Bone remodelling after total hip arthroplasty
using an uncemented anatomic femoral stem:
a three-year prospective study using bone
JJ Panisello, L Herrero, A Herrera, V Canales, A Martinez, J Cuenca
Department of Orthopedic Surgery, Miguel Servet University Hospital, Zaragoza, Spain
evaluations were very similar among the 3 groups.
Only densitometry could detect traceable changes
resulting from the differing biomechanics of the
neutral, varus, and valgus stem alignments.
Conclusion. The ABG II stem design made moderate
errors in alignment biomechanically tolerable.
Alignment defects had no clinical consequences and
resulted in minimal differences in bone remodelling.
Key words: arthroplasty, replacement, hip; bone density;
durapatite; densitometry; hip prosthesis
Bone densitometry (dual energy X-ray absorptiometry,
DEXA) has been used to analyse the bone-remodelling
pattern associated with different femoral stem
implant designs (cemented or uncemented) for the past
15 years.1–5 Varus or valgus malalignment of the
femoral stem has been reported to result in different
load transmission patterns than in neutral alignment,6
Address correspondence and reprint requests to: Juan Jose Panisello Sebastia, 3 Maria Espinosa Street, First floor, Door O, 50015
Zaragoza, Spain. Email: firstname.lastname@example.org
Purpose. To evaluate the clinical, radiological, and
densitometric changes in the bone-remodelling
patterns of femoral stems aligned in neutral, valgus,
or varus positions.
Methods. Between February and October 2000, 70
patients underwent unilateral total hip arthroplasty
for primary osteoarthritis using an uncemented
Anatomique Benoist Girard (ABG) II stem. 69 patients
(30 males and 39 females) with a mean age of 59 years
(range, 38–76 years) and a mean body weight of
79.3 kg (range, 29–110 kg) completed 3 years’ follow-
up on bone remodelling. The clinical, radiological,
and densitometric changes of the neutral, valgus, and
varus groups were evaluated, and the difference in
bone-remodelling patterns between the 3 groups was
Results. 54 patients had neutrally placed stems,
while varus and valgus malalignment occurred in 6
and 9 patients, respectively. Clinical and radiological
Journal of Orthopaedic Surgery 2006;14(1):32-7
Vol. 14 No. 1, April 2006Bone remodelling after THA using an uncemented femoral stem 33
with poor long-term results.7 Some implant designs
have proved to be more clinically and radiologically
tolerant of these alignment errors by virtue of the
specific geometries, load transmission patterns, and
the higher elasticity of the titanium alloys.8,9 We
assessed the clinical, radiological, and densitometric
results of 70 patients who underwent total hip
arthroplasty (THA) using an uncemented anatomic
femoral stem, and analysed the differences in bone-
remodelling patterns among implants aligned in a
neutral position, at >5Ο varus, and at >5Ο valgus.
MATERIALS AND METHODS
Between February and October 2000, 70 patients
underwent unilateral THA in the Miguel Servet
University Hospital using an uncemented anatomic
femoral stem (Anatomique Benoist Girard [ABG] II;
Stryker Howmedica, Staines, England). Patients were
followed up for 3 years on bone remodelling with
the ABG II stem. Patients diagnosed with unilateral
(the healthy opposite hip was used as a control)
primary osteoarthritis (not metaphyseal deformity
or severe osteoporosis) and given informed consent
were included. The local research ethics committee
approved this prospective controlled study. The
diagnoses of the patients were hip osteoarthrosis
(n=64), avascular necrosis (n=5), and rheumatoid
arthritis (n=1). 69 patients (30 males and 39 females)
with a mean age of 59 years (range, 38–76 years) and
a mean body weight of 79.3 kg (range, 29–110 kg)
completed the 3-year follow-up. The remaining one
patient was lost to follow-up who had a neurovascular
stroke and became wheelchair-bound 19 months
All patients were implanted with the ABG II stem
and uncemented cup (the cup was not part of the
study). The ABG II is an uncemented and anatomic
stem, with press fit metaphyseal fixation, made of
titanium, molybdenum, zirconium, and iron. The stem
features a scale-shaped design to increase stability at
the metaphyseal level, and is coated with a 70-microns-
deep hydroxyapatite layer. The tail of the implant is
short and thin so as to avoid endomedullary contact
with the diaphysis, and is ultrapolished to avoid bone
fixation at this level (Fig. 1). The ABG II evolved from
the earlier ABG I design, which has a longer and more
voluminous tail to facilitate neutral positioning.
