726

C OPYRIGHT Ó 2017 BY T HE J OURNAL OF B ONE AND J OINT S URGERY, I NCORPORATED

Intramedullary Nailing for Atypical Femoral Fracture

with Excessive Anterolateral Bowing
Young-Chang Park, MD, Hyung-Keun Song, MD, Xuan-Lin Zheng, MD, and Kyu-Hyun Yang, MD

Investigation performed at the Department of Orthopedic Surgery, Gangnam Severance Hospital, Yonsei University College of Medicine,
Gangnam-Gu, Seoul, South Korea

Background: Intramedullary nailing is the treatment of choice for atypical femoral fractures. However, several problems,
such as iatrogenic fracture and medial gap opening, can occur during intramedullary nailing when atypical femoral
fractures are associated with excessive anterolateral bowing. To overcome these problems, we have developed a new
grading system for anterolateral femoral bowing and a new technique for intramedullary nailing. The core of this new
technique is matching the anterior curvature of the femoral nail with the anterolateral bowing of the femur when the nail
passes the apex of the curvature, by rotating the nail externally.
Methods: From January 2005 through March 2016, 24 female patients (30 cases) who underwent a surgical procedure
for atypical femoral fracture with anterolateral bowing at 2 institutes were evaluated. The postoperative outcomes
(anterolateral bowing grade, anterior and lateral bowing angles, medial gap and posterior gap of the fracture site, iatrogenic
fracture, and time to initial medial callus formation and osseous union) were compared between the new technique
(18 cases) and the conventional technique (12 cases).
Results: With regard to the reliability of the new grading system, the interobserver and intraobserver reliability of the new
grading system demonstrated an almost perfect agreement (kappainter = 0.893, kappaintra = 0.883). For patients with
complete fractures, the differences between the preoperative and postoperative anterior and lateral bowing angles
were significantly less (p = 0.013 for both) in the new technique group. The medial and posterior gaps at the fracture site
were also significantly less in the new technique group (p = 0.013 for the medial gaps and p = 0.022 for the posterior gaps).
Iatrogenic fracture occurred only in the conventional technique group, affecting 2 cases. The time to initial medial callus
formation was significantly shorter (p = 0.033) in the new technique group compared with the conventional technique group.
Conclusions: Our new grading system for anterolateral femoral bowing is convenient and reliable. Furthermore, the new
intramedullary nailing technique with the current intramedullary nail system is appropriate for the repair of atypical femoral
fractures with excessive anterolateral bowing.
Level of Evidence: Therapeutic Level III. See Instructions for Authors for a complete description of levels of evidence.

Peer Review: This article was reviewed by the Editor-in-Chief and one Deputy Editor, and it underwent blinded review by two or more outside experts. The Deputy Editor
reviewed each revision of the article, and it underwent a final review by the Editor-in-Chief prior to publication. Final corrections and clarifications occurred during one or
more exchanges between the author(s) and copyeditors.

A
lthough bisphosphonates significantly decrease the risk of
osteoporotic fracture, recent reports suggest that atypical
femoral fractures are associated with the long-term use of
bisphosphonates1-4. The incidence of atypical femoral fracture is
low, ranging from 3 to 113 per 100,000 person-years; however, the
treatment of atypical femoral fracture is difficult owing to many
complications such as delayed union or nonunion1,2,5-9. Further-
more, Kang et al.10 reported that the incidence of subsequent
atypical femoral fracture in the contralateral femur after the oc-
currence of an atypical femoral fracture was 30%.
Atypical femoral fractures are often associated with antero-
lateral bowing, and the risk is 5 times higher in Asian ethnic
populations than in Caucasians11,12. The current surgical treatment
of choice for atypical femoral fracture is intramedullary nailing1,13.
However, many orthopaedic surgeons encounter considerable
difficulties during intramedullary nailing for a complete fracture,

Disclosure: There was no source of funding for this study. On the Disclosure of Potential Conflicts of Interest forms, which are provided with the online
version of the article, one or more of the authors checked “yes” to indicate that the author had a relevant financial relationship in the biomedical arena
outside the submitted work (http://links.lww.com/JBJS/C810).

