Femoral Shaft Fractures

Skeletal and skin traction can be used to regain length
to the limb. Skin traction has limited utility because of the inability
to apply sufficient and sustained forces without damaging the skin.
This makes the use of skin traction a poor choice as a definitive
treatment modality. However, in the initial stabilization of a patient
in the field before evaluation at a hospital, some form of noninvasive
traction using a temporarily applied splint may assist with aligning
the limb, providing patient comfort, and limiting additional soft
tissue injury caused by the mobile femoral shaft fracture segment.

Skeletal traction for the femur can be applied after
placement of an appropriate pin. This can be placed at the distal femur
or the proximal tibia. Small-diameter pins (2-mm) can be optimized with
the use of tensioned bows, obviating the need for large threaded
Steinman pins. A traction pin should be placed with sterile technique
if possible, and knowledge of the location of the relevant anatomical
structures in the region of the pin entry and exit locations should
minimize complications. The entry and exit tracts for the pin should be
appropriately anesthetized with local anesthesia. Safe pin placement is
usually from medial to lateral at the distal femur and from lateral to
medial at the proximal tibia. Placement at the distal femur as opposed
to the proximal tibia for fractures of the femoral shaft allows for
direct traction of the fractured bone and avoids distraction across the
potentially injured knee joint. Distal femoral pins should be placed in
an extra-articular location to avoid septic arthritis. Proximal tibia
pins are typically positioned at the level of the tibial tubercle and
placed in a bicortical location.

Temporary traction in anticipation of definitive
operative femoral fracture treatment may consist of the application of
weights attached to pulleys at the end of the bed, or some form of
balanced suspension. Ideally, traction applied to the distal femur or
proximal tibia is balanced against countertraction applied to the
proximal femoral segment. Proximal countertraction can be accomplished
with the weight of the patient or with Trendelenburg positioning. For
definitive treatment of femoral shaft fractures with traction, the
angle of limb, applied weight, and direction of the applied traction
will all require adjustment based on the fracture location.

Definitive treatment of femoral shaft fractures has been
successfully accomplished with Neufeld roller traction and
modifications thereof.³⁶,⁹¹,¹³⁹,¹⁷²
The main advantages with Neufeld traction are early knee motion,
enhanced patient mobility, and earlier discharge from the hospital.
Fracture position can be well maintained despite the enhanced knee and
patient mobility. In a study of 30 patients treated with Neufeld roller
traction either temporarily or definitively, Browner et al.³⁶ observed no nonunions and no infections. Similarly, Gates et al.⁹¹
reported on their experience with the use of hinged casts and roller
traction in 121 patients in developing countries. In their series,
hospital stay was short and all but one fracture healed.

Similar to traction techniques, cast braces are now used
uncommonly for the treatment of femoral shaft fractures. However, cast
braces can be used successfully, and the techniques of application for
the appropriate indications are important to understand. In unusual
circumstances, they can be used to augment intramedullary fixation if
unlocked implants are used.²⁵¹ For
distal and femoral shaft fractures, cast braces allow early patient
mobilization and may be used in combination with a period of traction.⁶⁴,¹¹⁶,¹⁸⁶ They are thought to work by converting the thigh into a semi-rigid hydraulic tube that maintains the alignment of the femur.¹⁸⁶
Typically, traction is initially used until there is some stability to
the fracture, usually for 6 to 8 weeks. Clinical resolution of fracture
site discomfort, radiographic evidence of callus formation, and lack of
fracture instability on clinical examination are all good indicators of
early healing. Like any form of closed management, success depends on
the technique of application and the proper implementation of weight
bearing. Cast braces cannot be used to correct angulatory or rotational
deformities.

The indications, techniques, and success of cast brace application have been described in detail by Mooney et al.¹⁸⁶
They described the application of a proximal thigh cast and a long leg
cast incorporating the foot combined with a hinge at the knee joint.
The authors reviewed the treatment of 150 patients with femoral shaft
fractures treated with cast braces. The average time in traction was
7.3 weeks, and the average time of casting was 7.2 weeks. There were no
nonunions or refractures. This compares to a group of patients treated
by the same authors with traction followed by spica casting. In these
patients the total treatment time was more than 24 weeks, and more than
10% of patients had either a nonunion or a refracture.¹⁸⁶ In another study reviewing an experience with cast braces for femoral fractures, Hardy¹¹⁶
reported acceptable results in106 patients. Union was predictable and
comminuted mid-diaphyseal fractures were observed to shorten more than
distal fractures.¹¹⁶ Similarly, Connolly et al.⁶⁴
followed 143 patients with femoral shaft fractures treated with
traction followed by cast bracing. Although shortening of up to 2 cm
was observed frequently, nonunion and clinically significant malunion
were uncommon. The conclusions of these studies demonstrate that cast
braces can be successfully used for the treatment of femoral shaft
fractures after a period of traction.

The current preference for closed interlocked nailing
over traction followed by cast bracing or spica casting of femoral
shaft fractures is generally considered obvious. This was studied, and
the improved outcomes with nailing were documented. Locked nailing was
associated with a much shorter hospital stay, decreased time to union,
and improved alignment when compared with traction plus casting.²⁹¹ Similarly, Johnson et al.¹³⁹
reported an average hospital stay of 31 days in patients treated with
roller traction compared with 20 days for patients treated with
intramedullary nails. The frequency of failure, defined as an unplanned
reoperation or unacceptable alignment, was reported in 66% of fractures
treated with roller traction compared with 4% of those treated with a
locked intramedullary nail.

Although
these results come as no surprise, they do emphasize the significant
improvements in treatment that have been observed with intramedullary
nailing.

External fixation as a definitive treatment method for
femoral shaft fractures has limited indications. However, temporary
external fixation is used with increasing frequency in some valuable
circumstances. External fixation is most easily accomplished with a
unilateral frame (Fig. 50-8) but circular
frames are designed specifically for the femur. Unfortunately, these
devices are poorly tolerated by the patient over the extensive time
necessary to ensure adequate healing of the femur.

External fixation has several advantages, especially as
a temporizing measure for the initial stabilization of a fractured
femur. The procedure is rapid, and a temporary external fixator can be
reliably applied in less than 30 minutes. This is particularly
important in the critically ill patient.⁴,³⁴,¹⁹⁹,²¹²,²⁴⁵
The vascular supply to the femur is not damaged to a significant degree
during the application of an external fixator, and this may be
important in high-energy and open injuries with significant damage to
the extraosseous and intraosseous blood supply. No additional foreign
material is introduced in the region of the fracture, which may be
particularly advantageous in open fractures and injuries with
infection. Further, external fixation allows access to the medullary
canal and the surrounding tissues in open fractures with significant
contamination. Dressing changes and local soft tissue care are
maintained with external fixation, although other methods may allow
improved access.

Most of the disadvantages of femoral external fixation
are related to this technique as a definitive treatment measure. Pin
tract infections occur commonly and are related to the time the fixator
is in place, the amount of soft tissues that the pins must traverse,
and the sterility at the time of the initial application.¹¹⁷
Loss of knee motion occurs commonly. Angular malunion and femoral
shortening occur more frequently than with other methods. There are
still concerns about the potential increased infection risk associated
with conversion of an external fixator to another definitive treatment
methods. Finally, unilateral external fixation has limited ability to
adequately stabilize the femoral shaft. This is largely because of the
large weight of the leg combined with the distance between the femoral
shaft and the bar of the external fixator.

The indications for external fixation (Table 50-2) are continuing to evolve.^*
Initially, extensive comminution and open fractures were considered to
be relative indications for the use of femoral external fixation as a
definitive treatment for femoral shaft fractures. However, as other
treatment methods such as intramedullary nailing have continued to
improve and low complication rates have been demonstrated with nails
even in the most complex injuries, the indications for the use of
external fixation have become more limited. Currently, the primary
indications include use as a temporary bridge to intramedullary nailing
(Fig. 50-9), use in the severely injured
patient who cannot tolerate reaming and/or placement of a medullary
implant as a form of damage control orthopaedics, use in a patient with
an ipsilateral arterial injury that require repair, and in patients
with severe soft tissue contamination in whom a second debridement
would be limited by other devices. Which patients are more suitable for
initial placement of an external fixator followed by secondary
conversion to a medullary implant continues to be better defined but
includes patients with a severe head injury, elevated Injury Severity
Score (ISS), associated thoracic trauma, or multiple extremity
injuries. In patients with an accompanying ipsilateral arterial injury,
an external fixator can be rapidly applied, produces temporary
stability of the limb, and can easily be converted to another form of
stabilization.¹²⁷

Application of a temporary unilateral frame is
relatively straightforward and based on the biomechanical principles of
external fixation. For patients requiring rapid application of a
temporary spanning fixator, the goals should be limb realignment,
patient comfort, and prevention of further soft tissue damage. Pins can
be placed anteriorly, anterolaterally, or laterally. Anterior pins
require placement through the extensor mechanism and should be avoided
if possible. Laterally placed pins require placement through the
iliotibial band and may similarly limit knee motion. To minimize this,
the iliotibial band should be incised in line with the leg at the time
of fixator placement. Pins can be placed from an anterolateral
location, anterior to the iliotibial tract. This may minimize knee
stiffness but is technically more difficult because the pins are out of
plane with the typically obtained fluoroscopic images that are used for
their placement.

In adults, the pins should be at least 5 mm in diameter.
A minimum of two pins per segment should be placed. A pin should be
placed near the fracture but out of the predicted fracture hematoma.
Another pin should be placed as far away from the fracture in each
segment. If the external fixator is planned as a temporizing measure, a
perfect reduction of the femur may be unnecessary. However, depending
on the original indications for placement of the external fixator, the
device may ultimately remain in position for longer than expected. As a
result, every attempt should be made to realign the limb and
reestablish the femoral length to facilitate the secondary
intramedullary nailing procedure.

Initially, external fixation was used definitively for
femoral shaft fractures with extensive comminution, associated open
wounds, or multiple life-threatening injuries. Several small series
reported fracture union, early mobilization, rare infections, and
limited impact on knee motion.⁴,⁶⁸,¹⁰⁵,²³⁵ Alonso et al.⁴
treated 24 patients with femoral shaft fractures with external
fixation. This was a temporary treatment in 14 patients, although in 10
patients this was definitive. Healing was ultimately observed in 88%,
but almost half of the patients had significant loss of knee motion.
Shortening was noted in two patients treated definitively with external
fixation. On the basis of their experience, the authors recommended the
use of external fixation in several circumstances including the
aggressive management of soft tissue injuries, for the severely head
injured or traumatized patient, and for infected femoral nonunions. The
limitations of external fixation as a definitive treatment method in
open femoral shaft fractures were further elucidated by Mohr et al.,¹⁸⁵
who reported a lengthy treatment period of more than 5 months, early
deep infections in 11%, late deep infections in 11%, shortening in 7%,
and restricted knee motion in 20%. Although all fractures healed, the
high incidence of complications and secondary surgeries in this group
of patients cannot be overlooked.

The use of temporary spanning external fixation in a large series of patients was reported by Nowotarski and colleagues.¹⁹⁹
Of more than 1500 femoral shaft fractures treated with intramedullary
nailing during the same time period, 54 patients were treated with
external fixation followed by planned conversion to an intramedullary
nail. The primary indications for this protocol included critically ill
patients (n = 46) or patients who required expedient femoral fixation
for repair of an ipsilateral vascular injury (n = 8). The average ISS
of these patients was 29, and open injuries were present in 19. The
average period of external fixation was 7 days with a range of 1 to 49
days. The majority of patients underwent conversion from an external
fixation to an intramedullary nail at a single operative procedure (n =
55), but this was done in a staged fashion in four patients who had
significant pin site drainage. In these four patients, traction with
intravenous antibiotics lasting an average of 10 days was instituted to
allow resolution in the intervening time period. Overall,

healing
was observed in 97% of patients within 6 months. One patient developed
an infected nonunion, and one patient had an aseptic nonunion. The
average knee motion was 107 degrees. The authors concluded that
immediate external fixation followed by early conversion to
intramedullary nailing for femoral shaft fracture was safe in selected
multiply injured patients.¹⁹⁹
The safety of temporary external fixation with delayed conversion to an
intramedullary implant was further corroborated in a retrospective
study of 173 patients with 192 fractures. The infection rates with
initial external fixation followed by conversion to an intramedullary
device were similar to that in patients treated with primary
intramedullary nailing in multiply injured patients. Pin site
contamination was observed more commonly when the fixator was left in
place for more than two weeks, suggesting a timely conversion to a
medullary implant whenver possible.¹¹⁷

More recently, the use of temporary femoral external
fixation has been investigated to further define the indications in
multiply injured patients. This has led to the concept of “damage
control orthopaedics” in selected patients who may be poor candidates
for early definitive intramedullary nailing of the femur. Scalea et al.²⁴⁵
reported their results with early external fixation followed by
definitive intramedullary nailing in 43 patients with contraindications
to primary nailing. This included patients with associated head injury,
hemodynamic instability, thoracoabdominal trauma, and multiple other
injuries. Similarly, Pape et al.²¹⁴
reported their conversion from early total care of patients to “damage
control orthopaedics” in the polytraumatized patient. This was
characterized by more frequent use of spanning and temporary external
fixation to treat lower extremity and femoral fractures followed by
delayed conversion to intramedullary nailing. The incidence of multiple
organ system failure was noted to decrease with this change in
treatment philosophy and was believed to be an adequate alternative in
patients at high risk for development of acute respiratory distress
syndrome (ARDS) or multiple organ system failure.²¹³,²¹⁴
In a prospective, randomized, multicenter study, a sustained
inflammatory response was seen after primary (<24 hours)
intramedullary femoral instrumentation, but not after initial external
fixation or after secondary conversion to an intramedullary implant.
These findings may become clinically relevant in patients at high risk
of developing complications. In addition, it confirms previous studies
that suggest damage control orthopedic surgery minimizes the additional
surgical impact induced by acute stabilization of the femur.²¹³
With time and further study, the indications for early external
fixation followed by delayed conversion to intramedullary nailing will
become better defined.

Operative stabilization with any technique offers
numerous advantages compared with nonoperative methods and includes
early patient mobilization and early functional rehabilitation of the
injured extremity. The use of plate fixation for the routine treatment
of femoral shaft fractures has decreased with the increased use of
intramedullary nails. Several distinct advantages to plating do exist,
including the ability to obtain an anatomical reduction in appropriate
fracture patterns and the lack of additional trauma to remote locations
such as the femoral neck, acetabulum, and distal femur. The main
disadvantages associated with plate fixation when compared with
intramedullary nailing are the need for an extensive surgical approach
with its associated blood loss, infectious complications, and soft
tissue insult. This can result in associated quadriceps scarring and
its effects on knee motion and quadriceps strength. Further, the impact
of the plate on the femur after fixation cannot be minimized. This
includes the decreased vascularization beneath the plate and the stress
shielding of the bone spanned by the plate. Minimally invasive
techniques of plate application, although not addressing all of the
issues associated with plating, do minimize the additional vascular
insult to the periosteal and medullary blood supply of the femur.⁸³ However, minimally invasive techniques have been associated with an increased incidence of malreduction.³¹⁵
Finally, because the plate is a load-bearing implant, implant failure
is expected if union does not occur. Although implant failure will
ultimately occur with any orthopaedic device in the case of nonunion, a
load sharing implant of sufficient size will likely have increased
longevity compared with a plate.

Almost any fracture of the femur can be successfully
plated. However, with the predictable results of intramedullary nails
for both the initial treatment of femoral fractures and the salvage of
other failed fixations, the use of plates as a primary treatment method
is limited (Table 50-3). That being stated,
certain clinical situations and patient factors may make plates
desirable. In most instances, the relative indications for plating are
the same as the relative contraindications for intramedullary nailing.
In patients with an extremely narrow medullary canal in whom
intramedullary nailing is impossible or difficult, plates remain an
excellent option. Plates are useful in fractures that occur adjacent to
or through a previous malunion, or in cases in which there is
obliteration of the canal because of infection or previous closed
management. Plates may be advantageous in fractures that have
associated proximal or distal extension into the pertrochanteric or
condylar regions because fracture reduction and stabilization may be
difficult with a medullary implant.

In patients with an associated vascular injury, the
exposure for the vascular repair frequently involves a wide exposure of
the medial femur. If rapid femoral stabilization is desired, a plate
can be applied quickly through the medial open exposure.

However,
in this circumstance the rate of nonunion is likely increased by the
combination of the injury and surgical exposure. One should consider
that any salvage procedures in this scenario will require a repeat
surgical exposure through a scarred medial approach around a vascular
repair.

