- Email: [email protected]
Femoral Stem Survivorship in Dorr Type A Femurs After Total Hip Arthroplasty Using a Cementless Tapered Wedge Stem: A Matched Comparative Study With Type B Femurs
Accepted Manuscript Femoral Stem Survivorship in Dorr Type A Femurs Following Total Hip Arthroplasty Using a Cementless Tapered Wedge Stem: A Matched Comparative Study with Type B Femurs Chan-Woo Park, MD, Hyeon-Jun Eun, MD, Sung-Hak Oh, MD, Hyun-Jun Kim, MD, Seung-Jae Lim, MD, Youn-Soo Park, MD PII:
S0883-5403(18)31117-3
DOI:
https://doi.org/10.1016/j.arth.2018.11.004
Reference:
YARTH 56900
To appear in:
The Journal of Arthroplasty
Received Date: 19 September 2018 Revised Date:
16 October 2018
Accepted Date: 2 November 2018
Please cite this article as: Park C-W, Eun H-J, Oh S-H, Kim H-J, Lim S-J, Park Y-S, Femoral Stem Survivorship in Dorr Type A Femurs Following Total Hip Arthroplasty Using a Cementless Tapered Wedge Stem: A Matched Comparative Study with Type B Femurs, The Journal of Arthroplasty (2018), doi: https://doi.org/10.1016/j.arth.2018.11.004. 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
1
Femoral Stem Survivorship in Dorr Type A Femurs Following Total Hip Arthroplasty Using a
2
Cementless Tapered Wedge Stem: A Matched Comparative Study with Type B Femurs
RI PT
3
Chan-Woo Park, MD, Hyeon-Jun Eun, MD, Sung-Hak Oh, MD, Hyun-Jun Kim, MD, Seung-Jae
5
Lim, MD, Youn-Soo Park, MD
6
Department of Orthopedic Surgery, Samsung Medical Center, Sungkyunkwan University School of
7
Medicine, Seoul, Korea
M AN U
SC
4
8
9
Concise Title: Stem Survivorship in Dorr Type A Femurs Following Primary THA
TE D
10
11
Please address all correspondence to:
13
Youn-Soo Park, MD
14
Department of Orthopedic Surgery, Samsung Medical Center
15
81 Irwon-ro, Gangnam-gu
16
Seoul 06351
17
South Korea.
18
Phone: 82-2-3410-3504
19
Fax: 82-2-3410-0061
20
Email: [email protected]
AC C
EP
12
1
ACCEPTED MANUSCRIPT
1
Femoral Stem Survivorship in Dorr Type A Femurs Following Total Hip Arthroplasty Using a
2
Cementless Tapered Wedge Stem: A Matched Comparative Study with Type B Femurs
AC C
EP
TE D
M AN U
SC
RI PT
3
1
ACCEPTED MANUSCRIPT
4
Abstract
5
Background: There is a lack of understanding on relationship between the femoral geometry and outcomes of total hip
7
arthroplasty (THA). We investigated clinical and radiographic outcomes of THA using a cementless tapered
8
wedge stem in patients with Dorr type A proximal femoral morphology, and compared with those of type B
9
femurs at a minimum follow-up of 5 years.
SC
Methods:
M AN U
10
RI PT
6
We analyzed 1089 hips (876 patients) that underwent THA using an identical cementless tapered wedge
12
stem. We divided all femurs into 3 types (Dorr type A, B and C). Type A and B femurs were statistically
13
matched with age, gender, body mass index and diagnosis by using propensity score matching. Clinical,
14
radiographic results and stem survivorship were compared between the matched two groups.
15
Results:
EP
TE D
11
A total of 611 femurs (56%) were classified as type A, 427 (39%) as type B, and 51 (5%) as type C. More
17
radiolucent lines around femoral stems were found in type A femurs (7.8%) than in type B femurs (2.5%)
18
(P<0.001). Patients with radiolucency showed worse Harris hip score (86.2 points) compared to those
19
without radiolucency (93.0 points) (P<0.001). The stem survivorship of type A femur (97.8%) was lower
20
than that of type B femur (99.5%) (P=0.041). The reasons for femoral revision in type A femurs were
21
periprosthetic fracture (67%), aseptic loosening (22%), and deep infection (11%).
AC C
16
2
ACCEPTED MANUSCRIPT
23
24
Conclusions: This study showed a higher rate of complications following THAs using a cementless tapered wedge stem in Dorr type A femurs than those performed in type B femurs.
RI PT
22
25
Keywords: Dorr classification, tapered wedge stem, periprosthetic femoral fracture, stem survivorship, total
27
hip arthroplasty
SC
26
M AN U
28
AC C
EP
TE D
29
3
ACCEPTED MANUSCRIPT
Abbreviations
31
THA, total hip arthroplasty; PFF, periprosthetic femoral fracture; BMI, body mass index; HHS, Harris hip
32
score; UCLA, University of California at Los Angeles; ONFH, osteonecrosis of the femoral head; BMD,
33
bone mineral density
RI PT
30
AC C
EP
TE D
M AN U
SC
34
4
ACCEPTED MANUSCRIPT
35
Introduction The use of cementless tapered wedge stems with a broaching procedure is common in contemporary total
37
hip arthroplasty (THA). Because of a recent trend in sparing the maximum bone stock and achieving
38
proximal fixation, tapered wedge stems are gaining more popularity[1-4]. However, with this type of stem,
39
the shape of the individual proximal femur is more important than with the traditional fit-and-fill stem.
40
Tapered wedge stems require tight initial fixation at the proximal metaphysis mediolaterally and three-point
41
fixation anteroposteriorly without the help of reaming[1,4].
M AN U
SC
RI PT
36
Anatomic variations in proximal femurs exist. The geometry of the proximal femur was previously
43
classified into 3 types by Dorr et al.[5]. Although various shapes of femoral implants have been introduced,
44
most were developed based on the femurs with an average canal dimension close to that of type B femurs. As
45
a result, each implant does not perfectly match all types of femur. For type A femurs, which have thick
46
cortices with champagne flute canals, this dimensional mismatch can be more severe. There is a high chance
47
of tight distal fixation with insufficient metaphyseal contact with this type of femur[6,7].
EP
TE D
42
There are also concerns of increased risks of periprosthetic femoral fracture (PFF) in THA with cementless
49
implants[8-11]. Recently, in a systemic review, Carli et al.[12] reported a three-fold increase in rates of PFF
50
with cementless tapered wedge stems. Although type A femurs have been considered safe in terms of
51
PFFs[13], recent studies have reported a high incidence of PFF in femurs with narrow canals as well[14,15].
52
AC C
48
Despite its significance, there is a lack of understanding regarding the relationship between the femoral
5
ACCEPTED MANUSCRIPT
geometry and outcomes of the femoral implant. We questioned whether a cementless tapered wedge stem for
54
THA would be as safe to use in Dorr type A femurs as in ordinary type B femurs. In this study, we compared
55
the clinical and radiographic outcomes of THA using a tapered wedge stem between Dorr types A and
56
B femurs after a minimum follow-up of 5 years.
RI PT
53
Materials and Methods
59
Patient cohort
M AN U
58
SC
57
This study was conducted under the approval of our institutional ethics committee review board. Between
61
January 2007 and December 2012, a total of 2108 consecutive THAs (1682 patients) were performed by a
62
single senior surgeon in our institution. We identified 1013 patients (1264 hips) who underwent primary
63
cementless THA utilizing an identical double tapered wedge stem. Femurs with any radiographic evidence of
64
proximal deformity below the level of anatomical neck were excluded, and patients with hereditary disease
65
affecting the skeletal system were also excluded from the study.
EP
TE D
60
Of 1013 patients, 27 patients (32 hips) died for reasons unrelated to the surgery, and 76 (107 hips) were
67
lost to follow-up before reaching a minimum of 5 years. Thirty-four patients (36 hips) were excluded
68
because of femoral deformities caused by infection, tumor, trauma, genetic disorder and previous surgeries.
69
Full clinical and radiographic data were available for 876 patients (1089 hips) who were enrolled in this
70
study.
AC C
66
6
ACCEPTED MANUSCRIPT
71
Surgical characteristics All surgeries were performed through the anterolateral approach by a single experienced surgeon.
73
Preoperative templating was conducted in all cases using standard radiographs with magnification markers.
74
Femoral preparations were performed by only the broaching technique.
RI PT
72
The same prosthesis was used in all surgeries. The femoral implant was a cementless Bencox stem
76
(Corentec, Cheon-An, South Korea) made of titanium alloy (Fig. 1). The geometry of this stem is similar to
77
the CLS Spotorno stem (Zimmer, Warsaw, IN, USA), which has a double tapered wedge design with a
78
rectangular cross-section and a slim diaphyseal part. The stem has a neck-shaft angle of 135° and a taper of
79
12-14 mm in diameter with a circular cross-section. Various stem sizes are available, ranging from 1 to 13.
80
The surface was treated by grit-blasting and the micro-arc oxidation method. This stem can be classified as
81
type 3C according to the classification suggested by Mont et al.[1]. The acetabular component was a
82
cementless Bencox cup (Corentec), which is made of titanium alloy and coated using the plasma spray
83
method. The operator used a ceramic-on-ceramic bearing in all cases.
84
Clinical assessments
AC C
EP
TE D
M AN U
SC
75
85
The Harris hip scores (HHSs) and University of California at Los Angeles (UCLA) activity scores were
86
recorded preoperatively. Clinical outcome was evaluated at each outpatient visit scheduled at 2, 6, and 12
87
months postoperatively and then annually. If patients experienced any postoperative pain, the location of
88
pain (e.g., the thigh, trochanteric area, groin, and buttocks) was recorded in their medical record. We
7
ACCEPTED MANUSCRIPT
identified major postoperative complications such as PFF, dislocation, aseptic loosening, infection, nerve
90
palsy, and any reoperations retrospectively.
91
Radiographic measurements
RI PT
89
All radiographs were measured twice each by 2 independent orthopedic surgeons who were fellows of hip
93
arthroplasty and had not participated in the surgery. The measurement was performed after removing all
94
identifying information of the patients. For continuous variables (e.g., the canal width), the average values
95
were used. For categorical variables, the value of the majority was used. The investigators used a picture
96
archiving and communication system (Centricity Enterprise Web V3.0, GE Medical Systems, Barrington, IL).
