Other Articles

Biological Responses to PEEK-Based Wear Debris


A study conducted by Rush University Medical Center in Chicago compares biological response to particles of PEEK with those of ultra high molecular weight polyethylene. Mark Brady of Invibio Biomaterial Solutions reports the findings.

An ageing and notably active population is driving unprecedented demand for high-performance biomaterials in the design and engineering of implantable medical devices. At the same time, heightened patient and surgeon expectations are challenging material suppliers and implant manufacturers to develop next-generation orthopaedic and spinal implants. They are meeting this challenge with innovative designs that advance surgical options and extend patient benefits. Moreover they are relying on advanced biomaterials with characteristics and performance properties exceeding those of conventional materials to realise their designs.

One high-performance thermoplastic, polyetheretherketone (PEEK) is increasingly used for a growing variety of implantable device applications, from spine (where it was first introduced) to orthopaedic, trauma, dental and cardiovascular. Its mechanical properties, flexible design options, radiolucency and ability to tailor its wear and modulus have broadened options for its use as an alternative to conventional materials.

Demanding applications

In spinal surgery, where PEEK has experienced evolving uses since its introduction in the 1990s in interbody fusion cages, the material’s radiolucency and tailored modulus close to that of bone offer significant advantages over conventional titanium. Since those early days, there has been increasing interest in the use of PEEK in more demanding applications such as standalone cages and dynamic stabilisation rods, as well as in next generation nucleus and total disc replacement. As an example of the growth in this sector, disc replacement surgery grew by 48.9% in the year 2009 to 2010 in the US.1

Carbon fibre (CF) reinforced PEEK materials have demonstrated low wear properties in self-mating wear couples2 as well as against hard counterfaces including ceramic3 and metal.4 These favourable low wear properties position CF reinforced PEEK as an attractive alternative to the use of ultra high molecular weight polyethylene against cobalt chrome molybdenum (UHMWPE/CoCrMo) in total and partial joint replacement.5

The biocompatibility of PEEK and PEEK compound bulk materials has long been established6 and have undergone rigorous testing to meet stringent criteria for their US Food and Drug Administration (FDA) Master Files. However, an inevitable consequence of using any biomaterial in articulating devices for motion preserving spinal implants or total joint replacements is the potential for generating wear debris.

The changing regulatory landscape

Increasingly, regulatory bodies including the US FDA are requiring a more comprehensive evaluation of the wear debris produced by medical devices. In addition to wear rates from simulator testing, characterisations of the particulate debris with regard to particle size and shape distribution, number and chemistry of particles are being requested. In some cases, debris representative of that produced during wear simulations must be assessed in small animal models to examine the local and systemic effects. Demonstrating a limited degree of wear and biological response to wear debris are critical factors in the regulatory approval of these devices.

Biological response to wear debris

Since the introduction of the Charnley hip prosthesis in the 1960s, the generation of significant wear debris has been an instrumental factor in the onset of aseptic osteolysis, and ultimately, in failure due to implant loosening.7 Indeed, aseptic osteolysis followed by implant loosening is the most common cause of failure of total joint replacements.8 Adverse biological responses to implant wear debris can result in an attempt to wall off foreign material through the formation of granulomatous tissue, characterised by an organised collection of macrophages, which may contain fused macrophages (multinucleated giant cells) at its core. Disruption of the contact between the implant and bone surface further contributes to bone loss and loosening of the implant.

Although a number of peri-prosthetic cell types, including fibroblasts and osteoblasts, are negatively affected by wear debris, the macrophages are responsible for the ensuing pro-inflammatory response. Professional phagocytes, macrophages are responsible for engulfing and digesting cellular debris and invading pathogens. However, when macrophages encounter polymeric debris, a process of frustrated phagocytosis occurs, whereby particles are engulfed, but the process of digestion is incomplete, which results in the release of pro-inflammatory cytokines including TNF-α, IL-1β, IL-6 and IL-8. Together with those released by osteoblasts and stromal cells, these potent mediators stimulate the differentiation and maturation of osteoclast precursors. Osteoclasts are the main bone resorbing cells of the body, and disruption of the delicate homeostasis with bone forming osteoblasts leads to progressive osteolysis. Revision of total hip arthroplasties due to periprosthetic bone loss have revealed that the number of macrophages is directly associated with the degree of osteolysis (Figure 1).9
Figure 1. A macrophage attempting to
engulf a UHMWPE particle

What influences the response to wear debris?

