The Journal of Arthroplasty 32 (2017) 2487e2495

Contents lists available at ScienceDirect

The Journal of Arthroplasty

journal homepage: www.arthroplastyjournal.org

Revision Arthroplasty

Modified Extended Trochanteric Osteotomy for the Treatment of

Vancouver B2/B3 Periprosthetic Fractures of the Femur
Andreas Ladurner, MD *, Pia Zurmühle, MD, Vilijam Zdravkovic, MD, Karl Grob, MD
Department of Orthopedics and Traumatology, Kantonsspital St. Gallen, St. Gallen, Switzerland

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

Article history: Background: Femoral component revision is the treatment of choice for Vancouver type B2/B3
Received 17 November 2016 periprosthetic femur fractures (PFFs). The purpose of this study was to report the clinical outcome of
Received in revised form revision total hip arthroplasty with the use of a modified extended trochanteric osteotomy (ETO) in
13 February 2017
PFF treatment.
Accepted 27 February 2017
Available online 8 March 2017
Methods: A total of 43 cases between 2000 and 2014 were analyzed. Clinical and radiographic evaluation
was performed with a mean follow-up of 40 months. Patient survival after revision surgery, complica-
tions, radiographic outcomes, and quality of life and hip function were assessed.
Keywords:
Results: Merle d'Aubigne
and Postel score averaged 15, and mean postoperative Harris hip score was 70.
total hip arthroplasty
periprosthetic femur fractures Radiographic evaluation revealed that the ETO and fractures healed in all but 1 patient within 9 months.
extended trochanteric osteotomy Component stability and apparent osseointegration were not coincident with healing of the osteotomy
dual modular tapered stem and fracture sites proximal to the inserted stem. Six patients (15%) developed postoperative complica-
outcome tions, which included the following: 1 nonunion with progressive subsidence, 2 hip dislocations, 2 deep
infections, and 1 breakage of the modular junction of the revision stem.
Conclusion: The modified ETO with a lateral approach to the hip for the treatment of PFF is compatible
with fracture healing, a low dislocation rate, and good clinical results. However, component stability and
apparent osseointegration are coincident with fracture healing only in the distal aspect of the inserted
stem. Absence of proximal osseointegration might lead to poor osseous support resulting in inadequate
fatigue strength at the junction of the dual modular revision stem.
© 2017 Elsevier Inc. All rights reserved.

The number of primary total hip arthroplasty (THA) is estimated
to increase markedly during the coming decades [1,2]. Conse-
quently, the number of revision of THA will probably follow this
development. Projection based on registry data indicates that
the number of revision cases will double by the year 2026 [3].
Periprosthetic femoral fracture (PFF) occurring after THA implan-
tation is one of the most common causes of revision surgery after
aseptic loosening, osteolysis, pain, and dislocation [4]. The reported
This study was approved by the local research ethics committee.
This research did not receive any specific grant from funding agencies in the public,
commercial, or not-for-profit sectors.
No author associated with this paper has disclosed any potential or pertinent
conflicts which may be perceived to have impending conflict with this work. For
full disclosure statements refer to http://dx.doi.org/10.1016/j.arth.2017.02.079.
* Reprint requests: Andreas Ladurner, MD, Department of Orthopedics and
Traumatology, Kantonsspital St. Gallen, Rorschacherstrasse 95, St. Gallen 9007,
Switzerland.
incidence of PFF can be estimated up to 15% [5] and varies according
to patient demographics, implants, and technologies used, as well
as the length of follow-up [2,3,6]. PFF often occurs in elderly
patients and predominantly results from low-energy trauma with
or without the presence of osteolytic changes around the femoral
component. It has been shown that up to 82% of all postoperative
PFFs are associated with loose implants [7].
Femoral component revision is the preferred method of treat-
ment for Vancouver B2 and B3 PFFs, as the femoral component is
loose [3,6e9]. The lack of metaphyseal support requires a revision
stem that bypasses the defect. Options for treatment include
cylindrical, nonmodular cobalt-chromium stems, tapered, fluted
modular titanium stems, long cemented stems, tumor prostheses,
and allograft composites. Modular stems allow the surgeon
increased axial and rotational implant control together with more
flexibility during the revision procedure to restore leg length and
enhance implant stability. Revising femoral components after PFF
can be technically demanding with several associated risks and
complications. Reoperation rates of 23% and a postoperative
complication rate of 18% have been reported [10]. In standard

http://dx.doi.org/10.1016/j.arth.2017.02.079
0883-5403/© 2017 Elsevier Inc. All rights reserved.
2488 A. Ladurner et al. / The Journal of Arthroplasty 32 (2017) 2487e2495
revision THA, exposure for implanting such femoral components
can be simplified with the use of an extended trochanteric
osteotomy (ETO) [9,11e13]. The outcomes of converting the PFF into
an ETO at the time of revision THA has proved successful in a small
series of patients [12,14,15]. Including the fracture as a part of the
ETO allows wide exposure and a direct access to the femoral
diaphysis for implantation of the revision components. Recent
literature demonstrates that the ETO has a relatively low rate of
nonunion and is associated with fewer intraoperative femoral
fractures or cortical perforations, as well as decreased surgical time
[16e18]. However, with the traditional ETO technique that uses a
posterior approach to the hip joint [19] and the release of the
external rotators, a dislocation rate of up to 30% has been reported
[3,4,13,14,20,21]. The purpose of this retrospective study was to
report the clinical outcome of revision THA with the use of a
modified ETO [9,13,18,22] for the treatment of Vancouver B2 and B3
PFFs.
