High 10-Year Survival Rate with an Anatomic Cementless Stem (SPS)
- 1.1k Downloads
Proximal cementless fixation using anatomic stems reportedly increases femoral fit and avoids stress-shielding. However, thigh pain was reported with the early stem designs. Therefore, a new anatomic cementless stem design was based on an average three-dimensional metaphyseal femoral shape. However, it is unclear whether this stem reduces the incidence of thigh pain.
We asked whether this stem design was associated with a low incidence of thigh pain and provided durable fixation and high function.
One hundred seventy-one patients (176 THAs) who had the anatomic proximal hydroxyapatite-coated stem implanted were reviewed. Eleven (6%) patients were lost to followup and 34 (20%) died without revision surgery. We used the Harris hip score (HHS) to assess pain and function. We evaluated femoral stem fixation and stability with the score of Engh et al. and also calculated a 10-year survival analysis. We assessed 126 patients (131 hips) at a mean followup of 10 years (range, 8–11 years)
At last followup, two patients described slight thigh pain that did not limit their physical activities. All stems appeared radiographically stable and one stem was graded nonintegrated but stable. Five patients had revision surgery: one on the femoral side (for posttraumatic fracture) and four on the acetabular side. Considering stem revision for aseptic loosening as the end point, survivorship was 100% (range, 95.4%–99.9%) at 10 years.
This anatomic cementless design using only metaphyseal fixation with a wide mediolateral flare, a sagittal curvature, and torsion, allowed durable proximal stem stability and fixation.
Level of Evidence
Level IV, therapeutic study. See Guidelines for Authors for a complete description of levels of evidence.
The survivorship rates for uncemented stems are as good as those for cemented stems, but depend on the stem’s design, material, and type of coating [4, 6, 7]. Primary stability of the stem is crucial to insure bone ingrowth. Surgeons can achieve initial stability by filling the femoral diaphysis or getting intimate contact between the stem and the anatomy of the proximal femur. Many designs of proximally fixed cementless femoral stems have been developed to achieve physiologic bone remodeling [8, 14]. The designers assumed the implant would achieve primary stability and minimize stress shielding or overload.
Using a proximal metaphyseal fit cementless custom anatomic stem in patients younger than 50 years, Flecher et al.  reported high functional scores and a 15-year survival rate of 93% with minimal osteolysis, even in patients with severe deformities. These data suggested anatomic cementless stems with metaphyseal fit and fill can provide durable pain relief and function in physically active patients.
We therefore asked whether these anatomic cementless stems provided (1) a low incidence of thigh pain, (2) stable fixation, (3) high long-term survivorship, (4) high functional scores, and (5) a low rate of wear.
Patients and Methods
One surgeon (AM) performed all operations using an anterolateral approach with the patient in the supine position. The femoral stem had a lateral flare designed to ensure primary craniocaudal stabilization. Thus, we always prepared the proximal femoral canal with a curette before rasping. First we removed cancellous bone in contact with the lateral cortex under the greater trochanter to prepare the place where the lateral flare of the stem was located, and second, we removed any excessive femoral “internal calcar septum”  (the internal vertical plate of condensed trabecular bone constituting the anatomic calcar).
Postoperative rehabilitation protocols included immediate weightbearing protected by crutches during the first 2 or 3 weeks according to patient tolerance. The physiotherapy was supervised and was performed each day for at least 1 month. The exercises focused on passive and then active recuperation of ROM. All patients received routine thromboprophylaxis with low molecular-weight heparin postoperatively for 21 days.