However, the ABG I surface is often susceptible to bone
growth and causes proximal bone malfunction.
Patients were operated on by 4 different surgeons
through the posterolateral approach. The acetabulum
was progressively reamed to the cup diameter
calculated preoperatively. The cup was inserted by
shaped in the metaphyseal area,
coated with a hydroxyapatite layer,
and has a thin short ultrapolished
An ABG II stem is scale-
Figure 2Bone densitometry showing the 7 Gruen zones.
34 JJ Panisello et al.Journal of Orthopaedic Surgery
press fit, usually with 3 pins. After osteotomy of the
femoral neck, a piece of cancellous bone was removed
using a box chisel to allow direct access to the femoral
canal, which was prepared by progressive broaching.
With the final broach in place, along with a modular
trial neck and head, a trial reduction was performed
to check hip stability. The trial insert was removed,
the cup interior was cleaned and dried, and the
definitive polyethylene layer and stem were placed.
The capsule was sutured, enclosing the femoral
head, and the short rotator muscles were attached to
the greater trochanter with transosseous sutures.
Fascia lata, subcutaneous tissues, and skin were
Patients were clinically evaluated according to
the Merle d’A ubigne and Postel (MA P) score10
measuring pain, joint mobility, and ability to walk.
Varus and valgus angulation (the difference between
the femoral diaphyseal axis and the stem’s main axis)
were measured and a difference of >5Ο was considered
varus or valgus. The 7 Gruen zones were inspected
for the presence of radiolucent lines, osteolysis,
cancellous condensation, cortical hypertrophy or
atrophy, reactive lines, and pedestal formation
according to the criteria by Engh et al.11 Bone stock
variation was measured in each of the 7 Gruen zones
in a 30x30–pixel square using a Hologic QDR 1000
densitometer (Hologic, Waltham [MA], US) [Fig. 2].
Patients were placed in a supine position on the
scanner table, with knees and hips in full extension.
The whole limb was in a neutral position supported
by a rigid plastic device fastened by Velcro tape to
guarantee measurement reliability.
The preoperative and postoperative first-year and
third-year MAP scores, radiographs, and bone stock
densitometric measurements among the neutral, varus,
and valgus groups were compared using parametric
or nonparametric tests, depending on the number of
patients and whether the samples were independent
or related. Chi squared test was used with Pearson
correlation for qualitative variables. A p value of
<0.05 was considered statistically significant.
Femoral stems of 54 patients (19 men and 35 women)
were neutrally aligned (Fig. 3a). Six patients (4 men
and 2 women) were varus aligned (Fig. 3b), with a
mean angle of 6Ο (standard deviation [SD], 1.94Ο)
immediately after surgery, and 6.2Ο (SD, 1.67Ο; range,
5Ο–9Ο) 3 years after surgery. Nine patients (7 men and
2 women) were valgus aligned (Fig. 3c), with a mean
angle of 6.1Ο (SD, 1.21Ο) immediately after surgery,
and 6.2Ο (SD, 1.34Ο; range, 5Ο–9Ο) 3 years after surgery.
There were no significant differences in age, weight,
atrophy in zones 1 and 7 and cortical thickening in zones 2 and 6. (b) A varus-aligned femoral stem with significantly more
cortical atrophy in zone 1 than the neural group. (c) A valgus-aligned femoral stem with significantly more cortical atrophy in
zone 7 than the neural group.
