The Role of Vibration and Drainage in Femoral
Impaction Bone Grafting
Benjamin J.R.F. Bolland, MRCS,* Andrew M.R. New, PhD,y Gopal Madabhushi, PhD,z
Richard O.C. Oreffo, DPhil,* and Douglas G. Dunlop, MD, FRCS Tr&Orth*
Abstract: Vibration is commonly used in civil engineering applications to efficiently
compact aggregates. This study examined the effect of vibration and drainage on
bone graft compaction and cement penetration in an in vitro femoral impaction bone
grafting model with the use of 3-dimensional micro-computed tomographic imaging.
Three regions were analyzed. In the middle and proximal femoral regions, there was
a significant increase in the proportion of bone grafts with a reciprocal reduction in
water and air in the vibration-assisted group (P b .01) as compared with the control
group, suggesting tighter graft compaction. Cement volume was also significantly
reduced in the middle region in the vibration-assisted group. No difference was
observed in the distal region. This study demonstrates the value of vibration and
drainage in bone graft compaction, with implications therein for clinical application
and outcome. Key words: impaction bone grafting, vibration, morselized allograft,
compaction, micro-computed tomographic imaging.
© 2008 Elsevier Inc. All rights reserved.
Impaction bone grafting (IBG) is a recognized
technique to reconstitute the extensive areas of
bone loss in the femur and acetabulum [1,2] often
encountered in revision hip surgery. Morselized
allograft continues to be the gold standard, provid-
ing good mechanical support and osteoconductive
potential. The success of the procedure is reliant on
sufficient compaction of the allograft to allow it to
support the prosthesis under physiologic loading
and to prevent excess subsidence and failure.
Impaction bone grafting is a technically demand-
ing procedure with a steep learning curve. Deter-
mining when the graft is adequately compacted is
potentially one of the most difficult factors for
surgeons to judge intraoperatively. The concern of
aggressive compaction leading to femoral fracture
(with rates as high as 27% reported ) must be
balanced against the risks of undercompaction of the
graft and subsidence of the prosthesis postopera-
tively. This is compounded by the fact that there is
no specific intraoperative indicator of graft compac-
tion completion. The difficulty in obtaining this fine
balance is reflected in the literature, in which a wide
range of outcomes are reported. Results from the
center from which the femoral IBG technique
originated have shown excellent mid-term out-
come, with 99% survival at an average 10-year
follow-up (in 226 patients who underwent a
femoral reoperation caused by symptomatic aseptic
loosening as the end point) . However, this has
not always been the experience of other centers. In
From the *Bone and Joint Research Group, Developmental Origins of
Health and Disease, University of Southampton, Southampton, United
Kingdom; yBioengineering Research Group, School of Engineering
Sciences, University of Southampton, Southampton, United Kingdom;
and zDepartment of Engineering, University of Cambridge, Cambridge,
Submitted May 27, 2007; accepted October 14, 2007.
Benefits or funds were received in partial support of the
research material described in this article. These benefits or funds
were received from Stryker UK Ltd, Newbury, UK (funded part
of salary toward a master's degree) as well as the Biotechnology
and Biological Sciences Research Council and Engineering and
Physical Sciences Research Council.
Reprint requests: Benjamin J.R.F. Bolland, MRCS, Bone and
Joint Research Group, Developmental Origins of Health and
Disease, University ofSouthampton, SO166YD Southampton, UK.
© 2008 Elsevier Inc. All rights reserved.
The Journal of Arthroplasty Vol. 23 No. 8 2008
Bristol, of 79 patients who underwent hip arthro-
plasties on whom femoral IBG had been performed
and followed up for just over 1 year, 9 (11%)
showed evidence of significant subsidence . This
was defined as subsidence greater than 10 mm and
in all cases occurred in the first 3 months after the
operation. Six hips required subsequent re-revision.
A series in Australia found a similar subsidence
value of 9 mm (range, 2-37 mm) at 24 months .
Shear strength is commonly measured when
evaluating the mechanical properties of bone grafts
because the mode of bone graft failure after IBG is
believed to be in shear. Techniques to improve the
shear strength of allografts have concentrated on
altering the composition and particle size of the
impacted material. Important factors include wash-
ing [7,8], grading , the addition of synthetic
materials [9-11] (eg, bone graft extenders), and the
degree of cement penetration . Since the
development of the technique by Slooff et al in
Nijmegen in the late 1970s for the acetabulum 
and subsequently by Ling et al in Exeter in the UK in
1987 for the femur , there have been few
modifications to either the technique itself or the
instrumentation used for impaction.
