Orthopaedics & Traumatology: Surgery & Research (2010) 96, 695—701
Irreducible Pipkin II femoral head fractures:
Is transgluteal approach the best strategy?
A.-P. Uzela,∗, G.-Y. Laflammeb, J.-L. Rouvillainc
aDepartment of Orthopaedic Surgery and Traumatology, Pointe-à-Pitre Teaching Hospital Center, Route de Chauvel,
97159 Pointe-à-Pitre, Guadeloupe
bUniversity of Montreal, Division of Orthopaedic Surgery, Sacré-Coeur Hospital,
5400 Gouin Ouest, Local J-3245, Montreal, H4J 1C5 Quebec, Canada
cDepartment of Orthopaedic Surgery and Traumatology, La Meynard Teaching Hospital Center,
97200 Fort-de-France, Martinique
Accepted: 13 April 2010
even less frequent. Truly irreducible fractures such as the two cases in this report must be
differentiated from incomplete reduction due to incarcerated bone or soft tissue interposition.
Opinions vary on the surgical approach to be used once the hip is reduced and the fragment
of the femoral head yet remains to be stabilized. Reports in the literature do not usually take
into account the specificity of irreducible lesions, which in our opinion should be treated by
the transgluteal approach (TGA) while reducible forms can be treated by the Hueter approach.
The transgluteal approach with the patient in the lateral decubitus position provides a direct
anterior view of the antero-infero-medial fracture site as well as dorsal access via the injuries
occasioned to dorsal soft tissues by the posterolateral dislocation. A lag screw can be used with
this approach, which is the only way to stabilize the ligament teres femoris attachment.
Level of evidence: Level IV retrospective historical study.
© 2010 Elsevier Masson SAS. All rights reserved.
Femoral head fracture-dislocations (FHFD) are rare, while irreducible cases are
∗Corresponding author. Tel.: +33 5 90 89 14 66, Secretary office:
+33 5 90 89 10 10x3461; fax: +33 5 90 89 17 44.
E-mail addresses: firstname.lastname@example.org,
email@example.com (A.-P. Uzel).
The incidence of femoral head fracture-dislocations (FHFD)
is between 8 and 26% . This entity is rare, and dislocations
that cannot be reduced are even less frequent, with possi-
ble serious effects to future hip function in these cases. The
main aetiology is a road accident [1—6]. The Pipkin  clas-
sification is most frequently used in these cases, but other
classifications are sometimes mentioned [8—10]. Initial X-
1877-0568/$ – see front matter © 2010 Elsevier Masson SAS. All rights reserved.
696 A.-P. Uzel et al.
rays are often of poor quality. Judet oblique views make it
possible to complete the evaluation. CT-scan is sometimes
difficult to obtain in emergencies, but should be systematic.
Truly irreducible dislocations, where the femoral head can-
not be reinserted into the acetabulum, as in our two cases,
must be differentiated from incomplete reduction caused
by a femoral head fragment or soft tissue interposition. .
Treatment of FHFD is difficult, and the surgical approach
remains controversial. The aim of our study was to pro-
vide technical and anatomical support for the transgluteal
approach (TGA) for the reduction and fixation of irreducible
Pipkin II fractures.
In 1996 and 1999 two patients with irreducible Pipkin
fractures were treated by the TGA by only incising the ante-
rior part of the gluteus medius muscle, which is digastric
with the lateralus vastus muscle. The first case involved a
24-year-old woman who presented with posterolateral dis-
location (Fig. 1) and the second case a 37-year-old man
who presented with posterolateral subluxation, with the
head indented on the posterior rim of the acetabulum
(Fig. 2). Both traumas occurred during road accidents. Close
reduction was attempted with the patient under general
anesthesia, curare induced muscle relaxation and a C-arm,
using a manœuvre associating flexion, adduction and lateral
disengagement of the hip with the patient in the decu-
bitus dorsal position with counter-traction of the pelvis
by an assistant. This was unsuccessful. With the trans-
gluteal approach partial tearing of the pelvitrochanteric
muscles was identified and the femoral head was found to be
indented on the posterior rim of the acetabulum in case2.
