Surgical Science, 2011, 2, 183-187
doi:10.4236/ss.2011.24040 Published Online June 2011 (
Copyright © 2011 SciRes. SS
Leg Length Discrepancy in Cementless Total Hip
Christopher N. Peck, Karan Malhotra, Winston Y. Kim
Department of Trauma and Orthopaedics, Salford Roya l Hospital NHS Foundation Trust,
Manchester, Britain
Received August 17, 2010; revised April 8, 2011; accepted April 18, 2011
The use of cementless total hip arthroplasty (THA) is on the increase. In order to achieve rotational and axial
stability larger implants may be required than originally templated for. This could potentially result in a lar-
ger leg length inequality. Our objective was to determine whether there is greater inequality in leg length
post-operatively in cementless THA as compared to cemented implants. 136 consecutive patients undergoing
elective THA between June 2007 and May 2008 were included. Post-operative digital radiographs were ex-
amined to determine leg length. Twenty seven patients (20%) underwent a cemented procedure and 109
(80%) a cementless procedure. In the cemented group the mean leg length discrepancy was 7.3 mm (range
19 mm short to 21 mm long). In the cementless group the mean measured leg length discrepancy was 6.3 mm
(range 18 mm short to 23 mm long). There was no significant difference between the two groups (P = 0.443).
This study shows that with accurate pre-operative templating, both cemented and cementless procedures
produce comparable and acceptable leg length discrepancies.
Keywords: Hip, Arthroplasty, Cementless, Leg Length
1. Introduction
Total hip arthroplasty is a very successful operation in
the management of end stage hip osteoarthritis with over
64,000 primary procedures being performed in the UK
last year [9]. Leg length discrepancy (LLD) is a well
documented complication of total hip replacement. It has
been shown to correlate strongly with patient dissatisfac-
tion [5,8,11,17] and is a leading cause of litigation [4].
Reports in the literature show a change in leg length fol-
lowing THA from 21 mm short to 35 mm long [5,7,15].
Studies have shown that a discrepancy greater than 10
mm can result in a limp, back pain, sciatica, stiffness, hip
dislocation, the need for a shoe raise and early failure
[7,13,15,17]. These problems are particularly problem-
atic in the young er population where the impact on qual-
ity of life may be more marked.
Cementless hip replacements are becoming increas-
ingly popular with the perceived advantages of increased
longevity and easier revision procedures. A potential
disadvantage of a cementless system is that to achieve
rotational and axial stability larger implants may be re-
quired intra-operatively than originally templated for.
This may result in inadvertent leg lengthening. The con-
verse may occur if stability is achieved with a smaller
implant. The use of a modular implant system means that
the surgeon can make adjustments to the final leg length
if necessary but this may affect the overall stability of the
The aim of this study was to determine if leg length
discrepancy exists to a greater degree in cementless THA
compared with cemented hip replacements.
2. Patients and Methods
This retrospective study was performed at a single insti-
tution, a large University Teaching Hospital in the
North-West of England. All surger i es were carried out by
one of five lower limb arthroplasty consultants. The in-
clusion criteria were all patients undergoing a primary
elective total hip arthroplasty (THA) between June 2007
and May 2008. Patients were excluded if the contralat-
eral hip had already been replaced.
Preoperative templating was done using an overlay
technique on hard film copies of digital radiographs. The
size of the acetabular cup was templated first and its cen-
184 C. N. PECK ET AL.
tre of rotation marked. Using this, the size of the femoral
stem was determined and the proposed neck cut that
would best reproduce the normal neck-shaft angle and
offset. All procedures were carried out under either a
general or spinal anaesthetic and were performed with
the patient in the lateral position through a posterior ap-
proach. Leg length was assessed intra-operatively by
comparing the relative position of the knees through the
drapes and the soft tissue tension during trial reduction.
