DIAPHYSEAL FRACTURES OF THE FEMUR

Ovid: Chapman’s Orthopaedic Surgery

Editors: Chapman, Michael W.
Title: Chapman’s Orthopaedic Surgery, 3rd Edition
> Table of Contents > SECTION
II – FRACTURES, DISLOCATIONS, NONUNIONS, AND MALUNIONS > Pelvis and
Femur > CHAPTER 20 – DIAPHYSEAL FRACTURES OF THE FEMUR

CHAPTER 20
DIAPHYSEAL FRACTURES OF THE FEMUR
Michael W. Chapman
M. W. Chapman: Department of Orthopaedics, University of California Davis, Sacramento, California, 95817.
This chapter is
dedicated to the memory of Richard C. Maurer, M.D., who wrote the
section on periprosthetic fractures in the second edition.

P.672


A. FEMORAL-SHAFT FRACTURES
There are major physiologic reasons to avoid prolonged
bed rest in fracture treatment. Immobilization is implicated in
complications ranging from constipation and atelectasis to osteoporosis
and pulmonary embolism, the last of which can cause death. For patients
with multiple injuries, the complications of immobilization can lead to
significantly increased morbidity and mortality (see Chapter 14).
Fractures of the femoral shaft are of special significance because of
the femur’s size and proximal position in the lower extremity,
necessitating long periods of bed rest with nonoperative treatment.
This problem has stimulated two different approaches to treatment: the
nonoperative approach of cast bracing, and operative treatment using
intramedullary nails or plates.
Early techniques of open, retrograde (meaning insertion
initially into the proximal fragment in a retrograde direction, as
opposed to retrograde insertion through the knee, which is addressed
later in this chapter) placement of medullary nails for stable,
mid-shaft fractures had an unacceptable rate of infection. With the
introduction of closed medullary nailing, infection became a rare
problem, and early mobilization was possible with stable fracture
patterns (18). With open reduction, adjunctive
cerclage wiring was used in unstable and comminuted cases. Because the
nail extended the full length of the femur, there was no stress riser
effect. Devascularization of a significant portion of the inner wall of
the femur occurs as a result of reaming, but this is rarely clinically
significant (see Chapter 11). Because in most
fractures the nail functions as a load-sharing device, rather than
bearing all of the load, axial loading across fractures with stable
patterns is encouraged, thus promoting callus formation.
Medullary nailing is the fixation of choice for
virtually all diaphyseal femur fractures. Reaming and nailing can be
performed immediately after injury for most open femur fractures. It
was once feared that this practice would lead to complications such as
infection, fat embolism, nonunion, or, in the case of the patient with
multiple injuries, adult respiratory distress syndrome (ARDS) and
multiple organ failure. It is now recognized that infection is not a
major problem and union rates are actually improved; fat embolism,
ARDS, and other problems are all substantially decreased or prevented
by femur stabilization soon after injury.
The current designs of interlocking nails permit
stabilization of proximal diaphyseal as well as subtrochanteric and
associated femoral neck fractures and very distal supracondylar
fractures. As experience with interlocking femoral nails has grown, it
has become apparent that interlocking of femoral fractures using proper
technique does not delay union and that the early removal of locking
screws (dynamization) to facilitate fracture union is not typically
required (7,8,11,12,39).
Further impetus for the use of interlocking nails has come with the
finding that a number of femur fractures are more comminuted than
initially recognized, necessitating the added fixation of locking
screws.
CLASSIFICATION
Two types of classification are necessary for fractures
of the femur: a grading system for soft-tissue injuries and a system
for classifying the fracture pattern (33,57,73).
GRADING OF SOFT-TISSUE INJURIES
The grading of soft-tissue injuries is more an art than a science. The system developed by Gustilo and Anderson (33) for open fractures and that of Tscherne for closed injuries, which has been modified by the AO group (55,67,73), are the most widely used. See Chapter 12 for details.
GRADING OF FRACTURE PATTERNS
The purpose of grading the fracture pattern is to assess
the intrinsic stability of the fracture and guide the selection of
treatment. In most diaphyseal femur fractures, this translates into a
decision about whether to lock the nail proximally, distally, or at
both locations. Pattern, location, and degree of comminution contribute
to overall fracture stability. For example, short oblique fracture
patterns are usually stable with standard (i.e., unlocked) nailing when
they occur near the isthmus and are comminuted not more than
Winquist-Hansen type II (Fig. 20.1) (18).
At more proximal or more distal locations, or if more comminuted, these
fractures are not stable under axial loading. On the other hand, even
in apparently stable patterns such as the short oblique fracture, there
can be a split extending proximally or distally from the fracture site
that cannot be seen on preoperative or even postoperative radiographs.
With weight bearing, this can shift, resulting in malposition. As a
result, I now statically lock all fractures of the femoral shaft
proximally and distally, impacting the fracture site to maximize bone
apposition and thereby eliminating any gap, when nailing
Winquist-Hansen type I and type II fractures. For the purposes of this
chapter, the Winquist-Hansen classification is simple, practical, and
useful for clinical decision making, particularly if it is applied to
the proximal, middle, and distal thirds of the diaphysis. The
Orthopaedic Trauma Association

P.673


has
adopted and expanded the classification of the AO Group. Their system
is much too detailed and complex for presentation here but is very
useful for exploring the nuances of fracture management and for
research. Interested readers are referred to the Special Edition of the
Journal of Orthopaedic Trauma devoted to this classification system (57).

Figure 20.1. The Winquist-Hansen classification of comminution in fractures of the femoral shaft.
BIOMECHANICS
The biomechanics and physiology of intramedullary nailing in the subtrochanteric area and the hip are addressed in detail in Chapter 11 and Chapter 19.
A few comments are required here, however. Measures of torsional
rigidity, bending stiffness, and resistance to compression (shortening)
have shown conclusively that first-generation interlocked nails using
either an oblique proximal cross-locking screw and transverse distal
cross-locking screws, or transverse cross-locking screws both
proximally and distally, are mechanically superior to unlocked nails
and to early designs such as the Brooker-Wills with internally deployed
distal fins (5,15,31,35,42).
Some interlocking nails are provided with cross-locking holes in both
the anteroposterior (AP) and mediolateral planes and/or have a slotted
hole available so that the fracture will compress with weight bearing.
In recent systems developed for insertion without
reaming, the maximum diameter is smaller, and therefore the
cross-locking screws have had to be of smaller diameter as well. In
spite of modifications in both design (solid nails) and materials
(titanium alloys), these smaller nails have an increased breakage rate.
In resisting torsional forces, one screw proximally and distally
appears to be just as effective as two screws (35).
For fractures in the distal third of the femur, where there is no
effective contact between the inner wall of the cortex and the nail in
the distal fragment, two transverse cross-locking screws are necessary
to provide adequate stability (15). Stability
in supracondylar fractures is enhanced with larger-diameter nails,
insertion to subchondral bone at the knee, and use of two cross-locking
screws (15,31,34).
In mid-diaphyseal fractures with stable patterns, Winquist-Hansen I and
II, one cross-locking screw proximally and distally is adequate. In
Winquist-Hansen III and IV patterns, use the maximum number of cross
screws available in the nail design being used (34).
TREATMENT OPTIONS
NONOPERATIVE TREATMENT
Nonoperative treatment is standard in children and young
adolescents. In mid to late adolescents and in some younger children
who have associated multiple injuries, particularly head injuries,
internal fixation with devices such as the Ender nail is necessary to
avoid injury to the growth plates (see Chapter 19 and Chapter 164).
In adults, the indications for treatment of femur fractures in balanced
suspension and skeletal traction and/or functional cast braces have
nearly disappeared due to the tremendous advantages of intramedullary
nailing, particularly since the sophistication of current techniques
has reduced the incidence of complications to very acceptable levels.
In addition to the systemic risks of nonoperative care already
mentioned, the prolonged bed rest and immobilization (3 to 6 months) is
not tolerable to most patients today, and it is accompanied by a
significant incidence of malunion, nonunion, and permanent joint
stiffness, particularly in the knee.
In regions of the world where sophisticated
intramedullary techniques are not available or where the risks of
surgery are unacceptable, nonoperative management remains the mainstay;
the techniques for it are discussed in detail in Chapter 10.
In recent years, I have not used balanced suspension and skeletal
traction as the sole method for treating a fracture of the femur in
adults. I use functional cast braces combined with intermittent
traction for supracondylar fractures of the femur in situations where
surgery is contraindicated, such as in the presence of serious systemic
illness in the elderly, coexisting infection in the region of the
fracture, or comminution and severe osteoporosis. Using a well-molded
cast-brace with intermittent tibial pin traction, it is possible to
mobilize these patients to a chair at bedside within the first week
after fracture and to obtain a reasonably good outcome in most cases (49,51).

P.674


EXTERNAL FIXATION
External fixation of closed fractures and of most open
fractures of the shaft of the femur is rarely indicated today but still
has a specific role in the armamentarium of the trauma surgeon (1,26,27,36). Indications for external fixation include the following:
  • Patients with multiple injuries who are
    hemodynamically unstable or who have a particular combination of
    injuries in which rapid stabilization of the femur with minimal
    soft-tissue trauma is essential
  • Some femoral-shaft fractures in children and adolescents (see Chapter 164)
  • High-grade or severely traumatized open
    fractures where external fixation may be a temporary or definitive
    method of stabilization
  • Some femur fractures with an associated
    vascular injury that requires repair, where internal fixation is either
    difficult or contraindicated
  • The rare fracture in which systemic or local factors contraindicate internal fixation
OPEN INTRAMEDULLARY NAILING
The standard of care for the vast majority of fractures
of the diaphysis of the femur today is locked intramedullary nailing
utilizing closed percutaneous techniques. Direct open reduction of the
fracture may be indicated in the following cases:
  • The fracture is not reducible by closed
    technique, or a bone fragment is incarcerated in the canal, which
    precludes passage of a guide pin or the nail.
  • The patient is a victim of
    multiple-system injuries and is on a regular operating table where
    convenience and speed are best facilitated by a modified open nailing
    (see Chapter 11, Chapter 12, and Chapter 14).
  • The fracture is open, and irrigation and debridement of it exposes the fracture site.
  • A preexisting deformity or implant is
    present in the femur, necessitating open osteotomy and/or hardware
    removal to make nailing possible.
When opening the fracture, it is usually possible to
make an incision directly over it and then to dissect through the
quadriceps where the femur has torn it, rather than make a separate
independent approach. The incision need be only large enough to admit
the fingers to facilitate passage of the guide pin across the fracture
site. Once this has been achieved, and if good fluoroscopic
visualization is available, it is usually possible to close the wound
and proceed with nailing as one would for closed technique. This has
the advantage of minimizing the amount of time the wound is open, and
it allows the medullary contents and reamed bone to accumulate around
the fracture site following reaming, which enhances union.
STANDARD NONLOCKED OR FLEXIBLE NAILS
Standard reamed or nonreamed nails, such as the original Küntscher nail (76),
are rarely used today in most developed countries, but because of the
simplicity and low cost of the procedure, it can be performed open with
either antegrade or retrograde technique in areas of the world where
more sophisticated locking nail techniques under fluoroscopic control
are not available. A typical case is illustrated in Figure 20.2. Flexible nonreamed nails such as the Ender nail (see Chapter 19)
can be inserted both antegrade and retrograde, and either singly or
multiply, and are used today most commonly in femur fractures in
children or early adolescents who require internal fixation. The Ender
nail did not provide enough control of malrotation and shortening to be
used routinely in fractures of the diaphysis in adults. Its simplicity
and low cost, particularly when combined with temporary skeletal
traction and cast bracing, make it a very useful technique in less
developed countries.
Figure 20.2. A: The classic indication for medullary nailing of a short, oblique, mid-isthmus fracture. B: A standard medullary nail is placed. C:
After removal 16 months later, there is a smooth, well-remodeled
callus, and the general shape of the bone is quite normal, with no
significant varus or valgus and some maintenance of an anterior bow.
Ideally, a medullary nail should restore this normal anterior bow and
result in a normal position of the knee. Because the pattern and
location of the fracture allowed secure fixation with a standard
medullary nail, no interlocking was performed, although I now lock
nearly all nails. The nail was intentionally left slightly short in
this fracture because an open epiphysis was present distally, even
though this patient was very close to the end of growth. Additionally,
cancellous bone near the knee is quite dense in young patients, making
it difficult to drive the nail farther.
ANTEGRADE, INTERLOCKED, REAMED NAILING USING CLOSED TECHNIQUE
The standard of care today for nearly all fractures of
the diaphysis of the femur is antegrade, interlocked, reamed nailing
using closed technique (10,11,12,14,16,17,39,45,61,72,77,78 and 79).
The main focus of this chapter will be on this technique.
Reconstruction nails providing fixation into the femoral head and neck
in combination with distal locking are not indicated in the vast
majority of diaphyseal fractures of the femur and are reserved for
subtrochanteric fractures, some pathologic fractures in the proximal
femur, and some concomitant ipsilateral fractures of the hip and
femoral shaft (see Chapter 19).
Brumback et al. (11,12,16) and Wiss et al. (78)
have all shown that the best results come from interlocked nailing that
maximizes bone apposition at the fracture site, with the use of closed
techniques and no direct intervention at the fracture site by open
means, and with static interlocking even in segmental fractures of the
femur.
REAMED VERSUS NONREAMED, LOCKED, CLOSED INTRAMEDULLARY NAILING
It is interesting to see how history repeats itself.
When Küntscher and others first developed intramedullary nailing prior
to World War II, nails were generally inserted without reaming. When
all the advantages of using larger, reamed nails became apparent,
unreamed nails for the femur nearly disappeared in developed countries.
As we began to carry out early reamed, locked nailing of fractures

P.675



P.676


of
the femur in patients with increasingly severe multiple injuries,
investigators at a number of important trauma centers in both Europe
and North America raised the issue of whether embolization of the
contents of the medullary canal to the lungs was a significant factor
in increasing morbidity (discussed in detail in Chapter 11).