Materials and Methods
We conducted a retrospective chart and radiographic review of
43 consecutive patients who had undergone treatment for
Vancouver B2 (n ¼ 40) and B3 (n ¼ 3) fractures with the use of a
modified ETO at our institution between January 2000 and January
2014. Institutional review board approval was obtained. The study
group consisted of 23 females and 20 males with a mean age at the
time of fracture treatment of 77.5 years (range, 58-95 years). The
right side was affected in 25 of 43 patients. All patients had a prior
THA performed either through an anterolateral or transgluteal
approach. In 34 cases (79%), the femoral component had been
uncemented, and in 9 cases (21%), cemented fixation was used.
The mean interval between THA and the PFF was 8.8 years
(range, 2 weeks to 29 years); 95% (41 patients) of the fractures
occurred because of a fall from a level height or a twisting move-
ment, and 5% (2 patients) were associated with high-energy
mechanism of injury. Preoperatively, 33 patients were ambulatory
with 22 maintaining an unassisted gait, 8 used canes, and 3 used
a walker. Two patients were wheelchair bound. However,
pretraumatic mobility could not be assessed retrospectively in
8 patients.
Clinical notes and radiographs were retrospectively evaluated
for 40 of 43 patients. Three patients were lost to follow-up. Patients
were planned for clinical and radiographic follow-up at 8 weeks,
6 months, 1 year, and 2 years after surgery (range, 12-114 months
postsurgery). Clinical assessment focussed on hip range of motion,
walking ability (including the need for walking aids), and hip
function. Twenty-nine of the 40 patients underwent additional
clinical and radiographic final follow-up visit including a stan-
dardized questionnaire incorporating the Harris hip score and
Merle d’Aubigne  and Postel pain and walking scores between
March and September 2015. Eleven patients had died before the
final follow-up evaluation of causes unrelated to revision THA
surgery. In those cases, the patient’s general practitioner was
interviewed about any further surgery on the index hip. According
to the general practitioner, no revision surgery was performed in
the deceased patients. Kaplan-Meier calculations were performed
and survivorship curves generated with aseptic loosening of the
femoral components and revision surgery as end points.
The following intraoperative and postoperative parameters
were assessed from the chart review: type and size of implant,
mode of trochanter fixation, the number of cable wires used, the
estimated blood loss, the need for intraoperative or postoperative
blood transfusion, and complications. Concomitant procedures
included revision of the acetabular component in 1 case due to
malpositioning and a polyethylene liner exchange in 1 case.
For radiologic evaluation, all patients underwent preoperative
X-ray examination including a standardized anteroposterior view
of the pelvis and a lateral view of the affected hip and total femur.
X-rays taken immediately postsurgery and at each subsequent visit
(8 weeks, 6 months, 1 year, 2 years, and [in 29 patients] at the time
of final review) were analyzed. Preoperative radiographs for all
patients were evaluated and classified according to the Vancouver
classification system [8,20]. The immediate postoperative radio-
graphs served as a baseline to which all other postoperative views
were compared. All radiographic measurements were made by
1 orthopedic surgeon who was not involved in the surgeries and
focussed on implant stability and fracture union.
Ingrowth of the femoral components over time was studied and
assessed according to the classification of Engh et al [23]. Fixed
landmarks on the prosthesis (eg, the modular junction) and the
femur (eg, tip of greater tuberosity and cerclage cables) were
measured, and subsidence was recorded if there was any change
between these distances. Stems were classified as osseointegrated
if there was increased density of bone adjacent to the stem and if
diverging radiolucent lines and prosthetic subsidence were absent.
Clinical (no pain with weightbearing, palpation, or stressing of
the site) and radiographic (bridging callus) evidence was used to
determine the time of fracture healing.
The overall outcome was graded using the system of Beals and
Tower [24]. Excellent results included a stable arthroplasty and
fracture union with minimal deformity or shortening. Fracture
healing with moderate deformity/shortening and a stable subsi-
dence of the femoral component classified the result as good.
A loose component (regardless of pain status), nonunion, sepsis,
a new fracture, or severe femoral deformity/shortening defined
a poor result.
Surgical Technique
A cementless, tapered fluted dual modular titanium revision
stem (Revitan stem, Zimmer, Winterthur, Switzerland) was selected
for all cases. This implant has a 2 taper and a reinforced junction to
reduce the risk of stem breakage and is available with a straight or
curved distal stem in combination with a spout or cylindrical
proximal component. In our series, the curved stem in combination
with a cylindrical proximal component was always used. Preoper-
ative planning was performed in every case. Templates were used
to estimate the diameter and length of the revision component
required.
A modified ETO was used in all cases [9,13,18,22]. In a lateral
decubitus position, a lateral subvastus approach to the femur was
performed (Fig. 1A). The lateral intermuscular septum perforating
vessels were ligated 2 cm above the exit through the lateral inter-
muscular septum. The intermuscular septum was regarded as an
anatomic limit for dorsal exposure. Therefore, the linea aspera was
not exposed. No muscle fibers were stripped from the bone. Bone
fragments were left attached to the surrounding tissue, and care
was taken to preserve the periosteum to the bone fragments [25].