Historically, much work has concentrated on the response to UHMWPE particulate debris because of its widespread use in hip and knee arthroplasty. Numerous explant studies have demonstrated that, although there is a wide range of particle sizes found in the peri-prosthetic tissue of failed total hip replacements, the vast number of particles fall into the micron and sub-micron range.10  A wealth of studies also show that macrophages are associated with particles of this size in peri-prosthetic tissues (particles within the phogocytosable size range) and that particles of this size are also the most biologically active in vitro with respect to cytokine production.11 A more accurate definition of the likely biological response also takes into account the total volume of debris, and together these represent the wear debris dose or load (the number of phagoctysoable particles per unit volume of tissue). It has been suggested that harder-wearing materials such as metals or ceramics be selected for articulating surfaces. Although these harder-wearing materials produce a lower wear volume, the particles are generally smaller than those produced from polymeric materials; thus, the potential for producing large numbers of particles remains.

A second determinant of biological activity of wear debris is thought to be related to shape, with more elongated, fibrous material generally being more pro-inflammatory than spherical or globular material. Less clear, however, is the ratio of length to width (aspect ratio) at which this becomes a critical issue.
Finally, chemical composition plays a key role in the pro-inflammatory potential of the wear debris. Which materials are more pro-inflammatory remains a topic of debate, yet there is a developing consensus that metals are generally more bioactive than polymers. Added to this are the growing concerns over metal ion release and the rare, but severe metal hypersensitivity reactions following some metal-on-metal hip replacements.

Biological response to PEEK in vitro

PEEK and CF reinforced PEEK compounds may offer an alternative as self-mating wear couples or as an alternative to polyethylene in metal-on-polymer bearings for articulating medical devices. Thus far, the inflammatory potential of PEEK polymer wear debris has not been directly addressed.
Invibio Biomaterial Solutions has recently performed a study in conjunction with Rush University Medical Center (Chicago, Illinois, USA, www.rush.edu) to determine the biological response to PEEK-OPTIMA particles compared with UHMWPE. The hypothesis was that PEEK particles of a size and morphology relevant to those produced during wear simulations would be less inflammatory to macrophages than UHMWPE. To test this hypothesis, human macrophages were challenged with increasing doses (particles per cell) of particles and the effect on cytotoxicity and production of pro-inflammatory cytokines measured.

PEEK-OPTIMA polymer and a commonly used, implantable grade UHMWPE underwent sterile cryo-milling, pulverisation and filtering to create predetermined particle sizes in the sub-micron range. These particles then underwent further analysis by low angle laser light spectroscopy to provide measurements including size and aspect ratio. Finally, the particles were verified as endotoxin-free, because the presence of endotoxin would induce an inflammatory response from the macrophages and mask any material-specific effects. By these methods PEEK-OPTIMA and UHMWPE particles were generated with mean sizes of 0.7 ΅m and 0.5 ΅m, respectively, and had a similar granular to flake-like shape (aspect ratios from 1.1 to 1.5, that is, round to oval).

Cultured human macrophages were exposed to increasing ratios of particles to cells (1:1, 10:1 and 100:1) for 24 and 48 hours (n=3 for each material, dose and time). Cytotoxicity was determined by lactate dehydrogenase (LDH) release from the cells (a measure of cell membrane damage). Culture media from particle-challenged cells were also collected and analysed for particle-induced IL-1β, IL-6, IL-8, MCP-1 and TNF-α expression by a pro-inflammatory Luminex multiplex array (Invitrogen, www.invitrogen.com).

Following particle challenge, UHMWPE particles elicited a significant increase in cytotoxicity after 24 hours compared with PEEK-OPTIMA (Figure 2a). This difference was observed for each ratio of particles to cells. No dose-related increase in cytotoxicity was evident for PEEK, but a dose response was observed in macrophages with UHMWPE particles. After 48 hours, LDH release was more comparable for the two materials, but notably, cytotoxicity attributed to the high dose (20:1) of UHMWPE particles remained significantly elevated (Figure 2b).