The morselized allograft shares many character-
istics with aggregate materials used in civil engineer-
ing applications . The behavior of these
aggregates under load has been studied extensively,
and this knowledge can be applied to improve the
mechanical characteristics of other aggregates, such
as bone graft. Vibration is commonly applied to
aggregates used in civil engineering applications to
improve the compaction (assembly) of the aggregate
particles and hence to increase the aggregate's
compressive and shear strengths . A recent
study investigated the effect of vibration and
drainage during the impaction process on hoop
strains (as a marker of fracture risk) and subsidence
(as a marker of prosthesis stability). It demonstrated
that vibration-assisted IBG leads to reduced peak
loads and hoop strains in the femur during graft
compaction and that the resulting graft is better able
to resist subsidence of the prosthesis .
What remains unclear is the effect this technique
has on graft compaction. Prosthetic stability can be
affected not only by graft compaction but also by the
degree of cement penetration. Greater cement
penetration into the impacted graft through to the
endosteal surface has been shown in vitro  and
in vivo  to be mechanically advantageous. The
mechanical advantage that cement penetration
confers, however, is offset against a deleterious
biological effect on allograft and cortical remodeling,
as demonstrated by Frei et al  in an in vivo bone
chamber model. Improving graft compaction is
mechanically advantageous and, with a thicker
bone graft mantle, creates a greater volume for
potential bone remodeling.
This study examined the effect of vibration and
drainage on bone graft compaction and cement
penetration in an in vitro femoral IBG model.
Materials and Methods
Bone Graft Preparation
Femoral heads were retrieved with the consent of
patients undergoing elective or traumatic hip sur-
gery at Southampton General Hospital and stored in
a −80 freezer for more than 6 months. Only tissue
that would otherwise have been discarded was used,
with the approval of the local ethics committee
(LREC 0091). Femoral heads were defrosted by
soaking in warm normal saline. All soft tissue,
osteophytes, and cartilage were removed using bone
nibblers and an oscillating saw (Stryker, Howme-
dica, UK). The heads were cut into halves and milled
using a 3-mm Aesculap bone mill. The resulting
morselized graft was washed with normal saline to
remove excess fat. Grafts from all femoral heads
were mixed together to reduce the effects of patient
variability. The same stock of morselized graft was
used for both study groups.
Standard “X-change” femoral IBG instrumenta-
tion (Stryker) consisting of distal and proximal
tamps or phantoms attached to a slap hammer
device was used in the control group. In the
vibration-assisted group, a vibration hammer
(Woodpecker vibration device, Minnesota Bram-
stedt Surgical Inc, St Paul, MN, USA; Fig. 1A)
designed to aid femoral broaching was adapted to
allow connection of impaction tamps/phantoms.
Multiple holes were drilled through the flanks of the
tamps into the central guidewire hole, providing
drainage portals for fat, marrow, and fluid (Fig. 1C
and D—standard and perforated phantoms).
Medium left third-generation composite femurs
manufactured from short glass fiber–reinforced
epoxy resin were used as the basis of the biomecha-
nical model (Model No. 3303, Sawbones Europe,
Malmo, Sweden). These models have been shown
to approximate the mechanical properties of the
human femur but with much less variability than
that found in cadaveric material . Twelve
models of the femur were prepared (6 in each
group): standard femoral graft compaction using
“X-change” proximal and distal impactors and a slap
1158 The Journal of Arthroplasty Vol. 23 No. 8 December 2008
hammer (Fig. 1B) for the control group) and
tors and a vibration hammer (Fig. 1A, C, and D) for
the experimental group.
The models were widened to a canal diameter of
22 mm to closely resemble the appearance and
composition of femurs encountered during revision
hip arthroplasty surgery with loss of all cancellous
bone and thinning of the cortex. The distal canal of
each femur was occluded 25 mm beneath the
anticipated position of the tip of the prosthesis
using bone cement.
Impaction bone grafting was carried out using a
standard protocol for the control and vibration-
assisted groups with the “X-change” instrumenta-
tion (Stryker UK Ltd, Newbury, UK). Compaction
of the graft was performed sequentially with a
measured volume of graft introduced into the
canal. Further portions of the graft were added to
the femur, with the graft being compacted before
the next portion of graft was added at each stage.
