International Orthopaedics (SICOT) DOI 10.1007/s00264-015-2739-1
ORIGINAL PAPER
The effect of femoral neck osteotomy on femoral component position of a primary cementless total hip arthroplasty Dimitris Dimitriou 1 & Tsung-Yuan Tsai 1 & Young-Min Kwon 1
Received: 25 February 2015 / Accepted: 4 March 2015 # SICOT aisbl 2015
Abstract Purpose The aim of this study was to quantify the femoral canal diameter and version at different femoral neck osteotomy locations, and to investigate the effect of the osteotomy plane on femoral component position in total hip arthroplasty (THA). Methods Preoperative and postoperative three-dimensional models were reconstructed in 15 patients (19 hips) who underwent primary cementless THA with tapered nonanatomical femoral stem. On the pre-operative models, the osteotomy plane was simulated at different levels (−5, 0, 5, and 10 mm from the femoral saddle [piriformis fossa]) and angles (30, 40, 50, and 60° from the femoral anatomical axis). Medullary canal version and mediolateral diameter were measured on the osteotomy surfaces. On the postoperative models, the femoral neck osteotomy plane, stem anteversion and alignment were measured. Results The average canal diameter ranged from 22.8 to 26.3 mm at different osteotomy levels and from 20.8 to 29.0 mm at different osteotomy angles. The average canal version ranged from 11.4 to 23.2° at different resection levels and from 12.8 to 21° at different resection angles. The femoral stem anteversion was correlated with neck osteotomy angle (R=0.72), whereas stem alignment in frontal plane (varus/valgus) was correlated with neck osteotomy level (R=0.87). Conclusions The femoral neck osteotomy plane in THA affects the postoperative stem position due to the complex
* Young-Min Kwon
[email protected] 1
Bioengineering Laboratory, Department of Orthopaedic Surgery, Massachusetts General Hospital and Harvard Medical School, 55 Fruit Street, GRJ 1215, Boston, MA 02114, USA
morphology of the proximal femoral medullary canal, suggesting that both femoral neck resection level and angle should be considered in optimizing femoral component alignment in THA patients. Keywords Total hip arthroplasty . Components positioning . Femoral anteversion . Offset
Introduction Component malposition in total hip arthroplasty (THA) is associated with aseptic loosening, impingement, dislocation, and early implant failure [1, 2]. Although numerous studies investigated the acetabular component position in THA, limited data are available on femoral stem position [3, 4]. In cementless THA, stem anteversion and frontal plane alignment (varus/valgus) are dictated by the shape of the proximal femur [5], as the femoral stem follows the medullary canal from the entrance point (femoral neck resection surface) to achieve proximal mediolateral metaphyseal and distal diaphyseal fixation [6]. Consequently, femoral canal morphology following neck osteotomy might affect the final stem position. A limited number of studies have investigated the femoral canal version and diameter at different levels and angles, using cross-sectional computed tomography (CT) images [7, 8]. However, the morphology of the femoral canal following neck resection might vary significantly from the axial CT images. Data regarding proximal femur anatomy following neck resection, and the effect of femoral neck osteotomy on the final stem position, in primary cementless THA using conventional metaphyseal stems remain largely unknown. Due to the complex anatomy of the proximal femur, we hypothesized that femoral canal morphology would vary significantly at different femoral neck resection planes and that
International Orthopaedics (SICOT)
femoral neck osteotomy would affect the femoral component position of a primary cementless THA. Therefore, the purpose of this study was: (1) to quantify the femoral canal version and mediolateral diameter on the surface of femoral neck, following different resection levels and angles; (2) to investigate the effect of the femoral neck osteotomy on the femoral stem orientation following primary cementless THA.
