Extract
Periprosthetic
patellar fractures may be the most common fractures complicating total knee
arthroplasty. Clinical and
radiographic examinations should be focused on determining the integrity of
the extensor mechanism and the fixation of the patellar component. The etiology of
periprosthetic patellar fractures is multifactorial and may be classified on
the basis of intraoperative or postoperative factors. Minimally displaced
fractures with an intact extensor mechanism and patellar component are best
treated nonoperatively with a short period of immobilization. Operative treatment of
periprosthetic patellar fractures often yields poor results with high
complication rates and little functional improvement. Operative intervention
should be reserved for fractures associated with extensor mechanism
dysfunction and patellar component loosening.
Periprosthetic
patellar fractures may be the most common fractures complicating total knee
arthroplasty. Clinical and
radiographic examinations should be focused on determining the integrity of
the extensor mechanism and the fixation of the patellar component. The etiology of
periprosthetic patellar fractures is multifactorial and may be classified on
the basis of intraoperative or postoperative factors. Minimally displaced
fractures with an intact extensor mechanism and patellar component are best
treated nonoperatively with a short period of immobilization. Operative treatment of
periprosthetic patellar fractures often yields poor results with high
complication rates and little functional improvement. Operative intervention
should be reserved for fractures associated with extensor mechanism
dysfunction and patellar component loosening.
Periprosthetic
patellar fractures may be the most common fractures complicating total knee
arthroplasty.
Clinical and
radiographic examinations should be focused on determining the integrity of
the extensor mechanism and the fixation of the patellar component.
The etiology of
periprosthetic patellar fractures is multifactorial and may be classified on
the basis of intraoperative or postoperative factors.
Minimally displaced
fractures with an intact extensor mechanism and patellar component are best
treated nonoperatively with a short period of immobilization.
Operative treatment of
periprosthetic patellar fractures often yields poor results with high
complication rates and little functional improvement. Operative intervention
should be reserved for fractures associated with extensor mechanism
dysfunction and patellar component loosening.
The patella is the largest sesamoid bone in the skeleton. Located within an
expansion of the quadriceps tendon, the patella allows an increased functional
lever arm of the quadriceps and enhances the mechanical advantage of the
extensor mechanism of the knee. In addition, it provides an articulating
surface with a low coefficient of friction, protects the native and prosthetic
knee from trauma, protects the quadriceps tendon and extensor mechanism from
frictional irritation, and affects the cosmetic appearance of the
knee1.
Because of its biomechanical importance, any problems involving the patella
or the patellar component of a total knee prosthesis can have a substantial
effect on overall knee function. In fact, patellar complications following
total knee arthroplasty have been a well-documented source of discomfort and
disability2-6.
Although infrequent, periprosthetic fractures of the patella remain a
challenge for even the most experienced joint reconstruction surgeons. This is
largely due to the discouraging results that are common following the
treatment of all but nondisplaced patellar fractures. Even with meticulous
anatomic fracture reduction, healing, and reconstitution of the extensor
mechanism, return to prefracture function is
rare7.
Most studies on periprosthetic patellar fractures have involved a small
number of patients, have lacked discrimination between intraoperative and
postoperative fractures and between those associated with revisions and those
associated with primary total knee arthroplasties, have not been consistent
with regard to follow-up assessment measures, have failed to identify the
pretreatment integrity of the patellar bone, and have involved use of
disparate treatment techniques. This list of confounders is further confused
by a lack of a standardized classification system for these fractures, a
standardized rating system for assessing knee function and predicting
prognosis following treatment, and a reliable management strategy. In this
article, we discuss periprosthetic patellar fractures after total knee
arthroplasty, review the classification systems, and put into perspective the
treatment algorithms that will potentially optimize the outcome of management
of this complication.
Periprosthetic patellar fractures may be the most common fractures
complicating total knee
arthroplasty7. The
reported prevalence ranges from as low as 0.11% to as high as
21.4%8,9.
According to epidemiological data in the Mayo Clinic Joint Registry, a
periprosthetic patellar fracture occurred in association with 0.68% of 12,000
primary total knee arthroplasties over a thirteen-year
period10. As a
result of the variability in the prevalence of this fracture, values reported
in the literature should be used only as a guideline to estimate the
prevalence of periprosthetic patellar fractures
(Table I).
In published series of patellar fractures, there is considerable
variability with regard to implant type, treatment of the posterior cruciate
ligament, prevalence of lateral releases performed, use of cement, inclusion
of revisions, and prevalence and method of patellar resurfacing, all of which
can independently affect the fracture prevalence and pattern.
Periprosthetic patellar fractures are more likely to occur postoperatively
than intraoperatively, and they are more frequent after revision total knee
arthroplasties7,8,11.
Berry11 reported
that the rate of postoperative patellar fracture after revision total knee
arthroplasty (1.8%) was more than double that after primary total knee
arthroplasty (0.7%) and was nine times higher than the rate of intraoperative
fracture during revision total knee arthroplasty (0.2%)
(Table
II)11.