Nonreamed nailing is attractive because of the decreased operating time and decreased blood loss (24,72).
In addition, the introduction of retrograde nailing through the
intercondylar notch of the knee was particularly useful with nonreamed
techniques. In a prospective randomized trial comparing unreamed to
reamed nails, Clatworthy et al. (24) found that
union was slower in the unreamed group, and a high incidence of implant
failure in the nonreamed nails led them to discontinue the study early.
Both they and Tornetta and Tiburzi (72) found
no advantage in fixing routine femoral fractures with nonreamed nails
and recommended the routine use of reamed nails because of quicker
union and fewer implant failures with the larger nails. They did not
believe that the quicker operating time and somewhat lower blood loss
in the nonreamed groups was a sufficient advantage to justify their
use. This remains controversial, however. Most surgeons recommend
unreamed nails for severely and multiply injured patients with
concomitant severe pulmonary compromise or injury (see Chapter 11).
RETROGRADE NAILING THROUGH THE KNEE
In North America, the group at Harbor View Hospital in Seattle (61,71)
introduced retrograde nailing of the femur through a lateral
epicondylar portal with the use of a Küntscher-style nail, primarily as
a means of fixing the femoral shaft in combined concomitant ipsilateral
fractures of the shaft and femoral neck. The technique has now been
abandoned in favor of using an entry through the intercondylar notch of
the femur, because the eccentric entry site made reduction difficult
and resulted in persistent varus deformity (64,68,71).
The current technique of retrograde nailing through the
knee has recently become popular because of the previously successful
experiences nailing tibias through this same approach and the challenge
of stabilizing a femur fracture in a quick, efficient manner in
patients with multiple injuries, whom orthopaedic surgeons frequently
found in the supine position on a regular operating table after general
surgical procedures. Devices used for retrograde nailing have included
specially designed short nails that provide additional screw fixation
in the distal fragment and were not designed to penetrate beyond the
isthmus (46), standard locked intramedullary nails designed for antegrade use (52),
and long retrograde nails designed for this technique. These can be
used reamed or nonreamed. The cross-locking screws are somewhat more
distal than standard nails, thereby increasing the range of
supracondylar fractures that can be handled. All current designs use AP
cross-locking proximally, as most surgeons view it as technically
easier to accomplish because of improved fluoroscopic visualization (30,32,35,37,52,53,58,63).
Retrograde insertion of a femoral nail through the intercondylar notch has some advantages over antegrade nailing:
  • Elimination of the need for a fracture table in most cases
  • A much easier starting point without the
    risks associated with proximal entry in the pyriformis fossa, which
    include injury to the sciatic nerve, fracture of the femoral neck,
    avascular necrosis of the femoral head, comminution of the proximal
    femur due to eccentric entry, and heterotopic bone at the nail entry
    site
Intercondylar retrograde insertion of nails requires
arthrotomy of the knee and partial reaming of the articular cartilage
at the roof of the intercondylar notch; therefore, potential risks
include the following:
  • Early knee dysfunction
  • Patellofemoral pain
  • Loss of knee motion
  • Arthrosis of the knee
  • Knee sepsis
  • The tendency to use smaller nails, which could increase the rate of nonunion and risk of implant breakage
  • Subtrochanteric fracture at the proximal tip of the nail
  • Less secure fixation distally,
    particularly in osteoporotic bone, since the more recently designed
    nails tend to place cross-locking screws in the cancellous bone of the
    metaphysis rather than through cortical bone at the metaphyseal
    diaphyseal junction
As the following brief review of the literature will
show, some of these concerns have been borne out through recent
clinical experience, and others have not proven to be a problem. Large
numbers of retrograde nails have been used only in the last 5 years.
Only time will tell whether this technique will continue to be used for
difficult fractures of the femur or, in the long run, replace antegrade
nailing. ElMaraghy et al. (29) studied the
effect of nail entry and reaming in the canine femur on arterial
profusion of the anterior and posterior cruciate ligaments. They
demonstrated that profusion in the anterior cruciate ligament is
decreased by 52% and in the posterior cruciate ligament by 49% acutely.
Whether this returns to normal with time and whether it has any
significant clinical effects on either healing or function of these two
ligaments is currently not known.
Morgan et al. (54) studied the
effect of the entry site on patellofemoral contact area and pressures.
When they used the intact femur for comparison, they found no
differences unless the nail was left 1 mm proud of the cartilage

P.677


and
this resulted in statistically significant increases in the mean
pressure at 120° of flexion, and in the maximum pressure at 90° of
flexion. Their studies suggest that the intercondylar entry hole does
not have any significant biomechanical effects on patellofemoral
function as long as good surgical technique is followed. To this point,
no study has been sufficiently large or long enough to state whether
this entry site will increase the incidence of later arthrosis, but,
based on the experience in sports medicine with various intraarticular
reconstructive procedures of the knee, this does not appear to be a
problem.

Because of the popularity of proximal cross-locking in the anterior posterior plane, Riina et al. (63)
studied the relationship of the femoral artery and nerve to the
proximal third of the femur. They demonstrated that the femoral artery
that lies medial to the femur has up to 15 branches in the proximal
third of the femur, but the closest lies 4 cm distal to the lesser
trochanter. The femoral nerve lies medial but has multiple branches
crossing the anterior femur, the first of which crosses 4 cm distal to
the pyriformis fossa; the fewest are above the lesser trochanter. Riina
et al. recommended that AP cross-locking always be performed above the
lesser trochanter to minimize risk to these structures.
A review of recently published larger studies (30,32,35,37,46,52,53,58)
showed only one infection, which resulted in a septic knee, among 271
fractures followed for between 1 and 2 years. The infection responded
to treatment and did not result in osteomyelitis. There were 10
nonunions (3.7%), which were more common with smaller unreamed nails
and for the most part responded to reamed exchange nailing or local
bone graft for bone deficiency. Only two rotary malunions were
reported. When there was no concomitant injury to the knee, all authors
reported that knee motion returned to normal or nearly normal, with
minimal long-term complaints. No cases of arthrosis were reported, but
of course the follow-up is quite short. Union seemed to be faster and
the incidence of nonunion lower when reamed nails larger than 10 mm in
diameter, with larger cross-locking screws, were used.
The results with retrograde nailing appeared to be
comparable to those reported with antegrade nailing, but there is
neither sufficient evidence nor a long-enough follow-up at the present
time to state whether retrograde nailing will replace antegrade nailing
as the method of choice for stabilization of fractures of the shaft of
the femur. Considering the fact that the published series thus far had
a higher incidence of open fractures and multiply injured patients, the
results are quite encouraging.
There is no question that a role for retrograde nailing
of the femur has now been firmly established. Current indications
include the following:
  • Multiple injuries, when nailing on a regular table in the supine position and use of an unreamed nail are advantageous
  • Combination fractures of the femoral
    shaft and tibial shaft, when nailing of both fractures through the same
    entry site is possible
  • Communicant ipsilateral fractures of the
    femoral neck and shaft, when independent fixation of the shaft and neck
    fractures is desired
  • Ipsilateral fractures of the acetabulum
    and pelvis, when avoidance of a surgical wound about the hip makes
    surgery of the pelvis and acetabulum on a delayed basis less risky
  • Pregnant or massively obese patients when proximal access to the hip or lateral decubitus positioning for nailing is difficult
  • Periprosthetic fractures in the distal femur when retrograde nailing is superior to open fixation or antegrade nailing
PLATE FIXATION
For fractures of the diaphysis, from the lesser
trochanter to 10 cm proximal to the knee joint, locked intramedullary
nailing is the treatment of choice and has replaced plate fixation.
Plating a mid-diaphyseal femur fracture involves a much larger surgical
procedure, more blood loss, and a higher incidence of nonunion and loss
of knee motion than medullary nailing (67). Plate fixation is indicated in the following cases:
  • For femur fractures in settings where
    inadequate experience with intramedullary nailing or nonavailability of
    intraoperative fluoroscopy or instruments and implants makes medullary
    nailing impractical
  • When a vascular repair has been done and
    plate fixation is feasible through the same surgical approach and/or is
    less risky to the vascular repair than intramedullary nailing
  • In ipsilateral concomitant fractures of
    the femoral neck when retrograde nailing or antegrade nailing using
    reconstruction nails is not available or when there is preexisting knee
    pathology that prevents the knee from being flexed beyond the 45° angle
    that is necessary to obtain the proper entry site
  • For nonunions or malunions of the femur
    when the medullary canal is obliterated or an open osteotomy and/or
    bone graft is required, when nailing will be more difficult than plate
    fixation, or when the devascularization of the femur caused by reaming
    is not desirable
  • When skin conditions or other factors at the hip and knee, such as a preexisting arthrodesis, preclude intramedullary nailing
OPEN FRACTURES
Acute open femoral fractures can be effectively treated by reamed or nonreamed medullary nailing without significantly

P.678



increased rates of infection or nonunion (4,10,45,75,77).
Perform an immediate, thorough, layer-by-layer irrigation and
debridement, and institute appropriate bactericidal intravenous
antibiotics as soon as possible. Then re-prepare and redrape so that a
clean operative field is available for intramedullary nailing. Gustilo
grades I, II, and IIA and AO grades IO1, IO2, and IO3 of open fractures
may be routinely treated in this fashion. Individually assess Gustilo
grade IIIB and AO grade IO4. If these more severe grades are highly
contaminated, have marginal soft tissues that cannot be debrided at the
initial debridement, or are neglected (more than 12 hours old), then
consider primary external fixation if the patient needs to be mobilized
immediately. If up to 2 weeks in skeletal traction is not
contraindicated in isolated fractures, then this is an alternative.
Locked intramedullary nailing can be done when the wound is closed and
there is no evidence of infection.

Early conversion from an external fixator to
intramedullary nailing does not carry an increased infection rate as
long as there has been no infection along the pin tracks. It may be
prudent to remove the external fixator, debride the pin tracks, and
close them (maintaining the patient in traction for a few days until
these wounds are sealed) before carrying out intramedullary nailing.
Gustilo grade IIIC and AO grades IO4 and NV3–5 fractures can be
initially stabilized with an external fixator, although plating may be
preferred if little additional periosteal stripping is necessary.
After immediate nailing, close the elective portions of
the wounds and leave the traumatic wound open, performing delayed
primary closure within 5 days if the wound is ready. The huge muscle
mass around the femur facilitates revascularization and provides
excellent coverage of bone and implants; therefore, flaps are rarely if
ever indicated except in fractures around the knee. More recent studies
have shown no infections or nonunions in 28 patients with Gustilo grade
I–IIIA open fractures (4), and no infections
and only two delayed unions in a series of 56 patients with open
fractures of the femur due to low-velocity gun-shot wounds treated with
immediate interlocked reamed nailing (77).
Routine bone grafting is not necessary to secure adequate rates of
union in open femur fractures treated by nailing unless there is
segmental bone loss.
Although unreamed nails have shown efficacy and
decreased rates of infection in open fractures of the tibia, no data
support their superiority over reamed nails for open fractures of the
femur.
Williams et al. (75) compared
primary to delayed intramedullary nailings in 42 patients with open
fractures, including 33% with grade III injuries, and had overall
infection and nonunion rates of 2.4%, with no significant difference
between the two groups (see Chapter 12 on open fractures).
SURGICAL TECHNIQUES
REAMED AND NONREAMED, LOCKED, ANTEGRADE, CLOSED INTRAMEDULLARY NAILING
Timing
There is a major advantage to performing closed nailing
as soon as possible in cases of multiple trauma and in isolated
fractures. In addition to the physiologic advantages of early fracture
stabilization and patient immobilization, it is technically easier to
nail a femur immediately after the injury. Muscle spasm, contracture,
hemorrhage, and swelling are less prominent, making the closed
reduction easier. The decreased length of hospitalization contributes
to the cost-effectiveness of acute trauma care.
Preoperative Considerations
Thoroughly evaluate the patient, treat all of her
injuries, and ensure hemodynamic stability. Obtain AP and lateral
radiographs of the entire fractured femur, including the entire hip
joint and knee joint. Look carefully for occult fractures of the
femoral neck. This requires a high-quality AP radiograph taken with the
hip in internal rotation to be certain that the femoral neck is in full
profile (usually done under anesthesia in the operating room). Obtain a
high-quality cross-table lateral view of the hip as well.
Place the fractured femur into skeletal traction as soon
as possible, as this is important to maintain length and to stabilize
the thigh, thereby reducing pain, reducing hemorrhage, and maintaining
overall alignment and length, which facilitates later nailing. Two
choices exist for skeletal traction. A transverse ¼-inch-diameter or
larger pin placed at the level of the tibial tubercle is most
frequently used. Its advantages are that it is not a potential
contaminator of the femoral canal (if nailing is to be delayed over 24
hours) and does not interfere with the tip of the nail as it is driven
to the distal end of the femur. Its disadvantages include a tendency to
cause knee extension as traction is applied on the fracture table, and
a limited ability to control a short distal fragment of femur.
A distal femoral pin, if placed anteriorly, can provide
better control in distal fractures without interfering with nail
placement. To position the traction pin, place a 0.062-inch or larger,
smooth Kirschner wire (K-wire) just distal to the superior patellar
pole, keeping it as anterior as possible. In this position, it is
likely to remain out of the path of the approaching nail, particularly
with nail designs that employ an anterior bow. For a distal fracture
that is to be nailed with the patient in the lateral decubitus
position, insert the traction pin somewhat obliquely so that it runs in
a distal–medial to proximal–lateral direction. The resultant pull
through a traction bow tends to counteract the valgus alignment caused
by gravity.
The major disadvantage of a distal femoral pin,

P.679



whether a K-wire or a Steinmann pin, is that if the distal fragment is
not placed into appropriate position for nailing, the pin holds it in a
fixed position, which can make reduction of the fracture and driving of
the nail exceedingly difficult. It also interferes with distal
cross-locking. The tibial pin allows the distal femoral fragment to
“float,” so alignment is simplified. This issue is addressed in more
detail during the following discussion on fracture reduction. I prefer
to use a proximal tibial Steinmann pin.

In rare cases in which the patient cannot be placed on a
fracture table or cannot undergo countertraction from the perineal
post, it is possible to apply traction and reduce the fracture by means
of a modified femoral distractor (Fig. 20.3).
The Synthes (Paoli, PA) Universal Femoral Distractor utilizes an
accessory targeting device that attaches to the proximal targeting
guide of their universal nail system. The targeting guide allows
placement of the proximal Schanz pin into the lesser trochanter through
a small stab incision (Fig. 20.3A). Place the
distal Schanz pin parallel to the femoral articular surface, distal to
the intended location of the nail tip, taking care not to traumatize
the adjacent femoral artery, vein, or nerve (Fig. 20.3B). With the fixator in place, distraction can be achieved as well as control of flexion/extension, varus/valgus, and rotation (Fig. 20.3C). Utilize a radiolucent table and an image intensifier (Fig. 20.3D).
Figure 20.3. Use of a femoral distractor to provide traction for medullary nailing. A: A Synthes accessory targeting device (left) and the proximal targeting guide to which it attaches (right). B: Placement of the distal Schanz pin. C: The fixator in place. D: Reduction of the fracture with distraction.
Baumgaertel et al. (6) found the
AO femoral distractor as efficacious as a fracture table and a better
time saver. Some surgeons even advocate nailing fractures on a regular
table, using hand traction and manipulation rather than the distractor,
as they found the distractor to be technically challenging and
time-consuming (79). This manual technique,
however, requires a very experienced surgical team and is technically
difficult, and the surgeon must be prepared to open the fracture if
reduction cannot be achieved. I recommend against this technique for
routine nailing in the hands of the average surgeon, and I advocate use
of a tibial pin on a fracture table.
Positioning of the Patient
Patients can be positioned in the lateral decubitus or supine position on a fracture table or on a radiolucent operating

P.680



table for closed intramedullary nailing. Most surgeons prefer a
fracture table and use the supine position because they are most
comfortable with this position, as they use it on a frequent basis for
hip fractures. The only advantage of the supine position is easier
positioning of the patient and assessment of rotation. Particularly for
heavily muscled or obese patients, the supine position makes access to
the nail entry point at the proximal end of the femur challenging, and
impingement of the instrumentation against the table or the side of the
patient makes fracture reduction and execution of the nailing much more
difficult.