Anteroposterior radiographs taken at postoperative 3 years showing (a) a neutrally aligned femoral stem with cortical
Vol. 14 No. 1, April 2006 Bone remodelling after THA using an uncemented femoral stem35
or height among the neutral, varus, and valgus groups.
Clinical outcomes of the 3 groups were not
significantly different in pain, joint mobility, or ability
to walk (Table 1). Nonetheless, thigh pain was present
in 2 varus patients (one was serious) caused by contact
between the prosthesis tail and the antero-external
Clinical evaluation using Merle d’Aubigne and Postel score10
Clinical evaluation Neutral group, n=54
Varus group, n=6
Valgus group, n=9
Ability to walk
Bone stock evaluation using bone densitometry
Neutral group, n=54
Varus group, n=6
Valgus group, n=9
year 1 year 3year 3 year 3
* Bone mineral density is expressed in mg of hydroxyapatite/cm2; p values are compared between neutral and varus groups, and neutral and
† NAnot applicable
36JJ Panisello et al. Journal of Orthopaedic Surgery
face of the medullar channel. No major clinical
Radiological results of bone-remodelling patterns
of the 3 groups at postoperative 3 years were similar;
no patients had instability sign: the neutral group
demonstrated cortical atrophy in zones 1 and 7 due to
stress shielding, and slight cortical thickening in
zones 2 and 6 (0.2–0.6 mm). The varus group had
significantly more cortical atrophy in zone 1 than the
neutral group (p=0.026). The valgus group had
significantly more cortical atrophy in zone 7 than the
neutral group (p=0.034). The differences may be due
to different stem alignments, as greater proximal
cortical atrophy occurs in zones receiving smaller
loads. In the neutral group, cancellous condensation
in zones 2, 3, 5 and 6 was present in 72%, 45%, 36%,
and 74% of patients, respectively. Osteolytic damage
(in the form of small granulomas) in zones 1 and 7
was present in 26.2% and 9.5% of patients, respectively.
Reactive lines in the distal stem in zones 3 and 5 were
present in 5% and 17% (33% for the valgus group) of
patients, respectively. No radiolucent lines were seen
in any zone in any group. These findings of the neutral
group were similar in the other 2 groups.
Densitometric bone stock findings (Table 2)
were similar to the radiological findings. In the
neutral group, significant loss of proximal bone
stock was found in zone 1 (5.57%) and zone 7
(15.92%) in postoperative year 3; increase of bone
stock was seen in other zones. Compared with the
neutral group, the varus group had greater loss in
zones 1 and 7 and greater increase in zone 6, while
the valgus group had greater increase in zone 3 and
greater loss in zone 7.
Varus or valgus alignment of femoral stems is often
associated with poor outcomes. Malalignment may
result in an anomalous load transmission pattern
and may lead to prosthetic loosening, pain, and
periprosthetic fractures.12 However, the effect of
malalignment could be effectively offset by the
stem geometry (wedge-shaped) and the use of
elastic alloys.8,9 The results of the use of the anatomical
A BG I stem was not worse in varus or valgus
The A BG II stem evolved from the A BG I to
obtain greater proximal bone fixation and avoid stress
shielding. Changes included shifting the load transfer
zone to the proximal stem and shortening the polished
stem tip with a smaller diameter distally. Both stems
and rasps were 2 cm shorter, with slightly modified
We failed to achieve neutral alignment of the
rasps and stems in the proximal femur when the
ABG II was first introduced into our hospital. The
number of malalignment decreased progressively
as surgical techniques improved and longer rasps
introduced to eliminate the problems of preparing
an aligned bed for the stem in the metaphyseal and
Clinical and radiological outcomes of the 3 groups
were not significantly different. The bone remodelling
patterns of the A BG II was very similar to the
A BG I.13,14 In the neutral group, proximal cortical
atrophy was in zones 1 and 7 due to stress shielding,
and slight cortical thickening in zones 2 and 6
(metaphyseal-diaphyseal zones) due to load
transmission from the implant to bone. The varus
group had significantly more cortical atrophy in
zone 1 than the neutral group because of shifting of
the load transmission to the medial and distal lateral
cortical area. The valgus group had significantly
more cortical atrophy in zone 7 than the neutral group
because of shifting of the load transmission to the
lateral and distal medial cortical area. All radiological
findings suggested stable and well-integrated stems.