Figure 2: Cytotoxicity of human macrophages challenged with PEEK-OPTIMA and UHMWPE at doses of 1, 10, and 20 particles per cell after a) 24 hours, and b) 48 hours

In general, challenge of macrophages with PEEK and UHMWPE materials resulted in a limited inflammatory response, with IL-6 levels in the culture media of particle-challenged cells remaining below the level of detection. After 24 hours, TNF-α levels were largely equivalent for both materials, although they were slightly elevated for UHMWPE at the highest particle load (20:1) (Figure 3a). Similarly, MCP-1 levels remained below the level of detection, with the exception of UHMWPE at 20:1 (Figure 3b). Expression of cytokines IL-1β (Figure 3c) and IL-8 (Figure 3d) were greatest for UHMWPE compared with PEEK at the low (1:1) and high (20:1) dose of particles, but was equivalent or greater for PEEK at a dose of 10:1.

After 48-hour challenge of macrophages with particles, expression levels of cytokines were largely equivalent for the two materials at the low dose, with the exception of MCP-1, which was increased for UHMWPE (Figure 3b). Despite the high degree of variability seen for PEEK, expression levels of IL-1β, TNF-α, MCP-1 and IL-8 in UHMWPE-challenged cells were raised compared with PEEK at doses of 10:1 and 20:1, and generally showed a greater sensitivity to the increase in particle load.

Figure 3: PEEK-OPTIMA and UHMWPE induced cytokine expression in macrophages following challenge with 1, 10, and 20 particles per cell at 24 and 48 hours. a) TNF- , b) MCP-1, c) IL-1 , d) IL-8

In summary, both materials elicited a relatively low inflammatory response from macrophages following particle challenge. Cytotoxic effects and inflammatory cytokine response were generally more evident following exposure to UHMWPE, and displayed a greater sensitivity to increasing dose of UHMWPE compared with PEEK. This in vitro study is in accordance with a previous in vivo study in which PEEK particles showed no adverse response in the spine.12 Despite the relative limitations of in vitro studies, this work suggests that PEEK-OPTIMA particles are more biocompatible than UHMWPE particles and demonstrate that PEEK-OPTIMA implant debris does not represent an inherent increased inflammatory risk over that of UHMWPE.

Biological response to CF reinforced PEEK compounds

Both clinical cases and animal models examining wear and biological response to CF reinforced PEEK have revealed only small amounts of debris and a limited inflammatory response, equivalent to that of UHMWPE.

As part of a study by Latif et al.,13 CF reinforced PEEK particles were used in a rat pouch model of biocompatibility. Although some minimal inflammation was observed, the biological response showed equivalence to that of PE. A clinical study using the ABG II Hip System (Stryker Orthopaedics, Mahwah, New Jersey, USA, www.stryker.com) was initiated in 2001, in which the ABG II acetabular inserts were fabricated from PEEK compounded with 30% pitch carbon fibres (MOTIS polymer, Invibio Biomaterial Solutions). After a mean follow-up of three years, none of the liners had been revised due to aseptic loosening and analysis of one liner, retrieved due to infection from a 55 year-old, highly active male patient, revealed head penetration of 0.13 mm after 28 months. Histology demonstrated a "low” amount of particles in the peri-prosthetic tissue from this patient with no necrotic areas, and only a limited inflammatory reaction.14 

More recently a mouse knee model was used to examine the biological response to CF reinforced forms of PEEK compared with UHMWPE. This study demonstrated the reproduction of sterile particles with a size distribution and morphology representative of those generated through knee wear simulators. The inflammatory response elicited by the particles was assessed and revealed that the biological reactivity to CF reinforced PEEK compounds were equivalent to that of UHMWPE.15 

PEEK-OPTIMA and CF reinforced compounds of this material have demonstrated biocompatibility in bulk forms and as particulate debris. The additional benefits with regard to wear properties and radiolucency presented by these biomaterials offers designers and engineers new options in the development of motion-preserving spinal implants and total joint replacement for the future.


