Parents seeking information and advice on the treatment of Perthes’ disease of the hip unknowingly
confront longstanding medical controversies that surround the management of this condition.  After
the initial diagnosis of the condition it is not unusual that parents will receive as many different
recommendations as the number of opinions obtained. So the parents, who are already upset with the
unexpected news that their child has a disease, are placed into the unusual position that they must
pick among widely different medical opinions and procedures.  This adds confusion to an already
troubled situation for parent and child.  With the intent to help parents, I am writing this information
so they can better understand the issues and hopefully make more informed decisions for their child’s

What is Perthes?

Legg-Calve-Perthes disease is an uncommon condition (1/5000) that affects one, or sometimes both,
hips of a child between the ages of 3-15.  Despite the coincidence, the condition isn’t named Legg nor
Calve because these areas are affected.  Rather these names apply because doctors Arthur Legg,
Jacques Calve, and Georg Perthes independently identified the disease in the early part of the 20th
century shortly after x-ray diagnosis tools became available.  The disease is commonly known as
Perthes or simply the abbreviation LCPD.

In adults, there is a condition similar to Perthes called avascular necrosis (AVN) or osteonecrosis of the
hip.  With both of these conditions, and for unknown reasons, the femoral head loses part (or all) of
its blood supply and part (or all) of the bone of the femoral head dies.  (Dead bone in the body is
called “necrotic” bone.)  The actual loss of blood circulation is an acute event and is the root cause of
Perthes.  The “loss of blood supply” event happens without the awareness of the patient and is usually
not associated with obvious symptoms.   In fact, the first symptoms of Perthes may not develop for
several months after the initial “loss of blood supply” episode. Once the femoral head dies it takes
roughly 1-2 years for the body to absorb the dead bone and then 1-3 additional years as the body
attempts to re-grow and re-form the femoral head.   This lengthy timeline and the body’s responses
during the timeline is part of what is called the “natural history” of Perthes.

During the 8 to aproximately18 months after the initial “loss of blood supply” episode, the disease
progresses due to the body’s normal repair mechanisms to remove and replace the dead bone of the
femoral head.  During this repair process the femoral head may be sufficiently weakened such that it
collapses leading to deformation of the joint surface.  This deformation creates a non-spherical femoral
head inside of a spherical socket and ultimately leads to joint stiffness, arthritis and pain.

How does a normal repair process designed to rid the body of dead bone go so awry.  To understand
this it is important to know something about the normal repair process.   The femoral head is made up
of a core of bone surrounded by cartilage. The cartilage of the femoral head gets its oxygen and
nutrition from the fluid in the hip joint.  The bone in the femoral head gets its oxygen and nutrition
from blood vessels that enter the femoral neck from outside the bone.  The femoral neck is separated
from the rest of the femur bone by the growth plate (growing zone) of the upper femur. When there
is blockage or damage to these blood vessels the bone in the femoral head dies due to loss of oxygen
and nutrition.  The cartilage on the outside of the femoral head which surrounds the underlying dead
bone does not die since it gets its oxygen and nutrition by diffusion from the fluid inside the hip joint.
The dead bone is removed by special bone eating cells (osteoclasts).  New living bone is produced by
special bone making cells (osteoblasts).  As the dead bone is removed from the femoral head, the
support of the cartilage of the joint is weakened.  If the surrounding cartilage structure is thick and
strong enough, there will be enough time for new bone to be deposited to replace the removed bone
and there will be no change in shape of the femoral head.  If too much support is removed and the
surrounding cartilage cannot withstand the load, then the weight bearing surface of the femoral head
will slowly collapse under the body’s weight and the normally spherical femoral head becomes flattened
to a more egg shaped configuration.  

To help explain the collapsing process, I like to use the analogy that compares the femoral head (that
has been damaged by a previous loss of blood supply episode), to the renovation of a building.  
Imagine an old building that, although still intact, has severely weakened internal support structures
that must be replaced with new internal supports in order to prevent the roof from collapsing.  If the
“demolition team” removes some old support beams and the rest of the building structure can
withstand this weakening, then the roof will not cave in and there will allow enough time for the
“construction team” to replace the old support beams with new beams.  If, however, too many of the
old support beams are removed before the construction team can replace the supports then the roof
will cave in.  A caved in roof is hard to rebuild and unfortunately can lead to a deformed new roof-
especially if the crashed roof has lots of weight on it like a snow covered roof.

