G Model

JINJ-6629; No. of Pages 7

Injury, Int. J. Care Injured xxx (2016) xxx–xxx

Contents lists available at ScienceDirect

Injury
journal homepage: www.elsevier.com/locate/injury

Surgical tips of intramedullary nailing in severely bowed femurs in

atypical femur fractures: Simulation with 3D printed model
Jai Hyung Park a, Yongkoo Lee b, Oog-Jin Shon c, Hyun Chul Shon d, Ji Wan Kim e,*
a
Department of Orthopedic Surgery, Kangbuk Samsung Hospital, Sungkyunkwan University, School of Medicine, Seoul, Republic of Korea
b
Korea Institute of Machinery & Materials, Daegu, Republic of Korea
c
Department of Orthopedic Surgery, Yeungnam University Hospital, Yeungnam University, College of Medicine, Daegu, Republic of Korea
d
Department of Orthopedic Surgery, Chungbuk National University Hospital, Chungbuk National University, College of Medicine, Cheong-Ju,
Republic of Korea
e
Department of Orthopaedic Surgery, Haeundae Paik Hospital, Inje University, College of Medicine, Busan, Republic of Korea

A R T I C L E I N F O A B S T R A C T

Article history: Introduction: The surgical management of atypical femoral fractures (AFFs) is complex in cases with
Accepted 22 February 2016 severe bowing of the femur, being associated with a high rate of failure. Our first aim was to use
preoperative templating and 3D printed model characterise the technical difficulties associated with use
Keywords: of current commercially available intramedullary nail (IMN) systems for the management of AFFs with
Femur severe bowing. Our second aim was to use outcomes of our 3D printing analysis to define technical
Atypical femur fracture criteria to overcome these problems.
3D print
Material and Methods: The modelled femur with 3D printing had an anterior bowing curvature radius of
Intramedullary nailing
Mismatch
772 mm and an angle of lateral bowing of 15.48. Nine commercially available IMN systems were
evaluated in terms of position of the nail within the medullary canal, occurrence of perforation of femoral
cortex by the distal tip of the nail, and location of the site of perforation relative to the knee joint. The
following IMN systems were evaluated: unreamed femoral nail (UFN), cannulated femoral nail (CFN),
Sirus nail, right and left expert Asian femoral nail (A2FN), right and left Zimmer Natural Nail (ZNN),
proximal femoral nail anti-rotation (PFNA), and Zimmer Cephalomedullary Nail (CMN).
Results: Along the sagittal plane, the UFN, CFN and Sirus systems were acceptably contained within the
medullary canal, as well as the ‘‘opposite side’’ A2FN and ZNN. Only the Sirus IMN system was contained
along the coronal plane. The distal part of the all other IMN systems perforated the anterior cortex of the
femur, at distances ranging between 2.8 and 11.7 cm above the distal end of the femoral condyles. Using
simulated fracture reduction in the 3D printed model, none of the 9 IMN systems provided acceptable
anatomical reduction of the fracture. A residual gap in fragment position and translation was provided by
the ‘‘opposite side’’ ZNN, followed by the UFN and Sirus systems.
Conclusion: Commercially available IMN systems showed mismatch with severely bowed femurs. Our
simulation supports that fit of these systems can be improved using an IMN system with a small radius of
curvature and diameter, and by applying specific operative procedures.
ß 2016 Elsevier Ltd. All rights reserved.

Introduction (ASBMR) described the major features of AFFs as including location

Since Odvina et al. [1] described severely suppressed bone
turnover with long-term use of Alendronate, in 2005, there has
been increasing interest in atypical femoral fractures (AFF). In
2010, the American Society for Bone and Mineral Research
* Corresponding author at: Department of Orthopaedic Surgery, Haeundae Paik
Hospital, Inje University, 1435, Jwa-dong, Haeundae-gu, Busan 612-862, Republic
of Korea. Tel.: +82 51 797 0990; fax: +82 51 797 0991.
E-mail address: bakpaker@hanmail.net (J.W. Kim).
in the subtrochanteric region or shaft of the femur; transverse
or short oblique orientation of the fracture line; atraumatic or
minimally traumatic mechanism of injury; non-comminuted
fracture pattern; localised periosteal or endosteal thickening of
the lateral cortex; and presence of a medial spike in complete
fractures [2]. An update to the classification by the ASBMR included
lower limb geometry and Asian ethnicity as potential contributing
risk factors for AFFs [3].
The geometry of the hip and of the proximal femur determines,
in part, the stresses that are experienced on the lateral cortex of the

http://dx.doi.org/10.1016/j.injury.2016.02.026
0020–1383/ß 2016 Elsevier Ltd. All rights reserved.

