Advertisement for orthosearch.org.uk
Orthopaedic Proceedings Logo

Receive monthly Table of Contents alerts from Orthopaedic Proceedings

Comprehensive article alerts can be set up and managed through your account settings

View my account settings

Visit Orthopaedic Proceedings at:

Loading...

Loading...

Full Access

Research

HOW MANY DIFFERENT TYPES OF FEMORA ARE THERE IN PRIMARY HIP OSTEOARTHRITIS? AN ACTIVE SHAPE MODELLING STUDY

British Orthopaedic Research Society (BORS)



Abstract

In uncemented total hip arthroplasty (THA), the optimal femoral component should allow both maximum cortical contact with proximal load transfer and accurate restoration of individual joint biomechanics. This is often compromised due to a high variability in proximal femoral anatomy. The aim of this on-going study is to assess the variation in proximal femoral canal shape and its association with geometric and anthropometric parameters in primary hip OA.

In a retrospective cohort study, AP-pelvis radiographs of 98 consecutive patients (42 males, 56 females, mean age 61 (range:45-74) years, BMI 27.4 (range:20.3-44.6) kg/m2) who underwent THA for primary hip OA were reviewed. All radiographs were calibrated and femoral offset (FO) and neck-shaft-angle (NSA) were measured using a validated custom programme. Point-based active shape modelling (ASM) was performed to assess the shape of the inner cortex of the proximal femoral meta- and diaphysis. Independent shape modes were identified using principal component analysis (PCA). Hierarchical cluster analysis of the shape modes was performed to identify natural groupings of patients. Differences in geometric measures of the proximal femur (FO, NSA) and demographic parameters (age, height, weight, BMI) between the clusters were evaluated using Kruskal-Wallis one-way-ANOVA or Chi-square tests, as appropriate.

In the entire cohort, mean FO was 39.0 mm, mean NSA was 131 degrees. PCA identified 10 independent shape modes accounting for over 90% of variation in proximal femoral canal shape within the dataset. Cluster Analysis revealed 6 shape clusters for which all 10 shape modes demonstrated a significantly different distribution (p-range:0.000-0.015). We observed significant differences in age (p=0.032), FO (p<0.001) and NSA (p<0.001) between the clusters. No significant differences with regard to gender or BMI were seen.

Our preliminary analysis has identified 6 different patterns of proximal femoral canal shape which are associated with significant differences in femoral offset, neck-shaft-angle and age at time of surgery. We are currently evaluating the entire dataset of 345 patients which will allow a comprehensive classification of variation in proximal femoral shape and joint geometry. The present data may optimise preoperative planning and improve future implant design in THA.