The Biotribology of PEEK-on-HXLPE Bearings Is Comparable to Traditional Bearings on a Multidirectional Pin-on-disk Tester

All-polymer bearings involving polyetheretherketone (PEEK) have been proposed for orthopaedic applications because they may reduce stress shielding, reduce weight of the implants, reduce wear and risk of osteolysis, and prevent release of metal ions by replacing the metal articulating components. Little is known about the biotribology of all-polymer PEEK bearings, including the effects of cross-shear, which are relevant for implant longevity, especially in the hip, and increased temperature that may affect lubricant proteins and, hence, lubrication in the joint.

Using pin-on-disk in vitro testing, we asked: (1) Can all-polymer bearing couples involving PEEK have a comparable or lower wear rate than highly crosslinked UHMWPE (HXLPE) on CoCr bearing couples? (2) Is the wear rate of PEEK bearing couples affected by the amount of cross-shear? (3) Is there a difference in wear mechanism and surface morphology for all-polymer bearing surfaces compared with UHMWPE (HXLPE) on CoCr?

We simultaneously tested a total of 100 pin-on-disk couples (n = 10 per bearing couple) consisting of three traditional metal-on-UHMWPE and seven polymer-on-polymer bearings for 2 million cycles under physiologically relevant conditions and in accordance with ASTM F732. Using analysis of variance, we analyzed the effect of bearing surface topography and cross-shear on wear rate. The changes in surface topography were evaluated using optical microscopy. Sample size was sufficient to provide 80% power to detect a difference of 1.4 mm³/MC in average wear rates of bearing couples.

The combined wear rates of all-polymer bearing couples were not different than traditional bearing couples. With the numbers available, the PEEK and HXLPE bearing couple had a mean wear rate (WR: mean ± SD) of 0.9 ± 1.1 mm³/MC (95% confidence interval [CI], 0.2–1.5 mm³/MC), which was not different than the wear rate of the CoCr and HXLPE bearing couple (1.6 ± 2.0 mm³/MC; 95% CI, 0.4–2.8 mm³/MC; mean difference = 0.73 mm³/MC, p = 0.36). Bearing couples with PEEK reinforced with a carbon fiber (CFR-PEEK) counterface had higher wear rates (14.5 ± 15.1 mm³/MC; 95% CI, 9.1–20.0 mm³/MC) than bearing couples with a PEEK (5.1 ± 3.7 mm³/MC; 95% CI, 3.7–6.4 mm³/MC) or CoCr (4.1 ± 2.7 mm³/MC; 95% CI, 3.2–5.1 mm³/MC) counterface (mean difference = 9.5 mm³/MC, p < 0.001; and mean difference = 10.4 mm³/MC, p < 0.001, respectively). PEEK and HXLPE were insensitive to the cross-shear scenario in the contact mechanics (WR: 0.3 ± 0.1 mm³/MC for PEEK pins [95% CI, 0.2–0.3 mm³/MC] [representing full cross-shear condition] and 0.0 ± 1.0 mm³/MC for PEEK disks [95% CI, −0.5 to 0.5 mm³/MC] [representing limited cross-shear condition], mean difference = 0.3 mm³/MC, p = 0.23; WR: 1.3 ± 1.0 mm³/MC for HXLPE pins [95% CI, 0.7–1.9 mm³/MC] [full cross-shear] and 2.1 ± 2.2 mm³/MC for HXLPE disks [95% CI, 0.8–3.3 mm³/MC] [limited cross-shear], mean difference = 0.8 mm³/MC, p = 0.24). Qualitatively, the surface morphology of UHMWPE appeared similar with PEEK or CoCr as a counterface, although it had a rougher appearance when coupled with carbon fiber-reinforced PEEK. No transfer film was detected on the specimens.

Our in vitro pin-on-disk data suggest that all-polymer bearings, especially PEEK-on-HXLPE bearing couples, may represent a viable alternative to traditional bearings with respect to their wear performance. Our results warrant further testing of all-polymer bearing couples in physiologically relevant joint simulator tests.

The in vitro pin-on-disk wear resistance of all-polymer bearings incorporating PEEK-on-HXLPE warrants further investigation using joint simulator testing for their validation as useful, metal-free alternatives to traditional CoCr-on-HXLPE bearings for use in orthopaedic applications.