Cơ xương khớp pdf

Laboratory testing has demonstrated that highly cross-linked polyethylene has markedly improved wear resistance compared with conventional polyethylene under a variety of conditions.. Am

Highly Cross-linked Polyethylene in Total Hip Arthroplasty

Abstract

Although total hip arthroplasty is a common and highly successful procedure, its long-term durability has been undermined by the cellular response to polyethylene wear debris and the subsequent effects on periprosthetic bone Research elucidating the effects of sterilization on polyethylene wear has facilitated the development

of a more wear-resistant material—highly cross-linked polyethylene Laboratory testing has demonstrated that highly cross-linked polyethylene has markedly improved wear resistance compared with conventional polyethylene under a variety of conditions Early clinical data have supported these results To make informed decisions about this already widely available and frequently used product, the practicing orthopaedic surgeon should have a basic understanding of the production process as well as knowledge of the most current laboratory and clinical data

one of the most successful sur-gical procedures ever developed Ce-mented and cementless component fixation provide excellent pain relief, return of function, and intermediate longevity in patients with degenera-tive conditions of the hip Ultra-high–molecular-weight polyethylene (UHMWPE) articulating with a metal head has been the predominant bear-ing surface since the inception of modern THA Although the success

of this bearing couple is well docu-mented, clinical studies and retrieval analyses have shown that polyethyl-ene wear and osteolysis are the ma-jor factors limiting the longevity of THA.1-5

Extensive research undertaken to elucidate the physical and biologic mechanisms behind polyethylene

wear and osteolysis6,7has led to the development of highly cross-linked UHMWPE The term highly cross-linked polyethylene is commonly used to describe this new generation

of polymers We use this term to de-scribe intentionally cross-linked ma-terial; however, different manufac-turers use proprietary methods to produce various levels of cross-linking in their components Labora-tory work and early clinical trials

cross-linked polyethylene is signifi-cantly more wear-resistant than con-ventional polyethylene A synthesis

of the manufacturing processes, in addition to laboratory and clinical data regarding highly cross-linked UHMWPE, may aid the orthopaedic surgeon in making an informed deci-sion regarding THA

Alexander C Gordon, MD

Darryl D D’Lima, MD

Clifford W Colwell, Jr, MD

Dr Gordon is Orthopaedic Surgeon,

Illinois Bone and Joint Institute, Morton

Grove, IL Dr D’Lima is Director,

Orthopaedic Research Laboratories,

Scripps Center for Orthopaedic

Research and Education, La Jolla, CA.

Dr Colwell is Director, Musculoskeletal

Center; Director, Scripps Center for

Orthopaedic Research and Education;

and Shiley Chair, Orthopaedic

Research, Scripps Center for

Orthopaedic Research and Education.

None of the following authors or the

departments with which they are

affiliated has received anything of value

from or owns stock in a commercial

company or institution related directly or

indirectly to the subject of this article:

Dr Gordon, Dr D’Lima, and Dr Colwell.

Reprint requests: Dr Gordon, Illinois

Bone and Joint Institute, 9000

Waukegan Road, Morton Grove, IL

60053.

J Am Acad Orthop Surg 2006;14:

511-523

Copyright 2006 by the American

Academy of Orthopaedic Surgeons.

Polyethylene Resins

and Manufacturing

Polyethylene Resin

Polyethylene is simply a

repeat-ing chain of ethylene monomer

mol-ecules; the modifiers low-density,

high-density, and

ultra-high–molec-ular weight refer to the molecultra-high–molec-ular

weight, chain length, and

arrange-ment of the polymer chains The

condensed polymers have crystalline

and amorphous regions, the

percent-age and arrangement of which affect

the properties of the material (Figure

1) In general, polymers with higher

percentages of crystalline regions

have higher elastic moduli and

dem-onstrate better resistance to crack

propagation, but they may be more

susceptible to the effects of

oxida-tion.8

Ruhrchemie AG, a predecessor

company of Ticona, began

commer-cial manufacture of UHMWPE resin

in the 1950s Ticona is currently the

leading manufacturer of

medical-grade UHMWPE resin, with plants

in Bishop, Texas and Oberhausen,

Germany All Ticona resins have the

designation “GUR,” followed by a

numeric modifier Their

medical-grade resins are named GUR 1020,

1120, 1050, and 1150 The first of the

four numerals (1) indicates that the

polymer is designated for ortho-paedic implantation The second nu-meral notes the presence (1) or ab-sence (0) of calcium stearate The third digit is an indicator of molecu-lar weight, and the fourth is an inter-nal corporate code