J Bone Joint Surg Am. 2017;99:726-35 d http://dx.doi.org/10.2106/JBJS.16.00760

 727
TH E JO U R NA L O F B O N E & JO I N T SU RG E RY J B J S . O RG
d
V O LU M E 99 -A N U M B E R 9 M AY 3, 2 017
d d
as well as in prophylactic nailing, because of excessive femoral
bowing. This is due to a mismatch between the degree of antero-
lateral bowing and the currently available intramedullary nail
systems. Furthermore, the presence of a hard endosteal callus at
the apex of the curve aggravates the difficulties during intramed-
ullary reaming and nail advancement. For these reasons, several
problems frequently occur, such as iatrogenic fracture, straight-
ening of the femur, medial gap opening, leg-length discrep-
ancy, eccentric position of the distal nail tip, delayed union, and
nonunion14-16 (Fig. 1).
A new technique of nailing or a newly designed implant to
overcome excessive anterolateral bowing is needed. Some sur-
geons have attempted to bend the nail according to the curvature
of the femur before the surgical procedure and have then inserted
the pre-bent nail or have introduced a contralateral femoral nail to
overcome the aforementioned mismatch15,16. However, these trials
involve off-label uses of the manufactured femoral nail, which
places the burden of responsibility solely on the surgeon should
complications arise. Alternatively, to overcome problems related
to anterolateral bowing, we developed a new on-label technique
with the current intramedullary nail systems. The core of this new
technique is matching the anterior curvature of the femoral nail
Fig. 1
Complications after conventional nailing. Figs. 1-A and 1-B Complete atypical femoral fracture with grade-III anterolateral bowing (dotted line in Fig. 1-A).
Figs. 1-C and 1-D Intraoperative fluoroscopy. When the nail tip passed the apex in the conventional technique, iatrogenic fracture (arrow) on the lateral
cortex occurred. Fig. 1-E Postoperative radiograph showing straightening of the femur, medial gap opening (arrowhead), and iatrogenic fracture (arrow).
I N T R A M E D U L L A R Y N A I L I N G F O R AT Y P I C A L F E M O R A L F R A C T U R E W I T H
E X C E S S I V E A N T E R O L AT E R A L B O W I N G
with the anterolateral bowing of the femur when the nail passes
the apex of the curvature, by rotating the nail externally.
The purposes of this study were to introduce a new grading
system for anterolateral femoral bowing, to develop a new in-
tramedullary nailing technique for patients with atypical femoral
fracture with excessive anterolateral bowing, and to compare the
postoperative outcomes between the new technique and the con-
ventional technique for intramedullary nailing.
Materials and Methods
New Grading System for Anterolateral Femoral Bowing
W e developed a new grading system for anterolateral femoral bowing ac-
cording to the position of the reference line at the apex of the curve of the
anterolaterally bowed femur on a true anteroposterior radiograph. The refer-
ence line is drawn from the tip of the greater trochanter to the center of the
4
intercondylar notch . Grade 0 (nearly straight) indicates that the reference line
is located in the middle one-third of the medullary canal at the apex of the
curve. Grade I (mild) describes a reference line located in the medial one-third
of the medullary canal. Grade II (moderate) refers to a reference line that begins
outside of the medullary canal medially and passes through the medial cortex.
Grade III (severe) describes a reference line that runs medial to the medial
cortex (Fig. 2). Excessive anterolateral bowing was defined as grade II or III. The
grade was measured on a radiograph of the preinjury (intact) ipsilateral lower
 728
TH E JO U R NA L O F B O N E & JO I N T SU RG E RY J B J S . O RG
d
I N T R A M E D U L L A R Y N A I L I N G F O R AT Y P I C A L F E M O R A L F R A C T U R E W I T H
V O LU M E 99 -A N U M B E R 9 M AY 3, 2 017
d d
E X C E S S I V E A N T E R O L AT E R A L B O W I N G

Fig. 2
Grading system for anterolateral bowing: grade 0, nearly straight (Fig. 2-A); grade I, mild (Fig. 2-B); grade II, moderate (Fig. 2-C); and grade III, severe (Fig. 2-D).

TABLE I American Society for Bone and Mineral Research Task Force 2013 Revised Case Definition of Atypical Femoral Fractures*†

Major Features Minor Features

Minimal or no trauma‡ Generalized cortical thickening of the femoral diaphyses

The fracture line originates at the lateral cortex, substantially transverse, Unilateral or bilateral prodromal symptoms
and may become oblique as it progresses medially
Medial spike in complete fractures; only lateral cortex in incomplete fractures Bilateral incomplete or complete femoral diaphysis fractures
Noncomminuted or minimally comminuted Delayed fracture healing
Localized periosteal or endosteal thickening of the lateral cortex at the
fracture site (“beaking” or “flaring”)

*Reproduced, with permission, from: Shane E, Burr D, Abrahamsen B, Adler RA, Brown TD, Cheung AM, Cosman F, Curtis JR, Dell R, Dempster
DW, Ebeling PR, Einhorn TA, Genant HK, Geusens P, Klaushofer K, Lane JM, McKiernan F, McKinney R, Ng A, Nieves J, O’Keefe R, Papapoulos S,
Howe TS, van der Meulen MC, Weinstein RS, Whyte MP. 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. 2014 Jan;29(1):1-23. Epub 2013 Oct 1. †Changes are in bold. The fracture
must be located along the femoral diaphysis from just distal to the lesser trochanter to just proximal to the supracondylar flare. At least 4 of 5
major features must be present. Minor features are not required but have sometimes been associated with these fractures. ‡As in a fall from a
standing height or less.
 729
TH E JO U R NA L O F B O N E & JO I N T SU RG E RY J B J S . O RG
d
I N T R A M E D U L L A R Y N A I L I N G F O R AT Y P I C A L F E M O R A L F R A C T U R E W I T H
V O LU M E 99 -A N U M B E R 9 M AY 3, 2 017
d d
E X C E S S I V E A N T E R O L AT E R A L B O W I N G

