Femoral Rotational Osteotomy (Proximal and Distal)

DEFINITIONS

Femoral anteversion is the angle in the transverse (rotational) plane between the neck of the femur and the distal femur, as defined by the intercondylar axis.

The term femoral torsion may be used in lieu of femoral anteversion and is favored by those who believe that the torsion often occurs in the femoral shaft rather than the neck.6, 23

Because the femoral neck is typically directed anteriorly, the vast majority of people have femoral anteversion. Femoral retroversion is rare and is present when the neck is directed posteriorly.

ANATOMY

Femoral anteversion is measured as the angle of the femoral neck relative to the distal femoral transcondylar axis. The transcondylar axis is measured along the posterior distal femoral condyles (FIG 1).

PATHOGENESIS

Femoral anteversion varies throughout growth and development, both prior to and after birth.

Femoral anteversion predominates in all age ranges and peaks at birth at a mean of 30 to 50 degrees, decreases to approximately 20 degrees by age 10 years, and 15 degrees by age 15 years.2, 7

During normal development, forces between the hip and anterior iliofemoral ligaments with the hip extended appear to lead to the decrease in femoral anteversion.

The natural remodeling process may be impaired in a variety of circumstances leading to persistent increased femoral anteversion in conditions including developmental dysplasia of the hip (DDH), cerebral palsy (CP), and Legg-Calvé-Perthes.7, 12, 20, 24, 25

FIG 1 • Femoral anteversion is the angle in the transverse plane by which the neck of the femur is directed (forward) relative to the transcondylar or coronal plane.

Femoral anteversion can have a genetic component and has a predilection to occur in certain families.

NATURAL HISTORY

Physiologic Anteversion

As noted, femoral anteversion decreases from 30 to 50 degrees at birth to approximately 20 degrees by age 10 years and 15 degrees by age 15 years.2, 7

Despite these mean values, variability is significant, and femoral anteversion is a common cause of persistent intoeing gait in

children.7, 10

In typically developing children, persistent femoral anteversion is rarely of functional consequence because they have normal balance, strength, and coordination.

Occasionally, the combination of persistent femoral anteversion and external tibial torsion (so-called “malignant malalignment”) may be seen in a child. If this combination results in significant patellofemoral pain and/or patellar instability, osteotomy may be indicated.5,

There is conflicting evidence linking abnormally increased or decreased femoral anteversion with osteoarthritis of the hip and knee.13,

26, 27

Cerebral Palsy

CP affects approximately 1 in 500 children in the United States and is associated with motor and cognitive delays.

CP has long been known to be associated with increased femoral anteversion throughout childhood, thought to be due to developmental delays, contractures and abnormal forces across the hip.3, 4, 9, 15, 20

More recent work has shown that the strongest correlation with femoral anteversion in children with CP is the Gross Motor Functional

Classification System (GMFCS) level.20 For children functioning at GMFCS I, anteversion is near normal, and anteversion increases in a stepwise fashion in GMFCS II and III children, being significantly persistently elevated in GMFCS III, IV, and V children.

Femoral anteversion (with resultant internal hip rotation during gait) is a common cause of intoeing in ambulatory children with CP and contributes to lever arm dysfunction and crouch gait in these children.15, 18, 19

In children with severe nonambulatory CP (GMFCS IV and V), increased anteversion and coxa valga are features of the hip at risk for subluxation or dislocation.20

Increased anteversion is a component of malignant malalignment syndrome, which has been implicated as a source of patellofemoral pain and instability.

P.462

FIG 2 • Classic “W-sitting” in a child with increased femoral anteversion. (Courtesy of Children's Orthopaedic Center, Los Angeles.)

PATIENT HISTORY AND PHYSICAL FINDINGS

Parents often note that children with anteversion often sit in a “W” position and may have difficulty sitting cross-legged (FIG 2). Intoeing is common in children with persistent femoral anteversion, including typically developing and special needs children.7, 10

Intoeing in typically developing children typically does not result in functional limitations, although it frequently results in significant functional deficits (including tripping and falls) in special needs children, such as those with CP.9, 19

Internal hip rotation during gait results in internal knee progression angle throughout the gait cycle. In most children with internal hip rotation during gait, the foot progression angle is also internal.

