J Orthop Sci (2008) 13:198–201

DOI 10.1007/s00776-008-1221-y

Original article

Measurement of extent of bone ongrowth and hydroxyapatite

absorption in retrieved acetabular cups
YOUNG-YOOL CHUNG1, JIN KIM2, CHAE HYUN LIM1, KI-SOO KIM1, and YEON SOO LEE3
1
Department of Orthopaedic Surgery, Kwangju Christian Hospital, Gwangju, Republic of Korea
2
Department of Parasitology, College of Medicine, Seo Nam University, Gwangju, Republic of Korea
3
Schools of Information and Mechatronics, Gwangju Institute of Science and Technology, 1 Oryong-dong, Buk-gu, Gwangju 500-712,
Republic of Korea

Abstract
Background. To understand why hydroxyapatite (HA)-coated
acetabular cups did not produce satisfactory results, we mea-
sured the extent of bone ongrowth and HA absorption of the
retrieved cups.
Methods. Between March 1992 and June 1998, a total of 289
patients underwent total hip arthroplasty with an HA-coated
ABG (Anatomical Benoist Giraud) prosthesis in our hospital.
We revised 71 cups in our hospital, 43 of which were included
in this study. The cups were in situ for an average of 80.4
months. We revised 21 cups for polyethylene wear and oste-
olysis, 20 for loosening, and 2 for recurrent dislocation. The
outer surfaces of the cups were subdivided by two circular
grooves into three areas: central, middle, peripheral. Bone
ongrowth and HA absorption was measured in terms of area
and thickness. The areas were graphically calculated using a
computer imaging analysis system. The thickness of the HA
coating remaining on the retrieved cup was measured by an
x-ray coating thickness measurement instrument.
Results. Of the 43 cups, 37 had bone ongrowth. Bone ongrowth
was observed on the central surface in 23 cups, on the middle
surface in 27 cups, and on the peripheral surface in 15 cups;
6 of the 37 cups had bone ongrowth only in the groove.
Altogether, 37 cups showed bone ongrowth on 6.43% of the
outer acetabular surface. HA absorption appeared in all 43
cups, and an average 63.21% of the HA coating area was
absorbed. there was a proportional relation between the rate
of HA absorption and implantation duration, but there was
no consistent trend between the thickness of the HA coating
and implantation duration.
Conclusions. We suggest that progressive HA absorption and
minimal bone ongrowth are related to the failure of HA-
coated acetabular cups at long-term follow-up after total hip
arthroplasty.
Offprint requests to: Y. Soo Lee
Received: August 21, 2007 / Accepted: February 7, 2008
Introduction
It is known that hydroxyapatite (HA), which was intro-
duced for biological fixation during the 1980s, achieved
very strong bone bonding for a short period after total
hip arthroplasty (THA) as a nontoxic, biocompatible,
and osteoconductive material in many experiments.1–3
There is a clinical report that a positive result was
achieved after THA for which an HA-coated acetabular
cup was used.4 However, it has been reported that the
clinical results of the THA for which the HA-coated
acetabular cup was used varied according to features of
HA, including thickness of the coating and the outer
surface structure of the cups.5–7
We conducted total hip arthroplasty using HA-coated
ABG (Anatomical Benoist Giraud; Howmedica, Stains,
UK) and found that there was a high degree of failure
in the HA-coated acetabular cups within 5 years.8 To
identify the reason why HA-coated acetabular cups
failed, this study measured the extent of bone ongrowth,
HA absorption, and HA thickness in the retrieved
cup.
Materials and methods
From March 1992 to June 1998, a total of 289 patients
underwent total hip arthroplasty with the HA-coated
ABG prosthesis in our hospital. Among them, 71 cups
were revised in our hospital, 43 of which were included
in this study.
The HA-coated ABG cup was hemispherical and
coated with HA of 60 ± 20 μm thickness on 2.5 μm grit
blasted surface. The chemical purity of the HA was
above 99.99%, and its maximum porosity was 2%. The
indications for THA in the 43 patients (31 men, 12
women) were as follows: 35 for avascular necrosis of
femoral head, 7 for osteoarthritis, and 1 for tuberculous
arthritis. One surgeon performed all the operations
Y-Y. Chung et al.: Bone and HA absorption in acetabular cups
using the posterolateral approach. Acetabular cups of
the same size as the last reamer were inserted. Screws
were used for fixation in four cups. The polyethylene
was sterilized with gamma radiation in the air. The
femoral head was composed of a cobalt-chrome alloy
and was 28 mm in diameter. The patients walked with
two crutches after removal of suction drainage. Antero-
posterior and lateral radiographs of the hip joint were
made at each visit.
The cups were in situ for an average of 80.4 months.
