Journal of Orthopaedic Science 21 (2016) 640e646

Contents lists available at ScienceDirect

Journal of Orthopaedic Science

journal homepage: http://www.elsevier.com/locate/jos

Original article

Biomechanical comparison of oblique and step-cut osteotomies used

in total hip arthroplasty with femoral shortening
€
Fatih Yıldız a, *, Onder I._ Kılıçog
lu b, Go
€ ksel Dikmen b, Ergun Bozdag
 c, Emin Sünbülog
lu c,
Meral Tuna c
a
Department of Orthopedics and Traumatology, Bezmialem Vakıf University School of Medicine, Istanbul, Turkey
b
Department of Orthopedics and Traumatology, Istanbul University School of Medicine, Istanbul, Turkey
c
Department of Biomechanics, Istanbul Technical University, Istanbul, Turkey

a r t i c l e i n f o a b s t r a c t

Article history: Background: Various types of shortening osteotomies and prosthesis are used for femoral reconstruction
Received 13 October 2015 in total hip arthroplasty of the high hip dislocation. This biomechanical study investigates whether step-
Received in revised form cut osteotomies result in better stability than oblique osteotomies and cylindrical femoral stems enhance
26 March 2016
stability of the osteotomy more than conical stems, and which osteotomy and prosthesis type maintain
Accepted 24 April 2016
Available online 9 June 2016
the stability better after cyclical loading.
Methods: Oblique and step-cut shortening osteotomies were compared under axial and rotational forces,
using synthetic femur models and conical or cylindrical femoral prostheses. The models underwent
cyclic loading for 10,000 cycles at 3 Hz (100e1000 N axial bending or 0.5e10 Nm torque). After the
completion of cyclic loading, the models were loaded until failure. Stiffness values before and after
cyclical loading, and failure loads were the outcome parameters. Relative displacements at the osteotomy
sites were also measured using 3-Dimensions Digital Imaging Correlation System.
Results: The mean failure load was significantly higher in conical prosthesis groups under axial forces. In
torsion tests, the mean stiffness of conical prosthesis groups after cyclical loading was higher in oblique
osteotomies. The other parameters were similar between the groups.
Conclusions: According to the results of the study, although some individual statistically significant
parameters were obtained, step-cut osteotomies, which are technically challenging procedures, were not
found biomechanically superior to oblique osteotomies, with neither conical nor cylindrical prostheses.
© 2016 The Japanese Orthopaedic Association. Published by Elsevier B.V. All rights reserved.

1. Introduction The literature reports use of both conical, metaphyseal press-fit

[12,13] and cylindrical, metaphyseal and diaphyseal press-fit [14,15]
In the treatment of high hip dislocation (HHD) secondary to femoral components after the osteotomies. Stability of the osteot-
acetabular dysplasia with total hip arthroplasty (THA), recon- omy, which is affected by the geometry of the osteotomy and type
structing rotational center of the hip in its anatomic level is crucial of the femoral component, should be increased in order to decrease
[1e4]. However, distraction of over 4e5 cm in the lower extremity the rate of nonunion. Metaphyseal and diaphyseal press-fit, straight
during the reduction of the femoral component may pose high risk femoral component designs are thought to have higher rotational
for the neurovascular structures. Femoral shortening osteotomies fixation rigidity [15].
are expected to minimize this risk [5e7]. Subtrochanteric osteot- This study compares the axial and rotational stability of the
omies may be transverse [8], oblique [9], step-cut [10] or V-shaped oblique and step-cut osteotomies to investigate whether the obli-
[11]. Currently, the choice is generally made on the pathoanatom- que osteotomy, which is relatively simple to perform, can provide
ical characteristics of the femur, and the preference and experience similar rotational stability as the step-cut osteotomy [16]. Effect of
of the surgeon. femoral component on the fixation rigidity was also tested with
conical or cylindrical prostheses. Hypotheses of the current study
* Corresponding author. Department of Orthopedics and Traumatology, Bez- were that the step-cut osteotomy results in higher strength than
mialem Vakif University, School of Medicine, Vatan Cad, 34093 Fatih, Istanbul,
the oblique osteotomy, cylindrical femoral stems result in increased
Turkey. Tel.: þ90 212 4531700x2505.
E-mail address: [email protected] (F. Yıldız). fixation rigidity of the osteotomy than conical stems, and the step-

http://dx.doi.org/10.1016/j.jos.2016.04.015
0949-2658/© 2016 The Japanese Orthopaedic Association. Published by Elsevier B.V. All rights reserved.
 F. Yıldız et al. / Journal of Orthopaedic Science 21 (2016) 640e646 641

cut osteotomy and cylindrical stems maintain the stability of the

osteotomy better after cyclical loading.