Foot progression angle may be neutral (or external) if femoral anteversion and internal hip rotation are combined with ipsilateral external tibial torsion and/or significant pes valgus deformity. (The combination of femoral anteversion and external tibial torsion is known as malignant malalignment.)

The rotational profile of the entire limb is checked with the child in the prone position.

In the presence of increased femoral anteversion, hip internal rotation markedly exceeds external rotation.

The amount of femoral anteversion can be assessed by the trochanteric prominence angle test (TPAT). The anteversion is the amount of hip internal rotation (from vertical) when the greater trochanter is palpated most laterally.3, 22

IMAGING AND OTHER DIAGNOSTIC STUDIES

A number of imaging techniques have been described to estimate femoral anteversion, including plain radiography, fluoroscopy, computed tomography (CT), ultrasonography, and magnetic resonance imaging (MRI).1, 11 None of these tests is routinely necessary in the assessment of children with femoral anteversion.

DIFFERENTIAL DIAGNOSIS

Increased anteversion is most commonly encountered in the following situations: Isolated “idiopathic” femoral anteversion

Malignant malalignment syndrome

CP DDH

In addition to femoral anteversion, other common causes of intoeing include the following: Tibial torsion

Foot deformity (including pes varus and/or metatarsus adductus)

Internal pelvic rotation (in some neuromuscular diseases)18

NONOPERATIVE MANAGEMENT

There is no nonoperative intervention which changes the rotational profile of the long bones.

Nonoperative treatment of intoeing is geared toward trying to get the child to position the affected hip(s) in more external rotation during gait. Such interventions may include mechanical devices such as twister cables or derotation straps or use of a home program to teach the child to actively externally rotate the hip(s).

SURGICAL MANAGEMENT

Indications

The most common indication is persistent femoral anteversion associated with intoeing interfering with function in children with CP. The anteversion and intoeing typically impact function by causing tripping and/or lever arm dysfunction.

Isolated idiopathic femoral anteversion rarely requires surgical correction. The rare exceptions are children at, or nearing, adolescence with severe anteversion accompanied by marked internal foot and knee progression angles resulting in functional limitations.

Malignant malalignment (the combination of femoral anteversion and external tibial torsion) results in anterior knee pain and/or patellar instability recalcitrant to nonoperative measures.

Preoperative Planning

Location of the Osteotomy

The first decision which should be made in planning femoral rotational osteotomy is whether to perform the osteotomy proximally or distally.

Benefits of a distal femoral osteotomy fixed with Kirschner wires (K-wires) include (1) smaller incision, (2) less soft tissue dissection, and (3) no need for reoperation for retained hardware (as the K-wires are removed in the office 1 month postoperatively).14, 15

Benefits of a proximal femoral osteotomy (typically fixed with a blade plate) include (1) rigid internal fixation and (2) the ability to correct coxa valga and hip subluxation by including a varus component when planning and performing the osteotomy.14, 15

P.463

Given the relative merits of the two techniques, distal femoral osteotomies are preferred for pure rotational correction in children before adolescence.

Proximal femoral osteotomies are typically preferred for children with coxa valga (often with associated hip subluxation, as in CP, DDH and/or Legg-Calvé-Perthes disease) and in children nearing adolescence (in order to avoid casting).

Even when a distal femoral osteotomy is being considered, an anteroposterior (AP) pelvis x-ray should be performed preoperatively to make sure that there is no significant hip uncovering requiring a proximal femoral osteotomy.

Amount of Correction

Preoperatively, the surgeon determines the amount of intraoperative rotational correction to be performed based on both physical examination measurements and the assessment of gait.

It is imperative that other deformities impacting transverse plane kinematic (rotational) data are thoroughly assessed.8, 14, 15 These include pelvic asymmetry and rotation, tibial torsion, and any foot deformities.

Remember that pes valgus contributes to outtoeing, and pes varus contributes to intoeing.

Physical examination (as described earlier) allows quantification of osseous deformity in these children. Radiographic imaging to assess rotation is not typically necessary.

Gait should be assessed thoroughly, preferably with a computerized gait study, in order to quantify the contributors to transverse plane alignment, including pelvic rotation, hip rotation, tibial torsion, and foot deformity. Although gait analysis is most commonly thought of for children with CP (or other neuromuscular maladies), it is also helpful to quantify gait deviations in typically developing children with significant torsional malalignment who are being considered for surgical intervention.