The distal end of the fracture was then identified. A prophylactic
cerclage cable was placed around the femur just distal to the extent
of the fracture site to prevent propagation, as the canal was pre-
pared for stem insertion later on. Depending on the integrity of the
greater trochanter (fracture line, amount of osteolysis), the prox-
imal osteotomy was either centered in the middle of the greater
trochanter or at the junction between the dorsal and anterior
two-thirds of the greater trochanter. Care was taken that any
valuable trochanteric bone fragment remained in contact with the
external rotators for final fixation. The posterior joint capsule was
preserved. An extended osteotomy of the greater trochanter was
made in a lateral to medial direction, centered over the anterior
 A. Ladurner et al. / The Journal of Arthroplasty 32 (2017) 2487e2495
Fig. 1. Surgical technique: (A) lateral subvastus approach to the femur, (B) performing the osteotomy by use of an oscillating saw, (C) opening the osteotomy and extraction of the
prosthesis, (D) assembly of the cylindrical proximal component in situ after the curved stem was fully seated.
edge of the prosthetic stem, and encompassing the anterior half of
the circumference of the femoral canal. The osteotomy was per-
formed with an oscillating saw and completed medially with an
osteotome (Fig. 1B). Whenever possible, the fracture line was
included into the osteotomy. The length of the trochanteric
osteotomy was planned to extend to the distal aspect of the fracture
line. The osteotomized anterior trochanteric fragment together
with the fractured bone fragments were reflected anteriorly along
with the attached abductors and vastus lateralis, intermedius, and
tensor vastus intermedius. Anterior dislocation of the prosthetic
hip was performed, the femoral component was extracted, and
debridement of the fracture fragments was performed removing
intervening soft tissue or bone cement (Fig. 1C). Whenever needed,
revision of the acetabular component was performed. Obstruction
of the femoral canal distal to the previous implanted stem was
removed by use of a medullary cavity reamer. The femoral bone
stock was further prepared using rasps of a progressively increasing
size until distal circular surface fixation in the isthmus region of the
femur was achieved. Trial femoral components were inserted
before implantation of the definitive component. An implantation
length in the intact distal femur twice the width of femoral
diaphysis was necessary to obtain the definitive distal curved
component. The modular portions were used to reproduce optimal
implant length and antetorsion. Assembly of the cylindrical prox-
imal component used was always done in situ after the curved stem
was fully seated (Fig. 1D). After reduction, the proximal fragments
were draped around the implant and stabilized using 2-mm Dall-
Miles cerclages (Fig. 2). Reduction and osteosynthesis of the
greater trochanter fragment was performed either with the use of a
cerclage wire passed around the subtrochanteric bone fragments or
through a transtrochanteric cannulated screw in combination
with 1.5-mm wires proximally (Figs. 2-4). Where the trochanteric
bone fragments were too weak due to pre-existing osteolysis, the
fragments were fixed by osteosutures only (Fig. 5).
2489
As recommended [26], the cerclage wires were always passed
around the bone distal to the lesser trochanter. A bone graft from
the greater trochanter (if available and left from the trochanteric
preparation) was placed at the fracture site to assist healing of the
osteotomy and fracture. Intraoperative radiographs were obtained
optionally. The postoperative regimen consisted of partial weight
bearing for 8 weeks (if possible) and progressive weightbearing,
thereafter.
Results
Forty-three patients were treated for Vancouver B2 and B3
fractures with the use of a modified ETO at our institution between
January 2000 and January 2014. Three were lost at follow-up
(1 died 4 days after surgery due to causes unrelated to surgery
and 2 patients were reassessed at a different institution.) Evalua-
tion of 29 patients at the final follow-up visit included a stan-
dardized questionnaire incorporating the Harris hip score and
Merle d’Aubigne  and Postel pain and walking scores. Eleven
patients had died before the final follow-up evaluation of causes
unrelated to revision THA surgery.
Overall mean length of follow-up was 40 months (range, 12-114
months), and the mean follow-up for the 11 patients who had died
before the final visit was 13.4 months (range, 6-33 months).
Radiographic Results
Radiographic evidence of fracture healing was achieved in all
but 1 patient. The mean time for ETO and fracture site healing in
39 of 40 patients was 9.1 months (Fig. 3). Osseointegration espe-
cially in the proximal part of the inserted stem was not coincident
with healing of the osteotomy and fracture sites. The earliest
osseointegration of the tapered fluted modular titanium revision
stem occurred in the dorsal and lateral aspect of the femur in
2490 A. Ladurner et al. / The Journal of Arthroplasty 32 (2017) 2487e2495
Fig. 2. Corresponding plain radiographs to the clinical case in Figure 1. (A) Preoperative and (B) postoperative radiography.
Gruen zones 3, 4, 5, 10, 11, and 12; followed by zones 2, 1, 9, and 13;
and finally in the area of Gruen zones 6, 7, 8, and 14 below the lesser
trochanter (Fig. 6). No clear evidence or absence of osseointegration
of the revision stem in the latter Gruen zones was noted in 19 of 40
cases (47.5%). The subtrochanteric ETO and fracture fragments were
fixed with an average of 3 cerclages. The average osteotomy length
at the time of revision was 135 mm (range, 68-188 mm). Depending
on the bone quality of the greater trochanter, the fixation of the ETO
fragment occurred with osteosuture only (5 of 40), intraosseous
cables in combination with cannulated screws (20 of 40; Fig. 2) or
subtrochanteric cable wire (15 of 40, Fig. 3). Dislocation of the
greater trochanter fragment of >1 cm occurred in 3 patients.