97
The radiographs obtained at the last follow-up were used for postoperative radiographic comparison. We
98
identified radiolucency of >1 mm around the femoral stem by using the Gruen zones[16] of the femur. Gruen
99
zones 1, 7, 8, and 14 were considered as proximal zones, and zones 3, 4, 5, 10, 11, and 12 as distal zones.
100
The alignment of the femoral component was classified as varus (>3°), valgus (>3°), or neutral. Subsidence
101
of the femoral component >4 mm was considered clinically meaningful. The mode of fixation of the stem
102
was categorized as bone ingrown, fibrous stable, or unstable using Engh et al.’s classification[17], and PFFs
103
were categorized based on the Vancouver classification[18].
104
Determination of bone types
AC C
EP
TE D
M AN U
SC
92
105
Femoral geometries were categorized according to the classification system of Dorr et al.[5] by using
106
standard preoperative anteroposterior radiographs. The calcar width was measured at the middle level of the
8
ACCEPTED MANUSCRIPT
lesser trochanter, and the canal width was measured at 10 cm below the lesser trochanter (Fig. 2). The
108
calcar-to-canal ratio was used to determine the bone types[19,20]. Femurs with a ratio of 0-0.500 were
109
considered as type A, 0.501-0.750 as type B, and 0.751-1 as type C (Fig. 3). The intraclass correlation
110
coefficient (ICC) was used to evaluate the reliability of the canal width ratio. The ICC for intraobserver
111
agreement was 91.4% (95% confidence interval [CI], 88.5%-94.3%), and ICC for interobserver agreement
112
was 88.7% (95% CI, 85.2%-92.2%).
SC
Statistical analysis
M AN U
113
RI PT
107
To reduce possible confounding factors, groups with types A and B femurs were statistically matched by
115
age, sex, body mass index (BMI), follow-up period, and diagnosis. One-way analysis of variance was used to
116
compare continuous variables, and the chi-square or Fisher’s exact test was used to compare categorical
117
variables among 3 groups. Groups with types A and B femurs were matched using propensity score matching,
118
as generated by logistic regression analysis. The paired t-test was used to analyze improvements in clinical
119
scores for each group. To compare 2 groups, the Student’s t-test and Wilcoxon rank sum test were used for
120
continuous variables. Survivorship analysis was performed with the Kaplan-Meier estimator using the end
121
point of stem revision for any reason. Statistical analysis was performed using SPSS statistics software,
122
version 24.0 (IBM Corp., Armonk, NY, USA), and P<0.05 was considered statistically significant.
AC C
EP
TE D
114
123
124
Results
9
ACCEPTED MANUSCRIPT
125
Patient characteristics There were 418 men (519 hips, 47.7%) and 458 women (570 hips, 52.3%) with a mean age of 52.5 years
127
(range, 16-81 year) at the index operation. The mean BMI was 24.4 kg/m2 (range, 15.3-37.9 kg/m2). Primary
128
diagnoses for THA included osteonecrosis of the femoral head (ONFH) (520 hips, 47.8%), osteoarthritis
129
(469 hips, 43.1%), and inflammatory arthritis, fracture, and others (100 hips, 9.2%). The mean follow-up
130
period was 84 months (range, 60-133 months).
131
Characteristics of bone types
M AN U
SC
RI PT
126
Among 1089 hips, 611 femurs (56.0%) were classified as type A, 427 (39.3%) as type B, and 51 (4.7%) as
133
type C. There were significant differences in age, sex, and diagnosis between the 3 groups (Table I). Patients
134
with type A femurs were the youngest (mean, 50.6 years), and those with type C femurs were the oldest
135
(mean, 64.7 years). There was a slight male predominance (54.7 %) in the type B femur group, whereas the
136
other groups showed female predominance. The most common diagnosis for patients with type C femurs was
137
osteoarthritis, whereas that for patients with the other 2 types was ONFH. There were also differences
138
between the 3 groups in implant size selection. After one-to-one propensity score matching was performed
139
for type A and B femurs, no meaningful difference was found in all patient characteristics (Table II). All
140
clinical and radiographic results, complications, and revision rates were compared between the matched 2
141
groups.
142
Clinical outcomes
AC C
EP
TE D
132
10
ACCEPTED MANUSCRIPT
In both groups, clinical scores improved significantly after THA (Table III). For type A femurs, the mean
144
HHS increased from 45.1 points preoperatively to 92.3 points postoperatively. The mean UCLA score
145
improved from 3.16 points to 6.44 points. After THA, type B femurs also showed improvement in HHSs
146
(from 44.0 to 92.9) and UCLA scores (from 3.10 to 6.46). Patients with type A femurs experienced more
147
thigh pain than those with type B femurs (P=0.008). No significant differences were found regarding
148
postoperative HHSs and UCLA scores at the final follow-up.
149
Radiographic features
M AN U
SC
RI PT
143
More femoral radiolucent lines were found in type A femurs than in type B femurs (7.8% versus [vs.]
151
2.5%) (P<0.001) (Table IV). Those were mostly limited to proximal zones (93.8%) (Fig. 4A-D). Stems in
152
type A femurs showed more neutral alignments than those in type B femurs (95.6% vs. 89.5%) (P<0.001).
153
More radiographic osteointegration was found in type B femurs than in type A femurs (94.9% vs. 90.7%)
154
(P=0.026). The patients with radiolucency felt more thigh pain (31%) and groin pain (16.7%) than those
155
without radiolucency (Table V), and the postoperative HHS was significantly lower in these patients
156
(P<0.001).
157
Complications and revisions
AC C
EP
TE D
150
158
There were more PFFs in patients with type A femurs than in those with type B femurs (3.2% vs. 1.0%)
159
(P=0.027) (Table VI). Of them, 4 fractures were caused intraoperatively, whereas no intraoperative fracture
160
occurred in patients with type B femurs. There were 9 postoperative Vancouver B2 fractures, all of which
11
ACCEPTED MANUSCRIPT
were revised using a distal fixating prosthesis (Fig. 5A-C). Two patients with type A femurs showed
162
radiolucency with persistent thigh pain and underwent revision with a different type of stem. One patient
163
developed periprosthetic infection and required 2-stage revision.
RI PT
161
Type A femurs (9 hips, 2.2%) showed a higher femoral revision rate than type B femurs (2 hips, 0.5%)
165
(P=0.034). In type A femurs, reasons for stem revision were PFF (6 hips, 66.7%), aseptic loosening (2 hips,
166
22.2%), and infection (1 hip, 11.1%). All type B femurs were revised because of the occurrence of PFFs.
167
Stem survivorship
M AN U
SC
164
Kaplan-Meier curves for stem survivorship with the end point of revision for any reason are shown in
169
Figures 6 and 7. Matched type A femurs (97.8%) showed a lower stem survival rate than matched type B
170
femurs (99.5%) at a mean follow-up of 7 years (log rank, P=0.041). Including unmatched cases, the overall
171
stem survival rate was 98.5% (type A femurs, 98.0%; type B femurs, 99.5%; and type C femurs, 96.1%).
172
173
Discussion
EP
TE D
168
The prevalence of type A femurs was higher than we expected. Patients’ age seemed to have a substantial
175
impact on the femoral morphology. Nash et al.[21] reported a prevalence rate for type A femurs of 16.9%
176
after analyzing patients with a mean age of 85 years. However, Issa et al.[19] reported a similar prevalence
177
(63.0%) to ours in a study of younger patients (mean age, 53.5 years). Kim et al.[22] reported even higher
178
incidence of type A femurs (85%) after investigating 871 patients with a mean age of 52.9 years. We also
AC C
174
12
ACCEPTED MANUSCRIPT
found a tendency of age progression in patients from type A femurs to types B and C femurs. This result is
180
supported by previous studies about age-related changes in the geometry of the femur based on CT
181
findings[23,24]. Another explanation for this result is that we studied on Asian patients. Several studies have
182
reported a smaller canal diameter in femurs of Asian women[25,26].
RI PT
179
PFFs following THA are devastating complications associated with cementless tapered wedge stems.
184
There have been conflicting reports regarding the risk of periprosthetic fractures in femurs with thick cortices
185
and narrow canals[12-15,18]. This discrepancy might have resulted from differences in the patient
186
characteristics. PFFs frequently occur in women of advanced age with compromised bone quality. Patients
187
with type A femurs were usually younger and had a higher bone mineral density (BMD)[4]. Without
188
considering the bias caused by patient factors, the result would be affected by the bone quality rather than the
189
geometry of the femur. We were unable to obtain data on preoperative BMD; thus, we focused more on the
190
pure dimensional difference of the femur by matching age, sex, BMI, and diagnosis.
EP
TE D
M AN U
SC
183
From the geometric viewpoint, type A femurs would not be a favorable bone type for the broaching
192
procedure. A sharp narrowing of the proximal canal carries a higher risk of intraoperative fracture during
193
impaction of the implant. To obtain an interface that fits in the proximal femur, the operator would have to
194
give a higher stress to the narrow canal. In this situation, a rectangular cross-sectional shape of the implant
195
can cause a sudden increase in hoop tension on the cortex due to the wedge effect[14], which may eventually
196
lead to a fracture. If the implant is oversized or malpositioned, the stress can be even greater. Postoperatively,
AC C
191
13
ACCEPTED MANUSCRIPT
a tight diaphyseal fixation[6,7] may contribute to an increased hoop stress when the sudden rotational or
198
axial force is delivered before sufficient osteointegration is made. There is also a possibility of overlooking
199
intraoperative damage on the cortex that progresses to an obvious fracture line caused by minor stresses. In
200
this study, most type A fractures were caused by low energy trauma, and they frequently occurred before 3
201
months postoperatively. However, more biomechanical evidence is needed to support this hypothesis.
RI PT
197
In the present study, more radiolucent lines were found in type A femurs than in type B femurs. These
203
lines were typically found in the proximal zones (Gruen zones 1, 7, 8, and 14) as shown in figures 4C and D.
204
This can serve as evidence for the mismatch between the metaphysis and diaphysis of the femur[6,7,27,28].