The amount of rotation performed at surgery is based on the static and dynamic measures. Typically, the distal fragment should be rotated 1.5 to 2.0 times the amount of abnormal internal hip rotation on gait preoperatively.15, 17 (For example, if there is 20 degrees of internal hip rotation preoperatively, the femur should be rotated 30 to 40 degrees intraoperatively.) Less derotation results in

undercorrection.

If a proximal femoral osteotomy is planned (based on patient age and/or radiographic findings of coxa valga and/or hip subluxation), plain pelvis radiographs facilitate the planning regarding any varus correction to be incorporated into the osteotomy.

FIG 3 • Infant, toddler, child, and adolescent size blade plates show relative sizes. Small fragment (3.5-mm cortical) are used with the infant and toddler plates, and large fragment (4.5-mm cortical) with the child and adolescent plates. (Courtesy of Children's Orthopaedic Center, Los Angeles.)

Hardware Choice

Proximal Femoral Osteotomy

Proximal femoral osteotomies are typically fixed with blade plates. A plate should be chosen with a long enough blade to ensure stable fixation while stopping short of the proximal femoral physis (FIG 3).

For small children, a rough guide for AO (Synthes, Paoli, PA) blade plate size follows:

Less than or equal to 18 kg: AO “infant” blade plate Eighteen to 24 kg: AO “toddler” blade plate

Greater than or equal to 25 kg: AO “child” blade plate

Distal Femoral Osteotomy

Three or four 2.4-mm (3/32 inches) K-wires are typically used for fixation.

Positioning

Proximal and distal femoral osteotomies are both easily accomplished with the patient supine on a radiolucent operating table. Supine positioning also facilitates the performance of numerous other concomitant procedures, particularly in children with CP or other neuromuscular diseases.

Some other authors advocate the prone position for proximal femoral osteotomies and cite the ability to perform the same intraoperative assessment of the torsional profile as the preoperative assessment.21

Prone positioning precludes pelvic osteotomy and hip flexor lengthening.

Approach

Proximal Femoral Osteotomy

A standard lateral approach is used.

The proximal end of the incision is at the level of the vastus ridge (origin of the vastus lateralis).

Distal Femoral Osteotomy

A direct lateral, subvastus approach is used.

The distal tip of the incision is at the level of the physis. With experience, in thin children, the surgery can be accomplished through a 4- to 5-cm incision, although larger incisions (typically 7 to 10 cm) are typical until ample experience is gained.

P.464

TECHNIQUES

Proximal Femoral Derotational Osteotomy with Blade Plate: Supine Technique

Positioning, Incision, and Exposure

The patient is positioned supine on the radiolucent table with a small “bump” placed under the sacrum.

The image intensifier is set up on the contralateral side of the patient with the image monitor at the foot of the bed. (The image intensifier is switched to the other side of the bed subsequently for bilateral procedures.)

Because most cases require bilateral surgery, both lower extremities are typically prepped and draped free. The prep is typically up to the level of the 12th rib, particularly if open hip reduction and/or pelvic osteotomy may be needed.

The proximal end of the incision is at the level of the vastus lateralis ridge.

Typically, when gaining experience with the procedure, the incision is 3 to 5 cm longer than the selected blade plate. In thin children, as experience is gained, the incision length is typically only slightly (˜1 to 2 cm) longer than the blade plate (TECH FIG 1A).

Electrocautery is used extensively in the dissection to minimize bleeding in these often small children, with correspondingly low blood volumes.

TECH FIG 1 • Exposure of the proximal femur through a standard lateral approach. A. The incision starts proximally at the vastus ridge and extends in line with the femur distal enough to accommodate the appropriate size plate. B. The tensor fascia lata (TFL) is exposed (and subsequently incised) in line with the skin incision. C. After the TFL is split, the vastus lateralis is visualized. Sometimes, adequate visualization of the vastus lateralis is facilitated by removal of the trochanteric bursa. The arrow points to the gluteus maximus inserting into the femur. D. Detaching the vastus lateralis from the trochanteric ridge in L-shaped cut. E. Exposure of the lateral surface of proximal femur, with the vastus lateralis reflected anteriorly. The pickups are holding the periosteum, and the cut is just anterior to the linea aspera and gluteus maximus insertion to facilitate subperiosteal dissection. F. A typical set of Crego elevators used for subperiosteal dissection. (Courtesy of Children's Orthopaedic Center, Los Angeles.)

