmission does occur proximally between the sleeve and the host bone. With the numbers available, we could not detect a difference in the presence of spot welds with respect to the age of the patients, the degree of osteoporosis, or the filling of the distal part of the canal by the stem (Table II). Furthermore, we could not detect a difference in the presence of spot welds with respect to neutral or slight varus orientation of the femoral components; however, implants that were placed in a varus or slight valgus position were observed to have a higher rate of spot welds (seven of eight hips and nine of ten hips, respectively). This finding may represent a sampling error because of the small number of patients in each group or it may demonstrate the response of the proximal part of the femur to greater stresses associated with implants that are placed in an off-axis orientation.
Engh et al.11 stated that another major radiographic sign of a stable implant was the absence of migration. According to this criterion, only one hip had radiographic instability in our series. Therefore, the overall prevalence of aseptic loosening in our series was 0.6 percent (one hip) after an average of 4.9 years of follow-up, which is comparable with the rate of 1 percent (two of 224 hips) described by D'Antonio et al.7 and is notably lower than the rate of 5 percent (five of 100 hips) reported by Heekin et al.14 in a study of a PCA stem with a similar duration of follow-up.
One unusual aspect of the present multicenter study was the diverse patient population. Although two centers reserved this femoral implant for use in younger patients with healthy bone stock, the other two centers used it independent of the age of the patient. Thus, there was a wide range of ages (twenty-two to ninety-three years) in the study cohort. However, the average Harris hip score and the degree of bone-remodeling were not notably different between any age-groups. Similarly, with the numbers available, no differences were detected in the clinical outcomes with regard to the degree of osteoporosis (the cortical index) at the time of the index arthroplasty, the percentage of fill of the distal aspect of the canal, or the orientation of the femoral component. Although these are results after intermediate-term follow-up, they suggest that this implant works very well in patients regardless of age or degree of osteoporosis. Our impression is that the tapered design of the proximal sleeve and the intimate contact between the sleeve and the host bone creates an optimum environment for bone ingrowth and long-term stability of the implant.
In conclusion, the S-ROM prosthesis provides excellent intermediate-term outcomes with respect to standard clinical and radiographic criteria. Importantly, the rate of osteolysis in this series was comparable with or lower than those of other reported studies of primary total hip arthroplasty. A longer duration of follow-up of these patients is necessary to determine whether the addition of metallic wear debris from the modular femoral stem will greatly influence the rate of osteolysis. However, these intermediate-term findings do not demonstrate a notable difference between the modular S-ROM femoral stem and other, nonmodular femoral prostheses. Additionally, there were no failures related to the modular junction in this series and no evidence of osteolysis distal to the modular stem-sleeve junction. The radiographic changes suggest a high degree of fixation at the distal sleeve with predictable and consistent remodeling within the femur.
Amstutz, H. C.; Nasser, S.; and Kabo, J. M.: Preliminary results of an off-the-shelf press-fit stem. The anthropometric total hip femoral component using exact-fit principles. Clin. Orthop.,249: 60-72, 1989.24960
1989
[PubMed]
