- Email: [email protected]
Comparison of Total Hip Arthroplasty With and Without Femoral Shortening Osteotomy for Unilateral Mild to Moderate High Hip Dislocation
Accepted Manuscript Comparison of total hip arthroplasty with and without femoral shortening osteotomy for unilateral mild to moderate high hip dislocation Huiwu Li, MD, Jiawei Xu, MD, Xinhua Qu, MD, Yuanqing Mao, MD, Kerong Dai, MD, Zhenan Zhu, MD PII:
S0883-5403(16)30523-X
DOI:
10.1016/j.arth.2016.08.021
Reference:
YARTH 55372
To appear in:
The Journal of Arthroplasty
Received Date: 29 May 2016 Revised Date:
30 July 2016
Accepted Date: 18 August 2016
Please cite this article as: Li H, Xu J, Qu X, Mao Y, Dai K, Zhu Z, Comparison of total hip arthroplasty with and without femoral shortening osteotomy for unilateral mild to moderate high hip dislocation, The Journal of Arthroplasty (2016), doi: 10.1016/j.arth.2016.08.021. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT Comparison of total hip arthroplasty with and without femoral shortening osteotomy for unilateral mild to moderate high hip dislocation Running title: THA for unilateral high hip dislocation
RI PT
Huiwu Li, MD 1; Jiawei Xu 1, MD; Xinhua Qu, MD 1;Yuanqing Mao MD 1; Kerong Dai, MD 1; Zhenan Zhu, MD 1
1. Department of Orthopaedics, Shanghai No. 9th People’s Hospital, Shanghai Jiao
SC
Tong University School of Medicine, Shanghai 200011, China [email protected]
Jiawei Xu MD
[email protected]
Xinhua Qu, MD
[email protected]
Yuanqing Mao, MD
[email protected]
Kerong Dai, MD
[email protected]
TE D
Zhenan Zhu, MD
M AN U
Huiwu Li, MD
[email protected]
EP
We confirm that this manuscript has not been published elsewhere and is not under
AC C
consideration by another journal. The authors have no conflicts of interest to declare. All authors have approved the manuscript and agree with submission to JOA. This study was performed in Shanghai No. 9th People’s Hospital. Please address all correspondence to: Zhenan Zhu, MD. Department of Orthopaedics, Shanghai No. 9th People’s Hospital, Shanghai Jiao Tong University School of Medicine, Shanghai 200011,
China.
Phone:
[email protected].
86-21-63139920,
Fax:
86-21-63139920,
Email:
ACCEPTED MANUSCRIPT Comparison of total hip arthroplasty with and without femoral shortening osteotomy
2
for unilateral mild to moderate high hip dislocation
3
Abstract
4
Background: This study is to compare the outcome between THA with and without femoral
5
shortening osteotomy for unilateral mild to moderate high hip dislocation in DDH patients.
6
Methods: The data on 42 hips in 42 patients who had undergone THA for unilateral mild to
7
moderate high hip dislocation were retrospectively reviewed after being prospectively
8
collected. In 22 patients, hips were reduced by soft tissue release and direct leverage using an
9
elevator, without the osteotomy. The remaining 20 patients were treated with a
10
subtrochanteric transverse shortening osteotomy. The mean follow-up of patients was 5 years
11
(SD 1.0) for the non-osteotomy group and 6.2 years (SD 1.6) for the osteotomy group.
12
Results: The Harris Hip Score (HHS) significantly improved in both groups. In the
13
non-osteotomy group, we observed a lower leg-length discrepancy compared with the
14
osteotomy group (0.4 cm and 2.2 cm, respectively). Four patients (18.2%) in the
15
non-osteotomy group and 15 patients (75%) in the osteotomy group developed a limp
16
(P<0.0001). 3 patients (13%) developed femoral nerve palsy in the non-osteotomy group but
17
they all recovered completely within 6 months after the surgery. Nineteen patients in the
18
non-osteotomy group showed knee valgus deformity immediately after the surgery, but only 4
19
cases in the osteotomy group.
20
Conclusion: Compared with THA with femoral shortening osteotomy, THA without the
21
osteotomy was associated with a lower number of patients who developed a limp at the end of
22
follow-up; however, the rehabilitation was slower and more difficult, and a larger number of
23
patients showed reversible nerve palsy and knee valgus deformity.
24
Key word: high hip dislocation; developmental dysplasia of the hip; total hip arthroplasty;
25
femoral shortening osteotomy; comparison study
AC C
EP
TE D
M AN U
SC
RI PT
1
26 27
Introduction
28
Total hip arthroplasty (THA) is an effective treatment for developmental dysplasia of the
29
hip (DDH) with high hip dislocation in adults. The durability of arthroplasty in these patients
30
is better with the restoration of an anatomic hip center [1]. One option for restoring the 1
ACCEPTED MANUSCRIPT anatomic hip center is a femoral shortening osteotomy, which facilitates hip reduction and
32
protects the nerve and artery [2]. However, this approach is considered to be complex and
33
time-consuming, and can lead to complications such as shortened limbs, nonunion at the
34
osteotomy site, loosened prosthesis, and hip dislocation [1-3]. For the patients with a mild to
35
moderately high dislocation of the hip, a number of surgeons have reported THA using
36
methods other than femoral shortening osteotomy, such as extensive soft tissue release,
37
intravenous injection of rocuronium, elevator leverage, or distraction nail [4-7]. These
38
techniques show advantages including the lengthening of the limb and reconstruction of the
39
true socket as well as the restoration of the anatomical course of the abductor and balance of
40
its muscular strength. However, complications such as nerve palsies, greater trochanter
41
avulsion fracture, and the feeling of the lengthening of the affected limb have been reported
42
[8-11].
M AN U
SC
RI PT
31
To the best of our knowledge, there are no studies compared the results of THA with and
44
without femoral shortening osteotomy for unilateral mild to moderately high hip dislocation
45
in patients with DDH. The aim of this study was to compare the results of the 2 techniques
46
with respect to hip function, leg-length discrepancy and complications in 2 comparable
47
cohorts of patients.
48
Patients and methods
50
Patients
EP
49
TE D
43
During the period from January 2005 to December 2009, more than 800 patients with
52
DDH underwent a THA in our hospital. A total of 66 hips were considered as having high
53
dislocation corresponding to type IV of the Crowe classification [12]. We classified the
54
Crowe IV dislocation into 2 subtypes, the mild to moderately high dislocation when the
55
migration of the femoral head was more than 1/5 of the pelvic height but less than 2/5 of the
56
pelvic height, and the extremely high dislocation when the migration was more than 2/5 of the
57
pelvic height (Fig. 1). In this study, the patients with an extremely high dislocation or with a
58
severe deformity of proximal femur were excluded from this study because the osteotomy
59
would have to be performed. Therefore, the patients with a mild to moderately high
60
dislocation were included in this study. We also excluded patients with bilateral high
AC C
51
2
ACCEPTED MANUSCRIPT dislocation of the hip. The surgeries were performed by the same groups of surgeons. From
62
January 2005 to September 2007, the femoral shortening osteotomy was performed for all
63
patients with mild to moderately high dislocation of the hip. Because of complex and
64
shortened limbs caused by shortening osteotomy, after September 2007, the patients were
65
reduced by soft tissue release and elevator direct leverage without femoral shortening
66
osteotomy. Forty-six patients were included in the study, of which 4 (4 hips) were lost to
67
follow-up, one hip with osteotomy, three hips without osteotomy. Finally, we retrospectively
68
reviewed prospectively collected data on 42 hips in 42 patients who underwent a cementless
69
THA for DDH with unilateral mild to moderately high hip dislocation. In 22 patients, the hip
70
was reduced by soft tissue release and direct leverage using an elevator without femoral
71
shortening osteotomy (from October 2007 to December 2009), while 20 patients were treated
72
with the osteotomy (from January 2005 to September 2007). The mean follow-up of patients
73
was 5.0 years (SD 1.0; 4 to 7 years) for the non-osteotomy group and 6.2 years (SD 1.6; 4 to 9
74
years) for the osteotomy group. Data on the sex, age, weight, height, body mass index, and
75
duration of follow-up of patients in the 2 groups are presented in Table 1.
M AN U
SC
RI PT
61
In the non-osteotomy group, the limb length discrepancies were 4.6 cm (SD 0.7, 3 to 6
77
cm). Three patients had osteoarthritis of the contralateral hip with dislocation classified as
78
Crowe type I, and 3 patients as Crowe type II. These patients had already undergone a THA
79
before this surgery. Four patients (18%) underwent pelvic osteotomy for the treatment of the
80
hip in childhood. Spinal comorbidities associated with DDH included multiple sclerosis in 5
81
patients (22.7%). In the osteotomy group, the limb length discrepancies were 4.8 cm (SD 1.1,
82
2.5 to 6.5 cm). Three patients had osteoarthritis of the contralateral hip with dislocation
83
classified as Crowe grade I, and 1 patient as Crowe grade III. These patients had already
84
undergone a THA before this surgery. Two patients (10%) underwent pelvic osteotomy for
85
the treatment of the hip in childhood. Spinal comorbidities associated with DDH included
86
multiple sclerosis in 3 patients (15%).