Please cite this article in press as: Park JH, et al. Surgical tips of intramedullary nailing in severely bowed femurs in atypical femur
fractures: Simulation with 3D printed model. Injury (2016), http://dx.doi.org/10.1016/j.injury.2016.02.026
G Model
JINJ-6629; No. of Pages 7

2 J.H. Park et al. / Injury, Int. J. Care Injured xxx (2016) xxx–xxx

femur [4]. The lateral cortex naturally sustains high levels of tensile
stress due to the natural bending of the femur in weight-bearing
[5,6]. Exacerbation of this natural bending with ‘‘bowing deformi-
ties’’ can significantly increase the magnitude of this tensile stress,
resulting in fractures involving the lateral cortex of the femur
[3]. This relationship between femoral ‘‘bowing’’ and tensile stress
contribute to the complexity in effectively managing fractures in
severely bowed femurs. In addition, current intramedullary nail
(IMN) systems have a poor fit with bowed femurs, increasing the
difficulty of the surgical management of these fractures [7–9].
Our study addressed two specific aims. Our first aim was to use
preoperative templating and to simulate with 3D printing to
characterise the technical difficulties associated with use of
current commercially available IMN systems for the management
of AFFs in femurs with severe bowing. Our second aim was to use
outcomes of our 3D printing analysis to define technical criteria to
overcome these problems.
Material and methods
Illustrative case
An 86-year-old woman bumped herself against a bedroom door
and sustained a left femoral shaft fracture. Antero-posterior (AP)
radiographs showed a fracture pattern typical of AFF, including
thickening of the lateral cortex, transverse orientation of the
fracture, and presence of a medial spike (Fig. 1A). Past history
revealed complaints of bilateral thigh pain over the last year, and a
36-month course of alendronate treatment. AP radiographs of the
right femur confirmed incomplete AFF, including thickening of
the lateral cortex, and presence of the ‘‘dreaded black line’’ in the
subtrochanteric region of the lateral bowing femur (Fig. 1B).
Lateral view radiographs demonstrated severe anterior bowing of
the femur, with a radius of curvature of 722 mm (Fig. 1C). A lateral
bowing angle of 15.48 was calculated from the lateral radiographs,
as the angle formed by the bisection of two lines drawn through
the long axis of the proximal and distal portions of the femoral
shaft.
Preoperative templating
In this case, we used the right femur, with the incomplete
fracture, to select an optimal IMN system. Alignment and
structure of the right femur was characterised by computed
tomography (CT), performed in a 16-slice CT scanner (SOMATOM
Sensation 16, Siemens, New York, NY), using a standard
algorithm for high-speed scan mode (120.0 kV, 348 mA).
Acquired images were reformatted into axial slices at 5-mm
intervals. The set of axial images was used for numerical,
3-dimensional (3D) reconstruction of the femur, using a standard
CT workstation (Siemens, New York, NY). The reconstructed
femur was 375 mm in length, measured from the tip of the
greater trochanter to the inferior surface of the lateral femoral
condyle. Rather than using manufacturers’ template for each nail
system, AP and lateral views of the reconstructed femur were
printed in real size, allowing positioning of the actual IMNs on
the printed images (Fig. 2A). The fit of the following IMN systems
was evaluated: unreamed femoral nail (UFN, diameter, 9 mm;

Fig. 1. Radiographs for the clinical case of the 86-year-old female with AFF. (A) Left femur AP showing lateral cortex thickening, transverse orientation, and medial spike.
(B) Right femur AP showing lateral cortex thickening and dreaded black line, and lateral bowed femur with 158. (C) Right femur lateral view showing severe anterior bowing
with 722 mm of radius of curvature.

Please cite this article in press as: Park JH, et al. Surgical tips of intramedullary nailing in severely bowed femurs in atypical femur
fractures: Simulation with 3D printed model. Injury (2016), http://dx.doi.org/10.1016/j.injury.2016.02.026
G Model
JINJ-6629; No. of Pages 7

J.H. Park et al. / Injury, Int. J. Care Injured xxx (2016) xxx–xxx 3

Fig. 2. The printed paper model of the real size of the femur is shown in (A). Perforation of the distal end of the UFN IMN, using the piriformis fossa as the proximal point of
origin, is shown in (B), where the label ‘‘D’’ indicates the distance from the point of perforation to the distal surface of the lateral condyle. (C) The fit for the UFN system, with
modified point of entry, indicated localisation of the nail within the medullary canal in the coronal plane, but with perforation in the sagittal plane. (D) Fit for the ‘‘opposite
side’’ A2FN, showing localisation of the distal tip with the medullary canal in the coronal plane, but with perforation in the sagittal plane. The label ‘‘d’’ indicates the distance
from the point of perforation to the distal surface of the lateral condyle. (E). Fit for the ‘‘opposite side’’ SNN system, again showing localisation of the distal tip within the
medullary canal in the coronal plane, but with perforation in the sagittal plane. The distance ‘‘d’’ was 2.8 cm, the least distance between the site of perforation and the knee
joint line of all 9 IMN systems.