The patient can be positioned supine or lateral on a
completely radiolucent table to allow unimpeded fluoroscopic imaging of
the femur from the hip to the knee. Lateral positioning eases the
retraction of the vastus lateralis in open techniques but may not be
applicable in the polytraumatized patient. In addition, proximal
imaging can be more difficult with the patient positioned laterally.
For supine positioning, the entire limb should be prepped in the
surgical field, including the groin to allow access to femoral vessels
if necessary. A small bump beneath the ipsilateral hip helps to
internally rotate the limb to neutral, simplifying the surgical
approach and making the intraoperative assessment of rotation easier. A
radiolucent ramp or several folded blankets placed beneath the thigh
and leg improves the lateral fluoroscopic imaging by elevating the leg
relative to the contralateral extremity.

Depending on the fracture configuration, an open or a
submuscular technique may be applicable. For simple fracture patterns
that allow an accurate cortical reduction of the majority of the
femoral shaft, an open technique with compression plating using AO
principles is advisable. However, in fracture patterns that have near
circumferential or segmental comminution, bridge plating techniques are
applicable. This can be accomplished with submuscular plate placement
or an open technique that leaves all the intercalary comminuted
segments undisturbed. No matter what technique is chosen, the
vascularity to the femoral shaft and the associated fracture segments
should be preserved.

For open compression plating of simple fracture
patterns, a lateral incision is used. The length of the incision should
be of adequate size to allow placement of a long plate directly on the
lateral femur without traumatic retraction of the muscular envelope.
The iliotibial band is sharply incised along the length of the
incision. The vastus lateralis is then atraumatically elevated from the
posterior limb of its fascia, from distal to proximal. The perforating
vessels should be sequentially identified, isolated, and ligated. The
muscle can then be elevated off the periosteum at the lateral femur. No
periosteal stripping is necessary, and no deep retractors placed over
the anterior femur should be used.

Reduction of the femur can be difficult, and several
useful adjuncts are available. The patient should be pharmacologically
paralyzed to allow for reestablishment of femoral length. Several
useful intraoperative measures are available to assist with the
indirect reduction of the femoral shaft. A femoral distractor can be
applied to facilitate restoration of length, alignment, and rotation.
Alternatively, a distal femoral traction pin can be applied, allowing
for manual distraction and rotational control. The plate can be applied
and fixed to one end of the femur followed by the use of a push screw
or the articulated tensionerdistractor. Finally, joysticks consisting
of 5-mm Schanz pins can be placed into each femoral segment to assist
with control of the alignment and rotation.

The size and length of the implant remain controversial.
Even in simple patterns, a 10-hole, 4.5-mm plate should be considered a
minimum length. The choice of a broad or a narrow plate depends on the
femoral diameter and the size of the patient. As the fracture
comminution increases, so should the plate length such that at least
five screw holes of plate length are present on each side of the
fracture (Fig. 50-10). For transverse
fractures, appropriate over-contouring of the plate should be
performed, and the fracture should be appropriately compressed. This
can be accomplished with a pull screw, the articulated tensioning
device, eccentric screw placement(s) within the holes, or combinations
thereof. For oblique fractures, overcontouring of the plate is
inadvisable and compression should be obtained with standard techniques
using the obliquity of the fracture relative to plate. A lag screw
through the plate increases the construct rigidity and should be used
if appropriate. The number of screws necessary for adequate
stabilization remains unknown, although eight cortices on each side of
the fracture have been recommended in the past.¹⁶⁶ The number of screws and the number of cortices are less important than the plate length.²⁶¹,²⁶² There are very few disadvantages to the use of a longer plate with fewer screws other than the need for a longer incision (Fig. 50-11).
However, screws can be placed percutaneously if a longer implant with a
smaller surgical approach is desired. Three screws in each segment
should maximize the

biomechanical
fixation. This should include a screw at the end of the plate
(preferably directed obliquely) and a screw as close as possible to the
fracture. A third screw in each segment further increases the torsional
stability of the bone implant construct.²⁶²

The technique of submuscular plating for the femur is similar to that in other long bones.²⁸⁶,²⁹³,³¹⁵
The goal is to minimize any associated trauma to the muscular envelope
and the periosteum surrounding the femur. This can be accomplished by
using small incisions at the palpable proximal and distal aspects of
the lateral femur. A plate of appropriate length can be contoured and
slid beneath the vastus lateralis along the length of the femur. A
plate that spans most of the femur exploits the palpable portions of
the femur proximally and distally and is actually easier to apply than
a shorter plate. The lack of an appropriate anterior bow in long plates
can make plate application difficult in simple patterns. However,
incomminuted fractures in which bridging fixation is planned and
submuscular techniques are most applicable, the use of a straight plate
is usually not problematic. After the plate is placed on the midlateral
aspect of the femur, the length, alignment, and rotation of the femur
should

be
restored either manually or with a femoral distractor. Screws can then
be fixed proximally and distally without difficulty. Additional screws
are placed through percutaneous stab incisions along the lateral femur
according to the aforementioned biomechanical principles.

External supports such as casts or cast braces are
unnecessary after a properly performed plate application for a femoral
fracture. Early active range-of-motion exercises should be encouraged
as soon as wound healing allows. Weight bearing should be limited to
the weight of the leg until there is radiographic evidence of healing
and resolution of fracture site discomfort. Radiographic evaluations at
6-week intervals are usually adequate for following the progression of
healing. If direct reduction and compression plating are used, primary
bone healing is predicted. As a result, radiographic evidence of
healing and, thus, weight bearing are frequently delayed for 4 to 5
months.⁹³,²²⁹
If bridge plating and indirect reduction techniques are used, healing
should progress similar to intramedullary nailing with callus formation
evident on radiographs. As a result, weight bearing can usually begin
earlier. Plate removal should be limited to symptomatic patients and
should be delayed for a minimum of 18 to 24 months because of the risk
of refracture.

The early experience with compression plating of femoral
shaft fractures using AO techniques were relatively consistent. Implant
loosening occurred in 6% to 11% of cases, nonunion in 2% to 8%, and
infection in 0% to 7%. Bone grafting was recommended for defects or in
cases with questionable fixation.⁵⁶,¹⁶⁷,²³⁸,²⁷¹ For the most part, direct reduction and stable fixation of the fracture were recommended. Ruedi and Luscher²³⁸
reported on the results of plating 126 comminuted fractures using
Association for the Study of Internal Fixation techniques. Plate
failures and nonunion were observed in less than 10% of cases, yet the
authors still recommended bone grafting in all fractures of the femoral
shaft fixed with a plate. Surprisingly, as early as 1979, bridge
plating of comminuted fractures was recommended in an effort to
maximize the vascularity and healing of these fragments.¹⁶⁹

Riemer et al.²²⁹
reported on 141 comminuted diaphyseal femur fractures treated with
immediate plate fixation. More than one third of these fractures were
open. Autogenous bone grafting, typically from the ipsilateral proximal
tibial metaphysis, was performed in 95% of patients. Ten plate failures
were observed (7%), and most healed after subsequent treatment with an
intramedullary nail. There was one infection in a patient with an open
fracture. The average time to union was 17 weeks. Similarly, Geissler
et al.⁹³ reported a 93% union rate
at 16 weeks in 71 plated femoral shaft fractures. Cancellous bone
grafting was used in 69% of patients and was recommended as a standard
supplement in this technique.

Currently, the routine use of cancellous bone grafting
in plated femoral shaft fractures may be unnecessary if indirect
reduction techniques and a biologically friendly surgical approach are
used.²³⁷,²⁸⁶,²⁹³,³¹⁵
Although simple fracture patterns continue to be amenable to direct
reduction and stable fixation, comminuted fractures can be treated by
completely avoiding the fracture fragments. By bridging the fracture
using incisions well away from the fracture, a plate can be slid
beneath the vastus lateralis and applied to the lateral femur with a
combination of percutaneously applied and openly applied screws. The
main goals of this technique are restoration of femoral length,
alignment, and rotation with preservation of blood supply to the
fracture and comminuted fragments. Early reports suggest that this
technique can be safely performed with results similar to other
indirect methods for femoral stabilization.²³⁷,²⁸⁶,²⁹³,³¹⁵

The use of traditional open reduction and submuscular
plating was retrospectively compared by Zlowodski et al. No bone grafts
were used in either group and a low infection rate was observed and was
unrelated to the surgical approach. However, the submuscular technique
was associated with a higher rate of suboptimal reductions (29%)
compared to a biologically respectful open approach.³¹⁵

Antegrade femoral nailing can be one of the most predictable procedures in orthopaedic traumatology (Table 50-4).
It is the most common treatment for femoral shaft fractures in adults,
and virtually all orthopaedic surgeons have had some exposure and
experience with the techniques necessary to successfully perform a
femoral nailing. However, there are numerous aspects to the procedure
that require emphasis to ensure a good result with minimal
complications. Many of the technical aspects of planning, positioning,
reduction, nailing, and interlocking are appreciated with increasing
experience.

Preoperative planning begins with an understanding of
the fracture pattern, which is dependent on an understanding of the
mechanism of injury and the applied force. Good-quality biplaner
radiographic images are necessary and should be reviewed to determine
the canal dimensions, femoral length, presence of comminution, femoral
morphology, and presence of nondisplaced fracture extensions that may
complicate treatment. Contralateral femoral radiographs can be obtained
if there is concern regarding the femoral morphology or in cases with
significant comminution, which makes estimates of femoral length
unreliable. The lateral radiograph should be scrutinized to determine
the femoral bow. Because virtually all femoral nails have a radius of
curvature that is greater (i.e., less curved) than the normal femoral,
increasing femoral bow can make nailing difficult using unmodified
implants. The radiographs should be viewed to determine whether there
is any associated preinjury osseus pathology such as metastatic
disease, primary bone tumor, osteomyelitis, or malunion.

As a part of the preoperative plan, the proper length
and diameter of a femoral nail should be anticipated before considering
operating on a patient. The femoral length can be determined by several
methods. Radiographs of the contralateral femur can be measured with a
ruler that is corrected for magnification. In fractures without
significant comminution, traction radiographs of the injured femur can
be used to estimate the length. Alternatively, a long ruler can be used
to measure the uninjured femur from the palpable greater trochanter to
the lateral epicondyle. The diameter of the intramedullary canal should
be estimated at the narrowest portion of the femoral canal at the
femoral isthmus. This is usually based on the lateral radiograph.
Although these measures are estimates only and a more accurate
assessment can be obtained intraoperatively, this allows the surgeon to
ensure that an adequate supply of femoral nails with the predicted
lengths and diameters are available.

Patients with small femoral canals (<10-mm) may require special implants specifically designed for these situations.

Some type of femoral traction should be maintained
during the time period extending from identification of the injury to
surgical stabilization. Because delay is unpredictable and frequently
unanticipated, skeletal traction should be placed as early as possible.
Traction maintains femoral length, prevents further soft tissue injury,
provides patient comfort, and may limit the amount of ongoing blood
loss into the thigh. A distal femoral traction pin is preferred to a
proximal tibial pin because of the more direct line of pull and the
high incidence of associated knee ligamentous injuries. A large
traction pin is painful, traumatic to place, and unnecessary. A small
pin (5/64-inch or 2-mm) can be placed from medial to lateral at the
distal femur and tensioned with a traction bow. The ideal location is
anterior in the distal metaphyseal region of the femur, allowing future
passage of the antegrade nail. Approximately 15 to 20 pounds of weight
(or up to 15% of body weight in larger patients) is usually adequate to
maintain femoral length in most patients if traction is applied
acutely. A pulley system at the end of the bed or, ideally, balanced
suspension should be used.

In closed fractures, preoperative prophylactic
antibiotics consisting of a first generation cephalosporin in patients
without an allergy should be used. Additional antibiotic coverage is
necessary in open fractures and is dependent of the level of
contamination, and the estimated associated soft tissue injury.

Patient positioning for treatment of a femoral shaft
fracture depends on numerous factors including the method of
stabilization, associated injuries, and surgeon comfort (Table 50-5).
In a patient with a known spine injury or in a patient who cannot be
adequately evaluated before femoral stabilization, protection of the
spine from further injury is vital. Lateral bending of the spine can be
minimized with supine positioning on a radiolucent table and femoral
stabilization with a retrograde nail, an external fixator, or a plate.
Supine positioning on a radiolucent table with antegrade nailing
requires some degree of lateral bending to allow access to the starting
point. This can be minimized with placement of a trochanteric entry
nail, which allows a more lateral starting point than with piriformis
entry nails.

Antegrade femoral nailing can be successfully performed
with the patient lying supine or lateral, each position having specific
advantages and disadvantages. In addition, nailing can be done on a
fracture table or a radiolucent table with or without traction. With
either position on either type of table, the accurate placement of a
distal femoral traction pin assists with the intraoperative assessment
of femoral rotation and attainment of

the
proper length. Even if a pin has been placed in the emergency
department for temporary application of traction, the pin position
should be confirmed fluoroscopically in the operating room and changed
if not in the proper location. An ideal pin should be perfectly
parallel to a condylar overlap lateral of the distal femoral condyles,
ensuring that the rotation of the distal femur is known. This is best
accomplished by placing the pin under fluoroscopic guidance on the
lateral view with the femur properly rotated. In addition, the pin
should be positioned anteriorly to allow passage of the nail posterior
to the pin. This eliminates the time-consuming and frustrating exercise
of trying to remove the traction pin as the femoral nail is passed
later in the procedure. Because of the natural tendency for a valgus
deformity when performing antegrade nailing in the lateral position,
consideration can be given to placing the pin somewhat angulated in the
coronal plane (proximal lateral to distal medial).

Supine positioning allows unencumbered access to the
entire patient by the anesthesiologists, general surgeons,
neurosurgeons, and other physicians involved with ongoing resuscitation
and treatment. This is especially important in patients with multiple
thoracic or abdominal injuries, spine fractures, pelvic ring injuries,
or multiple extremity fractures. Patients with multiple extremity
injuries can undergo simultaneous treatment if positioned supine.
Rotational and angular deformities may be minimized with supine
positioning compared with lateral positioning. The rotation of the leg
is more easily judged with the entire limb accessible. The main
disadvantage with supine positioning is difficulty with identifying the
proper starting point if a first generation antegrade nail is
anticipated. The gluteal muscles and the associated fat on the lateral
thigh make palpation of the greater trochanter difficult. Hip adduction
improves access to the proximal femur. With lateral positioning,
patient access is limited and pulmonary function is suboptimal.
However, identification of the starting point is improved because the
hip is flexed, angulating the torso away from the line of the femoral
shaft. As well, because of the natural adduction of the hip, the soft
tissues of the lateral hip tend to fall away from the anticipated
incision.

Supine positioning on the fracture table is time consuming but allows consistent and untiring intraoperative traction (Fig. 50-12).
Although not necessarily supported in the literature, use of the
fracture table may minimize the number of surgical assistants
necessary. A distal femoral traction pin is preferable to traction
applied to the foot with the fracture table. This allows direct control
of the distal femoral segment, ensuring an accurate assessment of
rotation without distracting across the knee joint. In addition, distal
femoral traction allows the knee to be flexed, minimizing the
intraoperative stretch and tension on the sciatic nerve. Alternatively,
a traction pin can be placed at the proximal tibial metaphysis. This
pin location prevents interference with the medullary implant, however,
traction is applied across the potentially injured knee joint and is
not in direct line with femoral axis. A combination of limb adduction
and tilting of the trunk away from the ipsilateral hip improves access
to the nail entry site. However, this requires that the patient’s
lumbar and thoracic spines are stable to positioning alterations.
Fluoroscopic imaging in the AP and lateral planes is usually without
difficulty. The uninjured leg can be positioned with the hip and knee
flexed to 90 degrees. However, numerous reports of unrecognized
compartmental syndrome in the uninjured leg during lengthy femoral
nailing procedures with the use of a well leg holder that supports the
calf emphasize the need to either check the leg intraoperatively every
hour or abandon this position completely.⁵,⁷⁶,¹⁷⁹ The relationship between compartmental pressures and time in a well-leg holder was evaluated by Tan et al.²⁶⁶
in a study of 10 patients undergoing femoral nailing procedures. The
authors demonstrated an immediate increase in pressure with positioning
in the well-leg holder that trended upward with time. The pressure
immediately returned to normal when the leg was taken out of the
holder. They further identified a correlation between leg pressure and
body mass index.²⁶⁶ The combination
of decreased perfusion caused by leg elevation and external pressure
from a calf support has been confirmed to cause a decrease in the
difference between diastolic blood pressure and the intramuscular
pressure when the uninjured leg is placed into a well-leg holder on a
fracture table.¹⁷⁹ On the basis of their study in healthy volunteers, Meyer et al.¹⁷⁹
demonstrated that positioning that leaves the calf free increases the
difference between the diastolic and intramuscular pressure,
potentially decreasing the risk of compartmental syndrome. This can be
accomplished by supporting the leg at the knee and the heel, leaving
the calf free to hang.¹⁷⁹
Alternatively, the uninjured leg can placed into a foot holder on the
traction table. The uninjured leg can be “scissored” with the hip and
knee extended. In this position, lateral fluoroscopic imaging of the
proximal femur is compromised but the safety of the uninjured leg is
increased.