87
Materials
AC C
EP
TE D
76
88
In the non-osteotomy group, the S-ROM stem and Duraloc acetabular cup with a ceramic
89
on ceramic bearing were used in 9 hips and with a polyethylene on cobalt-chromium bearing
90
in 2 hips (Depuy, Warsaw, Indiana); the Omnifit hydroxyapatite-coated stem and SecurFit 3
ACCEPTED MANUSCRIPT acetabular cup with a ceramic on ceramic bearing were used in 9 hips and with a polyethylene
92
on cobalt-chromium bearing in 2 hips (Stryker, Mahwah, New Jersey). Of them, 2 hips were
93
treated with a 22-mm cobalt-chromium femoral head, 2 hips with a 28-mm cobalt-chromium
94
femoral head, and 18 hips with a 28-mm ceramic femoral head. In the osteotomy group, the
95
S-ROM stem and Duraloc acetabular cup with a ceramic on ceramic bearing were used in 14
96
hips and with a polyethylene on cobalt-chromium in 6 hips (DePuy). Of them, 1 hip was
97
treated with a 22-mm cobalt-chromium femoral head, 5 hips with a 28-mm cobalt-chromium
98
femoral head, and 14 hips with a 28-mm ceramic femoral head.
99
Surgical technique
SC
RI PT
91
All of the operations were performed using a posterolateral approach. The procedure of
101
acetabular component insertion in THA with and without femoral shortening osteotomy was
102
the same. All osteophytes and hypertrophied soft tissue around the true acetabulum were
103
removed. The acetabular cups were inserted with a press-fit fixation.
M AN U
100
In the non-osteotomy group, the procedure has been previously described in detail [11].
105
We do not favor extensive muscle release to reduce the hip and always take great care to
106
retain the periarticular muscles. As reduction, the limb was held in adduction and internal
107
rotation as distal traction was applied to the femur by an assistant. If this was insufficient to
108
reduce the hip, direct leverage was applied to the greater trochanter using a Hohmann
109
retractor, which obtained purchase on the bone below the acetabulum. When the modular
110
femoral head reached the level of the acetabular component, the reduction was achieved by
111
external rotation of the leg. During the procedure, the knees were always held in flexion to
112
relax the sciatic nerve and reduce tension in soft tissue. This position was maintained for
113
several days after the surgery to avoid damage to the nerve.
EP
AC C
114
TE D
104
In the osteotomy group, at the beginning of the femoral osteotomy, the subtrochanteric
115
area of the femur is approached by elevating or splitting a short section of the vastus lateralis.
116
Next, a transverse proximal femoral osteotomy is performed in accordance with preoperative
117
templating. The femoral trial was then replaced into the proximal fragment, the head reduced
118
into the acetabulum, and traction applied to the distal femur by pulling on the leg. The bony
119
overlap at the osteotomy level was removed from the distal femur and the amount of resection
120
was minimized as much as possible. We usually place a cerclage cable or wire 4
ACCEPTED MANUSCRIPT 121
prophylactically before implant impaction. The final components were inserted with the stem
122
transfixing the osteotomy. Once fully seated, the osteotomy should be in apposition and
123
rigidly fixed. The rehabilitation program varied according to soft tissue tension. In general, patients
125
were allowed partial weight-bearing from 3 days to 1 week after the surgery and then full
126
weight-bearing at 1 to 6 weeks. If the soft tissue was very tight after leg lengthening, the hip
127
and knee flexion position was maintained for several days postoperatively and then extended
128
gradually over the course of 1 to 2 weeks. After regaining full extension, walking exercises
129
with crutches and full weight-bearing were initiated.
130
Evaluation method
SC
RI PT
124
Clinical and radiographic data were obtained for all patients before the surgery and at
132
follow-up examinations. The clinical evaluation was done based on the Harris Hip Score
133
(HHS), while patients’ satisfaction was subjectively described as excellent, good, moderate,
134
or unsatisfactory. The presence of a Trendelenburg sign was recorded. Length discrepancy
135
between the bilateral legs was assessed after the surgery by measuring the distance from the
136
anteriorsuperior iliac spine to the medial malleolus. The patient-perceived leg-length
137
discrepancy was also assessed by placing blocks under the shorter limb until the patient felt
138
comfortable. The function of the sciatic and femoral nerves was carefully examined but did
139
not include routine electrophysiological testing.
EP
TE D
M AN U
131
A routine radiographic examination included anteroposterior and lateral hip views. The
141
inclination and horizontal and vertical distance of the postoperative hip center were
142
determined using the radiographic technique described by Russotti and Harris [13]. The
143
translation of the tip of the greater trochanter was determined by comparing the change inthe
144
distance between the tip of the greater trochanter and interteardrop line before and after the
145
surgery. Pelvic obliquity was determined by measuring the iliolumbar angle. On the AP
146
radiographs of the pelvis and lower lumbar spine, 1 line was drawn connecting the apices of
147
both iliac crests and another line along the bottom of the fourth lumbar vertebra. The
148
iliolumbar angle was measured at the convergence of these 2 lines (Fig. 1) [14]. The stability
149
of the acetabular components was assessed radiographically using the method of DeLee and
150
Charnley [15] and that of the femoral components using the method of Gruen et al. [16].
AC C
140
5
ACCEPTED MANUSCRIPT 151
Serial radiographs were used to evaluate the union (bridging of the bone trabeculae) at the
152
osteotomy site. All samples were assessed by LHW and XJW independently. Agreement
153
between LHW and XJW was good (all Cohen’s unweighted κ> 0.90). Statistical analysis was conducted with SPSS for Windows Version 11.5 (SPSS Inc,
155
Chicago, IL, USA). The Student’s t-test was used to compare the pre- and postoperative HHS,
156
operation time, amount of bleeding, and radiographic parameters between the non-osteotomy
157
and osteotomy groups. The Mann-Whitney U test was used to assess patients’ satisfaction and
158
the Mantel-Haenszel chi-squared test to assess the patient-perceived leg-length discrepancy,
159
Trendelenburg sign, limp, knee valgus and external rotation deformity. Differences were
160
considered statistically significant at a P value of less than 0.05.
SC
RI PT
154
162
Results
163
Clinical results
164 165
M AN U
161
The clinical results are shown in Table 2. The non-osteotomy group showed a significantly shorter operation time (P=0.01) and a smaller amount of bleeding (P=0.03). A significant improvement in the HHS after the surgery was observed in both groups. In
167
the non-osteotomy group, the mean HHS improved from 48.4 points (SD 10, 28 to 74 points)
168
to 90.8 points (SD 4.7, 80 to 100) at the final follow-up examination (P<0.001). In the
169
osteotomy group, it improved from 47.0 points (SD 12.0, 24 to 70 points) to 87 points (SD
170
5.8, 70 to 96 points) at the final follow-up examination (P<0.001). The score improved later
171
during follow-up in the non-osteotomy group compared with the osteotomy group (Fig. 2).
EP
AC C
172
TE D
166
In the non-osteotomy group, patients’ satisfaction at the final follow-up examination was
173
reported as excellent for 19 patients, good for 2 patients, moderate for 1 patient, and
174
unsatisfactory for 0 patients. In the osteotomy group, the corresponding numbers were 11, 5,
175
3, and 1 patient, respectively (P=0.15).
176
In the non-osteotomy group, the affected limbs were lengthened 4.1 cm (SD 0.9, 2.0 to
177
5.0 cm). The average leg-length discrepancy at the final follow-up examination was 0.4 cm
178
(SD 0.7, -1.2 to 1.5 cm). In the osteotomy group, the affected limbs were lengthened 2.7 cm
179
(SD 1.2, 1.0 to 4.8 cm). The average leg-length discrepancy was 2.2 cm (SD 1.2, 0 to 5.0 cm,
180
Fig. 3). Six patients reported a feeling of leg-length discrepancy in the non-osteotomy group 6
ACCEPTED MANUSCRIPT 181
(the lengthening of the affected limb was felt by 5 patients, and the shortening by 1 patient).
182
In the osteotomy group, 17 patients reported a feeling of leg-length discrepancy (shortening
183
of the affected limb). There was no revision or marked loosening during the follow-up period in either group.
185
All patients were able to put on shoes and socks by themselves as well as walk and climb
186
stairs independently. In addition, all patients had a positive Trendelenburg sign before the
187
surgery. However, at the final follow-up examination, only 2 patients in the non-osteotomy
188
and 3 patients in the osteotomy group showed a positive Trendelenburg sign. Without a
189
heightening shoe, 1 patient (5%) in the osteotomy group showed a severe limp, 4 patients
190
(20%) a moderate limp, 10 (50%) a slight limp,
191
group, only 4 patients (18%) showed a slight limp (P<0.0001).
SC
RI PT
184
M AN U
and 5 (25%) no limp. In the non-osteotomy
Immediately after the surgery, 19 patients in the non-osteotomy group showed knee
193
valgus deformity, 6 of these patients showed not only valgus but external rotation deformity,
194
and 1 of these patients had patella dislocation when the knee flexed more than 40° (Fig. 4).
195
By the end of the follow-up, the knee valgus deformity resolved in 13 patients, and 6 patients
196
still showed slight valgus deformity. The patella dislocation resolved after 3 months with the
197
correction of the knee valgus. In the osteotomy group, the knee valgus deformity occurred in
198
4 cases and resolved in all patients by the end of the follow-up.
199
Radiographic evaluation
EP
TE D
192
All cups in both groups were built in the anatomical acetabulum. The mean inclination of
201
the cup, vertical distance of the hip center, and horizontal distance in the 2 groups are shown
202
in Table 3. There was no significant difference between the groups in the height of the tip of
203
the greater trochanter before the surgery (P=0.39) and in the shift of the greater trochanter
204
after the surgery (P=0.41).
205
AC C
200
Postoperative pelvic obliquity occurred in all patients. The immediate postoperative
206
obliquity was more apparent in the non-osteotomy group than in the osteotomy group. In the
207
non-osteotomy group, the difference of the iliolumbar angle before and after the surgery was
208
3.0° (SD 2.5, -1° to 7°). The pelvic obliquity corrected with time (Fig. 5). At the last
209
follow-up examination, the iliolumbar angle was 0.8° (SD 3.1, -4° to 7°). In the osteotomy
210
group, the difference was 2.5° (SD 2.0, -1.6° to 6°). At the last follow-up examination, the 7
ACCEPTED MANUSCRIPT 211
iliolumbar angle was -1.4° (SD 2.8, -7° to 3.2°) and had almost returned to the preoperative
212
level (Fig. 6). Solid union at the osteotomy site was achieved at a mean period of 4.2 months (SD 1.5, 3
214
to 9 months). There was no loosening or progressive radiolucency or osteolysis adjacent to
215
the acetabulum and femoral component observed at the last follow-up examination in any
216
case.