2. Materials and methods

This biomechanical study was performed by using a servohy-

draulic test device (MTS 858 Mini Bionix II, MN, USA) for axial and
torsional loading. Sixty-four synthetic identical right femur models
(Synbone 2230, Synbone AG, Switzerland) were divided into 4
groups each consisting 16 specimens: Group I (oblique osteotomy-
conical stem), group II (step-cut osteotomy-conical stem), group III
(oblique osteotomy-cylindrical stem) and group IV (step-cut
osteotomy-cylindrical stem). For conical and cylindrical femoral
stems, we used Synergy no.: 16 (Fig. 1A) and Primary Echelon no.:
17 (Fig. 1B) (Smith&Nephew, Memphis, TN, USA) for all models. In
the piloting phase of the experiment, femoral medullae were
reamed and broached with the largest broach possible after the
shortening osteotomy, and the appropriate sizes were determined
for the press-fit placement of the two types of stems.
In order to have a standard head resection and a shortening
osteotomy, osteotomy lines were measured and drawn on the
models, and cuts were made using a low profile electric saw. The Fig. 2. A. Femoral neck osteotomy and 3 cm of oblique subtrochanteric shortening, B.
neck osteotomy was performed 1.5 cm proximal to the trochanter Reconstructed femur using the oblique osteotomy, C. Cuts of the step-cut osteotomy, D.
After 3 cm of subtrochanteric shortening and femoral stem insertion.
minor and with an angle of 45 relative to the shaft. After reaming
and broaching, 3 cm of subtrochanteric resection was applied, using
one of the osteotomy types. For the oblique osteotomy, the first cut
of the construct could be placed perpendicular to base of the PVC
was performed with a 45 angle to the anatomical axis of the femur
pipe and its long axis would pass through the center of the PVC
and 20 mm distal to the level of trochanter minor. The second cut
pipe.
was parallel and 30 mm distal to the first one (Fig. 2A and B). For the
step-cut osteotomy, distal hemi-transverse cut of the osteotomy
2.1. Test protocol
was on the medial side and 60 mm distal to the trochanter minor
and proximal cut was on the lateral side at the level of the
Of the 16 models in each group, 8 were tested under axial forces
trochanter minor. Medial and lateral hemi-transverse osteotomies
and the remaining 8 under rotational forces. Any displacement at
were connected by a longitudinal cut, and with a transverse cut
the osteotomy site was detected using 3-Dimensional Digital Im-
between the first two, the step-cut osteotomy was completed
aging Correlation system (Vic3D-Correlated Solutions Inc.). This
(Fig. 2C and D). After insertion of the femoral stems in the medulla
system was capable to measure relative displacements between the
of the shortened models, the osteotomies were fixed with a 1.7-mm
fragments above and below the osteotomy site, as small as 1 mm in
steel cable and tensioned with a standard load of 550 N in the
medio-lateral (x), vertical (y) and antero-posterior (z) directions
midline of the osteotomy [17].
[18,19].
Two cross Kirschner wires were inserted to the femoral condyles
and the distal end of the femur specimens were embedded in
polyester putty using PVC pipes as scaffolds. A custom made 2.1.1. Axial load tests
centralizer was used during this procedure so that the femoral stem The force was transmitted to the femoral stem, which was kept
at 16 valgus simulating the single-leg stance phase of gait, through
a þ0 head (Smith&Nephew, Memphis, TN, ABD) and a cup similar
to the acetabular liner [18,19](Fig. 3). After a preload (3 Hz,
100e1000 N, 10 cycles) was applied for the stability of the con-
structions, the system was released and loaded again (initial
loading) up to 1000 N with a velocity of 50 N/sec in order to
measure the baseline stiffness values [18,19]. The specimens which
had not failed during initial loading tests were loaded with 1000 N
at 3 Hz for 10,000 cycles to simulate walking with partial weight
bearing after hip arthroplasty with femoral shortening [18e20].
After each 1000 cycles, stiffness of the models and displacements at
the osteotomy sites were recorded. In specimens where failure did
not occur in the system until this stage, finally, the models were
loaded to failure axially using displacement control mode at a
speed of 15 mm/min.