limb, when it was available. Otherwise, a radiograph of the contralateral lower
limb was used.
To evaluate the interobserver and intraobserver reliability of the new
grading system, we randomly selected 100 true anteroposterior radiographs of
femora with bowing grades 0 to III from our hospital database. Then 3 experi-
enced orthopaedic surgeons performed measurements twice at 8-week intervals.
The Fleiss generalized kappa coefficient was used for assessing interobserver and
17
intraobserver reliability . Interobserver reliability was calculated from the first
collected data. The calculated kappa coefficients were interpreted according to the
18
criteria of Viera and Garrett .
Patients
This study was approved by our institutional review board. All patients pro-
vided informed consent for participation.
From January 2005 through March 2016, 68 consecutive patients (88
cases) who underwent a surgical procedure for atypical femoral fracture at 2
institutes were included in this retrospective study. We confirmed that all pa-
tients met the revised case definition of atypical femoral fracture in the 2013
2
American Society for Bone and Mineral Research Task Force Report (Table I) .
Twenty-four patients (32 cases) were classified as grade 0 and 44 pa-
tients (56 cases) were classified as grades I to III. We excluded patients who were
classified as grade 0 (nearly straight), who did not undergo intramedullary
nailing, and whose alignment before injury was unclear because the contra-
lateral femur was not suitable for measurement of the curvature.
We identified a total of 33 patients (43 cases) who underwent intra-
medullary nailing. The new intramedullary nailing technique was applied in 15
patients with 18 fractures (10 complete and 8 incomplete), and the conven-
tional intramedullary nailing technique was applied in 21 patients with 25
fractures (22 complete and 3 incomplete). Three patients were in both treatment
groups because they underwent both procedures: the conventional technique
for the first fracture and the new technique for the subsequent fracture of the
contralateral femur. We used only the Zimmer Natural Nail (ZNN) System in the
new technique group. To compare postoperative outcomes between the new
technique group and the conventional technique group, we selected the ZNN

TABLE II Summary of Patient Demographic Characteristics for the New Technique Group*

Preop. Bowing
Duration of Angle (deg) Callus Pattern Ipsilateral
Patient Fracture Involved Level of Bisphosphonate Anterolateral at Lateral Total Knee
No. Type† Age‡ (yr) Side Fracture Use (yr) Bilaterality Bowing Grade Anterior Lateral Cortex Arthroplasty

1 Complete 76 R Middle 10 Yes I 14.6 8.1 Periosteal No

2 Complete 76 L Middle 3 Yes I 11 8.6 Periosteal No
3 Complete 67 R Middle None No II 14.8 9.4 Periosteal No
4 Complete 71 L Middle 5 No II 13 11.9 Endosteal Yes
and
periosteal
5 Complete 78 L Proximal 9 Yes II 13 9.1 Periosteal No
Complete 79 R Middle 9 Yes II 13 9.1 Periosteal No
6§ Complete 75 L Middle 5 Yes II 18.8 10.9 Endosteal No
and
periosteal
7 Complete 77 L Distal 10 Yes III 18.9 16.5 Endosteal No
and
periosteal
8 Complete 77 L Proximal 4 No III 22.7 17.3 Endosteal Yes
and
periosteal
9 Complete 80 L Middle 7 Yes III 31.1 21.2 Endosteal Yes
10 Incomplete 81 L Middle 10 Yes I 14 8 Endosteal No
and
periosteal
11 Incomplete 51 L Proximal 5 Yes I 11.5 8.1 Endosteal No
and
periosteal
12 Incomplete 81 R Middle None No II 14.8 9.9 Periosteal No
13 Incomplete 54 R Middle 10 Yes II 13.1 12.4 Endosteal No
and
periosteal
Incomplete 54 L Middle 10 Yes II 13 12.5 Endosteal No
and
periosteal
14§ Incomplete 77 R Middle 6 Yes II 18.8 10.9 Endosteal No
and
periosteal
15 Incomplete 66 L Middle 10 Yes II 15.7 11.8 Endosteal No
and
periosteal
16 Incomplete 76 L Middle 8 No III 26.9 24.1 Endosteal No

*The new technique group consisted of a total of 15 patients (18 cases); all patients were female. †Fracture type was divided into complete fracture and incomplete fracture; prophylactic nailing
was performed for incomplete fracture. ‡This was the patient age at the time of the surgical procedure. §Patient 6 (complete) and patient 14 (incomplete) are the same patient.
 730
TH E JO U R NA L O F B O N E & JO I N T SU RG E RY J B J S . O RG
d
I N T R A M E D U L L A R Y N A I L I N G F O R AT Y P I C A L F E M O R A L F R A C T U R E W I T H
V O LU M E 99 -A N U M B E R 9 M AY 3, 2 017
d d
E X C E S S I V E A N T E R O L AT E R A L B O W I N G