The cemented femoral implants used were the collarless
polished taper (CPT) (Zimmer, Swindon, UK) and the
Exeter stem (Stryker, Howmedica Osteonics, Berkshire,
UK). The cementless femoral implants used were the
Zimmer M/L Taper Hip Prosthesis with standard and
extended offsets and modular heads and the CLS
Zimmer cementless stem (Zimmer, Swindon, UK).
All post-operative radiographs were in the digital for-
mat on the Patient Archive and Communication System
(PACS, General Electric) and measurements were made
using the systems integrated measurement tools. An-
tero-posterior digital radiographs of the pelvis, centred
on the pubic symphysis with the hips internally rotated
15˚ were taken on the first post-operative day, prior to
mobilisation. Limb length inequality was calculated us-
ing the method described by Woolson et al. [16] and
subsequently by a number of other author s [5,8,11,12 ,15].
This is measured as the perpendicular distance from the
inter-teardrop line to the most prominent point on the
lesser trochanter of the femur. The distance on the
post-operative side was compared to the contralateral
limb which was assumed to be equal to the initial length
of the operated limb before development of joint disease.
Magnification was adjusted for by using a standard esti-
mation of 20% magnification consistent with the litera-
ture [3,10,14,15]. All measurements were made on the
initial post-operative radiographs and were made to the
nearest millimetre by a single observer not involved in
the surgical procedures. Each measurement was taken
twice, at the same sitting, and an average value used to
reduce intra-observer errors.
All data was analysed using Excel for Windows (Mi-
crosoft). Continuous parametric data were analysed using
the two sample T-test and nonparametric data were ana-
lysed using the chi squared test (p < 0.05 was considered
as significant).
3. Results
Between June 2007 and May 2008 166 patients under-
went a primary total hip arthroplasty (THA). Thirty pa-
tients had a pre-existing contralateral hip replacement
and so were excluded. Of the 136 patients in the study 27
(20%) underwent a cemented THA and 109 (80%) a ce-
mentless procedure. Fifty eight (43%) patients were male
and 78 (57%) were female. The mean age of patients was
66.7 years in the cemented group and 66.4 in the ce-
mentless (Table 1).
Post-operatively in the cemented group seven patients
(26%) had some degree of radiographic shortening, 19
(74%) had some degree of lengthening and one patient
(4%) had equal leg lengths. In the cementless group 32
patients (29%) had some degree of radiographic short-
ening, 67 (61%) had some degree of lengthening and 10
patients (9%) had equal leg lengths. There was no sig-
nificant difference between these proportions (P =
When corrected for an estimated magnification of 20%
the mean radiographic limb length inequality regard-
less of direction was 7.3 mm in the cemented group
and 6.3 mm in the cementless group (P = 0.496). In the
cemented group two patients (7%) had shortening greater
than 10 mm (mean 16.5 mm) and four patients (15%)
had lengthening greater than 10 mm (mean 17.3 mm). In
the cementless group five patients (5%) had shortening
greater than 10 mm (mean 14.0 mm) and 17 patients (16%)
had lengthening greater than 10mm (mean 14.0 mm).
There was no significa nt diff erence between these groups
(P = 0.331 for shortening, P = 0.140 for lengthening).
These results are summarised in Table 1.
4. Discussion
To be successful, hip replacement surgery needs to ac-
curately reconstruct the patient’s own anatomy and bio-
mechanics. This includes reproducing the centre of rota-
tion of the hip joint, the offset and leg length. This re-
quires the appropriate selection and orientation of im-
plants and if not done appropriately will lead to post op-
erative complications and patient dissatisfaction.
Leg length discrepancy can have a significant impact
on patient satisfaction and is one of the leading causes of
the litigation following THA in America [4,7]. Leg
lengthening is much less tolerated than leg shortening,
with lengthening over 10 mm being very poorly tolerated
by the patient [16]. Lengthening of the operated limb can
lead to joint stiffness, sciatic nerv e palsy, low back pain,
early failure of the prosthesis and the need for revision
surgery [7,13,15,17]. Shortening can impair hip abduc-
tion and increase the risk of dislocation [13].