Bringing this analogy back to the femoral head, the roof of the building is the hard, dense outer
surface of the femoral head that is called “subcondral bone” that supports the overlying cartilage of
the femoral head.  The supporting beams within the building are a low-density, honeycomb like bone
called “trabecular bone” that forms the core of the femoral head.  The old supports are dead trabecular
bone that died from lack of blood supply and are eaten away by osteoclast cells.  The new supports
are the reforming trabecular bone installed by the osteoblast cells.  The tons of wet snow on the
rooftop, represents the body’s weight and muscle forces that continuously apply load through the hip
joint tending to crush and permanently deform the weakened, fragile femoral head that while it is
trying to renovate its structure.

The best results in Perthes occur when the roof of the femoral head does not collapse or when it only
minimally deforms, and when there is plenty of time to repair the damage such as in a younger child.
The worst results in Perthes occur when the roof caves in badly and the older child doesn’t have
adequate growing time to repair the damage.

Unfortunately, if the roof deforms, the resulting egg shaped,  femoral head is no longer shaped as
sphere and therefore cannot move in all directions like a ball in a socket.  This reduces the normal
range of motion (ROM).  Furthermore if the shape of the ball (femoral head)  is mismatched with that
of the socket (acetabulum)  there will be abnormal wear and tear in the joint leading first to stiffness,
then to pain and eventually to loss of the joint cartilage (arthritis).  This early osteoarthritis often leads
to total hip replacement (THR) as an adult as early as the 30’s, but more likely the 40’s, 50’s or
beyond depending on the degree of deformation and joint shape mismatch left behind by the Perthes
disease process.  

Why does Perthes’ occur?

Unlike other joints in the body, the femoral head within the hip joint has a very unique, but precarious
blood supply where almost all the blood going to the femoral head is delivered by one main blood
vessel system that must enter the femoral head by traveling across the femoral neck into the hip joint
and finally into the head at its junction with the femoral neck.  Since the hip joint is designed to permit
motion in all planes, these blood vessels must move with the femoral head throughout its complete
range of motion.  In certain positions, these vessels may get kinked as the femoral neck presses
against the edge of the acetabulum.  Moreover, certain mechanical positions of the leg may promote
this kinking for prolonged periods of time.  For example, prone sleeping on the belly with hips
extended and toes pointed outward may restrict blood flow to the femoral head.

There is a lot of interest to as to why some children have loss of blood supply episodes and others do
not.  Research has identified some factors that may predispose children to Perthes disease.
Hypercoagulability (faster clotting of blood) is considered a risk factor.  Hypercoagulability means that
the blood clots abnormally quickly.  It is theorized that the combination of hypercoagulability (which
may be genetic) and mechanical episodes of kinking of the blood vessels of the femoral neck can lead
to clotting off of these arteries and loss of the normal blood flow to the femoral head.  Studies on
hypercoagulability are still inconclusive.  While some studies have shown that patients with Perthes’
have abnormal clotting factors compared with other children, other studies have refuted this claim.  
External environmental factors such as exposure to second hand smoke has also been implicated as a
causative factor.  Repetitive mechanical trauma  (e.g. gymnasts) to the hip joint may also predispose
children to loss of blood supply to the hip joint.  Researchers have also observed that many Perthes
children have delayed skeletal maturity at onset.  The bone age (measured from a hand x-ray) is often
2 years or more behind the chronologic age.  Finally, there are some underlying disease conditions that
increase the risk of Perthes or Perthes-like disorder, such as: sickle cell anemia; hypothyroidism;
epiphyseal dysplasia; Gaucher’s disease; graft versus host reaction after bone marrow transplant;
leukemia.  Such underlying disease conditions also predispose the child to the much less common
bilateral Perthes’ where both the left and right femoral heads are affected.  About 15% of children
afflicted with Perthes in one hip will get it in the other hip.

What is the long-term prognosis for Perthes’?