cases from the conventional technique group. The conventional technique group
finally consisted of 10 patients with 12 fractures (9 complete and 3 incomplete).
One patient was included in both groups, and, in total, 24 patients (30 cases)
were evaluated.
All patients were women, and the mean age (and standard deviation)
was 72.0 ± 9.68 years in the new technique group and 76.1 ± 9.04 years in the
conventional technique group (p = 0.255). The mean follow-up duration was
16.0 ± 17.9 months in the new technique group and 11.6 ± 8.2 months in the
conventional technique group. There was no difference (p = 0.872) in the
mean duration of bisphosphonate use between the new technique group (7.6
years) and the conventional technique group (7.4 years). For the new tech-
nique group, the preoperative anterolateral bowing grade was grade I in 4
cases, grade II in 10 cases, and grade III in 4 cases. For the conventional
technique group, the preoperative anterolateral bowing grade was grade I in 5
cases, grade II in 5 cases, and grade III in 2 cases. There was no difference (p =
0.612) in the preoperative anterolateral bowing grade between the 2 groups
before the surgical procedure. Of the 18 cases in the new technique group,
3 had previously undergone total knee arthroplasty on the ipsilateral knee
joint (Tables II and III). The total knee replacement implants in these pa-
tients were inserted on the basis of the original alignment with anterolateral
bowing.
Clinical and Radiographic Assessments
Regular follow-ups with radiographs were performed at 6-week intervals until
osseous union was achieved. We reviewed the electronic medical record charts
and radiographs for the following data: age at the time of the operation, sex,
affected side, injury mechanism, history of bisphosphonate intake, bilaterality,
and fracture location.
Radiographic review was performed by using a PACS (picture archiving
and communication system). All length measurements were adjusted relative to
the implant size, to take into consideration the magnification.
We measured the anterolateral bowing grade, anterior and lateral
bowing angles, degree of medial gap in the anteroposterior view and posterior
19
gap in the lateral view , presence of iatrogenic fracture, time to initial medial
callus formation, and time to osseous union except for the lateral cortex. The
bowing angle was measured as the angle between the bisecting lines of the
20
proximal and the distal shaft by using the method of Soh et al. .
Surgical Technique: New Technique
All surgical procedures were performed by the 2 senior authors. We used an
antegrade femoral GT (greater trochanter-starting) nail from the ZNN System.
This nail features a proximal lateralization of 3.1° and provides a dif-
ferent amount of anterior curvature for each nail length (length of 24 to 34 cm
and anterior bow radius of 1,270 mm, length of 36 to 42 cm and anterior
bow radius of 1,400 mm, and length of 44 to 48 cm and anterior bow radius of
1,520 mm).
In the new technique, we rotated the nail externally (clockwise for the
right femur and counterclockwise for the left femur) to a greater extent than
typical in the conventional technique, as the nail passed the apex of the curve.
Thereby, the anterior curvature of the nail accommodated the anterolateral
femoral curvature and the 3.1° proximal lateral bending of the nail added to the
formation of the anterior curvature after the nailing (Fig. 3).
The patient was positioned on the fracture table. We prepared the entry
of the intramedullary nail just inside the tip of the greater trochanter and
overreamed the medullary canal by 1.5 to 2.0 mm. The targeting guide faced
upward toward the ceiling when the nail tip was at the entry point. Then the
targeting guide was rotated externally to the horizontal position automatically,
while a greater trochanter-starting femoral nail was advanced into the med-
ullary canal. The new technique is identical to the conventional technique until
the distal nail tip approaches the apex of the curve.
We then rotated the targeting guide intentionally downward toward the
floor (external rotation) when the nail tip passed the curved portion of the
femur. In general, the targeting guide was directed toward the floor in cases of
grade-II or III femoral bowing. While advancing the nail to the end passing
through the apex, we gently derotated the nail internally about 45° under
fluoroscopic guidance. The reason for rotating the nail about 45° was to allow

TABLE III Summary of Patient Demographic Characteristics for the Conventional Technique Group*

Preop. Bowing Angle

Duration of (deg) Callus Pattern
Patient Fracture Involved Level of Bisphosphonate Anterolateral at Lateral
No. Type† Age‡ (yr) Side Fracture Use (yr) Bilaterality Bowing Grade Anterior Lateral Cortex

1 Complete 75 L Middle 5 No I 8.7 5.3 Endosteal

and periosteal
2 Complete 80 R Middle 5 No I 16 6.6 Periosteal
3 Complete 73 L Middle 10 Yes I 15.1 8.8 Periosteal
Complete 75 R Middle 12 Yes I 15.1 8.8 Endosteal
and periosteal
4 Complete 76 L Middle 7 Yes II 16.8 11.9 Periosteal
Complete 78 R Middle 9 Yes II 16.8 11.9 Endosteal
and periosteal
5 Complete 87 L Middle 8 No II 14.3 11.4 Periosteal
6 Complete 77 R Distal 6 No II 17.2 11.7 Endosteal
and periosteal
7 Complete 85 L Middle None Yes III 20.1 14.3 Endosteal
8 Incomplete 51 R Middle None Yes I 9.9 7.3 Endosteal
and periosteal
9 Incomplete 82 R Middle 7 Yes II 14.6 13.1 Endosteal
and periosteal
10 Incomplete 74 L Middle 5 No III 45.7 20.4 Periosteal

*None of the patients underwent ipsilateral total knee arthroplasty and all patients were female. †Fracture type was divided into complete fracture and incomplete fracture;
prophylactic nailing was performed for incomplete fracture. ‡This was the patient age at the time of the surgical procedure.
 731
TH E JO U R NA L O F B O N E & JO I N T SU RG E RY J B J S . O RG
d
I N T R A M E D U L L A R Y N A I L I N G F O R AT Y P I C A L F E M O R A L F R A C T U R E W I T H
V O LU M E 99 -A N U M B E R 9 M AY 3, 2 017
d d
E X C E S S I V E A N T E R O L AT E R A L B O W I N G

Fig. 3
Changes in nail shape (anteroposterior) depending on the degree of rotation. The picture at 0° shows the left side of the nail when inserted with the
conventional technique. By externally rotating the nail, the anterior curvature of the nail converts into an anterolateral curvature and the proximal lateral
bending supplements the anterior curvature.