Hercules Powder manufactured another UHMWPE resin known as the 1900 series Most recently pro-duced by Basell Polyolefins, the 1900 series line (1900 and 1900H) has re-mained the same as when Hercules produced it In 2002, Basell sold the

1900 resin technology and ceased production of this product Although new 1900 resin is not being pro-duced, some orthopaedic device manufacturers have stockpiled the material and continue to use it in their implants The 1900 series res-ins have a lower mean molecular weight and larger mean particle size than do the GUR 1050 resins, which may affect their clinical perfor-mance

Edidin et al9and Won et al10 stud-ied the effects of resin type and man-ufacturing method on the wear and degradation of the 1900 and GUR resins Won et al10analyzed retrieved tibial bearings from the Miller-Galante (MG) I and II (Zimmer, War-saw, IN) knee arthroplasty designs

The tibial bearings had the same

ge-ometry; both were gamma-sterilized

in air The MG-I bearings were made from direct compression-molded

1900 resin, and the MG-II compo-nents were manufactured from ex-truded GUR 415 stock The re-searchers found notably higher rates

of delamination and subsurface damage consistent with oxidation in the MG-II components They con-cluded that compression-molded

1900 resin was more resistant to ox-idation than GUR material

Edidin et al9 studied differences between 1900 and GUR resins in the laboratory setting using an

accelerat-ed aging technique to determine sus-ceptibility to oxidation They found that after accelerated aging and

gam-ma sterilization in air, compression-molded GUR 1050 and 1900 resins

as well as extruded 1050 stock de-graded similarly in mechanical test-ing The authors noted more rapid degradation of the 1900 resins than

of the GUR resins under similar test-ing conditions, but only minor dif-ferences in the post-aging mechani-cal properties of the three resins

Implant Fabrication

Implant manufacturers obtain UHMWPE as powdered resin or stock material from converting com-panies, such as Poly Hi Solidur (Fort

Figure 1

A,Molecular structure of ethylene and of ultra-high–molecular-weight polyethylene (UHMWPE) n = the degree of

polymerization B, Crystalline and amorphous regions of UHMWPE (Reproduced with permission from Kurtz SM: The

UHMWPE Handbook: Ultra-High Molecular Weight Polyethylene in Total Joint Replacement San Diego, CA: Elsevier

Academic Press, 2004, pp 4, 6.)

Wayne, IN) and Perplas Medical

(Bacup, Lancashire, UK)

Compo-nents are fabricated from the resin

by direct compression molding or

machined from converted stock

sup-plied by ram-extruded bars or

mold-ed sheets The mechanical

proper-ties of the final product are affected

by the specific temperature,

pres-sure, and cooling rate used in the

compression-molded components made from

1900 resin have demonstrated

excel-lent clinical performance despite

be-ing sterilized by gamma radiation in

air In his study of direct

compres-sion-molded components in the hip

and knee, Ritter12found osteolysis in

2.5% of hips at a mean of 21 years

and in 0% of knees at a mean of 8

years He concluded that the clinical

performance of direct

compression-molded 1900 resin is superior to that

of other polyethylene components

Because there is no clear consensus

on the best resin or fabrication

method for UHMWPE bearings in

THA, orthopaedic implant

manufac-turers decide which resin and

fabri-cation method best suits their

im-plants

Sterilization and Aging

of Conventional

Polyethylene

Polyethylene THA bearings are

ster-ilized by one of two general

meth-ods—surface treatment and

irradia-tion These methods, as well as

numerous other variables in the

ster-ilization process, have specific

ef-fects on the in vitro and in vivo

per-formance of UHMWPE acetabular

liners

Surface Treatment

The two commonly used surface

sterilization treatments are ethylene

oxide (EtO) gas and gas plasma

Al-though highly toxic, EtO is

well-suited for polyethylene because the

gas does not chemically react with

the component Safe and effective

EtO sterilization requires special

en-vironmental conditions during ster-ilization and appropriate timing to allow the gas to diffuse in and out of the component Gas plasma treat-ment is performed at a lower tem-perature and in a shorter time frame than EtO surface sterilization In gas plasma treatment, less toxic sub-stances (eg, peracetic acid, hydrogen peroxide gas plasma) are used to eliminate potential contamination

This method is newer than EtO, and data regarding its use are limited.11

Irradiation

Gamma radiation and its effects

on the mechanical properties of polyethylene have been well docu-mented, with a resultant large-scale overhaul of polyethylene production for THA Irradiation of polyethylene causes cleavage of the polymer chains, leading to the production of free radicals After radiation, the chains may bond at their original scission point or cross-link with one another When neither occurs, the cleaved end of the polymer chain re-mains a free radical When steriliza-tion and packaging of the compo-nent take place in the presence of oxygen, the free radicals generated

by the radiation are able to combine with oxygen molecules during stor-age and after implantation (Figure 2)

This leaves the component suscepti-ble to the effects of oxidation, which are now known to adversely affect its mechanical properties

In a retrieval analysis of compo-nents from multiple manufacturers, Sutula et al13investigated the sub-surface white band found in their re-trievals Infrared spectroscopy dem-onstrated that this subsurface white band corresponded to an area of high oxidation and was present only in components sterilized by gamma ir-radiation in air The appearance of this band was time-dependent, and all components in which the white band was observed had been steril-ized more than 3 years before the ob-servation The authors found that the presence of the subsurface white

band corresponded with decreased tensile strength, severe embrittle-ment of the subsurface zone, and an increased incidence of rim cracking and delamination in retrieved liners (Figure 3)

ex-amined the effects of sterilization method, calcium stearate addition, and thermal aging on the wear per-formance of UHMWPE in two hip simulator studies Before initiating the studies, all irradiated samples re-ceived a mean dose of 2.7 Mrad Despite differences in molecular weight and the presence or absence

of calcium stearate, gas plasma–ster-ilized components demonstrated wear rates comparable with each other Among components not sub-jected to accelerated aging, the EtO-sterilized samples had significantly higher wear rates than those

steril-ized with gamma radiation in air (P

= 0.0001) or in a vacuum (P = 0.0001).

Additionally, components that were

Figure 2

Effects of irradiation in an oxygen environment on UHMWPE

(Reproduced with permission from Greenwald AS, Bauer TW, Ries MD, Committee on Biomedical Engineering, Committee on Hip and Knee Arthritis: New polys for old: Contribution or

caveat? J Bone Joint Surg Am 2001;

83(suppl 2):27-31.)

gamma radiated in air had

signifi-cantly higher wear rates than did

those irradiated in a vacuum (P =

0.01) (Figure 4)

After thermal aging, all

gamma-irradiated cups, including those

ster-ilized with methods to decrease

oxidation (eg, ion implantation,

ni-trogen packaging, oxygen scavenger)

demonstrated oxidative degradation

and a subsequent increase in wear

The unsterilized and gas plasma–

treated cups wore at the same rates

before and after thermal aging, and

both types of cups showed no

oxida-tion After accelerated aging, cups

that were gamma-irradiated in air

had the highest wear rates of all test

specimens (Figure 5) The

investiga-tors concluded that prior to aging,

the cross-linking effect of

radia-tion—even in an oxygen

environ-ment—provided improvements in

wear compared with components

that were never sterilized or were

EtO-sterilized After oxidation and

Figure 5

Wear rates of six types of polyethylene for two cycle intervals after artificial aging for 14 days at 80°C All gamma-sterilized cups had a mean radiation dose of 2.7 Mrad (Reproduced with permission from McKellop H, Shen FW, Lu B, Campbell P, Salovey R: Effect of sterilization method and other modifications on the wear resistance of acetabular cups made of ultra-high molecular weight polyethylene: A

hip-simulator study J Bone Joint Surg Am 2000;82:1708-1725.)

Figure 3

Photographs of Charnley components

that were never implanted The section

without the band (top) was never

sterilized The section with the

pronounced white band (bottom) was

sterilized by gamma radiation in air

14 years earlier (Reproduced with

permission from Sutula LC, Collier JP,

Saum KA, et al: The Otto Aufranc

Award: Impact of gamma sterilization

on clinical performance of polyethylene

in the hip Clin Orthop Relat Res

1995;319:28-40.)

Figure 4

Wear rates of six types of polyethylene for two cycle intervals without artificial aging All gamma-sterilized cups had a mean radiation dose of 2.7 Mrad (Reproduced with permission from McKellop H, Shen FW, Lu B, Campbell P, Salovey R: Effect of sterilization method and other modifications on the wear resistance of acetabular

cups made of ultra-high molecular weight polyethylene: A hip-simulator study J

Bone Joint Surg Am 2000;82:1708-1725.)

embrittlement of the polymer,

how-ever, the advantage of irradiation is

lost

Sychterz et al16studied

steriliza-tion variables in a clinical setting

They reviewed radiographs of

pa-tients who had undergone

cement-less THA fixation whose

conven-tional acetabular liners had been

sterilized with (1) gamma radiation

in air, (2) gamma radiation in a

vac-uum and barrier-packaged, or (3) gas

gamma-irradiated in a vacuum and

those irradiated in air wore at

signif-icantly lower rates than did those

sterilized by gamma radiation in air

or gas plasma (P < 0.01) The authors

also concluded that the cross-linking

provided by gamma sterilization,

even in air, provided better wear

re-sistance than did gas plasma in

con-ventional polyethylene

Before the introduction of highly cross-linked polyethylene, two prod-ucts meant to be improvements on conventional polyethylene were mar-keted but subsequently discontin-ued—highly crystalline UHMWPE (Hylamer, DePuy, Warsaw, IN) and carbon fiber–reinforced polyethylene (Poly II, Zimmer) Hylamer has been more extensively studied than Poly

II in THA, with reports of Hylamer wearing at rates comparable to those

of conventional polyethylene.15,16 De-spite this finding, some studies17,18

indicate high wear rates and severe osteolysis in patients implanted with Hylamer liners sterilized by gamma radiation in air In a retrieval analy-sis, Collier et al19 noted that for a given level of oxidation, Hylamer lin-ers that were gamma-sterilized in air sustained more wear and damage than did conventional polyethylene

sterilized in the same manner They suggested that the increased crystal-linity of Hylamer makes it more sus-ceptible to oxidation than conven-tional polyethylene Most reports of Poly II are from the knee arthroplasty literature, but one report of carbon fi-ber polyethylene in THA discussed two instances of severe tissue reac-tion and prosthetic loosening associ-ated with this material.20

Highly Cross-linked Polyethylene

Manufacturing

The highly cross-linked compo-nents available for implantation are machined from ram-extruded bar stock of GUR 1050 resin Although the exact methods are proprietary and differ among manufacturers, the steps to produce cross-linked poly-ethylene follow the same general sequence: radiation cross-linking, thermal treatment, and terminal sterilization21(Figure 6)

The first step is a cross-link–induc-ing radiation dose of 2.5 to 10 Mrad provided by cobalt 60 (gamma) or an electron beam source This is fol-lowed by thermal treatment, in which the polyethylene is heated be-low, at, or above its melting temper-ature, depending on the manufac-turer This step is meant to quench free radicals, allowing the polyethyl-ene chains to preferentially cross-link, thus diminishing the chances for ox-idative degradation The heating methods, which are proprietary, may

be combined with electron beam ir-radiation because this process mea-surably heats the polymer The final step is terminal sterilization and bar-rier packaging Terminal sterilization

of these components is usually a sur-face treatment, but some manufactur-ers use a sterilizing dose of gamma ra-diation in an inert atmosphere

In a study attempting to deter-mine the effects of these specific steps, Muratoglu et al22found

high-er levels of free radicals and more post-aging oxidation in polymers

Figure 6

1050 Extruded rod

Machine cup Radiation

(1) 5 Mrad (2) 10 Mrad

7.5 Mrad radiation

125 C Warming oven

Warming oven

3 Mrad

Sterilize, N2

Heat above melt (>135 C) Heat anneal Electron beam9.5 Mrad Electron bea m10 Mrad

Heat anneal in

package

Machine cup Machine cup Heat above melt

(>135 C)

Sterilize (1) Gas plasma (2)

2.5 Mrad Sterilize,

N2/Vacuum

Machine cup Machine cup

Ethylene oxide sterilize

Gas plasma sterilize

Process Heat

stabilized

CISM (cold irradiated subsequent melt)

CIAN (cold irradiated adiabatic non-melt)

WISM (warm irradiated subsequent melt)

Longevity Durasul

Zimmer Stryker

Howmedica Osteonics

Crossfire (1) Marathon

(2) XLPE (1) DePuy/

Johnson & Johnson (2) Smith+Nephew

Duration

Product

Company Stryker

Howmedica

Osteonics

°

Heat above melt C)

WIAM (warm irradiated adiabatic melting)

(>135

Ethylene oxide

Zimmer

°

Processing steps for highly cross-linked polyethylene, by manufacturer

(Reproduced with permission from Greenwald AS, Bauer TW, Ries MD, Committee

on Biomedical Engineering, Committee on Hip and Knee Arthritis: New polys for

old: Contribution or caveat? J Bone Joint Surg Am 2001;83(suppl 2):27-31.)

treated with sub-melt temperature

annealing and terminal gamma

ster-ilization (Crossfire; Stryker

How-medica Osteonics, Mahwah, NJ)

than in those that were melted and

gas sterilized after cross-linking

radi-ation (Longevity; Zimmer) In a

retrieval analysis of explanted

cross-linked liners from different

manu-facturers, Bhattacharyya et al23

hy-pothesized that Crossfire would

show more in vivo oxidation than

melt-stabilized polyethylene, such

as Longevity or Durasul (Zimmer)

Within 3 years of implantation, the

authors found elevated oxidation

levels and one component with a

subsurface white band among the

Crossfire liners; they did not detect

any oxidation in the other two types

Bhattacharyya et al23concluded that

the free radicals formed by sub-melt

temperature annealing and gamma

sterilization can lead to in vivo

oxi-dation

Laboratory Studies

Laboratory studies have

demon-strated that higher degrees of

cross-linking improve wear resistance and

decrease particulate volume in a hip

femoral head size causes increased volumetric wear rates in hips im-planted with conventional

highly cross-linked polyethylene have demonstrated greatly dimin-ished wear compared with conven-tional polyethylene in liners articu-lating with 22-, 28-, 32-, and 46-mm heads

cross-linked liners with nominally cross-linked liners articulating with 28- and 32-mm femoral heads The highly cross-linked liners had been sub-melt temperature annealed and sterilized with gamma radiation; the nominally cross-linked liners were polyethylene that was

conventional-ly sterilized by gamma radiation in nitrogen The 28- and 32-mm highly cross-linked liners had significantly

(P < 0.001) less wear than did their

conventional counterparts, but the wear of 28- and 32-mm highly cross-linked cups did not differ

significant-ly (Figure 7) The authors concluded that larger femoral head size may

not be predisposed to increased wear

in highly cross-linked liners Muratoglu and colleagues27,28have extensively studied electron beam cross-linked, melt-annealed, and EtO-sterilized (Durasul) UHMWPE They studied the mechanical proper-ties, oxidation levels, effect of femo-ral head size, and wear rates com-pared with those of conventional polyethylene The authors found markedly less wear of the highly cross-linked liners compared with gamma-sterilized/inert implants for femoral head sizes ranging from 22 to

46 mm After weighing the compo-nents, they determined that there was no detectable wear from the highly cross-linked specimens and that the head penetration noted was solely the result of plastic deforma-tion This was corroborated by the presence of machining marks on the cross-linked liners after 20 million cycles; these marks had been worn away on the conventional polyethyl-ene specimens The mechanical and

showed no oxidation after acceler-ated aging and no evidence of free radicals, but it did demonstrate a de-crease in ultimate tensile strength (UTS) and yield strength compared with gamma-sterilized/inert polyeth-ylene Despite the inferior mechan-ical properties of the highly cross-linked polyethylene, the testing results fell well within American So-ciety for Testing and Materials (ASTM) standards for medical-grade UHMWPE However, ASTM stan-dards do not imply that a polyethyl-ene component is suitable for clini-cal use and do not include a specification for fracture toughness The diminished crack propagation re-sistance of cross-linked polyethylene may have clinical implications Other researchers have tested highly cross-linked polyethylene un-der more adverse conditions, such as wear in the presence of a third body

or a rough countersurface One study comparing gamma/nitrogen–cross-linked, barrier-packaged

polyethyl-Figure 7

Cumulative wear rates of highly cross-linked (X) and nominally cross-linked (O)

acetabular liners articulating with 28- and 32-mm heads (Reproduced with

permission from Hermida JC, Bergula A, Chen P, Colwell CW Jr, D’Lima DD:

Comparison of the wear rates of twenty-eight and thirty-two-millimeter femoral

heads on cross-linked polyethylene acetabular cups in a wear simulator J Bone

Joint Surg Am 2003;85:2325-2331.)

ene articulating with femoral heads

of differing surface roughness

report-ed significantly (P = 0.004) less wear

of the cross-linked liners.29These

roughened balls (surface roughness,

0.9 µm) during and after the initial

wear-in period had wear rates higher

than that of the conventional

poly-ethylene articulating with a smooth

surface, thus negating the effects of

cross-linking on wear

Although it is not known

wheth-er this degree of roughening occurs

in vivo, Minakawa et al30attempted

to quantify the third-body damage of

retrieved femoral heads They

deter-mined that cobalt-chrome heads

could suffer varying degrees of

dam-age; cobalt-chrome heads had a

mean surface roughness of 0.4 µm

on their most damaged areas The

authors found a single component

with damage >2.0 µm, which

sug-gests that the conditions in the

study by McKellop et al29could

oc-cur in vivo

Bragdon et al31compared the wear

resistance of gamma/nitrogen and

cross-linked polyethylene in an

en-vironment of

polymethylmethacry-late (PMMA) or alumina third-body

particles As expected, the

speci-mens with the alumina particles

wore much more than did those

with PMMA or without third-body

particles Although the authors

found that the presence of a very

hard third body (eg, alumina)

affect-ed the wear of cross-linkaffect-ed

polyeth-ylene, PMMA particles had a very

small effect The cross-linked liners

in this study demonstrated

signifi-cantly (P < 0.0001) less wear than did

conventional polyethylene in all

testing conditions Taylor et al32also

studied the effects of PMMA

parti-cles on wear of cross-linked

polyeth-ylene They found lower wear in the

cross-linked specimens than in

con-trols; however, they did note

signif-icant surface damage and wear rates

that were much higher than those

reported by Bragdon et al.31

Although much research has

fo-cused on the wear rates of cross-linked polyethylene, other reports have focused on the characterization

of the wear particles generated dur-ing these tests Ingram et al33 mea-sured the size of wear particles pro-duced by wearing 5- and 10-Mrad cross-linked polyethylene against smooth and rough surfaces, then tested their biologic activity by de-termining the levels of tumor

macro-phages cultured with the wear debris They found that increased levels of cross-linking, associated with wear from the rougher surface, led to a higher percentage of debris

in the submicron range and

against a smooth surface resulted in nanometer-sized particles in non-and cross-linked specimens, thus de-creasing their biologic activity

These data suggest that although ab-solute wear is decreased with cross-linking, the particles generated are biologically active and have the po-tential to induce osteolysis

cross-linked acetabular liners from six US orthopaedic implant manufacturers

to determine the effects of the differ-ing manufacturdiffer-ing techniques on the mechanical properties, crystal-linity, and pre- and post-aging oxida-tion levels of the various compo-nents (Table 1) Their goal was to determine the properties of

clinical-ly available poclinical-lyethylene liners and relate those properties to the wear rates published by the manufactur-ers The authors did not do a head-to-head comparison of wear rates

These cups were subjected to an ac-celerated aging protocol Before ag-ing, all test cups showed no to low initial oxidation rates; however, Du-rasul, Crossfire, and ArCom did have higher “as received” oxidation levels than did a standard reference poly-ethylene After accelerated aging, the Longevity, Crossfire, and ArCom liners demonstrated significantly

as-received counterparts (P < 0.01),

while the others had no change in oxidation level The Longevity liners had the lowest initial oxidation

lev-el of the six test specimens, and the authors thought that its increase in oxidation after aging was not enough

to affect its mechanical properties Mechanical testing demonstrated

a range of UTS (34 to 59 MPa) and a smaller range of tensile strength at yield point (19 to 24 MPa) for the as-received components (Table 2 ) Af-ter aging, ArCom, Durasul, and Crossfire liners demonstrated a de-crease in UTS, and ArCom, Reflec-tion, and Crossfire liners showed sig-nificant differences in yield point compared with their as-received counterparts (Table 3) All materials tested exceeded the ASTM standard (27 MPa) for UTS and tensile stress

at yield point (19 MPa) in non–cross-linked polyethylene

Collier et al34found that the UTS

of the materials was stratified by ra-diation dose; the components

receiv-ing >5 Mrad had significantly (P <

0.01) lower values than did the refer-ence polyethylene before aging In contrast, the tensile strength at yield point was stratified by the heating method rather than radiation dose The components that were heated at

or above their melting temperature

showed significantly (P < 0.01)

low-er values than did the reflow-erence ma-terial The investigators concluded that even intentionally cross-linked polyethylene is not immune to oxi-dation and free-radical formation; the varying oxidation levels and sus-ceptibility to oxidation after aging were dependent on the processing conditions The authors also stated that increasing the radiation dosages appears to produce lower wear rates,

as reported by the manufacturers, but also results in lower toughness

in-cludes a paragraph from each manu-facturer regarding the rationale for the manufacturing processes used in the production of its components Subsequent to the 2003 publica-tion of the study by Collier et al,34