We performed radiographic postoperative evaluations using AP views of the pelvis and hip and a true lateral view of the hip. We looked for radiolucencies and osteolysis in the seven zones described by Gruen et al.  and the corresponding seven zones on the lateral view. Radiolucencies (progressive and greater than 2 mm) and osteolysis (defined as areas of reduced bone density not present on the initial radiographs) were recorded . We evaluated the femoral component fixation and stability using the score described by Engh et al. , which included fixation score and stability scores. The fixation scale analyzed the appearance of the stem porous interface (reactive lines and lucencies) and the presence of spot welds. The absence of lines or lucencies and the presence of spot welds corresponded to bone ingrowth and were counted positively (+5 points) with a maximal score of 10. The stability scale was determined by comparing the postoperative and the last radiographs. We recorded the appearance of the smooth interface of the stem (lines or lucencies), presence of a pedestal with an unstable stem tip, calcar remodeling, deterioration of the integrated interface, stem migration, and presence of particle shedding. Positive scores were recorded if there were no lines or lucencies, no pedestal, no calcar hypertrophy, no interface deterioration, no migration, and no particle shedding. The maximal score using the criteria of Engh et al. was 27. Greater than 10, stem osteointegration was confirmed; from 0 to 10 points osteointegration was probable; between −10 to 0 the stem was considered nonintegrated but stable; and less than −10 points the stem was considered unstable. For the socket, we recorded radiolucencies and osteolysis in Zones 1 to 3 according to DeLee and Charnley . The loosening criteria for the cup were those defined by Hodgkinson et al. . Polyethylene wear was measured using ImagikaTM software (View Tech®, CMC Corp, Edison, NJ, USA). We used three points to measure the head diameter and three others for the cup, and then measured linear wear by calculating the distance from the center of the head to the center of the cup. An accuracy of 0.28 mm was reported for this technique when using a 28-mm head, a metal-back, and a design where the centers of the cup and the head coincided . The cup abduction angle was measured on the AP view of the pelvis using the ImagikaTM software. This angle was defined with respect to the landmarks of the inferior aspect of the obturator foramen and the long axis of the projected ellipse of the face of the cup .
We performed a 10-year survival analysis using the Kaplan-Meier technique (with 95% CI). A multivariate analysis was performed to analyze the wear considering the following parameters: age, sex, BMI, activity level (HHS), and cup abduction angle.
At the last followup, two (1.6%) patients described occasional thigh pain not limiting their physical activities or requiring medication.
We considered all stems stable and integrated and one stem as nonintegrated but stable. On the femoral side, no patient had radiolucencies greater than 2 mm wide. We noted a cortical pedestal without pain in one patient. The mean total score of Engh et al. was 20 ± 5, with a mean fixation score of 8 ± 4 and mean stability score of 12 ± 4. On the acetabular side, osteolysis zones were observed in 11 hips.
Details for patients who had revision surgery
Cause for revision
Posttraumatic femur fracture
Recurrent anterior dislocation
Delayed recurrent dislocation
Cup aseptic loosening
Delayed recurrent dislocation
Comparison of preoperative and 10-year followup Harris hip scores
Harris hip score parameter
Mean, SD, (range)
Mean, SD, (range)
9 ± 10 (0–30)
43 ± 4.8 (10–44)
17 ± 8 (0–33)
27 ± 8 (2–33)
6 ± 2 (0–12)
12 ± 3 (3–14)
1 ± 2 (0–4)
4 ± 0.3 (0–4)
Range of motion
4 ± 1 (2–5)
4.9 ± 0.2 (4–5)
37 ± 17 (4–67)
90 ± 13 (45–100)
The average linear wear was 0.09 mm per year (range, 0–0.26 mm per year).
Five patients had an early dislocation that did not require revision. We observed no other complications for this cohort and no zirconia head fractures were observed.
Proximal cementless fixation using anatomic stems is an attractive option for THA. However, an incidence of thigh pain of 5.7% to 17% was reported with the early designs [25, 35, 39] contrasting with the absence of pain reported by Flecher et al. [13, 14] with the anatomic custom stems. Therefore an anatomic cementless stem was designed, based on a computerized three-dimensional (3-D) analysis of a 300 CT scan database, which allowed defining an average metaphyseal femoral shape. We presumed this shape would provide a lower incidence of thigh pain than the previous designs, long-term stability and fixation, high 10-year Kaplan-Meier survivorship, high functional scores, and low rate of wear.
There were limitations to our study. First, we analyzed a small cohort with only 131 patients with a relatively older age at the time of surgery. The analysis of younger patients with high functional demand may be interesting because the mechanical constraints were expected to be higher. Second, we did not assess bone remodeling. A proper stem design allowing proximal fixation theoretically might help achieve more physiologic load transmission, enhancing more physiologic bone remodeling. A comparative study using DXA might assess whether the SPS stem effectively enhanced bone remodeling. Third, there was no control group.
Comparison of long-term followups of cementless polyethylene-bearing stems
Number of hips
Mean followup (years)
Survival rate for aseptic loosening
Thigh pain (%)
Stress shielding (%)
Bone ingrowth (%)
Aldinger et al. 
Baker et al. 
Flecher et al. 
Hennessy et al. 
Kim et al. 