interlocking fixation, thus avoiding the neurovascular structures. In excessive
femoral bowing, the degree of internal rotation decreases as the severity of
bowing increases (Fig. 4). Internal rotation of the targeting guide had to be
performed gently under fluoroscopy to avoid causing iatrogenic fracture and
medial gap widening. In that position, inserting the distal interlocking screws is
challenging because drilling should start from the posterolateral corner of the
femur. As the nail is settled in external rotation, it is impossible to insert the
neck-head screw through the nail. Thus, our technique is difficult to use in high
atypical subtrochanteric fractures if fixation of the proximal fragment with the
neck-head screw is necessary. Further details are available in the video gallery at
21
aaos.org .
Statistical Analysis
SPSS version 20 (IBM) was used for all statistical analyses. Continuous variables
are presented as the median with the 25% and 75% quartiles, and categorical
variables as the number and percentage. For continuous variables, the Mann-
Whitney U test was performed, whereas for categorical variables, a chi-square
test or the Fisher exact test was conducted. Significance was set at p < 0.05.
Results
Reliability of the New Grading System
T he interobserver and intraobserver reliability of the new
grading system demonstrated an almost perfect agreement
(kappainter = 0.893, kappaintra = 0.883).
Comparison of the New Technique and the
Conventional Technique
Changes Between the Preoperative and Postoperative Anterolateral
Femoral Bowing Grades
In the new technique group, the bowing grade was sustained (5
of 10) or downgraded by 1 level (5 of 10) after the operations for
complete fracture. By comparison, in the conventional technique
group, the bowing grade was sustained (0 of 9), downgraded
by 1 level (7 of 9), or downgraded by 2 levels (2 of 9). These
differences between the 2 groups were significant (p = 0.022).
Comparison of Postoperative Outcomes: Complete Fractures
There was no significant difference (p > 0.05) in the preoperative
anterior and lateral bowing angles between the 2 groups. The
new technique group displayed a smaller difference between
the preoperative and postoperative angles, which means that the
new technique was better at sustaining the original angle;
the difference was significant for both the anterior and lateral
bowing angles (p = 0.013 for both). The new technique group
also showed significantly smaller medial gaps (p = 0.013) and
posterior gaps (p = 0.022) in the immediate postoperative ra-
diograph. Iatrogenic fracture occurred only in the conventional
 732
TH E JO U R NA L O F B O N E & JO I N T SU RG E RY J B J S . O RG
d
V O LU M E 99 -A N U M B E R 9 M AY 3, 2 017
d d
Fig. 4
Radiographs of patient 8 from Table II. This patient had grade-III anterolateral bowing (dotted line in Fig. 4-A) and atypical subtrochanteric fracture
(Fig. 4-B). The patient underwent a surgical procedure with the new technique (Fig. 4-C) and demonstrated osseous union at 8 months after the surgical
procedure (Fig. 4-D).
technique group, in 2 (22%) of 9 cases (Table IV). Iatrogenic
fracture occurred in 1 grade-II patient and 1 grade-III pa-
tient, which resulted in 2 levels of downgrade after nailing
(Fig. 1).
The time to initial medial callus formation was signifi-
cantly shorter in the new technique group (p = 0.033). The
time to osseous union, defined as a bridging callus in 3 cortices
except for the lateral cortex, was shorter in the new technique
group; however, the difference was not significant (p = 0.190).
Delayed union was defined as a time to osseous union of >6
months, and occurred in 1 (10%) of 10 cases in the new
technique group and in 2 (22%) of 9 cases in the conventional
technique group. These cases healed without any additional
procedure. Nonunion, defined as no progression of healing,
did not occur in either group (Table IV).
Comparison of Postoperative Outcomes: Incomplete Fractures
Prophylactic intramedullary nailing was performed in 8 in-
complete fractures uneventfully in the new technique group
(Fig. 5). In the conventional technique group, 1 of 3 cases
needed osteotomy at the apex to perform conventional intra-
medullary nailing.
I N T R A M E D U L L A R Y N A I L I N G F O R AT Y P I C A L F E M O R A L F R A C T U R E W I T H
E X C E S S I V E A N T E R O L AT E R A L B O W I N G
Discussion
A s atypical femoral fracture is known to be a type of in-
sufficiency fracture2, the implant for fracture fixation, if
indicated, must span the entire femoral shaft to eliminate the
stress riser effect in the lateral cortex at the end of the im-
plant, as would occur with a short plate. For that reason,
intramedullary nailing is the treatment of choice for atypical
femoral fracture22,23. However, atypical femoral fractures are
frequently associated with excessive anterolateral bowing,
which has been a major obstacle in intramedullary nailing,
with an prevalence of 32% in this study and 50% in the study
by Hyodo et al.4. As deformity of the medullary canal is one
of the main contraindications of intramedullary nailing, a
mismatch between the medullary canal and the femoral
nail can be an important issue. After conventional nailing
for atypical femoral fracture with excessive anterolateral
bowing, the femur is straightened, which results in medial
fracture gap opening, possible iatrogenic fracture, and leg-
length discrepancy. Although we could not overcome this
problem completely, we confirmed through this study that
our new technique addresses this problem to an acceptable
level.
 733
TH E JO U R NA L O F B O N E & JO I N T SU RG E RY J B J S . O RG
d
I N T R A M E D U L L A R Y N A I L I N G F O R AT Y P I C A L F E M O R A L F R A C T U R E W I T H
V O LU M E 99 -A N U M B E R 9 M AY 3, 2 017
d d
E X C E S S I V E A N T E R O L AT E R A L B O W I N G