On the other hand, the lateral decubitus position is
much more technically exacting, it requires the surgeon and usually two
assistants to position the patient, and judging rotation can be more
difficult. In addition, in multiply injured patients and in those with
pulmonary injury or spine fractures, the lateral decubitus position is
a greater threat to the spine and compromises pulmonary function. In
the latter instances, the supine position is indicated. In addition,
the supine position is much more convenient for bilateral fractures and
carrying out simultaneous procedures. Because the nailing is much
easier, I much prefer the lateral position.
Lateral Positioning
  • For lateral positioning, I use a
    ¼-inch-diameter or stronger transverse Steinmann pin placed
    transversely through the proximal tibia and dressed with a
    circumferential sterile dressing. I use a Chick-Langren fracture table (Fig. 20.4).
    Figure 20.4. Lateral decubitus positioning on a Chick-Langren table using a proximal tibial traction pin. A:
    Posterior view. Note that the perineal post between the legs supports
    the proximal fragment and is positioned to allow the anterior-superior
    iliac spine to contact the vertical component of the post. The
    fractured femur is uppermost. B: Frontal view of the same setup. C: View from the head of the patient. (Disregard the dashed lines drawn on the patient.)
  • Before moving the patient to the fracture
    table, be certain that the table is level and properly located in the
    room with reference to the anesthesia equipment and lights, and that
    the C-arm has been placed on the appropriate side of the table. Make
    sure that all appropriate equipment is mounted on the table and in the
    appropriate position for the patient. Having everything “pre-prepared”
    greatly facilitates positioning.
  • While the anesthesiologist controls the
    airway and positions the upper extremities, have an assistant control
    both lower extremities, roll the torso, and secure the patient to the
    table.
  • Once the patient is in the lateral
    position, place the T-shaped pelvic perineal post and secure it so that
    the vertical portion lies just distal to the anterior superior iliac
    spine (Fig. 20.4A). Elevate the perineal post
    to help support the proximal fragment in a horizontal position and
    allow the genitalia to drape below the post. Make sure the groin and
    all bony prominences are well padded by the appropriate pads. Also be
    certain that the greater trochanter on the down side is supported by
    the table; the table is adjustable for this purpose.
  • Secure the tibial pin into the tibial pin
    traction holder attached to the traction arm of the fracture table, and
    manually pull sufficient traction to stabilize the patient on the
    pelvic perineal post and to bring the fracture out to length. In
    transverse fractures, slight distraction is an advantage initially.
    Place the femur parallel to the floor with the hip flexed approximately
    20° to 25° and the hip internally rotated just enough to place the heel
    slightly above the patella, which is about 10° to 15°. This compensates
    for the fact that patients in the lateral decubitus position tend to
    slump somewhat forward during the nailing, which can induce an external
    rotation deformity. On the Chick-Langren table, the traction assembly
    provides sufficient support that it is not necessary to support the leg
    distal to the traction pin unless the bone quality is poor; in that
    case, fashion a sling to the overhead arm of the table to support the
    leg. Allow the knee to flex freely, as this permits flexion of the knee
    as traction is applied, thereby avoiding stretch on the sciatic nerve.
  • Never apply traction through the foot
    with the knee extended and with the hip in the flexed position, because
    this will cause a sciatic nerve stretch injury even if used for as
    little as an hour. Excessive traction and inadequate padding cause
    pudendal nerve palsy. Maintain traction on the femur only until the
    guide pin is placed, and then loosen it as much as possible to avoid
    this problem.
  • Secure the lower, unoperated leg onto a
    well-padded leg holder, placing the leg holder just proximal to the
    knee to support it and padding the peroneal nerve well to avoid injury.
    I prefer circumferential tape over a blanket into which the leg is
    wrapped, as this provides good stability and can be a source of
    traction in the event of bilateral femur fractures. Initially, position
    the lower leg as close as possible to the upper, operated leg in all
    planes to facilitate use of the C-arm.
  • Secure the upper extremities to
    double-airplane splints, securing the uppermost upper extremity to
    stabilize the thorax in a directly vertical position. After positioning
    the patient, check all bony prominences and neurovascular structures
    with the anesthesiologist to be certain that everything is padded and
    protected.
  • Bring in the C-arm fluoroscope on the
    anterior side of the patient. Place it exactly at right angles to the
    femur and in a position where the entire length of the femur from the
    hip to the knee can be seen easily on both the AP and lateral views
    without repositioning the base of the C-arm.
  • Prepare the injured thigh from the iliac
    crest to the tibial pin circumferentially, and drape it so that the
    buttocks and all but the most medial aspect of the thigh down to the
    tibial pin are available.
  • Under fluoroscopic control, extend the
    down leg at the hip just enough so that the fracture site and proximal
    femur can be clearly seen on the lateral view. Keeping the lower
    extremities as close together as possible facilitates

    P.681



    P.682


    passage
    of the C-arm from the AP to the lateral position. Extending the
    uninvolved hip permits better visualization of the fractured femur on
    the lateral view. The least extension is required for supracondylar
    fractures, and the most for subtrochanteric fractures.

Supine Position on a Fracture Table
  • Place the patient in the supine position
    on the fracture table and secure the tibial pin into the traction
    apparatus on the table (Fig. 20.5). With the
    perineal post in place, reduce the fracture and secure the patient’s
    position on the table by applying manual traction. The difference from
    the lateral position is that the thigh is slightly adducted to expose
    the trochanter better (avoid excessive adduction, as this tightens the
    iliotibial band, which produces valgus malalignment of the fracture)
    and the leg is placed in neutral rotation. Raising the traction holder
    to flex the hip 15° to 20° greatly facilitates access to the entry
    site. To facilitate exposure of the hip, tilt the torso to the opposite
    side by using either a kidney rest or a tape-and-bolster arrangement.
    Avoid too much tilt, which also tightens the iliotibial band. On the
    Chick-Langren table, the support under the buttocks is in three
    sections, and the lateralmost section on the operative side can be
    lowered out of the way to facilitate exposure. The key to this
    positioning is the flexion of the hip, which allows the instrumentation
    to pass beneath the patient rather than impinging on his side, as
    occurs if the hip is left in full extension. Place the opposite leg in
    a 90°-90° leg holder, making certain that the pelvis is stabilized to
    the table or to the vertical post of the leg-holding device. Bring the
    C-arm between the legs, and position it to be certain that AP and
    lateral views from the hip to the knee can be acquired without moving
    the base of the machine.
    Figure 20.5. Supine positioning with use of a distal femoral traction pin.
  • The knee can be left to flex freely, as
    described for the lateral decubitus technique, or a small footstool can
    be brought in to support the foot. Since it is dependent, apply an
    elastic support hose.
  • Prepare and drape the thigh free, as already described for lateral decubitus nailing.
The Skin Incision and Trochanteric Starting Point
  • Begin the incision at the tip of the
    trochanter, and extend it proximally in line with the fibers of the
    gluteus maximus and the femoral shaft (Fig. 20.6).
    The length of the incision is determined by the soft-tissue thickness.
    Continue dissection sharply through the gluteus maximus fascia, and
    then split the gluteus fibers bluntly. Stay posterior to the gluteus
    medius, palpate the trochanteric tip, and with your palm facing upward,
    slide a finger down the posterior medial border of the trochanter. You
    will encounter a sharp bump at the mid-portion of the trochanter. This
    represents the old physeal line of the trochanter and is known as the
    quadrate tubercle (Fig. 20.6A). Then slide your finger directly

    P.683



    forward into a fossa located at the junction of the femoral neck with
    the underside of the posterior central aspect of the trochanter. By
    common usage this is called the pyriformis fossa, which is actually
    nearest to the insertion of the obturator externus. This is the usual
    entry site for a femoral nail because it is directly over the medullary
    canal (Fig. 20.6B). (See the section on INTRAMEDULLARY NAILING in Chapter 11.)

    Figure 20.6. A: Posterior view of the buttocks with the posterior aspect of the greater trochanter exposed. a, quadrate tubercle, the origin of quadratus femoris; b, insertion of the obturator externus tendon; c, insertion of the pyriformis tendon. B: Top view
    of the femur. Note that the entry site is directly over the medullary
    canal and tucked up under the greater trochanter, encroaching only
    minimally on the femoral neck.
  • In the past, a Küntscher awl was used to
    penetrate the cortex at this point. Some surgeons still prefer it, but
    most have now converted to using a guide pin over which is passed a
    reamer. This procedure is easier, can be done percutaneously, and is
    more accurate and easily visualized under fluoroscopy. The danger of
    accidentally slipping off the posterior aspect of the trochanter into
    vital structures such as the sciatic nerve is eliminated. Most any
    guide pin combined with a reamer used for nailing hip fracture
    suffices. I use the 3.2-mm guide pin for the Alta hip screw (Howmedica,
    Rutherford, NJ) and drill over it with either a 14 mm or an 8 mm
    reamer, depending on the size of the patient.
  • Place the tip of the guide pin on the tip
    of your finger, palpating the pyriformis fossa, and hold the tip upward
    against the underside of the greater trochanter at its junction with
    the femoral neck. Control the guide pin with a guide pin holder or a
    T-wrench. The guide pin can be used to palpate the anterior and
    posterior aspects of the neck to be certain that you are central or
    slightly posterior. Then use a mallet to drive the guide pin into the
    soft bone on the underside of the greater trochanter at its junction
    with the neck. Do not drive it in more than 15–20 mm. Then visualize
    the guide pin on a fluoroscope, confirming that the entry point is
    directly over the medullary canal on both the AP and lateral views.
    (See Hints and Tricks below.)
Reduction of the Fracture and Insertion of a Ball-Tipped Reaming Guide Pin
  • Reduce the fracture, and insert a long
    ball-tipped reaming guide pin down the medullary canal to the
    subchondral bone at the knee. Other than finding the appropriate
    starting point, this can be the most difficult and challenging portion
    of the procedure; it is described and illustrated in Chapter 11.
    Many different methods have been advocated for reducing femur fractures
    closed, many of which utilize various external manipulation devices,
    such as mobile concentric rings or a crutch with a strap. The technique
    described here utilizes manipulation of the proximal fragment with an
    intramedullary device, makes reduction of the fracture far simpler than
    by external manipulation, follows a logical progression, and minimizes
    operating time and x-ray exposure (43).
  • Key points are the positioning of the
    distal fragment in a neutral position in both the lateral and supine
    positions, as already described. Because of muscle pull, the proximal
    fragment typically lies flexed and adducted, more so in proximal
    fractures. Subtrochanteric fractures with the lesser trochanter intact
    are externally rotated as well. The key to reduction is aligning the
    proximal fragment with the distal rather than the reverse.
  • Use a fracture reduction tool (Stryker, Osteonics, Howmedica, Rutherford, NJ) or a long, 11- to 12-mm-diameter

    P.684



    intramedullary nail inserted into the proximal fragment. Place into the
    reduction tool the ball-tipped reaming guide pin. It is helpful to
    place a gentle, smooth bend of 10° to 20° in the distal 2.5 cm of the
    guide pin, as this facilitates manipulation of the guide pin around
    corners. Attach a T-handle to the proximal end of the guide pin.

  • Place this assembly into the proximal fragment, and manually, gently lodge the reduction tool into the proximal fragment.
  • Using the step image on the fluoroscope
    to minimize radiation exposure, push your hand anteriorly, which
    extends the proximal fragment, and bring it directly in line with the
    end of the distal fragment on the lateral view. Place the guide pin at
    the exit of the distal fragment, but do not cross the fracture site yet.
  • Making certain that you do not change
    position on the lateral view, now push your hand toward the floor,
    which abducts the proximal fragment and brings it directly in line with
    the distal fragment on the AP view. Place the guide pin directly in the
    center of the medullary canal on this view, then push it 2–4 cm into
    the proximal fragment. If your alignment is perfect, you will
    immediately feel crepitus as the guide pin encounters the rough
    endosteal surface of the medullary canal. With this feedback
    demonstrating that the guide pin is in the medullary canal, insert it
    down to subchondral bone at the knee, ensuring that it is in, on both
    views.
  • If you feel no crepitus or resistance to
    penetration of the guide pin, then most likely it is out into the soft
    tissues on the other view. In that case, realign the guide pin on the
    view on which you are working, and repeat the procedure for the
    following view.
  • Perform the reduction maneuver and
    insertion of the guide pin on step images, as this dramatically lowers
    radiation exposure in comparison with live imaging. If you are
    unsuccessful after two or three attempts to pass the guide pin, then it
    is likely that there is a small fragment in the medullary canal, or
    that there is some other reason, such as soft-tissue interposition,
    that the guide pin is not passing. In this circumstance, try inserting
    the guide pin under direct fluoroscopic control, and the image will
    tell you what is occurring.
  • If there is a small fragment in the canal, use hand T-reamers (AO Synthes, Paoli, PA) to clear the canal.
  • The usual cause of failure to enter the
    distal fragment is losing the position on one view by inadvertent
    movement of the hands when switching to the other view. This takes
    practice, and after several nailings using the technique, this problem
    nearly disappears.
  • In rare fractures, some external
    manipulation is necessary because of extreme swelling in the thigh or
    other impediments to manipulating the proximal fragment.
  • In supracondylar fractures of the femur,
    where an anterior traction pin has not been placed in the distal femur,
    the distal femoral fragment may remain quite flexed. Reduce this by
    pushing on the posterior proximal end of the distal fragment to reduce
    its flexed position and hold alignment while the guide pin is passed.
  • Ideally, the guide pin should be directly
    centered in the condyles of the femur on both AP and lateral views.
    This is not critical, however; if reasonable alignment of the fracture
    is maintained during reaming and if only the isthmus of the femur is
    reamed, the position of the guide pin distally is irrelevant unless you
    are nailing a very distal supracondylar fracture where there is a
    tendency for the distal fragment to lie in valgus and flexion.
Measuring for the Length of Nail
  • Ensure that the guide pin is distally
    where you want the distal end of the nail to end. In most systems, nail
    length is determined by placing an identical second guide pin at the
    tip of the trochanter and then measuring from the top of the guide pin
    in the femur to the top of the second guide pin. Make adjustments if a
    shorter nail is desired.
  • In the Alta system, the length of nail is
    measured directly off the fracture reduction tool, which for the Alta
    standard nail places the top of the nail flush with the top of the
    femoral neck rather than even with the tip of the trochanter. This
    allows for some adjustment of the position of the nail if a more
    proximal position of the cross-locking screws is desired.
Reaming the Medullary Canal
  • To minimize pressurization of the canal,
    which leads to embolization of bone marrow, and to avoid overheating
    and jamming of the reamers, it is important to use conical (as opposed
    to cylindrical) reamer heads that have deep flutes to deliver reamings
    proximally, and reverse cutting flutes.
  • Run reamers at the highest available RPM
    on a power reamer because this makes cutting most efficient, but push
    the reamer slowly down the medullary canal to minimize pressurization.
    Visualize the passage of the first couple of reamers across the
    fracture site to be certain that everything is going smoothly.
    Fluoroscopic imaging thereafter is usually not necessary unless
    problems are encountered.
  • Begin with the smallest reamer in the set
    and progress in 1 mm increments until the first reamer contacts the
    inner cortex of the canal. Then progress in 0.5 mm increments until the
    desired diameter is reached. Once cortical contact is made, feel the
    reamer head with your hand as it is removed after each pass to be
    certain that it is not overheating. If you encounter hard reaming with
    slow progress of the reamer, extract the reamer and clean the flutes of
    cortical bone to increase cutting efficiency and to avoid jamming the
    reamer when it is withdrawn. Do not allow the reamer to pull itself
    distally

    P.685


    because
    this may cause it to screw down the medullary canal, which inadequately
    reams the canal and may make extraction of the reamer difficult. For
    most reamed nailings, reaming of 2 mm after initial cortical contact
    suffices. Most systems provide reamers from approximately 8 to 17 mm in
    diameter. More reaming can be done if a larger nail is indicated. Do
    not ream the metaphyseal cancellous bone distal to the isthmus, because
    driving the rod into this bone increases the quality of fixation. In
    supracondylar fractures, it is necessary to enter only the mouth of the
    fragment to be certain that the nail will pass smoothly. Be certain
    that the guide pins used are straight and are not nicked. Imperfect
    guide pins can cause jamming between the guide pin and the reamer head
    when it is withdrawn, causing inadvertent removal of the guide pin and
    loss of reduction of the fracture.