Three sets of compaction using the distal impactors
were performed before exchanging to the proximal
impactors. In the control group, a standard techni-
que of applying 20 blows per portion of graft was
maintained. In the experimental group, the graft
was compacted by application of the vibration
hammer to the tamp/phantom for approximately
10 seconds. The end point of impaction in the
control group was defined by there being no
further movement of the tamp after 10 consecutive
blows with the slap hammer, whereas that in the
vibration-assisted group was defined by there being
no further movement of the tamp despite force
application to the vibration hammer. Preliminary
experiments confirmed the end point in the
vibration-assisted group to correlate with the end
point in the control group (ie, after vibration-
assisted compaction, the application of 10 blows
using the slap hammer resulted in no further
movement of the tamp). A single mix of bone
cement (Smartset CMW, DePuy CMW Ltd, Black-
pool, UK), prepared in a vacuum mixing system
(Cemvac, DePuy CMW Ltd), was inserted retro-
grade using a revision nozzle and a cement gun
and pressurized using a proximal cement seal.
These were followed by the insertion of a 44
No. 2 Exeter (Stryker UK Ltd, Newbury, UK)
femoral prosthesis (Fig. 2A).
Micro-Computed Tomographic Imaging
Micro-computed tomographic scans of the
impacted bone grafted femurs were obtained using
a bench-top micro-tomography system (X-TEK
Systems Ltd, Tring, Hertfordshire, UK) with a
photomultiplier detector. X-rays were generated
using an electron gun accelerating voltage of
145 kV, a beam current of 45 μA, and a tungsten
target. Owing to the model size, the femur was
scanned in the proximal, middle, and distal sections.
For the proximal scan, the femur was inverted; to
permit this, it was mounted on a polyethylene tube
with base and lid components that connected to the
adjustable sample platform of the scanner. After
being mounted on the sample platform, the femur
was centered in the x-ray beam, the electron beam
was focused, and the detector was calibrated under
no–x-ray-beam and uninterrupted–x-ray-beam
conditions. The samples were then scanned at
1600 angular positions integrating 16 frames at
each. After the scanning process, the raw data were
collected and reconstructed using Next Generation
Fig. 1. Femoral impaction instrumentation: (A) vibration hammer; (B) slap hammer from Stryker “X-change” femoral
impaction instrumentation; (C) standard and perforated proximal impactors (phantoms); and (D) circumferential drill
holes directed to the central canal of the phantom.
Vibration Femoral IBG ?
Bolland et al
Imaging version 1.4.59 software (X-TEK Systems
Ltd) with an average voxel size of 120 μm.
Micro-Computed Tomographic Analysis
The reconstructed images were visualized using
Volume Graphics Studio Max 1.2.1 software
(Volume Graphics, Heidelberg, Germany), and
3-dimensional views (Fig. 2B) were created along
with axial, sagittal, and coronal slices. Segmentation
tools allowed the extraction of cement, bone graft,
and femoral sawbone components individually (Fig.
3A-D). Along with the 3-dimensional view of the
individual component, a histogram plotting the
number of voxels against gray-scale values was
created. The sawbone, air, water, bone graft, and
cement were all scanned separately using identical
image settings to determine the gray-scale range
for the individual components. Regions from the
distal, middle, and proximal thirds of the femur
were analyzed. Subvolumes, 2 cm in height and 2
cm apart, commencing from the tip of the
prosthesis were selected and represented the distal,
middle, and proximal regions of interest (ROIs).
Referencing from the stem tip ensured that
reproducible volumes were selected and excluded
any variation introduced from the depth of stem
insertion (Fig. 4A-D).
Fig. 3. Three-dimensional reconstructions of impacted
bone graft models demonstrating visualization of (A)
femur and bone graft, (B) bone graft and cement mantle,
(C) bone graft only, and (D) cement mantle only.
Fig. 2. Femoral model for IBG. (A) Composite femur with Exeter stem cemented in situ. Proximal strain gauge is attached
to the medial calcar. (B) Three-dimensional computed tomographic reconstruction of the proximal femur.
Fig. 4. Regions of analysis. (A) Scout view of the femoral
model demonstrating the selected regions analyzed. (B–D)
Representative cross-sectional 3-dimensional reconstruc-
tions at the (B) proximal, (C) middle, and (D) distal ROIs.