2.1.2. Rotational load tests

The femoral stems were rotated around their longitudinal
anatomic axis in order to minimize bending moments in the
Fig. 1. A. Synergy no. 16 (conical cross-section 16 mm, stem length: 170 mm), the line
specimens during rotation [19]. For this reason, distal ends of the
of the oblique osteotomy and B. Primary Echelon no. 17 (conical cross-section 17 mm, specimens were fixed to the torsional load cell of the testing ma-
stem length: 150 mm) and the step-cut osteotomy lines. chine and the torque was applied through a custom made
642 F. Yıldız et al. / Journal of Orthopaedic Science 21 (2016) 640e646

loading tests and using torque-angle of rotation curves for the

rotational loading tests. The direction and amount of the relative
displacements at the osteotomy sites, which were not yet measured
before in the literature to our knowledge, were also recorded.
The data were analyzed by using MedCalc v11.6 statistical
package. Normality of data was tested using KolmogoroveSmirnov
test. Presence of any significant difference between groups was
tested using one-way ANOVA test, while comparisons among the
groups were performed with the post-hoc test of Stu-
denteNewmaneKeuls. The significance level was accepted as
p < 0.05.

3. Results

All specimens tested under axial forces completed the initial

loading and cyclic loading tests, and were loaded until failure in the
third stage. Four of the 32 models planned for use in rotational
loading experiments were excluded from the study because two
models have failed during insertion of the cylindrical prosthesis in
step-cut osteotomy groups and the other two were broken due to
programming errors of the test machine, which exerted out-of-
protocol force on the specimens. Thus, seven specimens in each
group were tested under rotational forces. In group I, during the
initial loading, fixation of the osteotomy failed without a fracture in
one model. One specimen was broken during cyclical loading and 5
specimens by failure torque. In group II, one model failed during
initial loading, three models failed during cyclical loading and the
remaining three by failure torque. In group III, three models failed
in the first 1000 cycles without a fracture and the remaining four by
failure torque. In group IV, one specimen failed during initial
loading, two during cyclical loading and four by failure torque.
The results of the outcome parameters were summarized in
Tables 1 and 2.
Comparing the osteotomy types (group I vs. II and group III vs.
IV), differences in the mean initial stiffness values were not sta-
tistically significant, for both axial (p ¼ 0.130) and torsional testing
(p ¼ 0.079). After cyclical loading, under axial forces, the difference
between the groups was also not significant (p ¼ 0.106). However,
under rotational forces, it was significantly higher only in group I
than group II (p ¼ 0.043). The mean failure loads did not vary
among different osteotomy types, for both axial (p > 0.1) and
Fig. 3. Experimental setup of the axial loading tests. torsional (p ¼ 0.53) testing protocols.
Comparing the prosthesis types (group I vs. III and group II vs.
IV), the mean stiffness values of the groups, before (Figs. 5A and 6A)
instrument, which was fixed to the neck of the femoral stems and after cyclical loading (Figs. 5B and 6B), were not significantly
(Fig. 4). Following preloading (3 Hz, 0.5e10 Nm, 10 cycles) the different for either testing. The mean failure torque was also found
system was released and the models were torsioned (initial torque) to be similar between the prostheses types (p ¼ 0.53), under
with a velocity of 0.2 Nm/s and 10 Nm of torque to measure initial rotational forces (Fig. 6C). But, under axial forces, the mean failure
stiffness values. Then, 10,000 cycles of torsion (0.5e10 Nm) was load was significantly higher in the conical prostheses groups than
applied at 3 Hz under torsional control. After each 1000 cycles, those with cylindrical (p < 0.001) (Fig. 5C).
stiffness values of the models and displacements at the osteotomy After axial cyclic loading, the mean amount of displacements in
site were recorded. At the end of 10,000 cycles of loading, torque groups I, II, III and IV were found to be similar in medio-lateral (x-
controlled loading was replaced by angle controlled loading in the axis) (p ¼ 0.136), and antero-posterior (z-axis) directions
specimens that did not yet fail, and torsion was exerted at a velocity (p ¼ 0.106), however, it was statistically less in step-cut osteotomy
of 10 /min to fail the models. groups in only the vertical (y-axis) (p < 0.001) direction (Fig. 5D). In
the torsional groups, they were not significantly different in medio-
2.2. Outcome measures lateral (x-axis) (p ¼ 0.317), vertical (y-axis) (p ¼ 0.293) and antero-
posterior (z-axis) directions (p ¼ 0.159) (Fig. 6D).
Failure was defined as a sudden drop in the forceedisplacement
curve at any time during axial or torsional loading, or a new fracture 4. Discussion
at any point in the model, and was accepted as a criterion to end the
experiment. The highest force/torque which caused failure in the In the treatment of HHD with THA, reconstructing the center of
reconstructed models defined as failure load/torque were recorded. rotation of the hip nearly in the true position and indications of
Stiffness values of the specimens before and after cyclical loading femoral shortening may not be debated much; however, there is
were measured using forceedisplacement curves for the axial disagreement about the level and type of shortening, whether
 F. Yıldız et al. / Journal of Orthopaedic Science 21 (2016) 640e646 643

