Biomaterials 21 (2000) 907}913

The in-vivo wear performance of prosthetic femoral heads

with titanium nitride coating
Manuela Teresa Raimondi, Riccardo Pietrabissa*
Laboratory of Biological Structure Mechanics, Dipartimento di Bioingegneria, Politecnico di Milano, Milano, Italy
Received 10 November 1998; accepted 21 October 1999

Abstract

This paper reports the study performed on four titanium nitride (TiN) coated prosthetic femoral heads collected at revision surgery
together with patient data. Surface topology has been examined using Scanning Electron Microscopy (SEM) and elemental analysis of
both coating and substrate have been evaluated using energy-dispersive X-ray spectrometry. Quantitative assessment of the surface
topography is achieved using contacting pro"lometry. The average R roughness value is calculated at "ve di!erent locations for each
!
femoral head. The UHMWPE counterface worn volume has been measured directly on the acetabular components. TiN fretting and
coating breakthrough occurred in two of the four components examined. In the damaged coating areas the surface pro"le is
macroscopically saw-toothed with average tooth height 1.5 lm. The average R value is 0.02 lm on the undamaged surfaces and
!
0.37 lm on the damaged ones. Failure of the coating adhesion resulted in the release of TiN fragments and of metallic particulate from
the substrate fretting corrosion and in the increase of the head surface roughness a!ecting counterface debris production. Our results
suggest that TiN-coated titanium alloy femoral heads are inadequate in the task of resisting third body wear mechanisms
in vivo. ( 2000 Elsevier Science Ltd. All rights reserved.

Keywords: Hip arthroplasty; Femoral head; Coating; Titanium nitride; UHMWPE; Wear

1. Introduction highly polished Co}Cr}Mo alloy-UHMWPE and ce-

Wear particles from total joint replacements are
thought to play a critical role in implant failure by
generating periprosthetic bone lysis phenomena, as a re-
sult of a debris-induced macrophage in#ammatory re-
sponse [1}7]. Increased durability of a hip replacement
therefore appears to be related to the reduction of the
arti"cial joint wear [6,8]. Dramatic wear rates are re-
ported in literature for several di!erent joint couplings,
namely stainless-steel alloys-polytetra#uorethylene [9]
(PTFE) and later titanium alloys-ultra high molecular
weight polyethylene (UHMWPE) [10,11]. In titanium
alloy components extensive fretting corrosion is reported
in the presence of third-body hard particles (such as
acrylic cement and/or metal particles) entrapped between
the articulating bearing surfaces. Lower polymeric wear
rates are currently achieved by the clinical application of
* Corresponding author. Tel.: #39-02-2399-3349; fax: #39-02-
2399-3360.
E-mail address: pietrabissa@biomed.polimi.it (R. Pietrabissa)
ramic-UHMWPE couplings [12,13].
Di!erent solutions aimed at improving the tribological
properties of the femoral head-acetabular cup system are
being developed. For the acetabular cup, metal [14,15]
or ceramic [16] bearing surfaces, improved UHMWPE
[17}19] and various carbon-"bre reinforced synthetic
materials [20] are under investigation. For the titanium
alloy femoral head, surface hardening and coating tech-
niques are being developed, such as nitrogen di!usion-
hardening [21], multi-layer titanium}niobium ceramic
coatings and titanium nitride (TiN) coatings [22,23].
The combination of a TiN-coating applied to a pre-
hardened titanium head is intended to combine in
a proper way the pro"table bulk material properties of
the Ti6Al4V alloy, extensively used in orthopaedic im-
plants for its biocompatibility and corrosion resistance,
with the superior fretting resistance of a hard nitride
layer [23]. The wear resistance of TiN-coated surfaces is
compared in friction and wear experimental studies with
those of standard materials [22,23]. Limited data are
available to evaluate the clinical performance of TiN-
coated femoral heads and concerns have been raised

0142-9612/00/$ - see front matter ( 2000 Elsevier Science Ltd. All rights reserved.
PII: S 0 1 4 2 - 9 6 1 2 ( 9 9 ) 0 0 2 4 6 - X
908 M. Teresa Raimondi, R. Pietrabissa / Biomaterials 21 (2000) 907}913

regarding the coating resistance to long-term wear re-

(mm3/yr)
Volume
lated damage mechanisms [24].

24

82

116
In this paper we present an analysis of four TiN-coated
femoral heads retrieved at revision after a period of

Undetectable
in vivo articulation against UHMWPE liners. The heads
Cup worn have been examined to assess the surface damage. The
(mm3)
worn volume is measured on the polymeric acetabular

143

654

931
components. A coating failure scenario is proposed and
discussed.