In a study of 75 patients Konyves et al. [5] showed
that the post-operative Oxford Hip Score was signifi-
cantly worse in patients with leg lengthening compared
to those with equal leg lengths. Most other studies confirm
that a discrepancy in leg length has a direct adverse ef-
fect on clinical and functional outcome [8,11,17]. In con-
trast, White et al. [15] showed no statistically significant
Copyright © 2011 SciRes. SS
Copyright © 2011 SciRes. SS
Table 1. Patient demographics and post-operative radiographic leg length discrepancy (LLD) in cemented and cementless
THA. (SD Standard Deviation).
Cemented THA Cementless THA Significance
Number of Patients 27 (20%) 109 (80%)
Mean Age (Range) 66.7 years (30 - 87) 66.4 years (37 - 88) p = 0.930
Males (M)/Females (F) M = 15 (56%)
F = 12 (44%) M = 43 (39%)
F = 66 (61%)
Overall mean LLD (SD) 7.3 mm (6.0) 6.3 mm (4.9 mm) p = 0.496
Mean (SD) Shortened LLD [Range] 6.8 mm (7.2)
[1 to 19 mm] 6.8 mm (4.0)
[1 to 18 mm] p = 0.956
Mean (SD) Lengthened LLD [Range] 8.0 mm (5.5)
[1 to 21 mm] 7.0 mm (5.0)
[1 to 23 mm] p = 0.475
difference in the Harris Hip Score and SF 36 Health Sur-
vey in patients with radiological leg lengthening or
shortening. However, their outcome scores were meas-
ured at six months after surgery so long-term problems
were not assessed. Furthermore it has been suggested
that the Harris Hip Score may not fully take into account
the patients’ subjective experience of pain [17]. The po-
sition of the acetabular component has a much smaller
impact on leg length, it has been shown that only 2% of
patients with a leg length discrepancy were attributed to
the position of the acetabular component [5].
True and apparent leg length should be measured pre-
operatively in all patients to gu ide the surgeon as to how
the joint should be reconstructed [7]. Intra-operatively,
with the patient in the lateral position, leg length can be
judged by comp aring the relative knee and an kle positio n
and by assessment of soft tissue tension. However, this is
not always accurate as any small change in the position
of the patient can lead to a big change in apparent leg
length and be misleading [11,16]. An additional part of
the preoperative plan is hip templating which is routin ely
used to guide the surgeon towards the size and placement
of implants, the neck length, offset, level of the femoral
osteotomy and the restoration of limb length required.
This can be done by the traditional overlay method or by
using a computer package for digital radiographs. These
methods have been shown to have variable correlation
with the actual implants used [2,14] and so may be de-
ceptive to the surgeon leading to difficulty in achieving
equal leg lengths.
Despite evidence of better short to medium term re-
sults for cemented implants there is an increasing ten-
dency towards cementless fixation with the perceived
advantages of increased longevity and easier revision
procedures. In 2008 33% of implants were cementless
compared to 21% in 2004 and this trend look s set to con-
tinue [9]. With cementless hip replacements rotational
and axial stability is achiev ed intra-op eratively by a pr ess
fit technique. We felt that in many cases a larger implant
was required to accomplish this th an was originally tem-
plated for with the possibility of inadvertently causing
leg length discrepancy. The majority of cementless
components increase in offset with increasing component
size and therefore lengthen the limb. Modular implant
systems may allow the surgeon to make adjustments to
the final leg length if necessary but this may affect the
overall stability of the joint. Final adjustments would be
impossible with a mono-block system. In a laboratory
study Barink et al. [1] assessed how well two different
cementless hip systems matched the final rasp position.
Using synthetic and cadaveric femurs the average
rasp-stem mismatch was within 2 mm in three different
planes, the mismatch being larger in the cadaveric fe-
murs. The authors felt that this mismatch was of low
clinical relevance. However, any mismatch may be mis-
leading when performing a trial reduction to determine
the final implant size. As far as we are aware this study
has not been repeated with other cementless stems leav-
ing any mismatch with th ese syste ms unknown.