1160 The Journal of Arthroplasty Vol. 23 No. 8 December 2008
Based on the determined gray-scale ranges for
separate components, the total number of voxels
representing the cement and bone graft for the distal
middle and proximal regions for the control and
experiment groups was quantified. The cement
volume was expressed both as a percentage of the
total ROI and in millimeters cubed (the resolution of
which the volume in millimeters cubed could be
calculated). The segmented bone graft volume also
represented proportions of air and water; therefore,
all 3 components were quantified separately and
measured as a percentage of the total segmented
volume. This also had the additional advantage of
eradicating any difference in the segmented volume
size under analysis.
Statistical analysis of the data was performed
using Student's t test and 2-sample unequal variance
(GraphPad Instat Software, GraphPad Software, Inc,
San Diego, CA, USA).
Computed Tomographic Calibration of Air,
Water, Bone Graft, and Cement Properties
The gray-scale ranges, derived from scanning
components separately, were 0 to 25 for air, 25 to
the bone graft volume in the absence of water (Fig. 5).
Regional Analysis of Cement Volume and Bone
Distal ROI. Cement volume: The volume of bone
cement represented in the distal ROI was 1.21 cm3
Fig. 5. Computed tomographic calibration histogram.
Gray-scale ranges represented by the individually ana-
lyzed components of the femoral IBG model (ie, air, water,
bone graft, and cement).
Fig. 6. Graphical representation of the mean proportions of bone graft, water, air (mean % ± SD), and cement volume
(mean cm3± SD) in the distal, middle, and proximal ROIs (n = 5).*P b .05;**P b .01.
Vibration Femoral IBG ?
Bolland et al
(SD, 0.26) or 32.1% (SD, 6.8) in the control group
and was 1.29 cm3(SD, 0.38) or 34.0% (SD, 9.1) in
the vibration-assisted group (Fig. 6). This difference
was not statistically different (P = .72).
Bone graft proportions: The distal ROI was com-
posed of a mean 96.2% (SD, 1.5%) bone graft, that
of 3.1% (SD, 1.2%) water, and that of 0.4% (SD,
0.3%) air in the control group as compared with its
composition being 94.9% (SD, 3.3%) bone graft,
3.9% (SD, 2.2%) water, and 1.0% (SD, 1.2%) air in
the vibration-assisted group (Fig. 6). For all compo-
nents, the difference between the 2 groups was not
statistically significant (bone graft, P = .43; water,
P = .49; air, P = .35).
Middle ROI. Cement volume: The volume of bone
cement represented in the middle ROI was 2.21 cm3
(SD, 0.25) or 62.2% (SD, 7.4) in the control group
and was 1.64 cm3(SD, 0.15) or 45.7% (SD, 4.4) in
the vibration-assisted group (Fig. 6). This difference
was statistically different (P b .01).
Bone graft proportions: The middle ROI was
composed of a mean 81.9% (SD, 6.0%) bone
graft, that of 12.5% (SD, 3.1%) water, and that of
5.5% (SD, 3.1%) air in the control group as
compared with its composition being 91.8% (SD,
2.3%) bone graft, 6.7% (SD, 1.5%) water, and 1.2%
(SD, 1.0%) air in the vibration-assisted group
(Fig. 6). The increase in the proportion of bone
graft along with a drop in the proportions of water
and air in the vibration-assisted group as compared
with the control group was significant in all cases
(bone graft, P b .05; water, P b .01; air, P b .05).
Proximal ROI. Cement volume: The volume of
bone cement represented in the proximal ROI was
group and was 2.53 cm3(SD, 0.25) or 43.2% (SD,
3.6) in the vibration-assisted group (Fig. 6). This
difference was not statistically different (P = .29).
Bone graft proportions: The middle ROI was
composed of a mean 63.8% (SD, 3.1%) bone
graft, that of 22.5% (SD, 1.8%) water, and that of
13.5% (SD, 1.9%) air in the control group as
compared with its composition being 68.8% (SD,
3.1%) bone graft, 19.4% (SD, 2.1%) water, and
10.8% (SD, 1.9%) air in the vibration-assisted group
(Fig. 6). The increase in the proportion of bone graft
along with a drop in the proportions of water and air
in the vibration-assisted group as compared with the
control group was significant in all cases (bone graft,
P b .05; water, P b .05; air, P b .05).