217
Complications
RI PT
213
In the osteotomy group, there was 1 intraoperative fracture of the distal femoral segment
219
and 1 delayed union (9 months) at the osteotomy site. In the non-osteotomy group, 3 patients
220
(14%) had femoral nerve palsies with numbness of the medial aspect of the lower limb and
221
weakness of knee extension. The limbs were lengthened 4.0, 4.1, and 4.6 cm respectively. All
222
of these patients were treated conservatively and recovered completely at 3 weeks, 2 months,
223
and 6 months after the surgery, respectively. No sciatic nerve injury, dislocation, or infection
224
was observed in any of the groups (Table 4).
M AN U
SC
218
225
Discussion
227
This is the first study that directly compared the results of THA with and without femoral
228
shortening osteotomy for unilateral mild to moderately high dislocations in patients with
229
DDH. The comparison focused on several aspects such as surgical techniques, HHS,
230
complications, patients’ satisfaction, and leg-length discrepancy.
231
Surgical techniques
EP
AC C
232
TE D
226
Various osteotomy techniques have been shown to facilitate hip reduction [17, 18].
233
However, a number of surgeons reported that they are complex and time-consuming. In the
234
present study, the mean operation time was approximately 16 minutes longer and bleeding
235
was more excessive by about 94 mL in the group undergoing subtrochanteric transverse
236
shortening osteotomy than in the non-osteotomy group. Nonetheless, hip reduction in THA
237
without femoral shortening osteotomy requires considerable skill and experience. Sener et al.
238
[19] proposed that for severe dislocation of the hip joint, the entire excision of the joint
239
capsule and scar tissue is a priority, and then the loosening of the superior parts of the gluteus
240
maximus, adductor, rectus femoris, and sartorius muscles is considered. Kawai et al.[8] 8
ACCEPTED MANUSCRIPT proposed an additional femoral neck cut performed until femoral head reduction was possible,
242
while Yan et al. [5] suggested an intravenous injection of rocuronium at 0.9 mg/kg 1 minute
243
before reduction, combined with a strong continuous traction of the affected limb. In the
244
present study, a direct leverage using an elevator, which obtained purchase on the bone below
245
the acetabulum, was applied to the greater trochanter. No matter what the techniques, we
246
believe that a long learning curve is necessary.
247
HHS
RI PT
241
The improvement of the HHS in the non-osteotomy group was relatively slower than that
249
in the osteotomy group. The excessive tightness of soft tissue caused by the lengthening of
250
the limb hampered the rehabilitation process. However, with the stretching of soft tissue, the
251
HHS in the non-osteotomy was comparable and perhaps even better than that in the
252
osteotomy group at the final follow-up examination.
253
Complications
M AN U
SC
248
To our knowledge, there is no documented maximum value for an extremity to be safely
255
lengthened without aneurological complication, and there is also no definition of mild,
256
moderate or extremely high dislocation. In this study, we classified the dislocation according
257
to our experience and regarded migration of the femoral head of more than 2/5 of the pelvic
258
height as an extremely high dislocation and the standard of whether the shortening osteotomy
259
was performed. Since 2007, all of the patients whose migration of the femoral head was less
260
than 2/5 of the pelvic height were treated successfully without femoral shortening osteotomy,
261
no matter how many centimeters the migration was. Although some authors have reported
262
that limb lengthening should be limited to 4 cm [20]. Kerboull et al. [21] published favorable
263
results in a consecutive series of 118 THAs performed for Crowe type IV hip dislocation. In
264
that series, only 2 hips required a shortening osteotomy and only 1 transient peroneal nerve
265
palsy occurred despite the fact that 30 limbs were lengthened by more than 4 cm. In fact, they
266
even reported that limb lengthening of up to 7 cm was possible. In this study, in the
267
non-osteotomy group, limb lengthening of more than 4 cm was performed in 10 hips, none of
268
which developed sciatic nerve palsy. However, there were 3 patients who presented with
269
femoral nerve palsy. We speculated that the femoral nerve is more vulnerable to stretch than
270
the sciatic nerve, which is associated with the anatomic characteristics of the femoral nerve,
AC C
EP
TE D
254
9
ACCEPTED MANUSCRIPT 271
namely, it is shorter and diverges into the muscles earlier than the sciatic nerve. Nevertheless,
272
all patients fully recovered without sequelae in 6 months. Another phenomenon observed mainly in the non-osteotomy group was knee valgus and
274
external rotation deformity. The excessively tight soft tissue, especially the iliotibial tract,
275
contributed to the deformity. In addition, 1 patella dislocation after the surgery occurred in the
276
non-osteotomy group. Our previous study showed that DDH is associated with morphological
277
changes in the knee joint [22]. Compared with the knees in patients with normal hips, the
278
femoral condyles in patients with dislocated hips were smaller and the trochlear groove was
279
shallower. The dysplasia of the knee combined with the valgus and external rotation caused
280
by a tightened iliotibial tract led to patella dislocation in this study. Although most of the knee
281
valgus and external rotation deformity resolved with relaxation of the iliotibial tract, the
282
change of the patella track after THA in patients with high DDH still deserves our attention.
283
Patients’ satisfaction and leg-length discrepancy
M AN U
SC
RI PT
273
Compared with patients in the osteotomy group, patients in the non-osteotomy group
285
showed relatively higher satisfaction even there is no significant difference. The main
286
complaint that affected patients’ satisfaction at the final follow-up examination in the
287
osteotomy group was a limp. It has been reported that most osteotomies of the femur leave the
288
leg significantly shortened [7]. Park et al. [17] reported 23 cases of Crowe type IV DDH
289
patients managed with transverse femoral shortening osteotomy. The average leg-length
290
discrepancy after the surgery in patients with unilateral dysplasia was 42 mm. The
291
postoperative limp was graded as none (17%), mild (26%), moderate (39%), or severe (18%),
292
which was in line with our results.
EP
AC C
293
TE D
284
Unlike in the osteotomy group, the most common complaint in the non-osteotomy group
294
in the first several months after surgeries was leg lengthening, even though the true average
295
leg-length discrepancy at the final follow-up examination was only 4 mm. This leg
296
lengthening was due to pelvic obliquity caused by excessively tight soft tissues. In this study,
297
an established tautness of soft tissues and muscles surrounding the THA joint caused an
298
inclination of the pelvis toward the operated side of the body. We used the change of the
299
iliolumbar angle preoperatively and postoperatively to reflect pelvic obliquity caused by
300
muscle tautness. Postoperative pelvic obliquity occurred in all patients; it led to relative leg 10
ACCEPTED MANUSCRIPT length inequality, as a result of which patients felt lengthening of the affected leg. As tension
302
in soft tissue gradually decreased, pelvic obliquity was corrected and the feeling of leg
303
lengthening slowly disappeared. This process was longer and more apparent in the
304
non-osteotomy group than in the osteotomy group. When pelvic obliquity was corrected, the
305
feeling of leg lengthening disappeared or alleviated in the non-osteotomy group, but the
306
feeling of leg shortening became gradually more apparent in the osteotomy group.
RI PT
301
Our study has several limitations. First, it was a retrospective evaluation of prospectively
308
followed patients. Second, the study group was relatively small, even though the study was
309
one of the largest series of patients with this uncommon deformity. Third, the implants in this
310
series were not identical for all arthroplasties. Finally, the follow-up period was short- to
311
mid-term. All of the above reasons may have affected the objectiveness of this study.
312
However, the fact that all operations were performed by the same group of surgeons may be
313
considered a strength of the study.
M AN U
SC
307
In conclusion, we determined that the intermediate-term clinical and radiographic results
315
of THA with or without femoral shortening osteotomy for DDH patients with a high
316
dislocation were satisfactory. Compared with THA with femoral shortening osteotomy, THA
317
without the osteotomy was associated with a shorter operative time and less bleeding, a lower
318
number of patients with a limp. However, the latter procedure also required higher surgical
319
skills and was associated with slower and more difficult rehabilitation as well as more cases
320
of reversible nerve palsy and knee valgus deformity; the long-term effects of valgus deformity
321
on the knee remain to be observed.
AC C
EP
TE D
314
This study was supported by Shanghai Natural Science Foundation (Grant No.
322 323
15ZR1424800) and Scientific Research Foundation for the Returned Overseas Chinese
324
Scholars, Ministry of Education of China (Grant No. 201550002).
325
References
326
1.
Pagnano MW, Hanssen AD, Lewallen DG, Shaughnessy WJ.The effect of superior
327
placement of the acetabular component on the rate of loosening after total hip
328
arthroplasty. J Bone Joint Surg Am. 1996;78-A:1004-1014.
329 330
2.
Zagra L, Bianchi L, Mondini A, Ceroni RG. Oblique femoral shortening osteotomy in total hip arthroplasty for high dislocation in patients with hip dysplasia. Int Orthop. 2015; 11
ACCEPTED MANUSCRIPT 39(9): 1797-802.
331 332
3.
Zhu J, Shen C, Chen X, Cui Y, Peng J, Cai G. Total hip arthroplasty with a non-modular
333
conical stem and transverse subtrochanteric osteotomy in treatment of high dislocated
334
hips. J Arthroplasty. 2015; 30(4): 611-614. 4.