Moreover a suprafoveal fragment of the femoral head, still
attached to the ligament teres femoris was found in both
cases. The cartilage on the intact part of the femoral head
II fracture with posterolateral dislocation.
Case1 X-ray obturator oblique view showing a Pipkin
luxation, with the head indented on the posterior rim of the
Case2 antero-posterior X-ray showing a Pipkin II sub-
was explored, and in both cases abrasions were identified
on the femoral head adjacent to the fracture line. Reinte-
gration of the femoral head was extremely difficult due to
a buttonhole effect from the capsulolabral incarceration.
In both cases a femoral head extractor was introduced into
the femoral neck by the greater trochanter to disengage the
femoral head under curare muscle relaxation then manu-
ally and instrumentally free the interposed soft tissues. The
hip was positioned in slight abduction with external rotation
by an assistant to align the femoral head with the intraar-
ticular fragment, which was left attached to the ligament
teres femoris. Reduction was stabilized with two pins, whose
direction was confirmed by frontal and profile slices with the
C-arm (hip at 90◦flexion and maximal abduction) making
drilling and osteosynthesis possible. In case1, the lag screw
system was placed in trochanteric region with one cannu-
lated screw 7mm in diameter (in this case, one of the two
pins broke and was left in place). In case2, the lag screw
fixation was inserted into the femoral head with two cannu-
lated screws 3.5mm in diameter whose heads were buried
under the cartilage.
The posterior capsular
was repositioned and sutured to the capsule through the
opening provided by the partial tearing of the pelvit-
rochanteric muscles from the dislocation. We did not use
suture anchors on these soft tissues because peripheral ele-
ments were still intact.
The rehabilitation protocol in both cases included longi-
tudinal traction for 3weeks so that healing of the soft tissues
could begin, as well as isometric muscle contractions. Walk-
ing rehabilitation without applying weight for a total of
three months was then begun. Post-operative follow-up at
4months showed no limping in either case.
fissure wasrepaired with
Transgluteal approach and irreducible Pipkin II fractures697
necrosis of the medial femoral head corresponding to the ini-
tial fragment, the rest of the head appears to be vascularised,
superior-pole osteoarthritis (The 7-mm screw was removed, a
pin that broke during surgery was left in place).
Case1 at the final follow-up, partial avascular
Case1: fracture union occurred in 3months. At 13years
of follow-up the Merle d’Aubigné and Postel Classification
(PMA) scores were 5/6/6 . On X-ray, density was found
to be restored in the medial half of the femoral head with
bone cysts suggesting avascular necrosis of the medial head
(which was slightly more extensive than the initial fracture
fragment), but the femoral head remained circumferential.
Superior-pole osteoarthritis was also noted with narrowing
of the joint space of more than 50% (Fig. 3).
Case2: at 4months of follow-up the PMA score 
was 6/6/6 and the patient had gone back to working
as a plumber. X-ray revealed the presence of Brooker
stage II ossifications . The follow-up consultation at
the 33rd month showed necrosis of the femoral head with
osteoarthritis of the hip with a score of 4/6/4. Total hip
arthroplasty was performed in June 2004 at 4½years of
follow-up (Fig. 4).
True irreducible dislocations as in our two cases must be
distinguished from incomplete reduction, which may have
many causes: rotation of the fracture fragment around the
ligament teres femoris, impaction of the soft tissues or an
osteochondral fragment . In both our cases a button-
hole effect associated with capsulolabral incarceration in
the acetabulum was the cause of complete irreducibility.
Moreover, in case2 the femoral head was indented on the
posterior acetabular rim.
cular necrosis of the femoral head.
Case2 at the 33rd month of follow-up global avas-
Management of capsulolabral tears
Management of these tears is an important prognostic factor
because the labrum plays a role in hip stability and lubrifi-
cation, as well as in the distribution of mechanical stress,
and resection results in an increase of mechanical loading on
the cartilage of more than 90%. . Thus, theoretically at
least, failure to stabilize the labrum is a negative prognostic
factor. In non-traumatic cases, Sadri  did not find better
results when the lubrum was stabilized with anchor sutures
than after resection of the damaged area. However, if the
tear is more than 1/3 of the circumference of the acetabu-
lar rim, fixation with at least three suture-anchors seems to
be preferable to resection. In a study on femoro-acetabular
impingement, Espinosa et al.  obtained better results in
the group, which underwent labral re-fixation.
