Abstract
Introduction
Primary stability is essential for long-term performance of cementless femoral components. There is debate as to whether collars contribute to primary stability. The results from experimental studies and finite element (FE) analysis have been variable and contradictory. Subtle differences in performance are often swamped by variation between cadaveric specimens in vitro, whereas FE studies tend to be performed on a single femur. However, FE studies have the potential to make comparisons of implant designs within the same cohort of femurs, allowing for subtle performance differences to be identified if present. This study investigates the effect of a collar on primary stability of a femoral prosthesis across a representative cohort of femurs.
Materials and Methods
FE models were generated from QCT scans of eight cadaveric femurs taken from the Melbourne Femur Collection (4 male and 4 female; BMI: 18.7 – 36.8 kg.m-2; age: 59 – 80 years) which were of joint replacement age. Heterogeneous bone material properties were assigned based on the CT greyscale information. Each femur was implanted with the collared and collarless version of Corail femoral stem (DePuy, Leeds, United Kingdom). The stems were sized and positioned so that the prosthesis filled the medullary canal with minimal gap between the prosthesis and the inner boundary of the cortical bone. The peak muscle and joint contact forces associated with level gait were applied and the distal femur was rigidly fixed. The forces were scaled based on the body weight for each subject. Micromotion, as well as microstrains at the bone-prosthesis interface were measured for each subject. Paired t-test was run to compare the micromotion and the microstrains measured for the collared and collarless prosthesis.
Results
There were no significant differences in micromotion (p > 0.005) and microstrains (p » 0.005) between collared and collarless prostheses. The mean of the median micromotions for the collared and the collarless prostheses were 19.4 microns and 20.5 microns, respectively. The mean of the median equivalent strains at the bone-implant interface for the collared and the collarless prostheses were 828.5 microstrains and 824.3 microstrains, respectively. The mean percentage of the area at the contact interface that experienced equivalent strains lower than 2000 microstrains was 69.9% for the collared and 70.0% for the collarless designs. The mean percentage of the contact area at the bone-prosthesis interface that experience equivalent strains greater than 7000 microstrains, the yield strain, was only 9.9% for the collared and 5.7% for the collarless designs.
Discussion and conclusions
There was considerable variation across the cohort of femurs, with a factor of two difference for both micromotion and interface strain While small differences were noted between the collared and collarless prostheses implanted in the same femur, these differences were minimal and were likely to have little affects on primary stability, at least for a level gait load case. More demanding load cases may result in greater differences between collared and collarless implants. The results suggest that the addition of a collar in routine cases may not enhance the primary stability of a cementless hip stem.