Heterotopic ossifications were diagnosed in 6 of 40 patients, and
according to the Brooker classification [27], they were classified
grade I in 1 of 6, grade II in 3 of 6, grade III in 1 of 6, and grade IV in
1 of 6 patients. Subsidence of the Revitan stem was noted in 3 of 40
cases in the first 6 postoperative months. In 1 case each, subsidence
was 6 and 8 mm, but they were found to stabilize without further
change at the latest follow-up. In 1 case, symptomatic subsidence of
32 mm occurred. A stem revision was required due to concomitant
failed fracture healing and aseptic loosening. In the remaining 37 of
40 cases, no measurable subsidence occurred; 39 of 40 femoral
components demonstrated evidence of osseointegration based on
the criteria of Engh et al [23].
Clinical Results
The mean postoperative Harris hip score was 70 of 100 points,
and Merle d'Aubigne
and Postel score averaged 15 of 18.
Postoperative mean range of motion was 98.2 flexion (range, 35
to 120 ) and 6 extension (range, 10 to 0 ). Mean internal and
external rotations were 13.4 (range, 10 to 40 ) and 37.4 (range,
15 to 75 ), respectively, whereas mean abduction was 34.4 (range,
5 to 70 ). At the latest available follow-up, 6 patients exhibited a
positive Trendelenburg sign; 21 of 40 patients (52.5%) walked
fluently without assistance, 18 of 40 patients were using a gait-
assisting device (11 required 1 or 2 canes and 7 used walkers),
and 1 patient was wheelchair bound due to general weakness.
According to the Beals and Tower classification, there were 31 of 40
excellent, 5 of 40 good, and 4 of 40 poor results.
The mean intraoperative blood loss was 900 mL. However, in
2 of 40 patients, the blood loss could not be reviewed retrospec-
tively; 21 of 40 patients (53%) suffered from symptomatic anemia
requiring erythrocyte concentrate transfusion (10 patients required
both intraoperative and postoperative transfusions, 2 patients had
intraoperative blood cell transfusion, and 9 had postoperative
transfusions.) Mean transfusion rate in those 21 patients were
3 erythrocyte concentrates.
Complications
There were a total of 6 of 40 patients who developed post-
operative complications (15%). Five (12.5%) required further revi-
sion surgery. One patient (2.5%) developed nonunion at the fracture
site due to progressive subsidence up to 32 mm. Two patients (5%)
had dislocations, 1 requiring revision of the acetabular component,
whereas the other patient was treated conservatively after repo-
sition. Two patients (5%) developed deep infections, and of these,
 A. Ladurner et al. / The Journal of Arthroplasty 32 (2017) 2487e2495
Fig. 3. Periprosthetic femur fracture (PFF) treatment in a 58-year-old patient. (A) PFF, (B) postoperative radiography, and (C) result 5 years postoperatively.
the one with an acute infection was successfully treated with lavage
and component exchange, whereas the second patient with infec-
tion 10 weeks postoperatively necessitated a full 2-stage exchange
arthroplasty. One patient (2.5%) sustained a fracture of the modular
junction of the tapered and distally well-fixed revision stem (Fig. 7).
Despite at first glance not evident radiologically, no ingrowth of
bone was seen intraoperatively in the proximal aspect of the
broken dual modular stem in the Gruen zones 6, 7, and 8. One direct
surgery-related complication was observed. Immediately post-
operative, the patient was diagnosed with active bleeding from the
descending branch of the lateral circumflex femoral artery, caused
by intraoperative erosion while placing a cerclage wire. The prob-
lem was solved by angiographic coiling. Prolonged serous wound
effusion of >14 days was noted in 3 cases, all of which resolved
spontaneously without further intervention.
Discussion
Treatment of PFFs can be challenging, and high rates of compli-
cations, reoperations, and early mortality are reported throughout
the literature [7,10,28,29]. Clinical outcomes are often poor and
substantially burden the patients. Treatment goals consist of a stable
joint, fracture healing, and the return to prefracture function.
Femoral component revision is the preferred method of treatment of
Vancouver B2 and B3 PFFs.
ETO is known to be a useful technique for complex revision
THA [12,13]. The 2 commonly used techniques are the standard
ETO via a posterior approach or the modified ETO via a lateral
approach. The traditional standard ETO technique as described by
Younger et al [19] uses the posterior approach to the hip, with
release of the external rotators, incision of the posterior joint
capsule, and posterior surgical dislocation. In the modified ETO
technique, a lateral approach with osteotomy of the greater
trochanter in its mid-portion is performed [9,13,18,22]. The
2491
modified ETO preserves the posterior capsule and external rotator
attachments. The anterior joint capsule is incised and the disloca-
tion and removal of the implant can be performed anteriorly
without impairment to the posterior joint capsule and external
rotators [9,18]. As the hip joint is advanced from the distal through
the extended trochanteric fragments violence to the abductor
muscles, for example, gluteus medius and minimus muscles can be
limited to the first 3 cm proximal to the tip of the greater trochanter
(Fig. 1D), leaving the superior gluteal nerve untouched. It can be
argued that performing an ETO in the setting of PFF will increase
the number of fracture fragments. However, the fracture line can be
included into the osteotomy if applicable. Our study results indicate
that healing of the ETO can still be expected. Both techniques, the
traditional and modified ETO provide wide exposure of the ace-
tabulum, facilitate femoral component exposure and removal, aid
in canal preparation and femoral reconstruction, and allow for
correction of proximal femoral deformity [16,18]. An advantage of
the ETO is that the insertion of the gluteus medius and minimus
along with the origin of the vastus lateralis, intermedius, and tensor
vastus intermedius can be preserved. The soft tissue attachment
with its blood supply and large surface area of contact helps in
healing the osteotomy.