Loosening
Loosening

Loosening
Acetabular Diagnosis

Infection
Revision

2. Materials and methods

component

Biarticular

Biarticular

Cemented

Cemented
all-poly

all-poly
Four hip prostheses have been harvested at revision
head

head

from four patients 18 to 96 months after hip arthroplasty.

All the femoral stems were built of Ti6Al4V alloy and
Self-locking,

were implanted with modular TiN-coated femoral heads

Stem type,

cementless
cemented

cemented

cemented

articulating on UHMWPE counterfaces. Clinical and

"xation

Moore,

MuK ller,

MuK ller,

radiographic data were available for all the patients, two

of which were total hip arthroplasty (THA) cases and two
were hemiarthroplasty cases. The THA hips were
Hemiarthroplasty

Hemiarthroplasty
Type of surgery

cemented MuK ller stems articulating on cemented all-poly

cups. The hemiarthroplasty hips were one cemented sel-
#ocking stem and one cementless Moore stem, both
THA

THA

articulating on biarticular heads of the Bateman's type.

The four hips have been replaced with cemented THA
Osteoarthritis

Osteoarthritis

components. The patient sex, height, weight, activity level

Neck fracture

Neck fracture

and age at primary and at revision surgery, the primary

Diagnosis
Primary

diagnosis, the reason for removal, the prosthesis type and

stability at revision have been recorded for each patient
Revision
(yr)

71

70

79

64
Primary
Age

(yr)

69

64

71

56
Activity

Normal

High

High
Low
level
Weight
(kg)

70

55

78

60
Height

1.60

1.60

1.68

1.50
(m)
Demographic data for the components

Patient

Sex

M
F

F
(months)
in situ
Diameter Time

18

72

96

96

Fig. 1. Location of the pro"lometer scans performed on the femoral

heads. Notations Sup, Inf, Ant and Post refer to the superior, inferior,
anterior and posterior areas, respectively. Two scans are performed on
Component

(mm)

each location, in directions parallel and perpendicular to the neck axis.

22

22

32

32
Table 1

On the apex the scans are taken on the antero-posterior and medio-
No.

lateral planes.
1

4
 M. Teresa Raimondi, R. Pietrabissa / Biomaterials 21 (2000) 907}913 909

Fig. 2. SEM images of the transition area between the gold-coloured and the discoloured zones. Coating failure and isolated fragments on the exposed
substrate (a); a double edge (b); #aked fragment and the scratched titanium alloy substrate (c); detail of the porous anchorage layer binding the coating
to the substrate (d).