Fig. 4. Test machine and experimental setup of the torsional loading tests. Serial imagings from the osteotomy sites were taken by the two camera stereo vision system to measure
the relative movements between the fragments, by using random assay patterns which were attached proximal and distal to the osteotomy (magnified area of the image).

fixation of the osteotomy is necessary or not, and characteristics of In a recent biomechanical study, four different subtrochanteric
the femoral stems. In addition to shorter surgery time, acceptable shortening techniques in THA were compared using cylindrical
rotational stability and nonunion rates, femoral shortening and (Securfit Plus) and conical prosthesis with ridges (Wagner Cone)
rotational correction may be much easier with oblique osteotomy [16]. Under axial loading the mean stiffness value (N/mm) of the
than in step-cut osteotomy [14]. On the other hand, some authors oblique osteotomy was found to be higher than the step-cut
claim that rotational strength of the step-cut osteotomy is superior osteotomy, for cylindrical (880.6 vs. 769.1, respectively) and
to the other types, especially when used with cylindrical femoral conical (1081.0 vs. 958.3, respectively) stems. Rotational stiffness
stems [10,15,21]. Stability of oblique and step-cut shortening values of the oblique osteotomy were also found to be higher than
osteotomies, and the effect of conical or cylindrical femoral stems the step-cut (57.4 vs. 54.4 for cylindrical and 70.9 vs. 59.7 for conical
to the fixation rigidity were compared from the biomechanical stems, respectively). However, they concluded no significant dif-
point of view. We found that none of the osteotomy or prosthesis ferences. Similarly, in our study, the difference in the mean stiffness
types were superior to another although some individual statisti- values of the groups before cyclical loadings was not significant for
cally significant data, which were thought to have no impact on the both axial and rotational loadings. However, after cyclical loading,
clinical outcome, was found in some groups. which was not tested in the previous study, under rotational forces,

Table 1
Findings of axial loading tests. Final displacements at the osteotomy sites were measured after 10,000 cycles.

Failure load Initial stiffness Stiffness after cyclical Displacement Displacement Displacement
(N) (N/mm) load (N/mm) X (mm) Y (mm) Z (mm)

Group I Min. 1980.88 326.92 345.25 0.30 0.36 0.58

Max. 2554.74 596.73 675.73 1.61 0.90 3.82
Mean 2225.77 491.11 517.35 0.80 0.64 1.22
SD. 203.79 76.31 93.82 0.41 0.20 0.99
Group II Min. 1841.90 312.77 379.41 0.20 0.00 0.11
Max. 2329.70 558.86 598.70 1.33 0.26 0.67
Mean 2109.12 441.72 465.32 0.93 0.12 0.39
SD. 139.49 71.82 65.31 0.36 0.08 0.20
Group III Min. 1695.11 357.42 393.23 0.50 0.33 0.11
Max. 2123.47 523.19 548.90 2.53 1.36 4.10
Mean 1866.86 414.44 431.86 1.01 0.89 0.79
SD. 134.14 46.08 47.20 0.60 0.33 1.26
Group IV Min. 1625.47 316.10 331.64 0.68 0.00 0.11
Max. 2085.13 597.27 620.18 1.64 0.47 0.40
Mean 1818.43 404.78 421.60 1.16 0.13 0.22
SD. 158.74 85.76 89.56 0.35 0.13 0.10
p Value <0.001 0.130 0.106 0.136 <0.001 0.106
Post-hoc test (I and II) vs. NS NS NS (I and III) vs. (II NS
p < 0.05 (III and IV) and IV)/I vs. III
644 F. Yıldız et al. / Journal of Orthopaedic Science 21 (2016) 640e646

Table 2
Results of the torsion tests. Final displacements at the osteotomy sites were measured after 10,000 cycles.