There are very few clinical or radiological studies in
the literature evaluating leg length discrepancy in ce-
mentless hip replacements. In a radiographic analysis
Leonard et al. [6] compared cemented and cementless
hip replacements for offset and limb length. They
showed that cementless procedures resulted in a greater
degree of leg lengthening with a mean of 5.6 mm com-
pared to 3.8 mm in the cemented group, this was statisti-
cally significant. Our results show an overall mean leg
length discrepancy of 7.3 mm in the cemented group and
6.3 mm in the cementless group which is consistent
with other published series [5,7,15]. In contrast to Leo-
nard et al. [6] we found no significant difference in ra-
diographic leg length discrepancy between cementless
and cemented implants. This may simply be due to the
spread of our results shown by the high standard devia-
tions, possibly due to the five different surgeons per-
186 C. N. PECK ET AL.
forming the hip replacements. However, no measure of
spread was provided in Leonard’s paper.
We acknowledge the limitations of the present study
which should be considered when in terpreting the results.
The surgeries were carried out by five different consult-
ants in lower limb reconstruction each with their own
variations on templating and intraoperative methods to
equalise leg lengths. No account was taken of the
pre-existing limb length which can have a direct impact
on the post operative limb length. Additionally, no as-
sessment was made of disease in the contralateral hip
which may also have a bearing on a surgeons desired
final leg length of the operated hip . Due to the retrospec-
tive nature of the study we were unable to compare the
size of the prosthesis used with the size of prosthesis
templated for. This would have given us a better under-
standing as to whether the size of final implants may
have changed leg length.
The general limitations of using radiographs to make
our measurements include the variation in positioning of
the pelvis with respect to the x-ray film and variations in
relative magnification due to distance from the film.
However, as these factors affect both hips and we are
comparing the difference between the two it is reason-
able to use plain radiographs to assess leg length. Our
methods of measuring leg length discrepancy have been
used in numerous other studies [5,8,11,12,15] and are
considered to be accurate and reliable. To calculate the
magnification the radiographic diameter of the prosthetic
femoral head can be compared to the actual diameter of
the prosthesis. This data was not available in the current
study so a standard magnification factor of 20% was
used to adjust the actual measurements taken from the
radiographs. However, this is the standard magnification
used in most templating systems and consistent with
other studies [3,10,14,15].
5. Conclusions
In conclusion, total hip arthroplasty is a safe, effective
and reproducible treatment for end stage degenerative
hip disease. The current trend is towards the use of ce-
mentless implants particularly in the younger patient.
Our study shows that there is no significant difference in
leg length discrepancy between cementless and cemented
implants. This suggests that with accurate pre-operative
templating both cemented and cementless procedures
produce comparable and acceptable leg length discrep-
6. Conflict of Inter e st S ta t e me nt
No benefits in any form have been received or will be
received from a commercial party related directly or in-
directly to the subject of this article.
7. References
[1] M. Barink, H. Meuers, M. Spruit, C. F. Fankhauser and N.
Verdonschot, “How Close Does an Uncemented Hip
Stem Match the Final Rasp Position?,” Acta Orthopae-
dica Belgica, Vol. 70, No. 6, 2004, pp. 534-539.
[2] L. W. Carter, D. O. Stovall and T. R. Young, “Determi-
nation of Accuracy of Preoperative Templating of Non-
cemented Femoral Prostheses,” The Journal of Arthro-
plasty, Vol. 10, No. 4, 1995, pp.507-513.
[3] K. S. Conn, M. T. Clarke and J. P. Hallett, “A Simple
Guide to Determine the Magnification of Radiographs
and to Improve the Accuracy of Preoperative Templat-
ing,” Journal of Bone and Joint Surgery, Vol. 84-B, No.