The tensor fascia lata is identified (TECH FIG 1B) and then split in line with the skin incision, thus exposing the vastus lateralis (TECH FIG 1C). Exposure of the vastus lateralis may be facilitated by removal of the trochanteric bursa.

The vastus lateralis is elevated transversely off the vastus ridge and extended in an “L” fashion in order to facilitate exposure and later closure (TECH FIG 1D).

Staying just anterior to the insertion of the gluteus maximus and the linea aspera with the dissection makes the dissection much easier as the periosteum easily peels off the femur.

Circumferential subperiosteal dissection is needed at the osteotomy site in order to allow appropriate rotational correction. This is usually best accomplished using a set of curved (Crego-type) elevators (TECH FIG 1E).

Preparation for Osteotomy and Fixation

A guide pin is inserted with the aid of the image intensifier. The guide pin insertion point is typically approximately 1 cm distal to the greater trochanteric apophysis and should be located in the mid-coronal plane of the femur (TECH FIG 2A).

Prior to pin insertion, the hip is internally rotated until the greater trochanter is most lateral, indicating that the femoral neck is parallel to the ground. With the hip in this position, if the guidewire is inserted parallel to the floor, it should be parallel to the axis of the femoral neck in the transverse plane.

P.465

The angle of the guidewire and chisel in the coronal plane is based on whether or not a varus component is needed for the osteotomy.

Although using fluoroscopy in the AP view, the guidewire should be inserted in the direction desired for the ultimate position of the blade plate chisel while maintaining the wire parallel to the ground to allow appropriate position in the femoral neck on the lateral view (TECH FIG 2B).

Because it is often difficult to read the depth directly off the blade plate chisel, it is easiest to insert the guidewire to the depth the seating chisel will be inserted. The depth of the wire in the femur should be equal to the length of the blade of the blade plate to be used. This is most easily accomplished by placing a second K-wire adjacent to the guidewire and advancing it to the lateral femoral cortex. The difference in the lateral prominence of the second wire relative to the first is the depth of the guidewire in the femur, which measures 38 mm in this picture (TECH FIG 2C).

The chisel chosen must be the one which matches the size of the blade plate (ie, infant, toddler, child, adolescent, or adult).

P.466

It is important to remember that the chisel used for child and adolescent plates is the same.

TECH FIG 2 • Fluoroscopy-assisted chisel insertion. A. A guidewire is inserted into the proximal femur in the mid-coronal plane, typically starting ˜1 cm distal to the vastus ridge. The hip is internally rotated until the greater trochanter is directly lateral prior to wire insertion and the wire inserted parallel to the floor. B. Position of the guide pin on the AP view. This pin position is more vertical than for isolated rotation, unless the plate has an angle matching the pin-shaft (and chisel-shaft) angles. C. The guide pin should be inserted to the depth to which the chisel will be inserted. The depth is confirmed by manually placing a second pin adjacent to the first pin and placing it against the lateral femoral cortex. The difference between the prominence of the second wire relative to the first is the depth of the guidewire in the femur, which measures 38 mm in this picture. D. Insertion of the seating chisel parallel to the guide pin. The chisel should be backed out several millimeters from the fully inserted position prior to osteotomy in children with strong bone, although it was rarely necessary in small children being treated with infant or toddler blade plates. E. Fluoroscopic imaging showing the chisel inserted parallel, and just cephalad to, the guide pin on the AP view. F. Position of the seating chisel in the center of the neck on the frog lateral view. Because a cannulated system was not used, the chisel does not need to be completely parallel to the pin on this view. (continued)

TECH FIG 2 • (continued) G. The chisel-shaft angle is the angle between the chisel and the shaft of the femur. For a pure rotational osteotomy, the chisel-shaft angle should equal the angle of the plate to be inserted. For varus osteotomies, the chisel-shaft angle exceeds the angle of the plate to be inserted, and for valgus osteotomies, the chisel-shaft angle is less than that of the plate. (A-F: Courtesy of Children's Orthopaedic Center, Los Angeles.)