Dangers of irreducibility
With a high-energy impact and a large femoral head frag-
ment, irreducibility can weaken the femoral neck, thus the
radiological evaluation must confirm that the femoral head
and neck are intact before attempting reduction. Type III
femoral neck fractures are often a result of accidents dur-
ing orthopedic reduction [2,3,7,16—18]. The review of the
literature of iatrogenic type III fractures by Sy et al. 
noted that 13/14 cases occurred during attempts to reduce
type II fractures. This type of lesion is therefore at high risk
of complications, which should be prevented by a precise
initial diagnosis, gentle manoeuvres during reduction, and
complete muscle relaxation under anesthesia.
Although the treatment of irreducible FHFD fractures can
only be surgical, opinions vary on the type of surgical
A.-P. Uzel et al.
Advantages and disadvantages of the different surgical approaches proposed for the treatment of femoral head fracture-dislocations.
AuthorsApproach Advantage 1 Advantage 2 Disadvantage 1 Disadvantage 2
Butler  Lateral approach with
Good access to the
anterior and posterior
part of the joint
Necrosis due to possible injury
to the circumflex artery during
Pseudarthrosis of the
Roeder et Delee 
and Epstein 
Posterior approach Good access to the
and the posterior wall
of the acetabulum
No MCA injury
Good for the
treatment of Pipkin IV
Difficult to control reductionNo direct fixation of the
Duquennoy et al. Watson-Jones
Visual control of
dislocation and direct
fixation of the
dislocation of the hip
Possible superior gluteal nerve
No dorsal access in the
presence of an
to reduce the hip and
free any incarceration
Stannard et al. Smith Petersen
No MCA injury Visual control of
reduction of the
dislocation and direct
fixation of the
dislocation of the hip
Visual control of
reduction of the
dislocation and direct
fixation of the
dislocation of the hip
Good for the
treatment of Pipkin
type IV fractures
Direct screw fixation
possible if the
femorus is incised
Can only be performed in the
decubitus dorsal position
No dorsal access in
irreducible disloations to
reduce the hip and free
Vielpeau et al.  Hardinge or Thomine
No MCA injury
Performed in decubitus lateral
Dorsal access possible
through soft tissue
damage caused by the
Ganz et al. Digastric
360◦access to the
acetabulum and the
Trochanteric osteosyntesis may
pseudarthrosis (but less
than in with a classic
because the gluteal
medius and the vastus
lateralus remain intact)
Nazarian and Muller
Ludloff approach Anatomical approach
with no incision of
Direct screw fixation,
the ligament teres
femoris is preserved
Control limited to the
Rarely used, near to the
MCA: Medial circumflex artery.
Transgluteal approach and irreducible Pipkin II fractures699
approach to be used to repair and stabilise this entity. To our
knowledge none of the authors except Vielpeau et al. ,
takes into account the specificity of the irreducible lesion,
which, for us, must be treated through the TGA, while
reducible cases can be treated by the Hueter approach. The
advantages and disadvantages of these different approaches
are reported in Table 1. Knowledge of the anatomy of the
medial circumflex femoral artery (MCA) is essential dur-
ing the trochanterotomy because the deep branch of this
artery can be damaged if a retractor placed in the medio-
posterior zone above the trochanter slips . The modified
trochanterotomy described by Siebenrock et al.  has
been proposed for acetabular fractures, and provides access
to the posterior and posterosuperior acetabulum, and to a
small zone of the anterior column above the anterior infe-
rior iliac spine. In order to protect the MCA, these authors
 do not incise the obturator externus or the quadratus
femoris muscles. When a posterior approach is used there
is a risk of damage to the deep branch of the MCA, the
main source of vascularization of the femoral head .