Densitometric findings of bone remodelling
patterns agreed with the radiological findings. Loss of
bone stock in zones 1 and 7 and increase of bone stock
in zones 2 to 6 were confirmed and consistent with
previous study.15 The 6% to 16% loss of proximal bone
stock in the present study was slightly lower than
other studies of 20% to 25% using different anatomical
implant designs.4,5,16,17 Wedge-shaped femoral stems
had excellent preservation of bone stock in medial
and distal zones, but 5% to 30% of bone stock loss in
the proximal zone remained, depending on the
implant used.3,18,19 Custom-made stems were used for
optimal metaphyseal adjustment, but were still
associated with 10% to 20% of bone stock loss in zones
1 and 7 within 3 years.20–24 Reduced-stiffness femoral
stems could reduce bone stock loss in the proximal
zone to 2% to 15% in 2-year studies,17,25 but the risks of
nonunion between layers and long-term failure
remained. Compared with the neutral group, the
varus group had greater loss of bone stock in zones 1
and 7 and greater increase of bone stock in zone 6,
while the valgus group had greater increase in zone 3
and greater loss in zone 7. This suggests that the
implant geometry (voluminous metaphyseal section,
thin and short tail) reduced negative biomechanical
effects on bone remodelling.
The validity of the present study may be limited
by the small number of the varus and valgus patients.
It is difficult to obtain sufficient malaligned cases,
Vol. 14 No. 1, April 2006Bone remodelling after THA using an uncemented femoral stem 37 Download full-text
multicentre study is therefore necessary to obtain
Bone densitometry is the most reliable tool to
evaluate bone remodelling after THA using different
femoral stem designs. Follow-up must be at least 3
1. Engh CH, McGovern TF, Bobyn JD, Harris WH. A quantitative evaluation of periprosthetic bone-remodeling after cementless
total hip arthroplasty. J Bone Joint Surg Am 1992;74:1009–20.
2. Kilgus DJ, Shimaoka EE, Tipton JS, Eberle RW. Dual-energy X-ray absorptiometry measurement of bone mineral density
around porous-coated cementless femoral implants. Methods and preliminary results. J Bone Joint Surg Br 1993;75:279–87.
3. Gibbons CE, Davies AJ, Amis AA, Olearnik H, Parker BC, Scott JE. Periprosthetic bone mineral density changes with femoral
components of differing design philosophy. Int Orthop 2001;25:89–92.
4. Nishii T, Sugano N, Masuhara K, Shibuya T, Ochi T, Tamura S. Longitudinal evaluation of time related bone remodeling
after cementless total hip arthroplasty. Clin Orthop Relat Res 1997;339:121–31.
5. Spittlehouse AJ, Smith TW, Eastell R. Bone loss around 2 different types of hip prostheses. J Arthroplasty 1998;13:422–7.
6. Mollan RA, Watters PH, Steele R, McClelland CJ. Failure of the femoral component in the Howse total hip arthroplasty. Clin
Orthop Relat Res 1984;190:142–7.
7. Vresilovic EJ, Hozack WJ, Rothman RH. Radiographic assessment of cementless femoral components. Correlation with
intraoperative mechanical stability. J Arthroplasty 1994;9:137–41.