Altogether, 21 cups were revised for polyethylene
wear and osteolysis, 20 for loosening, and 2 for recur-
rent dislocation. Loosening was defined as when the
acetabular cup was displaced more than 2 mm and
had an angular change of more than 5° on the follow-
up radiograph. When movement of the cups was
found during the revision, even though no loosening
was found on radiography, we declared that they
were loose. The 20 loose cups were found either
through radiography (5 cases) or revision surgery (15
cases).
The retrieved cups were divided by two circular
grooves into three surfaces: central, middle, and periph-
eral. We visually observed bone ongrowth and HA
absorption on the grooves and surfaces of the cups (Fig.
1A). Also, we obtained the extent of bone ongrowth
and HA absorption with an imaging analysis system
(analySIS-Pro 3.0; Soft Imaging System, Munster,
A
B retrieved acetabular cup showed bone ongrowth and HA
C (analysis-pro 3.0; Soft Imaging System, Munster, Germany)
199
Germany). The retrieved cups were divided into six
parts based on six screw holes on the middle surface and
were then photographed with a digital camera. We drew
a line connecting the center of the cup to the centers of
the screw holes on the middle surface in each digital
image to obtain a triangular image (Fig. 1B). Six figures
were needed per cup. We input each digital image into
the image analysis device to obtain the extent of bone
ongrowth and HA absorption (Fig. 1C). In some cases,
it was difficult to differentiate between bone ongrowth
and HA on the digital images, so we visually measured
the location and size of bone ongrowth in the cups and
compared them with the digital image to minimize
errors. For this study, the total area of the cups included
the area except the central hole and the holes of the 12
screws.
We used an x-ray coating thickness measurement
instrument (Fischerscope mms, Fischer Technology,
Windsor, CT, USA) to measure the thickness of the HA
on the retrieved cups. To minimize errors in measure-
ment, we decided a zero point by using samples with the
same components as the acetabular cup. The acetabular
cup was turned with an interval of 30° on the three sur-
faces, respectively, for measurement. The probe of the
instrument was made to meet the surface to measure at
a right angle, and measurement was repeated five times
per location. The average value obtained was decided
to be the thickness.
Fig. 1. A 56-year-old man underwent total hip arthroplasty
with the hydroxyapatite (HA)-coated ABG acetabular cup
for avascular necrosis of the femoral head 4 years ago. A The
acetabular cup was revised due to recurrent dislocation. The
absorption. B The retrieved cup was divided into six trian-
gular digital images to measure the extent of bone ongrowth
and HA absorption. C Extent of bone ongrowth and HA
absorption was calculated using an imaging analysis system
200
Results
We found that 37 (86%) of the 43 cups had bone
ongrowth, and 6 cups had no bone ongrowth. The six
cases that showed no bone ongrowth were revised
because of loosening. No fibrous bonding was found on
the surface of the retrieved cups. Bone ongrowth was
found on the central surface of 23 cups (53.4%), on the
middle surface of 27 cups (62.79%), on the peripheral
surface of 15 cups, and on the grooves of 36 of the 43
cups (83.7%). We observed bone ongrowth only on the
grooves of 6 of the 37 cups with bone ongrowth. The
average bone ongrowth of the 37 cases was 6.43%.
We found HA absorption in every case and more
than 50% absorption in 28 cases. The average absorp-
tion area was 63.21%, and the absorption rate increased
as implantation duration increased. The average HA
thickness on the cups was 55.8 ± 16.0 μm on the central
surface, 53.0 ± 15.7 μm on the middle surface, and 48.6
± 17.0 μm on the peripheral surface.
Discussion
It is known that nontoxic, biocompatible, osteoconduct-
ible HA coating creates strong bonding between the cup
and bone as bone ongrowth occurs without an interme-
diate step of fibrous tissue for a short period after
implantation.9,10 Geesink et al.2 reported biomechanical
and animal experiments using HA-coated cups, and in
1988 Furlong and Osborn1 carried out a THA using
HA-coated acetabular cups and presented the results.
Tonino et al.11 carried out a THA using HA-coated cups
and reported that bone ongrowth was found in all hips
during the next 2 years.
For THAs with HA-coated prostheses, the femoral
stems have produced satisfactory results. We conducted
femoral stem revision in only 1 of the 142 cases during
a 5-year follow-up after THA with HA-coated prosthe-
ses.8 However, the results of the HA-coated acetabular
cups varied greatly according to the kind of cup. Tonino
and Rhamy7 and Oosterbos et al.6 reported that they
found 100% bone ongrowth around the cups at 5- and
7-year follow-ups after THAs with HA-coated acetabu-
lar cups, with no revision necessary due to loosening.