Wu X, Li SH, Lou LM, Cai ZD. The techniques of soft tissue release and true socket
RI PT
335 336
reconstruction in total hip arthroplasty for patients with severe developmental dysplasia
337
of the hip. Int Orthop. 2012;36(9):1795-1801.
338
5.
Yan F, Chen G, Yang L, He R, Gu L, Wang F. A reduction technique of arthroplasty without
340
developmental high dislocation of hip: a case series of 28 hips. J Arthroplasty.
341
2014;29(12):2289-2293. 6.
femoral
shortening
osteotomy
for
the
treatment
of
M AN U
342
subtrochanteric
SC
339
Li H, Mao Y, Oni JK, Dai K, Zhu Z. Total hip replacement for developmental dysplasia
343
of the hip with more than 30% lateral uncoverage of uncemented acetabular components.
344
Bone Joint J. 2013; 95-B(9):1178-1183.
345
7.
Lai KA, Shen WJ, Huang LW,Chen MY.Cementless total hip arthroplasty and limb-length equalization in patients with unilateral Crowe type IV hip dislocation.J Bone
347
Joint Surg Am. 2005;87-A:339-345.
348
8.
TE D
346
Kawai T, Tanaka C, Kanoe H. Total hip arthroplasty for Crowe IV hip without subtrochanteric shortening osteotomy -a long term follow up study. BMC Musculoskelet
350
Disord. 2014;15:72.
352
9.
Hartofilakidis G, Karachalios T. Total hip arthroplasty for congenital hip disease. J Bone
AC C
351
EP
349
Joint Surg Am. 2004;86:242-250.
353
10. Eskelinen A, Helenius I, Remes V, Ylinen P, Tallroth K, Paavilainen T. Cementless total
354
hip arthroplasty in patients with high congenital hip dislocation. J Bone Joint Surg Am.
355
2006;88:80-91.
356
11. Zhao X, Zhu ZA, Xie YZ, Yu B, Yu DG. Total hip replacement for high dislocated hips
357
without femoral shortening osteotomy. J Bone Joint Surg Br. 2011;93(9):1189-1193.
358
12. Crowe JF, Mani VJ, Ranawat CS. Total hip replacement in congenital dislocation and
359 360
dysplasia of the hip. J Bone Joint Surg Am. 1979;61:15-23. 13. Russotti GM, Harris WH. Proximal placement of the acetabular component in total hip 12
ACCEPTED MANUSCRIPT 361
arthroplasty. A long-term follow-up study. J Bone Joint Surg Am. 1991;73:587-592.
362
14. Lee DY, Choi IH, Chung CY, Cho TJ, Lee JC. Fixed pelvic obliquity after poliomyelitis:
364 365
Classification and management. J Bone Joint SurgBr. 1997;79:190-196. 15. DeLee JG, Charnley J.Radiological demarcation of cemented sockets in total hip replacement. Clin Orthop Relat Res. 1976;121:20-32.
RI PT
363
366
16. Gruen TA, McNeice GM, Amstutz HC. “Modes of failure” of cemented stem-type
367
femoral components: a radiographic analysis of loosening.Clin Orthop Relat
368
Res.1979;141:17-27.
17. Park MS, Kim KH, Jeong WC. Transverse subtrochanteric shortening osteotomy in
370
primary total hip arthroplasty for patients with severe hip developmental dysplasia.J
371
Arthroplasty.2007;22:1031-1036.
M AN U
SC
369
372
18. Krych AJ, Howard JL, Trousdale RT, Cabanela ME, Berry DJ. Total hip arthroplasty
373
with shortening subtrochanteric osteotomy in Crowe type-IV developmental dysplasia. J
374
Bone Joint Surg Am. 2009;91:2213-2221.
376
19. Sener N, Tözün IR, Aşik M. Femoral shortening and cementless arthroplasty in high congenital dislocation of the hip. J Arthroplasty. 2002;17:41-48.
TE D
375
20. Eggli S, Hankelmayer S, Muller ME. Nerve palsy after leg lengthening in total
378
replacement arthroplasty for developmental dysplasia of the hip. J Bone Joint Surg Br.
379
1999;843-845.
EP
377
21. Kerboull M, Hamadouche M, Kerboull L. Total hip arthroplasty for Crowe type IV
381
developmental hip dysplasia: a long-term follow-up study. J Arthroplasty. 2001;16(8
382 383 384 385
AC C
380
Suppl 1):170-176.
22. Li H, Qu X, Wang Y, Dai K, Zhu Z. Morphological analysis of the knee joint in patients with hip dysplasia. Knee Surg Sports Traumatol Arthrosc. 2013;21(9):2081-2088.
386 387 388 389 390 13
ACCEPTED MANUSCRIPT Legends Fig. 1. The pelvis was divided into 5 equal parts by line 1 to 4. The mild to moderately high dislocation means the migration of the femoral head is more than 1/5 of the pelvic height but less than 2/5 of the pelvic height, that is, the bottom of the femoral head-neck junction
RI PT
locating the zone between line 2 and line 3. The extremely high dislocation means the migration was more than 2/5 of the pelvic height, that is, the bottom of the femoral head-neck junction locating beyond line 3. The degree of pelvic obliquity was reflected through the iliolumbar angle. Line OB was drawn to connect the apices of both iliac crests, and line OA
measured at the convergence of these 2 lines.
SC
was drawn along the bottom of the fourth lumbar vertebra. The iliolumbar angle AOB was
M AN U
Fig. 2. The improvement of the Harris Hip Score was slower in the non-osteotomy group than in the osteotomy group, but the nonosteotomy group achieved a comparable or better score at the final follow-up examination.
Fig. 3. A patient had undergone THA without femoral shortening osteotomy and had an equal leg length (a). A patient had undergone THA with femoral shortening osteotomy and had one
TE D
leg 4 cm shorter than the contralateral leg (b).
Fig. 4. A 45-year-old woman with Crowe type IV hip dislocation (a). A THA without femoral shortening osteotomy was performed, and the knee showed a valgus and external rotation
EP
deformity after surgery (b). Before THA, the CT scan showed that the patella was in the femoral trochlear groove but the femoral condyle was smaller and the trochlear groove was
AC C
shallow (c). After surgery, the patient had patella dislocation when the knee flexed more than 40° (d).
Fig. 5. A 46-year-old woman with Crowe type IV hip dislocation in one leg and THA in the other leg for Crowe type II hip dislocation (a). A THA without femoral shortening osteotomy was performed for a high hip dislocation. Pelvic obliquity in the operated side after THA (b). The obliquity improved 6 months after the surgery (c). Fig. 6. The immediate postoperative obliquity is more apparent in the non-osteotomy group than in the osteotomy group. In the non-osteotomy group, the difference in the iliolumbar angle preoperatively and postoperatively was 3°. Pelvic obliquity corrected with time. At the final follow-up examination, the iliolumbar angle was 0.8°. In the osteotomy group, the
ACCEPTED MANUSCRIPT difference was 2.5°. At the final follow-up examination, the iliolumbar angle of -1.4° almost
AC C
EP
TE D
M AN U
SC
RI PT
returned to the preoperative level.
ACCEPTED MANUSCRIPT Table 1. Characteristics of patients in the non-osteotomy and osteotomy groups Osteotomy group
No. of patients/ hips
22
20
Sex (M/F)
5/17
6/14
0.60
Average age (y) (SD)
51.0 (12.6)
54.9 (12.1)
0.30
Mean weight (kg) (SD)
62.0 (8.0)
64.5 (10.2)
0.37
Mean height (cm) (SD)
159.9 (7.0)
160.5 (7.2)
0.79
BMI (kg/m2) (SD)
24.2 (2.3)
25.1 (2.9)
0.27
Mean follow-up (y) (SD)
5.2 (1.0)
6.2 (1.5)
0.02
Limb length discrepancies
4.6 (0.7)
SC 4.8 (1.1)
AC C
EP
TE D
M AN U
SD: Standard deviation
P value
RI PT
Non-osteotomy group
0.39
ACCEPTED MANUSCRIPT Table 2. Comparison of the clinical results between the non-osteotomy and osteotomy groups Osteotomy
group
group
Operative time (min) (SD)
112.3 (18.2)
128.5 (22.1)
0.01
Perioperative bleeding (ml) (SD)
443.2 (117.8)
537.5 (162.9)
0.03
Patients satisfaction
0.15
19
11
Good
2
5
Moderate
1
Unsatisfactory
0
Patient-perceived leg-length discrepancy Trendelenburg sign
Knee valgus and external rotation
EP AC C
1
0.5 (0.6)
2.2 (1.2)
<0.0001
6/22
17/20
0.0001
2/22
3/20
0.56
4/22
15/20
<0.0001
4/20
<0.0001
TE D
Limp
3
M AN U
Leg-length discrepancy (cm) (SD)
SC
Excellent
deformity
P value
RI PT
Non-osteotomy
19/22
ACCEPTED MANUSCRIPT Table 3. Comparison of radiographic results between the non-osteotomy and osteotomy groups (SD) Non-osteotomy
Osteotomy
group
group
45.9 (6.1)
47.2 (6.8)
Horizontal
21.8 (3.3)
20.6 (2.5)
0.18
Vertical
21.1 (5.2)
18.6 (3.3)
0.10
55.4 (6.2)
57.4 (7.3)
0.39
35.7 (7.8)
0.41
The height of the tip of the GT before surgery (mm)
AC C
EP
TE D
GT: greater trochanter
32.8 (12.9)
M AN U
Translation of the GT (mm)
0.57
RI PT
Hip center (mm)
SC
Inclination (°)
P value
ACCEPTED MANUSCRIPT Table 4. Summary of complications in the non-osteotomy and osteotomy groups Osteotomy group
Femoral nerve palsy
3
0
Sciatic nerve palsy
0
0
Delayed union at the osteotomy site
0
1
Intraoperative fracture of the distal
0
1
AC C
EP
TE D
M AN U
SC
femoral segment
RI PT
Non-osteotomy group
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
S0883-5403(16)30523-X
DOI:
10.1016/j.arth.2016.08.021
Reference:
YARTH 55372
To appear in:
The Journal of Arthroplasty
Received Date: 29 May 2016 Revised Date:
30 July 2016
Accepted Date: 18 August 2016
Please cite this article as: Li H, Xu J, Qu X, Mao Y, Dai K, Zhu Z, Comparison of total hip arthroplasty with and without femoral shortening osteotomy for unilateral mild to moderate high hip dislocation, The Journal of Arthroplasty (2016), doi: 10.1016/j.arth.2016.08.021. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT Comparison of total hip arthroplasty with and without femoral shortening osteotomy for unilateral mild to moderate high hip dislocation Running title: THA for unilateral high hip dislocation
RI PT
Huiwu Li, MD 1; Jiawei Xu 1, MD; Xinhua Qu, MD 1;Yuanqing Mao MD 1; Kerong Dai, MD 1; Zhenan Zhu, MD 1
1. Department of Orthopaedics, Shanghai No. 9th People’s Hospital, Shanghai Jiao
SC
Tong University School of Medicine, Shanghai 200011, China [email protected]
Jiawei Xu MD
[email protected]
Xinhua Qu, MD
[email protected]
Yuanqing Mao, MD
[email protected]
Kerong Dai, MD
[email protected]
TE D
Zhenan Zhu, MD
M AN U
Huiwu Li, MD
[email protected]
EP
We confirm that this manuscript has not been published elsewhere and is not under
AC C
consideration by another journal. The authors have no conflicts of interest to declare. All authors have approved the manuscript and agree with submission to JOA. This study was performed in Shanghai No. 9th People’s Hospital. Please address all correspondence to: Zhenan Zhu, MD. Department of Orthopaedics, Shanghai No. 9th People’s Hospital, Shanghai Jiao Tong University School of Medicine, Shanghai 200011,
China.