Lee et al. 
Parvizi et al. 
Schramm et al. 
Suckel et al. 
Zenz et al. 
We achieved a high survival rate for the SPS stem at 10 years with high hip scores and no recorded loose stems. These findings compared well with those reported in the literature, especially when compared with the anatomic cementless stems [3, 5]. Other studies also reported high survival rates of proximally HA-coated, press-fit stems for different designs [19, 27, 32, 37]. Hennessy et al.  reported a 100% survival rate at 10 years with an extended, porous-coated, straight-stem with diaphyseal fixation (AML, DePuy, Warsaw, IN, USA). However, 28% of their patients had stress shielding and 2% reported thigh pain despite bone ingrowth being achieved by all patients. Parvizi et al.  reported no stem failures for the HA-coated Taperloc (Biomet, Warsaw, IN, USA) at 10 years followup. Lee et al.  also reported no stem failures with the quadrangular proximally coated stem Omnifit (Stryker, Mahwah, NJ, USA). Suckel et al. , using the Zweymüller stem (Alloclassic, Sulzer Orthopedics, Zimmer, Switzerland), found a similar survival rate of 98% at 15 years followup. The SPS design achieved a similar long-term survival rate when compared with these designs. The comparison of bone remodeling for all these designs may be useful in determining which ones more closely restored physiologic load transmission. The survivorship rate for the cup was 99.3%, which was similar to that reported in the literature for cementless cups .
The rate of patients with thigh pain was low in our cohort, comparing well with the lower rates of thigh pain reported in the literature for uncemented stems, which, although not always reported in studies, varies from 0% to 17% [1, 25, 35, 39]. Naumann et al.  reported 29% of patients had thigh pain during the first months after surgery and 5.6% had pain remaining afterward, for the first generation of the Zweymüller stem. They showed that pain was associated with stress shielding. Kim et al.  reported 10% of patients had thigh pain with the Profile anatomic stem, suggesting a lack of stability and fixation. Flecher et al. , however, reported no thigh pain with the anatomic custom stems, although the patients were younger than 50 years and with high functional demand. These findings suggested that a proper fit between the stem and the proximal femur may avoid thigh pain.
We observed a wear rate of approximately 0.09 mm per year, lower than the reported linear wear rates with metal-on-polyethylene components . The reported wear rate for zirconia-polyethylene bearing surfaces varied from 0.09  to 0.5 mm per year . Some authors [2, 22] reported high wear rates for zirconia-polyethylene. These authors linked these high wear rates to the zirconia transformation phase, which reportedly induced an increase in the surface roughness [2, 36]. However, in another study, low wear rates similar to ours were reported when using zirconia-polyethylene bearings . The main reason for revision in our series (three of the four cup revisions) was for late instability related to acetabular wear. Therefore, we now use alumina ceramic-on-ceramic bearing surfaces in younger or high-function patients.
To our knowledge, no other stem design has been based on a computerized technique to analyze the 3-D anatomy of the intracanalar proximal femur. We believe the anatomic SPS stem allowed correct transmission of physiologic loading to the metaphyseal cancellous bone, minimizing proximal stress-shielding and enhancing bone remodeling around the stem in the proximal femur. Proximal stress shielding, which has been associated with bone resorption around stems , could induce implant loosening  and should be avoided. The stable long-term fixation we observed with the SPS stem has supported this assumption because we observed no aseptic loosening and proximal fixation was achieved. However, we did not analyze bone remodeling with DXA; therefore, we could not draw a valid conclusion regarding the ability of the design to restore physiologic load transmission. Chen et al.  used DXA to analyze bone remodeling after THA, and proved that metaphyseal fixation allowed correct bone remodeling around the femoral stem. This may have explained our low rate of thigh pain (1.6%), despite the fact that no diaphyseal fixation was achieved. Thigh pain has been reported in as much as 16% of patients when using anatomic uncemented stems .
The SPS stem may tend to increase femoral anteversion consequently increasing the risk of anterior dislocation. Therefore, based on previously reported anatomic studies [33, 34], the extracanalar part of the SPS stem will be modified soon (SPS Evolution®). Our data confirm the medium-term durability of fixation with this implant. However, a longer followup is required to confirm continued durability and maintenance of function.
This article is distributed under the terms of the Creative Commons Attribution License which permits any use, distribution, and reproduction in any medium, provided the original author(s) and the source are credited.