Giannikas12 reported that the THA showed good results
clinically and radiographically at a 4.8 year follow-up.
However, Manley et al.5 reported that the rate of
mechanical failure of the HA-coated press-fit cup was
22%, with an average 7.9-year follow-up, and the rate
of revision was 11%. Reikeras and Gunderson13 also
reported that 39 of 191 cases needed revision because
of loosening after an average 7- to 10-year follow-up.
Rokkum et al.14 reported that 5 of 100 cases using HA-
coated threaded cups needed revision. We carried out
Y-Y. Chung et al.: Bone and HA absorption in acetabular cups
revisions in 4 of 60 cases during the 3-year follow-up
after THA using HA-coated press-fit acetabular cups
and in 45 of 142 cases during a 5-year follow-up.8
There are a variety of reasons why HA-coated ace-
tabular cups failed. Capello et al.15 described the reason
for loosening as follows: As HA-coated acetabular cups
are chemically bonded with bone without physical
bonding, it cannot endure stress occurring between the
cups and bone. Reikeras and Gunderson16 explained it
as follows: The high success rate of the HA-coated
threaded cup was due to mechanical stability and bio-
logical fixation, which contribute to overcoming of the
stress that occurs between the cup and bone. As the
acetabular cups used in this study were HA-coated on
the microstructure surface and are chemically connected
to the bone without a mechanical connection, it is sug-
gested that loosening may result from the stress occur-
ring in everyday life at a long-term follow-up.
Overgaard’s group17 reported that bone ongrowth
was found in 30% of the HA-coated area in an animal
study. In our study, we found the extent of bone
ongrowth was only an average of 6.43% of the whole
area of the retrieved cups. The extent of bone ongrowth
of the cups with osteolysis and/or loosening was mea-
sured to be lower than that of other cups. Of course, as
41 of the 43 cases were accompanied by osteolysis, we
could not identify the actual extent of bone ongrowth.
We did not know whether HA-coated acetabular cups
had low bone ongrowth or the bone ongrowth that was
measured as low was osteolysis. However, the extent of
bone ongrowth of the HA-coated acetabular cup was
much lower than that of porous coated cups.18,19
The amount of bone formed in the HA-coated ace-
tabular cups was scant compared to that of porous
coated cups. Strong bonding to the bone was not found
in the HA-coated cup, unlike porous coated cups. HA-
coated acetabular cups could be easily removed without
destruction of acetabulum during revision arthroplasty.
It is suggested that loosening of HA-coated cups, which
have an bone ongrowth pattern different from that of
porous coated cups, occurs at the long-term follow-up
because the bonding was easily destroyed by such
actions as osteolysis and stress. We think that if osteoly-
sis occurs around the acetabular cup loosening develops
more easily in the HA-coated cup than the porous-
coated one because of the specific patterns of bone
ongrowth of the HA-coated cup described above. of the
43 cases in this study, 20 were revised because of loosen-
ing. It is thought that the high revision rate of the cups
due to loosening may derive from a specific character
of bone formation of HA-coated acetabular cups.
This study found that the extent of HA absorption
was 63.21% for an average of 80.4 months’ follow-up.
The extent of HA absorption was in proportion to the
implantation duration. About 9.5% absorption was
Y-Y. Chung et al.: Bone and HA absorption in acetabular cups
found per year. If it is implanted for more than 10 years,
it is expected that absorption may occur in more than
95% of the coated area. Continuous absorption of HA
is believed to be another reason for failure. The change
of thickness on the whole surface was not in proportion
to the implantation duration. As the thickness differs
according to the area to be measured in the same cup,
the change of the thickness could not be statistically
connected with duration of implantation. According to
the insertion periods, absorption of HA occurs, but it
seems to be more affected by the surrounding environ-
ment, such as the presence of osteolysis.
In 15 of the 20 cases with loosening in this study, we
found that the cups had moved while revising them even
though loosening was not seen on radiography. This
study found that 15 cups with movement had bone for-
mation seen on radiography after hip arthroplasty. We
suggest that this kind of loosening occurred because HA
was continuously absorbed after implantation. Also it is
thought that bone ongrowth occurred early after implan-
tation, and the cups were bonded strongly; but as the
implantation duration increased, more HA absorption
occurred and the bonding between the cups and bone
weakened. We believe that HA absorption is the most
important reason for failure of the cups with a smooth
surface and suggest that new absorption-resistant HA
with an advantage of early bone formation should be
developed so the HA-coated acetabular cups can survive
longer.
Conclusions
This study suggests that the high failure rate of the HA-
coated acetabular cup after total hip arthroplasty is due
to the scarce amount of bone ongrowth and continuous
HA absorption. Therefore, for longer survival of the
HA-coated acetabular cups, the surface of the cups
should be improved to promote the rate of bone
ongrowth and stress endurance. HA that can last without
absorption after implantation is necessary.