Intramedullary nailing may be easier for the surgeon when the patient is positioned laterally (Fig. 50-13).
The initial positioning is more time consuming and there is limited
access to the patient. This is especially important in the patient with
associated thoracic trauma or pulmonary injury. The natural tendency is
to allow the femur to internally rotate excessively during lateral
nailing, and the initial patient positioning should combat this.
Internal rotation of the distal segment should be limited to 10 or 15
degrees. The uninjured leg is positioned with the knee and hip
extended, allowing unimpeded fluoroscopic imaging in both planes. The
knee height should be appropriately adjusted to resist the commonly
observed valgus deformity.

An alternative to positioning on the fracture table is supine

positioning on a radiolucent table with manual traction or intraoperative distraction¹⁴³,¹⁷⁴,²⁵⁴,²⁶⁰ (Fig. 50-14).
This technique has been used successfully and is the standard in most
trauma centers. The main disadvantage to the manual technique is the
need for a trained assistant. However, the strategic use of an
intraoperative femoral distractor or modification of the operating room
table to allow weights to be placed at the end of the bed minimizes the
need for assistance. For supine nailing on a radiolucent table, a bump
is placed medially beneath the hip and the patient is moved as
laterally as possible to maximize access to the lateral hip and flank.
The bump is placed beneath the pelvis and not the greater trochanter,
therefore maximizing access to the hip. The entire leg and ipsilateral
hip are prepped proximally past the level of the iliac crest. A distal
femoral traction pin can be placed as previously described to allow for
application of manual traction and to control rotation. Alternatively,
a femoral distractor can be applied, eliminating the need for a distal
femoral traction pin as well as an assistant.¹⁷⁴
Distally, a half pin can be placed anteriorly in the metaphysis from
lateral to medial. Proximally, a half pin can be placed from anterior
to posterior at the level of the lesser trochanter, medial to the
predicted path of the nail. This pin position is at least 2.5 cm from
the femoral nerve and 3.0 cm from the femoral artery based on cadaveric
dissections.¹⁷⁴ Alternatively, the
proximal pin can be placed from lateral to medial at the lesser
trochanter, posterior to the predicted path of the nail. The half pin
locations and technique have been described in detail by Baumgaertel
and coauthors.¹⁵ The success of femoral nailing without a fracture table has been confirmed in several recent studies. Karpos et al.¹⁴³
prospectively followed 32 consecutive femoral fractures that were
reduced and stabilized with manual traction. They reported good overall
alignment and decreased operative time compared with comparable femoral
nailing using a fracture table. In addition, more than one third of
patients were able to undergo multiple extremity procedures with the
same operative draping.¹⁴³ Similarly, Sirkin et al.²⁵⁴
reported similar results with femoral nailing on a radiolucent table
using manual traction only. Compared with the use of a fracture table,
they found the procedure to be more rapid, required fewer table
transfers, allowed accurate reduction, and produced no increased
morbidity.²⁵⁴ In a prospective study
comparing the use of a fracture table with manual traction, Stephen and
colleagues reported an increased incidence of malrotation and increased
operative time with the use of a fracture table.²⁶⁰

Obtaining and maintaining a reduction of the femoral
shaft can be difficult and is required during several key stages of
femoral nailing. The fracture must be reduced during guidewire passage,
during reaming, and when the nail is passed. There are a number of
techniques that can facilitate reduction, many or all of which may be
required. Many of these reduction maneuvers are dependent on the
patient’s position and the type of operative table.

Femoral length can be obtained as previously described
by using a femoral distractor, manual traction, or a fracture table.
The use of a femoral distractor has been described in detail as a
useful adjuvant for obtaining length in both acute fractures as well as
delayed nailings¹⁵,¹⁷⁴ (Fig. 50-15).
Although only one half pin is placed into each of the proximal and
distal segments of the femur, surprisingly good control of alignment
can be obtained. In fractures without segmental comminution, an
accurate assessment of the length after application of a femoral
distractor is easily obtained from fluoroscopic clues. If segmental
comminution exists, the distractor is useful for indirectly
reassembling the intercalary fragments by tensioning the soft tissue
attachments, frequently providing indications regarding length. If the
length cannot be determined fluoroscopically, comparison with the
uninjured femoral length or postoperative CT scanning may be necessary.

A variety of devices can be applied externally to assist
with reduction of the femur. Crutches, cooled mallets, large
retractors, and large reduction levers can all be used depending on the
patient position and the primary deformities. If supine nailing

on
a radiolucent table is used, soft bumps of varying heights placed under
the leg near the fracture can be used to correct the angular and
translational deformities seen on the lateral fluoroscopic projection.
In addition, a cannulated nail can be placed into the proximal segment
after it has been prepared with an awl or reamers, thus allowing
control of the proximal segment and passage of the guidewire. A short
small-diameter nail or a specifically designed cannulated device is
available from many nail manufacturers specifically for this purpose.

The use of percutaneously placed half pins to facilitate reduction has been described as well.⁹⁵
This technique allows correction of both angular and rotational
deformities simultaneously. Through small lateral (or anterior) stab
incisions, half pins can be placed either unicortically in the
midlateral aspect of the femur or posteriorly in the thick portion of
the linea aspera. These pins can be used to manipulate the fracture in
the coronal sagittal planes by attaching T-handle chucks to each. If
the sagittal plane translational deformity is largely corrected by
strategically placed bumps, lateral unicortical pins can be used to
correct the coronal translation and angulation with simple
manipulations. The reduction can be maintained while the guidewire is
passed as well as during reaming (Fig. 50-16). To avoid damage to the reamers, the tips of the pins should not be within the intramedullary canal.

The intraoperative assessment of rotation (Fig. 50-17) is difficult and is described in detail in the “Complications” section under the heading of “Rotational Malalignment.”

Antegrade Intramedullary Nails: Entry Site Choice and
Soft Tissue Impact. Current nail designs allow for placement of
antegrade intramedullary implants at the piriformis fossa and in the
region of the greater trochanter. Nails designed to enter at the
trochanteric tip are specifically designed with an additional proximal
lateral bend to facilitate nail placement through this eccentric
starting portal relative to the femoral canal. The piriformis nail
entry location has the advantage of predictably aligning with the
femoral canal in both planes. A trochanteric entry may be easier to
identify given the improved access to the greater trochanter. This may
be especially important in obese patients.²²⁸
However, proper placement of the implant requires the surgeon to
accurately match the intraoperative fluoroscopic entry angle with the
lateral bend of the specific nail. A mismatch can be associated with
iatrogentic proximal femoral fracture in shaft fractures, and
malreduction in subtrochanteric fractures.⁹²,¹⁴¹,²⁰⁴

The use of the term “piriformis fossa” for nail entry is
actually a misnomer. The actual location of the ideal entry point for a
medullary implant that is aligned with femoral canal is more accurately
termed the “trochanteric fossa.”²⁰⁸
However, given the general acceptance of the term “piriformis fossa”
when referring to the depression on the inner surface of the greater
trochanter (the “trochanteric fossa”), “piriformis fossa” will be used
in this chapter.

The insertion of any antegrade intramedullary nail is associated with injury to the local soft tissues at the hip.^*
The functional consequences are largely unknown, but the local damage
may contribute to the frequently observed incidence of hip pain (10% to
40%) following antegrade intramedullary nailing. As a result,
alternative entry sites at the proximal femur have been investigated.⁹⁰,¹⁸⁴,²²⁰
Patient and limb positioning may have an impact on the local soft
tissue injury during percutaneous antegrade piriformis entry femoral
nailing. In a cadaver study based on retrograde placement of a reamer
proximally to simulate the correct entry site for antegrade nailing,
Ozsoy and coauthors²⁰⁶ measured the
distance to and identified the potential risk of injury to the superior
gluteal nerve. The authors concluded that limited hip flexion and
adduction during nail insertion increases the risk of injury to the
superior gluteal nerve and the gluteus medius muscle. Therefore,
lateral patient positioning or freelegged may allow increased hip
flexion and adduction, thereby limiting these risks.²⁰⁶

The impact of the nail entry location on the local soft tissues has been investigated in detail by Dora et al.⁷⁵
Using a cadaver model and three different nail entry portals
(piriformis fossa, in the usual location of a first-generation nail;
anterior to piriformis fossa, in the usual location of a reconstruction
nail; and at the tip of the greater trochanter, in the location of some
trochanteric nails), the authors identified injury to multiple
structures. Specifically, the trochanteric entry portal was associated
with frequent injury to the piriformis tendon (80%) and the obturator
internus (25%), the reconstruction nail entry portal was associated
with less frequent injury to the piriformis tendon (50%) and the
gluteus mimimus (25%), and the piriformis fossa entry portal was
associated with frequent injury to the obturator internus (86%), the
piriformis tendon (71%), the obturator externus (29%) and branches from
the medial femoral circumflex artery.⁷⁵
In another cadaver study supporting a lateral nail entry point, Moein
et al. found increased injury to branches of the medial femoral
circumflex with a piriformis nail entry.¹⁸⁴
The impact of injury to branches of the medial femoral circumflex
during antegrade intramedullary nailing is unknown although the remote
possibility of avascular necrosis of the femoral head exists. However,
since the introduction of antegrade intramedullary nailing, this
complication has been limited to a few recent isolated case reports,¹⁰⁶,²²² indicating that avascular necrosis is extraordinarily uncommon and the relationship to nail entry is unspecified.

Trochanteric entry nails have been introduced largely
because of the improved ease of starting point identification and nail
insertion (Fig. 50-18). The local soft tissue impact of these

trochanteric entry femoral implants has been investigated with conflicting conclusions.⁹⁰,¹⁷³,¹⁸⁴,²²⁰
Trochanteric entry nails with large screws or helical blades into the
femoral head require a large (usually 17 mm) entry site reamer. In a
study using a cadaver model to investigate the impact of a large entry
portal at the tip of the greater trochanter, McConnell et al.¹⁷³
reported disruption of an average of 27% of the gluteus medius tendon.
However, a modified medial trochanteric portal placed just medial to
the trochanteric tip (using a 14-mm reamer) was associated with no
visible damage to the tendinous insertion of the gluteus medius in a
different cadaver model.²²⁰ Further,
the presence of an alternative entry site lateral and distal to the tip
of the greater trochanter has been demonstrated. This greater
trochanteric bald spot is covered by the subgluteus bursa, is without
tendinous insertion, and may represent an alternative entry site for
specific nail designs.⁹⁰

Surgical Technique: Piriformis Entry Nails. Both open
and percutaneous approaches for insertion of an antegrade nail are
possible. The traditional open approach is accomplished through a
longitudinal incision that begins at the proximal tip of the greater
trochanter and extends proximally over a variable distance depending on
the size of the patient. Curving the incision posteriorly in line with
the natural bow of the femur maximizes access to the proper starting
point. The gluteus maximus fascia is incised in line with the incision,
and the muscle fibers can be separated bluntly. The greater trochanter
can be palpated directly, as can the enveloping fascia of the gluteus
medius. The piriformis fossa (or more accurately named, the
trochanteric fossa²⁰⁸) can then be
palpated directly, allowing placement of a curved awl or terminally
threaded pin. Alternatively, a relatively percutaneous approach can be
used (Fig. 50-19). This technique requires
placement of a pin into the piriformis fossa followed by the use of an
appropriately sized cannulated drill with a diameter of 8 to 10 mm.
Frequently this incision (10 to 20 mm in length) is located 10 to 20 cm
proximal and slightly posterior to the palpable greater trochanter.

If a starting awl is used in an open approach, its position should be confirmed fluoroscopically on the AP and lateral planes (Fig. 50-20).
If the awl appears to be on top of the lateral aspect of the femoral
neck on the AP fluoroscopic projection, the starting point is too
anterior relative to the piriformis fossa. The piriformis fossa is
somewhat posterior to the femoral neck

on
the lateral fluoroscopic image and inferior to the cephalad neck of the
femur on the AP image. The awl can then be advanced in line with the
intramedullary canal on both views. If a cannulated drill is chosen to
initiate the starting portal, the guide pin can be advanced with a
drill or a mallet down the canal of the femur using fluoroscopic
imaging to confirm accurate placement. The starting point is critical
to ensure safe passage of a medullary implant, and a suboptimal
position should not be accepted. A starting point that is too anterior
increases the proximal femoral hoop stresses and increases the
likelihood of femoral bursting. With the patient positioned supine on a
radiolucent table, access to the proper starting point is improved with
maximal adduction of the leg. In more proximal fractures in which
manual adduction of the leg fails to adduct the proximal femoral
segment, a percutaneously placed spike pusher or Schanz pin at the
lateral aspect of the proximal femur can assist with presenting the
starting point.

The ideal radiographic starting point for placement of an antegrade piriformis entry nail has been reviewed in some detail.⁹²,⁹⁶
The medullary cavity proximal exit point was defined in one study using
16 cadaver femora. The location was determined by obtaining radiographs
in 12 sequential planes separated by 15 degrees axially in femurs that
were filled with a radioopaque substance. In 88% of specimens, the
entry point for a straight nail was identified at the medial border of
the greater trochanter, at the tendinous insertion of the piriformis,
and 2.1 cm anterior to the posterior border of the greater trochanter.
For nails with a radius of curvature of 100 cm the nail entry point is
only 0.7 mm anterior to the dorsal edge of the trochanter.⁹²
Because most available antegrade nails have a femoral bow that is
greater than 100 cm, logic suggests that the ideal starting point is
located somewhere between these reported locations.

After the starting hole has been made, a bulb-tipped
guidewire can be passed down the canal of the femur and confirmed with
biplanar fluoroscopic imaging. In young patients with dense bone,
T-handled hand reamers may be necessary to define the intramedullary
canal in the pertrochanteric region. This step can be avoided if a
cannulated drill technique is used for initiation of the starting
point, assuming that the drill is passed distally enough in line with
the medullary canal. The guidewire should be bent distally to allow
some directional control. A guidewire without a ball-tip should not be
used if reaming is planned. The guidewire is then advanced down the
canal of the femur to the level of the fracture. With pharmacologic
relaxation, the fracture is then reduced to allow passage of the
guidewire across the fracture and into the distal segment. Numerous
reduction maneuvers as previously outlined are useful for this easily
described yet practically difficult portion of the procedure. The bend
at the end of the guidewire can be exploited to assist with passage of
the guidewire (Fig. 50-21). The guide wire is
then advanced distally and should terminate centrally in the distal
femur. Any residual angulation should be corrected before advancement
of the guidewire distally. This is especially true in fractures that
are distal to the femoral isthmus, especially if the fracture is in the
distal third. A mallet is used to seat the guidewire into the dense
bone of the distal femur to ensure that it remains in position during
reaming. For distal fractures, a true AP and lateral image of the
distal femur should be obtained to ensure central placement of the
guidewire. Femoral length should then be confirmed before determining
the desired nail length. A second guidewire of equal length can be used
to determine the proper length of the nail necessary. Knowledge of the
specific nail characteristics and the normal level of insertion at the
piriformis fossa is necessary to determine the proper length.

Flexible reamers can then be passed over the guidewire
and down the canal of the femur. The fracture should be maintained in a
reduced position during the passage of each reamer to minimize
eccentric bone removal. The process of reaming produces elevated
intramedullary canal pressures and embolism of fat into the venous
system. This is related both to the technique and the design of the
reamers. Reamers should be advanced slowly and incrementally increased
in diameter to minimize the pressure and fat embolism. Reamers should
be sharp, have deep flutes, and have a narrow flexible shaft.²⁴,¹⁹⁴,²¹⁰,²¹⁸
With the currently available locked intramedullary implants, the
placement of large-diameter nails with an intimate fit along a long
length of the medullary canal is no longer necessary. Depending on the
biomechanical characteristics of the nail, the patient’s size, the
presence of comminution, and the desire for immediate weight bearing, a
minimum nail diameter can be chose by the surgeon. A nail diameter that
is coincident with isthmic cortical chatter felt during reaming is
probably adequate in most cases. The canal is overreamed by 1.0 to 1.5
mm to allow passage of the nail. Failure to enlarge the medullary canal
can result in nail incarceration or iatrogenic femoral comminution,
both of which are avoidable complications.

The ball-tipped guidewire can then be exchanged for a straight guidewire using a flexible exchange tube. However,

many currently available nailing systems allow removal of the
ball-tipped guidewire through the nail, eliminating this step in the
procedure. The selected nail dimensions are then confirmed, and the
nail is attached to the proximal locking jig and placed over the
guidewire. The anterior bow of the nail can be used to ease the initial
insertion into the starting hole at the piriformis fossa. By internally
rotating the nail by 90 degrees, the tip of the nail can be placed
around the medial greater trochanter. The rotation of the nail is then
corrected to match the rotation of the femur. The nail is advanced to
the level of the fracture. At this point, the length, alignment, and
rotation of the femur should be corrected accurately. The nail is
advanced across the fracture, and the reduction parameters should be
confirmed. The nail is then driven distally to the appropriate depth (Fig. 50-22).

Proximal Interlocking. Proximal interlocking is
simplified with the use of an external jig that allows placement of
screws using multiple sleeves. The configuration and location of these
screws are unique to each nailing system and may be transverse or

oblique
in direction. Static and dynamic screws may be available proximally or
distally. The choice to use one or two proximal interlocking screws
depends on the fracture stability after nailing, fracture location,
patient’s size, nail and screw diameters, and predicted postoperative
activity.

Distal Interlocking. Distal interlocking is usually
accomplished using a free-hand technique that is highly dependent on
fluoroscopic imaging. External targeting jigs remain under
investigation but have limited applicability because of targeting
inaccuracies. This is because of the long length of the implant and the
deformation that occurs with nail passage. Computer-assisted navigation
systems continue to be explored and will likely increase in popularity
in the near future.

Regardless of how distal interlocking is accomplished,
the alignment, especially in rotation, must be maintained until at
least one screw is placed. With supine nailing on a radiolucent table,
the usual error in a rotationally unstable pattern is to externally
rotate the distal segment in an attempt to align the nail with the
image intensifier. To avoid this, the bump should be removed from
beneath the ipsilateral hip and the entire leg should be rotated as a
single unit. There are numerous choices for a free-hand technique.
Radiolucent drills, tapered pins, and radiolucent handles are all
available to assist with distal interlocking. Regardless of the
technique, the goal is to expeditiously place one or two interlocking
screws through the nail with the minimum amount of radiation.

The image intensifier is first used to localize a
perfectly round circle of the interlocking screw hole. A small
longitudinal skin incision is made followed by an incision through the
iliotibial band in line with the fibers. A sharp drill bit, preferably
with a trochar tip, is angulated obliquely to allow imaging, and the
tip is placed in the exact center of the projected hole as confirmed by
the lateral fluoroscopic image. The drill is then positioned parallel
to the path of the beam and advanced. The drill is then removed,
leaving the drill bit in position to confirm the direction and
placement. Small adjustments can be made on the basis of the
interpretation of the fluoroscopic image. The drill is then advanced
through the nail, a depth gauge is used to determine the screws length,
and an interlocking screw can be placed. The same procedure is repeated
for a second interlocking screw if necessary.

Clinical Experience with Antegrade Intramedullary Nailing. Reamed Intramedullary Nailing Using a Piriformis Entry Technique. The early experience with intramedullary nailing was universally positive.^* Most of these reports used the techniques described by Kuntscher¹⁵⁵
and reamed nails were placed in a closed fashion without interlocking.
These implants relied on the endosteal contact between the nail and the
femur for control of rotation, angulation, and length. The perceived
advantages compared with other techniques included early joint
mobility, early weight bearing, shortened hospital stay, decreased
pain, and predictable union. Fracture comminution and segmental
fractures could be treated with antegrade reamed nails with reports of
100% union and no infections.²⁹⁶,²⁹⁷ In the classic series of 520 femoral fractures in 500 patients treated with reamed intramedullary nailing, Winquist et al.²⁹⁸
reported a union rate of 99.1%. However, shortening of greater than 2
cm occurred in 10 patients and rotation of greater than 20 degrees was
identified in 12 patients.²⁹⁸

As interlocking became available, the indications for
antegrade intramedullary nailing continued to expand and the results
have continued to improve. Brumback and coauthors described the
technical decision making, the need for static locking, and the
expected healing in a series of three articles reviewing locked
intramedullary nailing of femoral shaft fractures.³⁷,³⁹,⁴⁰,⁴² The previously reported problems with shortening and postoperative loss or rotational alignment were largely eliminated.^† Comminuted and rotationally unstable fractures healed in 98% of 112 reamed and interlocked nailings reported by Wiss et al.³⁰¹ Similar results were reported by Sojbjerg et al.²⁵⁶ and Kempf et al.¹⁴⁴
The use of interlocked nails in segmental femoral fractures was
similarly reviewed with healing observed in 97% of patients at an
average of 32 weeks without any additional interventions other than
nail dynamization.³⁰⁰ Brumback et
al. investigated the need for planned conversion from static to dynamic
fixation in a prospective study of 100 femoral shaft fractures treated
with intramedullary nailings.⁴² They
reported healing in 98% with static interlocking fixation and concluded
that the routine conversion to dynamic fixation was not necessary.⁴² Complications related to closed interlocked nailing was reported by Benirschke et al.¹⁸
in a review of 267 patients evaluated retrospectively and
prospectively. Intraoperative technical complications were observed
uncommonly, but functional outcomes were not as good as previously
appreciated; 37% of patients had pain and 39% had some limitations with
ambulation or standing. More recently, Wolinsky et al.³⁰⁴
reported their results with 551 femoral shaft fractures treated with
closed locked antegrade intramedullary nailing. The authors reported
union in 98.9%, infection in 1%, and no malunions with greater than 10
degrees of angulation.

The positive effects of reaming on fracture healing is
thought to be from a combination of altered blood flow to the bone and
the local muscles, and the deposition of marrow and cortical elements
at the site of the fracture.^‡ The impact of these factors
has been studied in animal models in both the tibia and the femur. More
recent studies suggest that reaming increases a number of growth
factors (platelet-derived growth factor, vascular endothelial growth
factor, insulin-like growth factor I, transforming growth factor
beta-1, and bone morphogenic protein-2) and this is hypothesized to
contribute to the positive healing effects of reaming.¹⁰¹ A potential negative effect of reaming includes injury to the inner two thirds of the bone.¹⁴⁵
Animal studies in the tibia suggest that this process is reversed by 12
weeks after reamed nailing and by 6 weeks after unreamed nailing.²⁴⁸
However, reaming has been shown to have a positive influence on the
surrounding muscle perfusion, which may have implications for the
extraosseous blood supply to the bone and the healing fracture.¹²⁵ In one study, periosteal blood flow was found to increase by a factor of six within 30 minutes of reaming.²²³ Total blood flow and callus blood flow increases have been similarly demonstrated in animal models.¹⁰⁹

Unreamed Femoral Nailing.
The role of unreamed intramedullary nailing for the treatment of
femoral shaft fractures remains unclear. Although the decreased damage
to the endosteal blood flow has been demonstrated in animal studies,
the clinical relevance with respect to infection and union remains
elusive. In addition, the ill effects of reaming on pulmonary function,
especially in the multiply injured patient, have not been definitively
demonstrated.^§ The early clinical results of unreamed nailing for femoral shaft fractures reported by Krettek et al.¹⁵¹
were favorable, with only minor implant-related complications reported.
Nonunion was reported in 5.1% of patients in a multicenter review of
unreamed nails reported by Hammacher et al.¹¹⁴
The authors conceded that the use of a small nail inserted without
reaming did not produce a decrease in the number of patients with adult
respiratory distress.¹¹⁴ Herscovici et al.¹²⁰
reported similar results in 125 fractures treated with unreamed nails.
Although this represented a mixture of antegrade and retrograde
nailings, only 93% healed after the index procedure.

Reamed versus Unreamed Antegrade Intramedullary Nailing.
The potentially negative effects of reaming for insertion of
intramedullary nails include elevated intramedullary pressures,
elevated pulmonary artery pressures, increased fat embolism, and
increased pulmonary dysfunction. The potential advantages of reaming
include the ability to place a larger implant, increased union, and
decreased hardware failure. The increased union is believed to be
attributable to both the enhanced biomechanical environment of a larger
implant and the enhanced biologic environment produced during reaming.
Reaming increases the periosteal blood flow and deposits osteoinductive
elements at the fracture, both of which may contribute to improved
healing.¹⁰⁹,²²³

Several studies evaluate specifically the clinical
results of treating femoral shaft fractures with intramedullary
interlocked nails with and without intramedullary reaming.^* Of interest are the different results observed with the retrospective and prospective studies comparing the two techniques.

In a retrospective study of 147 consecutive patients
treated with reamed (n = 50) and unreamed femoral nails (n = 97),
Giannoudis et al.⁹⁷ reported union
at an average of 26.9 weeks in the unreamed group compared with 20.5
weeks in the reamed group. The operative time was shorter with the
unreamed technique, and there was no increased risk of infection,
angular deformity, or limb length discrepancy. This led the authors to
conclude that the unreamed technique was more rapid and had no
associated clinical problems of significance. Similarly, Reynders and
Broos²²⁴ retrospectively reviewed
107 closed femoral fractures treated with reamed (n = 54) and unreamed
(n = 53) intramedullary nailings. The authors reported no difference in
union time but significantly less operative time with the unreamed
technique. This led to the recommendation that unreamed femoral nailing
should be used to treat acute closed femoral shaft fractures.

However, several prospective and randomized trials have
presented differing results with contrasting conclusions. In a
prospective and randomized study of 81 patients treated with a
stainless steel statically locked intramedullary nail inserted with and
without reaming, Tornetta and Tiburzi²⁷⁸
reported statistically similar operative times, transfusion
requirements, and time to union when the entire groups were compared.
Reaming was associated with more rapid callus formation yet higher
blood loss. Further analysis of the data revealed faster union in the
subpopulation of patients with distal fractures when reaming was used.
Intraoperative technical complications occurred more commonly with the
unreamed technique.²⁷⁸ In a subsequent study by the same authors,²⁷⁷
172 patients were prospectively randomized and their results were
compared. With the larger data set available, the authors were able to
show a significant improvement in the time to union with reaming (80
days) compared with those nails placed without reaming (109 days). This
led to the conclusion that there is no advantage to the routine use of
nailing without reaming.²⁷⁷ Clatworthy and coauthors⁶¹
reported their results of a prospective and randomized trial comparing
reamed and unreamed nails in 45 patients with femoral shaft fractures.
Of note, the study was stopped early because of the high rate of
implant failure in the unreamed group. They reported more significantly
more rapid union (28.5 versus 39.4 weeks) and a much lower proportion
of nonunions at 9 months (18% versus 57% nonunion) in patients treated
with reamed nails compared with those treated with unreamed nails.
Secondary procedures to achieve union were more common in the unreamed
group. Finally, in a multicenter, prospective, randomized clinical
trial from seven level-I trauma centers sponsored by the Canadian
Orthopaedic Trauma Society, 224 patients were enrolled to compare the
rate of nonunion after intramedullary nailing with and without reaming.
The authors reported nonunions in 7.5% of patients treated without
reaming compared with 1.7% of patients treated with reaming. They
determined the relative risk of nonunion was 4.5 times greater without
reaming and with the use of a relatively small-diameter nail.⁴⁸

In a systematic review and meta-analysis of lower extremity long bone fractures, Bhandari et al.²¹
attempted to determine the effects of reaming on the rates of nonunion,
implant failure, malunion, compartmental syndrome, pulmonary embolus,
and infection. Only prospective and randomized trials were used as part
of the analysis. Although the data represented both tibias and femurs
treated with intramedullary nails, the evidence suggests that reamed
nailing significantly reduces the rates of nonunion and implant failure
compared with unreamed nailing.²¹

Trochanteric Entry Antegrade Nails.
Trochanteric entry nails have been used with success for the treatment
of intertrochanteric, subtrochanteric, pertrochanteric, and combination
fractures of the proximal femur. Some of the potential advantages of
the trochanteric tip as a starting portal for placement of a medullary
implant include less operative time and less intraoperative
fluoroscopy, largely because of ease of identification of the
trochanter as a starting point. If a nail designed for placement into
the piriformis fossa is placed through the more readily accessible
trochanteric tip, the complications of iatrogenic fracture comminution
and varus malreduction are increased.⁹²,¹⁴¹,²⁰⁴
However, newer nail designs with appropriate proximal bends have
allowed for placement of a medullary implant through the trochanteric
tip. However, the initial nail designs used for stabilization of these
proximal injuries were characterized by a large proximal segment,
typically requiring a 17-mm

reamer
at the entry site. In a cadaver model designed to investigate the
impact of this large entry portal in the tip of the greater trochanter,
a significant portion (average of 27%) of the gluteus medius tendon
insertion was found to be disrupted.¹⁷³ The clinical significance of this is currently unknown.

Trochanteric entry nails for the treatment of routine
diaphyseal femur fractures is a relatively contemporary approach. Newer
nail design characteristics including a smaller proximal diameter of
the implant and multiple bends have allowed placement through the
trochanteric tip. The smaller entry portal has the theoretic advantage
of minimizing any additional soft tissue damage at the gluteus medius
tendon. Ricci et al.,²²⁸ in a
prospective nonrandomized study, compared the results after treatment
with antegrade nails placed either through the piriformis fossa or the
greater trochanter. In both techniques, alignment was maintained and no
iatrogenic fracture comminution occurred. The average fluoroscopy time
was significantly greater in the piriformis insertion group, and a
longer average operative time was required in these patients. These
differences were amplified in obese patients in whom identification of
the piriformis starting portal is difficult. Patient outcomes were
similar between the two groups. The authors suggested that a
specifically designed trochanteric entry nail represents a viable
alternative for the management of femoral shaft fractures.²²⁸

Background. Retrograde femoral nailing was first introduced by Kuntscher for the treatment of pertrochanteric femoral fractures.¹⁵⁶
A variety of techniques and implants were subsequently described
including flexible implants introduced through extra-articular distal
femoral portals with and without distal femoral locking capabilities.
These implants were used to treat both distal and proximal femoral
fractures, as well as to prophylactically stabilize proximal femoral
metastases.⁸¹,²⁴³,³¹⁴
An extra-articular medial condylar starting portal was suggested by
Swiontkowski et al. for the treatment of ipsilateral femoral neck and
shaft fractures, using antegrade femoral implants.²⁶⁴ Sanders et al.²⁴²
expanded the indications to include the multiply injured patient, other
ipsilateral fractures, and pregnancy. Insertion of both femoral nails
and tibial nails was possible using a medial condylar distal femoral
entry site, enabling fracture healing in difficult situations.

The use of short supracondylar nails placed through an
intercondylar distal femoral starting portal was introduced in 1991
with the genucephalic nail.¹¹⁹ This
implant was initially used to treat distal femoral fractures and as an
alternative for revision of failed fixations in the distal femur. The
use of this implant expanded to include treatment of intercondylar
fractures and periprosthetic fractures of the distal femur. Because of
the short length of the implant, the initial application to diaphyseal
injuries was limited.¹¹⁹,¹⁶⁸,²⁴⁷
Experience and difficulties with the shorter nails, combined with the
success of longer retrograde femoral implants placed through various
starting portals, ultimately led to the development of retrograde
femoral nails specifically designed to span the entire femur.¹⁰⁸,¹²¹,¹⁸²,²⁰²,²¹⁷
The major implant design changes that have facilitated the expansion of
indications include a more appropriate anterior bow, multiple distal
interlocking options, anterior to posterior proximal interlocking
options, and variable nail dimensions. These implants can now be used
in certain circumstances to treat femoral diaphyseal fractures.

Indications and Contraindications. The early experience
with the use of retrograde nails has produced a number of clinical
circumstances in which the implants may offer an advantage over other
techniques of femoral shaft stabilization.¹²¹,²⁴² However, these remain relative indications (Table 50-6, Fig. 50-23).
The major advantage with a retrograde entry portal is the ease in
properly identifying the starting point. In addition, distraction of
the fracture site is decreased with retrograde nailing compared with
the antegrade technique. In situations in which surgical time and
repositioning cannot be tolerated by the patient, retrograde nailing
may be more suitable than the antegrade technique. Ipsilateral injuries
such as femoral neck, pertrochanteric, acetabular, patellar, or tibial
shaft fractures may make retrograde nailing a desirable option. In the
case of an ipsilateral intracapsular femoral neck fracture, proper
fixation of the femoral neck should remain a priority and is best
accomplished with multiple cannulated screws, a sliding hip screw, or
combinations thereof. The femoral shaft fracture can be treated with a
retrograde nail or plate fixation, both of which have been shown to
predictably produce fracture union (Fig. 50-24).
Depending on the surgical technique and the planned operative approach
in a patient with an ipsilateral acetabular fracture, retrograde
nailing remains a consideration. This is most relevant if a planned
extensile or posterior approach to the acetabulum is planned because
the antegrade nail insertion site may be within the subsequent surgical
exposure. This is probably not a large clinical concern for most
surgeons experienced in the treatment of acetabular fractures. If a
retrograde technique is used in this situation, consideration should be
given to over-reaming the femoral canal significantly to allow for
atraumatic retrograde nail placement, avoiding further surgical impact
on the already injured acetabulum. In patients with ipsilateral tibial

shaft
fractures (i.e., a “floating knee”), a single percutaneous incision for
the placement of a retrograde femoral nail and an antegrade tibial nail
has been advocated and accomplished with good results. However, the
length of the incision for the percutaneous placement of an antegrade
femoral nail is limited, and the advantage of avoiding this remains
unknown. The presence of an ipsilateral patellar fracture requiring
surgical fixation certainly allows for easy identification of the
starting point for a retrograde femoral nail. However, similar to the
situation in patients with an ipsilateral tibial fracture, the
morbidity associated with the introduction of an additional surgical
incision at the hip is probably minimal.

Several other relative indications for retrograde nails
have been suggested. The presence of multiple other fractures and
multiple other organ system injuries suggest that rapid stabilization
of a patient in the supine position may be optimal. Although femoral
nailing certainly can be quickly accomplished with an antegrade
technique, the retrograde entry site may be performed more rapidly in
some circumstances. The rapid placement of a small-diameter unreamed
retrograde nail in the multiply injured patient as an alternative to
temporary external fixation has been described as “Damage Control
Nailing” and can be considered.¹²²
Bilateral femoral fractures can be treated with retrograde nails
without repositioning the patient, potentially decreasing the total
operative time in a critically injured patient. In patients with an
associated vascular injury, rapid treatment with an initially unlocked
retrograde implant can be performed, allowing fracture stabilization to
facilitate the subsequent vascular repair. In the morbidly obese
patient in whom the entry site for antegrade nailing is particularly
difficult, a retrograde nail may be a reasonable alternative and has
been associated with decrease operative time and radiation exposure.²⁸²
Similarly in pregnant patients, retrograde nails may be easier to place
and have the advantage of potentially decreased radiation exposure
around the hip. In patients with distal femoral periprosthetic
fractures above a total knee arthroplasty, a retrograde femoral nail
may offer a reasonable fixation option. The design of the femoral
component and the available space for placement of the retrograde
implant are necessary before consideration of retrograde medullary
fixation. The fixation of the medullary implant in the stress-shielded
distal femoral metaphysis may be poor, making consideration of other
strategies necessary. Finally, an ipsilateral through knee amputation
in a patient with an associated femoral shaft fracture is a good
indication for a retrograde nail.

Several contraindications for the use of retrograde
nails exist. Because of the need for knee flexion that allows nail
placement,²⁷⁹ patients with
restricted knee motion may be poor candidates for retrograde femoral
nailing. Similarly, patients with patellar baja may not have adequate
room for placement of a retrograde nail between the patella and
anterior tibial plateau with the knee flexed. Subtrochanteric
fractures, although successfully treated with retrograde nails in some
instances, may be difficult to control unless the nail is placed to an
extremely cephalad location. The presence of an associated open
traumatic wound may make retrograde nailing less desirable because
these injuries are associated with an increased risk of infection with
the potential for subsequent septic arthritis of the knee.

Surgical Technique. The patient is positioned supine on
a radiolucent table that allows unimpeded fluoroscopy from the knee to
the hip. A small bump beneath the ipsilateral flank may help avoid
external rotation of the leg. The entire leg and hip are prepped to the
iliac crest and medially to allow access to the femoral artery if
needed. Pharmacologic relaxation is necessary in all cases, especially
if manual traction is the primary reductive force. The C-arm is
positioned perpendicular to the long axis of the table on the
contralateral side to allow for ease with imaging and placement of
interlocking screws. Knee flexion of 34 to 52 degrees²⁷⁹
is necessary to allow identification of the proper entry portal and
placement of the nail. This can easily be accomplished with several
bolsters or a large radiolucent triangle. Femoral length and fracture
reduction can be accomplished manually, with a femoral distractor¹⁸³ or with percutaneous and unicortical Schanz pins used as joysticks.⁹⁵
A femoral distractor can be used to maintain length, alignment, and
rotation during the entire procedure. The proximal pin can be placed
from anterior to posterior, medial to the predicted nail path. Distally
the pin is placed from lateral to medial, anterior to the predicted
nail path. Percutaneous Schanz pins for manipulative reduction can be
placed from lateral to medial, unicortically, and close to the
fracture. As with any nailing procedure, the fracture should be reduced
during reaming and nail placement, and until interlocking is complete.

A longitudinal incision centered between the inferior
pole of the patella and the tibial tubercle is optimal. A smaller
incision of 2 to 3 cm at the location of the perfect starting portal is
certainly possible and preferred by many surgeons. The approach can be
accomplished medial to or through the central portion of the
infrapatellar tendon. The starting point can be palpated or identified
radiographically. The location for the proper starting point is
anterior to the posterior cruciate ligament origin and at or slightly
medial to the intercondylar sulcus in line with the canal of the femur.
This has been identified as 6.2 to 12 mm anterior to the posterior
cruciate ligament femoral attachment and through the articular
cartilage of the posterior intercondylar sulcus.⁵²,¹⁵²
Alternatively, the entry site can be identified radiographically on the
basis of the lateral fluoroscopic image. With the femur rotated into a
perfect condylar overlap lateral view, the proper entry site is at the
apex of Blumensaat’s line (Fig. 50-25).
However, the optimal entry side based on the lateral radiographic
projection is dependent on the implant design and may be more anterior
than previously suggested.¹⁹¹ Given
the anatomical valgus of the distal femur the entry angle should not be
perpendicular to the femoral articulation because this can produce a
varus reduction if the fracture is located distal to the femoral
isthmus. Instead, the entry angle should be in 5 to 9 degrees of valgus
as confirmed on the intraoperative AP fluoroscopic image, coincident
with the axis of the femoral canal. The starting point can be initiated
with either a sharp awl or a cannulated drill. Use of a cannulated
drill has the advantage of allowing for minor corrections in the entry
angle before committing with a large opening in the distal femoral
articular surface. Both reamed and unreamed techniques have been
described, and the relative advantages of each are probably similar to
those of antegrade techniques. For placement of a reamed nail, a
ball-tipped guidewire is placed across the fracture and into the
medullary canal of the opposite segment. The guidewire should be
advanced proximal to the lesser trochanter to ensure placement of a
long implant. Reaming is performed with the fracture reduced and the
knee flexed to the proper degree that minimizes damage to the inferior
pole of the patella, the proximal tibia, and the distal femoral
articulation. The canal should be overreamed by at least 1 mm greater
than the desired nail diameter. If a proximal femoral or ipsilateral
acetabular fracture is present, consideration should be given to
over-reaming by 1.5 to 2.0 mm to allow for atraumatic nail passage.

Most current nail designs have an anterior bow with a
radius of curvature that is greater than that of the normal human
femur. As a result, extension of the fracture may occur despite a
proper entry portal for the implant. A more posterior entry site, in

addition
to potentially damaging the femoral posterior cruciate ligament origin,
can result in accentuation of fracture site extension. In contrast, a
more anterior starting point will produce fracture site flexion and be
located in the patellofemoral articulation. The nail should be recessed
beneath the cartilage surface distally and above the lesser trochanter
proximally (Fig. 50-26).
In a study using a cadaveric femoral model tested with pressure
sensitive film, Morgan et al. demonstrated that placement of the nail 1
mm prominent at the knee joint resulted in significant increases in
patellofemoral joint contact pressures at 90 and 120 degrees of knee
flexion when compared with nails placed flush or recessed.¹⁹⁰
Distal interlocking is typically facilitated with a targeting jig that
allows placement of a variable number of screws from lateral to medial.
For some nail designs, endcaps are available that contact the ultimate
distal interlocking screw, preventing toggle of the screw-nail
junction. Locking of the distal interlocking bolts to the nail may be
beneficial for decreasing fracture site motion.²⁶⁸
In addition, bolts are available that can be placed on the medial ends
of the screws as they exit the medial metaphyseal femoral cortex,
potentially increasing the fixation in osteoporotic bone. The knee
joint should be vigorously irrigated after nail placement to remove any
debris that has accumulated from reaming. Proximal interlocking can be
performed from anterior to posterior or from lateral to medial
depending on the design of the nail. Given the need to maintain the
proper femoral rotation throughout the procedure, anterior to posterior
proximal interlocking is usually easier. Because of the potential for
injury to the neurovascular structures located medially, internal
rotation of the nail relative to the femur should be avoided. The major
structures at risk during placement of anterior to posterior
interlocking screws include the femoral artery and branches of the
profundus artery medially, and the sciatic nerve posteriorly. The
relative safe zone for placement of anterior to posterior interlocking
screws has been described at the level of the lesser trochanter. In
this location, the safe corridor was identified from 7 degrees medial
to 20 degrees lateral relative to the sagittal axis. This safe zone
decreased by 52% in patients with an ipsilateral acetabular fracture,
according to examination of CT scans.³⁵

There are several complications that are unique to
retrograde nailing. Injury to the local neurovascular structures with
proximal interlocking is possible and has been reported, including
injury to a branch of the profunda femoris artery.⁶⁶
An incision that allows direct palpation of the anterior femur combined
with avoidance of internal rotation of the nail minimizes this risk.

In
addition, locking performed above the level of the lesser trochanter
further minimizes the risk of injury to the local neurovascular
structures.²³¹
Although injury to the popliteal artery has been reported with
retrograde nailing, this complication appears to be more related to the
injury than to the implant or the retrograde technique.¹²
Heterotopic ossification in intra-articular and periarticular locations
has been reported after retrograde nailing, and this should remain a
concern in patients with associated risk factors.¹²³ Synovial metallosis has been observed after retrograde medullary nailing, but the impact of this is unknown.¹³⁶
The negative impact of the retrograde femoral reaming on the blood flow
to the distal femur and the cruciate ligaments has been demonstrated in
an animal model, but the clinical impact is unclear.⁷⁹
The use of retrograde nails for the treatment of proximal fractures
remains controversial. In a review of 17 subtrochanteric femur
fractures treated with retrograde nails, DiCicco et al.⁷³ reported healing with greater than 5 degrees of varus malalignment in 35% of patients. Similarly, Ricci et al.₂₂₆
reported more problems with angular malalignment of proximal third
femoral fractures treated with intramedullary nailing when compared
with middle third fractures. The exact distance between the proximal
extension of the fracture and the interlocking screws is unknown, but 5
cm is probably advisable as a minimum length of intact femoral
diaphysis.

Clinical Experience with Retrograde Medullary Nails. The
early clinical experience with the use of long retrograde nails placed
through an intercondylar starting point for the treatment of femoral
shaft fractures was largely limited to the multiply injured patient and
those patients with severe ipsilateral lower extremity injuries.¹⁰⁸,¹²¹,¹⁸²,²¹⁷ Patterson et al.²¹⁷
reported on 17 femur fractures in 16 patients treated with retrograde
nails and followed for an average of almost 2 years. Medullary nails
normally designed for antegrade implantation were used in these
patients. The indications for the use of a retrograde nail was an
ipsilateral femoral neck fracture in eight patients, an ipsilateral
through or above knee amputation in five patients, and other associated
injuries including a traumatic arthrotomy and anipsilateral acetabular
fracture. Fourteen of these 17 injuries were open fractures.
Complications were observed commonly and were likely related to the
severity of the initial injuries. The poor results were due to
nonunions, loss of knee motion, and infection; hardware removal was
required commonly. Despite this, the authors believed that knee
function was not adversely affected and that the intercondylar starting
point could be used successfully in difficult injuries.²¹⁷ Moed and Watson¹⁸²
described their experience with 22 femoral shaft fractures treated with
unreamed retrograde nails placed through the intercondylar starting
point. Union occurred in less than 4 months in 19 cases, and the major
complications were three nonunions and a rotational malunion.
Restriction of knee motion was not observed. Herscovici and Whiteman¹²¹
further described the indications for the use of retrograde nails based
on their experience with 45 fractures. They observed a rotational
abnormality in one patient and two nonunions that both healed after a
revision procedure. Knee motion averaged 129 degrees; however,
decreased flexion and an extensor lag were observed in a minority of
patients. The authors suggested that the relative indications for this
procedure may include ipsilateral proximal femoral fractures,
ipsilateral acetabular fractures, ipsilateral tibial fractures, morbid
obesity, and bilateral femoral shaft fractures. These initial studies,
although promising, suggest that knee-related morbidity does exist with
these implants and that union rates may be lower than in patients
treated with antegrade, statically locked, reamed medullary nailing.
However, the complexity of these injuries and the use of antegrade
nails placed in a retrograde fashion in the most severe injuries likely
have contributed to these observations.

As full-length nails specifically designed for
retrograde applications became more readily available, prospective
studies evaluating their use in all femoral fractures as well as in
comparison with antegrade techniques became possible. Early reports
demonstrated an increased number of nonunions and delayed unions
compared with historical reports of antegrade, reamed medullary nailing
techniques.¹⁸³,²⁰² Moed et al.¹⁸³
prospectively reported on a consecutive series of patients with femoral
shaft fractures treated with unreamed retrograde nails placed through a
patellar tendon splitting approach. Early nail dynamizations were
performed at 6 to 12 weeks. The authors observed a decrease rate of
nonunion (6%) and time to union (12.6 weeks) when compared with their
earlier study. There were minimal complaints of knee pain, and knee
function was rated as excellent using a clinical rating system. Ostrum
et al.²⁰² similarly reported on a
prospective and consecutive series of 61 fractures treated with a
reamed 10-mm retrograde nail placed through a medial parapatellar
incision. More than half of these fractures had significant (Winquist
type 3 or 4) comminution, and 26% were open injuries. The authors
observed a union rate of 95% when they included one planned bone
grafting for traumatic bone loss and five nail dynamizations. One case
of knee sepsis was observed, leading the authors to recommend caution
in patients with open fractures. Associated knee problems were not
encountered.

Antegrade and retrograde nailing techniques have been
directly compared and reported in one retrospective study, one
prospective, and two prospective and randomized studies.²⁰¹,²²⁶,²⁷⁷,²⁸²
In a retrospective study comparing 134 femoral fractures treated with a
retrograde technique and 147 fractures treated with an antegrade
technique, Ricci et al.²²⁶ compared
the rates of delayed union, nonunion, malunion, and complications at an
average of 23 months. Union occurred after the initial procedure in 89%
of the fractures treated with antegrade nails and 88% of those treated
with retrograde nails. The final rate of union after additional
interventions and the rate of malunions were the same between the two
groups. After excluding patients with ipsilateral hip or knee injuries,
the authors reported a significantly higher rate of pain at the
associated entry site for each group. In patients treated with
retrograde nails, 36% had knee pain compared with 9% of those treated
with antegrade nails. Ipsilateral hip pain was reported in 10% of
patients treated with antegrade nails compared with 4% of patients
treated with retrograde nails. This led the authors to conclude that
although union rates between the two treatment approaches are
comparable, complications related to the knee and hip are related to
the nail insertional site. Ostrum et al.²⁰¹
compared femoral fractures treated with antegrade and retrograde nails
in a prospective and randomized study of 100 patients. Small-diameter
(10-mm) nails were inserted after reaming in both groups. The authors
reported fewer nail dynamizations and more rapid healing with the
antegrade technique by almost

4
weeks. Knee pain was found to be equal in both groups, but thigh pain
was observed to be associated with antegrade nails. Final healing after
secondary procedures was similar between the two groups. In another
prospective and randomized study comparing antegrade and retrograde
femoral nailing for femoral shaft fractures, Tornetta and Tiburzi²⁷⁷
found no difference in operating time, blood loss, technical
complications, nail size, or the need for blood transfusions in 69
fractures. Reamed nails were used with both insertional techniques. The
authors reported an increased rate of rotational and length
discrepancies with the retrograde technique. In their study, the
antegrade procedures were performed using a fracture table, whereas the
retrograde procedures were accomplished with manual traction on a
radiolucent table. The time to union and the rate of union (100%) were
the same for both surgical approaches. Because of the limited length of
follow-up, no conclusions could be drawn regarding long-term knee
function or complications. In a more recent follow-up presentation,
Tornetta and coauthors reported similar results in knee range of
motion, stair-climbing ability, pain medication requirements, and Short
Form-36 outcomes in patients prospectively randomized to treatment with
either an antegrade or retrograde nail.²⁸⁰ In a prospective and nonrandomized multicenter study, Tucker and coauthors²⁸²
demonstrated similar outcomes in patients treated with an antegrade or
a retrograde technique. The emphasis of their study was primarily on
the difference in patients who were obese (defined as body mass index
greater than or equal to 30 kg/m²).
In obese patients, the retrograde technique was associated with
significantly decreased operative and radiation exposure times,
supporting the use of this technique in some obese patients.²⁸²

Operative stabilization of femoral shaft fractures
allows early patient mobilization, decreases pain, facilitates nursing
care, minimizes joint stiffness, and allows early functional
rehabilitation. Early mobilization avoids many of the complications
associated with prolonged recumbency such as pulmonary compromise,
pressure sores, and muscle deconditioning.

Patients usually experience significantly diminished
pain after femoral stabilization. As a result, they should be
encouraged to sit up and get out of bed immediately after fixation.
Because of the strength of a femur treated with a statically locked
intramedullary nail, there should be virtually no concern by the
patient or the physician regarding the stability of the mechanical
construct. This applies to femurs stabilized with both antegrade and
retrograde nailing techniques. Quadriceps and hamstrings exercises can
proceed according to the patient’s comfort. Unrestricted active and
passive range-of-motion exercises of the knee and hip can similarly be
instituted immediately after surgery. Restoration of motor strength is
dependent on the traumatic injury to the muscles, any associated
injuries, and the patient’s motivation. In patients with isolated
femoral shaft fractures, supervised outpatient physical therapy is
frequently unnecessary.

Weight bearing on the extremity is guided by a number of
factors including the patient’s associated injuries, the soft tissue
injury, and the location of the fracture. For fractures treated with
the currently manufactured intramedullary nails, immediate weight
bearing is safe from a mechanical standpoint. Brumback et al.⁴¹
reported the biomechanical and clinical results of simulated and actual
early weight bearing using commonly available implants. They found that
immediate weight bearing of segmentally comminuted midisthmal fractures
treated with a statically locked intramedullary nail using two distal
interlocking screws was safe and allowed healing without shortening or
nail fatigue failure. Similarly, Arazi et al.⁷
followed 24 patients with comminuted diaphyseal femur fractures for a
minimum of 1 year. Unrestricted weight bearing was allowed, and most
patients were able to bear weight between 2 and 4 weeks after surgery.
No nail-related mechanical failures occurred, and all fractures healed
without complications. An interlocking screw in each of two patients
demonstrated some bending, but this was without consequence. The
experience reported in these two studies supports the contention that
early weight bearing after reamed statically locked antegrade
intramedullary nailing is safe. Early weight bearing may encourage
callus formation and should be encouraged in applicable patterns.
Fractures with proximal or distal extension and associated ipsilateral
femoral fractures require individual modification to the weight-bearing
progression. If weight bearing is limited during the first 6 to 12
weeks, an ankle dorsiflexion splint or orthosis should be used to avoid
an equinus contracture.

Radiographic evaluations are usually obtained at 6
weeks, 12 weeks, and 6 months and should include two views of the
entire femur to include the proximal and distal interlocking screws as
well as the hip and knee joints. Groin pain in a patient who has
sustained a femoral shaft fracture should be evaluated at any point in
the postoperative course given the possibility of an associated
clandestine or iatrogenic femoral neck fracture. However, clandestine
femoral neck fracture has been observed in asymptomatic patients as
well.²³⁰ A femoral neck fracture was
ultimately identified in 10 patients without radiographic evidence
before plating of an ipsilateral diaphyseal femoral fracture. This
review emphasizes the need for dedicated hip radiographs
postoperatively and at any time a patient becomes symptomatic.²³⁰

The treatment of the femur fracture in the multiply
injured patient, especially those patients with associated chest
trauma, remains a topic of some contention. Early retrospective studies
suggested that immediate internal fixation of long bone fractures was
beneficial in patients with multiple injuries, and this included the
use of reamed intramedullary nails.¹⁰³,¹³⁷,²³³,²³⁴,²⁶⁵
Although the risk of fat embolism was recognized, most series suggested
that early femoral stabilization actually decreased the incidence of
severe fat embolism and pulmonary complications. Talucci et al.
reported five cases of fat embolism syndrome in patients treated with
delayed intramedullary nailing and no cases in patients treated with
immediate nailing.²⁶⁵ Johnson et al.¹³⁷
retrospectively evaluated 132 patients with multiple musculoskeletal
injuries with an ISS of 18 or higher. They found that early operative
stabilization of the majority of fractures was associated with a
decreased incidence of ARDS. Because of the retrospective nature of the
study, however, firm conclusions could not be determined. In a landmark
article, Bone et al.²⁵ addressed the impact of early stabilization of long bone fractures

on pulmonary complications, hospital stay, and the number of days in the intensive care unit.

In a prospective and randomized study, 178 patients were
treated with either early (<24 hours) or delayed femoral
stabilization. The patients were divided on the basis of the number of
associated injuries as either isolated or multiply injured. The
incidence of pulmonary complications including ARDS, fat embolism, and
pneumonia was lower in both groups of patients treated with immediate
femoral stabilization.

Although early stabilization of long bone injuries
appears to be clearly important in the multiply injured patient, the
impact of femoral nailing and especially reaming is less clear. This
has been especially contentious in the patient with a concomitant
pulmonary injury. Several investigators have therefore looked
specifically at the effects of reaming and intramedullary nailing on
pulmonary function in animal models.⁷⁷,²¹¹,³⁰²,³⁰⁶,³⁰⁷ In a hemorrhagic shock and lung contusion model, Pape et al.²¹¹
found increased pulmonary artery pressure and increased lung capillary
permeability in animals treated with reamed nails compared with
unreamed nails. In contrast, Wolinsky et al.³⁰²
found that femoral reaming did not affect pulmonary shunt or pulmonary
compliance in an ARDS sheep model if appropriately resuscitated.
Similarly, Duwelius et al.⁷⁷ found that pulmonary dysfunction did not increase in a pulmonary contusion model.

Given the contrasting findings in these laboratory
investigations, several clinical studies sought to address the issue of
reaming and intramedullary nailing on pulmonary function and mortality
in patients with an associated thoracic injury.²⁷,²⁹,⁵¹,⁵³,²⁰⁹,²⁸⁵ Pape et al.²⁰⁹
investigated whether early intramedullary stabilization of femoral
shaft fractures in multiply injured patients may be detrimental in the
subpopulation of patients with an associated thoracic injury. In a
retrospective review of 766 patients, 106 met the inclusion criteria
and all were multiply injured (ISS >18). The authors reported an
increased incidence of ARDS and mortality in the patients who had
associated severe chest trauma and were treated with immediate nailing.
However, this is in contradistinction to the study by Charash et al.,
who arrived at a different conclusion.⁵³
That is, pulmonary complications were decreased in all patients who
underwent early femoral stabilization with reamed nailing, including
those with thoracic trauma.

In an effort to look specifically at the impact of
reamed femoral nailing on the multiply injured patient and the patient
with thoracic trauma, Bosse et al.²⁷
designed a study looking at patients treated at two different centers.
Although the patient populations and the protocols for trauma care were
similar at the two centers, reamed intramedullary nailing was used in
95% of patients at one institution, whereas plating was used in 92% of
patients at the other institution. This allowed for comparison of the
effects of reamed nailing and not necessarily the impact of a femoral
fracture on outcomes. The overall incidence of ARDS was only 2% in
patients with a femoral fracture compared with 6% to 8% in patients
without a femur fracture but with a thoracic injury. The occurrence of
ARDS, pneumonia, pulmonary embolism, multiple organ system failure, and
death were similar regardless of the type of treatment or the presence
of a thoracic injury. This led to the conclusion that nailing with
reaming for the acute stabilization of femoral fractures in the
multiply injured patient with a thoracic injury did not increase the
occurrence of the aforementioned complications.

Currently, ongoing investigations continue evaluating
the impact of intramedullary instrumentation on the pulmonary
complications and mortality in the multiply injured patient.^*
In a prospective and randomized evaluation of patients treated with
early intramedullary nailing for their femoral fractures, no difference
was found in the incidence of ARDS in patients treated with reamed or
unreamed nails. The authors observed a low overall incidence of ARDS
and could not detect a difference given the small sample size of
multiply injured patients.⁴⁸
However, there continues to be a significant body of evidence that
suggests that in a specific subpopulation of patients, early
intramedullary nailing may be deleterious and is associated with
elevation of certain proinflammatory markers, especially interleukin
(IL)-6.^† Both the femoral fracture and the process of
reaming has been shown to increase IL-6 levels, which may indicate an
increase in the inflammatory response and the presence of an additional
insult due to the surgical treatment.¹⁹²
This IL-6 increase has been shown to be closely correlated with injury
severity and the development of a systemic inflammatory response state.⁹⁸,²¹⁶
This has led to the recommendation that early external fixation of long
bone fractures followed by delayed intramedullary nailing may minimize
the additional surgical impact in patients at high risk for developing
complications (see Chapter 9).

Open femoral shaft fractures are typically the result of
high-energy trauma. These patients frequently have multiple other
orthopaedic injuries and involvement of several organ systems. Vascular
injuries are commonly observed in association with open femoral shaft
fractures.

Because of the thick muscular envelope surrounding the
femoral shaft, an associated open traumatic wound suggests significant
initial displacement and associated soft tissue disruption. Unlike the
tibia, which is entirely subcutaneous over its anteromedial face, the
femur is completely covered in muscle except at the palpable proximal
and distal aspects. For that reason, a small skin disruption is usually
indicative of a much larger soft tissue injury. As a result, the usual
classification system of Gustilo and Anderson for open fractures cannot
necessarily be accurately applied.¹¹¹ Although previous articles have used this system for open femoral shaft fractures,^‡ no studies exist that compare its reliability and reproducibility.

The initial management of an open femoral fracture is
similar to that of other open long bone fractures and includes
antibiotics, debridement, and stabilization. Repeated examinations of
the open wound outside of the operating room should be avoided.
Antibiotics and tetanus prophylaxis should be administered immediately.
Early operative débridement, preferably within 6 to 8 hours, should be
performed depending on the patient’s other associated injuries. In the
operating room, any planned incisions for surgical stabilization of the
femur should be drawn on the skin to allow for intelligent surgical
extensions of the open wound for definitive debridement. If
appropriate, the traumatic wounds should be extended to allow a
complete examination of the bone ends and medullary canal. A separate
surgical incision for the purposes of debridement can be used if the
traumatic wound is located in a poorly extensile location.

All
devitalized cortical fragments should be removed with the intention of
not routinely reimplanting these segments. The bone ends should be
débrided and cleaned of any foreign material while avoiding further
soft tissue devitalization. Multiple liters of saline should be used to
clean the contaminated fragments. A thorough débridement should be
performed to decrease the overall risk of early and late infection. In
circumstances in which gross contamination has occurred and the initial
débridement is questionable, a second débridement at 48 hours should be
planned.

If the initial débridement is believed to be adequate
and was performed in a timely fashion, definitive stabilization of the
femoral shaft can proceed, usually with reamed intramedullary nailing.
Historically, traction and external fixation have been used to
definitively treat open femoral shaft fractures. However, these methods
have numerous disadvantages that have precluded their routine use.
Specifically, traction requires a period of immobilization and supine
positioning that is undesirable in the polytraumatized patient. The
delay in stabilization of the femoral shaft has been shown to be
associated with increased mortality, and every effort should be made to
stabilize the limb within 24 hours.²⁵
In addition, length and alignment are poorly controlled and fracture
instability contributes to early patient discomfort. Nursing care is
difficult, because patients must remain in a relatively fixed position.
Management of other associated injuries is further jeopardized by the
need for constant traction on the extremity. External fixation is a
reasonable temporizing measure in specific circumstances but is
associated with multiple difficulties as a definitive treatment method.
In patients with multiple injuries that preclude early operative
stabilization of their open femur fractures and in patients with severe
soft tissue injuries that require a second débridement, temporary
external fixation is a reasonable and rapid initial management approach
that allows delayed conversion to a medullary implant at 5 to 10 days.
External fixation as a definitive treatment is associated with pin
tract infections, shortening, and loss of knee motion.

Immediate or delayed intramedullary nailing of open femoral shaft fractures (Fig. 50-27),
including type III open wounds, has been found to have an acceptable
complication rate in several studies and remains the treatment of
choice in most instances.³⁷,¹⁶⁴,²⁰⁰,²³⁹,²⁹⁵ Lhowe and Hansen¹⁶⁴
reported on 42 patients after immediate reamed intramedullary nailing
of open femoral diaphyseal fractures who were followed for a minimum of
12 months. Union occurred in all patients, and infection, angular
malunion, and limb length inequality were observed infrequently.
Brumback et al.³⁷ retrospectively
identified 89 open femoral fractures, the majority of which were type
III open fractures. Immediate reamed nailing was performed in 56
patients, whereas 33 patients underwent delayed stabilization at 5 to 7
days. Fractures that had a delay in the initial debridement of more
than 8 hours were typically treated with staged nailing. All fractures
healed and no infections were observed in type I, II, and IIIA open
fractures. In the type IIIB injuries, infections were observed in 11%
and occurred in injuries treated immediately and those treated delayed.
Whether early nailing resulted in a higher infection rate was further
investigated by Williams et al., who prospectively assigned 42 patients
with open femoral fractures to immediate versus delayed intramedullary
nailing. They observed a low rate of infection and nonunion, and found
no relationship with timing or the associated soft tissue injury.²⁹⁵
The initial treatment of open femoral shaft fractures with immediate,
reamed intramedullary nailing has been confirmed by multiple other
studies.²⁰⁰,²³⁹

Fractures of the femoral diaphysis caused by
low-velocity missiles are observed relatively commonly in most urban
trauma centers. They are typically the result of handgun injuries, and
four classic types of fractures have been described in the femur. These
include the “drill-hole,” incomplete, butterfly, and spiral fractures.
The former is result of a metaphyseal injury in which the bullet passes
through the bone, producing minimal comminution and maintaining the
integrity of the remaining femur. Incomplete fractures occur when the
bullet grazes the cortex of the femur with a subcritical force.
Fracture comminution occurs when the missile impacts the diaphyseal
bone, resulting in the commonly observed butterfly fragmentations.
Spiral fractures typically occur in a slightly remote location relative
to the path of the bullet and likely represent an associated torsional
force caused by weight bearing at the time of impact.²⁵⁵

The results of immediate reamed intramedullary nailing of femoral shaft fractures caused by gunshots have been uniformly good.²⁰,¹⁶³,¹⁹⁸,²⁷⁶,³⁰⁸
This has led to significant cost savings and early patient mobilization
compared with the traditional approach of traction, antibiotics, and
delayed intramedullary fixation.¹⁶³,²⁹⁹,³⁰⁸
For fractures caused by low and midvelocity gunshots, treatment with
immediate nailing without a formal debridement of the comminuted
diaphyseal fragments and bullet particles has resulted in predictable
union with acceptable complications²⁰ (Fig. 50-28). Nowotarski and Brumback¹⁹⁸
retrospectively reviewed 39 femoral fractures caused by handgun
missiles, which were treated with early (<24 hours) intramedullary
nailing at an average follow-up of more than 1 year. Complications
included one delayed union, one nonunion, and one infection, all of
which healed with a single additional operative procedure. Similarly,
Tornetta and Tiburzi²⁷⁶ reported on
the results of antegrade intramedullary nailing of distal femur
fractures caused by gunshots. Despite close proximity to the distal
interlocking screws in all cases, fractures healed predictably and no
secondary bone grafting procedures were required. Because of the good
results in these and other studies combined with the cost savings by
decreasing the length of hospitalization, immediate intramedullary
nailing of gunshot femoral fractures is recommended.²⁰,¹⁶³,¹⁹⁸,²⁷⁶,²⁹⁹,³⁰⁸

Femoral shaft fractures caused by high-velocity missiles
and shotgun blasts are an entirely different injury than those caused
by handguns. Shotgun injuries are characterized by significant
contamination, devitalization, and soft tissue injury. Occasionally if
the injury occurred at close range, the wadding may be present within
the wound and represents a foreign body. These injuries frequently
require multiple débridements to decrease the rate of infection.
Similarly, high-velocity missile injuries caused by hunting and assault
rifles produce massive tissue injuries that require an aggressive and
radical débridement of all devitalized fragments, and serial
debridements are frequently necessary.

The approach to the patient with an associated femoral
shaft fracture and an ipsilateral vascular injury requires
communication and coordination between the vascular surgeon and the
orthopaedic surgeon. The goal of treatment should be skeletal

stabilization
combined with expeditious lower-extremity revascularization. In many
cases, management should include fasciotomies of the leg to avoid
time-dependent reperfusion compartmental syndrome. Prophylactic
compartment releases distal to the site of vascular injury should be
considered if there is greater than 3 hours of ischemia. Vascular flow
to the extremity should be reestablished within 6 hours if possible to
maximize limb salvage rates. Therefore, the orthopaedic management is
largely based on the time from injury to surgery. If stabilization can
proceed expeditiously and allows enough time for vascular repair,
definitive treatment with a nail or a plate can be performed primarily.
If the time from injury to presentation is unknown or lengthy, then
limb perfusion should be reestablished first. This can be accomplished
with either temporary shunting or definitive vascular repair. A
temporary shunt provides restoration of blood flow, which can then be
followed with definitive skeletal stabilization. This allows a stable
skeleton to simplify the definitive vascular repair. In addition, if a
definitive vascular repair precedes orthopaedic stabilization, there is
always the risk of injuring the vascular repair as the limb length is
reestablished. If vascular repair with a saphenous interposition graft
is performed before skeletal stabilization, some redundancy of length
is recommended to prevent disruption during re-establishment of femoral
length.

The femur can be temporarily stabilized with an external fixator.¹²⁷
This can be useful in cases in which definitive stabilization with a
plate or nail cannot be performed because of time constraints or other
injuries of the patient. Early exchange to an intramedullary nail can
then be performed as the patient’s overall clinical condition allows.¹²⁷
In most instances, definitive internal femoral stabilization can be
performed during the same surgical procedure as the vascular repair.
The choice of a plate or a nail depends on the surgical exposure, the
fracture location, and the associated injuries. In open femoral
fractures with an associated vascular injury, the surgical approach for
medial plating is usually accomplished by a combination of the
traumatic injury and vascular approach. A plate can usually be applied
quickly. Because of the potentially increased risk of nonunion and
infection in these patients, the surgeon must consider the possibility
that a repeat procedure in this location may be necessary.
Intramedullary nailing can be performed with the same indications and
concerns as any closed or open femoral fractures as applicable.

Starr et al.²⁵⁸
retrospectively evaluated 19 patients with femoral fractures requiring
stabilization and an associated vascular injury that required repair.
Internal fixation preceded vascular repair in 10 patients, whereas in
nine patients femoral stabilization followed vascular repair. Their
favorable results led them to conclude that internal fixation could
safely precede vascular repair, and vascular shunts should be used in
cases of prolonged ischemic time. Despite skeletal stabilization and
limb revascularization, open femur fractures with associated vascular
injuries have significant complications. The long-term results of 13
blunt femoral fractures with associated arterial injuries were reviewed
by DiChristina et al.⁷² In two
patients, double level arterial injuries were identified. In the eight
open fractures, two patients had above knee amputation, three patients
had persistent drainage, and no patients regained more than 90 degrees
of knee motion. In contrast, in five patients with closed fractures,
all regained full function and full knee motion. The authors again
stressed that femoral stabilization can be performed before arterial
repair if sufficient time is available for the vascular procedure. As
with most injuries, the outcomes were dependent on the associated soft
tissue injury, specifically the presence of an open fracture.⁷²

The timing for treatment of femoral shaft fractures in
patients with associated craniocerebral trauma remains controversial.
Patients with significant head trauma are at risk for secondary brain
injury as a result of surgical management of any injury including their
femoral shaft fracture. Transient hypotension is frequently implicated
as a potential cause of further cerebral insult. This may be because of
the inability of these patients with a head injury to adequately
autoregulate their cerebral perfusion pressure. However, these patients
are frequently multiply injured and at risk for pulmonary complications
related to a delay in fixation of their femoral shaft fracture. As a
result, the management of patients with complicated injuries requires
coordination between the relevant surgical services including the
general, neurologic, and orthopaedic surgeons.

Previous reports regarding the timing of femoral shaft
stabilization in patients with an associated head injury are difficult
to interpret. The retrospective nature of all of these studies
introduces significant selection biases and limits definitive
conclusions.⁶,¹³³,¹⁷⁵,²⁵⁷,²⁸¹
However, some useful information can be obtained from these studies.
Sixty-one patients with severe or moderate closed head injuries
(Glasgow Coma Scale score <9) with femur fractures were
retrospectively reviewed by Townsend et al.²⁸¹
They found that patients treated for their femur fractures within 2
hours were eight times more likely to experience a hypotensive event
than those patients treated after 24 hours. The majority (74%) of
patients with intracranial monitoring experienced cerebral perfusion of
less than 70 mm Hg. Similarly, Jaicks et al.¹³³
found a trend toward increasing intraoperative hypotension and hypoxia
in patients treated with early femoral stabilization. Although the
neurologic complication rate was similar between patients undergoing
early versus delayed femoral stabilization, the average discharge
Glasgow Coma Scale score was significantly lower in the early fixation
group. In a study of 17 patients with an average ISS of 35, Anglen et
al.⁶ monitored intracranial pressure
during reamed intramedullary nailing. The average cerebral perfusion
pressure decreased by 17 mm Hg, and 70% of patients had an average
intraoperative cerebral perfusion pressure of less than 75 mm Hg.
However, the impact of transient hypotension on neurologic outcome is
not known.

In contrast, Starr et al.²⁵⁷
were unable to identify a statistically significant predictor or
central nervous system complications in 32 patients with combined head
injuries and femoral fractures. They did find that a delay in femur
stabilization was a strong predictor of pulmonary complications,
concluding that a delay made pulmonary complications 45 times more
likely. McKee et al.¹⁷⁵ used a
retrospective, case control study to determine the effect of a femoral
shaft fracture treated with intramedullary nailing on the neurologic
disability and mortality in patients with associated head injuries.
They identified 46 patients with closed head injuries with femur
fractures and matched these with 99 patients with a similar ISS and
Glasgow Coma Scale score yet no femur fracture. The authors found no
significant difference in early mortality, length of hospital stay,
level of neurologic disability, or the results of cognitive testing.
However, the small number of patients may limit the conclusions of this
study. In a detailed cohort study evaluating the effect of timing of
nonneurological operative interventions, Wang et al.²⁸⁸
compared early (≤24 hours) and late (>24 hours) surgical treatment
on functional outcomes, morbidity, and mortality. At 6 months following
injury, patients treated early had significantly improved
neuropsychological scores. Additionally, a higher incidence of
pneumonia and a longer hospital stay was associated with a delay in
surgical treatment. This led the authors to conclude that in patients
with combined multisystem trauma and non-operative traumatic brain
injury, early treatment of orthopaedic injuries under general
anesthesia was not associated with worse neuropsychological or
functional outcomes.²⁸⁸

The imaging, identification, and treatment of combined
femoral neck and shaft fractures have been a source of controversy
since the first descriptions of the injury pattern. Concomitant femoral
neck fractures occur in 3% to 10% of patients with femoral shaft
fractures.²,²⁹⁰,³⁰³,³¹³ Many of the associated femoral neck

fractures are nondisplaced, and missed injuries have been reported in 30% to 57% of cases (Table 50-7).

In a systematic review of 65 publications reporting on 722 cases of ipsilateral hip and femoral shaft fractures, Alho²
identified the commonly observed fracture patterns and results of
treatment. The median age of the patients in the review was 34 years,
and men were injured in 78% of cases. The hip fracture diagnosis was
delayed in 30% of cases. Hip fracture treatment was uniformly
successful with 99% healing and only 5.1% complicated by aseptic
necrosis. Locked intramedullary nailing for treatment of the femoral
shaft yielded the most predictable result when compared with unlocked
nailing and plate fixations. There were no reported cases of infection
or nonunion with locked nailing of the femoral shaft in this systematic
review.

Several predictable and commonly observed proximal
femoral fracture patterns appear to be associated with ipsilateral
femoral shaft fractures. In his systematic review, Alho reported the
locations of the associated hip fracture as subcapital in 2%,
midcervical in 21%, basicervical in 39%, pertrochanteric in 14%, and
intertrochanteric in 24%.² In a
review of 52 patients with ipsilateral proximal femoral and shaft
fractures, the fracture line inferior exit point was observed to be at
the inferomedial femoral neck and proximal to the lesser trochanter.
Only three fracture configurations were seen. The most common pattern
was a basicervical neck fracture (AO 31-B2.1) in 55% followed by a
vertical midcervical intracapsular shear pattern (AO 31-B2.3) in 35%
and an intertrochanteric fracture through the greater trochanter (AO
31-A1.2) in 10%. This indicates that the majority of these proximal
femoral fractures are extracapsular in location. All of the observed
intracapsular femoral neck fractures were vertically oriented, making
fixation particularly difficult in displaced patterns. Of the fractures
that were initially missed, none were intertochanteric in location.²⁵²

Most articles suggest that the rates of femoral neck
nonunion and aseptic necrosis are lower in patients with combined
femoral neck and shaft fractures than in young patients with
high-energy femoral neck fractures in isolation.¹,²,¹⁹
This may be because of the commonly observed extracapsular location of
the neck fracture, the tendency for these fractures to be minimally or
nondisplaced, or a combination of the two. The delay in diagnosis of
the femoral neck fracture and treatment of the femoral neck after
stabilization of the femoral shaft fracture (i.e., in cases of
iatrogenic or occult femoral neck fractures) have not necessarily
produced complications.²,¹⁹

The delay in diagnosis of the femoral neck or
intertrochanteric fracture is the result of several concurrent factors.
Because of the presence of an ipsilateral fracture of the femoral
shaft, AP imaging of the injured extremity before stabilization of the
femoral shaft fracture is typically hampered by external rotation of
the proximal femoral segment.⁶⁹
Because these patients are frequently in severe discomfort as a result
of their shaft fracture, suboptimal imaging of the hip may
unfortunately be accepted. Even with careful positioning of the lower
extremity for quality images of the hip as a part of the routine
radiographic evaluation of a femoral shaft fracture, proximal segment
control is inadequate because of the shaft fracture. It has been noted
serendipitously that in patients with known or suspected ipsilateral
acetabular fractures, the obturator oblique radiograph frequently
demonstrates the femoral neck along its longest axis, assisting with
the radiologic evaluation.

Because of the high incidence of missed femoral neck
fractures in patients with femoral shaft fractures, review of all
available imaging studies at multiple time points in the patient’s
evaluation and treatment is recommended (Table 50-7).
First, dedicated hip radiographs should be obtained as a part of the
initial radiologic evaluation of any patient with a femoral neck
fracture. Second, if pelvic oblique radiographs are obtained because of
a suspected or known ipsilateral acetabular fracture, these should be
scrutinized for the presence of a femoral neck fracture. Third, if a CT
scan is obtained for evaluation of abdominal or pelvic trauma, this
should be reviewed because occult fractures are frequently demonstrated
on the relevant axial images (Fig. 50-29).
Fourth, intraoperative fluoroscopic images before initiating antegrade
nailing of a femoral shaft fracture should be obtained. Fifth, hip
fluoroscopic images and/or radiographs should be obtained after femoral
shaft stabilization with the hip in 10 to 15 degrees of internal
rotation. Finally, dedicated postoperative hip radiographs should be
obtained before leaving the operating room to confirm the integrity of
the femoral neck. By using a best-practice protocol consisting of a
dedicated internal rotation plain radiograph of the hip, a 2-mm CT scan
through the femoral neck, a fluoroscopic lateral of the femoral neck
before fixation, and postoperative orthogonal hip radiographs in the
operative room, Tornetta et al. demonstrated a substantial improvement
in the rate of missed femoral neck

fractures
in patients with femoral shaft fractures. The delay in femoral neck
fracture diagnosis was reduced by 91% by using these consistent
radiographic evaluations.²⁷³

Yang et al.³¹³
reviewed the impact of abdominal CT scan evaluations on the ability to
demonstrate nondisplaced fractures of the femoral neck. In a series of
152 femoral shaft nailings, 14 femoral neck fractures were ultimately
identified. Of these, eight were not visible on the initial screening
pelvic radiographs. On retrospective review of the abdominal CT scans,
six of these eight femoral neck fractures were shown to be present but
nondisplaced. The remaining two femoral neck fractures were not visible
on preoperative CT scans and were thus believed to be iatrogenic
according to the authors.

The delay in identification of the femoral neck fracture
may be accentuated if the fracture is truly nondisplaced. However,
there is always the possibility that the fracture is iatrogenically
produced during the manipulation, reduction, and placement of an
antegrade intramedullary nail. It has been suggested that iatrogenic
fractures do occur as evidenced by normal preoperative CT scans in
patients later found to have a nondisplaced fracture of the femoral
neck after intramedullary nailing.³¹³ This was investigated by Simonian et al.,²⁵³
who compared the proximal femoral anatomy and nail insertion depths in
four patients with iatrogenic femoral neck fractures with 311 patients
without this complication. They found a higher neck shaft angle of 139
degrees in the patients with femoral neck fractures than in those
without (125 degrees). Further, in the four patients with femoral neck
fractures, the nails were inserted below the tip of the greater
trochanter. This was not observed in patients with similarly elevated
neck shaft angles but without femoral neck fractures. This led the
authors to recommend avoiding deep insertion of the femoral nail beyond
the tip of the greater trochanter in patients with a valgus femoral
neck. Although these findings may be specific to the nail insertional
jig used, an awareness of this potential complication is necessary.
However, the majority of femoral neck fractures that are diagnosed
after femoral nailing are likely missed injuries. This is emphasized in
a report of “clandestine” femoral neck fractures in a group of patients
treated with plate fixations for their femoral shaft fractures. Riemer
et al. reported delayed identification of a femoral neck fracture in
31% of patients treated with plating of their femoral shaft fracture, a
procedure unlikely to produce significant associated proximal femoral
injury.²³⁰

The treatment of this combination of injuries continues
to be controversial. Experience has led many authors to suggest that
these two noncontiguous fractures should each be treated with an
implant that optimizes fracture healing while simultaneously
prioritizing the femoral neck fracture fixation (Fig. 50-30).
This is in large part because of the increased difficulty observed with
managing complications associated with inadequate or suboptimal femoral
neck fixation such as aseptic necrosis, nonunion, malunion, and failure
of fixation. The major complications associated with failure of the
available methods for treating the femoral shaft component of this
injury combination are generally familiar and more straightforward.
This has led to the recommendation that the femoral neck or
intertrochanteric fracture should be treated with multiple lag screws
or a sliding hip screw, with or without capsulotomy, as appropriate.
The femoral shaft component can be treated with whichever method is
most familiar and reliable, typically a retrograde femoral nail or
lateral plate fixation.²⁶³,²⁶⁴,²⁹⁰
Femoral plating combined with multiple screws or a sliding hip screw
for the femoral neck fracture has yielded good results in several
series.⁵⁵,⁹⁴

A number of technical aspects and decisions require
emphasis. The type of fixation for the femoral neck fracture should not
differ from isolated injuries in this location. Typically, multiple lag
screws are used to stabilize subcapital and transcervical fractures,
whereas a sliding hip screw is used for basicervical and
intertrochanteric fractures. The order of fixation is influenced by the
two treatment methods chosen combined with the urgency of early
reduction and stabilization of an intracapsular femoral neck fracture.
If the proximal femoral fracture is intracapsular in location and
displaced, consideration should be given to completing this portion of
the procedure first if the patient’s overall medical condition allows.
This is largely to allow for urgent preservation of the femoral head
blood supply. In cases in which a retrograde nail is desired for
stabilization of the femoral shaft fracture, the canal should be
overreamed to allow for atraumatic passage of the nail to avoid any
additional stress on the femoral neck. If a basicervical femoral neck
fracture is stabilized with a sliding hip screw, unicortical screws can
be initially placed in the side plate and exchanged for longer screws
that avoid the retrograde nail. This will allow for placement of a
retrograde nail as proximally as possible.

In the event that an ipsilateral nondisplaced femoral
neck fracture is identified after antegrade femoral nailing, strategic
lag screw fixations anterior to the nail can be successfully performed.
An additional hip joint capsulotomy can be performed to relieve any
intracapsular tamponade if the fracture location warrants. In cases in
which a displaced femoral neck fracture is identified after placement
of an antegrade nail, the ultimate treatment depends largely on the
fracture location. If the fracture communicates with the nail entry
site, it is unlikely that an anatomical reduction of the femoral neck
can be obtained with the nail still in position. Nail removal,
anatomical fixation of the femoral neck, and alternative fixation of
the femoral shaft should be performed.

Alternatively, a single implant such as a
cephalomedullary (reconstruction) nail has been used to treat combined
femoral neck and shaft fractures. This has been associated with good
results in several small series and has been recommended as an option.
However, in a retrospective review of ipsilateral neck and shaft
fractures treated with multiple different methods, Watson and Moed²⁹⁰
found that 75% of the femoral neck nonunions that occurred at their
institution developed after the use of a reconstruction nail. Jain et
al.¹³⁴ reported on 23 cases of
ipsilateral hip and shaft fractures treated with a single implant over
a 4-year period. Complications of the femoral neck fixation included
one nonunion, one varus malunion, and one aseptic necrosis, for an
overall rate of 13%. There are a number of technical and conceptual
difficulties with the use of a reconstruction nail for the treatment of
combined neck and shaft fractures. First, a medullary implant is
unlikely to be the treatment of choice for fractures of the femoral
neck. Second, the entry site for the reconstruction nail is frequently
in the same location as the femoral neck fracture. Even with an open
reduction of the femoral neck, there is the potential for further
displacement during nail placement. Third, it can be technically
challenging to seat the nail properly with the correct anteversion to
allow placement of two screws accurately into the femoral head and
parallel to the femoral neck. By avoiding a nail entry site that is
through or near the femoral neck fracture, a trochanteric entry
cephalomedullary implant to treat this combination of fractures may
alleviate some of these technical challenges.

In addition to fractures of the femoral neck, other
associated ipsilateral fracture patterns can further complicate the
management of femoral shaft fractures (Fig. 50-31).
Segmental comminution, characterized by circumferential loss of
cortical contact even after reduction, can be successfully managed with
closed intramedullary nailing.³⁰⁰ In a series of 33 segmental femur fractures treated with reamed antegrade nailing, Wiss et al.³⁰⁰
identified infrequent complications that included one nonunion and two
malunions. This led the authors to conclude that virtually all
fractures between the lesser trochanter and the femoral condyles can be
successfully managed with intramedullary nailing. Technical
difficulties with nailing segmentally comminuted fractures include the
intraoperative determination of the proper length and rotation of the
femur, as the visual clues at the fracture itself are less helpful.
Preoperative assessment of the contralateral femoral length can be used
as a guide to judge the length of the injured extremity. Segmental
femoral injuries characterized by multiple noncontiguous fractures
should be differentiated from fractures with segmental comminution and
deserve special mention. These fractures are particularly difficult to
treat with intramedullary nailing because of the need to simultaneously
maintain two reductions. Percutaneous Schanz pins placed as joysticks
into each segment can greatly assist with maintenance of reduction
during reaming and nail placement. For fractures with a “napkin ring”
segment of diaphyseal bone, the free segment should be stabilized with
either a percutaneous pin or a percutaneous clamp during reaming to
avoid spinning the fragment on its soft tissue attachments.

Distal femoral articular fractures can occur in
combination with femoral shaft fractures, posing additional challenges.
The articular fracture should be prioritized from an implant
standpoint, but not necessarily from a timing standpoint. The
stabilization of femoral shaft fractures in a critically ill patient
still takes priority over the operative fixation of periarticular
fractures. However, a thoughtful preoperative plan can ensure that an
accurate articular reduction of the distal femur can be accomplished as
well. If the patient can tolerate a longer procedure, the articular
injury should be reduced and stabilized first with whatever implants
are necessary, because there is more flexibility with stabilization of
the femoral shaft fracture. Screw fixations alone are frequently
adequate for coronal plane articular fractures of the distal femur and
will not interfere with passage of an antegrade nail. For medial and
lateral femoral condyle fractures, a combination of strategically
placed lag screws and plates will usually not interfere with nail
placement. For a supracondylar intercondylar distal femoral fracture
combined with a noncontiguous femoral shaft fracture, the articular
surface of the intercondylar component must first be reduced and
stabilized followed by stabilization of the remaining supracondylar and
shaft fractures. Depending on the locations of the distal and shaft
fractures, single (plate or nail) or multiple (plate and plate or plate
and nail) implants may be necessary. In general, combined plate and
screw fixations of the distal articular fracture should be performed.
Reasonable results have also been obtained by using a combination of
lag screws to reduce the distal articular injury with a nail to
stabilize both the femoral shaft and the distal supracondylar
components. However, in their review of 10 patients with intercondylar
distal femoral fractures combined with a femoral shaft fracture, Butler
et al. reported some articular displacement in 40% of cases in which
screws alone were used to stabilize the intercondylar component.⁴⁴

Combined noncontiguous fractures of the ipsilateral femoral

neck, femoral shaft, and distal femur have been reported. This
combination of injuries occurs in less than 0.5% of femoral fractures.
All femoral neck fractures in one series were vertical in orientation
and all distal femoral articular injuries were unicondylar. A
recommended treatment algorithm was presented by Barei et al.¹¹
and was based on their experience with seven patients. Open reduction
and internal fixation of the displaced femoral neck fracture should be
prioritized followed by fixation of the displaced distal articular
fracture. Finally, the diaphyseal fracture can be treated with a
technique that does not jeopardize either of the other two fixations.
Specifically, this can be with either a plate or an intramedullary nail
(antegrade or retrograde).

Combined injuries to the ipsilateral extensor mechanism,
hip joint, and femoral shaft, although rare, do occur. These are
frequently caused by a dashboard injury with direct impact on the knee
with the hip joint flexed. The extensor mechanism disruption may be a
patellar fracture, an infrapatellar tendon injury or a quadriceps
tendon rupture. A treatment protocol based on their experience with
managing 10 patients with this combination of injuries was reported by
Mosheiff et al.¹⁹³ They recommended
intramedullary nailing of the femur, followed by open reduction and
internal fixation of the acetabular fracture, followed by extensor
mechanism repair. Frequently, multiple stages were required, but this
protocol simplified each surgical episode and maximized patient
recovery.

Patients with bilateral femoral shaft fractures are
known to have a worse overall prognosis and a higher mortality than
patients with unilateral femur fractures.⁶⁵,¹⁹⁷
Several factors may contribute to this including increased associated
injuries, increased blood loss, a higher risk of adult respiratory
distress syndrome, or the increased systemic effects of multiple long
bone fractures. In addition, the presence of bilateral femur fractures
is associated with other complications including compartmental
syndrome, peroneal nerve palsy, and delayed union. The incidence of
bilateral femoral shaft fractures in patients treated at tertiary
referral trauma centers ranges from 7.3% to 9.6%. These patients tend
to have a higher ISS and a lower Glasgow Coma Scale score than patients
with unilateral femoral shaft fractures.⁶⁵,¹⁹⁷

In a review of 885 consecutive patients with blunt trauma who were admitted with femoral fractures, Copeland et al.⁶⁵
reported increased ISS, mortality, and ARDS in 85 patients with
bilateral injuries. The mortality rate in patients with bilateral femur
fractures was 25.9% compared with 11.7% in patients with unilateral
injuries. Similarly, ARDS was diagnosed in 15.7% of patients with
bilateral fractures compared with 7.3% of patients with unilateral
fractures. The increase in mortality was found to be related to the
associated injuries and various physiologic parameters as opposed to
the presence of bilateral fractures.⁶⁵
The increased mortality in patients with bilateral fractures was
confirmed in another study reviewing 689 patients with unilateral
injuries and 54 patients with bilateral injuries treated with reamed
intramedullary nailing for their femoral shaft fractures.¹⁹⁷
Patients with bilateral femur fractures were found to have a higher
average ISS and a longer length of stay than patients with unilateral
fractures. Mortality in patients with bilateral femur fractures was
5.6% compared with 1.5% in patients with unilateral femur fractures.
The relative risk of death in patients with bilateral fractures was 3.8
times higher than in patients with unilateral fractures. Multiple
linear regression and logistic regression with age and ISS as
covariants were used to compare patients with a similar degree of
trauma. This revealed an odds ratio of 3.3 for mortality in patients
with bilateral femur fractures compared with patients with unilateral
femur fractures. Because the ISS does not increase with the presence of
a second femoral shaft fracture, the ISS may not accurately reflect
mortality in patients with bilateral femur fractures.¹⁹⁷

Definitive treatment of bilateral femoral shaft
fractures is the same as that for unilateral femoral shaft fractures.
Reamed intramedullary nailing, whether antegrade or retrograde, can be
expected to produce similar rates of union compared with treatment in
unilateral injuries. Retrograde nails have been used successfully and
mortality appears to be similar to other studies using predominanly
antegrade nails.⁴⁹ The use of
temporary external fixation consistent with damage control orthopaedics
may be considered in these patients depending on their associated
injuries.

Obesity is a rapidly expanding health concern and is not
likely to decrease in the near future. There are several technical
aspects of antegrade intramedullary nailing that make the procedure
more difficult in obese patients. Osseous landmarks are difficult to
palpate, and femoral adduction is usually somewhat limited. In
addition, the common locations of excess adipose tissue usually limit
the surgeon’s ability to align a guide pin or an awl with the femoral
canal when attempting to localize the correct starting point. This
frequently leads to the need for a large surgical approach to simply
identify the osseous anatomy proximally. Although lateral positioning
greatly assists with identifying the starting point, this may not be a
practical position depending on the size of the patient. Reduction of
the femoral shaft is increasingly difficult with increasing patient
size, and the usual techniques may be impractical. As well, the
external jig that is used for proximal interlocking may not clear the
lateral thigh tissues. Because of the difficulties associated with
identifying the piriformis antegrade starting point, alternative nail
entry sites have been recommended in obese patients.²²⁸,²⁸² Ricci et al.²²⁸
demonstrated decreased surgical time and radiation exposure yet similar
functional outcomes in obese patients treated with a trochanteric entry
nail compared to a piriformis entry nail. Similar results have been
demonstrated using retrograde nailing as an alternative in obese
patients.²⁸² The authors found
similar decreases in operative and fluoroscopy time when a retrograde
nail was used compared to an antegrade nail placement. Although the
retrograde starting point may be easier to identify in obese patients,
proximal interlocking can be a lengthy experience given the distance
between the anterior skin of the thigh and the anterior femoral cortex.
Long drill bits, long screwdrivers, and a suture tied to the
interlocking screw may assist with proximal interlocking of a
retrograde nail.

The complications and surgical difficulties in seven
morbidly obese patients treated with reamed antegrade nailings were
reported by McKee and Waddell.¹⁷⁶
The patients had an average weight of more than 300 pounds with an
average body mass index of 49.2. The surgical procedures averaged 3.8
hours in duration with an average blood loss of 1100 mL. Numerous
complications were noted including two iatrogenic trochanteric
fractures, two wound infections, four deep venous thromboses, and one
fatal pulmonary embolism.

Femoral fractures in the elderly occur from both low and
high-energy mechanisms. More frequently, they are the result of falls
from standing. However, motor vehicle accidents and pedestrians hit by
cars occur in this patient demographic as well. Particular challenges
in this group of patients include fixation in relatively osteopenic
bone and management of the multiple associated medical problems
frequently observed.

In elderly patients, cortical thinning and decreased
bone mineral density occur and are more commonly observed in females.
As a result, fixation becomes particularly difficult. In addition,
preexisting arthritic conditions frequently limit the excursion of the
knee and hip joints, potentially prohibiting specific fixation options.
Of particular importance is the identification and treatment of
preexisting medical diseases that occur frequently in this patient
population. In a cohort of 102 elderly (average age of 81 years)
patients who sustained a femur fracture, Sartoretti et al.²⁴⁴
identified commonly observed comorbid conditions that included
cardiovascular disease in 80%, musculoskeletal disease in 75%,
gastrointestinal disease in 67%, psychiatric disease in 61%, urologic
disease in 55%, and pulmonary disease in 41%. The postoperative
mortality was 11%, and the mean hospitalization was 30 days.

Intramedullary nailing of femoral shaft fractures in
elderly patients is uniformly associated with high union rates yet a
high incidence of complications including mortality. In a series of 138
patients aged more than 65 years with femoral shaft fractures,
complications were observed in 46% and mortality in 20%.²⁸ Other authors have reported mortality rates that are similar to those in elderly patients with hip fractures.³,³⁰,¹⁸⁹ The mental status of the patient has been found to be a significant predictor of survival,²⁸ and the development of a new medical problem after injury has been associated with a poor result.³⁰
In surviving patients, fracture union can be expected after
intramedullary nailing for fractures of the femoral shaft. In a review
of 12 femoral fractures treated with statically locked intramedullary
nails, union was observed in all patients.

Several technical aspects of intramedullary nailing
specific to elderly patients must be observed. Because of the
relatively thin cortices, care must be taken to accurately identify the
starting portal. Iatrogenic comminution can occur with manipulation of
the femoral shaft despite relatively low applied forces. Interlocking
screws frequently have poor purchase. If a mismatch between the
sagittal bow of the femur and the nail, an iatrogenic distal femoral
fracture may occur if a long antegrade nail is chosen.²⁰³
A careful preoperative evaluation of the shape of the femur on both the
AP and lateral views is necessary before deciding on a medullary
implant in elderly patients. Because of the capacious canal, a
large-diameter implant may be necessary.

In response to these issues with obtaining stability in
osteopenic bone, many manufacturers have made specific design changes
in medullary implants. Some interlocking screws are designed with a
large washer on the lateral side and a large nut with washer that can
be threaded on the leading (medial) side of the screw. Some retrograde
nails allow placement of a locking bolt that prevents coronal plane
toggling of the most distal interlocking screw. Multiplanar distal
metaphyseal locking screws and spiral blades are available with some
nails and may assist with purchase in the capacious metaphysis.
Knowledge of the specific implant characteristics is important when
deciding the treatment plan in these challenging fractures.

The treatment of pathologic fractures and impending fractures caused by skeletal metastases is discussed in Chapter 20.
Because the proximal femur is the most commonly affected site of
metastases in the appendicular skeleton, the relevant management will
be discussed. The diagnosis of metastases should not be assumed despite
clinical suggestions and a history of carcinoma in another location.
Discussion with an orthopaedic oncologist will help to avoid the
complications associated with treatment of an unknown femoral lesion.¹³²

The goal of treatment of metastatic femoral lesions is
to provide femoral stability, thereby decreasing pain, allowing
mobilization, and facilitating nursing care. Fracture healing depends
on the type of tumor. Osseous healing may not occur in many of these
injuries despite stabilization. In 129 pathologic fractures of long
bones, Gainor and Buchert⁸⁹ observed
healing in 67% of fractures from multiple myeloma, 44% from renal
carcinoma, 37% from breast carcinoma, and in no fractures secondary to
lung carcinoma. Overall, the fracture healing rate was 35%, but this
increased to 74% in patients who survived longer than 6 months.
Although a significant percentage of these fractures were treated
without internal fixation, this emphasizes the difficulty with
obtaining union in these patients.

Prophylactic treatment, usually with an intramedullary
nail, for metastatic lesions of the femoral shaft is based loosely on
clinical symptoms, expected length of survival, and lesion size.
Guidelines for the size of a metastatic lesion that requires
prophylactic internal fixation has been suggested by Menck et al.¹⁷⁸
in a radiographic study of 69 patients. They recommended stabilization
if the ratio between the width of the metastasis and the bone diameter
was greater than 0.6, if the axial cortical destruction in the femoral
shaft was greater than 30 mm, or if the cortical destruction of the
femoral circumference was greater than or equal to 50%.

Stabilization can be best accomplished with plates or intramedullary implants.¹³²
If plate fixation is chosen, cement should be used to augment the
fixation in most instances. Locked implants can be considered given the
frequently present poor bone quality. Intramedullary nailing is optimal
in most femoral fractures resulting from metastatic disease because of
the biomechanical properties of the load-sharing implant. To span the
entire femur including the femoral neck, a cephalomedullary nail should
be used in pathologic fractures. Patients can safely bear weight after
stabilization. However, intramedullary nailing for metastatic disease
is associated with increased perioperative complications including fat
embolism and mortality. Barwood et al.¹⁴
observed intraoperative oxygen desaturation in 24% of patients with
death in 7% of patients treated with intramedullary nail fixation for
their overt or impending femoral fractures. Cole et al.⁶³
reported their experience with reamed and unreamed intramedullary
nailing in 73 femurs with metastatic disease. Both techniques were
found to sufficiently stabilize the femur, and implant failures were
uncommon. Death associated with the nailing procedure occurred in one
patient treated with a reamed nail and in one patient treated with an
unreamed nail. In general, second-generation reconstruction nails or
other cephalomedullary implants should be used for the treatment of
metastatic femoral fractures. Although a standard femoral nail is
adequate for lesions confined to the femoral diaphysis, minimization of
the need for future stabilization procedure can be

accomplished with the use of an implant that effectively protects the entire femur.

Recently, a new form of insufficiency fracture in the
proximal femoral shaft or subtrochanteric region has been described.
This fracture has been associated with the use of alendronate and is
thought to be associated with the prolonged suppression of bone
remodeling with this medication.¹⁰²,¹⁵⁹,¹⁶²,¹⁹⁶
Alendronate is a bisphosphonate that inhibits bone resorption by
osteoclastic suppression. Decreased bone turnover and impaired repair
of microscopic damage has been implicated in the propogation of these
subtrochanteric or proximal femoral shaft fractures. Several
characteristics were common to these fractures based on a review of 17
patients with low energy subtrochanteric and proximal femoral fractures
by Kwek et al.¹⁵⁸ and included:
cortical thickening on the lateral side of the subtrochanteric region,
a transverse fracture, and a medial cortical spike (Fig. 50-32). The majority of patients (76%) had prodromal pain and bilateral involvement (53%).¹⁵⁸
Given the potential for bilateral involvement, it is generally
recommended that contralateral femur radiographs be obtained to look
for early signs prior to fracture.

Neviaser et al.¹⁹⁶
additionally demonstrated this relationship between chronic alendronate
use, the presence of a simple transverse fracture, and unicortical
beaking in the area of cortical hypertrophy. Although the exact causal
relationship between alendronate therapy and these atypical fractures
has yet to be

confirmed,
recognition of this pattern is important especially given the prodromal
symptoms and consistency of bilateral involvement.¹⁰²,¹⁵⁹,¹⁶²,¹⁹⁶