length, 340 mm, Synthes, Oberdorf, Switzerland); cannulated Simulation using the 3D printed model
femoral nail (CFN, diameter, 10 mm; length, 340 mm, Synthes,
Oberdorf, Switzerland); Sirus nail (diameter, 9.3 mm; length, The actual fit of the 9 IMN systems was evaluated using the 3D
340 mm, Zimmer, Winterthur, Switzerland); right and left expert printed models. On the contrary to preoperative templating, left
Asian femoral nail (A2FN, diameter, 10 mm; length, 340 mm, PFNA and CMN were used (In preoperative templating, right PFNA
Synthes, Oberdorf, Switzerland); right and left Zimmer Natural and CMN used). The entry point of all IMN systems was tip of
Nail (ZNN, diameter, 9.3 mm; length, 340 mm, Zimmer, Win- greater trochanter (GT). The proximal reaming was performed
terthur, Switzerland); right proximal femoral nail anti-rotation with 16.5 mm reamer and medullary canal was reamed upto
(PFNA, diameter, 10 mm; length, 340 mm, Synthes, Oberdorf, 13.5 mm. The cancellous portion of the distal metaphysis was
Switzerland); and Zimmer Cephalomedullary Nail (CMN; diam- completely reamed to control the position of the nail either
eter, 10 mm; length, 340 mm, Zimmer, Winterthur, Switzerland).
The proximal point of insertion for straight IMNs (UFN and CFN)
was localised at the piriformis fossa, while for the other IMN
systems, the origin was localised at the greater trochanter. In the
AP view, the IMNs were placed along the proximal medullary
canal, passing just lateral to the medial cortex at the mid-portion
of the shaft. To evaluate the fit of the IMNs, we determined if
the distal end of the IMN was completely situated within the
medullary canal. When the distal end of an IMN perforated the
lateral cortex of the femur, the distance from the point of
perforation to the distal surface of the lateral femoral condyle
was measured (Fig. 2B). In the lateral view, the IMNs were
placed at the middle of the proximal metaphysis, passing
anteriorly to the posterior cortex at the mid-portion of the
shaft. Again, the fit of the IMNs was determined by the position
of the distal end within the medullary canal, and measurement of
the distance from the point of perforation, if present at the
anterior cortex, to the distal surface of the lateral femoral
condyle (Fig. 2D).

Production of the 3D printed model

The image of the left femur, based on the state of the fracture,
was divided into sub-images containing each fractured fragment.
Each bone fragment was reconstructed from CT scan image data
using Mimics 16 software (Materialise, Leuven, Belgium). The 3D
models were exported in STL-file format to a 3D printer (uPrint SE
Plus, Stratasys, Eden Prairie, US). A solid model was constructed
using ABSplus material, with a 0.245-mm layer thickness. The 3D
printer reconstruction of the proximal and distal fragments is
shown in Fig. 3. Fig. 3. The 3D printed model of the femur constructed from the CT scan images.

Please cite this article in press as: Park JH, et al. Surgical tips of intramedullary nailing in severely bowed femurs in atypical femur
fractures: Simulation with 3D printed model. Injury (2016), http://dx.doi.org/10.1016/j.injury.2016.02.026
G Model
JINJ-6629; No. of Pages 7