TABLE IV Comparison of Postoperative Outcomes Between the New Technique Group and the Conventional Technique Group*

New Technique Group (N = 18) Conventional Technique Group (N = 12)

Complete (N = 10) Incomplete (N = 8) Complete (N = 9) Incomplete (N = 3) P Value†

Anterior bowing‡ (deg)

Preoperative 14.7 (13.0 to 19.9) 14.0 (11.5 to 26.9) 16.0 (14.7 to 17.0) 14.7 (9.9 to 45.7) 0.842
Postoperative 12.5 (10.5 to 14.7) NA 8.0 (6.6 to 10.0) NA
Difference§ 4.4 (2.0 to 5.8) NA 7.0 (6.5 to 7.9) NA 0.013
Lateral bowing‡ (deg)
Preoperative 10.2 (9.0 to 16.7) 11.8 (8.0 to 24.1) 11.4 (7.7 to 11.9) 11.5 (7.3 to 20.4) 0.497
Postoperative 7.8 (6.5 to 15.2) NA 4.9 (4.3 to 6.8) NA
Difference§ 2.2 (1.3 to 3.7) NA 4.6 (3.1 to 6.7) NA 0.013
Gap#** (mm)
Medial 2.1 (1.2 to 3.9) NA 5.2 (4.9 to 6.3) NA 0.013
Posterior 3.5 (2.6 to 4.3) NA 6.2 (4.6 to 7.2) NA 0.022
Iatrogenic fracture†† 0 (0%) of 10 0 (0%) of 8 2 (22%) of 9 1‡‡ (33%) of 3
Time# (wk)
To medial callus 4.5 (4.0 to 5.3) NA 7.0 (5.0 to 13.0) NA 0.033
To osseous union 21.0 (19.0 to 24.5) NA 28.0 (22.5 to 31.0) NA 0.190
Delayed union†† 1 (10%) NA 2 (22%) NA
Nonunion†† 0 (0%) NA 0 (0%) NA

*NA = not applicable. †The p values are for patients in both groups with complete fractures. ‡The values are given as the median, with the
interquartile range in parentheses, for the complete fractures and as the median, with the full range in parentheses, for the incomplete fractures.
§For the anterior and lateral bowing angles, the difference is the preoperative angle minus the postoperative angle. #The values are given as the
median, with the interquartile range in parentheses. **This was the gap of the fracture site on immediate postoperative anteroposterior and
lateral radiographs. ††The values are given as the number of patients, with the percentage in parentheses. ‡‡Osteotomy was performed for
1 case (grade III) of iatrogenic fracture in the conventional technique group to enable intramedullary nailing.

The bowing deformity needs to be maintained or re-
stored because it would allow minimizing the medial gap, leg-
length discrepancy, and risk of iatrogenic fracture by restoring
the original curvature of the femur. As the “dreaded black
line,”24-26 known as a radiographic feature indicative of stress
fracture nonunion26 and a possible predictor of complete
fracture, is usually located in the lateral cortex, preservation of
a healthy biologic milieu on the medial side of the fracture site
is very important, as healing usually starts medially27. Pre-
sumably, minimizing the medial gap would result in the
outcome that time to initial medial callus formation was
significantly shorter in the new technique group than in the
conventional technique group. In addition to straightening of
the femur, iatrogenic fracture may occur owing to the in-
creased brittleness of the bone, especially in grade-II and III
femoral bowing. Poor nail-to-bone contact at the isthmus
might destabilize the fixation. The changes in the mechanical
axis and length are not desirable after fixation because si-
multaneous bilateral complete fractures are extremely rare.
Changes in the mechanical axis become more critical
with regard to total knee arthroplasty because excessive femoral
bowing is frequently associated with osteoarthritis of the knee
joints (up to 42%)28. For a successful total knee arthroplasty
(functional restoration and durable longevity), postoperative
alignment in the coronal plane should be neutral as postop-
erative malalignment is a common cause of failure in total knee
arthroplasty29,30. Consequently, keeping the original bowing
and alignment is especially important for patients with atypical
femoral fracture with anterolateral bowing who had undergone
total knee arthroplasty. If the original alignment is changed
after nailing, it may compromise the longevity of the total knee
arthroplasty.
Prophylactic nailing for impending atypical femoral
fracture is a method advocated by many surgeons because the
treatment after complete fracture is more difficult and the
result is less promising31,32. However, prophylactic nailing is
more challenging in cases of excessive anterolateral bowing.
The presence of endosteal callus increases the difficulties in
prophylactic nailing due to eccentric reaming. The limited
number of cases (3) of prophylactic nailing with the con-
ventional method in this study can be attributed to this rea-
son. With the new technique, we were able to perform
prophylactic nailing in cases of excessive anterolateral bowing
without any complications.
We introduced a new grading system for anterolateral
femoral bowing that is simple, is easily applied in the clinic, and
shows high interobserver and intraobserver reliability. We de-
picted a line from the tip of the greater trochanter to the center
 734
TH E JO U R NA L O F B O N E & JO I N T SU RG E RY J B J S . O RG
d
I N T R A M E D U L L A R Y N A I L I N G F O R AT Y P I C A L F E M O R A L F R A C T U R E W I T H
V O LU M E 99 -A N U M B E R 9 M AY 3, 2 017
d d
E X C E S S I V E A N T E R O L AT E R A L B O W I N G

