Abstract
Introduction
Many tests have been published which measure frictional torque [1–4] in THR. However, different test procedures were used in those studies. The purpose of this study was to determine the effect of test setup on the measured friction torque values.
Methods
Specimen Description Table 1 lists tested study groups (n≥3). Metal-on-Metal specimens were custom designed and manufactured, and are not approved for clinical use. The remaining groups consisted of commercially available products (Stryker Orthopaedics, NJ).
Test Model – A
A 50 mm outer diameter (OD), Solid-Back Trident PSL shell (Stryker Orthopaedics, NJ) was assembled into a test block and the Ti6Al4V trunnion was oriented parallel to the central axis of the articulating surface [Fig. 1]. A 2450N axial load was applied. The head underwent angular displacement of ± 20° about the central axis of the shell.
Test Model – B
The same shell and block as in Test Model A were used but positioned to simulate a 50° abduction angle [Fig. 1]. A Ti6Al4V trunnion was oriented to simulate a 130° neck angle. A 2450N side load was applied and the head underwent angular displacement of ± 20° about the central axis of the shell.
Test Model – C
A 54 mm OD, Solid-Back Trident PSL shell was assembled into a hip simulator and oriented with a 50° abduction angle [Fig. 1]. The head was assembled onto a stainless steel trunnion and oriented with a 130° neck angle. The load was held at 2450N and the shell underwent a ± 23° biaxial rocking motion.
All tests were conducted in a lubricated environment, using 25% bovine serum. Test Models A & B yielded a maximum static torque that was defined as the peak torque observed when the velocity of the femoral head approaches zero and the head changes direction. Test Model C exhibits continuous motion and yielded a maximum dynamic torque value. Test models were statistically compared using a single-factor ANOVA test and a Tukey post-hoc test at 95% confidence level.
Results
Sample group results, see Table 2, were compared between test models and it was found each model yielded significantly different (p < 0.05) values. Except for the 28 mm-SXL group where there was no significant difference between test models A & C.
Discussion
Each test method yielded unique results, as highlighted by the large difference seen between static and dynamic torque values given by test models B and C. Although, test model B yielded significantly greater static torque, further analysis of raw data indicated equivalent dynamic torque values when compared to model C. Additionally, the test methods did not consistently rank test groups. Two of the three tests showed similar torques when comparing the conventional and SXL materials, while the third model found a significant difference between the two groups. Results demonstrate that careful attention must be applied when selecting a test model.