Material and methods Patients The study was approved by the institution’s internal review board and each patient provided written informed consent prior to participation. Nineteen hips, in 15 patients (three men, 12 women) who underwent cementless THA for primary, end-stage hip osteoarthritis were evaluated in this study. Patients’ average age was 64.6 years (±8.1, range 53– 73 years), height 166.3 cm (±6.7, range 154.9–180.3 cm), weight 76.1 kg (±17.9, range 59.9–103.0 kg) and BMI 29.4 kg/m2 (±3.6, range 20.7–31.8 kg/m2). The average follow-up period was 10.6 months (±2.1, range 6.6– 13.1 months). Five patients had left THA, seven patients had right THA, while three patients received bilateral THA. No patient had a history of surgical complication, dislocation or component subluxation on the implanted hip. All patients received collarless tapered wedge cementless stems (Linear stem; DJO Global, Vista, CA, USA). THA was performed by a single surgeon using the posterolateral approach.
CT–based 3D modeling and measurements All patients were scanned following the same protocol using a 64-slice computed tomography (CT) scan (Sensation 64; Siemens, Erlangen, Germany) in the supine position, from the fourth lumbar vertebra to the mid-femur, with settings of 120 kV and 80 mA, before and following THA (Fig. 1a). The images were acquired along the axial direction with a slice thickness of 0.625 mm, an in-plane resolution of 0.95× 0.95 mm, and a matrix size of 512 × 512. Using a commercial program (Rhinoceros®; Robert McNeel and Associates, Seattle, WA, USA), preoperative and postoperative femoral surface models were reconstructed, in accordance to a previously validated and published protocol [9]. Briefly, the internal and external outlines of the femurs were reconstructed using a Gaussian filter with a gradient threshold and region growing. Each surface model was carefully inspected by an experienced orthopedic surgeon (Fig. 1a, b).
Measurement of anatomical parameters on the preoperative and postoperative femur An anatomical coordinate system of the femur was reconstructed, based on the International Society of Biomechanics recommendations [10] (Fig. 1a, b). The femoral osteotomy plane was simulated at different levels (−5, 0, 5, and 10 mm from the femoral saddle) for a fixed 45° neck osteotomy angle and at different neck osteotomy angles (30, 40, 50, and 60° from the femoral anatomical axis) for a fixed 0-mm neck osteotomy level. At each level and angle, the mid-cortical line, defined as the line connecting the midpoints between the anterior and posterior inner cortical lines, was identified (Fig. 2a, b). The femoral canal diameter along the mid-cortical line, defined as mediolateral femoral canal diameter (FCD), and the angle between the mid-cortical line and posterior condylar axis, defined as femoral canal version (FCV), were measured. On the postoperative model, the femoral neck osteotomy plane (FOP) was identified by fitting a three-dimensional (3D) plane on the femoral neck osteotomy surface, using the iterative closest point algorithm (MATLAB; The MathWorks, Natick, MA) [11]. The femoral neck osteotomy level (FOL) was defined as the closest distance between the femoral saddle (piriformis fossa) and the center of FOP. The femoral neck osteotomy angle (FOA) was defined as the 3D angle between FOP and femoral anatomical axis (Fig. 1a). Femoral anteversion, neck-shaft angle and femoral horizontal and vertical offset were measured on both preoperative and postoperative surface models (Fig. 1a). Sagittal stem alignment (stem tilt), defined as the angle between femoral anatomical axis and the long axis of the femoral stem on the sagittal plane was measured on the postoperative model (Fig. 1b). Similarly, coronal stem alignment (varus/valgus), defined as the angle between femoral anatomical axis and the long axis of the femoral stem on the coronal plane was also measured on the postoperative model. All measurements on the 3D neck osteotomy surface were repeated blindly by two observers in a 1-month period. Intraobserver and interobserver reliabilities of the measurements were evaluated using single-measure intraclass correlation coefficients (ICC) with a two-way random-effects model for absolute agreement. Statistical analysis The distributions of the variables were examined in exploratory data analyses and all variables met the criteria for a normal distribution. Descriptive statistics used mean, standard deviation and range. One-way repeated measured ANOVA was adopted to compare the femoral canal diameter and version, at different neck osteotomy