We have shown that in a femoral IBG model, the
application of vibration combined with perforated
impactors to allow drainage leads to improved bone
graft compaction as compared with the standard
technique and instrumentation used in present-day
This study has demonstrated that the compac-
tion of aggregates, such as bone graft, can be
improved with better alignment of the particles to
allow denser packing and greater interparticulate
surface area contact. In doing so, the volume in
between the particles, consisting of air and fluid, is
reduced. In the field of civil engineering, this has
been shown to occur with the application of
vibration [13,18]. Vibration gives the particulate
material a better chance to come into denser
packing provided that the void space reduction is
allowed by adequate drainage of the fluids in the
void space. In the presence of excess fluid in a
contained space, compaction effort is transmitted
not only to the bone graft but also to the fluid,
resulting in poorer impaction of the graft particles
. Drainage is used widely for in situ densifica-
tion of loosely deposited sand to alleviate the risk
of soil liquefaction under earthquake loading.
Improved closer packing increases the interparti-
culate contacts and the shear strength of the
particulate material [19,20].
In this study, the proportion of bone graft in the
vibration-assisted group was increased in both the
proximal and middle regions of the femur, with a
concomitant decrease in the proportions of air and
water. These findings are consistent with improved
denser packing and compaction of the graft. In the
distal region, there was no difference in the propor-
tions of bone graft, air, and water between the
2 groups. This has been supported by a previous
femoral IBG study on cadaveric femurs that also
demonstrated increased graft compaction in the
region where the distal impactors were used .
There are several possibilities for this. The distal
with the bone graft; therefore, the vibrating area will
have less contact and, consequently, less effect on
realigning the bone graft particles. An axial compres-
than a radial compressive force (which occurs with
the tapered proximal impactors). Consequently, the
bone graft distally is often very well compacted using
the standard technique, therefore making it more
difficult to detect a significant improvement. Finally,
much less than that in the proximal impaction,
resulting in the dissipation of fluid around the sides
and top of the distal impactors—thus nullifying the
potential benefits from the addition of drainage holes
to distal impactors.
1162 The Journal of Arthroplasty Vol. 23 No. 8 December 2008
It is important to note that the improvement in
bone graft compaction has not only a mechanical
advantage but also a potential biological advantage.
Greater bone graft compaction should result in less
cement penetration into the interparticulate spaces,
leaving a greater bone graft mantle for neovascular-
ization and bone remodeling to take place. Studies
have demonstrated a positive correlation between
increased cement penetration and delayed revascu-
larization of the endosteal surface—essential for
graft incorporation and remodeling .
The cement volume in the middle region of the
femur was reduced, which correlated with the
improved graft compaction also observed in this
region, in the vibration-assisted group. Similarly, in
the distal region, where there was no difference in
graft compaction, the cement volume also remained
unchanged between the 2 groups. However, in the
proximal region, the cement volume remained the
same despite an increase in the percentage of bone
graft, suggesting inferior compaction in comparison
with the middle region. Possible explanations
include the fact that the proximal bone graft not
only sustains the fewest impactions/exposure to
vibration but also is the most difficult to contain, an
essential requirement for good graft impaction.
Therefore, overall, across the 3 regions, there was
no increase in cement volume in the vibration-
assisted group. A previous study that used this
technique demonstrated that prosthetic stability is
enhanced, with less subsidence observed after
cyclical loading . Therefore, with no increase
in the cement volume, we postulate that the
improved prosthetic stability is a direct result of
improved bone graft compaction.
The underlying principle of IBG is to achieve
stability of the implant and subsequently allow the
restoration of living bone stock by bone ingrowth.
However, in achieving good mechanical stability,
the environment for the biological remodeling of
the graft may be compromised. Examples include
(a) washing (this removes fat, increasing shear
strength of the bone graft, but also removes growth
factors and cytokines that encourage ingrowth and
new bone formation [7,8]), (b) compaction energy
(the energy imparted during impaction has a
positive correlation with the shear strength of the
graft but a negative correlation with bone ingrowth
), and (c) cement penetration. The vibration
technique has improved compaction of the bone
graft, producing a greater bone graft mantle with
improved support for the femoral prosthesis. A
thicker graft mantle unaffected by the thermal
reaction and detrimental effects of bone cement
may result in improved endosteal blood supply,
neovascularization, and, ultimately, bone remodel-
ing and incorporation. However, further in vivo
studies are required to establish this and the
longer-term effects this technique has on bone
incorporation and remodeling.
This study has demonstrated that vibration
combined with perforated tamps to allow drainage
of excess fluid significantly improves bone graft
compaction as compared with the current surgical
technique. Further characterization of the vibration
hammer and perforated tamps, including the effects
of vibration frequency and amplitude, air pressure,
and the number, size, and position of the holes in
the tamps, is required to determine optimum
compaction conditions. These results offer new
clinical approaches and therapeutic implications
therein to a significant orthopedic problem.