Only few authors have described the ETO in periprosthetic
fractures. Levine et al [12] reported his outcomes performed in
14 periprosthetic fractures. He used the standard ETO technique
with posterior surgical dislocation. The ETO healed in all 14 hips
within 13 weeks. Although there was a total of 5 (36%) post-
operative complications in this series, none of the complications
were related to the ETO. Drexler et al [13] reported the use of a
modified ETO in 34 patients with Vancouver B2/B3 PFFs treated
with revision THA. They reported union in 33 of 34 hips with a
survival rate of 88.2% after a mean follow-up of 57 months.
Despite excellent rates of fracture union, the most common
complication after revision THA for a periprosthetic fracture is
2492 A. Ladurner et al. / The Journal of Arthroplasty 32 (2017) 2487e2495
Fig. 4. PFF treatment in a 78-year-old patient. (A) PFF, (B) postoperative radiography, refixation of the greater trochanter by using intraosseous cables in combination with can-
nulated screws, and (C) fracture healing 6 months after surgery.
dislocation. Postoperative dislocation using the standard ETO has
been reported to range from 7% to 30% [7,12]. In our series, only
2 patients (5%) sustained a hip dislocation. Although the modified
ETO showed a lower dislocation rate, a higher incidence of
trochanteric fracture and postoperative trochanteric dislocation
may occur when compared with the standard ETO [18,30]. In the
present study, dislocation of the greater trochanter fragment of >1
cm occurred in 3 patients (7.5%). Migration of the greater
trochanter with the use of ETO has been reported by others [17].
In 39 of 40 patients in the present series, there was healing of
both the femur fracture and the ETO. The aforementioned results are
similar to others [12,13,15,18,29], where a union rate of 91%-100%
with the use of ETO has been reported. Surgical techniques in the
present study emphasized fracture healing, the preservation of the
external rotator attachments, soft tissue handling, and the avoidance
of unnecessary exposure of the linea aspera where the vessels enter
the femur [25]. However, although osseointegration occurred rapidly
in zones 3-5 and 10-12, osseointegration was not evident radiolog-
ically in Gruen zones 6, 7, 8, and 14 at latest follow-up in 47.5%
(Fig. 5). This area corresponds to the modular junction of the pros-
thetic stem, and lack of bony support could be the reason for stem
failure in this area as described in the literature [31]. Therefore, the
shorter the proximal modular stem, the higher the chance that the
modular junction is situated in the zones where the osseointegration
is absent. The lack of proximal bone restoration might be due to the
fact that load transfer occurs through the tapered wedge portion of
the distal stem and the proximal portion of the stem is bypassed and,
therefore, does not stimulate bone ingrowths. Poor proximal osseous
support causing inadequate fatigue strength in the prosthesis has
been mentioned by others [32,33]. In 1 patient in the present
study, the breakage of the dual modular stem occurred where no
osseointegration was observed proximal to the junction (Fig. 7).
Subsidence rates show a wide variance throughout the
literature. Fink [34] reported no subsidence in a case series of
23 patients despite 2 patients who did not achieve bony ingrowth
fixation. Others observed subsidence occurring in up to 50% of
cases within 6 months of surgery [5,6]. Bohm and Bischel [35]
suggested that the degree of subsidence correlated with the
extent of preoperative femoral bone stock especially in the
diaphysis and with the quality of osseointegration and degree of
osteoporosis. In the present study, early femoral component
subsidence was noted in 3 cases (7.5%). One patient needed a
second revision due to subsequent aseptic loosening. In 2 patients,
the subsidence was subsequently nonprogressive and associated
with femoral component osseointegration.
All the patients in the present series were treated using a dual
modular, tapered and distally fixed, uncemented titanium alloy
(TiAl6Nb7) revision stem (Revitan, Zimmer GmbH, Winterthur,
Switzerland). Modular fluted tapered stems have been used to treat
Vancouver B2 and B3 fractures before [5,15,21,36].
Breakage of the modular junction as mentioned previously can
be a possible concern when using modular implants. Modularity
probably increases the complication rate and might make the
implants susceptible to corrosion, fretting, or fatigue fracture at the
modular junction [31e33,37]. Factors associated with an increased
risk of prosthetic fracture are high patient body weight or body
mass index, high patient activity level, small stem diameter, poor
proximal osseous support, the use of ETO, and varus orientation of
 A. Ladurner et al. / The Journal of Arthroplasty 32 (2017) 2487e2495
Fig. 5. PFF with stem breakage after fall from level height in a 76-year-old patient. (A) PFF, (B) treatment with stem exchange, refixation of greater trochanter with sutures due to
extensive bone loss, and (C) 12 years after stem exchange.
the stem [31e33]. Richards et al [38] reported 4 cases of stem
breakage at the modular junction in a series of 109 patients after a
mean follow-up of 37 months. The author stated that all of the stem
fractures occurred in older implant designs that are no longer in use
but that stem breakage was not observed with newer designs.