Bobyn, J. D.; Tanzer, M.; Krygier, J. J.; Dujovne, A. R.; and Brooks, C. E.: Concerns with modularity in total hip arthroplasty. Clin. Orthop.,298: 27-36, 1994.29827
1994
[PubMed]
Bourne, R. B.; Rorabeck, C. H.; Ghazal, M. E.; and Lee, M. H.: Pain in the thigh following total hip replacement with a porous-coated anatomic prosthesis for osteoarthrosis. A five-year follow-up study. J. Bone and Joint Surg.,76-A: 1464-1470, Oct. 1994.76-A1464
1994
Cameron, H. U.: The 3—6-year results of a modular noncemented low-bending stiffness hip implant. A preliminary study. J. Arthroplasty,8: 239-243, 1993.8239
1993
[PubMed]
Collis, D. K.: Cemented total hip replacement in patients who are less than fifty years old. J. Bone and Joint Surg.,66-A: 353-359, March 1984.66-A353
1984
D'Antonio, J. A.; Capello, W. N.; and Jaffe, W. L.: Hydroxylapatite-coated hip implants. Multicenter three-year clinical and roentgenographic results. Clin. Orthop.,285: 102-115, 1992.285102
1992
[PubMed]
D'Antonio, J. A.; Capello, W. N.; and Manley, M. T.: Remodeling of bone around hydroxyapatite-coated femoral stems. J. Bone and Joint Surg.,78-A: 1226-1234, Aug. 1996.78-A1226
1996
Dorr, L. D.; Takei, G. K.; and Conaty, J. P.: Total hip arthroplasties in patients less than forty-five years old. J. Bone and Joint Surg.,65-A: 474-479, April 1983.65-A474
1983
Engh, C. A.; Bobyn, J. D.; and Glassman, A. H.: Porous-coated hip replacement. The factors governing bone ingrowth, stress shielding, and clinical results. J. Bone and Joint Surg.,69-B(1): 45-55, 1987.69-B(1)45
1987
Engh, C. A.; Massin, P.; and Suthers, K. E.: Roentgenographic assessment of the biologic fixation of porous-surfaced femoral components. Clin. Orthop.,257: 107-128, 1990.257107
1990
[PubMed]
Engh, C. A.; Hooten, J. P., Jr.; Zetti-Schaffer, K. F.; Ghaffarpour, M.; McGovern, T. F.; Macalino, G. E.; and Zicat, B. A.: Porous-coated total hip replacement. Clin. Orthop.,298: 89-96, 1994.29889
1994
[PubMed]
Gruen, T. A.; McNeice, G. M.; and Amstutz, H. C.: "Modes of failure" of cemented stem-type femoral components. A radiographic analysis of loosening. Clin. Orthop.,141: 17-27, 1979.14117
1979
[PubMed]
Harris, W. H.: Traumatic arthritis of the hip after dislocation and acetabular fractures: treatment by mold arthroplasty. An end-result study using a new method of result evaluation. J. Bone and Joint Surg.,51-A: 737-755, June 1969.51-A737
1969
Heekin, R. D.; Callaghan, J. J.; Hopkinson, W. J.; Savory, C. G.; and Xenos, J. S.: The porous-coated anatomic total hip prosthesis, inserted without cement. Results after five to seven years in a prospective study. J. Bone and Joint Surg.,75-A: 77-91, Jan. 1993.75-A77
1993
Liang, M. H.; Katz, J. N.; Phillips, C.; Sledge, C.; Cats-Baril, W.; and The American Academy of Orthopaedic Surgeons Task Force on Outcome Studies: The Total Hip Arthroplasty Outcome Evaluation Form of The American Academy of Orthopaedic Surgeons. Results of a nominal group process. J. Bone and Joint Surg.,73-A: 639-646, June 1991.73-A639
1991
Noble, P. C.; Alexander, J. W.; Lindahl, L. J.; Yew, D. T.; Granberry, W. M.; and Tullos, H. S.: The anatomic basis of femoral component design. Clin. Orthop.,235: 148-165, 1988.235148
1988
[PubMed]
Schmalzried, T. P., and Harris, W. H.: Hybrid total hip replacement. A 6.5-year follow-up study. J. Bone and Joint Surg.,75-B(4): 608-615, 1993.75-B(4)608
1993
Schulte, K. R.; Callaghan, J. J.; Kelley, S. S.; and Johnston, R. C.: The outcome of Charnley total hip arthroplasty with cement after a minimum twenty-year follow-up. The results of one surgeon. J. Bone and Joint Surg.,75-A: 961-975, July 1993.75-A961
1993
Xenos, J. S.; Hopkinson, W. J.; Callaghan, J. J.; Heekin, R. D.; and Savory, C. G.: Osteolysis around an uncemented cobalt chrome total hip arthroplasty. Clin. Orthop.,317: 29-36, 1995.31729
1995
[PubMed]