(Table 1). At revision all the implants were loose and were
collected without disassembling the prosthetic heads
from the femoral stems.
Surface topology has been examined using scanning
electron microscopy (SEM) and elemental analysis of
both coating and substrate have been evaluated using
energy-dispersive X-ray spectrometry. Both analyses
have been carried out using a Stereoscan S260 electron
microscope (Leica Cambridge Ltd., Cambridge, UK).
Quantitative assessment of the surface topography has
been achieved using a Form Talysurf series 2 S4C induc-
tive gauge contacting pro"lometer (Rank Taylor Hobson
910 M. Teresa Raimondi, R. Pietrabissa / Biomaterials 21 (2000) 907}913
Fig. 3. X-ray spectra are consistent with TiN on the intact coating (a)
and on the intermediate layer (b), and consistent with the Ti6Al4V alloy
on the exposed substrate (c).
Ltd., Leicester, UK), equipped with a diamond tip, 2 lm
radius and 903 cone angle. Ten pro"les for each head
have been acquired in the areas represented in Fig. 1. The
adopted cut-o! is 0.25 mm, evaluation length is 1.25 mm
and sampling is 5000 points per pro"le. The form sub-
traction adopted is a least-squares circle. The roughness
pro"les have been obtained after data "ltering [25] and
the arithmetic mean roughness R has been calculated for
!
each roughness pro"le. At the di!erent locations the
average R value is calculated from the two values ob-
!
tained in the directions parallel and perpendicular to the
neck axis.
The prosthetic heads articulated in vivo on
UHMWPE counterfaces. The UHMWPE worn volume
has been measured directly on the polymeric compo-
nents. A cloud of points has been acquired on the worn
surface using a Werth Video Check IP 250/400 coordi-
nate measuring machine (Werth Messtechnik GmbH,
Giessen, G). The coordinate data have been input in
the commercial solid modelling code Catia (Dassault
Systèmes, Paris, F) which gave the 3-D model of the
worn surface. The solid model of the worn material has
been obtained as the space between the worn and the
original cup surfaces. The worn volume has been cal-
culated too.
3. Results
Gross examination of the articulating surfaces shows
multidirectional slight scratching (3 mm or less in length)
on all the retrieved heads. Two components show a dull
grey discoloration on the gold-coloured TiN coating,
located on the anterior head region and involving about
10% of the articulating surface in one head (component
no. 1 in Table 1), located on a wide band surrounding the
complete head equator and involving about 70% of the
articulating surface in another head (component no. 3 in
Table 1).
SEM images of the transition surfaces between the
gold-coloured and the discoloured zones are shown in
Fig. 2. Isolated TiN coating fragments have been ob-
served, which appear bonded to the exposed substrate.
X-ray spectrometry (Fig. 3) produces spectra consistent
with the TiN on the intact coating and with the Ti6Al4V
alloy on the substrate, indicating extensive abrasion of
the TiN coating. At a greater magni"cation (Fig. 2c)
double-edge boundaries are observed on the TiN frag-
ments indicating #aking of the multi-layer coating. De-
tailed inspection of the #akes shows a porous anchorage
layer binding the TiN coating to the substrate. X-ray
spectrometry produces spectra consistent with a TiN on
the intermediate layer. The titanium alloy exposed sub-
strate shows severe scratching whereas slight scratching
is visible on the coating fragments and on the undamaged
coating areas.
 M. Teresa Raimondi, R. Pietrabissa / Biomaterials 21 (2000) 907}913
Fig. 4. Pro"les acquired on the undamaged (a) and on the damaged (b) coating areas (raw data are treated with least square circle form subtraction).
Fig. 4 shows two pro"lometer scans acquired on the
undamaged and on the damaged coating areas. Only
form subtraction is performed on the raw pro"le data, to
reproduce the actual shape and size of the asperities. In
the damaged coating area the pro"le is saw-toothed with
average tooth thickness about 1.5 lm. Fig. 5 shows the
average values of the R parameter calculated on the
!
roughness pro"les, for each femoral head and at the vari-
ous head locations. The average R value is 0.02 lm on the
!
undamaged surfaces and 0.37 lm on the damaged ones.
The calculated worn volumes for each UHMWPE cup
are shown in Table 1.
4. Discussion
Two of the four examined components (nos. 2 and 4)
show a macroscopically intact TiN coating 6 to 8 years
after implant. The presence of slight scratching, visible to
the naked eye, suggests that third body abrasion was
however taking place prior to implant failure. Harman
911
et al. [24] have reported about wear debris originating
from a TiN-coated femoral head as delaminated surface
asperities, after one year of in situ operating. This may be
the mechanism initiating abrasion of the TiN coating;
additional hard particles (such as acrylic cement par-
ticles) may take part in this abrasive process at a second
stage.