Ultimately, the long-term prognosis depends on the mechanical shape of the femoral head as it
interfaces to the acetabulum within the hip joint.   However, the final mechanical interface won’t be
known until the disease runs through the lengthy repair process that can last up to 5 years after the
loss of blood supply episode.  So the short-term problem confronting orthopedic specialists is that
they need to examine the hip very early in the process and they need to estimate the final mechanical
configuration and then make informed judgment about the necessity and type of treatment.  

The best method to estimate the prognosis for child can be used very early in the disease process
(only a few months after the no blood episode).  This prediction technique is based on two pieces of
information:  1) the age of the child at onset and, 2) the degree of necrotic bone involvement within
the femoral head.  The older the child when diagnosed, the worse is the prognosis.  Children under
age six have the best prognosis even without treatment.  For children over age of six, the best
prognosis is between 6-8 years of age.  The prognosis worsens between ages 8-10 years and is often
quite poor after age 10.  The relationship between age at onset and prognosis is widely agreed upon
by pediatric orthopedic surgeons.

The other major prognostic factor evident early, is the extent of femoral “head involvement”.   The
smaller the amount of femoral head involved (with loss of blood supply) the lower the risk of long-term
femoral head deformation.  The simplest classification of Perthes’ (Salter-Thompson classification)
divides cases into “partial” vs. “whole head” involvement which corresponds to <50% and >50% of the
head affected respectively.  Obviously the prognosis is better with partial head involvement than with
whole head involvement.  Most agree that partial head involvement cases require no treatment.

What are the treatment alternatives for Perthes’?

If your child is older and has greater than 50% head involvement then treatment may be indicated.  
Treatment for Perthes’ can be generally divided into four categories:

1)        Range of motion (ROM) only treatment
2)        Trans-Neck-Head Tunneling (TNHT)  
3)        Containment Treatment.  (There are 5 different types of containment treatment)
4)        Distraction Treatment

1. ROM Treatment:  Everyone agrees that maintaining hip ROM is a good for patients with Perthes
because it helps prevent stiffness of the femoral head. It is thought to help the injured femoral head
maintain and regain its spherical shape. ROM exercises are a component of all treatment regimens.  
Nevertheless in children, over age six, with whole head involvement, it has been shown that any
treatment is better than ROM alone.  Although ROM has also been combined with prolonged bed rest
or non-weight bearing by crutches or wheelchair, this only provides symptomatic relief initially when
the child is first having pain and limp, and has no proven long-term benefit.  

2. Trans-Neck-Head Tunneling  (TNHT):  The TNHT method was developed in Portugal by Dr. Nuno
Lopes.  He has used it for over 20 years and has drilled over 200 hips.  This technique involves making
a 5 mm hole with a drill, or a device called a trephine, from the lateral side of the femur up the femoral
neck, across the growth plate, and into the center of the dead bone within the femoral head.  The
function of growth plate (physis) of the upper femur is to cause the femoral neck to grow in length
adding to the growth in length of the femur.  The growth plate acts as a barrier to blood vessel
communication between the bone of the upper femur and the femoral head and neck.  In adults when
the growth plate closes and disappears there is a communication between the circulation of the two
parts of the bone.  When there is loss of blood flow from the arteries entering the femoral neck from
the outside the circulation of the rest of the femur cannot provide blood flow to the femoral head
because the growth plate acts as a barrier.  By drilling a hole through the growth plate a portal of
communication is opened between the rest of the femur and the avascular femoral head.  The blood
vessels from the upper femur grow through the hole in the growth plate to resupply circulation to the
avascular femoral head.