Fig. 5
Radiographs of patient 16 from Table II. Figs. 5-A and 5-B This patient had grade-III anterolateral bowing (dotted line in Fig. 5-A) and incomplete
atypical femoral fracture with prodromal symptoms for 2 years (arrow in Fig. 5-A). Figs. 5-C, 5-D, and 5-E The patient underwent prophylactic nailing with
the new technique.

of the trochlea in a true anteroposterior view of the intact preserves the curvature more reasonably after intramedullary
femur. These 2 points represent the ideal end points of the nailing relative to the conventional technique. n
greater trochanter-starting femoral nail because we used this
nail for atypical femoral fractures. Preoperative planning and a
scanogram with an ipsilateral greater trochanter-starting fem-
oral nail in various angles of external rotation are now possible
before the operation (Fig. 3). Young-Chang Park, MD1
The limitation of this study was that it was not a ran- Hyung-Keun Song, MD2
Xuan-Lin Zheng, MD1,3
domized controlled trial, owing to the rarity of atypical femoral
Kyu-Hyun Yang, MD1
fractures with femoral bowing. A larger, multicenter, prospective
study is necessary to confirm the effectiveness of this method, 1Department of Orthopedic Surgery, Gangnam Severance Hospital, Yonsei
especially with regard to the longevity of total knee arthroplasty University College of Medicine, Gangnam-Gu, Seoul, South Korea
after nailing and the development of osteoarthritis of the knee
2Department of Orthopedic Surgery, Ajou University School of Medicine,
joint due to changes in the mechanical axis.
In conclusion, our new grading system for anterolateral Suwon, South Korea
femoral bowing is convenient and reliable. Furthermore, the 3Department of Orthopedic Surgery, The Second Hospital of Jilin
new intramedullary nailing technique with the current intra- University, Changchun, People’s Republic of China
medullary nail system is appropriate for the repair of atypi-
cal femoral fractures with excessive anterolateral bowing. It E-mail address for K.-H. Yang: kyang@yuhs.ac
 735
TH E JO U R NA L O F B O N E & JO I N T SU RG E RY J B J S . O RG
d
V O LU M E 99 -A N U M B E R 9 M AY 3, 2 017
d d
References
1. Shane E, Burr D, Ebeling PR, Abrahamsen B, Adler RA, Brown TD, Cheung AM,
Cosman F, Curtis JR, Dell R, Dempster D, Einhorn TA, Genant HK, Geusens P,
Klaushofer K, Koval K, Lane JM, McKiernan F, McKinney R, Ng A, Nieves J,
O’Keefe R, Papapoulos S, Sen HT, van der Meulen MC, Weinstein RS, Whyte M;
American Society for Bone and Mineral Research. Atypical subtrochanteric
and diaphyseal femoral fractures: report of a task force of the American
Society for Bone and Mineral Research. J Bone Miner Res. 2010 Nov;25
(11):2267-94.
2. Shane E, Burr D, Abrahamsen B, Adler RA, Brown TD, Cheung AM, Cosman F,
Curtis JR, Dell R, Dempster DW, Ebeling PR, Einhorn TA, Genant HK, Geusens P,
Klaushofer K, Lane JM, McKiernan F, McKinney R, Ng A, Nieves J, O’Keefe R,
Papapoulos S, Howe TS, van der Meulen MC, Weinstein RS, Whyte MP. 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. 2014
Jan;29(1):1-23. Epub 2013 Oct 1.
3. Mahjoub Z, Jean S, Leclerc JT, Brown JP, Boulet D, Pelet S, Grondin C, Dumont J,
Belzile EL, Michou L. Incidence and characteristics of atypical femoral fractures:
clinical and geometrical data. J Bone Miner Res. 2016 Apr;31(4):767-76. Epub
2016 Jan 5.
4. Hyodo K, Nishino T, Kamada H, Nozawa D, Mishima H, Yamazaki M. Location of
fractures and the characteristics of patients with atypical femoral fractures: analy-
ses of 38 Japanese cases. J Bone Miner Metab. 2016 Mar 29. [Epub ahead of print].
5. Black DM, Rosen CJ. Clinical practice. Postmenopausal osteoporosis. N Engl J
Med. 2016 Jan 21;374(3):254-62.
6. Weil YA, Rivkin G, Safran O, Liebergall M, Foldes AJ. The outcome of surgically
treated femur fractures associated with long-term bisphosphonate use. J Trauma.
2011 Jul;71(1):186-90.
7. 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 Aug;470(8):2295-301. Epub 2012 Jun 6.
8. Schilcher J. High revision rate but good healing capacity of atypical femoral
fractures. A comparison with common shaft fractures. Injury. 2015 Dec;46
(12):2468-73. Epub 2015 Oct 8.