4 J.H. Park et al. / Injury, Int. J. Care Injured xxx (2016) xxx–xxx

concentrically or eccentrically. The fracture fragments were
reduced to their anatomical alignment as best as possible with
the IMN, and the proximal locking and hip screws were inserted,
followed by insertion of the distal locking screws. The fit for each
IMN system was evaluated by measuring the remaining gap
between fracture fragments and the location of the fragments
along both the sagittal and coronal planes (Fig. 4A, B). Fluoroscopic
images were also obtained to determine the relative positions of
the IMNs within the medullary canal (Fig. 4C).
Results
Pre-operative templating on printed paper
The fit parameters of the 9 IMN systems, evaluated from the
printed model, are listed in Table 1. Of the 9 IMN systems,
acceptable fit of the IMN within the medullary canal was obtained
only in the Sirus system with an additional small anterior shift for
the Sirus system. For the other IMN systems, distal perforation was
identified. Perforation in the sagittal plane (i.e., lateral view) was
identified for all remaining 8 IMN systems, while perforation in the
coronal plane (i.e., AP view) varied among the remaining 8 IMN
systems. An acceptable AP position was obtained for the UFN, CFN,
Sirus (Fig. 2C), and A2FN (Fig. 2E). IMN systems was obtained
acceptable position by a small lateral displacement of the point of
entry at the GT. The point of perforation in the AP view varied
between 11 cm and 15.8 cm above the distal surface of the lateral
femoral condyle (i.e., the knee joint line), and between 2.8 cm and
11.7 cm in the lateral view. The location of the perforation of the
distal tip of the IMN above the knee joint line was related to the
radius of curvature of the IMN system. IMN systems with a smaller
radius of curvature (i.e., more bowing) perforated the lateral cortex
more distally than IMN systems with a greater radius of curvature;
the system with the lowest radius of curvature (i.e., Sirus, 950 mm)
produced no cortical perforation.
Simulation of the 3D printed model
Anatomical reduction was not attained with any of the 9 IMN
systems, which means the mismatching of the nail and the bowed
femur; however, some systems performed more adequately than
others. The measured gaps between fracture fragments and
translation of the fragments post-reduction are reported in
Table 2. In the coronal plane, the residual gaps between fracture
fragments ranged between 0 and 4 mm, while the translation
ranged between 0 and 6 mm. In contrast, the fracture gaps in the
sagittal plane were slightly larger, ranging between 1 and 7 mm,
while the residual translation in the sagittal plane was minimal,
between 0 and 1 mm. The diameter of the fracture site in the AP
and lateral views was 25 mm and 26 mm, respectively. For a 1 mm
gap, the tangent of an angle will be 0.040 in the coronal plane, and
0.038 in the sagittal plane. Therefore, a 1 mm increase of gap in the
coronal plane induces about a 2.38 valgus change in the alignment,
and that of posterior gap results in a 2.28 posterior angulation.
Among the 9 IMN systems, the ‘‘opposite side’’ ZNN provided the
best fit, with the least gap and translation (Fig. 4D–G). The UFN and
Sirus systems also provided an adequate fit. In terms of the
proximal locking, the UFN system used a conventional screw
locking method, while the Sirus and ZNN systems used a
cephalomedullary screw system.
Clinical recommendations based on results
1. Implant choice
a. A nail with the smallest radius of curvature
b. A nail with the smallest diameter
c. The opposite side of laterally bended nail (if not available, a
straight nail with GT entry point)
2. Reaming of thickened lateral cortex
3. Using poller screws & positioning the distal part of the nail
laterally and anteriorly in the distal femur
4. Back-strike technique to reduce gap, if the presence of gap

Fig. 4. Outcomes of the simulation analysis using the 3D printed model, indicating the distance of the medial gap (A) and of the posterior gap (B). Fluoroscopy images of the
simulated model are shown in (C). The AP view of the fit obtained using an ‘‘opposite side’’ ZNN system (i.e., right system used on the left) (D), with the corresponding
fluoroscopy image (E). The lateral view of the fit for the ‘‘opposite side’’ ZNN system is shown in (F) lateral view of ZNN with the corresponding fluoroscopy image (G).

Please cite this article in press as: Park JH, et al. Surgical tips of intramedullary nailing in severely bowed femurs in atypical femur
fractures: Simulation with 3D printed model. Injury (2016), http://dx.doi.org/10.1016/j.injury.2016.02.026
G Model
JINJ-6629; No. of Pages 7

J.H. Park et al. / Injury, Int. J. Care Injured xxx (2016) xxx–xxx 5

Table 1
Fit parameters for the 9 IMN systems using the printed model.

IMN system AP-view Location of perforation Lateral-view Location of perforation Radius of curvature
above knee joint line (cm) above knee joint line (cm) of IMN system (mm)

UFN 9  340 mm P Perforated 15.7 Perforated 8.0 1500

UFN 9  340 mm No perforation None Perforated 6.8 1500
CFN 10  340 mm P Perforated 15.8 Perforated 11.7 1500
CFN 10  340 mm No perforation None Perforated 7.4 1500
Sirus 9.3  340 mm No perforation None No perforation None 950
A2FN 9  340 mm Perforated 11.0 Perforated 3.7 963
A2FN 9  340 mm O No perforation None Perforated 3.7 963
ZNN 9.3  340 mm Perforated 11.5 Perforated 2.8 1275
ZNN 9.3  340 mm O No perforation None Perforated 2.8 1275
PFNA 10  340 mm Perforated 11.6 Perforated 7.4 1500
CM 10  340 mm Perforated 12.2 Perforated 4.8 1275

IMN, intra-medullary nail; O, opposite side system used (i.e., right system used on the left).