Failure torque Initial stiffness Stiffness after cyclical Displacement Displacement Displacement
(Nmm) (Nmm/degree) load (Nmm/degree) X (mm) Y (mm) Z (mm)

Group I Min. 15,066.16 804.70 1193.19 0.02 0.33 1.50

Max. 22,306.00 1600.36 1677.55 0.75 1.21 5.95
Mean 18,701.08 1260.47 1428.68 0.38 0.67 3.14
SD. 2978.89 256.45 168.30 0.30 0.38 1.62
Group II Min. 14,059.06 550.31 568.55 0.00 0.06 0.02
Max. 14,684.21 1333.11 1374.59 2.80 1.39 1.51
Mean 14,360.95 932.27 973.48 1.33 0.62 0.66
SD. 255.67 278.38 291.92 0.96 0.48 0.60
Group III Min. 12,239.78 985.97 1400.95 0.08 0.04 0.30
Max. 27,843.47 1457.16 1531.83 4.16 6.36 13.97
Mean 19,156.33 1255.61 1463.95 1.80 2.12 4.30
SD. 5714.01 164.44 53.54 1.68 2.53 5.60
Group IV Min. 13,027.38 994.05 919.10 0.01 0.00 0.10
Max. 22,814.28 1382.63 1498.41 5.85 1.39 2.05
Mean 17,431.31 1160.60 1177.97 1.76 0.57 0.84
SD. 4139.93 143.79 228.45 2.15 0.48 0.68
p-Value 0.53 0.079 0.043 0.317 0.293 0.159
Post-hoc test NS NS I vs. II NS NS NS
p < 0.05

NS: not significant.

it was significantly higher in oblique osteotomy (1428.6 Nmm/de- the other hand, successful results from conical prosthesis were also
gree) than the step-cut osteotomy (973.4 Nmm/degree) when used published [12e15,21]. Muratli et al. found that if the oblique
with a conical prosthesis although this difference was not seen in osteotomies were performed, the mean stiffness values of the
cylindrical prosthesis group. Higher failure loads of the oblique conical prosthesis groups, under both axial (p ¼ 0.028) and
osteotomy groups can be a reason for lower stiffness of the step-cut torsional forces (p ¼ 0.043), were significantly higher than the
osteotomies. cylindrical prosthesis [16] although it was different for the step-cut
Biomechanical features of femoral stem should play an impor- osteotomy groups (p ¼ 0.128 and p ¼ 0.398, respectively). However,
tant role on the fixation rigidity of the subtrochanteric osteotomies. in our study, the stiffness values before and after cyclical loading
For this reason, some authors recommend cylindrical prosthesis, on were higher in conical prosthesis groups but the differences were

Fig. 5. Comparisons of the groups under axial loading: A. Initial stiffness, B. Stiffness after cyclical loading, C. Failure load, D. Relative displacements at the osteotomy sites.
 F. Yıldız et al. / Journal of Orthopaedic Science 21 (2016) 640e646 645

Fig. 6. Comparisons of the groups under rotational forces: A. Initial stiffness, B. Stiffness after cyclical loading, C. Failure torque, D. Relative displacements at the osteotomy sites.

