In our tertiary arthroplasty referral center, 2,367 hip arthroplasties were performed between 1992 and 1994. In 1,123 hips, a custom femoral component was used: in 329 of these hips this was for severe femoral deformity. 61 patients (72 hips) were under the age of 40 years at the time of surgery and all were included in the present prospective study. The study was approved by the Institutional Review Board of the University of Heidelberg (346/ 2004 and 081/2005) and the study was carried out in accordance with the Helsinki Declaration of 1975.
The series comprised 72 uncemented total hip arthroplasties in 61 patients (33 men), all of whom were younger than 40 years (mean age 35 (22–40) years) with femoral deformities and osteoarthritis (Table). 5 men and 6 women had bilateral surgery at different sessions. Mean body weight was 77 (50–140) kg, the mean height was 171 (150–196) cm, and mean BMI was 26 (18–41). All patients were evaluated prospectively and operated on between 1992 and 1994 by 4 surgeons at our center.
Patient demographics, diagnoses, and classification of proximal femoral deformity
All patients had a femoral deformity according to the criteria of Berry (1999)
. The distortion of the femoral morphology was the indication for using the custom stem in this study. All patients received a CT3D-A custom-made femoral stem (OS Orthopedic Services GmbH, Mainhausen, Germany) () and a Double-Spherical-Pressfit cup (DSP-CUP) (OS Orthopedic Services) with a polyethylene (PE) insert and a 28-mm ceramic head (Ceramtech, Plochingen, Germany). PE inserts that had been gamma-irradiated in gas with a minimum of 6 mm thickness were used in all cases.
26-year-old woman with severe secondary OA of the left hip, femoral deformity, and varisation osteotomy. Radiographs (left to right) preoperatively, and 1 year and 15 years postoperatively.
Design of the custom CT3D-A stem uses computer technology based on CT images. For construction of the implant, the femur was first reconstructed in three dimensions. The hip stem is derived directly from the form of the bone cavity. The outer and inner surfaces of the hard bone of every patient are reconstructed three-dimensionally. The important area for the prosthesis design is the strong trabecular transition structure between corticalis and cancellous bone. Threshold contouring was used with 500 Houncefield units as a threshold value.
The finding of the contours is done automatically; however, any irregularities such as disturbing bone lamellae or possible interruptions in the bone can be corrected interactively by the design engineer.
This provisional implant was then modified progressively using a computer program supplied by the manufacturer, with which the stem is introduced in a virtual model and extracted from the femur while maintaining preselected areas of bone contact for the best filling and fitting in the proximal metaphysis. Distal diaphyseal fixation was avoided by reducing the diameter of the stem. The length of the stem ranged from 140 to 160 mm. The extramedullary part of the prosthesis is designed for restoration of a physiological CCD angle, neck offset, and anteversion. The final shape of the prosthesis was transferred to a computer-assisted machining (CAM) device and the stem was prepared from a titanium block (TiAl6V4). The macro-structure with a medial bridge and arched structure effectively strengthen both the axial and the rotational stability. A coating layer of hydroxyapatite (HA) (thickness, 80–150 µm) was applied to the proximal two-thirds of the implant. The program also constructs the corresponding rasp for each stem individually.
The 3-D preoperative planning was carried out during the design process by the engineer and was validated by the surgeon. This planning laid special emphasis on the neck osteotomy level, the final position of the implant in relation to the lesser and greater trochanters, and the values of helitorsion and neck anteversion.
An anterolateral approach was used for all hips. The femoral cavity was prepared using the custom broach that mimicked the shape of the stem. A pneumatic hammer was used in all cases for preparation of the femur to achieve compaction of the cancellous bone, so that the prosthesis would fit at the correct level. All 72 stems were implanted according to the preoperative plan without any problems or complications intraoperatively. No fracture of the femur, incorrect fit of the prosthesis, or incorrect torsion of the neck of the prosthesis was observed during the surgical procedures. All patients started walking on the day after surgery with full weight bearing as tolerated.
During follow-up, 2 patients died of unrelated causes (7 and 8 years after surgery). Thus, 59 patients (70 hips) were reviewed clinically and radiographically by two independent observers at a mean follow-up time of 14 (10–16) years. For 55 of these patients, the length of follow-up was 12 years or more. No patients were lost to follow-up. All results were analyzed clinically on the basis of preoperative and postoperative Harris hip score (HHS). We assessed patients subjectively by asking them how they felt about their procedure (dissatisfied, satisfied, or very satisfied). The pre- and postoperative activity levels of the patients were assessed according to Devane et al. (1997)
. Clinical assessment included limp, range of motion, and pain. Patients assessed their pain in the hip after surgery at the time of follow-up on a visual analog scale (VAS; 0–10). The range of motion of the hip joint was measured with a goniometer. The “full range of motion” of the hip joint was defined as flexion > 90°, abduction > 15°, adduction > 15°, internal rotation > 15°, and external rotation > 15°.
At the time of review, AP and lateral radiographs of the hip were compared with those taken immediately after surgery and with those taken regularly during the postoperative follow-up. Two independent, experienced orthopedic surgeons compared the radiographs for stem alignment, subsidence, radiolucent lines, bone hypertrophy, osteolysis, stress shielding, pedestal formation of the stem tip, heterotopic ossifications, and femoral and acetabular loosening. Radiolucent lines, endosteal ossifications, and osteolysis were recorded for the seven Gruen zones and bone hypertrophy was defined as thickening of the periprothetic diaphyseal bone. Osteolysis was defined as areas of localized bone resorption or endosteal erosion. Stress shielding was defined according to Engh et al. (1987)
. Pedestal formation was defined as a shelf of endosteal new bone at the stem tip partially or completely bridging the intramedullary canal. Radiographic failure was defined as subsidence of the stem of more than 2 mm, variation in the frontal stem axis of more than 2°, any osteolysis, or any radiolucency of more than 2 mm—or that progressed with time. Acetabular loosening was defined as continuous migration of > 5 mm or tilting of > 5° compared with baseline AP radiographs. Criteria for stability were the same as those described by Engh et al. (1990)
Kaplan-Meier survival analysis was performed with two endpoints: (1) revision procedures of the acetabular component and of the stem for aseptic loosening, and (2) revision procedures of the acetabular component and the stem for any cause. The statistical analysis was done with SAS software version 9.1 for Windows (SAS Institute Inc., Cary, NC).