Orthopaedics & Traumatology: Surgery & Research 102 (2016) 387390

Available online at

ScienceDirect
www.sciencedirect.com

Original article

Results of the Sugioka transtrochanteric rotational osteotomy for

osteonecrosis: Frequency and role of a defect of the quadratus femoris
muscle in osteonecrosis progression
T. Yamamoto , G. Motomura , K. Karasuyama , Y. Nakashima , T. Doi , Y. Iwamoto
Department of Orthopaedic Surgery, Graduate School of Medical Sciences, Kyushu University, 3-1-1 Maidashi, Higashi-ku, 812-8582 Fukuoka, Japan

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

Article history: Background: During transtrochanteric rotational osteotomy (RO), it is important to preserve the posterior
Received 17 July 2015 column artery (PCA), which is generally located in the adipose tissue underneath the quadratus femoris
Accepted 19 January 2016 muscle (QF). If there is a defect in the QF, the risk of injuring the PCA, subsequently resulting in total
necrosis of the femoral head, may increase. Therefore, we investigated: (1) the frequency of defects of
Keywords: the QF at the time of RO, and (2) clinical outcome of RO based upon a defect of the QF.
Osteonecrosis Hypothesis: The presence of defects of the QF at the time of RO could be detected pre-operatively by
Femoral head
magnetic resonance imaging.
Quadratus femoris
Rotational osteotomy
Methods: RO was performed in 124 hips between 2001 and 2010. In all, 95 of the hips were in male
patients and 29 in female patients. The mean age was 45.4 years (range: 1161 years) at the time of
surgery and MRI was performed before RO in all cases. We retrospectively evaluated the progression of
a collapse through 3 years after RO.
Results: MRI showed a defect in the QF in four hips (3.2%) (2 males, 2 females), all of which were conrmed
intra-operatively. Among the four patients, one (25%) underwent total hip arthroplasty because of varus
deformity of the osteotomy site due to total necrosis of the femoral head 1 year after RO. The 120 hips with
a normal QF showed no evidence of total necrosis or progression of necrosis of the femoral head, indicating
that the presence of defects of the QF may increase the risk of poor survivorship of this procedure.
Conclusions: Defects of the QF have been reported to occur in 12% of all patients, whereas in our study the
incidence in ON was approximately 3%. In ON patients with QF defects, pre-operative MRI evaluation of
the QF appears to be important when planning RO, followed by a carefully performed surgical procedure.
Level of evidence: IV; retrospective case series without control group.
2016 Elsevier Masson SAS. All rights reserved.

1. Introduction is to preserve the posterior column artery (PCA), a branch of the

Once subchondral collapse occurs in a patient with osteonecro-
sis (ON) of the femoral head, surgical treatment is necessary
to relieve the pain and to obtain good hip function [1,2]. First
developed by Sugioka in 1978 [3], transtrochanteric rotational
osteotomy (RO) is a joint-preserving procedure used to treat ON.
The procedure involves transposing an intact area to a weight-
bearing portion of the joint, resulting in transfer of the necrotic area
to a non-weight-bearing area. The survival rate has been reported
to be around 80%, while less than 20% of survival rate has also been
reported, indicating that this procedure is technically demand-
ing [48]. One of the most important aspects in performing RO
Corresponding author. Tel.: +81 92 642 5487; fax: +81 92 642 5507.
E-mail address: yamataku@ortho.med.kyushu-u.ac.jp (T. Yamamoto).
medial femoral circumex artery, which supplies nutrients to the
femoral head [9,10]. The PCA is generally located in the adipose
tissue underneath the quadratus femoris muscle (QF), which is a
at, quadrilateral muscle between the gemellus inferior and the
proximal margin of the adductor magnus. In order to expose the
obturator externus, minimum release of QF is recommended, but if
the QF is absent, the adipose tissue is directly exposed, which may
increase the risk of injuring the PCA during the procedure, eventu-
ally resulting in total necrosis of the femoral head. The QF is thus
an important anatomical signpost for performing RO safely.
However the anatomical variations of the QF muscle are not
accurately known. In addition, there is no information whether
the presence of defects of the QF at the time of RO inuences the
survivorship of this procedure. Therefore we investigated the fre-
quency of QF defects at the time of RO in ON patients and also
assessed their inuence on clinical outcomes of RO. The goals of