  • Now insert a plastic tube over the
    reaming guide pin down the medullary canal, making certain that it is
    across the fracture site, and exchange the reaming ball-tipped guide
    pin for a driving guide pin.
Selecting and Driving the Intramedullary Nail
  • In systems of reamed, locked nails
    manufactured from titanium alloys, an 11-mm-diameter nail for women and
    a 12-mm-diameter nail for men provide sufficient strength and
    stability. In stainless-steel systems, some manufacturers recommend a
    12 mm nail for women and a 13 mm nail for men. Smaller nails may be
    necessary in very small people, and larger-diameter nails may be
    necessary in large men or in the elderly, where the medullary canal
    tends to be capacious. The relationship between the maximum diameter
    reamed and the size of the nail is based on the manufacturer’s
    recommendations, but in most cases overreaming by 1 mm provides a
    suitable fit. Some manufacturers recommend overreaming even more,
    particularly in femurs that have a fairly marked anterior bow.
  • Assemble the nail on the driver and drive
    the medullary nail down the medullary canal across the fracture site,
    using the driver provided by the manufacturer. At this point, only
    minimal traction should be necessary, and in fact, traction on the
    extremity should have been released as soon as the fracture was reduced
    and the guide pin passed, assuming that traction is not necessary to
    maintain overall alignment or length. A nail that drives hard can be
    the result of an eccentric entry site or insufficient reaming. Do not
    persist in driving the nail because of the risk of nail incarceration
    or comminution of the femur. Reassess the entry site; oval it in the
    correct direction, if necessary, and consider reaming somewhat larger.
  • In comminuted fractures where
    determination of length may be difficult, a good guide is to
    preoperatively measure the opposite extremity from the prominence of
    the trochanter to the lateral epicondyle of the femur and relate this
    to the length of nail to be used.
  • Avoid distraction at the fracture site.
    If distraction occurs when you drive the nail, then drive it slightly
    deeper, cross-lock distally, and use the driver assembly to backslap
    the nail to approximate the fracture ends; then lock proximally. In
    supracondylar fractures, be certain to provide support beneath the
    fracture site medially as the nail is driven to ensure that it is
    centered in the distal fragment and that alignment is good.
  • Visualize the nail as it passes into the
    mouth of the distal fragment to ensure that it does not hook up on a
    cortex and cause comminution. Most nails today have a bullet-nose
    design, which minimizes the risk of comminution.
  • Visualize the nail fluoroscopically to be
    certain that it is in the appropriate location and that the fracture is
    both rotationally and axially in good alignment.
Cross-Locking the Nail
  • In nearly all nail systems today, distal
    cross-locking is best accomplished by freehand cross-locking under
    fluoroscopic control (discussed in Chapter 11).
    The jigs now available for distal cross-locking have had mixed success.
    Hand-held fluoroscopic guides and laser-guided systems facilitate
    freehand cross-locking.
  • P.686


  • Proximal cross-locking can be done with a jig or guide system in all nails.
  • When cross-locking is completed, verify
    on a lateral view that the screws have passed through the nail. They
    can appear on the AP view to be in, and yet have missed the nail.
  • If a screw misses the nail, then
    correcting it can be quite difficult because the drill point or screw
    tends to drop into the old screw hole. A quick and simple way to solve
    this problem is to expose the screw hole, widen it with a larger drill
    point so that the near hole in the nail is now exposed, and then insert
    the screw under direct vision. This results in loss of purchase in the
    near cortex, but, when combined with a second screw, it provides
    adequate fixation on the opposite cortex as long as bone quality is
    good.
  • In osteoporotic bone, if fixation is too
    distal, it will provide poor purchase because of the poor-quality
    cortical and cancellous bone in the very distal regions of the femur.
    In addition, very distal fixation may place the screws
    intraarticularly, which can irritate the knee joint. In these
    situations, use a shorter nail so that cross-locking occurs in the
    better-quality bone of the more proximal metaphysis.
  • It is almost always advisable to have
    three to five threads of the screw extending through the opposite
    cortex, because this greatly facilitates removing the medial section of
    screw if the screw breaks.
  • In osteoporotic bone, washers and backup nuts enhance fixation.
  • In stable patterns such as
    Winquist-Hansen I and II, where there is good cortical contact at the
    fracture, a single screw proximally and distally usually suffices. This
    saves operating time and money. If single screws are used where two or
    three cross-locking holes are available, always place the screw in the
    hole closest to the fracture site to eliminate a stress riser effect in
    the nail. This is particularly important in supracondylar fractures (15).
  • If you are using percutaneous techniques,
    measurement of screw length with the depth gauge through the nail can
    be difficult. You can measure screw lengths directly off the
    fluoroscope monitor screen by using the known width of the nail as a
    reference for measuring the AP width of the cortex at the site of the
    cross-locking screw in the femur. The only time this is not reliable is
    when cross-locking distal to the epicondyle, where the sloping surface
    of the distal femur may cause you to insert a screw that is too long.
  • If the nail has a cap screw, insert this
    to prevent bone growth into the nail and to make extraction of the nail
    easier. In addition, it prevents the oozing of marrow contents out the
    proximal end of the nail, which can lead to heterotopic bone formation.
  • Thoroughly irrigate the wound, debride
    any muscle injured by the procedure, and close the wounds in layers.
    Confirm the quality of reduction and placement of the implants on final
    AP and lateral radiographs before awakening the patient.
Special Considerations for Supine Nailing
The overall technique for supine nailing is very similar
to that for lateral nailing, and the following guidelines are quite
useful:
  • When approaching the entry site for the
    nail, do not use a direct mid-lateral incision, but use the same
    incision as for lateral decubitus nailing, beginning at the tip of the
    trochanter and passing proximally in line with the fibers of the
    gluteus maximus. A direct lateral approach goes through the gluteus
    medius, damaging it. The thick muscle envelope in this area makes
    access to and visualization of the proper entry site extremely
    difficult. In addition, entry this far superior makes it exceedingly
    difficult to avoid injury to the base of the femoral neck because of
    the overhanging trochanter. A more posterior approach is better.
  • In extremely obese people or where the
    greater trochanter overhangs significantly, entry into the pyriformis
    fossa in the supine position can still be difficult. This is also a
    challenge when there is a concomitant ipsilateral fracture of the
    femoral neck, where the guide pin tends to enter the fracture site
    rather than the pyriformis fossa, which is just a few millimeters
    lateral to the fracture line.
  • In these cases, place the guide pin
    directly into the tip of the greater trochanter. Ream the entry site
    into the greater trochanter and then use an instrument such as an
    Army-Navy retractor to push or pull the reamer medially while reaming,
    to create an oval entry hole that is centered directly over the
    medullary canal. This oval hole is larger than is needed for the nail,
    but I have seen no adverse consequences of this technique.
  • Position the operating table as high as
    is practical, and plan to have the instrumentation enter in line with
    the femur, which is posterior to the side of the patient. Note that
    this is nearly identical to the lateral decubitus position. This
    greatly facilitates manipulation of instrumentation and reduction of
    the fracture, as well as driving the nail. The only disadvantage is
    that it is awkward because of the inclination of the instrumentation
    toward the floor.
  • Cross-locking distally in the supine
    position can sometimes be difficult because the locks on the C-arm
    prevent achieving appropriate alignment of the x-ray tube with the
    cross-locking holes in the nail. As long as the fracture is well
    impacted and rotationally stable, it is often easier to rotate the leg
    to bring the cross-locking holes in line with the central beam of the
    fluoroscope than to try to line it up using the fluoroscope mechanics.

P.687


Postoperative Care
  • Always examine patients carefully in the
    operating room under anesthesia to be certain that alignment,
    particularly rotation, has been restored. Always try to achieve perfect
    alignment. Some asymmetry in rotation is acceptable as long as it does
    not exceed 15° and the patient can rotate in both directions beyond
    neutral. Shortening up to 10 mm is either asymptomatic or easily
    compensated for by an in-the-shoe lift. Shortening of 15 mm or more,
    although nicely corrected with a shoe lift, is quite noticeable to most
    patients and is likely to be the source of complaints. Shortening over
    20 mm is unacceptable to the vast majority of patients, and correction
    is required in most patients who have shortening over 25 mm. Except in
    distal supracondylar fractures of the femur, alignment in the sagittal
    plane is nearly always guaranteed by the contour of the nail. The most
    common malalignment in supracondylar fractures is recurvatum.
    Recurvatum of up to 10° or so is acceptable in most patients as long as
    they do not have lax joints or a preexisting hyperextension problem at
    the knee. Try to avoid varus or valgus malalignment of over 5°. Over
    10° often requires correction.
  • As soon as the patient will tolerate it,
    begin hip and knee range-of-motion and muscle-strengthening exercises
    for the entire extremity. Do not forget the hip abductors; persistent
    weakness in these due to the operative approach is common (3).
    It usually takes 6 weeks to gain enough joint motion and muscle
    rehabilitation that the patient can walk safely without assistive
    devices. In Winquist I and II stable patterns, full weight bearing can
    progress as quickly as possible, and most patients can bear full weight
    without assistive devices by 6 weeks. In Winquist III and IV patterns,
    bridging callus must be seen on two views before progression to full
    weight bearing is acceptable.
See Figure 20.7 for a typical case of locked, antegrade, closed intramedullary nailing.
Figure 20.7. Femoral fracture treated with antegrade, closed, locked intramedullary nailing. A: AP radiograph showing fracture. B: Lateral radiograph. C:
Postoperative AP radiograph. Note the cross-locking screws about the
lesser trochanter and, distally, the bicortical cross-locking screws.
The fracture is in excellent alignment. D: Lateral radiograph.
Antegrade Nailing on a Regular Radiolucent Table with a Universal Femoral Distractor
  • The lateral decubitus position with the
    fractured femur uppermost is preferable, although some use the supine
    position with bumps under the back and midsacrum to make the entry site
    for the nail available.
  • Prepare and drape the extremity completely free, including the buttocks for the entry site.
  • Apply the AO distractor (Fig. 20.3). Make the anterior insertion of a Schanz pin into the medial aspect of the

    P.688



    femur at the lesser trochanter carefully to avoid injury to the femoral nerve and artery.

  • Reduce the fracture and perform antegrade nailing as described previously.
  • When you are cross-locking, the
    distractor may be in the way, particularly distally. If the fracture is
    sufficiently stable, remove the distractor and proceed with
    cross-locking as usual. In cross-locking with the leg free on a
    radiolucent table, it is useful to have the assistant position and
    stabilize the leg for cross-locking rather than shift the head of the
    C-arm to align it.
RETROGRADE INTERCONDYLAR NAILING
  • Position the patient in the supine
    position on a fully radiolucent table. Place a 3–5 cm soft bump under
    the ipsilateral buttocks and a large bolster under the knee and thigh
    so that the knee sits flexed at approximately 45°.
  • If a universal distractor will be used to
    gain length and aid reduction, apply a Schanz pin proximally just above
    the lesser trochanter and distally through the femoral condyles
    posterior to the nail entry site, attach the distractor, and make
    appropriate adjustments. Many fractures can be nailed without a
    distractor, particularly if an unreamed technique is used, which is
    discussed below.
  • Make a midline longitudinal incision from
    the inferior pole of the patella to the tibial tubercle. Split the
    patellar tendon the full length of the incision, and reflect the fat
    pad inferiorly. Identify the intercondylar notch. One centimeter
    anterior to the anterior edge of the posterior cruciate ligament,
    insert a guide pin under fluoroscopic control (Fig. 20.8).
    On the AP view, make certain that this is in the midline and aligned
    with the femoral canal. Laterally, the guide pin must be aligned with
    the long axis of the distal fragment of the femur and be central in the
    canal. (This usually results in an entry hole that is 50% on
    nonarticular surface in the intercondylar notch and 50% into the
    articular cartilage.) Over this pin, ream with an appropriate-size
    end-cutting reamer. In most cases, this will remove a few millimeters
    of articular cartilage at the superior edge of the intercondylar notch.
    Ream until the medullary canal is entered. Use a protector to prevent
    injury to the patellar tendon, and irrigate and suck during reaming to
    remove any loose reamings from the knee joint.
    Figure 20.8. Entry site for retrograde femoral nailing. A: Placement of the guide pin for the reamer. B: Lateral view showing the guide pin directly in line with the medullary canal of the femur.
  • For unreamed nailing, select the
    appropriate diameter and length of nail. I prefer to use a standard
    Alta medullary nail, which provides two transverse cross-locking screws
    both proximally and distally. This is the same nail used for antegrade
    nailing. For unreamed applications, I always try to insert a 10 mm nail
    because it allows use of the 5 mm cross-locking screws, but a 9 mm nail
    is acceptable. Although specially designed retrograde nails are
    available at our institution, I prefer this nail because, when it is
    fully inserted, the two distal cross-locking screws are at the
    epicondyle and slightly above, placing them in cortical bone, which
    provides better fixation than more distal locking. I prefer
    medial-lateral rather than anterior-posterior locking proximally. I
    select a nail that will place the two proximal cross-locking screws in
    the region of the lesser trochanter, which is the safest zone for
    neurovascular structures and avoids the problem of a stress riser in
    the subtrochanteric area.
  • Insert the unreamed nail up to the
    fracture site, making certain that the fit is not too tight. With the
    aid of an assistant, reduce the fracture and, using the nail as a lever
    to assist, pass the nail across the fracture site and drive it into the
    proximal fragment.
  • For reamed technique, the method is identical to that

    P.689



    already described for reamed nailing. The fracture manipulation tool
    for controlling the distal fragment is equally advantageous here.

  • Nail length is critical because the nail
    must be countersunk below the subchondral bone distally, the fracture
    must be at appropriate length, and the proximal cross-locking should be
    at the level of the lesser trochanter.
  • First, cross-lock distally, using the
    cross-locking guide, then impact the fracture and cross-lock
    proximally, using either the Alta distal cross-locking guide or
    freehand technique.
  • With the other leg lying flat on the
    table, visualization of the cross-locking holes at the lesser
    trochanter is easy because the two thighs are at different levels. Even
    if overlap occurs, with a good C-arm both femurs can be visualized and
    the femur with the nail in it is easily identified and targeted.
  • When using systems that provide a cap
    screw, be certain to take this into account when setting the depth of
    the nail at the knee (Fig. 20.9).
    Figure 20.9. Retrograde nailing of a femur fracture. A: AP radiograph showing a fracture of the shaft of the femur in a patient with multiple injuries. B:
    Postoperative AP radiograph showing retrograde nailing through the
    intercondylar notch of the knee with proximal and distal cross-locking.
    C: Lateral radiograph.
  • After completion of the nailing,
    thoroughly irrigate the knee joint to remove any loose bone, do a
    meticulous interrupted closure of the patellar tendon, and close the
    subcutaneous fat and skin in a routine fashion.
Postoperative Care
Postoperative care is the same as that already described for antegrade nailing.
PLATE FIXATION OF THE FEMORAL SHAFT
Although a medullary nail is the preferred method of
operative fixation for most diaphyseal femoral fractures, there are
situations in which plate fixation is appropriate (7,47,48,55).
  • Place the patient in the supine or
    lateral decubitus position on a radiolucent table. Make a midlateral
    longitudinal incision over the fracture, and sharply incise the
    iliotibial tract. Beyond this, the exposure is guided by the muscle
    disruption and periosteal stripping that are already present. Carefully
    clear the major fragment ends of soft tissue and the lateral or
    anterolateral surface where the plate will be located.
  • Reduce the fracture, typically by means
    of traction on the distal fragment. Maintain the anterior bow by
    placing a bolster posterior to the femur as distraction is applied.
    This can be effectively accomplished by means of a femoral distractor,
    as described by Mast (Fig. 20.3).

    P.690



    If small, comminuted fragments are present, do not strip them in an
    attempt to produce an anatomic reduction. Tease them into the best
    position possible, but do not worry about fixation. Fix large butterfly
    fragments anatomically by means of lag screws. Maintain correct
    rotation by using the linea aspera and fracture configuration as a
    guide.

  • Apply a single, broad plate to the
    lateral or anterior surfaces, obtaining eight cortices of fixation in
    both proximal and distal fragments, preferably augmented with lag-screw
    fixation between the major fragments. Before fixation is completed,
    obtain compression if there is good cortical contact by using an
    articulated tensioning device. The resulting construct may leave
    several plate holes unfilled and some comminuted fragments unreduced,
    but it will have correct length and alignment (Fig. 20.10).
    Figure 20.10. Plate fixation of a femur fracture. A:
    Lateral radiograph showing a comminuted mid-shaft grade IIIC open femur
    fracture with an associated transection of the femoral artery. B:
    Postoperative AP and lateral radiographs. The artery has been repaired,
    and the femur has been fixed with interfragmentary screws and a broad
    plate.
  • Apply autologous cancellous bone graft
    for medial comminution. If graft can be introduced through the fracture
    to the medial cortex before final reduction, less stripping of bone is
    required. Otherwise, carefully pack graft over the anterior cortex to
    rest medially and wherever comminution is extensive. Close the wound in
    layers over a suction drain.
  • Postoperatively, begin immediate
    active-assisted range-of-motion exercises and mobilization identical to
    the postoperative program for patients with nailed fractures. Continue
    touchdown weight bearing for 6 weeks, and then advance to partial
    weight bearing once bridging callus seen on two views has developed.
    Begin full weight bearing when the fracture has healed.
Plate removal may be advisable in active young people,
particularly if they engage in a sport or have a vocation in which
heavy stress on the femur or trauma may occur. Usually, we wait at
least 18 months after the time of plating, and remove the plate only if
the bone is solidly healed and the callus has remodeled. (See Chapter 11
for advice on plate removal.) After removal, patients can progress to
full weight bearing with crutches as soon as they can tolerate it, but
they must limit their activities to just walking and nonimpact
activities such as swimming and stationary bicycle until 6 months after
removal to prevent refracture.
EXTERNAL FIXATION WITH AN AO FIXATOR OR WAGNER DEVICE
  • Although not essential, the use of a
    fracture table is quite helpful in the application of external
    fixation. An image intensifier speeds up the procedure. Position the
    patient and reduce the fracture with traction. It is important to
    realize that, although small amounts of angular deformity can be
    corrected after the pins have been placed, rotational correction may be
    impossible to correct in some fixators. Therefore, the initial
    reduction

    P.691


    should
    be as close to perfect as possible and should not be made with the
    expectation that the device can later be adjusted to compensate for an
    inadequate reduction.

  • For an AO or Wagner fixator, use Schanz
    pins 6 mm in diameter, for which 4.5 mm holes are drilled in both
    cortices. To minimize tie-down of the quadriceps, apply the fixator
    laterally. Make a lateral stab wound over the femur at the level of the
    lesser trochanter. Carry it down through the deep fascia, then use a
    small elevator to slide posterior to the quadriceps and anteriorly to
    the lateral intermuscular septum. Elevate a small area on the femur.
  • Drill the femur through a drill guide
    with a 4.5 mm drill point, and insert a Schanz pin bicortically. Place
    the fixator over the pin, adjust the fixator for length and alignment,
    and place the second Schanz pin through the distalmost hole of the
    fixator or through a guide if one is provided by the manufacturer.
    Place this pin at or just proximal to the lateral epicondyle.
  • Adjust the fixator, confirm proper
    position of the fracture, and insert the two innermost pins. Some
    prefer to use three pins in each fragment.
  • Achieve final reduction, lock the pin holders, and then apply compression across the fracture if possible.
  • Follow the instructions for pin care outlined in Chapter 11.
    Regularly check the tightness of the clamps that secure the pins. Begin
    partial weight-bearing and active range-of-motion exercises
    immediately. At 8 to 12 weeks, the device may be replaced by a cast
    brace if callus has appeared and the fracture location and pattern are
    appropriate. Otherwise, the fixator can remain in place until the bony
    union is strong enough to eliminate the need for a cast brace, which is
    usually at least 4 months after fracture (Fig. 20.11) (26).
    Figure 20.11. Femur and tibial shaft fractures treated in external fixation.
IPSILATERAL CONCOMITANT FRACTURES OF THE HIP AND SHAFT
Ipsilateral concomitant fractures of the hip and femoral
shaft are unusual and pose a difficult treatment problem. Most of these
are intracapsular fractures, with a high shear angle similar to a
Pauwels type III fracture of the femoral neck. However, the fracture
tends to begin distally in the base of the neck and extend across the
neck, exiting in a subcapital location. Many of these are initially
completely undisplaced and therefore are difficult to diagnose. In a
review of the literature, Swiontkowski et al. (71)
noted that in 10 articles on the subject, roughly one third of the
femoral neck fractures reported were not detected initially. Casey and
Chapman (18) postulated that most of the force
in an ipsilateral femoral neck and shaft fracture is dissipated in the
femoral shaft fracture, therefore often resulting in undisplaced or
minimally displaced fractures of the femoral neck, which therefore have
less soft-tissue damage and a much lower incidence of avascular
necrosis of the femoral head than would be expected.
In every fracture of the shaft of the femur, an
ipsilateral fracture of the femoral neck must be ruled out. Routine
radiographs of the femur must include the hip joint with a good-quality
AP and cross-table lateral views. From a practical viewpoint, many of
these radiographs are taken in the emergency room under duress and do
not give an ideal view of the femoral neck. For that reason, it is
wise, prior to nailing a femoral-shaft fracture, to obtain a
good-quality AP radiograph of the hip while it is in internal rotation,
to throw the neck into full profile, with the patient on the fracture
table and under anesthesia. A good-quality lateral must be taken as
well, which the fracture table optimizes. If you detect a femoral neck
fracture, you may need to alter treatment plans. When nailings are
completed, repeat the radiographs of the hip, because hairline
fractures not seen on initial films may appear after the vigor of a
nailing. Any patient who has unexpected proximal thigh or hip pain in
the postoperative period should have these radiographs repeated because
these fractures may appear even up to 6 weeks later in spite of
good-quality initial films. Computed tomography, magnetic resonance
imaging, and bone scans are more sensitive than

P.692


routine
radiographs, but it is not practical and probably not economically
feasible to order one or more of these studies on every patient with a
femoral shaft fracture.

If the femoral neck fracture is picked up early, is not
displaced, and is internally fixed in good position, the outcomes are
usually good. Union rates approaching 100% and rates of avascular
necrosis under 15% can be expected (71).
Because of the risk of displacement of these fractures, which worsens
the prognosis, they should be treated as surgical emergencies and
internally fixed as soon as possible. Most are either undisplaced or
can be reduced well on a fracture table with traction and gentle
internal rotation despite an unstable shaft fracture. For that reason,
fixation of the shaft is not always necessary to achieve a satisfactory
reduction of the femoral neck. One can choose between fixation of the
femoral neck or fixation of the shaft first, using independent fixation
or a single device that provides combined fixation of both fractures.
Alternatives for fixation include the following:
  • Combined fixation with a gamma nail or reconstruction type nail
  • Antegrade nailing of the femoral shaft
    with a standard nail or the AO “miss a nail,” followed by fixation of
    the femoral neck with independent screws in front of the nail (Fig. 20.12)
    Figure 20.12. Independent fixation of an ipsilateral concomitant fracture of the femoral shaft and neck. A:
    This basilar femoral neck fracture was not apparent on the initial
    radiographs after locked nailing of the femoral shaft fracture. The
    fracture subsequently displaced slightly, revealing its presence. The
    fracture reduced anatomically with closed reduction. Fixation was
    achieved with two cannulated screws placed in front of the
    intramedullary nail. B: Lateral
    radiograph. If this fracture had been noted at the time of initial
    nailing, then fixation with a retrograde nail and independent screw
    fixation would have been a better approach.
  • Independent fixation of the hip with
    multiple screws or a compression hip screw combined with independent
    fixation of the femoral shaft with an intercondylar retrograde nail or
    a plate
  • Other possible combinations
Use of gamma or reconstruction nails for such a
combination can be quite challenging and requires considerable
experience. For surgeons without extensive experience with these
devices and this combination of fractures, independent fixation is
probably easier and more reliable. In undisplaced fractures, multiple
screw fixation, as described in Chapter 19 for the femoral neck, usually works

P.693



well. If fractures are displaced, they tend to be much more unstable,
and in young, vigorous individuals, a device such as a sliding
compression hip screw, which provides better purchase on the shaft of
the femur, may be indicated. In most cases, for the femoral shaft I now
recommend a retrograde nail through the intercondylar notch at the
knee, and occasionally plate fixation in select cases.

Postoperative care is determined by the healing of the
femoral neck. This may require as long as 6 months, during which time
the patient must not bear more weight than the weight of the leg, using
assistive devices.
SIGNIFICANT CORTICAL LOSS
A substantial segment of cortex may be lost at the scene
of injury or be completely detached from soft tissues at the time of
initial debridement. Remove contaminated free cortical fragments when
debriding open fractures, because they may become infected (Fig. 20.13).
Although shortening of as much as 1.5 cm is usually tolerated, avoid
larger discrepancies and maintain length with an interlocking nail.
Many femurs heal even with a gap of several centimeters, particularly
in patients under 20 years of age. Cortical losses of 3 cm or more are
more likely to not heal. The initial nailing is not definitive and may
be performed using a smaller-diameter nail with little reaming, because
little weight bearing will be permitted. Several weeks after healing of
the soft-tissue envelope, the patient may undergo a bone-grafting
technique described by Chapman (see Chapter 30) (Fig. 20.13).
Remove the initial nail and ream the canal to accept a larger nail of
appropriate diameter. Using a chest tube, pass a slurry of iliac crest
cancellous graft and reamings down the medullary canal into the
fracture site, followed by placement of a new statically locked nail.
The graft incorporates over several months, after which time weight
bearing is progressively increased. Another alternative is a
microvascularized fibula (40).
Figure 20.13. A:
A grade III open fracture of the femur with approximately 10 cm of
cortical loss at the scene of injury. The fracture was initially
irrigated, debrided, and stabilized with a narrow-diameter, statically
locked medullary nail. Local wound care consisted of dressing changes,
with uneventful healing of the soft-tissue envelope. B:
At 1 month from injury, the original nail was removed and the canal
reamed to accept a 15 mm, statically locked nail. Simultaneously, a
large amount of cancellous bone was harvested from the iliac crest and
introduced into the fracture gap through a chest tube, using a
technique described by Chapman (22). C: At 1 month from grafting, early callus can be seen forming across the fracture site. D: At 3 months from grafting, further callus formation is evident. E: At 15 months from fracture, the femur has completely reconstituted and hardware has been removed.

P.694


RESULTS
Since the introduction of interlocked intramedullary
nails and current methods for irrigation and debridement of open
fractures, the results in fractures of the shaft of the femur are truly
remarkable. In closed fractures utilizing closed intramedullary
techniques with reamed nails, reported union rates are from 98% to
100%, with infection rates only from 0% to 1% (4,6,12,17). In low-grade open fractures, Gustilo I through IIIA, comparable results are achievable (3). In grade IIIB open fractures, the infection rate after initial nailing approaches 11% (10). Therefore, a more sanguine and cautious approach is indicated for severe open fractures.
PITFALLS AND COMPLICATIONS
Complications specific to fractures of the shaft of the
femur include malunion, nonunion, hardware breakage, infection,
neurovascular injury, avascular necrosis of the femoral head,
heterotopic ossification at the hip, and loss of joint motion.
MALUNION
With attention to detail, malunion can be avoided in the
vast majority of cases. Varus or valgus malunion is most common in high
subtrochanteric fractures, where varus is most commonly the result of
an eccentric entry point, which, when combined with a large medial bone
deficiency, results in loss of position of the proximal fragment during
the nailing. Be certain that the entry site is correct. In some cases,
it may be necessary to insert a Schanz pin into the proximal fragment
to control it during nailing, or to open the fracture to reduce and
stabilize it directly. In supracondylar fractures nailed in the lateral
position, valgus may occur because of the angulation induced by gravity
on the fracture, and recurvatum at the fracture can occur in the supine
position. Keeping the guide pin centralized, and controlling the distal
fragment as the nail is driven down into it, will usually prevent these
problems. Control with a Schanz pin on a T-handle or opening the
fracture may occasionally be necessary. Shortening most commonly occurs
when there is significant comminution, which makes judgment of length
difficult. Follow the recommendations in the section on “Selecting and Driving the Intramedullary Nail” to avoid leg-length discrepancy.
Rotational malunion is the most difficult to avoid.
Preoperative evaluation of the range of motion of the opposite hip and
attention to the angle of progression of the

P.695


foot
when the patient is lying in the supine position on the operating table
often provides good guidelines as to how the rotational alignment of
the fractured limb should be set. Yang et al. (80)
recommend preoperative evaluation of the anteversion angle of the
intact femur to provide more accurate assessment of rotation. With the
advent of nonreamed small locked intramedullary nails, torsional
distortion of these nails has been a new and unexpected cause of
rotational deformity. This can be avoided by using stiffer nonreamed or
larger reamed nails (41). Careful postoperative
examination of the patient on the operating table usually reveals any
deformity, particularly in rotation. Plan to correct this at the time
of the initial surgery rather than having to bring the patient back for
repeat surgery or reconstructive surgery.