None of the authors of this manuscript has received or will
receive any type of support from a commercial party related
to the subject of this article. And, this research was partially
supported by a research grant from the School of Medical
System Engineering (SMSE) in GIST, Korea.
References
1. Furlong RJ, Osborn JF. Fixation of hip prostheses by hydroxy-
apatite ceramic coatings. J Bone Joint Surg Br 1991;73:741–5.
201
2. Geesink RG, de Groot K, Klein CP. Bonding of bone to apatite-
coated implants. J Bone Joint Surg Br 1988;70:17–22.
3. Soballe K, Hansen ES, Brockstedt-Rasmussen H, Bünger C.
Hydroxyapatite coating coverts fibrous tissue to bone around
loaded implants. J Bone Joint Surg Br 1993;75:270–8.
4. Rossi P, Sibelli P, Fumero S, Crua E. Short-term results of
hydroxyapatite-coated primary total hip arthroplasty. Clin Orthop
1995;310:98–102.
5. Manley MT, Capello WN, D’Antonio JA, Edidin AA, Geesink
RG. Fixation of acetabular cups without cement in total hip
arthroplasty: a comparison of three different implant surfaces at
a minimum duration of follow-up of five years. J Bone Joint Surg
Am 1998;80:1175–85.
6. Oosterbos CJ, Rahmy AI, Tonino AJ. Hydroxyapatite coated hip
prosthesis followed up for five years. Int Orthop 2001;25:17–21.
7. Tonino AJ, Rahmy AI. The hydroxyapatite-ABG hip system: 5-
to 7-year results from an international multicentre study — the
International ABG Study Group. J Arthroplasty 2000;15:274–
82.
8. Kim KS, Ko SH, Chung YY, Song JY, Park BH, Choi JB, et al.
Total hip arthroplasty with hydroxyapatite-coated ABG hip pros-
thesis. J Korean Orthop Assoc 2003;38:568–72 (in Korean).
9. Cook SD, Thomas KA, Kay JF, Jarcho M. Hydroxyapatite-coated
titanium for orthopedic implant applications. Clin Orthop 1988;
232:225–43.
10. Tisdel CL, Goldberg VM, Parr JA, Bensusan JS, Staikoff LS,
Stevenson S. The influence of a hydroxyapatite and tricalcium-
phosphate coating on bone growth into titanium fiber metal
implants. J Bone Joint Surg Am 1994;76:159–71.
11. Tonino AJ, Romanini L, Rossi P, Borroni M, Greco F, Garcia-
Araujo C, et al. Hydroxyapatite-coated hip prosthesis: early
results from an international study. Clin Orthop 1995;312:211–
25.
12. Giannikas KA, Din R, Sadiq S, Dunningham TH. Medium-term
results of the ABG total hip arthroplasty in young patients.
J Arthroplasty 2002;17:184–8.
13. Reikeras O, Gunderson RB. Failure of HA coating on a grit-
blasted acetabular cup: 155 patients followed for 7–10 years. Acta
Orthop Scand 2002;73:104–8.
14. Rokkum M, Brandt M, Bye K, Hetland KR, Waage S, Reigstad
A. Polyethylene wear, osteolysis and acetabular loosening with
an HA-coated hip prosthesis: a follow-up of 94 consecutive
arthroplasties. J Bone Joint Surg Br 1999;81:582–9.
15. Capello WN, D’Antonio JA, Manley MT, Feinberg JR. Hydroxy-
apatite in total hip arthroplasty: clinical results and critical issues.
Clin Orthop 1998;355:200–11.
16. Reikeras O, Gunderson RB. Long-term results of HA coated
threaded versus HA coated hemispheric press fit cups: 287 hips
followed for 11 to 16 years. Arch Orthop Trauma Surg 2006;
126:503–8.
17. Lind M, Overgaard S, Nquyen T, Ongpipattanakul B, Bünger C,
Søballe K. Transforming growth factor-beta stimulates bone
ongrowth: hydroxyapatite-coated implants studied in dogs. Acta
Orthop Scand 1996;67:611–6.
18. Engh CA, Hooten JP Jr, Zettl-schaffer KF, Ghaffarpour M,
McGovern TF, Bobyn JD. Evaluation of bone ingrowth on proxi-
mally and extensively porous-coated anatomic medullary locking
prostheses retrieved at autopsy. J Bone Joint Surg Am 1995;77:
903–10.
19. Kim IK, Park JH. Bone ingrowth rate on retrieved human porous
coated acetabular cup. J Korean Orthop Assoc 1996;31:975–81
(in Korean).