9. Dell RM, Adams AL, Greene DF, Funahashi TT, Silverman SL, Eisemon EO, Zhou
H, Burchette RJ, Ott SM. Incidence of atypical nontraumatic diaphyseal fractures of
the femur. J Bone Miner Res. 2012 Dec;27(12):2544-50.
10. Kang JS, Won YY, Kim JO, Min BW, Lee KH, Park KK, Song JH, Kim YT, Kim GH.
Atypical femoral fractures after anti-osteoporotic medication: a Korean multicenter
study. Int Orthop. 2014 Jun;38(6):1247-53. Epub 2014 Jan 25.
11. Marcano A, Taormina D, Egol KA, Peck V, Tejwani NC. Are race and sex asso-
ciated with the occurrence of atypical femoral fractures? Clin Orthop Relat Res. 2014
Mar;472(3):1020-7. Epub 2013 Oct 29.
12. Maratt J, Schilling PL, Holcombe S, Dougherty R, Murphy R, Wang SC, Goulet JA.
Variation in the femoral bow: a novel high-throughput analysis of 3922 femurs on
cross-sectional imaging. J Orthop Trauma. 2014 Jan;28(1):6-9.
13. Unnanuntana A, Saleh A, Mensah KA, Kleimeyer JP, Lane JM. Atypical femoral
fractures: what do we know about them?: AAOS exhibit selection. J Bone Joint Surg
Am. 2013 Jan 16;95(2):e8: 1-13.
I N T R A M E D U L L A R Y N A I L I N G F O R AT Y P I C A L F E M O R A L F R A C T U R E W I T H
E X C E S S I V E A N T E R O L AT E R A L B O W I N G
14. 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
Aug;18(7):410-5.
15. Oh HC, Park SJ, Yoon HK. Surgical treatment in atypical diaphyseal femoral
fracture with anterior and lateral bowing. J Korean Orthop Assoc. 2014;49(6):485-9.
16. Park JH, Lee Y, Shon OJ, Shon HC, Kim JW. Surgical tips of intramedullary
nailing in severely bowed femurs in atypical femur fractures: simulation with 3D
printed model. Injury. 2016 Jun;47(6):1318-24. Epub 2016 Mar 4.
17. Fleiss JL. Measuring nominal scale agreement among many raters. Psychol
Bull. 1971;76(5):378-82.
18. Viera AJ, Garrett JM. Understanding interobserver agreement: the kappa sta-
tistic. Fam Med. 2005 May;37(5):360-3.
19. Yang KH, Patel A. Significance of fracture gap in open tibial fracture. Yonsei Med
J. 1995 May;36(2):130-6.
20. Soh HH, Chua IT, Kwek EB. Atypical fractures of the femur: effect of anterolat-
eral bowing of the femur on fracture location. Arch Orthop Trauma Surg. 2015
Nov;135(11):1485-90. Epub 2015 Aug 19.
21. Yang KH, Park YC, Kang DH, Moon HS, Zheng X. Intramedullary nailing for femoral
shaft fracture with excessive antero-lateral bowing. Video presented at the Annual
Meeting of the American Academy of Orthopaedic Surgeons; 2016 Mar 1-5; Orlando, FL.
http://www.aaos.org/store/product/?productid=8638525. Accessed 2016 Dec 7.
22. 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 Sep;30(5):561-7. Epub 2012 May 19.
23. Saleh A, Hegde VV, Potty AG, Lane JM. Bisphosphonate therapy and atypical
fractures. Orthop Clin North Am. 2013 Apr;44(2):137-51. Epub 2013 Feb 21.
24. Hamilton WC. Injuries in dancers. Presented at the Midwinter Trauma Society
Meeting; 1984 Dec 5; Vail, CO.
25. Clanton TO, Solcher BW, Baxter DE. Treatment of anterior midtibial stress
fractures. Sports Med Arthrosc. 1994;2(4):293-300.
26. Koh JS, Goh SK, Png MA, Kwek EB, Howe TS. Femoral cortical stress lesions in
long-term bisphosphonate therapy: a herald of impending fracture? J Orthop Trauma.
2010 Feb;24(2):75-81.
27. Aspenberg P, Schilcher J. Atypical femoral fractures, bisphosphonates, and
mechanical stress. Curr Osteoporos Rep. 2014 Jun;12(2):189-93.
28. Lasam MP, Lee KJ, Chang CB, Kang YG, Kim TK. Femoral lateral bowing and
varus condylar orientation are prevalent and affect axial alignment of TKA in Koreans.
Clin Orthop Relat Res. 2013 May;471(5):1472-83.
29. Ritter MA, Faris PM, Keating EM, Meding JB. Postoperative alignment of total
knee replacement. Its effect on survival. Clin Orthop Relat Res. 1994 Feb;299:153-6.
30. Lombardi AV Jr, Nett MP, Scott WN, Clarke HD, Berend KR, O’Connor MI. Primary
total knee arthroplasty. J Bone Joint Surg Am. 2009 Aug;91(Suppl 5):52-5.
31. Das De S, Setiobudi T, Shen L, Das De S. A rational approach to management of
alendronate-related subtrochanteric fractures. J Bone Joint Surg Br. 2010 May;92
(5):679-86.
32. Oh CW, Oh JK, Park KC, Kim JW, Yoon YC. Prophylactic nailing of incomplete atypical
femoral fractures. ScientificWorldJournal. 2013;2013:450148. Epub 2013 Jan 9.