not significant for the comparisons. These findings may the pathoanatomy of HHD, therefore derotation was not performed.
suggest that using a cylindrical prosthesis does not make much For the biomechanical studies, animal or cadaveric bones, synthetic
difference on the fixation rigidity of the reconstruction against bone models and computer programs can be used as test materials
rotational forces, which contradicts the common belief [13,15,22]. because obtaining cadaveric femorae with similar anatomical and
To our knowledge, this is the first study in the literature about biomechanical characteristics is rather difficult [26]. We were un-
subtrochanteric femoral shortening in HHD, investigating stabil- able to assess effect of the biological environment such as muscle
ities of the osteotomy and prosthesis types before and after cyclical loads, healing and remodelation on the outcomes. Two osteotomy
loadings and measuring the amounts and directions of the dis- and prosthesis types were investigated although there are many
placements at the osteotomy sites. Only vertical displacements of other types. Lastly, the effect of cable fixation on the strength of the
the step-cut osteotomy groups under axial forces were found to be osteotomies was not evaluated although the cables were applied to
statistically significant, however this differences were about half a all specimens in the same manner.
millimeter and may not be clinically or practically relevant.
Although cyclic loading in the experimental setup cannot entirely 5. Conclusions
reflect the clinical reality of how weight bearing effects the
behavior of the osteotomies, it can give a perspective to compare Clinical studies corroborate the view that successful results may
the osteotomy and prosthesis types, which were applied the same be obtained with all four reconstruction options [13]. According to
testing protocol. In a study comparing osteosynthesis of intercalary the results of our study and literature, although the data is not
allografts, all step-cut and sigmoid osteotomy models had fractures sufficient to conclude that one osteotomy or prosthesis type is su-
in the edges of the osteotomy in the first cycle during cyclical perior to another, using oblique or step-cut osteotomy with neither
loading but in the transverse osteotomy group, all the models a conical nor a cylindrical prosthesis do not change stability of
completed cyclical loading tests. The step-cut osteotomy models femoral reconstruction.
were broken in lower rotation angles than sigmoid osteotomy and
that, parallel to our findings, maximum torque values did not vary Author contributions statement
by the osteotomy type [23].
One of the important complications of femoral shortening All authors have contributed to the study design, experiments
osteotomies is intraoperative fracture around the osteotomy whose and interpretation of the date, and all authors have read and
prevalence is reported to be around 5e22% [24,25]. We experienced approved the final submitted manuscript.
this problem in two models of the step-cut osteotomy groups while
preparing the medulla and inserting the cylindrical prosthesis. One Conflict of interest
of the disadvantages of the step-cut osteotomy, this has been
attributed to stress accumulating in points where vertical and There is no conflict of interest.
horizontal osteotomies meet [23]. This risk is less in oblique
osteotomy as there will be no stress accumulation. Design of the Acknowledgment
cylindrical prostheses may be a potential source of fracture risk.
This study had a number of limitations. Synthetic femur models This study was financially supported by Turkish Society of Or-
with normal anatomy were used instead of real human femurs with thopaedics and Traumatology.
646 F. Yıldız et al. / Journal of Orthopaedic Science 21 (2016) 640e646

