END GAME

The Failure of Total Hip Replacement

Graham Isaac PhD, Development Manager Hips

What causes failure ?

Generally hips fail for Infection or Recurrent Dislocation, (both of which result in failure
after a relatively short period of time), or finally aseptic loosening the predominant cause of
long term failure. It results from a breakdown of fixation, the main cause of which is
considered to be the body's reaction to wear debris.

"The results show that UIIIVIWPE wear products alone can cause massive
osteolysis by triggering foreign body granuloma formation"
Willert et al : (1990)

".... a high wear rate precludes a long service life"

Isaac et al : (1992)

"Osteolysis caused by polyethylene wear particles is the ultimate cause of failure"

Eftekhar : (1993)

Conclusion:- Wear is a major problem.

How does wear appear ?

Because the majority of joints have a polyethylene acetabular component, wear usually
involves the generation of polyethylene debris. This wear occurs by the femoral head
'tunnelling' its way into the acetabular cup. The diameter of the 'tunnel' is the same as the
femoral head. This means that penetration of the femoral head correlates to wear and that
the volume of wear debris generated can be calculated from penetration (assuming that creep
is a relatively small amount).

What are the alternatives ?

There are two bearing surface combinations in use which do not have a polyethylene
component:- Ceramic-on-ceramic and metal-on-metal, both of which have identical surfaces
articulating against each other - which would be normally considered to be bad engineering
practice.

PROTECTED DOCUMENT. DOCUMENT SUBJECT TO PROTECTIVE ORDER . DEPUY050669539

 Ceramic-on-Ceramic

Whilst this combination can perform well in the laboratory it is clear from the literature that
in a clinical situation things can go wrong, and produce large volumes of wear debris.

"It is clear from these failures that incorrect angulation of the acetabular cup
leads to problems of exaggerated wear over relatively large surface areas."
Nevelos et al (1993)

Furthermore, the results of survivorship studies are certainly no better and in some cases
significantly worse than joints with a polyethylene component.

"Survivorship analysis depicted a 97.5 % rate at 5 years , a 89.4 % at 10 years and

86.2% at 11 years."
Sedel et al (1994)

Metal-on-Metal

It is clear from the literature that the survivorship of Cobalt Chrome, Metal-on-Metal
prostheses in the past has been far from satisfactory.

McKee-Farrar Prosthesis 27.5% at 20 years

August et al 1986
Stanmore Prosthesis 53% at 11 years
Dobbs 1980

This poor survivorship has been attributed to poor component design and poor tolerancing of
the femoral head and acetabular cups. Despite this a small but significant number of
components function for many years and upon retrieval the bearing surfaces are undamaged
and have worn relatively little. It was also clear that all successful components were so
called 'polar bearing', i.e. contact is made at the pole, not at the equator of the bearing
surface. Manufacturing methods have improved to the extent that all components can be
made thus, and together with an 'improved' Cobalt-Chrome material, a modern stem design,
and a cushioned cup had lead its proponents to suggest that metal-on-metal is the bearing
surface of the future. However, simulator testing of such components suggest that their
performance is as unpredictable as ever, working well for a period of time before suffering
a sudden catastrophic breakdown of the bearing surface accompanied by a release of a large
volume of wear debris. Why should this be?

"..... if the clearance between the ball and the cup has been selected and
manufactured optimally a phase of steady wear with stable friction and wear
ratios at a low level after 0.5 - 1 million cycles follows".
Streicher et al (1992)

PROTECTED DOCUMENT. DOCUMENT SUBJECT TO PROTECTIVE ORDER. DEPUY050669540

 Thus the metal-on-metal bearing surface relies upon the manufacture and maintenance of an
appropriate clearance between head and cup to give a high quality bearing surface after a
period of 'Bedding-in'. However, in a clinical situation the maintenance of the bearing
surface quality is far from certain.

Incidence of damage to Cobalt- Chrome femoral heads . 54 heads retrieved at

revision surgery. Fine scratches 1-10m deep were found:-

11 out of 23 cemented (48%)

25 out of 31 uncemented /hydrid (80%)
Jasty et al (1994)

So even when the surfaces are manufactured optimally there is a mechanism' whereby the
surface can become damaged, the bearing surface breakdown, and large volumes of wear
debris produced.

It is also clear that whilst volumetric wear rate provides a useful comparison of different
materials combinations the effect of that wear debris is very different.