Phone:
[email protected].
86-21-63139920,
Fax:
86-21-63139920,
Email:
ACCEPTED MANUSCRIPT Comparison of total hip arthroplasty with and without femoral shortening osteotomy
2
for unilateral mild to moderate high hip dislocation
3
Abstract
4
Background: This study is to compare the outcome between THA with and without femoral
5
shortening osteotomy for unilateral mild to moderate high hip dislocation in DDH patients.
6
Methods: The data on 42 hips in 42 patients who had undergone THA for unilateral mild to
7
moderate high hip dislocation were retrospectively reviewed after being prospectively
8
collected. In 22 patients, hips were reduced by soft tissue release and direct leverage using an
9
elevator, without the osteotomy. The remaining 20 patients were treated with a
10
subtrochanteric transverse shortening osteotomy. The mean follow-up of patients was 5 years
11
(SD 1.0) for the non-osteotomy group and 6.2 years (SD 1.6) for the osteotomy group.
12
Results: The Harris Hip Score (HHS) significantly improved in both groups. In the
13
non-osteotomy group, we observed a lower leg-length discrepancy compared with the
14
osteotomy group (0.4 cm and 2.2 cm, respectively). Four patients (18.2%) in the
15
non-osteotomy group and 15 patients (75%) in the osteotomy group developed a limp
16
(P<0.0001). 3 patients (13%) developed femoral nerve palsy in the non-osteotomy group but
17
they all recovered completely within 6 months after the surgery. Nineteen patients in the
18
non-osteotomy group showed knee valgus deformity immediately after the surgery, but only 4
19
cases in the osteotomy group.
20
Conclusion: Compared with THA with femoral shortening osteotomy, THA without the
21
osteotomy was associated with a lower number of patients who developed a limp at the end of
22
follow-up; however, the rehabilitation was slower and more difficult, and a larger number of
23
patients showed reversible nerve palsy and knee valgus deformity.
24
Key word: high hip dislocation; developmental dysplasia of the hip; total hip arthroplasty;
25
femoral shortening osteotomy; comparison study
AC C
EP
TE D
M AN U
SC
RI PT
1
26 27
Introduction
28
Total hip arthroplasty (THA) is an effective treatment for developmental dysplasia of the
29
hip (DDH) with high hip dislocation in adults. The durability of arthroplasty in these patients
30
is better with the restoration of an anatomic hip center [1]. One option for restoring the 1
ACCEPTED MANUSCRIPT anatomic hip center is a femoral shortening osteotomy, which facilitates hip reduction and
32
protects the nerve and artery [2]. However, this approach is considered to be complex and
33
time-consuming, and can lead to complications such as shortened limbs, nonunion at the
34
osteotomy site, loosened prosthesis, and hip dislocation [1-3]. For the patients with a mild to
35
moderately high dislocation of the hip, a number of surgeons have reported THA using
36
methods other than femoral shortening osteotomy, such as extensive soft tissue release,
37
intravenous injection of rocuronium, elevator leverage, or distraction nail [4-7]. These
38
techniques show advantages including the lengthening of the limb and reconstruction of the
39
true socket as well as the restoration of the anatomical course of the abductor and balance of
40
its muscular strength. However, complications such as nerve palsies, greater trochanter
41
avulsion fracture, and the feeling of the lengthening of the affected limb have been reported
42
[8-11].
M AN U
SC
RI PT
31
To the best of our knowledge, there are no studies compared the results of THA with and
44
without femoral shortening osteotomy for unilateral mild to moderately high hip dislocation
45
in patients with DDH. The aim of this study was to compare the results of the 2 techniques
46
with respect to hip function, leg-length discrepancy and complications in 2 comparable
47
cohorts of patients.
48
Patients and methods
50
Patients
EP
49
TE D
43
During the period from January 2005 to December 2009, more than 800 patients with
52
DDH underwent a THA in our hospital. A total of 66 hips were considered as having high
53
dislocation corresponding to type IV of the Crowe classification [12]. We classified the
54
Crowe IV dislocation into 2 subtypes, the mild to moderately high dislocation when the
55
migration of the femoral head was more than 1/5 of the pelvic height but less than 2/5 of the
56
pelvic height, and the extremely high dislocation when the migration was more than 2/5 of the
57
pelvic height (Fig. 1). In this study, the patients with an extremely high dislocation or with a
58
severe deformity of proximal femur were excluded from this study because the osteotomy
59
would have to be performed. Therefore, the patients with a mild to moderately high
60
dislocation were included in this study. We also excluded patients with bilateral high
AC C
51
2
ACCEPTED MANUSCRIPT dislocation of the hip. The surgeries were performed by the same groups of surgeons. From
62
January 2005 to September 2007, the femoral shortening osteotomy was performed for all
63
patients with mild to moderately high dislocation of the hip. Because of complex and
64
shortened limbs caused by shortening osteotomy, after September 2007, the patients were
65
reduced by soft tissue release and elevator direct leverage without femoral shortening
66
osteotomy. Forty-six patients were included in the study, of which 4 (4 hips) were lost to
67
follow-up, one hip with osteotomy, three hips without osteotomy. Finally, we retrospectively
68
reviewed prospectively collected data on 42 hips in 42 patients who underwent a cementless
69
THA for DDH with unilateral mild to moderately high hip dislocation. In 22 patients, the hip
70
was reduced by soft tissue release and direct leverage using an elevator without femoral
71
shortening osteotomy (from October 2007 to December 2009), while 20 patients were treated
72
with the osteotomy (from January 2005 to September 2007). The mean follow-up of patients
73
was 5.0 years (SD 1.0; 4 to 7 years) for the non-osteotomy group and 6.2 years (SD 1.6; 4 to 9
74
years) for the osteotomy group. Data on the sex, age, weight, height, body mass index, and
75
duration of follow-up of patients in the 2 groups are presented in Table 1.
M AN U
SC
RI PT
61
In the non-osteotomy group, the limb length discrepancies were 4.6 cm (SD 0.7, 3 to 6
77
cm). Three patients had osteoarthritis of the contralateral hip with dislocation classified as
78
Crowe type I, and 3 patients as Crowe type II. These patients had already undergone a THA
79
before this surgery. Four patients (18%) underwent pelvic osteotomy for the treatment of the
80
hip in childhood. Spinal comorbidities associated with DDH included multiple sclerosis in 5
81
patients (22.7%). In the osteotomy group, the limb length discrepancies were 4.8 cm (SD 1.1,
82
2.5 to 6.5 cm). Three patients had osteoarthritis of the contralateral hip with dislocation
83
classified as Crowe grade I, and 1 patient as Crowe grade III. These patients had already
84
undergone a THA before this surgery. Two patients (10%) underwent pelvic osteotomy for
85
the treatment of the hip in childhood. Spinal comorbidities associated with DDH included
86
multiple sclerosis in 3 patients (15%).
87
Materials
AC C
EP
TE D
76
88
In the non-osteotomy group, the S-ROM stem and Duraloc acetabular cup with a ceramic
89
on ceramic bearing were used in 9 hips and with a polyethylene on cobalt-chromium bearing
90
in 2 hips (Depuy, Warsaw, Indiana); the Omnifit hydroxyapatite-coated stem and SecurFit 3
ACCEPTED MANUSCRIPT acetabular cup with a ceramic on ceramic bearing were used in 9 hips and with a polyethylene
92
on cobalt-chromium bearing in 2 hips (Stryker, Mahwah, New Jersey). Of them, 2 hips were
93
treated with a 22-mm cobalt-chromium femoral head, 2 hips with a 28-mm cobalt-chromium
94
femoral head, and 18 hips with a 28-mm ceramic femoral head. In the osteotomy group, the
95
S-ROM stem and Duraloc acetabular cup with a ceramic on ceramic bearing were used in 14
96
hips and with a polyethylene on cobalt-chromium in 6 hips (DePuy). Of them, 1 hip was
97
treated with a 22-mm cobalt-chromium femoral head, 5 hips with a 28-mm cobalt-chromium
98
femoral head, and 14 hips with a 28-mm ceramic femoral head.