205
During stem insertion, tight distal fixation can prevent ideal compaction of cancellous bone in the
206
metaphyseal region. Additionally, this phenomenon is partially due to undersizing of the implant[29].
207
Operators tend to choose smaller stems to prevent causing intraoperative fracture of narrow femoral
208
canals[14]. Wangen et al.[28] reported 22 cases (18%) of radiolucent lines in 93 tapered wedge stems.
209
However, they could not find a correlation with the hip score, pain, function, and loosening rate. Probably
210
because of a larger volume, we found inferior clinical scores and more thigh pain in patients with radiolucent
211
lines than in those without radiolucent lines. We performed 2 femoral revisions in patients with a
212
combination of radiolucency and progressive thigh pain. Micromotion in the proximal region can be
213
generated by bending moment when the axial loading is delivered[6]. This can cause thigh pain, especially
214
when a higher weight is applied. Herein, we found several patients with proximal radiolucency complaining
AC C
EP
TE D
M AN U
SC
202
14
ACCEPTED MANUSCRIPT
of thigh pain only when they were lifting a heavy object. Some other patients experienced varying degrees of
216
pain caused during the few initial steps when they were walking. Longer follow-ups will be necessary to
217
clarify whether the symptoms in these patients will eventually require additional revisions.
RI PT
215
The current study has several important take-away messages. A type A femur is not a rare type of femur
219
among young individuals, and it may not be a favorable candidate for tapered wedge stems. There are
220
possibilities of unexpected early failures and unpleasant clinical outcomes. For this type of femur, we
221
recommend that surgeons pay more attention when choosing, templating, sizing, and implanting tapered
222
wedge stems. Experience with various femoral implants would help determine the best suitable implant. A
223
distal reaming technique can also be helpful to reduce the proximal/distal discrepancy and ensure optimal
224
metaphyseal filling of the implant. Although not included in this series, we performed distal reaming on a
225
few recent cases with narrow femoral canal, especially when a tight distal fixation was perceived before
226
reaching the desired size of the broach. However, a caution should be made not to ream out the host bone
227
excessively and disturb natural contact area of tapered wedge stems.
EP
TE D
M AN U
SC
218
To the best of our knowledge, this study is the first matched comparative study about the relationship
229
between the femoral bone type and stem survivorship. The strengths of this study include the large volume,
230
completeness of clinical and radiographic data, reduction of confounding factors, and homogeneity of the
231
implant choice.
AC C
228
15
ACCEPTED MANUSCRIPT
There are also some limitations to the present study. First, this was a retrospective study. However, a
233
prospective randomization would be not feasible because of the inherent nature of the femoral geometry.
234
Second, the femoral geometry was classified using only the Dorr classification. Although this classification
235
is simple and widely accepted, it is based on plain radiographs, which can be inaccurate, and it does not
236
reflect multiple aspects of femoral geometry. Future studies using more precise modalities, such as CT, will
237
be needed. Third, a large number of young patients with osteonecrosis were included. These patient
238
characteristics might have affected the overall outcome of our study. Fourth, the mean follow-up period of 7
239
years was relatively short, particularly for a cohort of young patients. Follow-up studies with a longer
240
observation period will be necessary for a long-term comparison. Lastly, this was a single-center study, in
241
which all operations were performed by a single senior surgeon using the same type of implant. There may
242
be a surgeon factor as well as an implant factor that affected the current study’s result.
TE D
M AN U
SC
RI PT
232
In conclusion, this study showed a lower stem survivorship and higher complication rates following THAs
244
using a cementless tapered wedge stem in Dorr type A femurs than those performed in type B femurs. The
245
most common reason for stem revision in type A femurs was the periprosthetic femoral fracture.
247
AC C
246
EP
243
Conflict of interest: None.
248
16
ACCEPTED MANUSCRIPT
249
References
250
1.
251
arthroplasty. J Bone Joint Surg Am 2011;93:500. 2.
253 254
McLaughlin JR, Lee KR. Uncemented total hip arthroplasty using a tapered femoral component in obese patients: an 18-27 year follow-up study. J Arthroplasty 2014;29:1365.
3.
RI PT
252
Khanuja HS, Vakil JJ, Goddard MS, et al. Cementless femoral fixation in total hip
Grant TW, Lovro LR, Licini DJ, et al. Cementless Tapered Wedge Femoral Stems Decrease
255
Subsidence in Obese Patients Compared to Traditional Fit-and-Fill Stems. J Arthroplasty
256
2017;32:891.
257
4.
Oba M, Inaba Y, Kobayashi N, et al. Effect of femoral canal shape on mechanical stress distribution and adaptive bone remodelling around a cementless tapered-wedge stem. Bone
259
Joint Res 2016;5:362. 5.
261 262
Dorr LD, Faugere MC, Mackel AM, et al. Structural and cellular assessment of bone quality of proximal femur. Bone 1993;14:231.
6.
M AN U
260
SC
258
Ishii S, Homma Y, Baba T, et al. Does the Canal Fill Ratio and Femoral Morphology of Asian
263
Females Influence Early Radiographic Outcomes of Total Hip Arthroplasty With an
264
Uncemented Proximally Coated, Tapered-Wedge Stem? J Arthroplasty 2016;31:1524. 7.
266 267
Cooper HJ, Jacob AP, Rodriguez JA. Distal fixation of proximally coated tapered stems may predispose to a failure of osteointegration. J Arthroplasty 2011;26:78.
8.
Colacchio ND, Robbins CE, Aghazadeh MS, et al. Total Hip Intraoperative Femur Fracture:
TE D
265
268
Do the Design Enhancements of a Second-Generation Tapered-Wedge Stem Reduce the
269
Incidence? J Arthroplasty 2017;32:3163.
270
9.
Lim SJ, Lee KJ, Min BW, et al. High incidence of stem loosening in association with periprosthetic femur fractures in previously well-fixed cementless grit-blasted tapered-
272
wedge stems. Int Orthop 2015;39:1689.
273
10.
EP
271
Song JSA, Dillman D, Wilson D, et al. Higher periprosthetic fracture rate associated with use of modern uncemented stems compared to cemented stems in femoral neck fractures. Hip
275
Int 2018:1120700018772291.
276
11.
AC C
274
Zhang Z, Zhuo Q, Chai W, et al. Clinical characteristics and risk factors of periprosthetic
277
femoral fractures associated with hip arthroplasty: A retrospective study. Medicine
278
(Baltimore) 2016;95:e4751.
279
12.
Carli AV, Negus JJ, Haddad FS. Periprosthetic femoral fractures and trying to avoid them:
280
what is the contribution of femoral component design to the increased risk of periprosthetic
281
femoral fracture? Bone Joint J 2017;99-b:50.
282
13.
Gromov K, Bersang A, Nielsen CS, et al. Risk factors for post-operative periprosthetic
283
fractures following primary total hip arthroplasty with a proximally coated double-tapered
284
cementless femoral component. Bone Joint J 2017;99-b:451.
17
ACCEPTED MANUSCRIPT 285
14.
Bonnin MP, Neto CC, Aitsiselmi T, et al. Increased incidence of femoral fractures in small
286
femurs and women undergoing uncemented total hip arthroplasty - why? Bone Joint J
287
2015;97-b:741. 15.
290
intraoperative fracture of the femoral bone. Ortop Traumatol Rehabil 2010;12:237. 16.
291 292
components: a radiographic analysis of loosening. Clin Orthop Relat Res 1979:17. 17.
293 294
Engh CA, Bobyn JD, Glassman AH. Porous-coated hip replacement. The factors governing bone ingrowth, stress shielding, and clinical results. J Bone Joint Surg Br 1987;69:45.
18.
295 296
Gruen TA, McNeice GM, Amstutz HC. "Modes of failure" of cemented stem-type femoral
RI PT
289
Dorman T, Sibinski M, Kmiec K, et al. Cementless hip joint alloplasty complicated by
Duncan CP, Masri BA. Fractures of the femur after hip replacement. Instr Course Lect 1995;44:293.
19.
SC
288
Issa K, Stroh AD, Mont MA, et al. Effect of bone type on clinical and radiographic outcomes of a proximally-coated cementless stem in primary total hip arthroplasties. J Orthop Res
298
2014;32:1214.
299
20.
M AN U
297
Dossick PH, Dorr LD, Gruen T, et al. Techniques for preoperative planning and
300
postoperative evaluation of noncemented hip arthroplasty. Techniques in Orthopaedics
301
1991;6:1.
302
21.
Nash W, Harris A. The Dorr type and cortical thickness index of the proximal femur for predicting peri-operative complications during hemiarthroplasty. J Orthop Surg (Hong Kong)
304
2014;22:92.
305
22.
306 307
TE D
303
Kim YH, Park JW, Kim JS. Long-Term Results of Third-Generation Ceramic-on-Ceramic Bearing Cementless Total Hip Arthroplasty in Young Patients. J Arthroplasty 2016;31:2520.
23.
Ito M, Nakata T, Nishida A, et al. Age-related changes in bone density, geometry and biomechanical properties of the proximal femur: CT-based 3D hip structure analysis in
309
normal postmenopausal women. Bone 2011;48:627.
310
24.
EP
308
Yuen KW, Kwok TC, Qin L, et al. Characteristics of age-related changes in bone compared between male and female reference Chinese populations in Hong Kong: a pQCT study. J
312
Bone Miner Metab 2010;28:672.
313
25.
314 315
26.
27.
Lass R, Kolb A, Skrbensky G, et al. A cementless hip system with a new surface for osseous integration. Int Orthop 2014;38:703.
28.
320 321
Umer M, Sepah YJ, Khan A, et al. Morphology of the proximal femur in a Pakistani population. J Orthop Surg (Hong Kong) 2010;18:279.
318 319
Cho HJ, Kwak DS, Kim IB. Morphometric Evaluation of Korean Femurs by Geometric Computation: Comparisons of the Sex and the Population. Biomed Res Int 2015;2015:730538.
316 317
AC C
311
Wangen H, Nordsletten L, Boldt JG, et al. The Corail stem as a reverse hybrid survivorship and x-ray analysis at 10 years. Hip Int 2017;27:354.
29.