TiN fretting and coating breakthrough occurred in
two of the four examined components. In the presence of
hard particles entrapped between the articulating
counterfaces high local stresses can develop following
micro-contact of two hard materials under load (Hertz-
ian forces). In this occurrence, the peak stresses are
located below the contact surface (pellicular loads), and
failure of coating adhesion is likely to occur as a conse-
quence of the coating brittleness. Detachment of TiN
macroscopic fragments accelerates this failure mecha-
nism and extensive breakthrough of the coating is pos-
sible even at low loads.
Component no. 1 articulated in vivo with a biarticular
head on a patient with a low activity level and a coating
912 M. Teresa Raimondi, R. Pietrabissa / Biomaterials 21 (2000) 907}913
Fig. 5. The average values of the R parameter calculated, for each femoral head, at the superior, inferior, anterior, posterior and apex locations.
!
breakthrough initiation site was already present 1.5 years
after implant. As a matter of fact the entrapment of third
body particulate is favoured in the inner bearing of biar-
ticular joints [26]. This occurrence, associated to a rela-
tively high patient body weight, may have been the
critical failure factors for the coating of this component.
Component no. 3 articulated on a "xed socket on a high
demand heavy patient and the coating was almost com-
pletely disappeared from about 70% of the head surface
after 8 years of in situ operating. Extensive fretting cor-
rosion has been observed on the titanium alloy substrate
and this might have accelerated the periprosthetic bone
resorption and the implant failure.
The UHMWPE wear rates reported in Table 1 are
consistent with the wear data published by other authors
with reference to failed arthroplasties [13,27}29]. A com-
parison between the wear rates measured in the biarticu-
lar inner couplings and those measured in the THA
couplings is not possible, since the expected range of
relative motion is greater for the THA joints [30,31].
Various authors [32}35] have shown that a statistical
correlation is present between increased harder surface
roughness and UHMWPE worn volume. The correla-
tion is exponential according to several authors [36,37].
The shape and dimension of the asperities in the
damaged areas (Fig. 4) and the calculated R values
! Medical Mater Engng 1996;6:165}72.
(Fig. 5) suggest that the coating breakthrough can lead to
an increase of polymeric debris production rate. Further-
more, the coating fragments may favour third body wear
mechanisms. Failure of the coating adhesion results in
the following events, all proved [1}8] to adversely a!ect
the implant clinical outcome:
1. release of TiN fragments in the periprosthetic tissues;
2. release of metallic particulate from the titanium sub-
strate fretting corrosion;
3. increase of the head surface roughness a!ecting the
counterface debris production rate.
Our results suggest that TiN coated titanium alloy
femoral heads are inadequate in the task of resisting
in vivo third body wear mechanisms. Their use should
not be advocated.
Acknowledgements
The authors are grateful to Professor Giuseppe Silva of
Dipartimento di Meccanica for making available both
the SEM equipment and his competence and to Carlo
Santambrogio of Taylor Hobson S.p.A. for his help in the
pro"lometry studies.
References
[1] Harris WH. The problem is osteolysis. Clin Orthop 1995;
311:46}53.
[2] Jasty M, Bragdon C, Jiranek W, Chandler H, Maloney W, Harris
WH. Etiology of osteolysis around porous-coated cementless
total hip arthroplasties. Clin Orthop 1994;308:111}26.
[3] Buechel FF, Drucker D, Jasty M, Jiranek W, Harris WH. Os-
teolysis around uncemented acetabular components of co-
balt}chrome surface replacement hip arthroplasty. Clin Orthop
1994;298:202}11.
[4] Mohanty M. Cellular basis for failure of joint prosthesis. Bio-
[5] Revell PA, Al-Sa!ar N, Kobayashi A. Biological reaction to
debris in relation to joint prostheses. Proc Inst Mech Engng [H]
1997;211:187}97.
[6] Wroblewski BM. Wear of the high-density polyethylene socket in
total hip arthroplasty and its role in endosteal cavitation. Proc
Inst Mech Engng [H] 1997;211:109}18.
[7] McGee MA, Howie DW, Neale SD, Haynes DR, Pearcy MJ. The
role of polyethylene wear in joint replacement failure. Proc Inst
Mech Engng [H] 1997;211:65}72.
[8] Kadoya Y, Kobayashi A, Ohashi H. Wear and osteolysis in
total joint replacements. Acta Orthop Scand Suppl 1998;278:
1}16.
[9] Harris WH. The "rst 32 years of total hip arthroplasty. One
surgeon's perspective. Clin Orthop 1992;274:6}11.
 M. Teresa Raimondi, R. Pietrabissa / Biomaterials 21 (2000) 907}913
[10] Agins HJ, Alcock NW, Bansal M, et al. Metallic wear in failed
titanium-alloy total hip replacements. A histological and quanti-
tative analysis. J Bone Jt Surg [Am] 1988;70:347}56.
[11] McKellop HA, Sarmiento A, Schwinn CP, Ebramzadeh E. In vivo
wear of titanium-alloy hip prostheses. J Bone Jt Surg [Am]