99
Surgical technique
SC
RI PT
91
All of the operations were performed using a posterolateral approach. The procedure of
101
acetabular component insertion in THA with and without femoral shortening osteotomy was
102
the same. All osteophytes and hypertrophied soft tissue around the true acetabulum were
103
removed. The acetabular cups were inserted with a press-fit fixation.
M AN U
100
In the non-osteotomy group, the procedure has been previously described in detail [11].
105
We do not favor extensive muscle release to reduce the hip and always take great care to
106
retain the periarticular muscles. As reduction, the limb was held in adduction and internal
107
rotation as distal traction was applied to the femur by an assistant. If this was insufficient to
108
reduce the hip, direct leverage was applied to the greater trochanter using a Hohmann
109
retractor, which obtained purchase on the bone below the acetabulum. When the modular
110
femoral head reached the level of the acetabular component, the reduction was achieved by
111
external rotation of the leg. During the procedure, the knees were always held in flexion to
112
relax the sciatic nerve and reduce tension in soft tissue. This position was maintained for
113
several days after the surgery to avoid damage to the nerve.
EP
AC C
114
TE D
104
In the osteotomy group, at the beginning of the femoral osteotomy, the subtrochanteric
115
area of the femur is approached by elevating or splitting a short section of the vastus lateralis.
116
Next, a transverse proximal femoral osteotomy is performed in accordance with preoperative
117
templating. The femoral trial was then replaced into the proximal fragment, the head reduced
118
into the acetabulum, and traction applied to the distal femur by pulling on the leg. The bony
119
overlap at the osteotomy level was removed from the distal femur and the amount of resection
120
was minimized as much as possible. We usually place a cerclage cable or wire 4
ACCEPTED MANUSCRIPT 121
prophylactically before implant impaction. The final components were inserted with the stem
122
transfixing the osteotomy. Once fully seated, the osteotomy should be in apposition and
123
rigidly fixed. The rehabilitation program varied according to soft tissue tension. In general, patients
125
were allowed partial weight-bearing from 3 days to 1 week after the surgery and then full
126
weight-bearing at 1 to 6 weeks. If the soft tissue was very tight after leg lengthening, the hip
127
and knee flexion position was maintained for several days postoperatively and then extended
128
gradually over the course of 1 to 2 weeks. After regaining full extension, walking exercises
129
with crutches and full weight-bearing were initiated.
130
Evaluation method
SC
RI PT
124
Clinical and radiographic data were obtained for all patients before the surgery and at
132
follow-up examinations. The clinical evaluation was done based on the Harris Hip Score
133
(HHS), while patients’ satisfaction was subjectively described as excellent, good, moderate,
134
or unsatisfactory. The presence of a Trendelenburg sign was recorded. Length discrepancy
135
between the bilateral legs was assessed after the surgery by measuring the distance from the
136
anteriorsuperior iliac spine to the medial malleolus. The patient-perceived leg-length
137
discrepancy was also assessed by placing blocks under the shorter limb until the patient felt
138
comfortable. The function of the sciatic and femoral nerves was carefully examined but did
139
not include routine electrophysiological testing.
EP
TE D
M AN U
131
A routine radiographic examination included anteroposterior and lateral hip views. The
141
inclination and horizontal and vertical distance of the postoperative hip center were
142
determined using the radiographic technique described by Russotti and Harris [13]. The
143
translation of the tip of the greater trochanter was determined by comparing the change inthe
144
distance between the tip of the greater trochanter and interteardrop line before and after the
145
surgery. Pelvic obliquity was determined by measuring the iliolumbar angle. On the AP
146
radiographs of the pelvis and lower lumbar spine, 1 line was drawn connecting the apices of
147
both iliac crests and another line along the bottom of the fourth lumbar vertebra. The
148
iliolumbar angle was measured at the convergence of these 2 lines (Fig. 1) [14]. The stability
149
of the acetabular components was assessed radiographically using the method of DeLee and
150
Charnley [15] and that of the femoral components using the method of Gruen et al. [16].
AC C
140
5
ACCEPTED MANUSCRIPT 151
Serial radiographs were used to evaluate the union (bridging of the bone trabeculae) at the
152
osteotomy site. All samples were assessed by LHW and XJW independently. Agreement
153
between LHW and XJW was good (all Cohen’s unweighted κ> 0.90). Statistical analysis was conducted with SPSS for Windows Version 11.5 (SPSS Inc,
155
Chicago, IL, USA). The Student’s t-test was used to compare the pre- and postoperative HHS,
156
operation time, amount of bleeding, and radiographic parameters between the non-osteotomy
157
and osteotomy groups. The Mann-Whitney U test was used to assess patients’ satisfaction and
158
the Mantel-Haenszel chi-squared test to assess the patient-perceived leg-length discrepancy,
159
Trendelenburg sign, limp, knee valgus and external rotation deformity. Differences were
160
considered statistically significant at a P value of less than 0.05.
SC
RI PT
154
162
Results
163
Clinical results
164 165
M AN U
161
The clinical results are shown in Table 2. The non-osteotomy group showed a significantly shorter operation time (P=0.01) and a smaller amount of bleeding (P=0.03). A significant improvement in the HHS after the surgery was observed in both groups. In
167
the non-osteotomy group, the mean HHS improved from 48.4 points (SD 10, 28 to 74 points)
168
to 90.8 points (SD 4.7, 80 to 100) at the final follow-up examination (P<0.001). In the
169
osteotomy group, it improved from 47.0 points (SD 12.0, 24 to 70 points) to 87 points (SD
170
5.8, 70 to 96 points) at the final follow-up examination (P<0.001). The score improved later
171
during follow-up in the non-osteotomy group compared with the osteotomy group (Fig. 2).
EP
AC C
172
TE D
166
In the non-osteotomy group, patients’ satisfaction at the final follow-up examination was
173
reported as excellent for 19 patients, good for 2 patients, moderate for 1 patient, and
174
unsatisfactory for 0 patients. In the osteotomy group, the corresponding numbers were 11, 5,
175
3, and 1 patient, respectively (P=0.15).
176
In the non-osteotomy group, the affected limbs were lengthened 4.1 cm (SD 0.9, 2.0 to
177
5.0 cm). The average leg-length discrepancy at the final follow-up examination was 0.4 cm
178
(SD 0.7, -1.2 to 1.5 cm). In the osteotomy group, the affected limbs were lengthened 2.7 cm
179
(SD 1.2, 1.0 to 4.8 cm). The average leg-length discrepancy was 2.2 cm (SD 1.2, 0 to 5.0 cm,
180
Fig. 3). Six patients reported a feeling of leg-length discrepancy in the non-osteotomy group 6
ACCEPTED MANUSCRIPT 181
(the lengthening of the affected limb was felt by 5 patients, and the shortening by 1 patient).
182
In the osteotomy group, 17 patients reported a feeling of leg-length discrepancy (shortening
183
of the affected limb). There was no revision or marked loosening during the follow-up period in either group.
185
All patients were able to put on shoes and socks by themselves as well as walk and climb
186
stairs independently. In addition, all patients had a positive Trendelenburg sign before the
187
surgery. However, at the final follow-up examination, only 2 patients in the non-osteotomy
188
and 3 patients in the osteotomy group showed a positive Trendelenburg sign. Without a
189
heightening shoe, 1 patient (5%) in the osteotomy group showed a severe limp, 4 patients
190
(20%) a moderate limp, 10 (50%) a slight limp,
191
group, only 4 patients (18%) showed a slight limp (P<0.0001).
SC
RI PT
184
M AN U
and 5 (25%) no limp. In the non-osteotomy
Immediately after the surgery, 19 patients in the non-osteotomy group showed knee
193
valgus deformity, 6 of these patients showed not only valgus but external rotation deformity,
194
and 1 of these patients had patella dislocation when the knee flexed more than 40° (Fig. 4).
195
By the end of the follow-up, the knee valgus deformity resolved in 13 patients, and 6 patients
196
still showed slight valgus deformity. The patella dislocation resolved after 3 months with the
197
correction of the knee valgus. In the osteotomy group, the knee valgus deformity occurred in
198
4 cases and resolved in all patients by the end of the follow-up.
199
Radiographic evaluation
EP
TE D
192
All cups in both groups were built in the anatomical acetabulum. The mean inclination of
201
the cup, vertical distance of the hip center, and horizontal distance in the 2 groups are shown
202
in Table 3. There was no significant difference between the groups in the height of the tip of
203
the greater trochanter before the surgery (P=0.39) and in the shift of the greater trochanter
204
after the surgery (P=0.41).
205
AC C
200
Postoperative pelvic obliquity occurred in all patients. The immediate postoperative
206
obliquity was more apparent in the non-osteotomy group than in the osteotomy group. In the
207
non-osteotomy group, the difference of the iliolumbar angle before and after the surgery was
208
3.0° (SD 2.5, -1° to 7°). The pelvic obliquity corrected with time (Fig. 5). At the last
209
follow-up examination, the iliolumbar angle was 0.8° (SD 3.1, -4° to 7°). In the osteotomy
210
group, the difference was 2.5° (SD 2.0, -1.6° to 6°). At the last follow-up examination, the 7
ACCEPTED MANUSCRIPT 211
iliolumbar angle was -1.4° (SD 2.8, -7° to 3.2°) and had almost returned to the preoperative
212
level (Fig. 6). Solid union at the osteotomy site was achieved at a mean period of 4.2 months (SD 1.5, 3
214
to 9 months). There was no loosening or progressive radiolucency or osteolysis adjacent to
215
the acetabulum and femoral component observed at the last follow-up examination in any
216
case.