NONUNION
As mentioned earlier, nonunion is uncommon, and the treatment is addressed in Chapter 30.
For the most part, nonunions occur in open fractures or in those with
significant comminution or segmental bone loss, particularly if they
are treated with small nonreamed nails. Any distraction in the fracture
site, despite the tremendous healing capability of the femur, can also
lead to nonunion. In segmental bone loss, early bone grafting, either
open or with closed intramedullary technique, is indicated as soon as
the soft-tissue envelope has recovered. In open fractures or where
nonreamed nails were used, when no callus has formed by 12 weeks after
injury, consider conversion to a large reamed locked nail. In delayed
unions or nonunions where the problem seems to be distraction in the
fracture site, dynamization may be sufficient if rotational stability
will be maintained. Otherwise, exchange reamed locked nailing and
compression are best.
NAIL OR SCREW BREAKAGE
Breakage may occur early in small-diameter nonreamed
nails that use small screws, and late in larger reamed nails when the
patient has continued to bear weight in the face of a nonunion. Removal
of the side of a screw where the screw head is, presents little
problem. Removal of the opposite segment of the screw may require an
independent medial approach and, occasionally, overreaming of the
screw. Removal of the proximal end of the nail with the extractor
presents few problems. Removal of the remainder of a cannulated nail
can be done in many ways, but I find that inserting a ball-tipped guide
pin through the tip of the nail and then jamming it with a
non-ball-tipped guide pin and extracting the nail with a T-handle on
the ball-tipped guide pin usually works best.
INFECTION
Acute infections with antibiotic-sensitive organisms
occurring within a few weeks after nailing, where the nail is stable
and there is no involvement of the medullary canal, may respond to
simple irrigation and debridement of the wound, delayed primary
closure, and intravenous antibiotics. When the medullary canal is
involved, particularly with more difficult or resistant organisms, nail
removal, debridement of the infected wound and medullary canal by
reaming, intravenous antibiotics, and placement of
antibiotic-impregnated beads into the medullary canal are usually
required. Stability can be maintained by either longitudinal traction,
if early renailing is anticipated, or external fixation. If the patient
responds quickly to antibiotics, then early renailing after resolution
of the infection may be possible. Otherwise, the fracture may need to
be managed definitively in an external fixator (see Chapter 135).
NEUROVASCULAR INJURY
Palsies in the upper extremities and in the unaffected
lower extremity can be avoided by using appropriate positioning and
padding. Pudendal nerve palsy is correlated with excessive traction;
you can minimize it by using a large, well-padded perineal post and
minimizing the amount and duration of traction used for intramedullary
nailing on a fracture table (9). Sciatic nerve
palsies usually present as a total common peroneal nerve palsy and
partial posterior tibial nerve palsy. Although direct surgical injury
of a nerve in the incision in antegrade nailing is possible, it is
exceedingly rare. In very small individuals, especially when the hip is
flexed 20° or more, the sciatic nerve may drift quite close to the
trochanter. If there is any question about the location of the sciatic
nerve, lengthen the wound and explore it before placing sharp
instruments for entering the proximal femur. In most cases, sciatic
nerve paresis is caused by a combination of traction and excessive
flexion of the hip with the knee in extension. Avoid this problem by
never putting the foot into a traction stirrup but rather applying
traction through a proximal tibial or distal femoral pin, allowing the
knee to flex freely to accommodate for stretch on the sciatic nerve.
AVASCULAR NECROSIS
Avascular necrosis of the femoral head secondary to
antegrade femoral nailing is exceedingly rare but has been reported. It
is most likely caused by interruption of the blood supply to the
femoral head as a result of encroachment of the nail entry onto the
superior aspect of the femoral neck (2). Avoid
this complication by avoiding reamed nails in very small patients or in
children with open physes. If antegrade nailing is indicated, use of
flexible nails such as the Ender nail through the tip of the trochanter
provides a good alternative.

P.696


HETEROTOPIC OSSIFICATION ABOUT THE HIP
In a review of 80 patients undergoing antegrade reamed nailing, Brumback et al. (13)
found an incidence of 26% of heterotopic ossification. They found that
irrigation with pulsatile lavage did not affect either this incidence
or the severity. Notably, the nail used in their study had an open top
that would permit the ooze of marrow contents into the hip region
postoperatively. It is rare that this heterotopic ossification is
sufficient to be symptomatic, but it is a significant impediment to
nail removal, and infiltration into the gluteus medius musculature can
result in significant damage to the muscle at the time of attempted
removal of the ossification or of a medullary nail. It has been my
experience that heterotopic ossification can be minimized by the
following:
  • Taking care to carefully protect the
    gluteus maximus and medius as well as other soft tissues about the hip
    during antegrade nailing
  • Thoroughly irrigating and debriding the
    wound at the terminus of the procedure, to remove any marrow contents
    or bone that might predispose to heterotopic ossification
  • Using an intramedullary nail that has a
    cap screw that closes the nail off and prevents retrograde oozing of
    marrow contents postoperatively. If the heterotopic ossification is
    symptomatic, the treatment is removal of it after it has matured, which
    is usually at least 1 year after injury.
JOINT STIFFNESS AND MUSCLE WEAKNESS
Since the advent of modern techniques, loss of motion in
the hip and knee after fractures of the shaft of the femur is uncommon
unless there is concomitant ipsilateral injury of these two joints. It
is important, however, to institute an early range-of-motion and
muscle-strengthening program because neglect of rehabilitation can lead
to stiffness and persistent weakness. A particular problem is weakness
of the abductors in antegrade nailing, apparently due to the traumatic
effect of the nailing on the gluteus medius and maximus muscles as well
as a tendency to neglect those muscles during rehabilitation. Always
include abductor strengthening exercises in the rehabilitation program.
AUTHOR’S PERSPECTIVE
For closed fractures of the shaft of the femur from just
above the lesser trochanter to 2.5 cm proximal to the lateral
epicondyle, and in both closed and open fractures grades I through
IIIA, my preferred treatment is reamed locked nailing with an Alta
titanium intramedullary nail—using an 11 mm nail in women and a 12 mm
nail in men—which provides two transverse cross-locking screws
proximally and distally. For routine diaphyseal fractures, I place two
proximal transverse cross-locking screws proximally and distally, with
two proximal transverse cross-locking screws at the level of the lesser
trochanter.
In high subtrochanteric fractures, I move the rod more
proximally, up to 1 cm proximal to the tip of the greater trochanter,
and cross-lock into the femoral neck. In grade IIIB open fractures,
when I feel that initial intramedullary nailing is the best approach, I
use a nonreamed nail, planning conversion to a reamed nail as soon as
the risk of infection has passed and the soft-tissue envelope has
recovered. When I believe the risk of infection is too high, I use
external fixation.
For retrograde nailing through the intracondylar notch
of the knee, I prefer to use a standard Alta nail rather than a special
retrograde nail, and I always try to use a nail 10 mm or larger so that
the largest transverse cross-locking screws can be used. This is now my
treatment of choice for patients with severe multiple injuries in whom
rapid atraumatic fixation of the fracture in the supine position on a
regular radiolucent table is necessary. I also prefer this technique
for fixation of shaft fractures below a femoral neck fracture, using
independent fixation of the femoral neck with either multiple screws or
a sliding compression hip screw, depending on the configuration of the
fracture. I rarely plate a diaphyseal fracture of the femur, reserving
this technique for patients in whom the configuration of the femur or
absence of a medullary canal precludes intramedullary nailing, or in
whom a vascular repair has been done that is least threatened by plate
fixation as opposed to nailing. I reserve single-plane half-pin
external fixation for grade IIIB (and higher) open fractures and the
occasional polytrauma patient in whom retrograde nailing is not
possible.
B. PERIPROSTHETIC FRACTURES OF THE FEMUR
Fracture of the femur associated with a prosthetic implant is not unusual and is increasing in incidence (21,23,25,38,44,50,59,60,62,66,69,74). The second leading cause of revision hip arthroplasty at the Mayo Clinic is now periprosthetic fracture (44).
The overall prevalence is between 0.1% and 1.1% of total hip
prostheses. Periprosthetic fractures can occur acutely at the time of
initial implantation of the prosthesis or later. Factors that
predispose to periprosthetic fractures include the following:
  • Intraoperative creation of cracks, defects, or windows in the bone that are not bypassed with a stem of sufficient length
  • P.697


  • Extravasation of cement out of bone defects that results in an unhealed stress riser
  • Severe cortical erosion from a loose femoral component
  • Hip surgery prior to prosthetic replacement, because of weakening of the bone or because of holes left from previous implants
These fractures usually occur in elderly patients with
osteoporosis who have had total joint replacement of the hip or knee.
Because the bone is weak, and on one side of the fracture the medullary
canal is occupied by the prosthesis, internal fixation becomes a major
challenge. Many surgeons have recommended nonoperative treatment for
these fractures, but experience has shown that nonoperative treatment
leads to a high rate of local and systemic complications (38).
Local complications include fracture nonunion, femoral shortening,
femoral deformity, and prosthetic loosening. These complications
require major secondary surgery for correction. Prolonged bed rest and
immobilization can produce life-threatening complications. These risks
for nonoperative treatment, combined with the high likelihood of the
need for future revision arthroplasty, have led to a recommendation of
internal fixation for all displaced fractures associated with femoral
prostheses, whenever clinically possible.
CLASSIFICATION
Fractures of the femur associated with an implanted
femoral prosthesis can be classified on the basis of type and location
of prosthesis, the type and configuration of fracture line, and the
status of prosthesis fixation (i.e., whether it is stable or loose).
TYPE AND LOCATION OF PROSTHESIS
The following situations may be encountered:
Surface Replacement of the Proximal Femur
Fractures associated with surface replacements of the
hip usually occur at the femoral neck just distal to the prosthesis.
Treat by converting the surface replacement to a standard total hip
replacement arthroplasty. In the unusual case in which a fracture
occurs at or distal to the intertrochanteric region, you may treat it
by the usual operative methods for similar fractures not associated with a prosthesis.
Stemmed Replacement of the Proximal Femur
Stemmed prostheses of the proximal femur may be further
subdivided into those in which the femoral stem and cement, or the stem
alone occupies the entire medullary canal, and those in which the
femoral stem occupies only a portion of the medullary canal. In
fractures in which the entire medullary canal is filled with a
prosthetic stem, with or without cement, placement of a new prosthesis
with a longer stem provides intramedullary fixation (Fig. 20.14). An alternative method is to fix the fracture with a plate, supplemented with bone grafting (Fig. 20.15, Fig. 20.16).
Secure the plate with bicortical screws in the femur distal to the
prosthesis, and with unicortical or tangential screws proximally where
the medullary canal is filled by the prosthetic stem (Fig. 20.16). Another method of securing the plate proximally is to use cerclage wires or cable systems (Fig. 20.17).
If an uncemented stem occupies only a portion of the femoral canal, it
is sometimes possible to gain adequate fixation by passing small
flexible intramedullary rods, such as Ender or Rush rods, across the
fracture site and into the medullary canal adjacent to the prosthetic
stem (Fig. 20.18).
Figure 20.14. Use of a long-stemmed femoral prosthesis and cerclage wires to hold the reduction of a long oblique fracture.
Figure 20.15. Plate and screw fixation of a fracture at the tip of a proximal femoral prosthesis.
Figure 20.16.
Three possible methods of plate and screw placement to be used if the
intramedullary stem prevents central bicortical screw placement. The
goal is to obtain a solid purchase on eight cortices on each side of
the fracture.
Figure 20.17. Use of cerclage wires to secure fixation of a plate to the proximal femur segment.
Figure 20.18. A:
Flexible small intramedullary rods are used to internally fix fractures
of the femoral shaft distal to the tip of an uncemented proximal
femoral prosthesis that fills the entire canal. B: Possible intramedullary placement if the prosthesis stem is not cemented and does not fill the entire canal.
Femoral Prosthesis Associated with Knee-Replacement Arthroplasty
Fractures around total knee prosthesis can include
femoral prosthesis associated with knee replacement arthroplasty;
distal femoral surface replacement without a stem; cemented distal
femoral replacement with an intramedullary stem; and uncemented distal
femoral replacement with an intramedullary stem. Fractures of the
distal femur just proximal to a total knee replacement are most often
associated

P.698



P.699


with
anterior notching of the femur caused by poor placement of the cutting
guides during total knee arthroplasty. The notching causes weakening of
the femur just above the prosthesis, and delayed fracture propagating
from the stress riser is not an uncommon complication.

Distal femoral prostheses not associated with a femoral
stem are essentially surface replacement prostheses. In fractures that
occur proximal to a nonstemmed prosthesis, alternatives include a
condylar blade plate or compression screw with side plate screws (Fig. 20.19), retrograde nails (Fig. 20.20), and antegrade nails (Fig. 20.21).
In fractures associated with a distal femoral prosthesis that has an
intramedullary stem, a revision can be done using a longer stem (Fig. 20.22) or a plate-and-screw fixation with tangential placement of the screws to avoid the stem.
Figure 20.19. Blade plate and screw fixation above a distal femoral surface replacement.
Figure 20.20. Ender or Rush rod fixation above a distal femoral surface replacement.
Figure 20.21. Antegrade closed nail fixation of a supracondylar fracture of the femur proximal to a total knee prosthesis. A: Preoperative AP radiograph showing the fracture. B: Preoperative lateral radiograph. C:
Postoperative lateral radiograph showing fixation with an antegrade
Alta nail whose tip was cut off to allow more distal placement. D: AP radiograph.
Figure 20.22.
A custom long-stemmed distal femoral component used to internally fix a
comminuted distal femoral fracture above a short-stemmed distal femoral
prosthesis.
TYPE AND CONFIGURATION OF FRACTURE LINE
Proximal to or at the Tip of the Prosthetic Stem
The most common fractures of the femur associated with
femoral prostheses are transverse or short oblique fractures at, or
proximal to, the tip of the femoral stem

P.700



P.701



[Johansson type I (38)].
Long oblique or spiral fractures in this location are less common. The
fractures usually result from trauma at a stress riser created by the
transition between the stiff one-prosthesis segment and adjacent, less
stiff bone. If the fracture is transverse or short oblique, there is a
high risk of displacement, shortening, and nonunion. Internal fixation
is recommended. If the fracture has a spiral or long oblique course,
displacement may be minimal, and if there is no associated loosening of
the prosthesis, treatment can be nonoperative, consisting of traction
followed by fracture bracing and crutches.