Periprosthetic patellar fractures are typically diagnosed on the basis of a
combination of the history, physical examination, and plain radiographs. While
there may be a history of trauma, there is often no episode preceding the
pain. Patients may report anterior knee pain, especially with certain
activities such as ascending or descending stairs.
Patients may not show the signs or symptoms typically seen with acute
traumatic patellar fractures in the native knee and they may have no symptoms
at all (Fig. 1).
Bourne12 reported
that, in his series, three of four patients with a vertical, laterally based
periprosthetic fracture without disruption of the extensor mechanism were
asymptomatic. A majority of these fractures are discovered incidentally on
radiographs during a routine postoperative follow-up examination. More than
80% of patients with a periprosthetic patellar fracture seen by Tria et
al.3 (nine knees)
and Insall et al.13
(eight knees) were asymptomatic and were diagnosed on the basis of routine
follow-up radiographs. In a series of eighty-five periprosthetic patellar
fractures, 44% caused either minimal or no symptoms at the time of
diagnosis10.
When patients are symptomatic, the classic clinical manifestation is
anterior knee pain with patellar tenderness on direct palpation. There may be
an effusion as well as extensor weakness, instability, or difficulty ascending
or descending stairs, with a fear of
falling14. With
this scenario, it is necessary to evaluate the knee with formal
anteroposterior, lateral, and skyline radiographs
(Figs. 2-A and 2-B). Although
most fractures are diagnosed with use of radiographs, a technetium-99m bone
scan may be employed to determine whether a fracture is old or new or to
diagnose occult fractures. Old asymptomatic fractures do not require
treatment; however, it is important to realize that bone scans may be positive
for up to two years following a periprosthetic patellar
fracture14.
Classification systems in orthopaedic surgery allow effective communication
between clinicians, help them to make management decisions, provide insight
regarding projected outcomes, and assist in the advancement of clinical
research. Ultimately, a universal system for categorizing injuries must focus
on the variables that influence treatment strategy. Important components
affecting the management of periprosthetic patellar fractures include the
degree of fracture fragment displacement, the stability of the fixation of the
patellar prosthesis, the location and pattern of the fracture, the integrity
of the extensor mechanism, and the quality and vascularity of the remaining
bone stock.
Presently, there is no universally accepted validated classification system
that can provide functional outcome measures or be used as an adjunct to
clinical treatment algorithms. Several classification systems, involving use
of disparate radiographic and clinical variables, have been utilized in an
attempt to determine the appropriate treatment and predict
prognosis7,10,15,16.
Several predisposing factors may increase the risk of a periprosthetic
patellar fracture occurring during or after a total knee arthroplasty. It is
important to classify these factors on a temporal basis (intraoperative or
postoperative) in the setting of primary or revision total knee
arthroplasty.
The risk of a patellar fracture in association with a total knee
arthroplasty is lower intraoperatively than it is in the postoperative period.
Predisposing intraoperative factors include overzealous clamping of the
patella during the resurfacing, overreaming of the patella, slippage of the
reamer, aggressive bone resection with <10 to 15 mm of patellar bone stock
left remaining, thermal injury and bone necrosis due to polymethylmethacrylate
cement, and revision of the patellar component, particularly in a patient with
poor bone stock (Figs. 2-A and
2-B)7,8,12,17,18.
Trauma
Traumatic causes are categorized as direct or indirect on the basis of the
mechanism by which the patella fractures. Atraumatic causes are more numerous
and are subcategorized as fatigue/stress, events leading to a dysvascular
patella, and other etiologies such as patient, implant, and technical factors.
Periprosthetic patellar fractures can result from direct trauma such as a fall
onto the knee or a dashboard injury with a direct impact on the patella.
Alternatively, the fracture can be caused by an indirect mechanism such as an
eccentric quadriceps muscle contraction. With traumatic fractures, the history
of injury and symptoms can be clearer than those with fractures of atraumatic
etiology.
Anatomic Factors
An important consideration when determining the etiology and treatment of
periprosthetic patellar fractures is the integrity of the remaining patellar
bone stock after the total knee arthroplasty. When the remaining patella is
<10 mm thick, a new patellar component should not be
implanted19.
Additionally, because of the risk of intraoperative fracture, it is not
advisable to revise an unworn, well-positioned all-polyethylene patellar
component during a revision total knee
arthroplasty13,20.
Patellar fractures can occur as a result of fatigue failure of the remaining
patellar bone stock, particularly in the setting of osteonecrosis
(Fig. 3). This weakness in the
bone occurs as a result of the degree of patellar resection done for patellar
resurfacing at the time of the index procedure and secondarily as a result of
the altered stresses on the remaining patella imparted by the patellar
prosthesis or during
revision10,14,21.
Technical Factors
Several authors have evaluated devascularization of the patella during the
index procedure as a possible etiology of patellar fractures after total knee
arthroplasty, although a clear link has not been absolutely established. The
vascular supply to the patella may be disrupted secondary to lateral
retinacular release for the treatment of patellar maltracking, resection of
the infrapatellar fat
pad22, and the
standard median parapatellar arthrotomy.
Sacrifice of a branch of the superior lateral geniculate artery during
lateral release has been reported to be a risk factor for patellar
osteonecrosis and subsequent patellar fracture following total knee
arthroplasty. Scuderi et
al.23 reported a
lack of isotope uptake in nine of sixteen knees in which a lateral release had
been performed at the time of a primary total knee arthroplasty compared with
only three of twenty knees treated without a lateral release. While this
finding demonstrated a correlation between lateral release and loss of the
blood supply to the patella, it was clinically unimportant, with only one
patellar fracture occurring in that series. Tria et
al.3 reported
eighteen patellar fractures after 504 primary total knee arthroplasties. All
patients with a patellar fracture had undergone a lateral release at the time
of the index procedure, again raising the question of whether lateral release
was associated with patellar avascularity. This concept has been supported by
clinical1,16
and histological6
findings, with demonstration of osteonecrosis of the patella following lateral
release. However, Ritter and
Campbell24 were
unable to find a connection between lateral retinacular release and
osteonecrosis or patellar fracture. In a review of 555 primary total knee
arthroplasties, they found that a lateral release had not been performed in
seventeen of eighteen that were associated with a patellar fracture.
Osteolysis and loosening of the patellar component place the patella at
risk for fracture because of poor bone quality. Both increased osseous
resection and disruption of the anterior cortex during patellar preparation
can compromise the remaining bone stock. Inadequate osseous resection,
overstuffing of the anterior compartment (as a result of an under-resected
patella or an anteriorized femoral component), and patellar maltracking can
also increase the risk of a postoperative patellar fracture as a result of
transfer of higher contact stresses to the patella. Femoral and tibial
component malalignment, either rotational or angular, can promote abnormal
loading of the patellofemoral compartment, resulting in subluxation,
dislocation, or fracture of the
patella21.
Patient Factors
It has been suggested that several patient factors increase the risk of
patellar fracture after primary total knee arthroplasty. These include
osteoporosis, bone cysts, poor bone stock, rheumatoid arthritis, male gender,
increased activity, and an excessive range of
motion1,7,8.
Rheumatoid arthritis may be an independent risk factor for periprosthetic
patellar fractures, particularly in the presence of a very thin and eroded
patella. It may be best to avoid resurfacing in these
cases7. However,
Grace and Sim8 found
no significant difference in the fracture prevalence between knees with
osteoarthritis (12%, six of fifty) and those with rheumatoid arthritis (18%,
six of thirty-three). This was confirmed in another study that showed no
significant difference between the two groups, with a patellar fracture
occurring in six of seven and nine of eleven knees,
respectively1.
Nevertheless, resurfacing should be avoided in cases in which severe patellar
erosion has left <10 mm of patellar
bone19.
It is arguable whether gender is an independent risk factor for patellar
fracture. Seventy-three (62%) of 117 patellar fractures seen in association
with nearly 17,000 primary total knee arthroplasties listed in the Mayo Clinic
Joint Registry were in
men11. It has been
speculated that male gender is associated with higher activity levels and
greater body weight, leading to increased extensor mechanism force and
patellofemoral
stresses25.
However, other authors have reported a predominance of patellar fractures in
women, attributing this phenomenon to the higher overall prevalence of
osteoporosis in
women1,15,26.
Implant Design Issues
Certain features of the implant design have been implicated as risk factors
for patellar fracture after patellar resurfacing at the time of total knee
arthroplasty. A central peg has been shown to act as a stress riser that
increases the risk of fracture both during implantation and
postoperatively1,7.
Designs with a larger central peg also create focal stress concentrations due
to the increased bone resection required for
implantation1,6.
Cementless or press-fit implants, especially metal-backed implants, lead to
higher contact stresses across the patellofemoral joint, which are also
considered to be a risk factor for
fracture1,7,16.
Femoral component geometry may also affect the risk of patellar fracture after
total knee arthroplasty.
Bourne12 and
Windsor et al. 27
found that posterior-stabilized total knee prostheses had increased contact
stress across the femoral component, resulting in increased patellofemoral
contact stresses and an increased risk of patellar fracture
(Table III).
Periprosthetic fractures are more likely after revision total knee
arthroplasty, which may be considered an independent risk factor for patellar
fracture7. Engh and
Ammeen14 found that
removal of a stable patellar component from osteoporotic bone can lead to
intraoperative patellar fracture during revision surgery. This is particularly
true during the revision of a well-fixed metal-backed patellar component when
stress-shielding has resulted in osteoporosis and diminished bone stock.
Again, it is advisable not to resurface the patella during revision surgery
when the residual patellar thickness is <10 mm; this is found in
approximately 30% to 40% of knees undergoing revision total knee
arthroplasty19.
Berry11 found
that men had an increased rate of patellar fracture following revision total
knee arthroplasty. Of fifty-three fractures seen in association with more than
2900 revision total knee arthroplasties, thirty (57%) were in men. In a
separate study, Berry and Rand found that patellar fracture was one of the
most common complications following isolated revision of the patellar
component in forty-two
knees2. There were
five late patellar fractures, four of which were in patients who had had a
lateral retinacular release during either a primary total knee arthroplasty or
a revision of the patellar component. Because of the risk of patellar
fracture, all-polyethylene patellar components should not be revised unless
they are loose, substantially worn, or
malpositioned20.
Stress-shielding from a metal-backed patellar component or osteolysis from
metallosis can predispose a patella to fracture at the time of implant
removal, and patients should be informed about this risk.
The location and pattern of the fracture, the integrity of the extensor
mechanism, the stability of the patellar component, and the quality of the
remaining bone stock are considered when determining the appropriate treatment
of a periprosthetic patellar fracture. Treatment options include nonoperative
management; open reduction and internal fixation, occasionally in combination
with partial or total patellectomy; revision total knee arthroplasty with
replacement of all three components; or isolated revision of the patellar
component. Complete reconstruction of the extensor mechanism with an allograft
may be needed when the fracture and the extensor mechanism are
irreparable.
An asymptomatic, nondisplaced patellar fracture that is not associated with
an extensor lag and is associated with an intact patellar component may
require no specific treatment other than observation or immobilization in a
cast or
brace1,10,12,14,26.
A symptomatic, nondisplaced transverse fracture that is associated with an
intact patellar component and an intact extensor mechanism can be treated in a
cylinder cast or a knee brace locked in extension for six weeks, and
weight-bearing can be allowed immediately (Figs.
4-A, 4-B, and
4-C). Vertical patellar
fractures are associated with lower rates of disruption of the extensor
mechanism and thus can be treated with cast immobilization if they are
symptomatic.
A displaced transverse fracture through the middle third of the patella
should be repaired, but tension band wiring may be difficult to achieve in the
presence of a patellar component (Figs.
5-A,5-B,5-C,5-D)12,14.
If the component is loose, it should be removed to facilitate stabilization of
the fracture and repair of the extensor mechanism. A cylinder cast can be worn
for eight to twelve weeks, until there is radiographic and clinical evidence
of fracture-healing.
The treatment for a fracture of the superior or inferior pole of the
patella is similar to that for a quadriceps tendon or patellar tendon
disruption, respectively. The clinical outcome of a stable repair of the
extensor mechanism with use of a locking stitch, with or without partial
patellectomy, is more predictable than that after use of pins, wires, or
screws. Sutures are routed through vertically oriented drill holes in the
patella and are secured over a bone bridge over the proximal or distal edges
of the patella; a tension band technique may be used to further protect an
avulsion fracture of the inferior pole by routing the wire through the tibial
tubercle14. If the
knee cannot be passively flexed to 75° without disrupting or putting undue
tension on the surgical repair, then consideration should be given to
augmentation of the extensor mechanism with a semitendinosus or iliotibial
band tendon autograft, allograft, or
xenograft14.
Reconstruction of the extensor mechanism with an allograft is an
alternative that has been used to treat disruption of the extensor mechanism
following total knee arthroplasty. The indications for this treatment modality
have been extrapolated to include a periprosthetic patellar fracture
associated with disruption of the extensor mechanism. Burnett et
al.28 compared two
separate techniques of allograft reconstruction of the extensor mechanism in
patients followed for a minimum of two years. Thirteen patients in whom the
allograft had been tightly tensioned in full extension exhibited an average
postoperative extensor lag of 4.3° (range, 0° to 15°) with an
average Hospital for Special Surgery knee
score29 of 88
points. This group was compared with a second group of seven patients in whom
the allograft had been minimally tensioned; all of those patients had a
clinical failure, with an average postoperative extensor lag of 59°
(range, 40° to 80°) and an average Hospital for Special Surgery knee
score of 52 points. The difference between the two groups was significant (p
< 0.0001). Neither group had a loss of postoperative knee flexion, which
was 104° and 108°, respectively (p < 0.549). Although this study
included a small number of patients, allograft reconstruction of the extensor
mechanism with tensioning of the graft to treat a potentially catastrophic
complication following total knee arthroplasty yielded favorable clinical
results.
Loose components should be removed routinely; stable components should be
removed only if removal is necessary to achieve adequate fracture
stabilization. A patellar implant should not be used to resurface a patella
that has undergone previous operative fixation of a fracture because the
implant may weaken the repair and impart greater stress on the extensor
mechanism; this may increase the risk of nonunion or refracture. Hemispherical
trabecular metal patellar components (Zimmer, Warsaw, Indiana) have been used
for primary fracture fixation during patellar component revision with some
success, but further study is
necessary30.
Windsor et al.27
classified periprosthetic patellar fractures as vertical, transverse, or
comminuted, with subdivision into nondisplaced and displaced types. In the
majority of vertical fractures, cemented patellar buttons remain well fixed to
one of the two minimally displaced fracture fragments. On the basis of this
finding, vertical and comminuted fractures, regardless of displacement, and
transverse fractures that are displaced <2 cm can be treated with six weeks
of immobilization in a cylinder cast or a brace locked in extension. According
to current protocols, nonoperative treatment is not acceptable for fractures
with 2 cm of displacement unless there is no compromise of the extensor
strength or the range of
motion1. When a
fracture is associated with disruption of the extensor mechanism and an
extensor lag as seen on clinical examination, open reduction and internal
fixation should be considered, with the methods varying according to the
location of the fracture. Fixation with metal cerclage wire with retinacular
repair can be employed when the patellar component precludes adequate exposure
of the fracture site; this treatment should be followed by cast or brace
immobilization. Specific details of the operative repair depend on the quality
of the bone stock, the location and pattern of the fracture, and the stability
of the patellar component.
When the patellar component is loose, removal without replacement is often
necessary since the remaining bone stock is usually inadequate to accommodate
a new implant. Removing the component and leaving an osseous shell
(patelloplasty) may predispose the patient to the development of residual
anterior knee discomfort and crepitus, and patients should be advised about
this possibility. Severely comminuted patellar fractures with substantial
displacement and an extensor lag may require a partial patellectomy with
repair of the disrupted retinaculum and removal of the patellar component.
Alternatively, an allograft reconstruction of the extensor mechanism may be
necessary (Figs.
5-A,5-B,5-C,5-D).
Brick and Scott1
recommended surgical intervention for periprosthetic patellar fractures with
displacement in excess of 2 cm, an extensor lag, and loosening or displacement
of the patellar implant. Open reduction and internal fixation should not be
repeated after an initial attempt fails. Partial patellectomy, with repair of
the quadriceps tendon or patellar tendon, should be considered when there are
small displaced proximal or distal fragments. Total patellectomy should be
considered a salvage option for severely comminuted fractures and failed
fixation.
Ortiguera and
Berry10 recommended
a treatment algorithm for periprosthetic patellar fractures based on a useful
contemporary classification (Table
IV). Type-I fractures (implant and extensor mechanism intact) are
treated nonoperatively. Type-II fractures (implant intact and extensor
mechanism disrupted) require repair of the extensor mechanism with partial
patellectomy or open reduction and internal fixation of the fracture. Type-III
fractures (implant loose) are subdivided into Types IIIa (good remaining
patellar bone stock) and IIIb (poor remaining patellar bone stock). If the
patient is sufficiently symptomatic, operative intervention should be
considered, with patelloplasty being an option. Type-IIIa fractures may be
treated with component revision or component resection arthroplasty, and
Type-IIIb fractures can be treated with implant removal and patelloplasty or
total patellectomy.
Tria et al.3
reported their experience with eighteen patients who had a total of ten
nondisplaced and eight displaced patellar fractures. Nine of the ten
nondisplaced fractures were treated nonoperatively with analgesics and some
decrease in physical activity, but no immobilization. The tenth nondisplaced
fracture required excision of the fracture fragments because of nonunion after
conservative treatment. Of the eight displaced fractures, four were associated
with an intact extensor mechanism and patellar component, with minimal
symptoms, and were treated nonoperatively with decreased activity and no
immobilization. Three of the remaining four displaced fractures were
associated with disruption of the extensor mechanism. Two patients underwent
implant removal, and one had a patellectomy. The fourth patient refused
surgery. The average recovery of flexion was 110° (range, 90° to
125°) compared with 112° on the contralateral side. No other outcomes
were reported in that series.
Grace and Sim8
reported on twelve periprosthetic patellar fractures. Four of the twelve
fractures were minimally displaced (<5 mm), had little to no comminution,
and were associated with an intact patellar button. These four were all
treated nonoperatively with either a long leg knee immobilizer or a cylinder
cast for six weeks, and three of the four patients had a satisfactory result.
The remaining eight fractures (in seven patients) were displaced (>5 mm)
and had excessive comminution or a loose patellar button. All of these
fractures were managed operatively; a total patellectomy was done in four; a
partial patellectomy, in two; tension band wiring, in one; and circumferential
wiring, in one. Five of seven patients had a satisfactory result according to
The Hospital for Special Surgery knee rating
scale29. The
average postoperative flexion arc was 87°. In eleven of the entire group
of twelve cases, there was less than a 5° difference in the extensor lag
when compared with the prefracture state. The authors reported complications
in five of the twelve cases, including ipsilateral quadriceps rupture,
secondary fatigue fracture, a wound infection, fibrous nonunion, and synovial
cyst formation.
Hozack et al.26
reported on twenty-one fractures and concluded that nonoperative treatment of
nondisplaced fractures yielded satisfactory results whereas patellectomy
produced good but not excellent results, with decreased quadriceps strength in
three of four cases. The result was poor for four of six displaced fractures
treated with patellectomy, two of the four treated with fragment excision, and
both of the fractures treated with internal fixation. Quadriceps strength was
preserved with nonoperative treatment of nondisplaced fractures, but it
decreased when patellectomy was done for persistent instability and pain.
Goldberg et
al.15 reported on
thirty-six fractures. Fourteen that were classified as Type I (a marginal
fracture with an intact extensor mechanism and implant-bone interface) and two
that were classified as Type IIIb (a fracture of the inferior pole of the
patella without rupture of the patellar tendon) were treated nonoperatively
with knee immobilization and partial weight-bearing with crutches, followed by
a rehabilitation program with active range-of-motion and
quadriceps-strengthening exercises. In all sixteen cases, the result was good
or excellent according to the University Hospitals of Cleveland quantitative
functional knee
score31, with an
average arc of motion of 100°. There were six Type-II fractures
(disruption of the extensor mechanism or implant-bone interface), and all were
treated operatively. Of eight Type-IIIa fractures, seven underwent repair of a
ruptured patellar tendon associated with superior patellar migration; one
patient refused surgery and had a poor result. There were six Type-IV
fracture-dislocations, and all were treated operatively. In the total series
of thirty-six cases, twenty-two had an overall good or excellent result
(average arc of motion, 100°) and fourteen (four treated nonoperatively
and ten treated operatively) had a fair or poor result, with an average arc of
motion of 80° and an extensor lag of >10°. On the basis of this
study, Goldberg et al. concluded that patients with a fracture that is not
associated with a dislocation of the patella, loosening of the implant, or
complete disruption of the extensor mechanism usually will recover function
with nonoperative management.
Brick and Scott1
reported that all six of their patients who had a nondisplaced fracture
treated with immobilization for six weeks had a satisfactory result. However,
eleven of fifteen displaced fractures treated with internal fixation failed to
heal. They concluded that nonoperative treatment may be successful for 50% to
80% of periprosthetic fractures of the patella, depending on the variables
discussed earlier.
In a study of seventy-eight fractures by Ortiguera and
Berry10,
thirty-eight were Type I and thirty-seven of those were treated with
observation, a brace, or a cast. In this group, 82% (thirty-one) of the cases
remained asymptomatic, one patient required excision of a symptomatic
nonunion, arthrofibrosis developed in one, and the patellar implant eventually
loosened in one. One Type-I fracture was initially treated operatively. Eleven
of twelve Type-II fractures were treated with repair of the extensor mechanism
and partial patellectomy or open reduction and internal fixation. One was
treated with a brace, and the patient was still asymptomatic at the five-year
follow-up evaluation, despite a 5° extension lag. The result was a failure
for five of the six fractures managed with open reduction and internal
fixation and repair of the extensor mechanism. The results were not specified
for the five fractures that were treated with partial patellectomy and repair
of the extensor mechanism. There was a 50% complication rate in association
with the management of Type-II fractures, with a revision required in five of
the eleven cases. Seven of the twelve knees with a Type-II fracture had
instability, pain, or weakness at the time of the last follow-up. There were
twenty-eight Type-III fractures (associated with loosening of the patellar
implant), which were treated operatively if the patient was symptomatic. The
type of operation selected was based on the quality of the remaining bone
stock. Of the twelve Type-IIIa fractures (implant loose, with good remaining
bone stock), four were treated with observation; five, with open reduction and
internal fixation (with resection of the patellar component); two, with
component resection and patelloplasty; and one, with patellar component
revision. Six Type-IIIb fractures (implant loose, poor remaining bone stock)
underwent partial or complete patellectomy, five were treated with component
resection and patelloplasty, and one was treated with patellar component
revision. There were eight complications following treatment of Type-III
fractures, including failure of fixation, an extensor lag in excess of
15°, nonunion, infection, instability, and arthrofibrosis. Three knees
required a revision operation.
Keating et al.32
reviewed the results following 177 fractures and found that operative
treatment was generally associated with a high complication rate and
nonoperative treatment generally led to good results. However, these results
were confounded by the severity of the fracture, the integrity of the extensor
mechanism, and the quality of the residual patellar bone. Of twenty-two Type-1
fractures (all vertical fractures and fractures associated with an intact
extensor mechanism and a stable implant) and twenty-two Type-2A fractures
(horizontal fractures with separation of the fracture fragments of <1 cm
associated with disruption of the extensor mechanism and a stable or unstable
implant), all but two were treated nonoperatively, with good results.
According to the Knee Society Scoring
system33, the
patients with a Type-1 or 2A fracture had, on the average, pain scores of 44
and 49 points, knee scores of 85 and 92 points, and 120° and 117° of
flexion, respectively, and no extensor lag. One patient in each group
underwent excision of an extruded patellar button, and a deep infection
developed in one of them. Fourteen of seventeen Type-2B fractures (those with
separation of the fracture fragments of >1 cm) were treated nonoperatively.
A residual extensor lag was present in three of those cases. Operative
treatment was performed for three of the Type-2B fractures. Two were treated
with open reduction and internal fixation, and there was postoperative
nonunion despite good clinical results. In the final group of 114 Type-3
fractures (an intact extensor mechanism and an unstable implant), only six
were treated operatively. Two of the patients had a subsequent deep wound
infection, and one had secondary component loosening six months
postoperatively. The remainder of the Type-3 fractures were treated
nonoperatively, and at the time of follow-up the average pain score was 43
points, the average knee score was 83 points, and the average range of knee
flexion was 116°.
Laskin34
reported his experience with the management of patellar fractures in the
setting of revision total knee arthroplasty. Because a majority of fractures
occur in an avascular patella, attempting repair is often futile. As a result,
Laskin recommended removal of the implant and performance of a patelloplasty.
During this procedure, all small fragments of bone are removed except for the
small fragment adjacent to the patellar tendon. The remaining patellar bone
stock is thinned to a wafer, and the retinaculum is then repaired. At the
two-year follow-up evaluations after twelve patelloplasties that had been done
at the time of revision total knee arthroplasties, two of ten patients
complained of anterior knee pain, only one patient could ascend and descend
stairs, the average extensor lag was 10°, and the average range of knee
flexion was 120°.
Nonoperative treatment, usually with an initial period of immobilization,
of a minimally displaced fracture associated with an intact patellar component
and extensor mechanism can often produce satisfactory results with low
morbidity. In an overwhelming number of cases, operative treatment of
periprosthetic patellar fractures yields relatively poor results. Suboptimal
outcomes with high complication rates have typically been reported in the
literature, and little improvement in functional outcomes has been
demonstrated even with newer operative techniques.
Operative intervention is still required to treat some fractures that are
associated with loosening of the patellar component, considerable fracture
displacement, and disruption of the extensor mechanism. Specifically, Type-II
fractures require repair of the extensor mechanism with treatment of the
fracture by patellectomy or open reduction and internal fixation. Type-III
fractures necessitate surgical intervention with revision of the patellar
component, resection of the component, patelloplasty, or total patellectomy.
In addition, early or late allograft reconstruction of the extensor mechanism
should be considered. All patients should be advised about the risk of failure
of surgical reconstruction of periprosthetic patellar fractures.
Avascularity of the patella may also affect treatment. Newer strategies for
fracture fixation, such as the use of trabecular metal implants that allow
simultaneous fracture fixation and resurfacing and the use of bone graft in
conjunction with bone morphogenetic proteins, may help to redefine operative
management of these challenging fractures.
Brick GW, Scott RD. The patellofemoral
component of total knee arthroplasty. Clin Orthop Relat Res.
1988;231:
163-78.231163Â
1988Â
[PubMed] Â
Berry DJ, Rand JA. Isolated patellar
component revision of total knee arthroplasty. Clin Orthop Relat
Res. 1993;286:
110-5.286110Â
1993Â
Â
Tria AJ, Harwood DA, Alicea JA, Cody RP.
Patellar fractures in posterior stabilized knee arthroplasties. Clin
Orthop Relat Res. 1994;299:
131-8.299131Â
1994Â
Â
Clayton ML, Thirupathi R. Patellar
complications after total condylar arthroplasty. Clin Orthop Relat
Res. 1982;170:
152-5.170152Â
1982Â
Â
Le AX, Cameron HU, Otsuka NY, Harrington
IJ, Bhargava M. Fracture of the patella following total knee arthroplasty.
Orthopedics. 1999;22:
395-9.22395Â
1999Â
[PubMed] Â
Scott RD, Turoff N, Ewald FC. Stress
fracture of the patella following duo-patellar total knee arthroplasty with
patellar resurfacing. Clin Orthop Relat Res.
1982;170:
147-51.170147Â
1982Â
[PubMed] Â
Rorabeck CH, Angliss RD, Lewis PL.
Fractures of the femur, tibia, and patella after total knee arthroplasty:
decision making and principles of management. Instr Course
Lect. 1998;47:
449-58.47449Â
1998Â
Â
Grace JN, Sim FH. Fracture of the
patella after total knee arthroplasty. Clin Orthop Relat Res.
1988;230:
168-75.230168Â
1988Â
[PubMed] Â
Cameron HU, Fedorkow DM. The patella in
total knee arthroplasty. Clin Orthop Relat Res.
1982;165:
197-9.165197Â
1982Â
[PubMed] Â
Ortiguera CJ, Berry DJ. Patellar
fracture after total knee arthroplasty. J Bone Joint Surg Am.
2002;84:
532-40.84532Â
2002Â
[PubMed] Â
Berry DJ. Epidemiology: hip and knee.
Orthop Clin North Am.
1999;30:
183-90.30183Â
1999Â
[PubMed][CrossRef] Â
Bourne RB. Fractures of the patella
after total knee replacement. Orthop Clin North Am.
1999;30:
287-91.30287Â
1999Â
[PubMed][CrossRef] Â
Insall JN, Lachiewicz PF, Burstein AH.
The posterior stabilized condylar prosthesis: a modification of the total
condylar design. Two to four-year clinical experience. J Bone Joint
Surg Am. 1982;64:
1317-23.641317Â
1982Â
Â
Engh GA, Ammeen DJ. Periprosthetic
fractures adjacent to total knee implants: treatment and clinical results.
Instr Course Lect. 1998;47:
437-48.47437Â
1998Â
[PubMed] Â
Goldberg VM, Figgie HE, Inglis AE,
Figgie MP, Sobel M, Kelly M, Kraay M. Patellar fracture type and prognosis in
condylar total knee arthroplasty. Clin Orthop Relat Res.
1988;236:
115-22.236115Â
1988Â
[PubMed] Â
Healy WL, Wasilewski SA, Takei R,
Oberlander M. Patellofemoral complications following total knee arthroplasty.
Correlation with implant design and patient risk factors. J
Arthroplasty. 1995;10:
197-201.10197Â
1995Â
[CrossRef] Â
Berman AT, Reid JS, Yanicko DR, Sih GC,
Zimmerman MR. Thermally induced bone necrosis in rabbits. Relation to implant
failure in humans. Clin Orthop Relat Res.
1984;186:
284-92.186284Â
1984Â
[PubMed] Â
Willert HG, Ludwig J, Semlitsch M.
Reaction of bone to methacrylate after hip arthroplasty: a long-term gross,
light microscopic, and scanning electron microscopic study. J Bone
Joint Surg Am. 1974;56:
1368-82.561368Â
1974Â
Â
Reuben JD, McDonald CL, Woodard PL,
Hennington LJ. Effect of patella thickness on patella strain following total
knee arthroplasty. J Arthroplasty. 1991;
6: 251-8.6251Â
1991Â
[PubMed][CrossRef] Â
Lonner JH, Mont MA, Sharkey PF, Siliski
JM, Rajadhyaksha AD, Lotke PA. Fate of the unrevised all-polyethylene patellar
component in revision total knee arthroplasty. J Bone Joint Surg
Am. 2003;85:
56-9.8556Â
2003Â
Â
Figgie HE, Goldberg VM, Figgie MP,
Inglis AE, Kelly M, Sobel M. The effect of alignment of the implant on
fractures of the patella after condylar total knee arthroplasty. J Bone
Joint Surg Am. 1989;71:
1031-9.711031Â
1989Â
Â
Lynch AF, Rorabeck CH, Bourne RB.
Extensor mechanism complications following total knee arthroplasty. J
Arthroplasty. 1987;2:
135-40.2135Â
1987Â
[CrossRef] Â
Scuderi G, Scharf SC, Meltzer LP, Scott
WN. The relationship of lateral releases to patella viability in total knee
arthroplasty. J Arthroplasty.
1987;2:
209-14.2209Â
1987Â
[PubMed][CrossRef] Â
Ritter MA, Campbell ED. Postoperative
patellar complications with or without lateral release during total knee
arthroplasty. Clin Orthop Relat Res. 1987;
219: 163-8.219163Â
1987Â
[PubMed] Â
Tharani R, Nakasone C, Vince KG.
Periprosthetic fractures after total knee arthroplasty. J
Arthroplasty. 2005;20(4
Suppl 2): 27-32.2027Â
2005Â
[CrossRef] Â
Hozack WJ, Goll SR, Lotke PA, Rothman
RH, Booth RE. The treatment of patellar fractures after total knee
arthroplasty. Clin Orthop Relat Res. 1988;
236: 123-7.236123Â
1988Â
[PubMed] Â
Windsor RE, Scuderi GR, Insall JN.
Patellar fractures in total knee arthroplasty. J Arthroplasty.
1989;4Suppl:
S63-7.4S63Â
1989Â
[PubMed][CrossRef] Â
Burnett RS, Berger RA, Paprosky WG,
Della Valle CJ, Jacobs JJ, Rosenberg AG. Extensor mechanism allograft
reconstruction after total knee arthroplasty. A comparison of two techniques.
J Bone Joint Surg Am.
2004;86:
2694-9.862694Â
2004Â
[PubMed] Â
Insall JN, Ranawat CS, Aglietti P, Shine
J. A comparison of four models of total knee-replacement prostheses. J
Bone Joint Surg Am. 1976;58:
754-65.58754Â
1976Â
Â
Nelson CL, Lonner JH, Lahiji A, Kim J,
Lotke PA. Use of a trabecular metal patella for marked patella bone loss
during revision total knee arthroplasty. J Arthroplasty.
2003;18(7 Suppl 1):
37-41.1837Â
2003Â
[PubMed][CrossRef] Â
Goldberg VM, Heiple KG, Ratnoff OD,
Kurczynski E, Arvan G. Total knee arthroplasty in classic hemophilia. J
Bone Joint Surg Am. 1981;63:
695-701.63695Â
1981Â
Â
Keating EM, Haas G, Meding JB. Patella
fracture after post total knee replacements. Clin Orthop Relat
Res. 2003;416:
93-7.41693Â
2003Â
[CrossRef] Â
Insall JN, Dorr LD, Scott RD, Scott WN.
Rationale of the Knee Society clinical rating system. Clin Orthop Relat
Res. 1989;248:
13-4.24813Â
1989Â
Â
Laskin RS. Management of the patella
during revision total knee replacement arthroplasty. Orthop Clin North
Am. 1998;29:
355-60.29355Â
1998Â
[CrossRef] Â
Thompson FM, Hood RW, Insall J. Patellar
fractures in total knee arthroplasty. Orthop Trans.
1981;5:
490-1.5490Â
1981Â
Â
Chun KA, Ohashi K, Bennett DL, El-Khoury
GY. Patellar fractures after total knee replacement. AJR Am J
Roentgenol. 2005;185:
655-60.185655Â
2005Â
Â