217
Complications
RI PT
213
In the osteotomy group, there was 1 intraoperative fracture of the distal femoral segment
219
and 1 delayed union (9 months) at the osteotomy site. In the non-osteotomy group, 3 patients
220
(14%) had femoral nerve palsies with numbness of the medial aspect of the lower limb and
221
weakness of knee extension. The limbs were lengthened 4.0, 4.1, and 4.6 cm respectively. All
222
of these patients were treated conservatively and recovered completely at 3 weeks, 2 months,
223
and 6 months after the surgery, respectively. No sciatic nerve injury, dislocation, or infection
224
was observed in any of the groups (Table 4).
M AN U
SC
218
225
Discussion
227
This is the first study that directly compared the results of THA with and without femoral
228
shortening osteotomy for unilateral mild to moderately high dislocations in patients with
229
DDH. The comparison focused on several aspects such as surgical techniques, HHS,
230
complications, patients’ satisfaction, and leg-length discrepancy.
231
Surgical techniques
EP
AC C
232
TE D
226
Various osteotomy techniques have been shown to facilitate hip reduction [17, 18].
233
However, a number of surgeons reported that they are complex and time-consuming. In the
234
present study, the mean operation time was approximately 16 minutes longer and bleeding
235
was more excessive by about 94 mL in the group undergoing subtrochanteric transverse
236
shortening osteotomy than in the non-osteotomy group. Nonetheless, hip reduction in THA
237
without femoral shortening osteotomy requires considerable skill and experience. Sener et al.
238
[19] proposed that for severe dislocation of the hip joint, the entire excision of the joint
239
capsule and scar tissue is a priority, and then the loosening of the superior parts of the gluteus
240
maximus, adductor, rectus femoris, and sartorius muscles is considered. Kawai et al.[8] 8
ACCEPTED MANUSCRIPT proposed an additional femoral neck cut performed until femoral head reduction was possible,
242
while Yan et al. [5] suggested an intravenous injection of rocuronium at 0.9 mg/kg 1 minute
243
before reduction, combined with a strong continuous traction of the affected limb. In the
244
present study, a direct leverage using an elevator, which obtained purchase on the bone below
245
the acetabulum, was applied to the greater trochanter. No matter what the techniques, we
246
believe that a long learning curve is necessary.
247
HHS
RI PT
241
The improvement of the HHS in the non-osteotomy group was relatively slower than that
249
in the osteotomy group. The excessive tightness of soft tissue caused by the lengthening of
250
the limb hampered the rehabilitation process. However, with the stretching of soft tissue, the
251
HHS in the non-osteotomy was comparable and perhaps even better than that in the
252
osteotomy group at the final follow-up examination.
253
Complications
M AN U
SC
248
To our knowledge, there is no documented maximum value for an extremity to be safely
255
lengthened without aneurological complication, and there is also no definition of mild,
256
moderate or extremely high dislocation. In this study, we classified the dislocation according
257
to our experience and regarded migration of the femoral head of more than 2/5 of the pelvic
258
height as an extremely high dislocation and the standard of whether the shortening osteotomy
259
was performed. Since 2007, all of the patients whose migration of the femoral head was less
260
than 2/5 of the pelvic height were treated successfully without femoral shortening osteotomy,
261
no matter how many centimeters the migration was. Although some authors have reported
262
that limb lengthening should be limited to 4 cm [20]. Kerboull et al. [21] published favorable
263
results in a consecutive series of 118 THAs performed for Crowe type IV hip dislocation. In
264
that series, only 2 hips required a shortening osteotomy and only 1 transient peroneal nerve
265
palsy occurred despite the fact that 30 limbs were lengthened by more than 4 cm. In fact, they
266
even reported that limb lengthening of up to 7 cm was possible. In this study, in the
267
non-osteotomy group, limb lengthening of more than 4 cm was performed in 10 hips, none of
268
which developed sciatic nerve palsy. However, there were 3 patients who presented with
269
femoral nerve palsy. We speculated that the femoral nerve is more vulnerable to stretch than
270
the sciatic nerve, which is associated with the anatomic characteristics of the femoral nerve,
AC C
EP
TE D
254
9
ACCEPTED MANUSCRIPT 271
namely, it is shorter and diverges into the muscles earlier than the sciatic nerve. Nevertheless,
272
all patients fully recovered without sequelae in 6 months. Another phenomenon observed mainly in the non-osteotomy group was knee valgus and
274
external rotation deformity. The excessively tight soft tissue, especially the iliotibial tract,
275
contributed to the deformity. In addition, 1 patella dislocation after the surgery occurred in the
276
non-osteotomy group. Our previous study showed that DDH is associated with morphological
277
changes in the knee joint [22]. Compared with the knees in patients with normal hips, the
278
femoral condyles in patients with dislocated hips were smaller and the trochlear groove was
279
shallower. The dysplasia of the knee combined with the valgus and external rotation caused
280
by a tightened iliotibial tract led to patella dislocation in this study. Although most of the knee
281
valgus and external rotation deformity resolved with relaxation of the iliotibial tract, the
282
change of the patella track after THA in patients with high DDH still deserves our attention.
283
Patients’ satisfaction and leg-length discrepancy
M AN U
SC
RI PT
273
Compared with patients in the osteotomy group, patients in the non-osteotomy group
285
showed relatively higher satisfaction even there is no significant difference. The main
286
complaint that affected patients’ satisfaction at the final follow-up examination in the
287
osteotomy group was a limp. It has been reported that most osteotomies of the femur leave the
288
leg significantly shortened [7]. Park et al. [17] reported 23 cases of Crowe type IV DDH
289
patients managed with transverse femoral shortening osteotomy. The average leg-length
290
discrepancy after the surgery in patients with unilateral dysplasia was 42 mm. The
291
postoperative limp was graded as none (17%), mild (26%), moderate (39%), or severe (18%),
292
which was in line with our results.
EP
AC C
293
TE D
284
Unlike in the osteotomy group, the most common complaint in the non-osteotomy group
294
in the first several months after surgeries was leg lengthening, even though the true average
295
leg-length discrepancy at the final follow-up examination was only 4 mm. This leg
296
lengthening was due to pelvic obliquity caused by excessively tight soft tissues. In this study,
297
an established tautness of soft tissues and muscles surrounding the THA joint caused an
298
inclination of the pelvis toward the operated side of the body. We used the change of the
299
iliolumbar angle preoperatively and postoperatively to reflect pelvic obliquity caused by
300
muscle tautness. Postoperative pelvic obliquity occurred in all patients; it led to relative leg 10
ACCEPTED MANUSCRIPT length inequality, as a result of which patients felt lengthening of the affected leg. As tension
302
in soft tissue gradually decreased, pelvic obliquity was corrected and the feeling of leg
303
lengthening slowly disappeared. This process was longer and more apparent in the
304
non-osteotomy group than in the osteotomy group. When pelvic obliquity was corrected, the
305
feeling of leg lengthening disappeared or alleviated in the non-osteotomy group, but the
306
feeling of leg shortening became gradually more apparent in the osteotomy group.
RI PT
301
Our study has several limitations. First, it was a retrospective evaluation of prospectively
308
followed patients. Second, the study group was relatively small, even though the study was
309
one of the largest series of patients with this uncommon deformity. Third, the implants in this
310
series were not identical for all arthroplasties. Finally, the follow-up period was short- to
311
mid-term. All of the above reasons may have affected the objectiveness of this study.
312
However, the fact that all operations were performed by the same group of surgeons may be
313
considered a strength of the study.
M AN U
SC
307
In conclusion, we determined that the intermediate-term clinical and radiographic results
315
of THA with or without femoral shortening osteotomy for DDH patients with a high
316
dislocation were satisfactory. Compared with THA with femoral shortening osteotomy, THA
317
without the osteotomy was associated with a shorter operative time and less bleeding, a lower
318
number of patients with a limp. However, the latter procedure also required higher surgical
319
skills and was associated with slower and more difficult rehabilitation as well as more cases
320
of reversible nerve palsy and knee valgus deformity; the long-term effects of valgus deformity
321
on the knee remain to be observed.
AC C
EP
TE D
314
This study was supported by Shanghai Natural Science Foundation (Grant No.
322 323
15ZR1424800) and Scientific Research Foundation for the Returned Overseas Chinese
324
Scholars, Ministry of Education of China (Grant No. 201550002).
325
References
326
1.
Pagnano MW, Hanssen AD, Lewallen DG, Shaughnessy WJ.The effect of superior
327
placement of the acetabular component on the rate of loosening after total hip
328
arthroplasty. J Bone Joint Surg Am. 1996;78-A:1004-1014.
329 330
2.
Zagra L, Bianchi L, Mondini A, Ceroni RG. Oblique femoral shortening osteotomy in total hip arthroplasty for high dislocation in patients with hip dysplasia. Int Orthop. 2015; 11
ACCEPTED MANUSCRIPT 39(9): 1797-802.
331 332
3.
Zhu J, Shen C, Chen X, Cui Y, Peng J, Cai G. Total hip arthroplasty with a non-modular
333
conical stem and transverse subtrochanteric osteotomy in treatment of high dislocated
334
hips. J Arthroplasty. 2015; 30(4): 611-614. 4.
Wu X, Li SH, Lou LM, Cai ZD. The techniques of soft tissue release and true socket
RI PT
335 336
reconstruction in total hip arthroplasty for patients with severe developmental dysplasia
337
of the hip. Int Orthop. 2012;36(9):1795-1801.