The seating chisel is inserted immediately cephalad to the guidewire (TECH FIG 2D). Several AP and lateral fluoroscopic images are obtained during chisel insertion to make sure that the position is optimal on AP and frog lateral views.

Care must be taken to insert the chisel such that the face of the chisel is perpendicular to the shaft of the femur in the sagittal plane (in order to avoid flexion or extension at the osteotomy site).

The chisel is inserted to the tip of the guidewire and checked for position in AP and lateral planes (TECH FIG 2E,F). This is an easier way to insert the chisel to the correct depth than if one attempts to read the depth off the underside of the chisel when it is in the femur.

For a pure rotational osteotomy, the blade plate should be placed in a direction so that the “chisel-shaft angle” (the angle between the chisel and the shaft of the femur) equals the angle of the blade plate to be used (TECH FIG 2G).

Although many authors advocate using a 90-degree blade plate for proximal femoral rotational osteotomies, a plate with a larger angle (≥100 degrees) typically allows the surgeon to insert the blade deeper into the femoral neck, enhancing fixation, and facilitating early weight bearing.

For an osteotomy which requires a varus component, the chisel should be inserted at an angle which will result in the necessary varus correction. (For example, if 20 degrees of varus is desired, the chisel-shaft angle should exceed the angle of the blade plate by 20 degrees. In other words, if a 90-degree blade plate were going to be used, the guidewire and blade plate chisel should be inserted at an angle resulting in a 110-degree chisel-shaft angle.)

If a cannulated system is used, the guidewire must be inserted in precisely the desired position for the chisel and blade plate. As a result, this may add time to the surgery as the guide pin location must be exact.

In ambulatory children with healthy bone, the seating chisel should be intermittently backed during insertion to facilitate chisel removal prior to blade plate insertion. The chisel should be backed out from its deepest insertion point prior to the osteotomy

for the same reason.

Osteotomy

The level of the transverse osteotomy is marked with electrocautery perpendicular to the femoral shaft. The distance from the chisel insertion site is based on which side plate will be used: 1 cm for infant plates; 1.2 cm for toddler plates; and 1.5 cm for child, adolescent, and adult plates (TECH FIG 3A).

If a femoral shortening will be performed (typically indicated when performing bilateral varus derotational osteotomies [VDROs] in children with CP), then a second osteotomy site should be marked distal and parallel to the first osteotomy site. Typically, when shortening is performed, it should be 1.5 to 2.0 cm.

Two smooth derotation pins (1.6- or 2.0-mm K-wires) are placed in an anterior to posterior direction parallel to each other in the transverse plane, one just distal to the blade plate chisel and the second distal to the most distal transverse osteotomy (TECH FIG 3B).

A transverse osteotomy is made using an oscillating saw at the osteotomy site marked 1 to 1.5 cm distal to the chisel based on plate size (as mentioned earlier) (TECH FIG 3C).

A large Crego elevator is used to protect the anterior and medial structures, and a medium or large Chandler elevator protects the posterior structures.

For children with strong bone, frequent irrigation is used to minimize the risk of thermal necrosis of the bone at the

P.467

osteotomy site. The periosteum is elevated circumferentially at this level to allow placement of protective retractors during the osteotomy.

TECH FIG 3 • Identification of osteotomy site(s) and completion of osteotomy(ies). A. The site for the osteotomy is localized with the use of a ruler. The distance of the osteotomy from the chisel insertion site is 1.0 cm for infant blade plates, 1.2 cm for toddler plates, and 1.5 cm for child and adolescent plates. B. Two parallel AP guide pins proximal and distal to the transverse osteotomy line(s) are used to judge the magnitude of derotation. In this case, there are two transverse osteotomies to be made to allow femoral shortening in this child with CP and hip subluxation. The location for the osteotomies are denoted with arrows. C. Transverse osteotomy (perpendicular to the long axis of the femur). This is the only osteotomy needed if the osteotomy is being done for derotation only. D. When varus is a necessary component of the osteotomy (as in most children with CP and other neuromuscular disorders), this cut starts halfway across the femur and should be perfectly parallel to the chisel in all planes if a

90-degree plate is being used. This medial closing wedge osteotomy is not needed if pure derotation is performed without varus. (Courtesy of Children's Orthopaedic Center, Los Angeles.)