http://dx.doi.org/10.1016/j.otsr.2016.01.017
1877-0568/ 2016 Elsevier Masson SAS. All rights reserved.
388 T. Yamamoto et al. / Orthopaedics & Traumatology: Surgery & Research 102 (2016) 387390
Table 1
Clinical ndings of the 4 patients with a defect of the quadratus femoris muscle.
Patient number Gender Age Etiology
1 Female 47 Corticosteroids
2 Male 49 Alcohol
3 Female 36 Corticosteroids
4 Male 29 Alcohol
ARO: anterior rotational osteotomy; PRO: posterior rotational osteotomy; Rt.: right; Lt.: left.
the current study were to evaluate: (1) the frequency of defects of
the QF at the time of RO, and (2) clinical outcome and survival of
RO according to the presence of defects of the QF. Our hypothesis
was that the presence of defects of the QF at the time of RO could
be detected pre-operatively by magnetic resonance imaging.
2. Patients and methods
2.1. Patients
The institutional review board at our university approved this
study. Between 2001 and 2010, transtrochanteric RO was per-
formed in 124 hips (95 in males, 29 in females) of 109 patients
whose average age was 45.4 years (range: 1161 years) at the time
of surgery. (During the same period, prosthetic hip replacement
was performed in 200 ON hips, transtrochanteric varus osteotomy
in 34 ON hips, but core decompression was done in none of the ON
cases in our hospital [11,12]). RO was indicated and performed in
patients who: (1) had a necrotic area in the anterior or posterior
portion of the femoral head; (2) had an intact healthy area in the
anterior or posterior portion; and, most importantly, (3) were pre-
dicted to have a postoperative intact area ratio over 34% after the
RO [3,4]. Cases who fullled these criteria have been reported to
have a low risk of progression of a collapse [4].
Based on the Japanese Investigation Committee Classication
System of ON, 77 hips were stage 3A, 43 stage 3B, and 4 stage 4
(3A: collapse less than 2 mm, 3B: collapse over 3 mm), and one hip
was categorized as type B, 18 as type C1, and 105 as type C2 [13].
(According to the Ficat and Arlet classication, 121 hips were in
stage III and 4 in stage IV, while 121 hips in ARCO stage 3, and
4 in ARCO stage 4.) The etiology of ON was idiopathic in 5 hips,
alcohol abuse in 50, post-trauma in 9 (including 11-year-old girl
with ON after slipped capital femoral epiphysis), and corticosteroid
use in 60. The presence of a QF defect was mainly investigated in
all 124 hips based on intra-operative ndings. In addition, pre-
operative MRI was examined whether only high-signal adipose
tissue is observed without any evidence of muscle structure.
2.2. Methods
The basic RO surgical procedure has remained unchanged
since it was described by Sugioka [3]. Briey, it includes
three osteotomies: (1) osteotomy of the greater trochanter; (2)
intertrochanteric osteotomy, which passes from superolateral
to inferomedial on the anteroposterior (AP) view; and (3) an
osteotomy that passes from the proximal are of the lesser
trochanter inferolaterally toward the inferomedial extent of the
primary osteotomy. During these osteotomies, preservation of the
PCA (located in adipose tissue underneath the QF) is the most
important procedure, so minimum release of QF in order to expose
the obturator externus is recommended. To obtain rigid xation,
the K-MAX Adjustable Angle Hip Screw (K-MAX AA Hip Screw;
Japan Medical Materials, Osaka, Japan) as well as two screws are
used. Postoperatively, the patients were not allowed to bear weight
Defect side Necrosis side Pre-operative stage Procedure Total necrosis of the
and type femoral head (months)
Bilateral Bilateral Stage 3A, type C2 Lt. PRO
Left Left Stage 3B, type C2 Lt. ARO
Bilateral Bilateral Stage 3A, type C2 Lt. PRO
Left Bilateral Stage 3B, type C2 Lt. ARO + (10 months)
for 5 weeks, after which gradual weight-bearing was permitted.
Full weight-bearing was permitted after 6 months.
2.3. Methods of assessment
We evaluated radiographic evidence of total necrosis of the
femoral head 3 years after the procedure based on AP and frog lat-
eral views of all 124 hips. QF defects were pre-operatively assessed
on MRI and checked during surgery.
2.4. Statistical methods
The prevalence of total necrosis of the femoral head in patients
with a QF defect and those with a normal QF were statistically
assessed using Fishers exact test. Statistical differences were con-
sidered signicant when P < 0.05.
3. Results
The pre-operative MRI examinations revealed a QF defect in 4
of the 124 hips with ON (3.2%) (2 males, 2 females) (Table 1). The
etiology of the ON was alcohol abuse-related in two patients and
corticosteroid-related in the other two. All the four cases were cate-
gorized as type C2 according the Japanese Investigation Committee
Classication System of ON. The QF defects were conrmed intra-
operatively in all four patients and no false negative cases were
found. Anterior RO was performed in two patients and posterior
RO in two.
One of the four (25%) patients with a QF defect showed varus
deformity of the osteotomy site resulting from total necrosis of the
femoral head 10 months after the RO, resulting in conversion to a
hip prosthesis (Fig. 1). The other three patients showed no evidence
of total necrosis of the femoral head 3 years after RO. On the other
hand, among the 120 RO cases in which the QF was normal, no
hips showed progression of a collapse caused by total necrosis of
the originally intact area 3 years after RO. This incidence was lower
than that for the patients with a QF defect (P = 0.0323).
4. Discussion
A QF defect has been observed in 12% of general patients
in anatomical studies [14,15]. In the present study, the rate was
approximately 3% in patients with ON who underwent RO at our
institution. In addition, one of four patients with a QF defect under-
went total necrosis of the femoral head, whereas cases with normal
QF anatomy showed no evidence of total necrosis. This relatively
high rate of total necrosis may indicate that when the QF is not
observed another type of operation (e.g., transtrochanteric varus
osteotomy) should be considered, although the number with a
defect of QF muscle was only 4 hips and a defect of QF itself may
not have been the direct cause of total necrosis. If the RO is the only
joint-preserving procedure that is deemed appropriate, it is nec-
essary to verify the location of the PCA pre-operatively using MR
angiography or CT angiography to conrm the location of the PCA.
If the PCA is not observed in its normal location, RO is not indicated.
 T. Yamamoto et al. / Orthopaedics & Traumatology: Surgery & Research 102 (2016) 387390 389