Discussion had the lowest radius of curvature of 950 mm yielded the smallest
posterior gap of 1 mm, with the ‘‘opposite side’’ A2FN system
The treatment of AFFs is a challenging, with severe bowing of producing the next best reduction with a posterior gap of 3 mm.
the femur increasing the risk fracture for AFF [3], and adding The higher gap produced by the A2FN system may be due to its
complexity to the selection of appropriate IMN systems. AFFs are angulation in both the sagittal and coronal planes, which might
associated with a high rate of complication rates with operative decrease its fit despite a small radius of curvature. With the
fixation with open reduction and internal fixation technique [8], exception of the A2FN system, the posterior gap for all other IMN
and plating of AFF has been shown to poor outcomes [8,10]. For systems was found to be proportional to the radius of curvature of
these reasons, the ASBMR has recommended the use of full-length the IMN. Therefore, we can conclude that the IMN system with the
reconstruction nail systems for the management of AFFs [2]. How- smallest radius of curvature will provide the optimal fit in the
ever, severe anterior bowing of the femur can result in mismatch bowed femur. Damron et al. [12] also recommended that an
between the IMN and the alignment of the femur. Such mismatch increase in the curvature (smaller radius) of IMN implants could
is a risk factor for anterior cortical perforation of the distal femur reduce the risk for e anterior cortical perforation at the distal tip of
with subtrochanteric fractures [7], and leg length discrepancy with the implant. The selection of an optimal IMN system may vary
fractures of the femoral shaft [11], and complete fracture of the across populations. As an example, studies have demonstrated
femoral shaft with prophylactic nailing for incomplete AFF [9]. As difference in the average radius of anterior curvature of adult
customizable IMN systems for severely bowed femurs are not femurs among their populations. In the United States, the average
currently available, procedures should be taken to select an radius was 1200 mm (range, 530–3260 mm), with 95% of femurs
optimal IMN alignment to match, as closely as possible, the native having a radius between 720 mm and 2160 mm [7]. Comparatively,
alignment of the femur. the average radius of curvature in Scotland was 1580 mm (range,
Using a combination of 3D printing and modelling, we 660–2680 mm) [13], and 1040 mm (range, 658–1905 mm) in
demonstrated that IMN with the smallest radius of curvature Japanese women [14]. Maratt et al. [15] further analysed the radius
and diameter provided the most adequate accommodation to of curvature of the population in the United States by race,
severe anterior bowing. In practice, these desired alignment reporting an average radius of curvature of 1455 mm for
parameters of IMN can be achieved by using ‘‘opposite side’’ lateral Caucasians, with 95% for the femurs having radii between
bended nail, as an example using a right ZNN for fracture reduction 820 and 2780 mm, while the curvature was higher in African
of the left femur, or straight IMN systems. Of the 9 commercially Americans at 1545 mm and smaller for Asian American at
available IMN systems tested in this study, the ‘‘opposite side’’ ZNN 1297 mm. They also reported that males had slightly straighter
system provided the best clinical results by minimising both the femur than females. Oh et al. defined an apparent bowing
inter-fragment gap and the translation of fragments. deformity by an anterior radius of less than 600 to 700 mm, and
The anterior radius of curvature of the 9 IMN systems varied lateral bowing by an angle of less than 58 [16]. Analysis of the
between 950 mm and 1500 mm, compared to the 722 mm radius available intra-medullary femoral IMN systems demonstrated that
of the 3D printed model of the femur. The Sirus IMN system, which all systems had a greater radius of curvature than that of the bowed
Table 2 femurs. Therefore, selecting an IMN system with the smallest
Fit parameters for the 9 IMN systems in the sagittal and coronal planes using the 3D radius of curvature would minimise the ‘‘mismatch’’ in fitting of
printed bone model. the IMN system to an anteriorly bowed femur.
IMN system Coronal plane Sagittal plane Our results also demonstrated the importance of the diameter
(AP view) (lateral view) of the IMN system, with larger diameter nails, with the same radius
Gap Translation Gap Translation of curvature of thinner nails, will have a poorer fit. t. As an example,
(mm) (mm) (mm) (mm) the CFN system with a diameter of 10 mm produced an additional
UFN 9  340 mm E 1 1 2 0 3 mm in translation of fracture site in the coronal plane, and a
CFN 10  340 mm E 1 4 2 1 1 mm translation in the sagittal plane, when compared to the UFN
Sirus 9.3  340 mm 1 2 1 0 system which had the same length and radius of curvature as the
A2FN 9  340 mm 4 1 3 0 CFN, but with a diameter of 9 mm. IMN systems with smaller
A2FN 9  340 mm O 2 1 3 1
ZNN 9.3  340 mm 3 1 2 0
diameter are a more favourable choice to avoid perforation of the
ZNN 9.3  340 mm O 0 1 2 0 distal tip or reduce translation of fracture site. The small diameter
PFNA 10  340 mm 3 0 7 0 of IMN with minimal contact with cortex obtains the three point
CM 10  340 mm 3 0 5 0 contact of the IMN with canal, resulting in stability of the fracture
IMN, intra-medullary nail; O, opposite side system used (i.e., right system used on reduction. The ASBMR has also recommended that the medullary
the left). canal should be over-reamed, to a diameter at least 2.5 mm larger