1990;72:512}7.
[12] Sychterz CJ, Moon KH, Hashimoto Y, Terefenko KM, Anderson
C, Bauer TW. Wear of polyethylene cups in total hip arthroplasty.
A study of specimens retrieved post mortem. J Bone Jt Surg [Am]
1996;78:1193}200.
[13] Jasty M, Goetz DD, Bragdon CR, et al. Wear of polyethylene
acetabular components in total hip arthroplasty. An analysis of
one hundred and twenty-eight components retrieved at autopsy
or revision operations. J Bone Jt Surg [Am] 1997;79:349}58.
[14] Amstutz HC, Campbell P, McKellop H, et al. Metal on metal
total hip replacement workshop consensus document. Clin Or-
thop 1996;329(Suppl. 1):S297}303.
[15] Semlitsch M, Willert HG. Clinical wear behaviour of ultra-high
molecular weight polyethylene cups paired with metal and ce-
ramic ball heads in comparison to metal-on-metal pairings of hip
joint replacements. Proc Inst Mech Engng [H] 1997;211:73}88.
[16] Mandrino A, Moyen B, Ben Abdallah A, Treheux D, Orange D.
Aluminas with dispersoids. Tribologic properties and in vivo
aging. Biomaterials 1990;11:88}91.
[17] Huber J, Plitz W, Walter A, Re"or HJ. Comparison between
Chirulen, Hylamer and Enduron*tribological aspects. In: Trans-
actions Fifth World Biomaterials Congress, May 29}June 2 1996,
vol. 2. Toronto, Canada: University of Toronto Press, 1996.
p. 786.
[18] Muratoglu OK, O'Connor DO, Bragdon CR, Jasty M, Harris
WH. E!ect of cross-linking on the wear behaviour of ultra high
molecular weight polyethylene (UHMWPE) used in total joint
replacements. J Biomech 1998;31(Suppl. 1):114.
[19] Bragdon CR, O'Connor DO, Muratoglu OK, et al. A hip simula-
tor evaluation of di!erent forms of highly cross-linked polyethy-
lene. J Biomech 1998;31(Suppl. 1):167.
[20] Birken LMO, Berzins A, von Lacroix F, Schneider E. Composites
as a bearing partner in total knee replacement: failure analysis of
poly-II components with respect to the material design. J Bio-
mech 1998;31(Suppl. 1):48.
[21] Rodriguez D, Gil FJ, Planell JA. Wear resistance of the nitrogen
di!usion hardening of the Ti6Al4V alloy. J Biomech 1998;
31(Suppl. 1):49.
[22] Pappas MJ, Makris G, Buechel FF. Titanium nitride ceramic "lm
against polyethylene. A 48 million cycle wear test. Clin Orthop
1995;317:64}70.
[23] Ward LP, Subramanian C, Stra!ord KN, Wilks TP. Sliding wear
studies of selected nitride coatings and their potential for long-
913
term use in orthopaedic applications. Proc Inst Mech Engng [H]
1998;212:303}15.
[24] Harman MK, Banks SA, Hodge WA. Wear analysis of a retrieved
hip implant with titanium nitride coating. J Arthroplasty
1997;12:938}45.
[25] International Standard ISO 11562 Geometrical Product Speci-
"cation (GPS)*surface texture: pro"le method*metrological
characteristics of phase correct "lters. Edition 1996-12-01.
[26] Kusaba A, Kuroki Y. Femoral component wear in retrieved hip
prostheses. J Bone Jt Surg [Br] 1997;79:331}6.
[27] Kabo JM, Gebhard JS, Loren G, Amstutz HC. In vivo wear of
polyethylene acetabular components. J Bone Jt Surg [Br]
1993;75:254}8.
[28] Muratoglu O, Liu A, Jasty M, Bragdon CR, Elder JR, Harris WH.
Oxidative degradation and embrittlement of UHMWPE. Analy-
sis of 107 components. In: Transactions Fifth World Biomaterials
Congress, May 29-June 2 1996, vol. 2. Toronto, Canada: Univer-
sity of Toronto Press, 1996. p. 808.
[29] Hall RM, Unsworth A, Siney P, Wroblewski BM. Wear in re-
trieved Charnley acetabular sockets. Proc Inst Mech Engng [H]
1996;210:197}207.
[30] Izumi H, Torisu T, Itonaga I, Masumi S. Joint motion of bipolar
femoral prostheses. J Arthroplasty 1995;10:237}43.
[31] Mess D, Barmada R. Clinical and motion studies of the Bateman
bipolar prosthesis in osteonecrosis of the hip. Clin Orthop
1990;251:44}7.
[32] Wroblewski BM, McCullagh PJ, Siney PD. Quality of the surface
"nish of the head of the femoral component and the wear rate of
the socket in long-term results of the Charnley low-friction ar-
throplasty. Proc Inst Mech Engng [H] 1992;206:181}3.
[33] Brummitt K, Hardaker CS, McCullagh PJ, Drabu KJ, Smith RA.
E!ect of counterface material on the characteristics of retrieved
uncemented cobalt}chromium and titanium alloy total hip re-
placements. Proc Inst Mech Engng [H] 1996;210:191}5.
[34] Hall RM, Siney P, Unsworth A, Wroblewski BM. The e!ect of
surface topography of retrieved femoral heads on the wear of
UHMWPE sockets. Med Engng Phys 1997;19:711}9.
[35] McNie C, Barton DC, Stone MH, Fisher J. Prediction of plastic
strains in ultra-high molecular weight polyethylene due to micro-
scopic asperity interactions during sliding wear. Proc Inst Mech
Engng [H] 1998;212:49}56.
[36] Weightman B, Light D. The e!ect of the surface "nish of alumina
and stainless steel on the wear rate of UHMW polyethylene.
Biomaterials 1986;7:20}4.
[37] Lancaster JG, Dowson D, Isaac GH, Fisher J. The wear of
ultra-high molecular weight polyethylene sliding on metallic and
ceramic counterfaces representative of current femoral surfaces in
joint replacement. Proc Inst Mech Engng [H] 1997;211:17}24.