Dr. Lopes has observed that drilling leads to more rapid reconstruction of the femoral head  in early
Perthes. Using MRI and ultrasound to screen patients at risk he has identified a pre-Perthes stage of
disease.  He claims that TNHT at this very early stage can lead to complete resolution of the problem
and avert the resorption and reossification phases of Perthes.  His indications for this procedure are
children with whole head involvement who are identified at the pre-Perthes’ (MRI changes only) or very
early Perthes’ stage (e.g. subchondral fracture).  According to Lopes this method is almost universally
successful at the pre Perthes’ stage and about 50% successful at the early Perthes’ stage so that no
further surgical treatment is required. I have been using this method since 2001. My results in early
Perthes are similar to Lopes. External bracing is indicated after TNHT to protect the fragile head.   
Interestingly, TNHT may have a special role to prevent the progression on the opposite hip. The
second hip is identified on the MRI in the prePerthes stage which is when it is completely curable by

For further information on this please refer to Dr. Lopes web site (
pt/nfrancac/).  The attractive aspect of this method is that it is minimally invasive and technically
simple for any orthopedic surgeon to perform. TNHT is similar to a procedure called “core
decompression” has been used in AVN cases in adults for over thirty years. In adults, like children,
drilling only works well in very early stages of disease.

The main risks of TNHT are fracture and growth arrest of the femoral neck.  However, since the hole
diameter is small compared to the overall cross-section of the neck, the risk of fracture is minimal.  
Neither Dr. Lopes nor I have experienced any fractures after TNHT.  Growth arrest of the proximal
femoral growth plate can occur if a bony bridge develops across the drill channel joining the femoral
head and neck with upper femur bone thus tethering further growth by the growth plate.  If a bridge
develops it can stop the growth of the upper femur or if it is too near the edge of the growth plate
this can also lead to a deformity since one side of the growth plate will continue to grow while the
other side might be arrested.  Dr. Lopes research has shown that as long as the drill hole is less than
10% of the area of the growth plate, then the growth plate will not stop growing since the growing
forces within the growth plate will be strong enough to prevent growth arrest. Our results are too
early to comment on this, but certainly we are wary of this potential complication of an otherwise
minimally invasive surgical method. It is important to note that closure of the growth plate of the
femoral neck is a well recognized and common complication of Perthes’ even when no treatment is
performed.  Therefore it may be difficult to separate growth arrest due to drilling versus those due to
Perthes’.  I think this method is promising and may have an expanded role on its own or in
conjunction with other treatments.  

3. Containment Treatment:  In a normal hip joint, the acetabulum only covers two thirds of the
spherical surface of the femoral head.  In a healthy hip joint, this partial coverage is not an issue,
however, if the femoral head has been severely weakened by the death of the underlying honeycomb
bone then there is a serious mechanical problem that contributes to the collapse of the weakened
bone.  To understand the mechanical problem, you must understand the acetabulum is oriented at 45°
to the pelvis facing forward and outward, the front (anterior) and side (lateral).   Thus, the lateral rim
of the acetabulum is the point where the femoral head enters and exits the socket during normal use.  
At the point of mechanical transition from the perfectly molded acetabulum where the load of weight
bearing and muscle compressive forces are distributed over the large surface area of the femoral head
and acetabulum, and the outside uncovered and unloaded femoral head, there is a high stress point
called a “stress riser”.  This “stress riser” and is essentially a high pressure point on the femoral head
which contributes to the collapse of the weakened femoral head.

Returning to the analogy of the building under reconstruction... think of the “stress riser” as a spot
near the center of the roof where the wet snow is much thicker and heavier causing high crushing
pressure on the roof in that spot.  The biggest concern about the  “stress riser” is that the most
common part of the femoral head to be affected by the avascular necrosis is the “antero-lateral
quadrant” which is in exactly the spot where the stress riser is greatest!  This is bad news and is often
where the collapse of the weakened head begins.    

Containment treatment refers to moving more of the uncovered femoral head (especially the affected
part) inside the acetabulum where there is no “stress riser” pressure point.  Mechanically this is
accomplished by moving the femoral head into “abduction” and “flexion”.  “Abduction” is the movement
of the limb to the side (outward like the leg in Jumping Jack position) so that the femoral head tilts into
the socket where it is safe from excessive pressure.   “Flexion” is the movement of the thigh towards
the chest.  In the abducted and flexed position, the antero-lateral (front and outside) quadrant of the
femoral head is covered and protected by the acetabulum. One of the prerequisites for containment
treatment is that the hip must be easily moved into an abducted and flexed position.  If the hip cannot
go into that position easily, then containment treatment is not an option. Containment can be
achieved by external bracing or by several types of surgeries as outlined:   