"CoCr produced a more florid foreign body and chronic inflammatory reaction
and was associated with decreased ingrowth of bone compared to particles of
polyethylene"
Goodman (1994)

and just to emphasise the point

"... the only combination of materials which is likely to give rise to toxic levels
of metal under clinical conditions is Cobalt-Chrome articulating against itself".

Rae (1981)

Therefore whilst CoCr components may be manufactured optimally there is no guarantee that
this quality can be maintained and if the surfaces do breakdown the debris produced has been
shown to be more harmful then polyethylene.

What is the solution?

What is the problem?

The more appropriate question is what is the problem? Consider the long term results with
metal-on-polyethylene joints.

85% non-revision at 20 years Schulte et al (1993)

84% survivorship at 20 years Kavanagh et al (1994)
89.3% survivorship at 20 years Neuman et al (1994)
86% non-revisions at 20 years Johnsson et al (1994)
89% non-revisions at 10 years Onsten et al (1994)
91% non-revisions at 11 years Malchau et al (1.993)

PROTECTED DOCUMENT. DOCUMENT SUBJECT TO PROTECTIVE ORDER. DEPUY050669541

 Whilst the problem of wear is present, the results are still excellent when compared with the
results from metal-on-metal given above. When dealing with patients, a modification to
something that is very successful would appear to be a safer and more appropriate solution.

Problem :- Femoral Heads

A major problem is in vivo damage to metal femoral heads. (Isaac et al, 1992; Jasty et at,
1994). Relatively small amounts of damage can produce dramatic effects:-

".... a single transverse scratch can increase the wear rate of the polyethylene to
a remarkable extent."
Dowson et al (1987)

However, ceramic femoral heads have been shown to behave in significantly different ways:

"A ceramic femoral head is not only less prone to damage by particles of bone
cement in three body wear processes , but any damage produced is less likely to
cause increased wear rates of UHMWPE acetabular cups."

Cooper et al (1991)

This has been shown to have significant effects on clinical wear rates:

Wear rate , metal head :- 0.10mm per year

Wear rate , ceramic head:- 0.03mm per year
Schuller and Marti (1990)

It has also been shown that the new generation of Zirconia ceramics perform significantly
better as a bearing surface than the current generation of Alumina ceramics offering a further
45 % reduction in polyethylene wear (Kumar et al, 1991)

It has further been shown that femoral head-size affects volumetric wear rates.

"Volumetric wear rates , ....., were found to increase in a linear manner with
component diameter."
Kabo et al (1993)

The increased fracture toughness of Zirconia means that a ceramic 22.225mm head is now
available.

PROTECTED DOCUMENT. DOCUMENT SUBJECT TO PROTECTIVE ORDER. DEPUY050669542

 Solution :- Zirconia 22.25mm Head

Simulator tests

Zirconia v. PE (22.225mm) 3mm3 per year (PE wear debris)

Derbyshire et al (1994)

CoCr v. CoCr (28mm) 5.1mm3 per year (CoCr wear debris)

Sulzer Technical Publication (1993)

Zirconia clinical experience:- analysis of samples after 2 - 3 years in vivo - no change in

mechanical properties or surface finish (Cales and Stefani 1994).

Problem :- Variable Quality Polyethylene

"Statistically significant correlations were observed between the extents of defects,

cracking , delamination , total wear and duration in vivo .
Wrona et al (1994)

Solution :- Enhanced Polyethylene - Hylame

Still polyethylene but with enhance mechanical properties.

Wear resistance 30% improvement

Compressive strength 75% improvement
Oxidation resistance 100% improvement
Dupont Data on File

Hylamer clinical experience:- 5 year in vivo no reported problems.

Conclusion

The use of Zirconia Ceramic heads and Hylamer enhanced polyethylene offers significant
reductions in the amount of wear debris without significantly affecting the modus operandi
of an already highly successful system of joint replacement. The alternatives have either
mixed results (Ceramic-on-Ceramic) or are building upon past failures (Metal-on-Metal) with
partial solutions in the hope of future success.

- which option would you favour?