During hip dislocation treated by simple closed reduction,
the incidence of necrosis is 11%. This increases to 31% after
surgical management of dislocation-fractures , in par-
ticular due to iatrogenic damage to the vascularization of
the femoral head, especially the MCA . For Epstein 
there is damage to femoral head vascularization during pos-
terior dislocation, and he does not advise using anterior
approaches which would impede the contribution of the lat-
eral circumflex femoral artery LCA which is still patent,
although its vascular contribution is limited . Khan et
al.  compared blood flow in the femoral head during
hip resurfacing by the anterolateral and posterior approach
after injection of cefuroxime. The concentration of the
antibiotic was significantly higher during the anterolateral
approach, with a significant reduction in vascularization of
the femoral head during the posterior approach . The
modified Watson-Jones approach , which is performed
with the patient in the lateral decubitus position, preserves
the blood supply to the femoral head and provides good
anterior access, but in cases of irreducible posterior dislo-
cation, it would be difficult to implement. With the Ludloff
approach  which is more frequently used by pediatric
orthopedic surgeons, in particular for realignment of con-
genital dislocations of the hip, direct fixation of the femoral
head fragment with a screw is possible but only if the poste-
rior dislocation has been reduced. The approach described
by Ganz et al.  includes a trochanteric flip osteotomy of
15mm of the trochanter and preserves continuity with the
gluteus medius muscle and the vastus lateralus muscle. This
approach makes it possible to surgically dislocate the hip in
the decubitus lateral position and obtain 360◦exposure of
the acetabulum and the femoral head to clearly identify all
We used a TGA with the patient in the lateral decubitus
position. This approach provided direct access to the antero-
infero-medial femoral bone fragment because the incision
exposes the anterior hip joint. Dorsal access is possible
without touching the posterior joint and especially with-
out making existing injuries worse. Indeed our experience
has shown that posterolateral dislocation causes damage to
the pelvitrochanteric muscles, and with this approach any
posterior incarceration of the capsule or the labrum can be
freed manually or with an instrument, which we did in both
our cases. The hip was then disengaged by introducing a
femoral head extractor through the greater trochanter into
the femoral neck. Visualisation is also better than with a
posterior approach because of the anatomy of the acetab-
ulum: the anterior acetabular wall is not as wide, and the
acetabulm and femoral neck are anteverted.
Several variations of the TGA, which can be performed
with the patient in the lateral or dorsal decubitus have been
described including the Hardinge approach . These vari-
which can be more or less anterior in greater trochanteric
region and in the muscle. With the Hardinge approach 
the incision line is in the middle of the gluteus medius. With
our approach only the anterior part of the gluteus medius
was incised with the periostum to maintain a maximum
amount of material for reinsertion. In this way it is similar to
the approach by Thomine et al.  who limit the incision
line to the anterior part of the gluteus medius where it con-
verges with its tendon without touching the vastus lateralis
Screw fixation technique
Direct screw fixation to preserve the attachment of the lig-
ament teres femoris is only possible with the Ludloff 
approach. With other approaches, only lag screw fixation,
which begins in the trochanteric region or the femoral head
makes it possible to preserve the attachment of the liga-
ment teres femoris to the femoral head fragment. Direct
mediolateral screw fixation is impossible because the sag-
gital direction of the fracture line and the medial soft
tissues make it impossible to obtain proper inclination of
the screw even with the hip in maximum lateral rotation.
With the Ganz approach  surgical hip dislocation is pos-
sible and the fracture can be screwed directly but only
after incising the ligamentum teres femoris. In type II frac-
tures, reduction of the fragment is easier when it is still
attached to the ligamentus teres femoris . From an
anatomical point of view, the vessels of the ligament teres
femoris are always present but vary in size [28—31]. For
Sevitt and Thompson  the arteries of the ligament teres
femoris are absent or minimal in the femoral head and any
vascularization is limited to a part of the subfoveal zone.
It should be noted that if this artery is cut during total
hip replacement, coagulation is often necessary, and there
is a non-negligible amount of blood flow. We used non-
absorbable cannulated screws whose heads were buried in
case2, but bioabsorbable polylactide pins have also been
used with success .
Role of hip arthroscopy
Whatever the approach used, arthroscopy of the hip during
surgery makes it possible to evaluate chondral lesions and
to remove any existing foreign bodies. Its prognostic value
is essential, because the presence of impactions and/or
abrasions to the cartilage can compromise the outcome of
the hip. It is essential in reducible dislocations requiring
screw fixation by an anterior approach and where overall
access to the hip is limited, while in our cases, the TGA and
700 A.-P. Uzel et al.
the access provided by the pelvitrochanteric muscle lesions
made direct evaluation possible.