Fig. 1. A 29-year-old man with alcohol-related osteonecrosis; a: on A-P view, subchondral collapse more than 3 mm is seen on the left hip; b: on a frog lateral view, the
posterior part of the femoral head is intact and anterior RO was indicated; c: the T1-weighted MRI axial view shows that the quadratus femoris muscle (QF) (circled) on the
right side is normal (arrows). However, on the left side (circled), only high-signal adipose tissue is observed (arrowheads); d: intra-operatively, the QF was not observed, and
only adipose tissue was seen (arrowheads). The other obturator muscles, including the piriformis, obturator internus, gemellus, and obturator externus, were normal. GT:
greater trochanter; LT: lesser trochanter; GM: gluteus medius; PR: piriformis; GS: gemellus superior; GI: gemellus inferior; VL: vastus lateralis; e: RO was performed with
care so as not to damage the PCA or the adipose tissue. Postoperative intact area ratio was obtained around 50%; f: however, bone scintigraphy performed 5 weeks after
RO showed a cold spot throughout the femoral head, indicating that there was no blood supply to it. Ten months after RO, varus deformity of the osteotomy site followed
by a penetration of screw due to total necrosis of the femoral head was observed (black bar in e and f), when the Harris Hip Score was 43 points. Progression of a collapse
is also observed (arrow) at the lateral portion of the femoral head; g: on a frog lateral view, anterior part of the femoral head showed a collapse (arrow). The histological
examination of the resected femoral head revealed total necrosis of the femoral head.