For pure rotational osteotomies, the transverse osteotomy cut is the only one needed. For VDROs, two or three osteotomy cuts are needed:

The transverse osteotomy is described earlier.

In children in whom a femoral shortening is being performed (particularly in children with spastic CP), a second transverse osteotomy cut is made parallel to the first cut.

In all cases of varus osteotomy, a medial closing wedge osteotomy of the proximal fragment is performed; this osteotomy does not start at the lateral cortex but rather approximately 50% of the distal from lateral to medial. If a 90-degree plate is to be used, this osteotomy should be parallel to the blade plate chisel in all planes (TECH FIG 3D).

If a plate other than a 90-degree blade plate is to be used, then the osteotomy needs to be adjusted accordingly. For an 80-degree plate, the osteotomy should angle 10 degree toward the chisel. For plates with angles greater than 90 degrees, the osteotomy should angle away from the plate (10 degrees away for 100-degree plates, 20 degrees away for 110-degree plates, 30 degrees for 120-degree plates, etc.).

Derotation and Fixation

The blade plate chisel is removed and replaced by the previously chosen blade plate (TECH FIG 4A).

The position of the proximal femur is controlled (typically with anteriorly and posteriorly placed Hohmann retractors) prior to chisel removal.

The blade plate can often be inserted at least part-way into the path created by the blade plate chisel with gentle manual pressure. A mallet can be used to gently tap the plate into the femur along the same path. Forceful striking with the mallet should be avoided in order to minimize the risk of creating a new path and penetrating the cortex medially, anteriorly, and/or posteriorly.

The plate is impacted into final position with the impactor and mallet (TECH FIG 4B).

TECH FIG 4 • Blade plate insertion and osteotomy fixation. A. After chisel removal, the blade plate is inserted in the same path. Whenever possible, inserting the plate as far as possible with gentle manual pressure ensures correct insertion angle and alignment. The mallet is then used to gently tap the inserter and advance the plate deeper into place. B. After the insertion handle is removed, an impactor is used to complete seating of the blade plate. It is common that the proximal fragment is quite flexed (analogous to proximal femur fractures) following these osteotomies; this will be important to remember in the next step. (continued)

P.468

TECH FIG 4 • (continued) C. The osteotomy is reduced (with appropriate derotation), stabilized with a Verbrugge bone-holding clamp, and the amount of rotation can be assessed by measuring the angle between the two pins with a sterile goniometer. Of note, reduction is often facilitated by flexing the hip to 90 degrees prior to reduction. D. The blade plate is held against the femur with a Verbrugge clamp. E. Fixation following cortical screw insertion. (Courtesy of Children's Orthopaedic Center, Los Angeles.)

The distal femoral fragment is approximated to the plate with a bone-holding clamp such as a Verbrugge clamp. Reduction of the osteotomy is facilitated by flexing the hip to approximately 90 degrees.

Prior to application of the Verbrugge clamp, the distal femoral fragment is externally rotated until the angle between the previously parallel guidewires matches the amount of derotation desired (TECH FIG 4C). A sterile goniometer is used to confirm the amount of correction.

Screws are used to fix the plate to the distal femoral fragment (3.5-mm cortical screws for infant and toddler blade plates and 4.5-mm screws for the child, adolescent, and adult plates). At least one of the screws in the distal fragment is inserted in compression to hasten bone healing (except for infant plates, which cannot be compressed) (TECH FIG 4D, E).

After fixation with one or two distal screws, the arc of hip rotation is evaluated. External rotation should typically exceed internal rotation at this time.

Wound Closure

The vastus lateralis is approximated to the vastus ridge and repaired with two figure-8 sutures of size 0 absorbable suture, such as Vicryl (Ethicon, West Somerville, NJ). A running size 0 absorbable suture is used to close the posterior aspect of the vastus lateralis (TECH FIG 5).

The fascia lata is closed with a running size 0 absorbable suture.

The subcutaneous tissue is closed with simple, interrupted, inverted 2-0 absorbable sutures and a running 3-0 absorbable monofilament, such as Monocryl (Ethicon, West Somerville, NJ), is used to close the subcuticular layer.