Avascular necrosis of the femoral head is the most serious
complication after surgically treated dislocation, and this
developed fairly rapidly in both our cases. Although 13years
and 9months after surgery the clinical results are satisfac-
tory in our first case, the medial part of the femoral head
appears necrotic with osteoarthritic lesions. Nevertheless,
the fragment was realigned and union was obtained. We do
not believe that the surgical approach played a role in the
outcome of these two hips because at worst this approach
would result in injury to the LCA, which does not play a
major role in vascularization of the femoral head . Pos-
terior dislocation seems to be the cause. Indeed, in case1,
the femoral head is healthy except for the medial fragment.
However, we cannot compare our results to other series in
the literature because there are only two cases. Stannard
et al.  combined their series with that of Swiontkowski et
al.  and Marchetti et al. . Their results showed that
avascular necrosis developed in eight out 40cases treated
by the posterior approach while only two cases of necrosis
occurred in 32cases treated by the anterior approach. These
authors concluded that there was a greater chance of devel-
oping necrosis with the posterior approach; with results that
were nearly significant (p=0.09). The study by Vielpeau et
al.  found an unfavorable outcome in 19cases out of 26,
all types of FHFD combined, with 16cases of osteoarthri-
tis and three cases of severe avascular necrosis. An analysis
of this series after at least 5years of follow-up reported
osteoarthritis in approximately 20% of cases. .
Surgical treatment of femoral head fracture-dislocations
must not worsen existing injuries. There is a high risk of
femoral head fractures in Pipkin II irreducible type dislo-
cations during reduction. The transgluteal approach with
the patient in the lateral decubitus position provides direct
antero-infero-medial and dorsal access to the fracture site
because of musculocapsular lesions caused by the disloca-
tion. The only way to preserve the ligament teres femoris is
by using a lag screw through the greater trochanter or the
femoral head itself.
Conflict of interest statement
No conflicts of interest.
 Vielpeau C, Lanoe E, Delbarre JC, Hulet C. Fracture-luxation
de la tête fémorale. Ann Orthop Ouest 2000;32:61—5.
 Duquennoy A, Decoulx J, Capron JC, Torabi DJ. Les luxations
traumatiques de la hanche avec fracture de la tête fémorale.
À propos de 28observations. Rev Chir Orthop 1975;61:209—19.
 Roeder LF, Delee JC. Femoral head fractures associated with
posterior hip dislocations. Clin Orthop 1980;147:121—30.
 Butler JE. Pipkin type II fractures of the femoral head. J Bone
Joint Surg (Am) 1981;63:1292—6.
 Samaran P, Bellumore Y, Bonnevialle P, Mansat M. Fracture de
la tête fémorale. Revue d’une série de 15cas. Rev Chir Orthop
 Stannard JP, Harris HW, Volgas DA, Alonso JE. Functional out-
come of patients with femoral head fractures associated with
hip dislocations. Clin Orthop 2000;377:44—56.
 Pipkin G. Treatment of grade IV fracture-dislocation of the hip.
J Bone Joint Surg (Am) 1957;39:1027—42.
 Brumback RJ, Kenzora JE, Levitt LE, Burgess AR, Poka A. Frac-
tures of the femoral head. Hip 1987:181—206.
 Yoon TR, Rowe SM. Clinical and radiological outcome of femoral
head fractures. Acta orthop Scand 2001;72:348—53.
 Lafosse JM, Besombes C, Chiron P, Tricoire JL. Luxation et frac-
ture de la tête fémorale. Hip 2002:67—80.
 Merle d’Aubigné R. Cotation chiffrée de la fonction de la
hanche. Rev Chir Orthop 1970;56:481—6.
 Brooker AF, Bowerman JW, Robinson RA, Riley LH. Ectopic ossi-
fication following total hip replacement J Bone Joint Surg (Am)
 Ferguson SJ, Bryant JT, Ganz R, Ito K. An in vitro investigation
of the acetabular labral seal in hip joint mechanics. J Biomech
 Sadri H. Conflit fémoro acétabulaire. In: Duparc J, editor.
