Materials and Methods

Thirty-one hips in 31 patients were performed total hip arthroplasty with standard length tapered wedge-shaped (TW) cementless stem (CTi-II: Corin, Cirencester, UK). There were 6 males and 25 females. The average age at operation was 69 years old. The average body mass index was 23.9 kg/m2. Primary diagnoses were secondary osteoarthritis due to developmental dysplasia of the hip in 29 hips. Femoral canal shapes were normal in 21, stovepipe in 6 and champagne-flute in 4 hips. Bone qualities were type A in 6, B in 19 and C in 6 hips.

The patients underwent computed tomography (CT) preoperatively and postoperatively. We constructed preoperative three dimensional (3D) femur surface models from preoperative CT data with individual bone mineral density (BMD) mapping. The postoperative 3D femur and rough stem surface models were obtained from postoperative CT data. The coordinates of the postoperative femur were transformed to fit the preoperative femur model. A precise stem model constructed using computer-assisted design data was matched to the transformed rough stem model using the iterative closest point algorithm. We obtained a patient-specific model with the proximal bone geometry, allocation of BMD and stem alignment. We estimated the average of axial and rotational micromotion (MM) at stem-bone interface and the ratio of area (MM � 40 micrometers) on the porous surface in order to analyze primary stability of TW stem with several lengths (standard (100 %), 75 %, 50 %, 40 % and 30 % length).

[Materials and Methods]

One hundred hips (92 patients) with DDH were performed preoperative simulation for total hip arthroplasty by 3D digital template system (ZedHip: Lexi, Tokyo, Japan). The average age was 63.5 years old. There were 12 males and 80 females. The average bone mass index was 21.5 kg/m². Femoral canal shape was normal in 71, champagne-flute in 16 and stovepipe in 13 hips. Bone quality was classified into type A in 23, type B in 74 and type C in 3 hips. Preoperative computed tomography data were used for 3D digital template and reconstructed to 3D femoral model. Short TW stem (Taperloc Complete Microplasty: Biomet, Warsaw, IN) model constructed from computer-assisted design was matched to the reconstructed femoral model. Short TW stem model was in principle implanted according to the femoral neck anteversion with neutral alignment (varus and valgus < 2 degrees, flexion and extension < 2 degrees) at the coronal and sagittal plane of the femur. Stem size was determined in order to obtain the largest intramedullary matching at the coronal plane. Area of stem fitting with the cortical bone was investigated at 10 mm intervals above and below of mid minor trochanter. Intramedullary matching pattern was classified into proximal mediolateral metaphyseal fit, proximal flare fit and diaphyseal fit at multiple reconstructed planes of the 3D femoral model according to stem fitting area.

[Methods]

One-hundred seventy-nine Japanese patients who underwent TKA in Osaka Koseinenkin Hospital from April, 2009 to December, 2011 were included in this study. The genders and diagnoses were 25/ 154 patients (male/ female) and 143/ 36 patients [osteoarthritis (OA)/ rheumatoid arthritis (RA)], respectively. The mean age, height and body weight were 73.2 years, 152.2 cm and 56.7 kg, respectively. We measured the AP length, ML width, aspect ratio angle of distal femur. The measurement points on the medial condyle was 10 mm and on the lateral condyle was 8 mm from the lowest points of the medial and lateral posterior condyle to stimulate the optimal cutting thickness. We evaluated the relationship between AP length and ML width, or aspect ratio and ML width in diagnosis, alignment and gender. We also compared the measurements with the dimensions of femoral implants current used in Japan.

Continuous variables were assessed using the Mann-Whitney's U test. Relationship between AP length and ML width, or aspect ratio and ML width were assessed using simple regression analysis. Regression lines were assessed using analysis of covariance. For all tests, probability values (p values) of < 0.05 were considered to indicate statistical significance.