Dundon JM, Felberbaum DL, Long WJ. Femoral stem size mismatch in women undergoing
18
ACCEPTED MANUSCRIPT 322
total hip arthroplasty. J Orthop 2018;15:293.
323 324 325 326
RI PT
327 328 329 330 331
SC
332 333 334
M AN U
335 336 337 338 339 340
344 345 346 347 348 349 350 351 352
EP
343
AC C
342
TE D
341
353 354 355 356 357 358
19
ACCEPTED MANUSCRIPT Legends to figures
360
Fig. 1. A double tapered wedge stem with a grit-blasted surface treatment. (Bencox; Corentec, Cheon-An,
361
South Korea)
362
Fig. 2. Radiographic features of different bone types by the Dorr classification (type A, B and C)
363
Fig. 3. Determination of bone types using femoral canal width (FW) and calcar width (CW). The value of
364
calcar-to-canal ratio (FW/CW) was used for the classification.
365
Figs. 4. Postoperative radiographs of a 49-year old female with type A femur who had undergone THA due
366
to secondary osteoarthritis. Figs. 4-A and 4-B Two-month postoperative images show minimal proximal
367
radiolucent lines. Figs. 4-C and 4-D More obvious proximal radiolucent lines were found with hypertrophic
368
distal cortices 65 months after the index operation.
369
Figs. 5. A 65-year old man with type A femur underwent THA owing to secondary osteoarthritis. Fig. 5-A
370
Immediate postoperative radiograph demonstrates stable stem fixation. Fig. 5-B Femoral radiograph shows
371
periprosthetic fracture (Vancouver B2) resulted from a minor trauma 1 month after the surgery. Fig. 5-C
372
Isolated stem revision was performed using a distal fixating modular stem.
373
Fig. 6. Stem survival curves for matched type A and B femurs (log rank, P=0.041).
374
Fig. 7. Stem survival curves for type A, B and C femurs (log rank, P=0.047).
AC C
EP
TE D
M AN U
SC
RI PT
359
20
ACCEPTED MANUSCRIPT Legends to figures
2
Fig. 1. A double tapered wedge stem with a grit-blasted surface treatment. (Bencox; Corentec, Cheon-An,
3
South Korea)
4
Fig. 2. Radiographic features of different bone types by the Dorr classification (type A, B and C)
5
Fig. 3. Determination of bone types using femoral canal width (FW) and calcar width (CW). The value of
6
calcar-to-canal ratio (FW/CW) was used for the classification.
7
Figs. 4. Postoperative radiographs of a 49-year old female with type A femur who had undergone THA due
8
to secondary osteoarthritis. Figs. 4-A and 4-B Two-month postoperative images show minimal proximal
9
radiolucent lines. Figs. 4-C and 4-D More obvious proximal radiolucent lines were found with hypertrophic
M AN U
SC
RI PT
1
distal cortices 65 months after the index operation.
11
Figs. 5. A 65-year old man with type A femur underwent THA owing to secondary osteoarthritis. Fig. 5-A
12
Immediate postoperative radiograph demonstrates stable stem fixation. Fig. 5-B Femoral radiograph shows
13
periprosthetic fracture (Vancouver B2) resulted from a minor trauma 1 month after the surgery. Fig. 5-C
14
Isolated stem revision was performed using a distal fixating modular stem.
15
Fig. 6. Stem survival curves for matched type A and B femurs (log rank, P=0.041).
16
Fig. 7. Stem survival curves for type A, B and C femurs (log rank, P=0.047).
AC C
EP
TE D
10
1
ACCEPTED MANUSCRIPT Table I. Baseline characteristics of three groups Type A (N = 611)
Type B (N = 427)
Type C (N = 51)
495
340
41
50.6 ± 13.6
54.0 ± 13.8
64.7 ± 12.0
0.001
RI PT
1
0.001
Number of patients Patient age* (yr) Gender† Male
266 (43.6%)
234 (54.7%)
Female
345 (56.4%)
193 (45.3%)
289 (47.4%)
Osteoarthritis
282 (46.1%)
Fracture & Others
40 (6.6%)
Ethnicity† 608 (99.5%)
Others BMI* (kg/m²)
Stem size* (number)
<0.001
16 (31.4%)
162 (38.1%)
25 (49.0%)
50 (11.7%)
10 (19.6%) 0.709 51 (100%)
3 (0.5%)
1 (0.2%)
0 (0%)
24.4 ± 3.4
24.3 ± 3.3
24.1 ± 3.1
0.901
83.8 ± 23.7
85.1 ± 25.1
75.7 ± 21.8
0.520
4.4 ± 1.7
5.6 ± 1.7
6.4 ± 2.0
0.002
EP
Follow-up duration* (mo)
32 (62.7%)
426 (99.8%)
TE D
Asian
215 (50.2%)
M AN U
ONFH‡
19 (37.3%)
SC
Diagnosis†
P Value
*The values are given as the mean and the standard deviation. †The values are given as the number of hips
3
with the percentage in parentheses. ‡ ONFH = Osteonecrosis of femoral head.
4 5
AC C
2
6 7 8 9
1
ACCEPTED MANUSCRIPT Table II. Baseline characteristics of matched two groups Type A (N = 408)
Type B (N = 408)
328
328
53.3 ± 13.7
53.6 ± 13.4
Number of patients Patient age* (yr)
P Value
0.947
RI PT
1
Gender†
1.000
219 (53.7%)
219 (53.7%)
Female
189 (46.3%)
189 (46.3%)
SC
Male
Diagnosis†
0.091
197 (48.3%)
Osteoarthritis
181 (44.4%)
156 (38.2%)
30 (7.4%)
44 (10.8%)
BMI* (kg/m²)
24.3 ± 3.4
24.4 ± 3.3
0.555
Follow-up duration* (mo)
87.8 ± 18.5
85.2 ± 20.5
0.058
5.6 ± 1.8
<0.001
TE D
Fracture & Others
Stem size* (number)
208 (51.0%)
M AN U
ONFH
4.5 ± 1.7
*The values are given as the mean and the standard deviation. †The values are given as the number of hips
3
with the percentage in parentheses. ‡ ONFH = Osteonecrosis of femoral head.
5 6 7 8
AC C
4
EP
2
9 10 11
1
ACCEPTED MANUSCRIPT 1
Table III. Clinical outcomes of matched two groups Type A (n=408)
Type B (n=408)
P Value
Preoperative
45.3 ± 13.1
44.0 ± 14.6
0.176
Postoperative
92.3 ± 9.17
92.9 ± 7.36
0.242
Preoperative
3.11 ± 0.97
3.00 ± 0.95
0.947
Postoperative
6.44 ± 1.28 19 (4.7%)
Postoperative groin pain†
20 (4.9%)
6.46 ± 1.13
0.772
6 (1.5%)
0.008
M AN U
Postoperative thigh pain†
SC
UCLA activity score*
RI PT
Harris hip score*
12 (2.9%)
0.149
2
*The values are given as the mean and the standard deviation. †The values are given as the number of hips
3
with the percentage in parentheses.
7 8 9 10 11 12
EP
6
AC C
5
TE D
4
13 14 15 16
1
ACCEPTED MANUSCRIPT Table IV. Radiographic features of matched two groups Type B (n=408)
P Value
32 (7.8%)
10 (2.5%)
<0.001
30 (7.4%)
7 (1.7%)
0 (0%)
2 (0.5%)
2 (0.5%)
1 (0.2%)
Radiolucent lines Proximal zones only Distal zones only Proximal & distal zones Stem position
<0.001
Neutral
390 (95.6%)
365 (89.5%)
Varus
16 (3.9%)
43 (10.5%)
Valgus
2 (0.5%) 16 (3.9%)
Stem stability
5 6 7 8
0.571 0.026
387 (94.9%)
Fibrous stable
36 (8.8%)
21 (5.1%)
Unstable
2 (0.5%)
0 (0%)
Subsidence
7 (1.7%)
1 (0.2%)
0.069
Heterotrophic ossification
33 (8.1%)
46 (11.3%)
0.124
TE D
370 (90.7%)
The values are given as the number of hips with the percentage in parentheses.
EP
4
13 (3.2%)
Bone ingrowth
AC C
3
0 (0%)
M AN U
Cortical hypertrophy
2
RI PT
Type A (n=408)
SC
1
9 10 11 12
1
ACCEPTED MANUSCRIPT 1
Table V. Clinical outcomes of femurs with and without radiolucency With radiolucency (n=42)
Without radiolucency (n=774)
P Value
Preoperative
42.5 ± 12.7
44.8 ± 13.9
0.299
Postoperative
86.2 ± 15.7
RI PT
Harris hip score*
Preoperative
3.14 ± 0.95
Postoperative
6.00 ± 1.61 13 (31%)
Postoperative groin pain†
7 (16.7%)
3.05 ± 0.96
0.549
6.47 ± 1.18
0.067
12 (1.6%)
<0.001
25 (3.2%)
0.001
M AN U
Postoperative thigh pain†
<0.001
SC
UCLA activity score*
93.0 ± 7.6
2
*The values are given as the mean and the standard deviation. †The values are given as the number of hips
3
with the percentage in parentheses.
7 8 9 10 11 12
EP
6
AC C
5
TE D
4
13 14 15 16
1
ACCEPTED MANUSCRIPT Table VI. Major complications and revision rates of matched two groups Type B (n=408)
P Value
13 (3.2%)
4 (1.0%)
0.027
4 (1.0%)
0 (0%)
0.124
4 (1.0%)
0 (0%)
9 (2.2%)
4 (1.0%)
Vancouver AG
1 (0.2%)
1 (0.2%)
Vancouver B1
1 (0.2%)
Vancouver B2
7 (1.7%)
Periprosthetic femoral fracture Intraoperative Vancouver A2
2 (0.5%)
0 (0%)
0.499
Aseptic cup loosening
2 (0.5%)
1 (0.2%)
1.000
Deep infection
3 (0.7%)
1 (0.2%)
0.624
0 (0%)
0 (0%)
1.000
3 (0.7%)
1 (0.2%)
0.624
9 (2.2%)
2 (0.5%)
0.034
3 (0.7%)
1 (0.2%)
0.624
0 (0%)
1.000
TE D
Dislocation Stem revision
EP
Cup revision Revision of both components
4 5
1 (0.2%)
The values are given as the number of hips with the percentage in parentheses.