338
5.
Yan F, Chen G, Yang L, He R, Gu L, Wang F. A reduction technique of arthroplasty without
340
developmental high dislocation of hip: a case series of 28 hips. J Arthroplasty.
341
2014;29(12):2289-2293. 6.
femoral
shortening
osteotomy
for
the
treatment
of
M AN U
342
subtrochanteric
SC
339
Li H, Mao Y, Oni JK, Dai K, Zhu Z. Total hip replacement for developmental dysplasia
343
of the hip with more than 30% lateral uncoverage of uncemented acetabular components.
344
Bone Joint J. 2013; 95-B(9):1178-1183.
345
7.
Lai KA, Shen WJ, Huang LW,Chen MY.Cementless total hip arthroplasty and limb-length equalization in patients with unilateral Crowe type IV hip dislocation.J Bone
347
Joint Surg Am. 2005;87-A:339-345.
348
8.
TE D
346
Kawai T, Tanaka C, Kanoe H. Total hip arthroplasty for Crowe IV hip without subtrochanteric shortening osteotomy -a long term follow up study. BMC Musculoskelet
350
Disord. 2014;15:72.
352
9.
Hartofilakidis G, Karachalios T. Total hip arthroplasty for congenital hip disease. J Bone
AC C
351
EP
349
Joint Surg Am. 2004;86:242-250.
353
10. Eskelinen A, Helenius I, Remes V, Ylinen P, Tallroth K, Paavilainen T. Cementless total
354
hip arthroplasty in patients with high congenital hip dislocation. J Bone Joint Surg Am.
355
2006;88:80-91.
356
11. Zhao X, Zhu ZA, Xie YZ, Yu B, Yu DG. Total hip replacement for high dislocated hips
357
without femoral shortening osteotomy. J Bone Joint Surg Br. 2011;93(9):1189-1193.
358
12. Crowe JF, Mani VJ, Ranawat CS. Total hip replacement in congenital dislocation and
359 360
dysplasia of the hip. J Bone Joint Surg Am. 1979;61:15-23. 13. Russotti GM, Harris WH. Proximal placement of the acetabular component in total hip 12
ACCEPTED MANUSCRIPT 361
arthroplasty. A long-term follow-up study. J Bone Joint Surg Am. 1991;73:587-592.
362
14. Lee DY, Choi IH, Chung CY, Cho TJ, Lee JC. Fixed pelvic obliquity after poliomyelitis:
364 365
Classification and management. J Bone Joint SurgBr. 1997;79:190-196. 15. DeLee JG, Charnley J.Radiological demarcation of cemented sockets in total hip replacement. Clin Orthop Relat Res. 1976;121:20-32.
RI PT
363
366
16. Gruen TA, McNeice GM, Amstutz HC. “Modes of failure” of cemented stem-type
367
femoral components: a radiographic analysis of loosening.Clin Orthop Relat
368
Res.1979;141:17-27.
17. Park MS, Kim KH, Jeong WC. Transverse subtrochanteric shortening osteotomy in
370
primary total hip arthroplasty for patients with severe hip developmental dysplasia.J
371
Arthroplasty.2007;22:1031-1036.
M AN U
SC
369
372
18. Krych AJ, Howard JL, Trousdale RT, Cabanela ME, Berry DJ. Total hip arthroplasty
373
with shortening subtrochanteric osteotomy in Crowe type-IV developmental dysplasia. J
374
Bone Joint Surg Am. 2009;91:2213-2221.
376
19. Sener N, Tözün IR, Aşik M. Femoral shortening and cementless arthroplasty in high congenital dislocation of the hip. J Arthroplasty. 2002;17:41-48.
TE D
375
20. Eggli S, Hankelmayer S, Muller ME. Nerve palsy after leg lengthening in total
378
replacement arthroplasty for developmental dysplasia of the hip. J Bone Joint Surg Br.
379
1999;843-845.
EP
377
21. Kerboull M, Hamadouche M, Kerboull L. Total hip arthroplasty for Crowe type IV
381
developmental hip dysplasia: a long-term follow-up study. J Arthroplasty. 2001;16(8
382 383 384 385
AC C
380
Suppl 1):170-176.
22. Li H, Qu X, Wang Y, Dai K, Zhu Z. Morphological analysis of the knee joint in patients with hip dysplasia. Knee Surg Sports Traumatol Arthrosc. 2013;21(9):2081-2088.
386 387 388 389 390 13
ACCEPTED MANUSCRIPT Legends Fig. 1. The pelvis was divided into 5 equal parts by line 1 to 4. The mild to moderately high dislocation means the migration of the femoral head is more than 1/5 of the pelvic height but less than 2/5 of the pelvic height, that is, the bottom of the femoral head-neck junction
RI PT
locating the zone between line 2 and line 3. The extremely high dislocation means the migration was more than 2/5 of the pelvic height, that is, the bottom of the femoral head-neck junction locating beyond line 3. The degree of pelvic obliquity was reflected through the iliolumbar angle. Line OB was drawn to connect the apices of both iliac crests, and line OA
measured at the convergence of these 2 lines.
SC
was drawn along the bottom of the fourth lumbar vertebra. The iliolumbar angle AOB was
M AN U
Fig. 2. The improvement of the Harris Hip Score was slower in the non-osteotomy group than in the osteotomy group, but the nonosteotomy group achieved a comparable or better score at the final follow-up examination.
Fig. 3. A patient had undergone THA without femoral shortening osteotomy and had an equal leg length (a). A patient had undergone THA with femoral shortening osteotomy and had one
TE D
leg 4 cm shorter than the contralateral leg (b).
Fig. 4. A 45-year-old woman with Crowe type IV hip dislocation (a). A THA without femoral shortening osteotomy was performed, and the knee showed a valgus and external rotation
EP
deformity after surgery (b). Before THA, the CT scan showed that the patella was in the femoral trochlear groove but the femoral condyle was smaller and the trochlear groove was
AC C
shallow (c). After surgery, the patient had patella dislocation when the knee flexed more than 40° (d).
Fig. 5. A 46-year-old woman with Crowe type IV hip dislocation in one leg and THA in the other leg for Crowe type II hip dislocation (a). A THA without femoral shortening osteotomy was performed for a high hip dislocation. Pelvic obliquity in the operated side after THA (b). The obliquity improved 6 months after the surgery (c). Fig. 6. The immediate postoperative obliquity is more apparent in the non-osteotomy group than in the osteotomy group. In the non-osteotomy group, the difference in the iliolumbar angle preoperatively and postoperatively was 3°. Pelvic obliquity corrected with time. At the final follow-up examination, the iliolumbar angle was 0.8°. In the osteotomy group, the
ACCEPTED MANUSCRIPT difference was 2.5°. At the final follow-up examination, the iliolumbar angle of -1.4° almost
AC C
EP
TE D
M AN U
SC
RI PT
returned to the preoperative level.
ACCEPTED MANUSCRIPT Table 1. Characteristics of patients in the non-osteotomy and osteotomy groups Osteotomy group
No. of patients/ hips
22
20
Sex (M/F)
5/17
6/14
0.60
Average age (y) (SD)
51.0 (12.6)
54.9 (12.1)
0.30
Mean weight (kg) (SD)
62.0 (8.0)
64.5 (10.2)
0.37
Mean height (cm) (SD)
159.9 (7.0)
160.5 (7.2)
0.79
BMI (kg/m2) (SD)
24.2 (2.3)
25.1 (2.9)
0.27
Mean follow-up (y) (SD)
5.2 (1.0)
6.2 (1.5)
0.02
Limb length discrepancies
4.6 (0.7)
SC 4.8 (1.1)
AC C
EP
TE D
M AN U
SD: Standard deviation
P value
RI PT
Non-osteotomy group
0.39
ACCEPTED MANUSCRIPT Table 2. Comparison of the clinical results between the non-osteotomy and osteotomy groups Osteotomy
group
group
Operative time (min) (SD)
112.3 (18.2)
128.5 (22.1)
0.01
Perioperative bleeding (ml) (SD)
443.2 (117.8)
537.5 (162.9)
0.03
Patients satisfaction
0.15
19
11
Good
2
5
Moderate
1
Unsatisfactory
0
Patient-perceived leg-length discrepancy Trendelenburg sign
Knee valgus and external rotation
EP AC C
1
0.5 (0.6)
2.2 (1.2)
<0.0001
6/22
17/20
0.0001
2/22
3/20
0.56
4/22
15/20
<0.0001
4/20
<0.0001
TE D
Limp
3
M AN U
Leg-length discrepancy (cm) (SD)
SC
Excellent
deformity
P value
RI PT
Non-osteotomy
19/22
ACCEPTED MANUSCRIPT Table 3. Comparison of radiographic results between the non-osteotomy and osteotomy groups (SD) Non-osteotomy
Osteotomy
group
group
45.9 (6.1)
47.2 (6.8)
Horizontal
21.8 (3.3)
20.6 (2.5)
0.18
Vertical
21.1 (5.2)
18.6 (3.3)
0.10
55.4 (6.2)
57.4 (7.3)
0.39
35.7 (7.8)
0.41
The height of the tip of the GT before surgery (mm)
AC C
EP
TE D
GT: greater trochanter
32.8 (12.9)
M AN U
Translation of the GT (mm)
0.57
RI PT
Hip center (mm)
SC
Inclination (°)
P value
ACCEPTED MANUSCRIPT Table 4. Summary of complications in the non-osteotomy and osteotomy groups Osteotomy group
Femoral nerve palsy
3
0
Sciatic nerve palsy
0
0
Delayed union at the osteotomy site
0
1
Intraoperative fracture of the distal
0
1
AC C
EP
TE D
M AN U
SC
femoral segment
RI PT
Non-osteotomy group
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT