Total Hip Replacement (THR) surgeries are considered as an efficient method to treat degenerative joint diseases. However, the survivorship of THR decreases dramatically after 15 years of use (Ferguson et al., 2018). A typical clinical wear rate of ultra-high molecular weight polyethylene (UHMWPE) cups, of 100~300 μm/year, will lead to loosening, inflammation and ultimately failure (McKellop et al., 1995). Efforts are seriously needed to prolong the survival of the hip prostheses, which is important for optimizing patient care and medical costs (Chu, 2015). Consequently, the bio-tribological performance of orthopedic implants has been gaining increasing attention. The tribological evaluation of such kinds of implants is regarded as an essential evaluation of hip prostheses.

Hip joint wear testers, as called hip joint simulator, is developed for the wear evaluation of hip prostheses during a specified duration, under body-simulated conditions referring to the dynamic and kinematic characteristics of human hip joint (Cheng and Shan, 2012). Wear rates and particles in the hip simulator studies are similar to those observed in vivo experience (Hua and Zhang, 2008; Saikko and Shen, 2010; Oliveira et al., 2013; Wang et al., 2013). Currently, ISO organization has published several standards on the wear tests of hip prostheses (ISO, 2009, 2014, 2016).

For a 5-million-cycle hip simulator wear test, the duration usually lasts over 3 months. And the wear rates, obtained by interval measuring, are regarded as the most important results in such kinds of tests. Online information like friction torque and lubricant performance cannot be well-studied. Therefore, besides the material removal, other properties of the prostheses articulating surfaces are needed to understand the dynamic performance during the wear tests. Frictional force monitoring is a valid complementary factor to represent the true tribological behavior. However, among the numerous simulator designs, there are only a few including friction measurement (Bowsher and Shelton, 2001; Liao et al., 2003), which also provides very important data in the validation of numerical analysis of hip prostheses (Gao et al., 2018)

In this study, a new tribological hip joint simulator (HSMS₄₂₀₃) with an improved friction measurement system is designed, built and validated. An online friction measuring module has been newly assembled in the hip joint simulator. Validation tests with UHMWPE and x-linked UHMWPE are performed and discussed. The purpose of this paper is to provide the features and development of this novel hip joint wear tester.

After more than 10 million cycles, the HSMS₄₂₀₃ ran smoothly and no malfunction occurred. Cumulative gravimetric wear for the two samples (UHMWPE and x-linked UHMWPE acetabular cups) exhibited a linear relationship with an increase in the number of motion cycles. Volumetric wear per cycle (mm³/cycle) was therefore calculated and shown in Figure 3. The correlation coefficients of linear regression of the mean volumetric wear values were 0.99757 (UHMWPE), and 0.99953 (x-linked UHMWPE). The slopes of the linear regression, also known as the wear rates, were 24 mm³/10⁶ of UHMWPE, and 16 mm³/10⁶ of x-linked UHMWPE. The initial two samples' morphologies are shown in Figures 4A,B, while the wear surfaces after 5 million cycles are shown in Figures 4C,D.

Continuous dynamic friction was acquired by the senor in the friction measurement module. Except for the running-in effect at every start, most data were stable and showed periodic variation with two maxima and two minima per cycle (Figure 5). The maxima were observed when the upper part of the Z-shaped lever is at the horizontal positions, and the minima were at the top and bottom positions. Also, the frictional force can be distinguished from the two tested samples. For the UHMWPE acetabular cups, they vary from 0~32 N to 0~28 N for the x-linked one.

The duration of an ISO14242 standard wear test for artificial hip joints is 5 million cycles (ISO, 2014). Therefore, the reliability of the hip joint simulator is very important so as to achieve a non-stop testing condition for 1 million cycles at least. The HSMS₄₂₀₃ was still in healthy condition after 10 million cycles. This confirmed that the strong reliability requirement of the simulator could be met. In some retrieval analysis of artificial joints, the clinical volumetric wear is of the order of 10 mm³/10⁶ cycles (Lachiewicz et al., 2009; Wang et al., 2013; Hua et al., 2014). Herein, the volumetric wear loss of conventional UHMWPE (24 mm³/10⁶) and x-linked UHMWPE (16 mm³/10⁶) are within the range. This indicates that the clinical wear of artificial hip joints could be reperformed in vitro by the HSMS₄₂₀₃.

The linearity of wear rates on the testing cycles is highly dependent (Figure 3). This indicates that this method can provide a good prediction of the wear loss. The wear rate of x-linked UHMWPE is almost 60% of UHMWPE. In wear testing of orthopedic implants, a reliable method requires the capability to distinguish between original design purpose and obtain a clear comparison. It was found that the difference in the tribological performance of artificial hip joints can be distinguished in HSMS₄₂₀₃, even between conventional and modified biomaterials, i.e., UHMWPE. Furthermore, it was also found that the volumetric wear of the x-linked UHMWPE acetabular cup is lower than the normal UHMWPE acetabular cup at the product level, which is in accordance with the wear tested at material level (Hua et al., 2014). As shown in Figure 4, after 5 million cycles test, the UHMWPE acetabular cup's wear surfaces are burnished, which are typical surface morphologies under multi-direction hip motion (Saikko, 2005). The surface of the x-linked UHMWPE cup is not as well-burnished, but more cross scratches have been observed, meaning that less material is worn out, referring to its lower wear rate.

For the whole test period, the frictional force tends to be stable, except for every start after the interval. It can be two times larger than the stable value. The cause of this phenomenon could be that at the start the head center is not self-centered yet. So if the slide track is deviated, even edge-effects can exist, making the loading pressure vibrate or go up sharply. Another possible cause could be that after the running-in period, the friction pair is burnished, and therefore, can form better lubrication conditions.

The frictional force in stable period is of the same order as Saikko (2009) and can be obviously distinguished between the two tested samples. The frictional force of the x-linked UHMWPE acetabular cup is nearly 10% lower than the UHMWPE's, which shows that the trend is in accordance with the volumetric wear, making the new friction measurement module valid for the biotribological investigation of hip prostheses. Correspondingly, friction of the implant product proves to be a key factor affecting the wear performance in joint articulation.

In future work, more studies can be carried out via this newly developed hip joint simulator (HSMS₄₂₀₃). The new lubricants, such as bionic joint fluids (Hua et al., 2007), could be evaluated both for friction and wear under a specified duration, under body-simulated conditions. Moreover, product-level tests for some advanced surface technologies aimed at decreasing friction in hip joints, such as super-lubrication technology (Jiao et al., 2018), could be performed.

A new wear test apparatus, hip joint simulator (HSMS₄₂₀₃) has been developed for the biotribological investigation of artificial hip joints. The HSMS₄₂₀₃ is in healthy condition after 10-million-cycle tests. The highly linear wear rates observed in the experiments agree well with the clinical results. The difference in wear performance of artificial hip wear couples (CoCrMo on UHMWPE and CoCrMo on x-linked UHMWPE) can be distinguished. Moreover, the HSMS₄₂₀₃ provides a potential solution toward the friction monitoring to obtain more dynamic data in artificial joint wear tests.

ZH corresponding author, who has designed the hip simulator. HJ codesigner of the hip simulator. FT performed the experiment. XW and XX instructed the experiments in the manuscript. LG and XH instructed the friction measuring in the experiment. ZJ instructed the design of hip simulator. PD co-performed of the validation test and did the wear and friction measurement.