Please cite this article in press as: Park JH, et al. Surgical tips of intramedullary nailing in severely bowed femurs in atypical femur
fractures: Simulation with 3D printed model. Injury (2016), http://dx.doi.org/10.1016/j.injury.2016.02.026
G Model
JINJ-6629; No. of Pages 7

6 J.H. Park et al. / Injury, Int. J. Care Injured xxx (2016) xxx–xxx

than the IMN diameter, to compensate for the narrow medullary anterior bowing of the femur. Our 3D printed model processes
diameter [2]. proved successful to define optimal selection and positioning of
Our templating test confirmed that use of a straight IMN, with a the IMN to improve fit between the IMN systems and the native
point of entry at the greater tuberosity, or use of an ‘‘opposite side’’ curvature of the femur (Fig. 5).
lateral bended nail is appropriate to overcome lateral bowing. For The limitations of our study must be acknowledged in
straight IMNs (e.g., UFN and CFN in this study), the point of entry evaluating its outcomes. Foremost, we used actual IMN systems
should be at the GT, and the distal portion of the IMN should be instead of manufacturers’ template for each system. Convention-
placed along the lateral cortex of the distal femur. In contrast, IMNs ally, pre-operative templating is used to confirm the size of the
with a lateral bend (e.g., ZNN and A2FN in this study) could not be IMN required [26,27], but the magnification of the template varies
appropriately placed, but use of the IMN for the ‘‘opposite side’’ between manufacturers. Templating is not always accurate [28],
was applicable to fit within the medullary canal in AP (Fig. 4). In and it was difficult to acquire the templates for all 9 IMN systems
addition, we observed that an anterior transition of the point of evaluated in our study. Therefore, we used a patient’s CT data for
entry at the GT displaced the position of the distal portion of the
IMN along the anterior cortex of the distal femur, which make an
increase in the radius of intramedullary curvature of the femur,
compared to the neutral point of entry defined on pre-operative
templating (Fig. 2C). However, an anterior transition of the point of
entry may result in a burst fracture of the proximal femur during
the IMN procedure [17–19]; therefore, this clinical scenario was
not simulated using our 3D printed model. Eccentric placement of
the IMN in the distal metaphysis of the femur can be unstable and,
in this case, poller screws can be used to increase stability in
clinical practice.
We tested these surgical recommendations in our simulation
analysis using a 3D printed model. The severe lateral curvature of
our test femur could almost be compensated using opposite side
ZNN system and the UFN system with a point of entry at the greater
trochanter, yielding a gap of 1 mm in the coronal plane. However,
the severity of the anterior curvature could not be adequately
compensated by the systems in the sagittal plane, with a 2 mm
posterior gap in position. We further demonstrated that the point
of entry could be adjusted to improve the fit of these systems to the
severe anterior bowing. Our results are supported by other studies.
Johnson and Tencer [20] reported that point of entry, fracture
point, reaming, and the diameter, design, and material of the IMN
system all influenced fit/mismatch and risk for perforation.
The geometry of the extremity must also be considered as a
potential contributor to altered stresses on the lateral cortex of the
femur which may, in combination with other pathological changes
of the bone itself, predispose to development of an AFF. The native
geometry of the lower limb in Asian populations is a clearly stated
as risk factor for AFFs [16,21]. Accordingly, the ASBMR revised their
definition of AFF in 2012 to include stress fractures caused by a
bowing deformity in the classification of AFFs [3]. In Japanese
populations, bowing deformity of the femoral shaft and the
mechanical axis of the lower limb are considered risk factors for
femoral stress and AFFs of the femoral shaft [22,23]. Koch reported
a geometrical description of the femur and an analysis of the
magnitude and distribution of stresses induced by loading of the
lower limb during gait [24]. This report indicated that the highest
magnitudes of tensile stress occur in the superior aspect of the
neck of the femur and over the lateral aspect of the proximal 20–
30% of the femoral length from GT. In patients with bowing
deformity, the deformity should be considered as the main factor
of stress concentration in the mid-portion of the femoral shaft [16].
Significant and severe bowing of the femur is not only a risk for
AFF, but also makes the management of these fractures challeng-
ing. A higher failure rate of reduction in severely bowed femurs has
been reported in studies [8,25]. While these studies reported
impairment in bone healing process as the primary cause of poor
outcome [25], the surgical technique, and specifically adaptation of Fig. 5. A 76-year-old woman with history of alendronate treatment over a period of
the procedure to different lower limb geometries to the effective- 60 months. (A) Radiographs of an incomplete right AFF and complete left AAA. The
alignment measures for the right femur were 98 of lateral bowing and a 823 mm
ness of the reduction of the AFF, should be re-evaluated. Oh et al.
radius of the anterior curvature. (B) Post-operative radiographs, 7 months after
[9] reported a mismatch between the curvature of the IMN and surgery, showing bone union. To adjust for the native bowing of the femurs, the AFF
that of the femur to be a determinant of intraoperative in the right femur was managed using a left ZNN IMN system, with a left ZNN
complication; in their study, this mismatch accentuated the system used for the right femur.