Van Houwelingen et al [36] showed similar results with 5 stem
fractures in a series of 48 patients after a mean follow-up of 84
months. Also, this series analyzed the original standard stem design
that has now been modified. Overall implant survivorship was 90%.
Norman et al [31] reported 2 cases of a Revitan stem failure with
breakage of the modular junction in 2 active male patients. In their
analysis, undersizing of the stem diameter at the modular junction
relative to the characteristics of the patients was stated as a
possible cause. Nasr and Keene [37] reported a single case of a
fractured connection taper in a Revitan stem, highlighting that
fractures can still occur with modern modular prostheses. The
failure of the Revitan stem in the present study was initiated by a
traumatic impact onto the hip because of fall from a bicycle,
which might have caused a fatigue crack of the modular junction.
Therefore, although midterm survivorship of modular titanium
stems are considered high, continued surveillance of stem
junctional fatigue is required.
Clinical Results
The average Harris hip score in the present series was 70 of 100
points, and Merle d'Aubigne
and Postel score averaged 15 of 18.
2493
Although this is a relatively low score and does not represent a
return to a high level of function, it is more likely related to the
significant medical comorbidities, frail nature, and limited preop-
erative activity levels in these patients. Abdel et al [21] reported a
mean Harris hip score of 83, 4.5 years postoperatively in a series of
44 patients. Several studies have also confirmed these low post-
operative Harris hip scores ranging from 59 to 71 after treatment
for PFFs [12,15,39e41]. The main objectives when treating peri-
prosthetic fractures are fracture union and reduced pain. Although
unassisted ambulation is always a goal for these patients, in the
present series, only 18 of 40 patients (48%) walked fluently without
assistance. Six of 40 patients showed a positive Trendelenburg sign
at final follow-up. The relatively high rate of limping might be
contributed to general weakness of a rather old patient cohort and
may be due to multiple surgical interventions before PFF. Several
studies have shown that typically 50% patients will require a
postoperative assistive device and maintain a limited ambulatory
status [3,13,38]. The presence of a positive Trendelenburg sign was
not mentioned in previous studies.
Complications
Revision surgery was necessary in 12.5% of our patients (5 of 40).
The most common cause was infection (2 patients, 5%), with
1 necessitating a 2-stage exchange, and the other causes were
1 subsidence, a recurrent dislocation, and a stem breakage. The
latter 2 were associated with another trauma and not directly
2494 A. Ladurner et al. / The Journal of Arthroplasty 32 (2017) 2487e2495

Fig. 6. (A and B) Radiographic osseointegration of the implant over time (anteroposterior view, axial view). Zone of primary osseointegration (Gruen zones 3, 4, 5, 10, 11, and
12). Zone of secondary osseointegration (Gruen zones 1, 2, 9, and 13). Zone of tertiary (latest) osseointegration (Gruen zones 6, 7, 8, and 14).

related to the previous surgery. The Swedish national hip registry
shows a postoperative complication rate (requiring revision
surgery) of 18% after PFF treatment [3]. Springer et al [42] reviewed
116 patients (118 hips) after periprosthetic fracture treatment
and reported a revision rate for loosening, nonunion, recurrent
dislocation, refracture, or infection of 17% at the latest follow-up of
5.4 years postoperatively. Implants used were cemented stems,
proximally porous-coated or extensively porous-coated stems,
allograft-prosthesis composite, or tumor prosthesis. Abdel et al [21]
reported a series with comparable follow-up time (mean, 4.5 years)

Fig. 7. Stem breakage after PFF treatment, solved by stem exchange. (A) PFF, (B) stem exchange via a modified extended trochanteric osteotomy, (C) stem breakage at the modular
junction 5 years after PFF treatment, and (D) treatment with stem exchange.
 A. Ladurner et al. / The Journal of Arthroplasty 32 (2017) 2487e2495
and number of cases (n ¼ 44). Treatment modality was similar to
our series, using a modular fluted, tapered stems and (in many
cases) an ETO. The reoperation rate in this series was 18% (7 of 44),
mostly due to recurrent dislocation (5 of 8 revisions).
Conclusion
In conclusion, the modified ETO with a lateral approach to the
hip for the treatment of PFF is compatible with fracture healing, a
low dislocation rate, and good clinical results. However, component
stability and apparent osseointegration are coincident with fracture
healing in the distal aspect of the inserted stem only and not
proximal. Absence of proximal osseointegration might lead to poor
osseous support resulting in inadequate fatigue strength at the
junction of the dual modular revision stem.
Limitations of the present study include the retrospective design
and relatively short time to follow-up. Furthermore, patients’
preoperative medical condition may have some influence on the
postoperative outcome.
References
[1] Kurtz S, Ong K, Lau E, Mowat F, Halpern M. Projections of primary and revision
hip and knee arthroplasty in the United States from 2005 to 2030. J Bone Joint
Surg Am 2007;89:780e5. http://dx.doi.org/10.2106/JBJS.F.00222.
[2] Lindahl H. Epidemiology of periprosthetic femur fracture around a total hip
arthroplasty. Injury 2007;38:651e4. http://dx.doi.org/10.1016/j.injury.2007.