International Orthopaedics (SICOT) Fig. 1 Three-dimensional measurements on the postoperative femur. Definitions of the femoral neck osteotomy level (FOL), femoral neck osteotomy angle (FOA), femoral neck osteotomy plane (FOP), femoral anatomical axis (FAA) are shown. a Anterior view of the right femur. b Medial view of the right femur. c Superior view of the right femur
levels and angles. Simple linear regression was used to detect potential correlations between femoral neck osteotomy plane (level and angle) and post-operative femoral parameters. Multiple linear regression was performed to identify the factors that affect final component position following THA. The models were adjusted for gender, age, height and BMI. Level of significance was set at a=0.05. The statistical analyses were conducted using Statistica software (StatSoft, Tulsa, OK, USA)
Results
Fig. 2 Example of simulation of femoral neck osteotomy at different neck osteotomy planes. a Simulation of three different neck osteotomy levels. b Simulation of three different angle osteotomies. The midcortical
line is shown by yellow dotted line, whereas the posterior condylar axis is shown by green dotted line. Femoral canal version (the angle between mid-cortical line and posterior condyle axis) is shown by red dotted line
Femoral canal diameter (FCD) measurements Significant differences were observed between femoral canal diameter (FCD) at different neck osteotomy levels and angle (Fig. 3a, b). The average FCD ranged from 22.8 to 26.3 mm at different neck osteotomy levels and from 20.8 to 29 mm at different neck osteotomy angles, respectively (Table 1). The FCD decreased as the neck osteotomy levels and angles
International Orthopaedics (SICOT)
Fig. 3 a, b Femoral canal diameter at different osteotomy levels and osteotomy angles. c, d Femoral canal version at different osteotomy levels and angles. Data are represented as average and standard deviation. Statistical significant differences (p < 0.05) are shown with red lines
increased (Fig. 3a, b). Intra-observer ICC and interobserver ICC were high for both femoral canal diameter and version (0.94, 0.91 and 0.89, 0.87, respectively).
FCV increased as the neck osteotomy level and angle increased from −5 to 10 mm and from 40 to 60°, respectively (Fig. 3c, d).
Femoral canal version (FCV) measurements
Femoral anteversion measurements
With respect to femoral canal version (FCV), significant differences were observed at different neck osteotomy levels and angles (Fig. 2c, d). The average FCV ranged from 11.4 to 23.2° at different neck osteotomy levels and from 13.3 to 21° at different neck osteotomy angles (Table 2). The average
The average pre-operative and postoperative femoral anteversion were 11.5 ± 13.1 ° (range, −14.4 to 35.2 °) and 10.7 ± 13.4° (range, −22.9 to 32.8 °), respectively. The average neck osteotomy level and angle were 4.0 ± 5.1 mm (range, −3.1 to 10.1) and 48.3 ± 7.8 ° (range, 33.2–51.2 °),
Table 1 Femoral canal diameter at different osteotomy levels and different osteotomy angles
Variable 1
Variable 2
Osteotomy level (mm) −5 0 −5 5 −5 10 0 5 0 10 5 10 Osteotomy angle (°) 30 40 30 50 30 60 40 50 40 60 50 60
Mean 1
Mean 2
p value
26.3 26.3 26.3 25.2
25.2 23.9 22.8 23.9
Mediolateral diameter (mm) 0.14 2.6–0.4 0.001* 1.1–3.8 0.0009 1.6–5.4 0.01* 2.8–2.4
25.2 23.9
22.8 22.8
29.0 29.0 29.0 28.0 28.0 28.0
28 23.0 20.8 23.0 20.8 20.8
*p < 0.05, indicating statistically significant difference
95 % CI
0.03* 0.2–4.7 0.14 −0.4 to 2.5 Mid-cortical diameter (mm) 0.29 −0.9 to 2.9 0.002* 2.4–9.5