AC C
3
2 (0.5%)
Aseptic stem loosening
Nerve palsy
2
0.162
1 (0.2%)
M AN U
Postoperative
RI PT
Type A (n=408)
SC
1
6 7 8
1
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
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
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(18)31117-3
DOI:
https://doi.org/10.1016/j.arth.2018.11.004
Reference:
YARTH 56900
To appear in:
The Journal of Arthroplasty
Received Date: 19 September 2018 Revised Date:
16 October 2018
Accepted Date: 2 November 2018
Please cite this article as: Park C-W, Eun H-J, Oh S-H, Kim H-J, Lim S-J, Park Y-S, Femoral Stem Survivorship in Dorr Type A Femurs Following Total Hip Arthroplasty Using a Cementless Tapered Wedge Stem: A Matched Comparative Study with Type B Femurs, The Journal of Arthroplasty (2018), doi: https://doi.org/10.1016/j.arth.2018.11.004. 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
1
Femoral Stem Survivorship in Dorr Type A Femurs Following Total Hip Arthroplasty Using a
2
Cementless Tapered Wedge Stem: A Matched Comparative Study with Type B Femurs
RI PT
3
Chan-Woo Park, MD, Hyeon-Jun Eun, MD, Sung-Hak Oh, MD, Hyun-Jun Kim, MD, Seung-Jae
5
Lim, MD, Youn-Soo Park, MD
6
Department of Orthopedic Surgery, Samsung Medical Center, Sungkyunkwan University School of
7
Medicine, Seoul, Korea
M AN U
SC
4
8
9
Concise Title: Stem Survivorship in Dorr Type A Femurs Following Primary THA
TE D
10
11
Please address all correspondence to:
13
Youn-Soo Park, MD
14
Department of Orthopedic Surgery, Samsung Medical Center
15
81 Irwon-ro, Gangnam-gu
16
Seoul 06351
17
South Korea.
18
Phone: 82-2-3410-3504
19
Fax: 82-2-3410-0061
20
Email: [email protected]
AC C
EP
12
1
ACCEPTED MANUSCRIPT
1
Femoral Stem Survivorship in Dorr Type A Femurs Following Total Hip Arthroplasty Using a
2
Cementless Tapered Wedge Stem: A Matched Comparative Study with Type B Femurs
AC C
EP
TE D
M AN U
SC
RI PT
3
1
ACCEPTED MANUSCRIPT
4
Abstract
5
Background: There is a lack of understanding on relationship between the femoral geometry and outcomes of total hip
7
arthroplasty (THA). We investigated clinical and radiographic outcomes of THA using a cementless tapered
8
wedge stem in patients with Dorr type A proximal femoral morphology, and compared with those of type B
9
femurs at a minimum follow-up of 5 years.
SC
Methods:
M AN U
10
RI PT
6
We analyzed 1089 hips (876 patients) that underwent THA using an identical cementless tapered wedge
12
stem. We divided all femurs into 3 types (Dorr type A, B and C). Type A and B femurs were statistically
13
matched with age, gender, body mass index and diagnosis by using propensity score matching. Clinical,
14
radiographic results and stem survivorship were compared between the matched two groups.
15
Results:
EP
TE D
11
A total of 611 femurs (56%) were classified as type A, 427 (39%) as type B, and 51 (5%) as type C. More
17
radiolucent lines around femoral stems were found in type A femurs (7.8%) than in type B femurs (2.5%)
18
(P<0.001). Patients with radiolucency showed worse Harris hip score (86.2 points) compared to those
19
without radiolucency (93.0 points) (P<0.001). The stem survivorship of type A femur (97.8%) was lower
20
than that of type B femur (99.5%) (P=0.041). The reasons for femoral revision in type A femurs were
21
periprosthetic fracture (67%), aseptic loosening (22%), and deep infection (11%).
AC C
16
2
ACCEPTED MANUSCRIPT
23
24
Conclusions: This study showed a higher rate of complications following THAs using a cementless tapered wedge stem in Dorr type A femurs than those performed in type B femurs.
RI PT
22
25
Keywords: Dorr classification, tapered wedge stem, periprosthetic femoral fracture, stem survivorship, total
27
hip arthroplasty
SC
26
M AN U
28
AC C
EP
TE D
29
3
ACCEPTED MANUSCRIPT
Abbreviations
31
THA, total hip arthroplasty; PFF, periprosthetic femoral fracture; BMI, body mass index; HHS, Harris hip
32
score; UCLA, University of California at Los Angeles; ONFH, osteonecrosis of the femoral head; BMD,
33
bone mineral density
RI PT
30
AC C
EP
TE D
M AN U
SC
34
4
ACCEPTED MANUSCRIPT
35
Introduction The use of cementless tapered wedge stems with a broaching procedure is common in contemporary total
37
hip arthroplasty (THA). Because of a recent trend in sparing the maximum bone stock and achieving
38
proximal fixation, tapered wedge stems are gaining more popularity[1-4]. However, with this type of stem,
39
the shape of the individual proximal femur is more important than with the traditional fit-and-fill stem.
40
Tapered wedge stems require tight initial fixation at the proximal metaphysis mediolaterally and three-point
41
fixation anteroposteriorly without the help of reaming[1,4].
M AN U
SC
RI PT
36
Anatomic variations in proximal femurs exist. The geometry of the proximal femur was previously
43
classified into 3 types by Dorr et al.[5]. Although various shapes of femoral implants have been introduced,
44
most were developed based on the femurs with an average canal dimension close to that of type B femurs. As
45
a result, each implant does not perfectly match all types of femur. For type A femurs, which have thick
46
cortices with champagne flute canals, this dimensional mismatch can be more severe. There is a high chance
47
of tight distal fixation with insufficient metaphyseal contact with this type of femur[6,7].
EP
TE D
42
There are also concerns of increased risks of periprosthetic femoral fracture (PFF) in THA with cementless
49
implants[8-11]. Recently, in a systemic review, Carli et al.[12] reported a three-fold increase in rates of PFF
50
with cementless tapered wedge stems. Although type A femurs have been considered safe in terms of
51
PFFs[13], recent studies have reported a high incidence of PFF in femurs with narrow canals as well[14,15].
52
AC C
48
Despite its significance, there is a lack of understanding regarding the relationship between the femoral
5
ACCEPTED MANUSCRIPT
geometry and outcomes of the femoral implant. We questioned whether a cementless tapered wedge stem for
54
THA would be as safe to use in Dorr type A femurs as in ordinary type B femurs. In this study, we compared
55
the clinical and radiographic outcomes of THA using a tapered wedge stem between Dorr types A and
56
B femurs after a minimum follow-up of 5 years.
RI PT
53
Materials and Methods
59
Patient cohort
M AN U
58
SC
57
This study was conducted under the approval of our institutional ethics committee review board. Between
61
January 2007 and December 2012, a total of 2108 consecutive THAs (1682 patients) were performed by a
62
single senior surgeon in our institution. We identified 1013 patients (1264 hips) who underwent primary
63
cementless THA utilizing an identical double tapered wedge stem. Femurs with any radiographic evidence of
64
proximal deformity below the level of anatomical neck were excluded, and patients with hereditary disease
65
affecting the skeletal system were also excluded from the study.
EP
TE D
60
Of 1013 patients, 27 patients (32 hips) died for reasons unrelated to the surgery, and 76 (107 hips) were
67
lost to follow-up before reaching a minimum of 5 years. Thirty-four patients (36 hips) were excluded
68
because of femoral deformities caused by infection, tumor, trauma, genetic disorder and previous surgeries.
69
Full clinical and radiographic data were available for 876 patients (1089 hips) who were enrolled in this
70
study.
AC C
66
6
ACCEPTED MANUSCRIPT
71
Surgical characteristics All surgeries were performed through the anterolateral approach by a single experienced surgeon.
73
Preoperative templating was conducted in all cases using standard radiographs with magnification markers.
74
Femoral preparations were performed by only the broaching technique.
RI PT
72
The same prosthesis was used in all surgeries. The femoral implant was a cementless Bencox stem
76
(Corentec, Cheon-An, South Korea) made of titanium alloy (Fig. 1). The geometry of this stem is similar to
77
the CLS Spotorno stem (Zimmer, Warsaw, IN, USA), which has a double tapered wedge design with a
78
rectangular cross-section and a slim diaphyseal part. The stem has a neck-shaft angle of 135° and a taper of
79
12-14 mm in diameter with a circular cross-section. Various stem sizes are available, ranging from 1 to 13.
80
The surface was treated by grit-blasting and the micro-arc oxidation method. This stem can be classified as
81
type 3C according to the classification suggested by Mont et al.[1]. The acetabular component was a
82
cementless Bencox cup (Corentec), which is made of titanium alloy and coated using the plasma spray
83
method. The operator used a ceramic-on-ceramic bearing in all cases.
84
Clinical assessments
AC C
EP
TE D
M AN U
SC
75
85
The Harris hip scores (HHSs) and University of California at Los Angeles (UCLA) activity scores were
86
recorded preoperatively. Clinical outcome was evaluated at each outpatient visit scheduled at 2, 6, and 12
87
months postoperatively and then annually. If patients experienced any postoperative pain, the location of
88
pain (e.g., the thigh, trochanteric area, groin, and buttocks) was recorded in their medical record. We
7
ACCEPTED MANUSCRIPT
identified major postoperative complications such as PFF, dislocation, aseptic loosening, infection, nerve
90
palsy, and any reoperations retrospectively.
91
Radiographic measurements
RI PT
89
All radiographs were measured twice each by 2 independent orthopedic surgeons who were fellows of hip
93
arthroplasty and had not participated in the surgery. The measurement was performed after removing all
94
identifying information of the patients. For continuous variables (e.g., the canal width), the average values
95
were used. For categorical variables, the value of the majority was used. The investigators used a picture
96
archiving and communication system (Centricity Enterprise Web V3.0, GE Medical Systems, Barrington, IL).
97
The radiographs obtained at the last follow-up were used for postoperative radiographic comparison. We
98
identified radiolucency of >1 mm around the femoral stem by using the Gruen zones[16] of the femur. Gruen
99
zones 1, 7, 8, and 14 were considered as proximal zones, and zones 3, 4, 5, 10, 11, and 12 as distal zones.