Please cite this article in press as: Park JH, et al. Surgical tips of intramedullary nailing in severely bowed femurs in atypical femur
fractures: Simulation with 3D printed model. Injury (2016), http://dx.doi.org/10.1016/j.injury.2016.02.026
G Model
JINJ-6629; No. of Pages 7
J.H. Park et al. / Injury, Int. J. Care Injured xxx (2016) xxx–xxx
templating, which provided additional information on the shape
and size of the femur and printed out the AP and lateral images as
real size. The IMN systems have different angulations and were not
in full contact with the printed images that could have introduced
some error in measurement. This limitation was corrected using
the 3D print model of the femur, providing more accurate
measurement of the location of the IMN within the medullary
canal. Our study did not address the effects of femoral bowing on
the mechanical strength of the IMN, an issue which may be of
specific concern when using ‘‘opposite side’’ IMNs. Considering the
goal of fracture management is to restore the anatomy back to its
normal or near normal as possible, there is a clear clinical rational
for using an ‘‘opposite side’’ IMN; however, mechanical testing is
required to determine the effects at the level of the implant itself,
including its longevity in prospective clinical studies.
Conclusion
Commercially available IMN systems are poorly designed to
address the problem of AFF management in severely bowed
femurs. Our simulation supports that fit of these systems can be
improved using an IMN system with a small radius of curvature
and diameter, and by applying specific operative procedures.
Conflicts of interest
The authors declare no conflict of interest in relation to the
preparation of this paper.
Acknowledgement
This work was supported by the Medical Research Funds from
Kangbuk Samsung Hospital.
References
[1] Odvina CV, Zerwekh JE, Rao DS, Maalouf N, Gottschalk FA, Pak CY. Severely
suppressed bone turnover: a potential complication of alendronate therapy.
J146?Clin Endocrinol Metab 2005;90:1294–301.
[2] Shane E, Burr D, Ebeling PR, Abrahamsen B, Adler RA, Brown TD, et al. Atypical
subtrochanteric and diaphyseal femoral fractures: report of a task force of the
American Society for Bone and Mineral Research. J Bone Miner Res: Off J Am
Soc Bone Miner Res 2010;25:2267–94.
[3] Shane E, Burr D, Abrahamsen B, Adler RA, Brown TD, Cheung AM, et al. Atypical
subtrochanteric and diaphyseal femoral fractures: second report of a task force
of the American Society for Bone and Mineral Research. J Bone Miner Res: Off J
Am Soc Bone Miner Res 2014;29:1–23.
[4] Crossley K, Bennell KL, Wrigley T, Oakes BW. Ground reaction forces, bone
characteristics, and tibial stress fracture in male runners. Med Sci Sports Exerc
1999;31:1088–93.
[5] Donnelly E, Meredith DS, Nguyen JT, Gladnick BP, Rebolledo BJ, Shaffer AD,
et al. Reduced cortical bone compositional heterogeneity with bisphosphonate
treatment in postmenopausal women with intertrochanteric and subtrochan-
teric fractures. J Bone Miner Res: Off J Am Soc Bone Miner Res 2012;27:672–8.
Please cite this article in press as: Park JH, et al. Surgical tips of intramedullary nailing in severely bowed femurs in atypical femur
fractures: Simulation with 3D printed model. Injury (2016), http://dx.doi.org/10.1016/j.injury.2016.02.026
7
[6] Koh JS, Goh SK, Png MA, Ng AC, Howe TS. Distribution of atypical fractures and
cortical stress lesions in the femur: implications on pathophysiology. Singap
Med J 2011;52:77–80.
[7] Egol KA, Chang EY, Cvitkovic J, Kummer FJ, Koval KJ. Mismatch of current
intramedullary nails with the anterior bow of the femur. J Orthop Trauma
2004;18:410–5.