02.048.
[3] Malchau H, Herberts P, Eisler T, Garellick G, So€derman P. The Swedish Total
Hip Replacement Register. J Bone Joint Surg Am 2002;84eA Suppl:2e20.
[4] Ohly NE, Whitehouse MR, Duncan CP. Periprosthetic femoral fractures in total
hip arthroplasty. Hip Int 2014;24:556e67. http://dx.doi.org/10.5301/hipint.
5000155.
[5] Munro JT, Garbuz DS, Masri BA, Duncan CP. Tapered fluted titanium stems in
the management of Vancouver B2 and B3 periprosthetic femoral fractures.
Clin Orthop Relat Res 2014;472:590e8. http://dx.doi.org/10.1007/s11999-
013-3087-3.
[6] Hernandez-Vaquero D, Fernandez-Lombardia J, de Los Rios JL, Perez-Coto I,
Iglesias-Fernandez S. Treatment of periprosthetic femoral fractures with
modular stems. Int Orthop 2015;39:1933e8. http://dx.doi.org/10.1007/
s00264-015-2958-5.
[7] Lindahl H, Malchau H, Ode n A, Garellick G. Risk factors for failure after
treatment of a periprosthetic fracture of the femur. J Bone Joint Surg Br
2006;88:26e30. http://dx.doi.org/10.1302/0301-620X.88B1.17029.
[8] Brady OH, Garbuz DS, Masri BA, Duncan CP. The reliability and validity of the
Vancouver classification of femoral fractures after hip replacement.
J Arthroplasty 2000;15:59e62.
[9] Grob K. Revisionsprothetik am Hüftgelenk. Die Rolle der erweiterten
Trochanterosteotomie. Lead Opin Zeitschrift Für Orthop€ aden, Rheumatol Und
Unfallchirurgen 2008;1:60e3.
[10] Lindahl H, Malchau H, Herberts P, Garellick G. Periprosthetic femoral fractures
classification and demographics of 1049 periprosthetic femoral fractures from
the Swedish National Hip Arthroplasty Register. J Arthroplasty 2005;20:
857e65. http://dx.doi.org/10.1016/j.arth.2005.02.001.
[11] Sambandam SN, Duraisamy G, Chandrasekharan J, Mounasamy V. Extended
trochanteric osteotomy: current concepts review. Eur J Orthop Surg Trau-
matol 2016;26:231e45. http://dx.doi.org/10.1007/s00590-016-1749-z.
[12] Levine BR, Della Valle CJ, Lewis P, Berger RA, Sporer SM, Paprosky W.
Extended trochanteric osteotomy for the treatment of Vancouver B2/B3
periprosthetic fractures of the femur. J Arthroplasty 2008;23:527e33. http://
dx.doi.org/10.1016/j.arth.2007.05.046.
[13] Drexler M, Dwyer T, Chakravertty R, Backstein D, Gross AE, Safir O. The
outcome of modified extended trochanteric osteotomy in revision THA for
Vancouver B2/B3 periprosthetic fractures of the femur. J Arthroplasty
2014;29:1598e604. http://dx.doi.org/10.1016/j.arth.2014.03.020.
[14] Stiehl JB. Extended osteotomy for periprosthetic femoral fractures in total hip
arthroplasty. Am J Orthop (Belle Mead NJ) 2006;35:20e3.
[15] Mulay S, Hassan T, Birtwistle S, Power R. Management of types B2 and B3
femoral periprosthetic fractures by a tapered, fluted, and distally fixed stem.
J Arthroplasty 2005;20:751e6. http://dx.doi.org/10.1016/j.arth.2004.11.020.
[16] Jando VT, Greidanus NV, Masri BA, Garbuz DS, Duncan CP. Trochanteric
osteotomies in revision total hip arthroplasty: contemporary techniques and
results. Instr Course Lect 2005;54:143e55.
[17] Mardones R, Gonzalez C, Cabanela ME, Trousdale RT, Berry DJ. Extended
femoral osteotomy for revision of hip arthroplasty: results and
2495
complications. J Arthroplasty 2005;20:79e83. http://dx.doi.org/10.1016/
j.arth.2004.10.014.
[18] Lakstein D, Kosashvili Y, Backstein D, Safir O, Gross AE. Modified Extended
trochanteric osteotomy with preservation of posterior structures. Hip Int
2010;20:102e8.
[19] Younger TI, Bradford MS, Magnus RE, Paprosky WG. Extended proximal
femoral osteotomy. A new technique for femoral revision arthroplasty.
J Arthroplasty 1995;10:329e38.
[20] Brady OH, Garbuz DS, Masri BA, Duncan CP. Classification of the hip. Orthop
Clin North Am 1999;30:215e20.
[21] Abdel MP, Lewallen DG, Berry DJ. Periprosthetic femur fractures treated with
modular fluted, tapered stems. Clin Orthop Relat Res 2014;472:599e603.
http://dx.doi.org/10.1007/s11999-013-2936-4.
[22] Lakstein D, Kosashvili Y, Backstein D, Safir O, Lee P, Gross AE. The long
modified extended sliding trochanteric osteotomy. Int Orthop 2011;35:13e7.
http://dx.doi.org/10.1007/s00264-009-0890-2.