100
The alignment of the femoral component was classified as varus (>3°), valgus (>3°), or neutral. Subsidence
101
of the femoral component >4 mm was considered clinically meaningful. The mode of fixation of the stem
102
was categorized as bone ingrown, fibrous stable, or unstable using Engh et al.’s classification[17], and PFFs
103
were categorized based on the Vancouver classification[18].
104
Determination of bone types
AC C
EP
TE D
M AN U
SC
92
105
Femoral geometries were categorized according to the classification system of Dorr et al.[5] by using
106
standard preoperative anteroposterior radiographs. The calcar width was measured at the middle level of the
8
ACCEPTED MANUSCRIPT
lesser trochanter, and the canal width was measured at 10 cm below the lesser trochanter (Fig. 2). The
108
calcar-to-canal ratio was used to determine the bone types[19,20]. Femurs with a ratio of 0-0.500 were
109
considered as type A, 0.501-0.750 as type B, and 0.751-1 as type C (Fig. 3). The intraclass correlation
110
coefficient (ICC) was used to evaluate the reliability of the canal width ratio. The ICC for intraobserver
111
agreement was 91.4% (95% confidence interval [CI], 88.5%-94.3%), and ICC for interobserver agreement
112
was 88.7% (95% CI, 85.2%-92.2%).
SC
Statistical analysis
M AN U
113
RI PT
107
To reduce possible confounding factors, groups with types A and B femurs were statistically matched by
115
age, sex, body mass index (BMI), follow-up period, and diagnosis. One-way analysis of variance was used to
116
compare continuous variables, and the chi-square or Fisher’s exact test was used to compare categorical
117
variables among 3 groups. Groups with types A and B femurs were matched using propensity score matching,
118
as generated by logistic regression analysis. The paired t-test was used to analyze improvements in clinical
119
scores for each group. To compare 2 groups, the Student’s t-test and Wilcoxon rank sum test were used for
120
continuous variables. Survivorship analysis was performed with the Kaplan-Meier estimator using the end
121
point of stem revision for any reason. Statistical analysis was performed using SPSS statistics software,
122
version 24.0 (IBM Corp., Armonk, NY, USA), and P<0.05 was considered statistically significant.
AC C
EP
TE D
114
123
124
Results
9
ACCEPTED MANUSCRIPT
125
Patient characteristics There were 418 men (519 hips, 47.7%) and 458 women (570 hips, 52.3%) with a mean age of 52.5 years
127
(range, 16-81 year) at the index operation. The mean BMI was 24.4 kg/m2 (range, 15.3-37.9 kg/m2). Primary
128
diagnoses for THA included osteonecrosis of the femoral head (ONFH) (520 hips, 47.8%), osteoarthritis
129
(469 hips, 43.1%), and inflammatory arthritis, fracture, and others (100 hips, 9.2%). The mean follow-up
130
period was 84 months (range, 60-133 months).
131
Characteristics of bone types
M AN U
SC
RI PT
126
Among 1089 hips, 611 femurs (56.0%) were classified as type A, 427 (39.3%) as type B, and 51 (4.7%) as
133
type C. There were significant differences in age, sex, and diagnosis between the 3 groups (Table I). Patients
134
with type A femurs were the youngest (mean, 50.6 years), and those with type C femurs were the oldest
135
(mean, 64.7 years). There was a slight male predominance (54.7 %) in the type B femur group, whereas the
136
other groups showed female predominance. The most common diagnosis for patients with type C femurs was
137
osteoarthritis, whereas that for patients with the other 2 types was ONFH. There were also differences
138
between the 3 groups in implant size selection. After one-to-one propensity score matching was performed
139
for type A and B femurs, no meaningful difference was found in all patient characteristics (Table II). All
140
clinical and radiographic results, complications, and revision rates were compared between the matched 2
141
groups.
142
Clinical outcomes
AC C
EP
TE D
132
10
ACCEPTED MANUSCRIPT
In both groups, clinical scores improved significantly after THA (Table III). For type A femurs, the mean
144
HHS increased from 45.1 points preoperatively to 92.3 points postoperatively. The mean UCLA score
145
improved from 3.16 points to 6.44 points. After THA, type B femurs also showed improvement in HHSs
146
(from 44.0 to 92.9) and UCLA scores (from 3.10 to 6.46). Patients with type A femurs experienced more
147
thigh pain than those with type B femurs (P=0.008). No significant differences were found regarding
148
postoperative HHSs and UCLA scores at the final follow-up.
149
Radiographic features
M AN U
SC
RI PT
143
More femoral radiolucent lines were found in type A femurs than in type B femurs (7.8% versus [vs.]
151
2.5%) (P<0.001) (Table IV). Those were mostly limited to proximal zones (93.8%) (Fig. 4A-D). Stems in
152
type A femurs showed more neutral alignments than those in type B femurs (95.6% vs. 89.5%) (P<0.001).
153
More radiographic osteointegration was found in type B femurs than in type A femurs (94.9% vs. 90.7%)
154
(P=0.026). The patients with radiolucency felt more thigh pain (31%) and groin pain (16.7%) than those
155
without radiolucency (Table V), and the postoperative HHS was significantly lower in these patients
156
(P<0.001).
157
Complications and revisions
AC C
EP
TE D
150
158
There were more PFFs in patients with type A femurs than in those with type B femurs (3.2% vs. 1.0%)
159
(P=0.027) (Table VI). Of them, 4 fractures were caused intraoperatively, whereas no intraoperative fracture
160
occurred in patients with type B femurs. There were 9 postoperative Vancouver B2 fractures, all of which
11
ACCEPTED MANUSCRIPT
were revised using a distal fixating prosthesis (Fig. 5A-C). Two patients with type A femurs showed
162
radiolucency with persistent thigh pain and underwent revision with a different type of stem. One patient
163
developed periprosthetic infection and required 2-stage revision.
RI PT
161
Type A femurs (9 hips, 2.2%) showed a higher femoral revision rate than type B femurs (2 hips, 0.5%)
165
(P=0.034). In type A femurs, reasons for stem revision were PFF (6 hips, 66.7%), aseptic loosening (2 hips,
166
22.2%), and infection (1 hip, 11.1%). All type B femurs were revised because of the occurrence of PFFs.
167
Stem survivorship
M AN U
SC
164
Kaplan-Meier curves for stem survivorship with the end point of revision for any reason are shown in
169
Figures 6 and 7. Matched type A femurs (97.8%) showed a lower stem survival rate than matched type B
170
femurs (99.5%) at a mean follow-up of 7 years (log rank, P=0.041). Including unmatched cases, the overall
171
stem survival rate was 98.5% (type A femurs, 98.0%; type B femurs, 99.5%; and type C femurs, 96.1%).
172
173
Discussion
EP
TE D
168
The prevalence of type A femurs was higher than we expected. Patients’ age seemed to have a substantial
175
impact on the femoral morphology. Nash et al.[21] reported a prevalence rate for type A femurs of 16.9%
176
after analyzing patients with a mean age of 85 years. However, Issa et al.[19] reported a similar prevalence
177
(63.0%) to ours in a study of younger patients (mean age, 53.5 years). Kim et al.[22] reported even higher
178
incidence of type A femurs (85%) after investigating 871 patients with a mean age of 52.9 years. We also
AC C
174
12
ACCEPTED MANUSCRIPT
found a tendency of age progression in patients from type A femurs to types B and C femurs. This result is
180
supported by previous studies about age-related changes in the geometry of the femur based on CT
181
findings[23,24]. Another explanation for this result is that we studied on Asian patients. Several studies have
182
reported a smaller canal diameter in femurs of Asian women[25,26].
RI PT
179
PFFs following THA are devastating complications associated with cementless tapered wedge stems.
184
There have been conflicting reports regarding the risk of periprosthetic fractures in femurs with thick cortices
185
and narrow canals[12-15,18]. This discrepancy might have resulted from differences in the patient
186
characteristics. PFFs frequently occur in women of advanced age with compromised bone quality. Patients
187
with type A femurs were usually younger and had a higher bone mineral density (BMD)[4]. Without
188
considering the bias caused by patient factors, the result would be affected by the bone quality rather than the
189
geometry of the femur. We were unable to obtain data on preoperative BMD; thus, we focused more on the
190
pure dimensional difference of the femur by matching age, sex, BMI, and diagnosis.
EP
TE D
M AN U
SC
183
From the geometric viewpoint, type A femurs would not be a favorable bone type for the broaching
192
procedure. A sharp narrowing of the proximal canal carries a higher risk of intraoperative fracture during
193
impaction of the implant. To obtain an interface that fits in the proximal femur, the operator would have to
194
give a higher stress to the narrow canal. In this situation, a rectangular cross-sectional shape of the implant
195
can cause a sudden increase in hoop tension on the cortex due to the wedge effect[14], which may eventually
196
lead to a fracture. If the implant is oversized or malpositioned, the stress can be even greater. Postoperatively,
AC C
191
13
ACCEPTED MANUSCRIPT
a tight diaphyseal fixation[6,7] may contribute to an increased hoop stress when the sudden rotational or
198
axial force is delivered before sufficient osteointegration is made. There is also a possibility of overlooking
199
intraoperative damage on the cortex that progresses to an obvious fracture line caused by minor stresses. In
200
this study, most type A fractures were caused by low energy trauma, and they frequently occurred before 3
201
months postoperatively. However, more biomechanical evidence is needed to support this hypothesis.
RI PT
197
In the present study, more radiolucent lines were found in type A femurs than in type B femurs. These
203
lines were typically found in the proximal zones (Gruen zones 1, 7, 8, and 14) as shown in figures 4C and D.
204
This can serve as evidence for the mismatch between the metaphysis and diaphysis of the femur[6,7,27,28].
205
During stem insertion, tight distal fixation can prevent ideal compaction of cancellous bone in the
206
metaphyseal region. Additionally, this phenomenon is partially due to undersizing of the implant[29].
207
Operators tend to choose smaller stems to prevent causing intraoperative fracture of narrow femoral
208
canals[14]. Wangen et al.[28] reported 22 cases (18%) of radiolucent lines in 93 tapered wedge stems.
209
However, they could not find a correlation with the hip score, pain, function, and loosening rate. Probably
210
because of a larger volume, we found inferior clinical scores and more thigh pain in patients with radiolucent
211
lines than in those without radiolucent lines. We performed 2 femoral revisions in patients with a
212
combination of radiolucency and progressive thigh pain. Micromotion in the proximal region can be
213
generated by bending moment when the axial loading is delivered[6]. This can cause thigh pain, especially
214
when a higher weight is applied. Herein, we found several patients with proximal radiolucency complaining
AC C
EP
TE D
M AN U
SC
202
14
ACCEPTED MANUSCRIPT
of thigh pain only when they were lifting a heavy object. Some other patients experienced varying degrees of
216
pain caused during the few initial steps when they were walking. Longer follow-ups will be necessary to
217
clarify whether the symptoms in these patients will eventually require additional revisions.
RI PT
215
The current study has several important take-away messages. A type A femur is not a rare type of femur
219
among young individuals, and it may not be a favorable candidate for tapered wedge stems. There are
220
possibilities of unexpected early failures and unpleasant clinical outcomes. For this type of femur, we
221
recommend that surgeons pay more attention when choosing, templating, sizing, and implanting tapered
222
wedge stems. Experience with various femoral implants would help determine the best suitable implant. A
223
distal reaming technique can also be helpful to reduce the proximal/distal discrepancy and ensure optimal
224
metaphyseal filling of the implant. Although not included in this series, we performed distal reaming on a
225
few recent cases with narrow femoral canal, especially when a tight distal fixation was perceived before
226
reaching the desired size of the broach. However, a caution should be made not to ream out the host bone
227
excessively and disturb natural contact area of tapered wedge stems.
EP
TE D
M AN U
SC
218
To the best of our knowledge, this study is the first matched comparative study about the relationship
229
between the femoral bone type and stem survivorship. The strengths of this study include the large volume,
230
completeness of clinical and radiographic data, reduction of confounding factors, and homogeneity of the
231
implant choice.
AC C
228
15
ACCEPTED MANUSCRIPT
There are also some limitations to the present study. First, this was a retrospective study. However, a
233
prospective randomization would be not feasible because of the inherent nature of the femoral geometry.
234
Second, the femoral geometry was classified using only the Dorr classification. Although this classification
235
is simple and widely accepted, it is based on plain radiographs, which can be inaccurate, and it does not
236
reflect multiple aspects of femoral geometry. Future studies using more precise modalities, such as CT, will
237
be needed. Third, a large number of young patients with osteonecrosis were included. These patient
238
characteristics might have affected the overall outcome of our study. Fourth, the mean follow-up period of 7
239
years was relatively short, particularly for a cohort of young patients. Follow-up studies with a longer
240
observation period will be necessary for a long-term comparison. Lastly, this was a single-center study, in
241
which all operations were performed by a single senior surgeon using the same type of implant. There may
242
be a surgeon factor as well as an implant factor that affected the current study’s result.
TE D
M AN U
SC
RI PT
232
In conclusion, this study showed a lower stem survivorship and higher complication rates following THAs
244
using a cementless tapered wedge stem in Dorr type A femurs than those performed in type B femurs. The
245
most common reason for stem revision in type A femurs was the periprosthetic femoral fracture.
247
AC C
246
EP
243
Conflict of interest: None.
248
16
ACCEPTED MANUSCRIPT
249
References
250
1.
251
arthroplasty. J Bone Joint Surg Am 2011;93:500. 2.
253 254
McLaughlin JR, Lee KR. Uncemented total hip arthroplasty using a tapered femoral component in obese patients: an 18-27 year follow-up study. J Arthroplasty 2014;29:1365.
3.
RI PT
252
Khanuja HS, Vakil JJ, Goddard MS, et al. Cementless femoral fixation in total hip
Grant TW, Lovro LR, Licini DJ, et al. Cementless Tapered Wedge Femoral Stems Decrease
255
Subsidence in Obese Patients Compared to Traditional Fit-and-Fill Stems. J Arthroplasty
256
2017;32:891.
257
4.
Oba M, Inaba Y, Kobayashi N, et al. Effect of femoral canal shape on mechanical stress distribution and adaptive bone remodelling around a cementless tapered-wedge stem. Bone
259
Joint Res 2016;5:362. 5.
261 262
Dorr LD, Faugere MC, Mackel AM, et al. Structural and cellular assessment of bone quality of proximal femur. Bone 1993;14:231.
6.
M AN U
260
SC
258
Ishii S, Homma Y, Baba T, et al. Does the Canal Fill Ratio and Femoral Morphology of Asian
263
Females Influence Early Radiographic Outcomes of Total Hip Arthroplasty With an
264
Uncemented Proximally Coated, Tapered-Wedge Stem? J Arthroplasty 2016;31:1524. 7.
266 267
Cooper HJ, Jacob AP, Rodriguez JA. Distal fixation of proximally coated tapered stems may predispose to a failure of osteointegration. J Arthroplasty 2011;26:78.
8.
Colacchio ND, Robbins CE, Aghazadeh MS, et al. Total Hip Intraoperative Femur Fracture:
TE D
265
268
Do the Design Enhancements of a Second-Generation Tapered-Wedge Stem Reduce the
269
Incidence? J Arthroplasty 2017;32:3163.
270
9.
Lim SJ, Lee KJ, Min BW, et al. High incidence of stem loosening in association with periprosthetic femur fractures in previously well-fixed cementless grit-blasted tapered-
272
wedge stems. Int Orthop 2015;39:1689.
273
10.
EP
271
Song JSA, Dillman D, Wilson D, et al. Higher periprosthetic fracture rate associated with use of modern uncemented stems compared to cemented stems in femoral neck fractures. Hip
275
Int 2018:1120700018772291.
276
11.
AC C
274
Zhang Z, Zhuo Q, Chai W, et al. Clinical characteristics and risk factors of periprosthetic
277
femoral fractures associated with hip arthroplasty: A retrospective study. Medicine
278
(Baltimore) 2016;95:e4751.
279
12.
Carli AV, Negus JJ, Haddad FS. Periprosthetic femoral fractures and trying to avoid them:
280
what is the contribution of femoral component design to the increased risk of periprosthetic
281
femoral fracture? Bone Joint J 2017;99-b:50.
282
13.
Gromov K, Bersang A, Nielsen CS, et al. Risk factors for post-operative periprosthetic
283
fractures following primary total hip arthroplasty with a proximally coated double-tapered
284
cementless femoral component. Bone Joint J 2017;99-b:451.
17
ACCEPTED MANUSCRIPT 285
14.
Bonnin MP, Neto CC, Aitsiselmi T, et al. Increased incidence of femoral fractures in small
286
femurs and women undergoing uncemented total hip arthroplasty - why? Bone Joint J
287
2015;97-b:741. 15.
290
intraoperative fracture of the femoral bone. Ortop Traumatol Rehabil 2010;12:237. 16.
291 292
components: a radiographic analysis of loosening. Clin Orthop Relat Res 1979:17. 17.
293 294
Engh CA, Bobyn JD, Glassman AH. Porous-coated hip replacement. The factors governing bone ingrowth, stress shielding, and clinical results. J Bone Joint Surg Br 1987;69:45.
18.
295 296
Gruen TA, McNeice GM, Amstutz HC. "Modes of failure" of cemented stem-type femoral
RI PT
289
Dorman T, Sibinski M, Kmiec K, et al. Cementless hip joint alloplasty complicated by
Duncan CP, Masri BA. Fractures of the femur after hip replacement. Instr Course Lect 1995;44:293.
19.
SC
288
Issa K, Stroh AD, Mont MA, et al. Effect of bone type on clinical and radiographic outcomes of a proximally-coated cementless stem in primary total hip arthroplasties. J Orthop Res
298
2014;32:1214.
299
20.
M AN U
297
Dossick PH, Dorr LD, Gruen T, et al. Techniques for preoperative planning and
300
postoperative evaluation of noncemented hip arthroplasty. Techniques in Orthopaedics
301
1991;6:1.
302
21.
Nash W, Harris A. The Dorr type and cortical thickness index of the proximal femur for predicting peri-operative complications during hemiarthroplasty. J Orthop Surg (Hong Kong)
304
2014;22:92.
305
22.
306 307
TE D
303
Kim YH, Park JW, Kim JS. Long-Term Results of Third-Generation Ceramic-on-Ceramic Bearing Cementless Total Hip Arthroplasty in Young Patients. J Arthroplasty 2016;31:2520.
23.
Ito M, Nakata T, Nishida A, et al. Age-related changes in bone density, geometry and biomechanical properties of the proximal femur: CT-based 3D hip structure analysis in
309
normal postmenopausal women. Bone 2011;48:627.
310
24.
EP
308
Yuen KW, Kwok TC, Qin L, et al. Characteristics of age-related changes in bone compared between male and female reference Chinese populations in Hong Kong: a pQCT study. J
312
Bone Miner Metab 2010;28:672.
313
25.
314 315
26.
27.
Lass R, Kolb A, Skrbensky G, et al. A cementless hip system with a new surface for osseous integration. Int Orthop 2014;38:703.
28.
320 321
Umer M, Sepah YJ, Khan A, et al. Morphology of the proximal femur in a Pakistani population. J Orthop Surg (Hong Kong) 2010;18:279.
318 319
Cho HJ, Kwak DS, Kim IB. Morphometric Evaluation of Korean Femurs by Geometric Computation: Comparisons of the Sex and the Population. Biomed Res Int 2015;2015:730538.
316 317
AC C
311
Wangen H, Nordsletten L, Boldt JG, et al. The Corail stem as a reverse hybrid survivorship and x-ray analysis at 10 years. Hip Int 2017;27:354.
29.
Dundon JM, Felberbaum DL, Long WJ. Femoral stem size mismatch in women undergoing
18
ACCEPTED MANUSCRIPT 322
total hip arthroplasty. J Orthop 2018;15:293.