[8] Prasarn ML, Ahn J, Helfet DL, Lane JM, Lorich DG. Bisphosphonate-associated
femur fractures have high complication rates with operative fixation. Clin
Orthop Relat Res 2012;470:2295–301.
[9] Oh CW, Oh JK, Park KC, Kim JW, Yoon YC. Prophylactic nailing of incomplete
atypical femoral fractures. ScientificWorldJournal 2013;2013:450148.
[10] Das De S, Setiobudi T, Shen L, Das De S. A rational approach to management of
alendronate-related subtrochanteric fractures. J146?Bone Jt Surg Br Vol
2010;92:679–86.
[11] Oh HC, Park SJ, Yoon HK. Surgical treatment in atypical diaphyseal femoral
fracture with anterior and lateral bowing. J Korean Orthop Assoc 2014;49:
485–9.
[12] Damron TA, Palomino KE, Roach S. Long Gamma nail stabilization of patho-
logic and impending pathologic femur fractures. Univ Pa Orthop J
1999;12: 13–20.
[13] Bruns W, Bruce M, Prescott G, Maffulli N. Temporal trends in femoral curvature
and length in medieval and modern Scotland. Am J Phys Anthropol
2002;119:224–30.
[14] Maehara T, Shinohara K, Yamashita K, Bun H, Kaneda D, Ikuma H. The
morphology of the femur in elderly Japanese female: analysis using 3D-CT.
Kossetsu (Fract) 2012;34:451–5.
[15] Maratt J, Schilling PL, Holcombe S, Dougherty R, Murphy R, Wang SC, et al.
Variation in the femoral bow: a novel high-throughput analysis of 3922 femurs
on cross-sectional imaging. J Orthop Trauma 2014;28:6–9.
[16] Oh Y, Wakabayashi Y, Kurosa Y, Fujita K, Okawa A. Potential pathogenic
mechanism for stress fractures of the bowed femoral shaft in the elderly:
mechanical analysis by the CT-based finite element method. Injury
2014;45:1764–71.
[17] Johnson KD, Tencer AF, Sherman MC. Biomechanical factors affecting fracture
stability and femoral bursting in closed intramedullary nailing of femoral shaft
fractures, with illustrative case presentations. J Orthop Trauma 1987;1:1–11.
[18] Papadakis SA, Zalavras C, Mirzayan R, Shepherd L. Undetected iatrogenic
lesions of the anterior femoral shaft during intramedullary nailing: a cadaveric
study. J Orthop Surg Res 2008;3:30.
[19] Tencer AF, Sherman MC, Johnson KD. Biomechanical factors affecting fracture
stability and femoral bursting in closed intramedullary rod fixation of femur
fractures. J Biomech Eng 1985;107:104–11.
[20] Johnson KD, Tencer A. Mechanics of intramedullary nails for femoral fractures.
Unfallchirurg 1990;93:506–11.
[21] Marcano A, Taormina D, Egol KA, Peck V, Tejwani NC. Are race and sex
associated with the occurrence of atypical femoral fractures? Clin Orthop
Relat Res 2014;472:1020–7.
[22] Oh Y, Wakabayashi Y, Kurosa Y, Ishizuki M, Okawa A. Stress fracture of the
bowed femoral shaft is another cause of atypical femoral fracture in elderly
Japanese: a case series. J Orthop Sci: Off J Jpn Orthop Assoc 2014;19:579–86.
[23] Sasaki S, Miyakoshi N, Hongo M, Kasukawa Y, Shimada Y. Low-energy diaphy-
seal femoral fractures associated with bisphosphonate use and severe curved
femur: a case series. J Bone Miner Metab 2012;30:561–7.
[24] Koch JC. The laws of bone architecture. Am J Anat 1917;21:177–298.
[25] Weil YA, Rivkin G, Safran O, Liebergall M, Foldes AJ. The outcome of surgically
treated femur fractures associated with long-term bisphosphonate use.
J146?Trauma 2011;71:186–90.
[26] Krettek C, Blauth M, Miclau T, Rudolf J, Konemann B, Schandelmaier P.
Accuracy of intramedullary templates in femoral and tibial radiographs. J
Bone Jt Surg Br Vol 1996;78:963–4.
[27] Iqbal M, Saravanan R, Konchwala A, Sakellariou A. Preoperative templating of
tibial nails – is it worthwhile? Injury 2000;31:449–50.
[28] King RJ, Craig PR, Boreham BG, Majeed MA, Moran CG. The magnification of
digital radiographs in the trauma patient: implications for templating. Injury
2009;40:173–6.