Smith & Nephew increased the

cross-linking radiation dose from 5

to 10 Mrad; Stryker Howmedica

Os-teonics introduced a new

cross-linked polyethylene (X3) that is

se-quentially irradiated and annealed;

and Biomet began production of

Ar-Com XL, an intentionally

cross-linked and heat-stabilized version of

its direct compression-molded

Ar-Com product No published clinical

or laboratory results are available on

these updated products

Concern about the loss of fracture

toughness and more brittle nature of

highly cross-linked polyethylene has

been the topic of numerous reports

Baker and colleagues36-38conducted

several studies to elucidate the fa-tigue resistance and fracture tough-ness of cross-linked polyethylene

Their hypothesis was that the

high-er degree of cross-linking, leading to

a restriction of chain mobility in the amorphous regions of polyethylene, would decrease the plasticity of the polymer, resulting in a material that

is less resistant to crack propagation

The true stress at break point and the resistance to crack propagation were inversely related to the cross-linking radiation dose and were at-tributed to a decrease in plasticity at the fracture tip The results of these studies suggest that for clinical situ-ations in which stress

concentra-tions and surface defects may exist,

a lower degree of cross-linking may

be safer

Clinical Evaluation and Retrieval Analysis

Extensive laboratory data exist on the wear resistance of cross-linked polyethylene, but few clinical ies are available The published stud-ies (Table 4) are generally consistent with the laboratory data, but long-term follow-up is not yet available Three randomized, prospective eval-uations of highly cross-linked poly-ethylene with 2-year clinical

follow-up have been published.39-41

In a study of cemented THA,

Di-Table 1

Cross-linked Material Tested by Collier et al 34

Material

Radiation Source

Dose to

ArCom (Biomet,

Warsaw, IN)

1900H Direct compression-molded

or machined from molded bar

Gamma 2.5 to 4 Mrad None

Marathon

(DePuy,

Warsaw, IN)

1050 Machined from extruded

bar

(150°C)

Reflection XLPE

(Smith &

Nephew,

Memphis, TN)

(subsequently changed to 10 Mrad)

At melt temperature (136°C)

Durasul

(Zimmer,

Warsaw, IN)

1050 Machined from

compression-molded sheet

Electron beam

9.5 Mrad Above-room-temperature

pre-heat before electron beam; melt anneal; controlled heat and cooling rates; warm irradiation with adiabatic melting Crossfire

(Stryker

Howmedica

Osteonics,

Mahwah, NJ)

(subsequently irradiated with 3 Mrad)

Below melt temperature (>120°C)

Longevity

(Zimmer)

1050 Compression molded and

machined

pre-heat before electron beam; process between cold irradiation with subsequent melt and warm irradiation with adiabatic melting Adapted with permission from Collier JP, Currier BH, Kennedy FE, et al: Comparison of cross-linked polyethylene materials for

orthopaedic applications Clin Orthop Relat Res 2003;414:289-304.

gas et al39 compared Durasul with

conventional polyethylene sterilized

by gamma irradiation in nitrogen

Head penetration into the liner was

evaluated with radiostereometric

analysis (RSA) at 1 and 2 years

post-operatively Head penetration seen

on supine radiographs was similar

between groups at 1 and 2 years, but

was approximately 50% less in the

highly cross-linked group in the

same time period as seen on

stand-ing radiographs Additionally, no

dif-ference was found between groups

with respect to component

migra-tion or the appearance of radiolucent

lines Most of the head penetration

in the first year was attributed to plastic deformation of the socket

In another study,40these same in-vestigators presented their 3-year data with the cemented liners and introduced a new study of bilateral hybrid THAs The cemented cup study results were similar to those

in the previous report, with lower head penetration rates in the highly cross-linked group at 3-year

follow-up The hybrid hip study used an RSA method to compare Longev-ity liners with polyethylene that was compression-molded,

gamma-irradiated in nitrogen, and

implant-ed into a cementless cup During the first year, head penetration rates of the two polyethylenes were not sig-nificantly different, but at 2 years,

significantly (P < 0.0005) less head

penetration was observed in the cross-linked components The au-thors concluded that the similar

ear-ly head penetration rates generalear-ly reflect creep and not wear

Using a digital radiographic tech-nique in a randomized, prospective evaluation with 2-year follow-up,

with polyethylene irradiated in

ni-Table 2

Mechanical Properties of As-Received Acetabular Liners

Cross-linked

Material

Yield Point (MPa)

Probability Value*

Ultimate Tensile Strength (MPa)

Probability Value* Elongation (%)

Probability Value*

Reflection

XLPE

HSS

Reference

UHMWPE35

*Probability values are for the t-test between the cross-linked materials and the Hospital for Special Surgery (HSS) reference

ultra-high–molecular-weight polyethylene (UHMWPE).