Around the Prosthesis
For fractures to occur around the prosthesis, a portion
of the prosthesis must be loose. Johansson type II fractures occur
around the prosthesis and extend to beyond the distal tip of the
prosthesis with dislodgement of the stem from the distal canal. The
prosthesis may have been loose before injury, or the fracture itself
may break the prosthesis from its fixation to the surrounding bone. In
these situations, the fracture must be reduced and internally fixed,
and the prosthesis must be revised (Fig. 20.23).
Figure 20.23. Use of a long-stemmed proximal femoral prosthesis with cerclage wires to contain fragments around the stem.
At a Distance from the Prosthesis
Fractures that occur at a distance from the prosthesis
(Johansson type III) may be treated by plate fixation. However, the
interval of bone between the end of the plate and the tip of the
prosthetic stem is subject to high stresses and is a potential site of
future fracture.
STATUS OF PROSTHESIS FIXATION
If the prosthesis remains tight and well fixed to bone,
aim fracture management at reestablishing bony alignment and achieving
union. If the prosthesis is loose, fracture fixation is necessary, and
the prosthesis must be reimplanted with a longer stem and recemented
or, if a porous-coated long-stemmed prosthesis is used, bone-grafted.
Frequently, cerclage wires or cables are required to “gather in” bone
fragments or to oppose oblique fractures around the prosthesis (Fig. 20.23). Chandler and Tigges (21)
recommend reinforcing the fracture site with allograft when a proximal
femoral fracture occurs at or distal to a well-fixed femoral stem, or
proximal to a well-fixed knee replacement. They recommend the following
technique.
Fixation with Allograft Reinforcement
  • Expose the fracture, reduce it anatomically, and temporarily stabilize it with cerclage wires or cables (Fig. 20.24A, Fig. 20.24B).
    Figure 20.24. Allograft strut treatment for a proximal femoral fracture at the end of a prosthesis. A: Fracture of the femoral shaft just distal to the stem of a prosthesis. B: The fracture has been reduced and temporarily fixed. C: An allograft split into two equal pieces. D: Machining the endosteal surface of a graft. E: Placement of the two grafts on the femur. F: Use of hose clamps to secure the allografts to the femur. G: Passing the cerclage wires around the struts. H: Tightening the wires. I:
    The hose clamps have been removed, and the autologous graft has been
    applied. (Reproduced with permission from Chandler HP, Tigges RG. The
    Role of Allografts in the Treatment of Periprosthetic Femoral
    Fractures. In: Cannon WD, ed. Instructional Course Lectures, Vol. 47. Rosemont, IL: American Academy of Orthopaedic Surgeons, 1998.)
  • Obtain an allograft femoral or tibial
    shaft of appropriate size, harvest the diaphyseal portion, and split it
    into two equal medial and lateral shells in the sagittal plane to
    provide biological bone plates. Fresh frozen allografts are preferred,
    but freeze-dried allografts are also thought to be effective (Fig. 20.24C).
  • Machine the endosteal inside surfaces of
    the grafts with a high-speed burr to conform them to the outside
    contours of the recipient femur (Fig. 20.24D).
  • Place these on the medial and lateral
    sides of the femur, preserving the soft-tissue attachments to the linea
    aspera, which contain the extraosseous blood supply to the femur. Make
    the struts long enough to extend at least 10 cm beyond the fracture
    both proximally and distally (Fig. 20.24E).
  • Apply the allograft struts to the fracture, and compress them into place with commercial hose clamps (Fig. 20.24F).
  • With a wire passer, pass double #16 stainless-steel cerclage wires around the struts at 2–4 cm intervals (Fig. 20.24G).
  • Tighten the wires with a Harris wire tightener (Johnson & Johnson, Raynham, MA) (Fig. 20.24H). When the allograft has been secured with the wires, remove the hose clamps.
  • Graft the fracture with autologous cancellous bone from the iliac crest to hasten healing (Fig. 20.24I).
Use protected weight bearing to 60 pounds (27 kg) until 12 weeks, then advance weight bearing.
For femoral fractures distal to a well-fixed femoral stem

P.702



P.703



or proximal to a well-fixed knee replacement, Chandler and Tigges (21)
recommend retrograde intramedullary nails—assuming that the knee
prosthesis permits insertion of the nail, and recognizing that in total
hip replacement there is the potential for a stress riser between the
proximal end of the medullary nail and the distal end of the stem of
the prosthesis (21,56,65,70).

Where rods cannot be used, plate fixation is indicated.
Pack autologous cancellous bone around the fracture site. However,
comminution and the osteopenic bone frequently found in this group of
patients result in poor screw fixation. To circumvent this problem,
harvest an allograft from either a tibia or a femur (probably better
from a comparable femur). Chandler and Tigges (21)
back up the plate on the opposite cortex with an allograft strut
harvested from the identical location on the allograft femur so that
the strut fits the host femur accurately (Fig. 20.25, Fig. 20.26). Their technique includes the following.
Figure 20.25.
An allograft shell placed on the opposite cortex to reinforce the
fixation and provide a structural bone graft. (From Chandler HP, Tigges
RG. The Role of Allografts in the Treatment of Periprosthetic Femoral
Fractures. J Bone Joint Surg Am 1997;79:1422.)
Figure 20.26.
The use of an allograft in combination with a buttress plate to
reconstruct a comminuted supracondylar fracture of the femur
immediately above a total knee prosthesis in an elderly patient whose
bone quality precluded effective fixation with the screws.
  • Expose the distal femur and total knee through a medial parapatellar incision.
  • Reduce the fracture, and temporarily hold
    it reduced with cerclage wires or a K-wire while the plate is fitted to
    one cortex and the allograft strut to the opposite cortex.
  • Stabilize the plate and strut to the host
    femur with cerclage cables or wires in the area of an intramedullary
    stem, and use screws elsewhere (Fig. 20.17).
Their postoperative care is the same as that already described.
THE VANCOUVER SYSTEM
Duncan and Masri (28) have modified Johansson’s (38) classification system to create the Vancouver System. Their classifications for fractures about a total hip prosthesis follow.
  • Type A: Trochanteric fractures
  • Type B: Fractures about the stem or tip
    • B1: Well-fixed stem
    • B2: Loose stem
    • B3: Associated bony deficiency or destruction
  • Type C: Fractures distal to the tip
Lewallen and Berry’s (44) recommendations for these include the following:
  • Type A. Treat simple avulsion fractures
    of the greater or lesser trochanter in the presence of a well-fixed
    stem nonoperatively. For severe periprosthetic osteolysis, replace the
    component that is the source of the microparticles.
  • P.704


  • Type B1. These usually can be treated with internal fixation.
  • Type B2. Revise the loose component.
  • Type B3. Revise the component.
  • Type C. Internal fixation usually suffices.
Lewallen and Berry (44)
emphasize that in all types of fractures associated with a loose
femoral component, the best treatment is revision unless medical
problems or advanced age contraindicates this extensive surgery.
SURGICAL TECHNIQUES
Femur fractures in which a prosthesis is in place are
long and difficult surgical cases requiring excellent support and
careful preoperative planning. If possible, have autogenous blood and a
cell saver available. A laminar flow unit or ultraviolet radiation may
reduce the infection rate.
  • Expose the proximal femur according to
    your preference. If the proximal femoral prosthesis is to be revised,
    trochanteric osteotomy greatly facilitates exposure and removal of the
    prosthesis and cement. Extend the exposure distally to expose the
    femoral shaft and the full extent of the fracture. This facilitates
    fracture reduction, fixation, cementing, packing of bone graft, and
    passage of the prosthetic stem across the fracture under direct vision.
  • Occasionally, longitudinal splitting of
    the proximal fragment by the fracture at the time of surgery can prove
    to be an advantage, and by opening the proximal femoral segment “like a
    book,” you can easily remove the prosthesis and cement. Repair the
    split proximal fragment with cerclage wires or cables.
  • It is best to approach distal femoral
    prostheses associated with fractures by the anteromedial parapatellar
    incision if the prosthesis is to be revised. If the prosthesis is tight
    and there is enough bone stock proximal to the prosthesis to use
    standard means of internal fixation, you may use lateral or combined
    medial and lateral incisions over the distal femur for a blade plate or
    condylar medullary rods.
  • Occasionally, fractures proximal to a
    total knee prosthesis can be managed by closed reduction on a fracture
    table and either antegrade or retrograde fixation with a locked
    medullary nail, utilizing closed technique (Fig. 20.20, Fig. 20.21).
  • Since the procedure to replace the
    femoral prosthesis in the presence of a fracture is essentially a
    revision, the technique is very similar to revision arthroplasty, with
    the exception of needing to address the fracture. The technical details
    of revision arthroplasty for the hip are addressed in Chapter 106, and those for the fracture about the total knee, in Chapter 109. Specifics about the fixation have already been addressed in the preceding descriptions.
POSTOPERATIVE CARE AND REHABILITATION
Ideally, fracture fixation must be sufficiently solid so
that the patient can be out of bed and ambulatory by the 4th or 5th
postoperative day. In patients with good bone stock and strong
intramedullary fixation, weight bearing can progress as with any
fracture fixed with an intramedullary device. For patients who have had
a trochanteric osteotomy and reattachment, use at least 6 weeks of
touchdown weight bearing to avoid displacement of the trochanter. If
the femoral prosthesis has been exchanged, the usual postoperative
precautions associated with joint replacement to prevent dislocation
and other complications are necessary.
RESULTS
Johansson et al. (38) reported
on 35 patients with 37 periprosthetic fractures following total hip
replacement after a follow up averaging 3.9 years (range, 2–9 years).
Their

P.705


results
were dismal, with unsatisfactory outcomes in 22 fractures (59%). The
most common problem was prosthetic loosening, which occurred in ten,
followed by malunion with loosening in three, nonunion in three,
massive heterotopic bone in one, infection in three, and two
unsatisfactory outcomes in primary Girdlestone resections. Seven
immediate postoperative fractures that were treated in traction all had
unsatisfactory results. Johansson et al. had a total of 30
complications (60%) in their 22 patients. Improvement in techniques to
handle these difficult problems have shown markedly improved results.
Chandler et al. (20)
in 1993 achieved anatomic union in 100% of periprosthetic fractures
treated with their technique of onlay grafting, with all returning to
their functional status prior to fracture. In a later report in 1996 (19), using the combination of a metal plate and an allograft strut, they achieved anatomic union in 21 of 22 patients.

In a recent series from the Mayo Clinic, Lewallen and Berry (44)
analyzed outcomes in revision total hip arthroplasty for 97
postoperative periprosthetic femoral fractures in 94 patients, with an
average followup of 5 years (range, 2–14 years). Eighty-five percent of
the fractures healed, but stable long-term fixation of the femoral stem
was achieved in less then 50% of the hips. Overall, 20% had only
minimal symptoms in spite of the radiographic evidence of loosening;
however, 33% had loosening with severe pain. Use of a proximally
porosis-coated implant without cement was associated with the poorest
results. An important observation in their series was the occurrence of
a spontaneous fracture, in the absence of a traumatic event, in more
the 50% of their patients who had Vancouver type B3 fractures in which
the bone loss and comminution were severe. This points out that a
periprosthetic fracture is a very serious occurrence in the typical
elderly patient with a total hip or total knee arthroplasty. Although
nonoperative care for minimally displaced acute periprosthetic
fractures in traction has reasonably good outcomes, nearly all of these
require major surgery, with significant risks and a high rate of
complications. Although much progress has been made toward solving this
problem in the last 20 years, more is needed.
PITFALLS AND COMPLICATIONS
Operative complications of treating fractures associated
with femoral prostheses usually stem from difficulties of cement and
prosthesis removal or inadequate preoperative planning. A variety of
prosthetic alternatives, including cemented and uncemented devices of
various stem sizes and lengths, must be available to deal with
unexpected technical problems.
Always have available a long-stemmed power burr and
intramedullary reamers and curets for removal of cement. It is
important to know the techniques for removal of broken hardware from
the medullary canal, including the use of an image intensifier, to
avoid cortical perforation during reaming. Preoperative planning should
include careful silhouetting of the femur and fracture with appropriate
templates to determine prosthesis size and to match components for the
revision. Appropriate plates, screws, and intramedullary rods must be
available. It is not uncommon to change techniques from those planned
because of intraoperative findings at the time of surgery. Multiple
options must be open to achieve optimal fixation.
Postoperative complications usually arise from
inadequate prosthetic fixation or refracture. If cement fixation is
insufficient, the prosthesis will become loose, particularly in
patients in whom the original prosthesis loosened because of
periprosthetic cortical lysis. In patients in whom a prosthesis is left
in place and plates and screws or cerclage techniques are used to
stabilize the fracture, fixation is always less then ideal. Take care
to delay weight bearing until the fracture has united, because
excessive and premature weight bearing may lead to loss of fixation,
displacement, and nonunion, requiring reoperation. Patients who have a
stress riser between the end of the prosthesis and an adjacent plate
are always at risk of refracture and must be cautioned to avoid trauma
to the leg.
ACKNOWLEDGEMENT
This chapter in the second edition was written by
Sigvard T. Hansen, Jr., and David W. Lhowe. Portions of their original,
excellent chapter are included here.
REFERENCES
Each reference is categorized according to the following
scheme: *, classic article; #, review article; !, basic research
article; and +, clinical results/outcome study.
+ 1. Alonzo J, Geissler W, Hughes JL. External Fixation of Femoral Fractures. Indications and Limitations. Clin Orthop 1989;241:83.
+ 2. Astion
DJ, Wilber JH Scoles PV. Avascular Necrosis of the Capital Femoral
Epiphysis after Intramedullary Nailing for a Fracture of the Femoral
Shaft. A Case Report. J Bone Joint Surg Am 1995;77:1092.
+ 3. Bain GI, Zacest AC, Paterson DC, et al. Abduction Strength Following Intramedullary Nailing of the Femur. J Orthop Trauma 1997;11:93.

P.706


+ 4. Baixauli
F Sr, Baixauli EJ, Sanchez-Alepuz E, Baixauli F Jr. Interlocked
Intramedullary Nailing for Treatment of Open Femoral Shaft Fractures. Clin Orthop 1998;350:67.
! 5. Bankston AB, Keating EM, Saha S. The Biomechanical Evaluation of Intramedullary Nails in Distal Femoral Shaft Fractures. Clin Orthop 1992;276:277.
+ 6. Baumgaertel F, Dahlen C, Stiletto R, Gotzen L. Technique of Using the AO-Femoral Distractor for Femoral Intramedullary Nailing. J Orthop Trauma 1994;8:315.
+ 7. Bostman
O, Varjonen L, Vainionpaa S, et al. Incidence of Local Complications
after Intramedullary Nailing and after Plate Fixation of the Femoral
Shaft Fractures. J Trauma 1989;29:639.
+ 8. Browner B. Pitfalls, Errors, and Complications in the Use of Locking Küntscher Nails. Clin Orthop 1986;212:192.
+ 9. Brumback RJ, Ellison TS, Molligan H, et al. Pudenal Nerve Palsy Complicating Intramedullary Nailing of the Femur. J Bone Joint Surg Am 1992;74:1450.
+ 10. Brumback RJ, Ellison PS, Poka A, et al. Intramedullary Nailing of Open Fractures of the Femoral Shaft. J Bone Joint Surg Am 1989;71:1324.
+ 11. Brumback
RJ, Reilly JP, Poka A, et al. Intramedullary Nailing of Femoral Shaft
Fractures. Part I: Decision-making Errors with Interlocking Fixation. J Bone Joint Surg Am 1988;70:1441.
+ 12. Brumback
RJ, Uwagie-Ero S, Lakatos RP, et al. Intramedullary Nailing of Femoral
Shaft Fractures. Part II: Fracture-healing with Static Interlocking
Fixation. J Bone Joint Surg Am 1988;70:1452.
+ 13. Brumback
RJ, Wells JD, Lakatos R, et al. Heterotopic Ossification about the Hip
after Intramedullary Nailing for Fractures of the Femur. J Bone Joint Surg Am 1990;72:1067.
# 14. Bucholz RW, Jones A. Current Concepts Review, Fractures of the Shaft of the Femur. J Bone Joint Surg Am 1991;73:1561.
+ 15. Bucholz
RW, Ross RW, Lawrence KL. Fatigue Fracture of the Interlocking Nail in
the Treatment of Fractures of the Distal Part of the Femoral Shaft. J Bone Joint Surg Am 1987;69:1391.
+ 16. Butler
MS, Brumback RJ, Ellison TS, et al. Interlocking Intramedullary Nailing
for Ipsilateral Fractures of the Femoral Shaft and Distal Part of the
Femur. J Bone Joint Surg Am 1991;73:1492.
+ 17. Cameron CD, Meek RN, Blachut PA, et al. Intramedullary Nailing of the Femoral Shaft: A Prospective, Randomized Study. J Orthop Trauma 1992;6:448.
+ 18. Casey MJ, Chapman MW. Ipsilateral Concomitant Fractures of the Hip and Femoral Shaft. J Bone Joint Surg Am 1979;61:502.
+ 19. Chandler
HP, Danylchuk K. Treatment of Distal Femoral Fractures Proximal to a
Total Knee Replacement or Distal to the Stem of a Total Hip Replacement
in Osteopenic Patients. Presented as a poster exhibit at the annual
meeting of the American Academy of Orthopaedic Surgeons, Atlanta,
Georgia, February 25, 1996.
+ 20. Chandler
HP, King D, Limbird R, et al. The Use of Cortical Allograft Struts for
Fixation of Fractures Associated with Well-fixed Total Joint
Prostheses. Semin Arthroplasty 1993;4:99.
# 21. Chandler
HP, Tigges RG. The Role of Allografts in the Treatment of
Periprosthetic Femoral Fractures. An Instructional Course Lecture. J Bone Joint Surg Am 1997;79:1422.
+ 22. Chapman MW. Closed Intramedullary Bone Grafting and Nailing of Segmental Defects of the Femur. J Bone Joint Surg Am 1980;62:1004.
+ 23. Clancy GJ, Smith RF, Madenwald MB. Fracture of the Distal End of the Femur below Hip Implants in Elderly Patients. J Bone Joint Surg Am 1983;75:491.
+ 24. Clatworthy MD, Clark DI, Gray DH, Hardy AE. Reamed versus Unreamed Femoral Nails. A Randomized, Prospective Trail. J Bone Joint Surg Br 1998;80:485.
+ 25. Culp RW, Schmidt RG, Hanks G, et al. Supracondylar Fracture of the Femur following Prosthetic Knee Arthroplasty. Clin Orthop 1987;222:112.
+ 26. Dabezies EJ, D’Ambrosia R, Shuji H, et al. Fractures of the Femoral Shaft Treated by External Fixation with the Wagner Device. J Bone Joint Surg Am 1984;66:360.
+ 27. De Bastiani G, Aldegheri R, Renzi Brivio L. The Treatment of Fractures with a Dynamic Axial Fixator. J Bone Joint Surg Br 1984;66:538.
# 28. Duncan CP, Masri BA. Fractures of the Femur after Hip Replacement. Instr Course Lect 1995;44:293.
! 29. ElMaraghy
AW, Schemitsch EH, Richards RR. Femoral and Cruciate Blood Flow after
Retrograde Femoral Reaming: A Canine Study Using Laser Doppler
Flowmetry. J Orthop Trauma 1998;12:253.
+ 30. Gellman
RE, Paiement GD, Green HD, Coughlin RR. Treatment of Supracondylar
Femoral Fractures with a Retrograde Intramedullary Nail. Clin Orthop 1996;332:90.
+ 31. George
CJ, Lindsey RW, Noble PC, et al. Optimal Location of a Single Distal
Interlocking Screw in Intramedullary Nailing of Distal Third Femoral
Shaft Fractures. J Orthop Trauma 1998;12:267.
+ 32. Gregory
P, DiCicco J, Karpik K, et al. Ipsilateral Fractures of the Femur and
Tibia: Treatment with Retrograde Femoral Nailing and Unreamed Tibial
Nailing. J Orthop Trauma 1996;10:309.
# 33. Gustilo RB, Anderson JT. Prevention of Infection in the Treatment of 1025 Open Fractures of Long Bones. J Bone Joint Surg Am 1976;58:452.
! 34. Hajek
PD, Bicknell HR Jr, Bronson WE, et al. The Use of One Compared with Two
Distal Screws in the Treatment of Femoral Shaft Fractures with
Interlocking Intramedullary Nailing: A Clinical and Biomechanical
Analysis. J Bone Joint Surg Am 1993;75:519.
+ 35. Herscovici D Jr, Whiteman KW. Retrograde Nailing of the Femur Using an Intercondylar Approach. Clin Orthop 1996;332:98.
# 36. Hughes JL. External Skeletal Fixation. Part V. Wagner Apparatus: A Portable Traction Device for the Femur. Instr Course Lect 1984;33:158.
+ 37. Janzing
HMJ, Stockman B, Van Damme G, et al. The Retrograde Intramedullary
Nail: Prospective Experience in Patients Older Than Sixty-five Years. J Orthop Trauma 1998;12:330.

P.707


+ 38. Johansson JE, McBroom R, Barrington TW, Hunter GA. Fracture of the Ipsilateral Femur in Patients with Total Hip Replacement. J Bone Joint Surg Am 1981;63:1435.
+ 39. Johnson
KD, Tencer AF, Blumenthal S, et al. Biomechanical Performance of Locked
Intramedullary Nails in Comminuted Femoral Shaft Fractures. Clin Orthop 1986;206:151.
+ 40. Jupiter JB, Bour CJ, May JW. The Reconstruction of Defects in the Femoral Shaft with Vascularized Transfers of Fibular Bone. J Bone Joint Surg Am 1987;69:365.
+ 41. Krettek C, Miclau T, Blauth M, et al. Recurrent Rotational Deformity of the Femur after Static Locking of Intramedullary Nails. J Bone Joint Surg Am 1997;79:4.
! 42. Kyle
RF, Schaffhausen JM, Bechtold JE. Biomechanical Characteristics of
Interlocking Femoral Nails in the Treatment of Complex Femoral
Fractures. Clin Orthop 1991;267:169.
+ 43. Levin PE, Schoen RW, Browner BD. Radiation Exposure to the Surgeon During Closed Locked Intramedullary Nailing. J Bone Joint Surg Am 1987;69:761.
# 44. Lewallen DG, Berry DJ. Periprosthetic Fracture of the Femur after Total Hip Arthroplasty. In Instructional Course Lectures. J Bone Joint Surg Am 1997;9:1881.
+ 45. Lhowe DW, Hansen ST. Immediate Nailing of Open Fractures of the Femoral Shaft. J Bone Joint Surg Am 1988;70:812.
+ 46. Lucas
SE, Seligson D, Henry SL. Intramedullary Supracondylar Nailing of
Femoral Fractures. A Preliminary Report of the GSH Supracondylar Nail. Clin Orthop 1993;296:200.
+ 47. Magerl F, Wyss A, Brunner C, Binder W. Plate Osteosynthesis of Femoral Shaft Fractures in Adults. Clin Orthop 1979;138:62.
# 48. Mast J, Jakob R, Ganz R. Planning and Reduction Technique in Fracture Surgery. Berlin: Springer-Verlag, 1989.
# 49. Mays J, Neufelf AJ. Skeletal Traction Methods. Clin Orthop 1975;102:141.
+ 50. McLauchlan
GJ, Robinson CM, Singer BR, Christie J. Results of an Operative Policy
in the Treatment of Periprosthetic Femoral Fracture. J Orthop Trauma 1997;11:170.
+ 51. Meggitt BF, Juett DA, Smith JD. Cast-bracing for Fractures of the Femoral Shaft. A Biomechanical and Clinical Study. J Bone Joint Surg Br 1981;63:12.
+ 52. Moed
BR, Watson JT. Retrograde Intramedullary Nailing, without Reaming, of
Fractures of the Femoral Shaft in Multiply Injured Patients. J Bone Joint Surg Am 1995;77:1520.
+ 53. Moed BR, Watson JT, Cramer KE, et al. Unreamed Retrograde Intramedullary Nailing of Fractures of the Femoral Shaft. J Orthop Trauma 1998;12:334.
+ 54. Morgan E, Ostrum RF, DiCicco J, et al. Effects of Retrograde Femoral Intramedullary Nailing on the Patellofemoral Articulation. J Orthop Trauma 1999;12:12.
# 55. Muüller ME. Manual of Internal Fixation, 3rd ed. New York: Springer-Verlag, 1990.
+ 56. Murrell
GA, Nunley JA. Interlocked Supracondylar Intramedullary Nails for
Supracondylar Fractures after Total Knee Arthroplasty. A New Treatment
Method. J Arthroplasty 1995;10:37.
# 57. Orthopaedic Trauma Association: Fracture and Dislocation Compendium. J Orthop Trauma 1996;1(suppl 10):31.
+ 58. Ostrum RF, DiCicco J, Lakatos R, Poka A. Retrograde Intramedullary Nailing of Femoral Diaphyseal Fractures. J Orthop Trauma 1998;12:464.
+ 59. Pankovich AM, Tarabishy I, Barmada R. Fractures below Non-cemented Femoral Implants, Treatment with Ender Nailing. J Bone Joint Surg Am 1981;63:1024.
+ 60. Parrish TF, Jones JR. Fracture of the Femur following Prosthetic Arthroplasty of the Hip: Report of Nine Cases. J Bone Joint Surg Am 1964;46:241.
+ 61. Patterson BM, Routt ML Jr, Benirschke SK, Hansen ST Jr. Retrograde Femoral Nailing: A Procedure for Severe Injuries. J Orthop Trauma 1993;7:155.
+ 62. Ponzer S, Tidermark J, Toörnkvist H. Retrograde Nailing of Femoral Fractures Distal to a Moore Prosthesis. J Orthop Trauma 1998;12:588.
+ 63. Riina
J, Tornetta P III, Ritter C, Geller J. Neurologic and Vascular
Structures at Risk during Anterior-posterior Locking of Retrograde
Femoral Nails. J Orthop Trauma 1998;12:379.
+ 64. Riquelme
AG, Rodriguez AJ, Mino GL, Sanmartin RM. Treatment of the Femoral and
Tibial Fractures with Grosse and Kempf Locking Nails. Clin Orthop 1992;283:86.
+ 65. Ritter MA, Keating EM, Faris PM, Meding JB. Rush Rod Fixation of Supracondylar Fractures above Total Knee Arthroplasties. J Arthroplasty 1995;10:213.
+ 66. Roffman M, Mendes DC. Fractures of the Femur after Total Hip Arthroplasty. Orthopaedics 1989;12:1067.
+ 67. Ruedi T, Luscher JN. Results after Internal Fixation of Comminuted Fractures of the Femoral Shaft with DC Plates. Clin Orthop 1979;138:74.
+ 68. Sanders R, Koval KJ, DiPasquale T, et al. Retrograde Reamed Femoral Nailing. J Orthop Trauma 1993;7:293.
+ 69. Scott RD, Turner RH, Leitzes SM, Aufranc OA. Femoral Fractures in Conjunction with Total Hip Replacement. J Bone Joint Surg Am 1975;57:494.
+ 70. Smith
WJ, Martin SL, Mabrey JD. Use of Supracondylar Nail for Treatment of a
Supracondylar Fracture of the Femur following Total Knee Arthroplasty. J Arthroplasty 1996;11:210.
+ 71. Swiontkowski MF, Hansen ST, Kellam J. Ipsilateral Fractures of the Femoral Neck and Shaft. A Treatment Protocol. J Bone Joint Surg Am 1984;66:260.
+ 72. Tornetta
P III, Tiburzi D. The Treatment of Femoral Shaft Fractures Using
Intramedullary Interlocked Nails with and without Intramedullary
Reaming: A Preliminary Report. J Orthop Trauma 1997;11:89.
# 73. Tscherne H, Gotzen L, eds. Fractures with Soft Tissue Injuries. Berlin: Springer-Verlag, 1984.
+ 74. Wang G, Miller TO, Stamp WG. Femoral Fracture following Hip Arthroplasty. J Bone Joint Surg Am 1985;67:956.

P.708


+ 75. Williams MM, Askins V, Hinkes EW, Zycy GA. Primary Reamed Intramedullary Nailing of Open Femoral Shaft Fractures. Clin Orthop 1995;318:182.
+ 76. Winquist R, Clawson DK, Hansen ST. Closed Intramedullary Nailing of Femoral Fractures. J Bone Joint Surg Am 1984;66:529.
+ 77. Wiss DA, Brien WW, Becker V Jr. Interlocking Nailing for the Treatment of Femoral Fractures Due to Gunshot Wounds. J Bone Joint Surg Am 1991;73:598.
+ 78. Wiss DA, Brien WW, Stetson WB. Interlocked Nailing for Treatment of Segmental Fractures of the Femur. J Bone Joint Surg Am 1990;72:724.
+ 79. Wolinsky
PR, McCarty EC, Shyr Y, Johnson KD. Length of Operative Procedures:
Reamed Femoral Intramedullary Nailing Performed with and without a
Fractures Table. J Orthop Trauma 1998;12:485.
+ 80. Yang KH, Han DY, Jahng JS, et al. Prevention of Malrotation Deformity in Femoral Shaft Fracture. J Orthop Trauma 1998;12:558.

This website uses cookies to improve your experience. We'll assume you're ok with this, but you can opt-out if you wish. Accept Read More