We thank Eric Bonner, Robert Barnes, and John
Lester for their invaluable technical assistance
throughout the project; Stryker UK for providing
the IBG kit; DePuy for supplying bone cement and
cement mixing kits; and Finsbury Orthopaedics for
the loan of the vibration device.
1. Slooff TJ, Huiskes R, van HJ, et al. Bone grafting in
total hip replacement for acetabular protrusion. Acta
Orthop Scand 1984;55:593.
2. Gie GA, Linder L, Ling RS, et al. Contained morselized
allograft in revision total hip arthroplasty. Surgical
technique. Orthop Clin North Am 1993;24:717.
3. Ornstein E, Atroshi I, Franzen H, et al. Results of hip
revision using the Exeter stem, impacted allograft
bone, and cement. Clin Orthop Relat Res 2001:126.
4. Halliday BR, English HW, Timperley AJ, et al. Femoral
impaction grafting with cement in revision total hip
replacement. Evolution of the technique and results.
J Bone Joint Surg Br 2003;85:809.
5. Eldridge JD, Smith EJ, Hubble MJ, et al. Massive early
subsidence following femoral impaction grafting.
J Arthroplasty 1997;12:535.
6. Sharpe P. Impaction grafting or cement alone for
femoral revision hip replacement. In: Keene GS,
Howie DW, Graves SE, et al, editors. Conference
proceedings. Sydney, Australia: ANZORS; 1998.
7. van der DS, Weernink T, Buma P, et al. Rinsing
morselized allografts improves bone and tissue
ingrowth. Clin Orthop Relat Res 2003:302.
8. Dunlop DG, Brewster NT, Madabhushi SP, et al.
Techniques to improve the shear strength of impacted
bone graft: the effect of particle size and washing of
the graft. J Bone Joint Surg Am 2003;85-A:639.
Vibration Femoral IBG ?
Bolland et al
9. Arts JJ, Gardeniers JW, Welten ML, et al. No Download full-text
negative effects of bone impaction grafting with
bone and ceramic mixtures. Clin Orthop Relat Res
10. Blom AW, Cunningham JL, Hughes G, et al. The
compatibility of ceramic bone graft substitutes as
allograft extenders for use in impaction grafting of the
femur. J Bone Joint Surg Br 2005;87:421.
11. van Haaren EH, Smit TH, Phipps K, et al. Tricalcium-
phosphate and hydroxyapatite bone-graft extender
for use in impaction grafting revision surgery. An in
vitro study on human femora. J Bone Joint Surg Br
12. Frei H, Mitchell P, Masri BA, et al. Allograft impaction
and cement penetration after revision hip replace-
ment. A histomorphometric analysis in the cadaver
femur. J Bone Joint Surg Br 2004;86:771.
13. Smith GN. Elements of Soil Mechanics. 6th ed.
Oxford, Blackwell Science; 1990.
assisted bone-graft compaction in impaction bone
grafting of the femur. J Bone Joint Surg Br 2007;89:
15. Frei H, Mitchell P, Masri BA, et al. Mechanical
characteristics of the bone-graft–cement interface
after impaction allografting. J Orthop Res 2005;
16. Frei H, O'Connell J, Masri BA, et al. Biological and
mechanical changes of the bone graft-cement inter-
face after impaction allografting. J Orthop Res
17. Heiner AD, Brown TD. Structural properties of a new
design of composite replicate femurs and tibias.
J Biomech 2001;34:773.
18. Lambe TW, Whitman RV. Soil mechanics—SI version.
London, John Wiley & Sons, 1979.
19. Coelho PALF, Haigh SK, Madabhushi SPG, et al. Post-
earthquake behaviour of footings when using densi-
fication as a liquefaction resistance measure (Invited
paper). Ground Improv J 2007;11:45.
20. Brennan AJ, Madabhushi SPG. Liquefaction and
drainage in stratified soil. J Geotech Geoenviron Eng
21. de Waal MJ, Slooff TJ, Huiskes R, et al. Vascular
changes following hip arthroplasty. The femur in
goats studied with and without cementation. Acta
Orthop Scand 1988;59:643.
22. Tagil M, Aspenberg P. Impaction of cancellous bone
grafts impairs osteoconduction in titanium chambers.
Clin Orthop Relat Res 1998:231.
1164 The Journal of Arthroplasty Vol. 23 No. 8 December 2008