Final Thought

"All prostheses fail sometime . It is a race between the life of the patient and the
life of the prosthesis."
Mallory 1988

PROTECTED DOCUMENT. DOCUMENT SUBJECT TO PROTECTIVE ORDER. DEPUY050669543

 References

August, Adlam, Pynsent: The McKee Farrar Hip Arthroplasty a Long Term Study.
JBJS, 68B:520, 1986

Cales, Stefani: Mechanical Properties and Surface Analysis of Retrieved Zirconia Hip Joint
Heads after an Implantation time of two to three years.
J. Mater Sci., Mater Med, 5:376, 1994

Cooper, Dowson, Fisher, Jobbins: Ceramics Bearing Surfaces in Total Artificial Joints:
Resistance to Third Body Wear Damage from Bone Cement Particles.
J.Med Engng Tech, 15:63, 1991

Derbyshire, Fisher, Dowson, Hardaker, Brummitt: Comparative Study of the Wear of

UHMWPE with Zirconia Ceramic and Stainless Steel Femoral Heads in Artificial Hip Joints.
Med Eng Phys, 16:229, 1994

Dobbs: Survivorship of Total Hip Replacement

JBJS, 62B: 168, 1980

Dowson, Taheri, Wallbridge: The Role of Counterface Imperfections in the Wear of

Polyethylene.
Wear, 119:277, 1987

Eftekhar: Joint Hip Arthroplasty, 1993

Goodman: The Effects of Micromotion and Particulate Material on Tissue Differentiation

Acta Orthop Scan, S 258:65 1994

Isaac, Wroblewski, Atkinson, Dowson: A Tribological Study of Retrieved Hip Prostheses

Clin. Orthop, 276:115, 1992

Jasty, Bragdon, Lee , Hanson, Harris: Surface Damage to Cobalt-Chrome Femoral Head
Prostheses
JBJS, 76B : 73, 1994

Johnson, Franzen, Nilsson: Combined Survivorship and Multivariate Analysis of Revisions

in 799 Hip Prostheses
J. Arthro, 9:3, 1994

Kabo, Gebhard, Loren, Amstutz: In vivo Wear of Polyethylene Acetabular Components

JBJS, 75B: 254, 1993

Kavanagh, Wallrichs, Dewitz, Berry, Currier, Ilstrup, Coventry: Charnley Low Friction
Arthroplasty of the Hip. Twenty year Results with Cement
J. Arthro, 9:229, 1994

PROTECTED DOCUMENT. DOCUMENT SUBJECT TO PROTECTIVE ORDER. DEPUY050669544

 Kumar, Oka, Ikeuchi , Shimizu, Yamamuro , Okumura, Kotoura : Low Wear Rate of
UHMWPE Against Zirconia Ceramic (Y-PSV) in Comparison to Alumina Ceramic and SUS
316L Alloy
J.Bio Mater Res, 25: 813, 1991

Malchau, Herberts, Ahnfelt: Prognosis of Total Hip Replacement in Sweden

Acta Orthop Scan, 64: 497, 1993

Mallory: Mallory Head Hip System

Current Concepts of Joint Replacement , Ed. Greenwald 1988

Neumann, Freund, Sorenson : Long Term Results of Charnley Total Hip Replacement
JBJS, 76B: 245, 1994

Nevelos , Evans, Harrison , Rainforth : Examination of Alumina Ceramic Components from

Total Hip Arthroplasties
Proc I Mech E, 207 : 155, 1993

Onsten, Besjakov , Carlsson: Improved Radiographic Survival of the Charnley Prosthesis in

Rheumatoid Arthritis and Osteoarthritis
J. Arthro, 9 : 3, 1994

Rae: The Toxicity of Metals used in Orthopaedic Prostheses

JBJS, 63B : 435, 1981

Schuller, Marti : Ten Year Socket Wear in 66 Hip Arthroplastics : Ceramic versus Metal
Heads
Acta Orthop Scan, 61:240, 1990

Schulte, Callaghan , Scott, Johnston : The Outcome of Charnley Total Hip Arthroplasty with
Cement after a Minimum 20 year Follow-up
JBJS, 75A: 961, 1993

Sedel, Nizard , Kerboull, Witroet : Alumina-Alumina Hip Replacement in Patients Younger

than 50 years old.
Clin. Orthop . 298:175, 1994

Streicher, Weber , Schoen, Semlitsch : Wear Resistant Couplings for Longer Testing
Articulating Total Joint Replacement
Biomaterial-Tissue Interfaces, Advances in Biomaterials 10, Ed . Doherty: 179. 1992

Willert, Bertram , Buchhorn: Osteolysis in Alloarthroplasy of the Hip

Clin. Orthop ., 258:95, 1990

Wrona, Mayor, Collier , Jensen: The Correlation between Fusion Defects and Damage in
Tibial Polyethylene Bearings
Clin. Orthop ., 299:92, 1993

PROTECTED DOCUMENT . DOCUMENT SUBJECT TO PROTECTIVE ORDER . DEPUY050669545