THE 3D FEMORAL NECK AXIS- A NOVEL METHOD TO EXPLORE THE HEAD-NECK RELATIONSHIP IN NORMAL AND CAM HIPS

Abstract

A robust frame of reference is required to accurately characterize pathoanatomy in the proximal femur and quantify the femoral head-neck relationship. A three dimensional (3D) femoral neck axis (FNA) could serve such a purpose, but has not yet been established in the current literature.

The primary aim of this study was to develop and evaluate a reliable method of determining the 3D femoral neck axis. Secondly, we wanted to quantify the translational relationship between the femoral head and neck in normal and cam type hips.

Pelvic computed tomographic scans (CT) and radiographs were retrieved from our database of patients who had undergone navigated hip surgery or CT colonography. All patients had given informed consent for their medical files and imaging to be used for research purposes, as approved by the institutional review board.

Pre-operative scans were performed using the Siemens Sensation 64 slice scanner (Siemens Medical Solutions, Erlangen, Germany). The Imperial Protocol developed at the authors’ orthopaedic unit was applied, allowing acquisition of Digital Imaging and Communications in Medicine (DICOM) files of 0.75mm thickness.

Normal and cam type hips (n=30) were identified for analysis. ‘Normal’ hips (n=15) were defined in asymptomatic patients with no previous history of hip disease, and, no obvious abnormality on radiographs or CT. The ‘cam’ hip type (n=15) was defined by the presence of an anterior osseous bump at the head-neck junction, and an alpha angle greater than 50° on hip radiographs.

DICOMs were converted to 3D stereolith (STL) images using validated commercial image processing and analysis software (3-Matics, Materialise Group, Leuven, Belgium).

In order to determine the 3D-FNA, a best fit sphere was applied to the femoral head with a root mean square error of less than 0.5mm. The border between sphere and femoral neck defined the head -neck junction. The bone surface was marked here (including the anterior bump in cam hips) and at the neck base, providing two anatomical rings that defined the superior and inferior limits of the femoral neck. The centre point of each ring was calculated. A line connecting these points defined the femoral neck axis, and was verified on a DICOM viewer in sagittal, axial and coronal planes. The offset between the femoral head centre and neck axis was measured.

The 3D image and axis were further analysed to examine the femoral head-neck relationship, using customized software developed at our institution and previously validated in previous research projects.

To standardize rotational alignment, the femoral neck was aligned vertically in two planes by creating an axis between the tip of the greater trochanter and the center of the lesser trochanter. The aligned proximal femur was viewed end on, and the version of the head relative to the neck determined by calculating the angle between the head centre and a vertical marker placed at the 12 o’clock position. Angles below 180° demonstrated anteversion, while those above 180° demonstrated retroversion.