References [14] Kilicoglu OI, Turker M, Akgul T, Yazicioglu O. Cementless total hip arthroplasty
with modified oblique femoral shortening osteotomy in Crowe type IV
congenital hip dislocation. J Arthroplasty 2013 Jan;28(1):117e25.
[1] Masonis JL, Patel JV, Miu A, Bourne RB, McCalden R, Macdonald SJ,
[15] Takao M, Ohzono K, Nishii T, Miki H, Nakamura N, Sugano N. Cementless modular
Rorabeck CH. Subtrochanteric shortening and derotational osteotomy in pri-
total hip arthroplasty with subtrochanteric shortening osteotomy for hips with
mary total hip arthroplasty for patients with severe hip dysplasia: 5-year
developmental dysplasia. J Bone Jt Surg Am Vol 2011 Mar 16;93(6):548e55.
follow-up. J Arthroplasty 2003 Apr;18(3 Suppl. 1):68e73.
[16] Muratli KS, Karatosun V, Uzun B, Celik S. Subtrochanteric shortening in total
[2] Kelley SS. High hip center in revision arthroplasty. J Arthroplasty 1994
hip arthroplasty: biomechanical comparison of four techniques. J Arthroplasty
Oct;9(5):503e10.
2014 Apr;29(4):836e42.
[3] Pagnano W, Hanssen AD, Lewallen DG, Shaughnessy WJ. The effect of superior
[17] Schwab JH, Camacho J, Kaufman K, Chen Q, Berry DJ, Trousdale RT. Optimal
placement of the acetabular component on the rate of loosening after total hip
fixation for the extended trochanteric osteotomy: a pilot study comparing 3
arthroplasty. J Bone Jt Surg Am Vol 1996 Jul;78(7):1004e14.
cables vs 2 cables. J Arthroplasty 2008 Jun;23(4):534e8.
[4] Erdemli B, Yilmaz C, Atalar H, Guzel B, Cetin I. Total hip arthroplasty in
[18] Sariyilmaz K, Dikici F, Dikmen G, Bozdag E, Sunbuloglu E, Bekler B,
developmental high dislocation of the hip. J Arthroplasty 2005 Dec;20(8):
Yazicioglu O. The effect of strut allograft and its position on Vancouver type B1
1021e8.
periprosthetic femoral fractures: a biomechanical study. J Arthroplasty 2014
[5] Dunn HK, Hess WE. Total hip reconstruction in chronically dislocated hips.
Jul;29(7):1485e90.
J Bone Jt Surg Am Vol 1976 Sep;58(6):838e45.
[19] Tuncay I, Yildiz F, Bilsel K, Uzer G, Elmadag M, Erden T, Bozdag E. Biome-
[6] Hess WE, Umber JS. Total hip arthroplasty in chronically dislocated hips.
chanical comparison of 2 different femoral stems in the shortening osteotomy
Follow-up study on the protrusio socket technique. J Bone Jt Surg Am Vol 1978
of the high-riding hip. J Arthroplasty 2015 Dec 17.
Oct;60(7):948e54.
[20] Dennis MG, Simon JA, Kummer FJ, Koval KJ, Di Cesare PE. Fixation of peri-
[7] Crowe JF, Mani VJ, Ranawat CS. Total hip replacement in congenital
prosthetic femoral shaft fractures: a biomechanical comparison of two tech-
dislocation and dysplasia of the hip. J Bone Jt Surg Am Vol 1979 Jan;61(1):
niques. J Orthop Trauma 2001 MareApr;15(3):177e80.
15e23.
[21] Sener N, Tozun IR, Asik M. Femoral shortening and cementless arthroplasty in
[8] Togrul E, Ozkan C, Kalaci A, Gulsen M. A new technique of subtrochanteric
high congenital dislocation of the hip. J Arthroplasty 2002 Jan;17(1):41e8.
shortening in total hip replacement for Crowe type 3 to 4 dysplasia of the hip.
[22] Onodera S, Majima T, Ito H, Matsuno T, Kishimoto T, Minami A. Cementless
J Arthroplasty 2010 Apr;25(3):465e70.
total hip arthroplasty using the modular S-ROM prosthesis combined with
[9] Huo MH, Zatorski LE, Keggi KJ. Oblique femoral osteotomy in cementless total
corrective proximal femoral osteotomy. J Arthroplasty 2006 Aug;21(5):664e9.
hip arthroplasty. Prospective consecutive series with a 3-year minimum
[23] Cascio BM, Thomas KA, Wilson SC. A mechanical comparison and review of
follow-up period. J Arthroplasty 1995 Jun;10(3):319e27.
transverse, step-cut, and sigmoid osteotomies. Clin Orthop Relat Res 2003
[10] Paavilainen T, Hoikka V, Solonen KA. Cementless total replacement for
Jun;411:296e304.
severely dysplastic or dislocated hips. J Bone Jt Surg Br Vol 1990 Mar;72(2):
[24] Krych AJ, Howard JL, Trousdale RT, Cabanela ME, Berry DJ. Total hip arthro-
205e11.
plasty with shortening subtrochanteric osteotomy in Crowe type-IV devel-
[11] Hotokebuchi T, Sonohata M, Shigematsu M, Mawatari M. A new device for a
opmental dysplasia: surgical technique. J Bone Jt Surg Am Vol 2010
V-shaped subtrochanteric osteotomy combined with total hip arthroplasty.
Sep;92(Suppl. 1 Pt 2):176e87.
J Arthroplasty 2006 Jan;21(1):135e7.
[25] Eskelinen A, Helenius I, Remes V, Ylinen P, Tallroth K, Paavilainen T.
[12] Karachalios T, Hartofilakidis G. Congenital hip disease in adults: terminology,
Cementless total hip arthroplasty in patients with high congenital hip dislo-
classification, pre-operative planning and management. J Bone Jt Surg Br Vol
cation. J Bone Jt Surg Am Vol 2006 Jan;88(1):80e91.
2010 Jul;92(7):914e21.
[26] Khanna G, Bourgeault CA, Kyle RF. Biomechanical comparison of extended
[13] Reikeras O, Haaland JE, Lereim P. Femoral shortening in total hip arthroplasty
trochanteric osteotomy and slot osteotomy for femoral component revision in
for high developmental dysplasia of the hip. Clin Orthop Relat Res 2010
total hip arthroplasty. Clin Biomech 2007 Jun;22(5):599e602.
Jul;468(7):1949e55.