1. 2010 Spinal Surgery Update,
www.orthopedicnetworknews.com, 2 November 2010.
2. S.C. Scholes, A. Unsworth, "The Wear Performance of PEEK-OPTIMA Based Self-Mating Couples,” Wear, 268, 3–4, 380–387 (2009).
3. S.C. Scholes, A. Unsworth, "The Wear Properties of CFR-PEEK-OPTIMA Articulating Against Ceramic Assessed on a Multidirectional Pin-on-Plate Machine,” Proc. Inst. Mech. Eng. H, 221, 3, 281–289 (2007).
4. S.C. Scholes, A. Unsworth, "Wear Studies on the Likely Performance of CFR-PEEK/CoCrMo for Use as Artificial Joint Bearing Materials,” J. Mater. Sci. Mater. Med., 20, 1, 163–170 (2009).
5. T.M. Grupp et al., "Biotribology of Alternative Bearing Materials for Unicompartmental Knee Arthroplasty,” Acta Biomater., 6, 9, 3601–3610 (2010).
6. D.F. Williams et al., "Potential of Polyetheretherketone (PEEK) and Carbon-Fibre-Reinforced PEEK in Medical Applications,” J. Mater. Science Letters, 6, 188–190 (1987).
7. H.G. Willert, M. Semlitsch, "Reactions of the Articular Capsule to Wear Products of Artificial Joint Prostheses,” J. Biomed. Mater. Res. B Appl. Biomater., 11, 2 ,157–164 (1977).
8. M.J. Archibeck  et al., "The Basic Science of Periprosthetic Osteolysis,” J. Bone Joint Surg. Am., 82, 10, 1478 (2000).
9. T.P. Schmalzried et al., "Periprosthetic Bone Loss in Total Hip Arthroplasty. Polyethylene Wear Debris and the Concept of the Effective Joint Space, J. Bone Joint Surg. Am., 74, 6, 849–863 (1992).
10. E. Ingham, J. Fisher, "The Role of Macrophages in Osteolysis of Total Joint Replacement,” Biomater., 26, 11, 1271–1286 (2005).
11. T.R. Green et al., "Polyethylene Particles of a Critical Size are Necessary for the Induction of Cytokines by Macrophages In Vitro,” Biomater., 19, 24, 2297–2302 (1998).
12. C.H. Rivard et al., "In Vivo Biocompatibility Testing of PEEK Polymer for a Spinal Implant Aystem: A study in Rabbits,” J. Biomed. Mater. Res., 62, 4, 488–498 (2010).
13. A.M. Latif et al., "Pre-Clinical Studies to Validate the MITCH PCRtrade mark Cup: A Flexible and Anatomically Shaped Acetabular Component with Novel Bearing Characteristics,” J. Mater. Sci. Mater. Med., 19, 4, 1729–1736 (2008).
14. N. Pace et al., "Technical and Histologic analysis of a Retrieved Carbon Fiber-Reinforced Polyetheretherketone Composite Alumina-Bearing Liner 28 Months after Implantation,” J. Arthroplasty, 23, 1, 151–155 (2008).
15. S. Utzschneider et al., "Inflammatory Response against Different Carbon Fiber-Reinforced PEEK Wear Particles Compared with UHMWPE In Vivo,” Acta Biomater., 6, 11, 4296–4304 (2010).

Dr Mark Brady is Product Development Project Manager at Invibio Biomaterial Solutions,
Invibio Ltd, Technology Centre, Hillhouse International, Thornton Cleveleys, FY5 4QD, UK, tel. +44 (0)1253 898 000,
e-mail: mbrady@invibio.com


If you found this article interesting please
Sign Up Now for your free subscription to
Med-Tech Innovation

Related Articles

Choosing Materials for Medical Devices


Using the right materials can be critical to medical device performance. Dr Sarah Egan of Granta Design outlines technologies that can help designers with the important materials selection process.... read more >

Supercritical Carbon Dioxide Sterilisation


A gentle sterilisation process has been commercialised for the sterilisation of advanced biomaterials. This article describes the technique and relevant applications... read more >

Rapid Ethnographic Evaluation


It is accepted that ethnographic enquiry, the study of user behaviour in its social and cultural context, forms part of effective user-centric design, but the technique can be difficult and costly to implement. Dr Andrew Walters of PDR at the University of Wales Institute Cardiff describes developments that are making the technique more accessible.... read more >

Copyright © 2014 Med-Tech Innovation
Submit Editorial | Media Kit | Privacy Policy | Terms of Service | Site map