Session 1 - Biology of Biomaterials - Gateway Ballroom103, Sun 9:30 AM - 11:00 AM 47th Annual Meeting, Orthopaedic Research Society, February 25 - 28, 2001, San Francisco, California 0001

WEAR PARTICLE ANALYSES OF CONVENTIONAL AND CROSSLINKED UHMWPE TESTED IN AN ANATOMIC HIPSIMULATOR

+*Scott, M (E-Smith and Nephew); *Widding, K (E-Smith and Nephew); **Ries, M (E-Smith and Nephew); ***Shanbhag, A (A-Smith and Nephew)+*Smith and Nephew, Inc., Memphis, TN. 1450 Brooks Rd., Memphis, TN 38116, 901-399-5553, Fax: 901-399-6020, marcus.scott@smith-nephew.com

INTRODUCTION: Numerous forms of crosslinked UHMWPE, whichdemonstrate dramatic reductions in hip simulator gravimetric wear, have beendeveloped [1, 2] and used clinically with the intent to reduce particle-inducedosteolysis. It is generally believed that gravimetrically measured reductions inwear translate into reductions in particle generation. The purpose of this studywas to investigate the relationship between gravimetric wear volume and wearparticle characteristics (size, surface area, and volume). We compared oneconventional and two variations of crosslinked UHMWPE in this study.

METHODS: Anatomic hip simulator (AMTI, Watertown, MA) tests werecarried out to 10 million cycles on the following materials: (i) conventionalUHMWPE (C-PE), (ii) 5 MRad crosslinked UHMWPE (5-XPE), and (iii) 10MRad crosslinked UHMWPE (10-XPE). Ram extruded GUR 1050 material(PolyHi Solidur, Ft. Wayne, IN) was the starting material for all tests.Crosslinking was carried out at gamma irradiation doses of 5 and 10 MRad(SteriGenics, Gurnee, IL), followed by melt annealing (2 hrs at 150 oC) andslow cooling. Acetabular liners (32 mm ID) were machined from bar stock,followed by EtO sterilization. The 5-XPE and 10-XPE liners were artificiallyaged at 70 oC and 5 atm O2 for 3 weeks prior to testing [3]. Hip simulatortesting (n=3 for each group) was carried out against 32 mm CoCrMo heads in100% bovine serum. The testing was interrupted periodically for weightmeasurements and serum replacement. Wear particles were harvested fromtest serum using a previously validated acid digestion/ vacuum filtrationprotocol [4]. The particles deposited on the 0.05-µm filter membranes werecharacterized under a scanning electron microscope (SEM) at magnificationsof 1,000X and 20,000X. A minimum of 20 random, non-overlapping fieldsand 100 particles were imaged to ensure that the detected particles wererepresentative of the entire particle population within each serum sample. Foreach material, the mean particle diameter was determined and the followingparameters were calculated per million cycles: (i) number of particles, (ii)surface area of particles, and (iii) volume of debris generated. Particlediameter, surface area, and volume were calculated assuming sphericalgeometry. ANOVA and Duncan’s multiple range tests were used todetermine significant differences (α = 0.05) in mean particle diameter,number of particles, surface area of particles, and volume of debris generatedbetween the material conditions.

RESULTS: The gravimetric wear rates decreased as the crosslinkingradiation dose increased. For the C-PE, the wear rate was 36.9 mg/Mcycles,which decreased to 9.0 mg/Mcycles for the 5-XPE, which further decreased to–1.1 mg/Mcycles for the 10-XPE (Table 1). Based on SEM micrographs, theC-PE particles were predominantly submicron spheroids, with occasionalfibrils 5 to 10 µm in length (Figure 1). The 5-XPE and 10-XPE particles wereexclusively submicron spheroids (Figure 1).

C-PE 5-XPE 10-XPE

Fig. 1. SEM micrographs of UHMWPE particles extracted from serum(______ = 1 µµµµm).

In addition to the highest gravimetric wear rate, the C-PE material exhibitedthe largest particle diameter, surface area, and volume of debris generated(p<0.05 for all combinations of pairs, Table 1). Particle diameter and thesurface area and volume of particles decreased with increasing crosslinkingradiation dose. Interestingly, the 5-XPE material generated the highestnumber of particles, resulting in twice the number of particles per Mcyclesthan C-PE (Table 1). The 10-XPE material generated less than half thenumber of particles per Mcycle compared to C-PE.

TABLE I. Gravimetric Wear Rate, Particle Diameter, Surface Area, Volume,and Number for the Tested Materials: Mean ± 95% Confidence Interval

C-PE 5-XPE 10-XPEGravimetric Wear Rate(mg/Mcycle)

36.94±0.48

9.02±0.55

-1.11±0.22

Particle Diameter (µµµµm) 0.206±0.012

0.118±0.003

0.091±0.003

Particle Surface Area(m2/Mcycle)

1.17±0.13

0.69±0.12

0.09±0.02

Particle Volume(mm3/Mcycle)

95.84±14.69

22.45±3.45

4.16±1.17

Particle Number perMcycles (1x1012)

5.76±0.96

12.05±1.88

2.28±0.25

DISCUSSION: Increasing the crosslinking radiation dose resulted in a morewear resistant polyethylene as tested in our anatomic hip simulator, consistentwith previous reports [1, 2]. The surface area and volume of particlesdecreased with increasing radiation dose. Particle size (diameter) alsodecreased with increasing radiation dose. Due to different particle sizedistributions, a unique relationship between gravimetric wear and particlenumber existed for each tested material. As a result, the reduction ingravimetric wear for the 5-XPE material did not translate into a reduction inparticle number when compared with the C-PE material.

The mass loss due to wear for the 10-XPE liners was masked by differences influid absorption between the soak control and test specimens. As a result, the10-XPE liners showed a net weight gain. Particle analysis, however, showedthat small, but measurable volumes of wear particles were generated. Wearparticle analysis may thus provide a more direct measurement of the volumeand number of particles generated from highly crosslinked UHMWPE andmay be used to supplement gravimetric measurements for low wear materials.

Macrophage response to particulate wear debris is believed to be an importantfactor in osteolysis. It is well established that the cellular response isdependent upon particle number, size, surface area, and material chemistry,among other factors [5-7]. In the current study, differences in particlenumber, size, and surface area were observed among conventional andcrosslinked UHMWPE. Cell culture testing is currently underway todetermine the inflammatory response to these materials.

REFERENCES: [1] McKellop, J Ortho Res, 17(2), 157-67 (1999). [2]Muratoglu, 45th ORS Trans, 829 (1999). [3] Sanford, ORS Trans, 119 (1995).[4] Scott, 6th World Biomat Cong Trans., 177 (2000). [5] Shanbhag, ClinOrtho, 342, 205-17 (1997). [6] Green, Biomaterials, 19(24), 2297-302(1998). [7] Gonzalez, J Biomed Mater Res, 30(4), 463-73 (1996).

**Univ. of California, San Francisco, San Francisco, CA.***Biomaterials Laboratory, Massachusetts General Hospital, HarvardMedical School, Boston, MA.