Cahier d’Enseignement de la Sofcot, Vol. 94. Paris: Elsevier
Masson SAS; 2007. p. 85—101.
 Espinosa N, Rothenfluh DA, Beck M, Ganz R, Leunig M. Treat-
ment of femoro-acetabular impingement: preliminary results
of labral refixation. J Bone Joint Surg (Am) 2006;88:925—35.
 Zehi K, Karray S, Litaiem T, Douik M. Fracture-luxation
de la tête fémorale. À propos de cas. Acta Orthop Belg
 EpsteinHC. Posterior fracture-dislocations
long term follow-up. J Bone Joint Surg (Am) 1974;56:
 Sy MH, Kinkpé CV, Dakouré PWH, Diémé CB, Sané AD, Ndiaye
A, et al. Fracture du col fémoral compliquant la réduc-
tion orthopédique d’une luxation de hanche. À propos de
4observations. Rev Chir Orthop 2005;91:171—9.
 Gautier E, Ganz K, Krugel N, Gill T, Ganz R. Anatomy of the
medial femoral circumflex artery and its surgical implications.
J Bone Joint Surg (Br) 2000;82:679—83.
 Siebenrock KA, Gautier E, Ziran BH, Ganz R. Trochanteric
flip osteotomy for cranial extension and muscle protection
in acetabular fracture fixation using a Kocher-Langenbeck
approach. J Orthop Trauma 1998;12:387—91.
 Khan A, Yates P, Lovering A, Bannister GC, Spencer RF. The
effect of surgical approach on blood flow to the femoral head
during resurfacing. J Bone Joint Surg (Br) 2007;89:21—5.
 Bertin KC, Röttinger H. Anterolateral mini-incision hip replace-
ment surgery: a modified Watson-Jones approach. Clin Orthop
 Nazarian S, Muller ME. Voies d’abord de la hanche. Tech-
niques Chirurgicales-Orthopédie-Traumatologie. Encycl Méd
Chir, 44—600. Paris: Elsevier; 1998, 36p.
 Ganz R, Gill TJ, Gautier E, Ganz K, Krugel N, Berlemann U. Sur-
gical dislocation of the adult hip a technique with full access to
the femoral head and acetabulum without the risk of avascular
necrosis. J Bone Joint Surg (Br) 2001;83:1119—24.
 Hardinge K. The direct lateral approach to the hip. J Bone Joint
Surg (Br) 1982;64:17—9.
 Biga N, Dujardin F, Duparc F, Thomine JM. Voie d’abord de la
hanche par hémimyotomie antérieure (voie de Thomine). Ann
Orthop Ouest 2003;35:45—6.
 Sudanese A, Ferrari D, Toni A. Fragment fracture of the
femoral head: a description of two cases. Chir Organi Mov
of the hip:
Transgluteal approach and irreducible Pipkin II fractures701 Download full-text
 Tucker FR. Arterial supply to the femoral head and its clinical
importance. J Bone Joint Surg (Br) 1949;31:82—93.
 Chandler SB, Kreuscher PH. A study of the blood supply of the
ligamentum teres and its relation to the circulation of the head
of the femur. J Bone Joint Surg (Am) 1932;14:834—46.
 Howe WW, Lacey T, Schwartz RP. A study of the gross anatomy
of the arteries supplying the proximal portion of the femur and
the acetabulum. J Bone Joint Surg (Am) 1950;32:856—66.
 Sevitt S, Thompson RG. The distribution and anastomoses of
arteries supplying the head and neck of the femur. J Bone Joint
Surg (Br) 1965;47:560—73.
 Prokop A, Helling HJ, Hahn U, Udomkaewkanjana C, Rehm
KE. Biodegradable implants for Pipkin fractures. Clin Orthop
 Swiontkowski MF, Thorpe M, Seiler JG, Hansen ST. Operative
management of displaced femoral head fractures: case-
matched comparison of anterior versus posterior approaches
for Pipkin I and Pipkin II fractures. J Orthop Trauma
 Marchetti ME, Steinberg GG, Coumas JM. Intermediate-term
experience of Pipkin fracture-dislocations of the hip. J Orthop