8. Khalily C, Lester DK. Results of a tapered cementless femoral stem implanted in varus. J Arthroplasty 2002;17:463–6.
9. Schneider U, Breusch SJ, Thomsen M, Wirtz DC, Lukoschek M. Influence of implant position of a hip prosthesis on alignment
exemplified by the CLS shaft [in German]. Unfallchirurg 2002;105:31–5.
10. d’Aubigne RM, Postel M. Functional results of hip arthroplasty with acrylic prosthesis. J Bone Joint Surg Am 1954;36:451–
11. Engh CA, Massin P, Suthers KE. Roentgenographic assessment of the biologic fixation of porous-surfaced femoral components.
Clin Orthop Relat Res 1990;257:107–28.
12. Gill TJ, Sledge JB, Orler R, Ganz R. Lateral insufficiency fractures of the femur caused by osteopenia and varus angulation:
a complication of total hip arthroplasty. J Arthroplasty 1999;14:982–7.
13. Rogers A, Kulkarni R, Downes EM. The ABG hydroxyapatite-coated hip prosthesis: one hundred consecutive operations
with average 6-year follow-up. J Arthroplasty 2003;18:619–25.
14. 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.
15. Theis JC, Beadel G. Changes in proximal femoral bone mineral density around a hydroxyapatite-coated hip joint arthroplasty.
J Orthop Surg (Hong Kong) 2003;11:48–52.
16. Niinimaki T, Junila J, Jalovaara P. A proximal fixed anatomic femoral stem reduces stress shielding. Int Orthop 2001;25:85–8.
17. Karrholm J, Anderberg C, Snorrason F, Thanner J, Langeland N, Malchau H, et al. Evaluation of a femoral stem with reduced
stiffness. A randomized study with use of radiostereometry and bone densitometry. J Bone Joint Surg Am 2002;84:1651–8.
18. Korovessis P, Piperos G, Michael A, Baikousis A, Stamatakis M. Changes in bone mineral density around a stable uncemented
total hip arthroplasty. Int Orthop 1997;21:30–4.
19. Brodner W, Bitzan P, Lomoschitz F, Krepler P, Jankovsky R, Lehr S, et al. Changes in bone mineral density in the proximal
femur after cementless total hip arthroplasty. A five-year longitudinal study. J Bone Joint Surg Br 2004;86:20–6.
20. Zerahn B, Storgaard M, Johansen T, Olsen C, Lausten G, Kanstrup IL. Changes in bone mineral density adjacent to two
biomechanically different types of cementless femoral stems in total hip arthroplasty. Int Orthop 1998;22:225–9.
21. Martini F, Sell S, Kremling E, Kusswetter W. Determination of periprosthetic bone density with the DEXA method after
implantation of custom-made uncemented femoral stems. Int Orthop 1996;20:218–21.
22. Wixson RL, Stulberg SD, Van Flandern GJ, Puri L. Maintenance of proximal bone mass with an uncemented femoral stem
analysis with dual-energy X-ray absorptiometry. J Arthroplasty 1997;12:365–72.
23. Trevisan C, Bigoni M, Randelli G, Marinoni EC, Peretti G, Ortolani S. Periprosthetic bone density around fully hydroxyapatite
coated femoral stem. Clin Orthop Relat Res 1997;340:109–17.
24. Martini F, Lebherz C, Mayer F, Leichtle U, Kremling E, Sell S. Precision of the measurements of periprosthetic bone mineral
density in hips with a custom-made femoral stem. J Bone Joint Surg Br 2000;82:1065–71.
25. Glassman AH, Crowninshield RD, Schenck R, Herberts P. A low stiffness composite biologically fixed prosthesis. Clin
Orthop Relat Res 2001;393:128–36.
years as biomechanical adaptation to the prosthesis
occurs mostly in this period. Longer-term follow-up
may detect bone loss in proximal zones caused by
osteolysis or late bone remodelling due to involutional
bone loss. The ABG II stem design made moderate
errors in alignment biomechanically tolerable.
A lignment defects had no clinical consequences
and resulted in minimal differences in bone