323 324 325 326
RI PT
327 328 329 330 331
SC
332 333 334
M AN U
335 336 337 338 339 340
344 345 346 347 348 349 350 351 352
EP
343
AC C
342
TE D
341
353 354 355 356 357 358
19
ACCEPTED MANUSCRIPT Legends to figures
360
Fig. 1. A double tapered wedge stem with a grit-blasted surface treatment. (Bencox; Corentec, Cheon-An,
361
South Korea)
362
Fig. 2. Radiographic features of different bone types by the Dorr classification (type A, B and C)
363
Fig. 3. Determination of bone types using femoral canal width (FW) and calcar width (CW). The value of
364
calcar-to-canal ratio (FW/CW) was used for the classification.
365
Figs. 4. Postoperative radiographs of a 49-year old female with type A femur who had undergone THA due
366
to secondary osteoarthritis. Figs. 4-A and 4-B Two-month postoperative images show minimal proximal
367
radiolucent lines. Figs. 4-C and 4-D More obvious proximal radiolucent lines were found with hypertrophic
368
distal cortices 65 months after the index operation.
369
Figs. 5. A 65-year old man with type A femur underwent THA owing to secondary osteoarthritis. Fig. 5-A
370
Immediate postoperative radiograph demonstrates stable stem fixation. Fig. 5-B Femoral radiograph shows
371
periprosthetic fracture (Vancouver B2) resulted from a minor trauma 1 month after the surgery. Fig. 5-C
372
Isolated stem revision was performed using a distal fixating modular stem.
373
Fig. 6. Stem survival curves for matched type A and B femurs (log rank, P=0.041).
374
Fig. 7. Stem survival curves for type A, B and C femurs (log rank, P=0.047).
AC C
EP
TE D
M AN U
SC
RI PT
359
20
ACCEPTED MANUSCRIPT Legends to figures
2
Fig. 1. A double tapered wedge stem with a grit-blasted surface treatment. (Bencox; Corentec, Cheon-An,
3
South Korea)
4
Fig. 2. Radiographic features of different bone types by the Dorr classification (type A, B and C)
5
Fig. 3. Determination of bone types using femoral canal width (FW) and calcar width (CW). The value of
6
calcar-to-canal ratio (FW/CW) was used for the classification.
7
Figs. 4. Postoperative radiographs of a 49-year old female with type A femur who had undergone THA due
8
to secondary osteoarthritis. Figs. 4-A and 4-B Two-month postoperative images show minimal proximal
9
radiolucent lines. Figs. 4-C and 4-D More obvious proximal radiolucent lines were found with hypertrophic
M AN U
SC
RI PT
1
distal cortices 65 months after the index operation.
11
Figs. 5. A 65-year old man with type A femur underwent THA owing to secondary osteoarthritis. Fig. 5-A
12
Immediate postoperative radiograph demonstrates stable stem fixation. Fig. 5-B Femoral radiograph shows
13
periprosthetic fracture (Vancouver B2) resulted from a minor trauma 1 month after the surgery. Fig. 5-C
14
Isolated stem revision was performed using a distal fixating modular stem.
15
Fig. 6. Stem survival curves for matched type A and B femurs (log rank, P=0.041).
16
Fig. 7. Stem survival curves for type A, B and C femurs (log rank, P=0.047).
AC C
EP
TE D
10
1
ACCEPTED MANUSCRIPT Table I. Baseline characteristics of three groups Type A (N = 611)
Type B (N = 427)
Type C (N = 51)
495
340
41
50.6 ± 13.6
54.0 ± 13.8
64.7 ± 12.0
0.001
RI PT
1
0.001
Number of patients Patient age* (yr) Gender† Male
266 (43.6%)
234 (54.7%)
Female
345 (56.4%)
193 (45.3%)
289 (47.4%)
Osteoarthritis
282 (46.1%)
Fracture & Others
40 (6.6%)
Ethnicity† 608 (99.5%)
Others BMI* (kg/m²)
Stem size* (number)
<0.001
16 (31.4%)
162 (38.1%)
25 (49.0%)
50 (11.7%)
10 (19.6%) 0.709 51 (100%)
3 (0.5%)
1 (0.2%)
0 (0%)
24.4 ± 3.4
24.3 ± 3.3
24.1 ± 3.1
0.901
83.8 ± 23.7
85.1 ± 25.1
75.7 ± 21.8
0.520
4.4 ± 1.7
5.6 ± 1.7
6.4 ± 2.0
0.002
EP
Follow-up duration* (mo)
32 (62.7%)
426 (99.8%)
TE D
Asian
215 (50.2%)
M AN U
ONFH‡
19 (37.3%)
SC
Diagnosis†
P Value
*The values are given as the mean and the standard deviation. †The values are given as the number of hips
3
with the percentage in parentheses. ‡ ONFH = Osteonecrosis of femoral head.
4 5
AC C
2
6 7 8 9
1
ACCEPTED MANUSCRIPT Table II. Baseline characteristics of matched two groups Type A (N = 408)
Type B (N = 408)
328
328
53.3 ± 13.7
53.6 ± 13.4
Number of patients Patient age* (yr)
P Value
0.947
RI PT
1
Gender†
1.000
219 (53.7%)
219 (53.7%)
Female
189 (46.3%)
189 (46.3%)
SC
Male
Diagnosis†
0.091
197 (48.3%)
Osteoarthritis
181 (44.4%)
156 (38.2%)
30 (7.4%)
44 (10.8%)
BMI* (kg/m²)
24.3 ± 3.4
24.4 ± 3.3
0.555
Follow-up duration* (mo)
87.8 ± 18.5
85.2 ± 20.5
0.058
5.6 ± 1.8
<0.001
TE D
Fracture & Others
Stem size* (number)
208 (51.0%)
M AN U
ONFH
4.5 ± 1.7
*The values are given as the mean and the standard deviation. †The values are given as the number of hips
3
with the percentage in parentheses. ‡ ONFH = Osteonecrosis of femoral head.
5 6 7 8
AC C
4
EP
2
9 10 11
1
ACCEPTED MANUSCRIPT 1
Table III. Clinical outcomes of matched two groups Type A (n=408)
Type B (n=408)
P Value
Preoperative
45.3 ± 13.1
44.0 ± 14.6
0.176
Postoperative
92.3 ± 9.17
92.9 ± 7.36
0.242
Preoperative
3.11 ± 0.97
3.00 ± 0.95
0.947
Postoperative
6.44 ± 1.28 19 (4.7%)
Postoperative groin pain†
20 (4.9%)
6.46 ± 1.13
0.772
6 (1.5%)
0.008
M AN U
Postoperative thigh pain†
SC
UCLA activity score*
RI PT
Harris hip score*
12 (2.9%)
0.149
2
*The values are given as the mean and the standard deviation. †The values are given as the number of hips
3
with the percentage in parentheses.
7 8 9 10 11 12
EP
6
AC C
5
TE D
4
13 14 15 16
1
ACCEPTED MANUSCRIPT Table IV. Radiographic features of matched two groups Type B (n=408)
P Value
32 (7.8%)
10 (2.5%)
<0.001
30 (7.4%)
7 (1.7%)
0 (0%)
2 (0.5%)
2 (0.5%)
1 (0.2%)
Radiolucent lines Proximal zones only Distal zones only Proximal & distal zones Stem position
<0.001
Neutral
390 (95.6%)
365 (89.5%)
Varus
16 (3.9%)
43 (10.5%)
Valgus
2 (0.5%) 16 (3.9%)
Stem stability
5 6 7 8
0.571 0.026
387 (94.9%)
Fibrous stable
36 (8.8%)
21 (5.1%)
Unstable
2 (0.5%)
0 (0%)
Subsidence
7 (1.7%)
1 (0.2%)
0.069
Heterotrophic ossification
33 (8.1%)
46 (11.3%)
0.124
TE D
370 (90.7%)
The values are given as the number of hips with the percentage in parentheses.
EP
4
13 (3.2%)
Bone ingrowth
AC C
3
0 (0%)
M AN U
Cortical hypertrophy
2
RI PT
Type A (n=408)
SC
1
9 10 11 12
1
ACCEPTED MANUSCRIPT 1
Table V. Clinical outcomes of femurs with and without radiolucency With radiolucency (n=42)
Without radiolucency (n=774)
P Value
Preoperative
42.5 ± 12.7
44.8 ± 13.9
0.299
Postoperative
86.2 ± 15.7
RI PT
Harris hip score*
Preoperative
3.14 ± 0.95
Postoperative
6.00 ± 1.61 13 (31%)
Postoperative groin pain†
7 (16.7%)
3.05 ± 0.96
0.549
6.47 ± 1.18
0.067
12 (1.6%)
<0.001
25 (3.2%)
0.001
M AN U
Postoperative thigh pain†
<0.001
SC
UCLA activity score*
93.0 ± 7.6
2
*The values are given as the mean and the standard deviation. †The values are given as the number of hips
3
with the percentage in parentheses.
7 8 9 10 11 12
EP
6
AC C
5
TE D
4
13 14 15 16
1
ACCEPTED MANUSCRIPT Table VI. Major complications and revision rates of matched two groups Type B (n=408)
P Value
13 (3.2%)
4 (1.0%)
0.027
4 (1.0%)
0 (0%)
0.124
4 (1.0%)
0 (0%)
9 (2.2%)
4 (1.0%)
Vancouver AG
1 (0.2%)
1 (0.2%)
Vancouver B1
1 (0.2%)
Vancouver B2
7 (1.7%)
Periprosthetic femoral fracture Intraoperative Vancouver A2
2 (0.5%)
0 (0%)
0.499
Aseptic cup loosening
2 (0.5%)
1 (0.2%)
1.000
Deep infection
3 (0.7%)
1 (0.2%)
0.624
0 (0%)
0 (0%)
1.000
3 (0.7%)
1 (0.2%)
0.624
9 (2.2%)
2 (0.5%)
0.034
3 (0.7%)
1 (0.2%)
0.624
0 (0%)
1.000
TE D
Dislocation Stem revision
EP
Cup revision Revision of both components
4 5
1 (0.2%)
The values are given as the number of hips with the percentage in parentheses.
AC C
3
2 (0.5%)
Aseptic stem loosening
Nerve palsy
2
0.162
1 (0.2%)
M AN U
Postoperative
RI PT
Type A (n=408)
SC
1
6 7 8
1
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
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
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