There are several limitations in this study. First, it is unknown
whether a defect of QF muscle itself was a direct cause of total
necrosis of the femoral head in this study. It may be possible that
other unknown factors may have inuenced on postoperative total
necrosis. Secondly, we focused only on ON patients for whom RO
was indicated. Further study that includes other type of osteotomy,
prosthetic replacements and non-operative cases is necessary to
evaluate the precise prevalence of a QF defect in osteonecrosis
cases. Thirdly, the number of cases with a defect of QF muscle was
only 4 cases. In this study, statistical analysis was made based on
this small number of cases and more documented cases are neces-
sary. Finally, the relation between a QF defect and variants of the
PCA could not be analyzed in this study. Imaging studies using MR
angiography or CT angiography are necessary.
The obturator externus, which is located closely adjacent to the
adipose tissue containing the PCA, must be completely dissected
during RO in order to obtain sufcient rotation [3]. Also, the release
of QF is necessary to expose this obturator externus, which should
be minimum not to injure the PCA. In patients with a QF defect,
the adipose tissue containing PCA is directly exposed. These con-
ditions may increase the risk of injury of the PCA during surgery.
Several variants of the PCA have been reported in ON cases, includ-
ing the observation that the PCA is fed by a branch of the internal
iliac artery, not from the medial femoral circumex artery [10,16].
390 T. Yamamoto et al. / Orthopaedics & Traumatology: Surgery & Research 102 (2016) 387390
Because all of the soft tissue containing a branch of the internal iliac
artery needs be circumferentially resected around the femoral neck
during an RO, such an abnormality may cause total necrosis of the
femoral head because of disruption of the blood supply, which may
subsequently result in varus deformity of the osteotomy site due to
total necrosis of the femoral head. One of the four patients in this
study underwent total necrosis of the femoral head despite a care-
fully performed procedure to prevent damage to the PCA and the
medial femoral circumex artery, indicating that the PCA may not
have been branched from the medial femoral circumex artery. In
some cases of a defect of QF, it is possible that the feeding origin of
the PCA is different from that in the normal hip. Thus, pre-operative
evaluation of PCA based on MR angiography and/or CT angiography
would be recommended.
5. Conclusions
QF defects were observed in approximately 3% of our ON
patients. Special considerations are necessary when planning RO
on these patients followed by carefully performed surgery.
Disclosure of interest
T. Yamamoto perceives grants from AMED in relation with this
work, Y. Nakashima declares that he has no competing interest but
perceives fees from Kyocera and Zimmer outside the current study.
G. Motomura, K. Karasuyama, T. Doi and Y. Iwamoto declare that
they have no competing interest.
Acknowledgments
This work was supported in part by a research grant for
intractable diseases from the AMED and a Grant-in-Aid in Scientic
Research (No. 15K10479) from the Japan Society for the Promotion
of Science.
References
[1] Mankin HJ. Non-traumatic necrosis of bone (osteonecrosis). N Engl J Med
1992;326:14739.
[2] Mont MA, Hungerford DS. Non-traumatic avascular necrosis of the femoral
head. J Bone Joint Surg Am 1995;77:45974.
[3] Sugioka Y. Transtrochanteric anterior rotational osteotomy of the femoral head
in the treatment of osteonecrosis affecting the hip; a new osteotomy operation.
Clin Orthop Relat Res 1978;130:191201.
[4] Miyanishi K, Noguchi Y, Yamamoto T, Irisa T, Suenaga E, Jingushi S, Sug-
ioka Y, Iwamoto Y. Prediction of the outcome of transtrochanteric rotational
osteotomy for osteonecrosis based on the postoperative intact ratio. J Bone
Joint Surg Br 2000;82:5126.
[5] Sugioka Y, Yamamoto T. Transtrochanteric posterior rotational osteotomy for
osteonecrosis. Clin Orthop Relat Res 2008;466:11049.
[6] Rosset P, Castaing J. Posterior circumex artery: anatomical study. Technical
interences for circular capsulotomy in Sugiokas femoral osteotomy. Rev Chir
Orthop 1988;74:2038.
[7] Courpied JP. Transtrochanteric rotation osteotomy for femoral head necrosis.
Long-term results. Rev Chir Orthop 1994;80:694701.
[8] Dean MT, Cabanela ME. Transtrochanteric anterior rotational osteotomy for
avascular necrosis of the femoral head: long-term results. J Bone Joint Surg Br
1993;75:597601.
[9] Gautier E, Ganz K, Krugel N, Gill T, Ganz R. Anatomy of the medial
femoral circumex artery and its surgical implications. J Bone Joint Surg Br
2000;82:67983.
[10] Atsumi T, Yamano K, Muraki M, Yoshihara S, Kajihara T. The blood supply
of the lateral epiphyseal arteries in Perthes disease. J Bone Joint Surg Br
2000;82:3928.
[11] Bellot F, Havet E, Gabrion A, Meunier W, Mertl P, de Lestang M. Core
decompression of the femoral head for avascular necrosis. Rev Chir Orthop
2005;91:11423.
[12] Mukisi-Mukaza M, Manicom O, Alexis C, Bashoun K, Donkerwolcke M,
Burny F. Treatment of sickle cell diseases hip necrosis by core decompres-
sion: a prospective case-control study. Orthop Traumatol Surg Res 2009;95:
498504.
[13] Sugano N, Atsumi T, Ohzono K, Kubo T, Hotokebuchi T, Takaoka K. The
2001 revised criteria for diagnosis, classication, and staging of idiopathic
osteonecrosis of the femoral head. J Orthop Sci 2002;7:6015.
[14] Gray H. Anatomy of the human body. 29th edition Philadelphia: Lea & Febiger;
1973. p. 501.
[15] Stibbe EP. Complete absence of the quadratus femoris. J Anat 1929;64(Pt 1):97.
[16] Atsumi T. Hemodynamic study of the idiopathic necrosis of the femoral
head using super-selective angiography. Nihon Seikeigea Gakkai Zasshi
1983;57:35372.