2, 2002, pp. 269-272.
[4] A. A. Hofmann and M. C. Skrzynski, “Leg Length Ine-
quality and Nerve Palsy in Total Hip Arthroplasty: A
Lawyer Awaits,” Orthopaedics, Vol. 23, No. 9, 2000, pp.
[5] A. Konyves and G. C. Bannist er, “The Importance of Leg
Length Discrepancy after Total Hip Arthroplasty,” Jour-
nal of Bone and Joint Surgery, Vol. 87-B, No. 2, 2005, pp.
[6] M. Leonard, P. Magill, P. Kiely and G. Khayyat, “Radio-
graphic Comparison of Cemented and Uncemented Total
Hip Arthroplasty and Hip Resurfacing,” European Jour-
nal of Orthopaedic Surgery and Traumatology, Vol. 17,
No. 6, 2007, pp. 583-586.
[7] W. J. Maloney and J. A. Keeney, “Leg Length Discrep-
ancy after Total Hip Arthroplasty,” The Journal of Ar-
throplasty, Vol. 19, No. 4, 2004, pp. 108-110.
[8] S. B. Murp hy and T. M. E cke r , “Ev a lua ti o n of a New L eg
Length Measurement Algorithm in Hip Arthroplasty,”
Clinical Orthopaedics and Related Research, Vol. 463,
2007, pp. 85-89.
[9] National Joint Registry 6th Annual Report, 2009.
[10] M. J. Oddy, M. J. Jones, C. J. Pendergrass, J. R. Pilling
and J. A. Wimhurst, “Assessment of Reproducibility and
Accuracy in Templating Hybrid Total Hip Arthroplasty
Using Digital Radiographs,” Journal of Bone and Joint
Surgery, Vol. 88, No. 5, 2006, pp. 581-585.
[11] C. S. Ranawat, R. R. Rao, J. A. Rodriguez and H. S.
Bhende, “Correction of Limb-Length Inequality During
Total Hip Arthroplasty,” The Journal of Arthroplasty,
Vol. 16, No. 6, 2001, pp. 715-721.
[12] N. Sugano, T. Nishii, H. Miki, H. Yoshikawa, Y. Sato,
and S. Tamura, “Mid-Term Results of Cementless Total
Copyright © 2011 SciRes. SS
Copyright © 2011 SciRes. SS
Hip Replacement Using a Ceramic on Ceramic Bearing
with and Without Computer Navigation,” Journal of
Bone and Joint Surgery, Vol. 89, No. 4, 2007, pp. 455-460.
[13] K. T. Suh, S. J. Cheon and D. W. Kim, “Comparison of
Preoperative Templating with Post-Operative Assessment
in Cementless Total Hip Arthroplasty,” Acta Orthopae-
dica Scandinavica, Vol. 75, No. 1, 2004, pp. 40-44.
[14] A. Unnanuntana, D. Wagner and S. B. Goodman, “The
Accuracy of Preoperative Templating in Cementless To-
tal Hip Arthroplasty,” The Journal of Arthroplasty, Vol.
24, No. 2, 2009, pp. 180-186.
[15] T. O. White and T. W. Dougall, “Arthroplasty of the Hip
Leg Length Is Not Important,” Journal of Bone and Joint
Surgery, Vol. 84-B, No. 3, 2002, pp. 335-338.
[16] S. T. Woolson, J. M. Hartford and A. Sawyer, “Results of
a Method of Leg-Length Equalization for Patients Un-
dergoing Primary Total Hip Replacement,” The Journal
of Arthroplasty, Vol. 14, No. 2, 1999, pp. 159-164.
[17] V. Wylde, S. L. Whiteh ouse, A. H. Tay l or, G. C. Banni ster
and A. W. Blom, “Prevalence and Functional Impact of
Patient-Perceived Leg Length Discrepancy after Hip Re-
placement,” International Orthopaedics, Vol. 33, No. 4,
2009, pp. 905-909.