[23] Engh CA, Massin P, Suthers KE. Roentgenographic assessment of the biologic
fixation of porous-surfaced femoral components. Clin Orthop Relat Res 1990:
107e28.
[24] Beals RK, Tower SS. Periprosthetic fractures of the femur. An analysis of 93
fractures. Clin Orthop Relat Res 1996:238e46.
[25] Grob K, Manestar M, Lang A, Ackland T, Gilbey H, Kuster MS. Effects of ligation
of lateral intermuscular septum perforating vessels on blood supply to
the femur. Injury 2015;46:2461e7. http://dx.doi.org/10.1016/j.injury.2015.
09.029.
[26] Grob K, Monahan R, Gilbey H, Yap F, Filgueira L, Kuster M. Distal extension of
the direct anterior approach to the hip poses risk to neurovascular structures:
an anatomical study. J Bone Joint Surg Am 2015;97:126e32. http://dx.doi.org/
10.2106/JBJS.N.00551.
[27] Brooker AF, Bowerman JW, Robinson RA, Riley LH. Ectopic ossification
following total hip replacement. Incidence and a method of classification.
J Bone Joint Surg Am 1973;55:1629e32.
[28] Tangsataporn S, Safir OA, Vincent AD, Abdelbary H, Gross AE, Kuzyk PRT. Risk
factors for subsidence of a modular tapered femoral stem used for revision
total hip arthroplasty. J Arthroplasty 2015;30:1030e4. http://dx.doi.org/
10.1016/j.arth.2015.01.009.
[29] Abdel MP, Cottino U, Mabry TM. Management of periprosthetic femoral
fractures following total hip arthroplasty: a review. Int Orthop 2015;39:
2005e10. http://dx.doi.org/10.1007/s00264-015-2979-0.
[30] MacDonald SJ, Cole C, Guerin J, Rorabeck CH, Bourne RB, McCalden RW.
Extended trochanteric osteotomy via the direct lateral approach in revision
hip arthroplasty. Clin Orthop Relat Res 2003:210e6. http://dx.doi.org/
10.1097/01.blo.0000096818.67494.7b.
[31] Norman P, Iyengar S, Svensson I, Flivik G. Fatigue fracture in dual modular
revision total hip arthroplasty stems: failure analysis and computed tomog-
raphy diagnostics in two cases. J Arthroplasty 2014;29:850e5. http://
dx.doi.org/10.1016/j.arth.2013.09.008.
[32] Buttaro MA, Mayor MB, Van Citters D, Piccaluga F. Fatigue fracture of a
proximally modular, distally tapered fluted implant with diaphyseal fixation.
J Arthroplasty 2007;22:780e3. http://dx.doi.org/10.1016/j.arth.2006.07.007.
[33] Lakstein D, Eliaz N, Levi O, Backstein D, Kosashvili Y, Safir O, et al. Fracture of
cementless femoral stems at the mid-stem junction in modular revision hip
arthroplasty systems. J Bone Joint Surg Am 2011;93:57e65. http://dx.doi.org/
10.2106/JBJS.I.01589.
[34] Fink B. Revision arthroplasty in periprosthetic fractures of the proximal femur.
Oper Orthopa €die Und Traumatol 2014;26:455e68. http://dx.doi.org/10.1007/
s00064-014-0305-4.
[35] Bo€hm P, Bischel O. Femoral revision with the Wagner SL revision stem:
evaluation of one hundred and twenty-nine revisions followed for a mean of
4.8 years. J Bone Joint Surg Am 2001;83eA:1023e31.
[36] Van Houwelingen AP, Duncan CP, Masri BA, Greidanus NV, Garbuz DS. High
survival of modular tapered stems for proximal femoral bone defects at 5 to
10 years followup. Clin Orthop Relat Res 2013;471:454e62. http://dx.doi.org/
10.1007/s11999-012-2552-8.
[37] Nasr PJ, Keene GS. Revision of a fractured uncemented revision stem using a
custom designed punch and retrograde through-knee approach. Case Rep
Orthop 2015;2015:485729. http://dx.doi.org/10.1155/2015/485729.
[38] Richards CJ, Duncan CP, Masri BA, Garbuz DS. Femoral revision hip arthro-
plasty: a comparison of two stem designs. Clin Orthop Relat Res 2010;468:
491e6. http://dx.doi.org/10.1007/s11999-009-1145-7.
[39] Maury AC, Pressman A, Cayen B, Zalzal P, Backstein D, Gross A. Proximal
femoral allograft treatment of Vancouver type-B3 periprosthetic femoral
fractures after total hip arthroplasty. J Bone Joint Surg Am 2006;88:953e8.
http://dx.doi.org/10.2106/JBJS.E.00120.
[40] Lindahl H, Garellick G, Regne r H, Herberts P, Malchau H. Three hundred and
twenty-one periprosthetic femoral fractures. J Bone Joint Surg Am 2006;88:
1215e22. http://dx.doi.org/10.2106/JBJS.E.00457.
[41] Berry DJ. Treatment of Vancouver B3 periprosthetic femur fractures with a
fluted tapered stem. Clin Orthop Relat Res 2003:224e31. http://dx.doi.org/
10.1097/01.blo.0000096821.67494.f6.
[42] Springer BD, Berry DJ, Lewallen DG. Treatment of periprosthetic femoral
fractures following total hip arthroplasty with femoral component revision.
J Bone Joint Surg Am 2003;85eA:2156e62.