A)        Abduction Bracing by external bracing systems worn by the child
B)        Femoral Varus Osteotomy surgery
C)        Pelvic Osteotomy surgery
D)        Shelf Augmentation surgery
E)        Combination Pelvic and Femoral surgery

A) Abduction Bracing:  This is a non-operative method of containment.  It is most useful in children
between ages 6-8 years.  While there are numerous types of braces, the best known is the Scottish
Rite abduction brace which has two thigh cuffs and a bar in between them.  Another version of this
type of brace that I prefer is two thigh cuffs connected by a pelvic band with no bar between the
knees.  While it is more flexible, I feel this is an advantage in that it moves the hip through a greater
range of motion as the thighs come together and apart during walking.  For the brace to be effective,
it must to be worn 20 hours a day until reossification of the lateral pillar (outer rim) of the femoral
head. In a child aged 6-8 years old this can mean at least a year and half.  Furthermore, the brace is
only effective if ROM is maintained during treatment.  To assist in ROM, it is sometime necessary to
perform an adductor tenotomy where some hip tendons are cut and actually lengthened. In general,
bracing has gotten a bad reputation because it is often used in the face of a stiff hip and as
mentioned, stiffness is a contraindication to bracing.  Also, in older children, bracing may not be
practical since it must continue for two years or more to be effective and older children often refuse to
wear the brace for cosmetic reasons leading to battles between the patient and parents.

B) Femoral Varus Osteotomy:  The word “varus” means “inward”.  The word “osteotomy” means to
cut bone.  It might be easiest to understand that a varus osteotomy of the upper femur is an
operation where the bone of the upper femur is cut and a wedge shaped piece of is removed so that
the neck of the femur tilts inwards moving the normally uncovered portion of the head inside the
acetabulum.  In affect, the varus osteotomy achieves good coverage, like that obtained with bracing,
but without the need for the brace.

While this method is very good at getting good coverage and containment, it unfortunately does two
bad things to the hip joint.  First it alters the hip mechanics by changing the tension of the hip
abductor muscles as they attach to the “other” top of the femur known as the greater trochanter.  
These muscles, as they attach to the greater trochanter, are responsible for maintaining the pelvis
level when standing on one leg (which occurs with every step).  Therefore the varus osteotomy has
the unfortunate side affect of causing a limp, called a Trendelenburg limp, even after femoral head has
healed from Perthes.  As a result, the varus osteotomy may need to be “undone” surgically at a later
date to correct the Trendelenburg limp.  Secondly, the varus shortens the femur producing or
contributing to a common complication of Perthes called “leg length discrepancy” (LLD).  Children with
Perthes often develop a LLD due to premature closure of the upper femoral growth plate and loss of
height in the femoral head due to flattening.  The surgically induced varus adds to LLD.  Because of
Trendelenberg limp and LLD complications, varus osteotomy is my least preferred surgery for Perthes’.

C) Pelvic Osteotomy:  Pelvic osteotomy is an operation where the pelvic bone is cut in a manner that
permits reorientation of the acetabulum over top the affected anterior and lateral parts of the femoral
head thus improving containment.   The most common pelvic osteotomy is also known as the
“innominate” osteotomy (the innominate bone is one of the three bones of the pelvis).  This surgery is
also known as the Salter osteotomy because it was pioneered by Dr Robert Salter of Toronto,
Canada.  This is a more complex operation than the femoral varus osteotomy with more potential for
blood loss, however, when properly done by an experienced surgeon, excellent coverage can be
achieved.  Some surgeons find it difficult to achieve extensive coverage with this operation, but
because I have been trained by Dr. Salter, I have a personal preference and a great deal of experience
with this procedure.  Its advantages are that the distortion of anatomy produced by this operation are
not clinically significant or problematic as they are with femoral varus osteotomy.  In fact, this
operation tends to add length to the femur that helps compensate for the LLD produced by Perthes’.  
Critics of this procedure claim that the surgery increases the pressure on the soft and reforming
femoral head.  However, if the psoas and adductor tendons are lengthened, as recommended by Dr
Salter, this effect is not significant.   There is another type of pelvic osteotomy called the “triple
osteotomy” where the pelvis is cut in three places and the acetabulum reattached in a manner to
improve coverage over the femoral head.

D) Shelf procedure:  A shelf is a ledge of bone that is added to the existing acetabulum; overtop the
anterior and lateral aspects of the femoral head.  This helps cover and contain the femoral head to
more evenly distribute the body’s weight.  After the shelf has healed, it is believed that it helps prevent
superior (upward) and lateral (outward) subluxation (partial dislocation) of the femoral head from the
acetabulum.  Some consider it the best surgery to prevent subluxation.  The shelf can be done
through a fairly small incision and requires a bone graft.  It does not distort the orientation of the
upper femur or the acetabulum.  Its only down side is that eventually the capsule of the hip that lies
between the femoral head and the shelf is transformed to fibrocartilage. Fibrocartilage may not last
and I have seen cases in which arthritis develops prematurely under the shelf..  One of the most
common techniques for a shelf is called the Staheli shelf procedure.

E) Combined Femoral and Pelvic osteotomies: To achieve more extensive coverage some surgeons
recommend combined femoral and pelvic osteotomies.  This is a very big operation with large potential
blood loss and morbidity.  Combination osteotomies are not common.

For all containment surgeries, hip stiffness must be treated prior to surgery because the hip must be
mobile before one of the osteotomy treatments.  Some surgeons achieve this by performing an
adductor tenotomy (lengthening of the adductor longus tendon) followed by application of Petrie casts
(casts which are long leg casts with a bar between them to keep the legs in wide abduction and are
applied to both legs). After the hip loosens up then the osteotomy surgeries can be performed.

4. Distraction Treatment: This method was initially developed in Italy for a variety of stiff hip problems,
using an external mechanical apparatus which is attached to the patients femur and pelvic bones.  
Distraction treatment is also called arthrodiatasis.  Starting in 1989, I applied distraction treatment for
Perthes disease.  The word distraction simply means pulling apart. I apply distraction using a
mechanical device we call an external fixator (EF).  The EF is attached to the body by connecting to
6mm diameter pins that are drilled through the skin into the femur and pelvis through the skin.  Once
installed, the EF device carries the weight of the body thereby relieving the bones of their normal task
of carrying weight.  In this way, the EF takes the crushing pressures off the injured femoral head and
allows it a chance to re-grow more rapidly and presumably more spherically. The EF removes the
weight off the roof.    

The EF spans the hip joint allowing flexion and extension motion and at the same time holds the femur
in 15° of abduction to provide containment.  However the EF does restrict rotation, abduction, and
adduction motions of the hip.  The EF has special adjustment mechanism that permits gradual
distraction of the hip joint by turning a screw daily during the distraction phase.  The gradual
distraction gently pulls the femur away from the socket to reduce pressure on the head, yet allows
surrounding soft tissues time to stretch.  Distraction is the only method of treatment that truly
unloads the hip joint and allows the injured femoral head a chance to re-grow without significant
pressure bearing down.  Non-weight bearing methods (e.g. crutches or wheelchair) do not take the
load off the hip because of the large muscles and ligaments around the hip keep strong pressures on
the femoral head.  For example, when standing on one leg, we say that the hip joint is sustaining “4X
body weight”.  Such high pressures can be observed by simply turning over in bed.  As another
example, when a patient uses a bed pan, pressures equal to “1-2X body weight” can be induced.  
Given these examples, we know that crutches and wheelchair do not remove high pressures off the hip

One of the repair responses of the femoral head is for the cartilage surrounding the head to be
stimulated to grow especially laterally in the region where it is uncovered and under no pressure.  This
contributes the femoral head getting larger (coxa magna) and the femoral head getting more elliptical
or flatter (coxa plana).  My hypothesis is that if the pressure can be removed from the rest of the
depressed and deformed femoral head it might be stimulated to grow and resume its spherical shape
especially if the spherical acetabulum is there to mold its growth.  This is one of the effects we hope to
achieve by distraction.  The second effect is to reduce the subluxation of the femoral head. The
subluxation is proximal (upward) and lateral (sideward) and anterior (forward).  The distraction we do
is downward which most readily reduces the upward subluxation.  Also, the lateral subluxation is often
reduced by the abducted position of the hip during distraction.  It is difficult to assess and treat the
anterior subluxation.  

To reduce the pressure on the hip and the EF, we perform an adductor and psoas tenotomy (tendon
lengthening) when the EF is surgically applied.  The distraction proceeds at 1/2mm per day until the
upward subluxation is reduced about 1 cm past a preset line which is observed in hip x-rays and is
called Shentons line. During the distraction process, sometimes the hip goes into flexion and refuses
to distract so an additional external rod is connected to the EF which helps corrects the legs tendency
to go into flexion.  

Typically, it takes between 6 and 8 weeks for the proper amount of distraction to be achieved.  After
the distraction process has been completed, the fixator is left in place so that the patient has the EF
for a total of 4 to 5 months. Daily physical therapy (PT) is important to maintain hip flexion and
extension mobility while wearing the EF.  After 4 months for children under 13 and 5 months for
children over 13, the EF is removed under general anesthesia.  After EF removal, an abduction brace is
worn for six weeks day and night and PT and ROM exercises are resumed.  Six weeks after removal
the brace is worn at night only.  It takes several additional months to get all the hip motion and
muscular strength back.   

What is the best age, stage and indications for distraction treatment?  We have applied distraction
treatment successfully for all ages between 6 and 15 yrs.  While it works just as well for younger or
older children, its biggest advantage compared to other methods is for the older child age 9 and older
when the other methods have high failure rates.  Distraction does not work if applied too early during
the subchondral fracture phase, therefore, we recommend that the femoral head at least be in the
early fragmentation-collapse phase. Distraction also works well even at the late resorption and re-
ossification phases.  The best indication for distraction treatment is the child with evidence of collapse,
flattening and subluxation.  

What are the advantages of distraction treatment: There are several advantages of distraction over
the surgical containment methods: 1) distraction does not alter the anatomy of the femur or pelvis
and is a smaller surgical procedure than containment surgery; 2) distraction directly reduces
subluxation of the hip; 3) distraction stretches out the contracted capsule (soft tissues) of the hip
joint; 4) distraction improves the sphericity of the collapsed femoral head; 5) distraction improves the
mobility of a stiff hip; 6) distraction does not require invasive surgery of the hip joint itself.

What are the disadvantages of distraction treatment:  1) The biggest disadvantage of distraction
treatment is the need to wear an external fixator with pins going through the skin for four months.  
The pin sites can become painful and infected.  Pin site infection is easy to treat with oral antibiotic and
good hygiene and rarely leads to deep infection.  Most patients tolerate the EF very well during which
time patients must maintain only partial weight bearing for four to five months using crutches or a
walker in order to minimize undue mechanical stress on the pins.

What are the results of distraction treatment:  I have achieved 95% good or excellent results with
distraction treatment for Perthes, independent of age.  This is the only method where the results do
not correlate with age.  For this reason, it is the treatment of choice in patients over the age of 10 and
at least as good as other methods prior to age 10.  Distraction significantly improves hip range of
motion and especially functional sphericity (hip motion like a sphere).  It also significantly reduces the
subluxation of the hip joint and the spherical shape of the femoral head.  These are considered
positive prognostic factors for longevity of the hip joint. The final shape of the femoral head is the
single most prognostic factor for developing early arthritis of the hip.  The more spherical femoral head
joint will allow good ROM that leads to good longevity for the hip joint.  

Have any other centers performed distraction treatment or corroborated my results?  There have been
several centers around the world that have achieved similar results to mine.  These include centers in
Brazil, Turkey, Israel, India and Portugal.  The group in Sao Paolo, Brazil compared distraction
treatment to containment treatments and concluded that it produced the same or better results. The
group in Turkey initially published poor results with distraction treatment attempted with the Ilizarov
device with a different treatment protocol than mine.  Since their publication they have changed to our
protocol and are now enjoying a similar success rate to mine.  Similar results have been achieved for
the groups in Israel and in Portugal.  Dr. Lopes in Portugal has applied distraction treatment to cases
that are not good candidates for TNHT or who have failed TNHT.  In the US, a few centers have tried
distraction in a handful of cases usually with good success.  In Israel two separate centers have
reported excellent results.

I hope this method will eventually be as available as other surgical methods for the treatment of
Perthes’.  The fact that distraction it is not being used by many in the US is mostly caused by a lack of
training and exposure to this method.

What are the long term results from distraction treatment. Now that I have followed some of my
patients up to 20 years after distraction I realize that in many cases the femoral head is what we call a
Stulberg 3. In the past we considered a Stulberg 3 femoral head shape an excellent results since
arthritis would not develop usually until after age 50. Nevertheless in recent years I have observed that
some patients develop some pain related to impingement and some wear of the hip due to the
aspherical nature of the femoral head. Until 3 years ago we did not have any treatment for this. We
now do in the form of femoral head reshaping osteotomy.

Femoral Head Reshaping Osteotomy

This is the newest surgical treatment and one of the most exciting for Perthes. The femoral head after
healing often is ellipsoid rather than spherical shaped. The acetabulum however often remains
spherical. This causes a larger than normal femoral head to deform the labrum cartilage and the joint
cartilage. Tears in the cartilage lead to pain and wear and arthritis. Even worse than the ellipsoid
femoral head is the saddle shaped femoral head. In both the femoral head is longer in the frontal plane
while it is usually still cylindrical in the sagital plane. I use an osteotomy developed by Dr. Reinhold
Ganz in Switzerland. To my knowledge I became the first surgeon outside of Switzerland to do this
procedure and have performed over 15 of these during the past 3 years. As yet no other center in the
US has performed this procedure.  The central third of the femoral head is removed to change the
shape of an ellipse to a circle. This reduces the size of the femoral head in the frontal plane and
restores the sphericity and containment of the femoral head. I have used this procedure to salvage the
hip in cases who became symptomatic after previous distraction.

How to decide what to do for your child:

If you obtain multiple opinions regarding your child’s Perthes’ you may get as many different opinions
as the number of people you consult.  This is not unusual with Perthes.  It also does not imply that
one opinion is more right than the others.  To help you make your decision for your child let’s review
what most orthopedic surgeons agree about and where the areas of controversy are.  

What everyone agrees upon:

1)        Range of motion is good for the hip
2)        The long-term prognosis is worse with increasing age
3)        Children under age 6 rarely require treatment
4)        Partial head involvement does not require treatment
5)        Whole head involvement in children over age six, have a better prognosis with any type of
treatment than with no treatment
6)        There is no evidence that one method of containment is better than any other method.


1)        Best method of containment treatment:  femoral vs pelvic osteotomy vs brace vs shelf
2)        TNHT- Unproven—should it be used in early cases or is the risk of premature growth arrest
too high
3)        Distraction treatment: better than containment?  Best treatment in older child?
4)        Children should not be treated after age 10 since results of containment are poor?

How to reconcile the controversies for your own child

1)        Best method of containment?;  if you can maintain ROM and tolerate the external brace for 1-
2 years (less in younger children) then go with brace; If you cannot tolerate the brace, then look at
the pros and cons listed above and see what you feel most comfortable with.  Discuss with your
surgeon. Discuss with your child.
2)        TNHT?: go to Dr. Lopes web site.  At this point, it is a low risk procedure and if it works you
win.  If it does not work there is minimal risk.  
3)        Distraction Treatment?: all investigators results seem to indicate that distraction is at least as
good as containment methods.  If planning a containment surgery then consider this as an
alternative.  For children over age 9, distraction may become the method of choice.  For failure with
containment method, distraction can still salvage the hip.  This method may not be available in your
area, therefore you need to decide if you are prepared to travel to obtain this treatment.  
4)        > 10 years don’t treat?: probably true for containment methods.  Not true for distraction.  
Nothing to lose with distraction by doing treatment.  Failed distraction treatment no worse than failed
no treatment.

I hope this helps clarify some of the issues surrounding this controversial disease.  Certainly, I
recommend that you continue to learn all that you can about this condition and then consider the
various factors as you face the important decisions you must make for your child.   Obtain different
opinions; speak with other Perthes parents about their experiences.  This can help guide your process.

Dror Paley, MD
Perthes’ for Parents
by Dror Paley, MD, FRCSC
Back to orthopedic conditions