Adapted with permission from Collier JP, Currier BH, Kennedy FE, et al: Comparison of cross-linked polyethylene materials for

orthopaedic applications Clin Orthop Relat Res 2003;414:289-304.

Table 3

Mechanical Properties of Acetabular Liners After 28 Days of Artificial Aging

Cross-linked

Material

Yield Point (MPa)

Probability Value*

Ultimate Tensile Strength (MPa)

Probability Value* Elongation (%)

Probability Value*

Reflection

XLPE

*Probability values are for the t-test between the “as received” and aged cross-linked material properties

Adapted with permission from Collier JP, Currier BH, Kennedy FE, et al: Comparison of cross-linked polyethylene materials for

orthopaedic applications Clin Orthop Relat Res 2003;414:289-304.

trogen and barrier-packaged The

au-thors noted a marked (40% to 50%)

decrease in the two-dimensional

lin-ear, two-dimensional volumetric,

and three-dimensional linear wear

rates in the highly cross-linked

group Head penetration seen in the

first year after implantation was

mostly caused by plastic

deforma-tion, not by true wear

Heisel et al42performed a

nonran-domized study comparing Marathon

cross-linked polyethylene with

con-ventional polyethylene sterilized by

gamma irradiation in air; they found

an 81% decrease in volumetric wear

in the cross-linked group after 2

years Using regression analysis to

control for the differences between

groups, they determined that the

type of polyethylene was the only

significant variable influencing

vol-umetric wear rates

The study with the longest

follow-up to date, published by Dorr

gamma/nitrogen polyethylene after 5

years of clinical use In a

retrospec-tive study of 37 Durasul hips

matched to historical controls, direct

radiographic measurements were

used to calculate the mean annual

head penetration rates; the

investiga-tors found that a digital

measure-ment technique did not provide

accu-rate data The “bedding-in” period for

Durasul was approximately 2 years,

while that of the conventional poly-ethylene was 1 year From 2 to 5 years, the linear wear rate of Durasul was approximately 50% less and the annual head penetration rate was 60% to 75% less than conventional polyethylene during the same period

In a prospective, nonrandomized study using RSA, Rohrl et al44 com-pared wear rates of cemented stems articulating with either cemented gamma/air or Crossfire polyethylene

In contrast with other studies, the bedding-in period was only 2 months, and the wear rates were linear for both groups thereafter From 2 to 24 months, an 85% reduction in wear was noted in the Crossfire group, and cross-linked polyethylene

demon-strated significantly (P < 0.001) lower

wear rates than did gamma/air poly-ethylene without increased migra-tion or radiolucencies

In all of the aforementioned stud-ies, wear rates were lower for cross-linked polyethylene than for controls

Larger differences between conven-tional and cross-linked polyethylene were found when the controls were gamma-sterilized in air versus in an inert environment, again demonstrat-ing the inferior wear characteristics

of gamma/air polyethylene

In addition to the clinical studies,

in vivo behavior of highly cross-linked polyethylene after a

relative-ly short service life has been studied

using retrieval analysis Bradford et

al45studied 21 cross-linked Durasul liners revised 2 to 24 months after implantation Pitting, scratches, and surface cracking were common find-ings, but no liners demonstrated bur-nishing or severe wear (Figure 8) The authors postulated that the cracking was likely the result of the diminished ductility and fatigue re-sistance of the polymer and

conclud-ed that the in vivo wear patterns of highly cross-linked polyethylene dif-fer from those occurring in a hip simulator The significance of this finding is that hip simulators did not accurately predict the in vivo perfor-mance of a given material

Other researchers attribute a dif-ferent significance to the surface find-ings of explanted cross-linked liners, however Muratoglu et al46also stud-ied liners not revised for wear with a service life of 2 weeks to 10 months They used a melt-recovery technique

to test their hypothesis that the sur-face scratching represented plastic de-formation, not true wear Their most common findings were light and heavy surface scratches; a few spec-imens had polished areas The melt-recovery process was used to recover the machining marks, if present Five

of the seven liners treated with this process had complete or near-complete recovery of the original ma-chining marks The authors

con-Table 4

Summary of Highly Cross-linked Polyethylene Clinical Studies

Author

Cross-linked Polyethylene (Fixation)

Conventional Polyethylene

Follow-up (years)

Wear Reduction of Cross-linked

Polyethylene

(cemented)

Longevity (hybrid)

60% to 75% less head penetration