Abstract
Background: Patellar complications following endoprosthetic
reconstruction can occur as a result of anatomic, physiologic, and surgical
reasons. Patellar impingement on tibial polyethylene is a complication of
distal femoral replacement, and it is frequently related to inaccurate
restoration of the joint line and to soft-tissue contracture. The purpose of
our study was to determine the prevalence and type of patellar complications
following distal femoral replacements after excisions of bone tumors.
Methods: The results of reconstruction with use of a rotating-hinge
endoprosthesis following excision of a distal femoral tumor in forty-three
patients were retrospectively reviewed. Patients were followed clinically and
radiographically for a minimum of forty-eight months or until death. Pain
status, functional scores, and the range of motion were determined from a
prospectively maintained database. The ratio of the patellar tendon length to
the height of the patellar tendon insertion, as described by Insall and
Salvati, was calculated. In addition, we attempted to determine whether the
position of the patella was associated with anterior knee pain or with the
functional scores derived with use of the International Society of Limb
Salvage (ISOLS) scoring system.
Results: Thirty-five patellar complications, including eleven cases
of impingement, occurred in twenty-seven patients (63%). We found no
difference, on the basis of our sample size, with regard to the presence of
patellar pain, the range of motion, or the Insall-Salvati ratio between the
patients with and those without impingement. The ratio of the patellar tendon
length to the height of the patellar tendon insertion averaged 0.9 in the
group with impingement and 1.4 in the group without impingement (p = 0.07).
The ISOLS score averaged 21.2 points in the group with impingement compared
with 24.2 points in the group without impingement (p = 0.01). Patella baja
occurred in nine patients. The average ISOLS score (and standard deviation)
was 20.1 ± 4.4 points for the patients with patella baja compared with
24.8 ± 3.9 points in the group with a normal patellar position (p =
0.004). Patellar fracture occurred in two patients, and osteonecrosis occurred
in two patients. These patients were treated nonoperatively.
Conclusions: Patellar complications are common after distal femoral
resection and endoprosthetic reconstruction. Patellar impingement on the
polyethylene tibial bearing surface is a more common and important
complication of distal femoral replacement than has been reported to date.
Patella baja is also a relatively common complication, which has a negative
impact on knee function.
Level of Evidence: Prognostic Level IV. See Instructions
to Authors for a complete description of levels of evidence.
Devascularization of the patella and the patellar tendon during tumor
resection about the knee is inevitable and can be detrimental, causing
osteonecrosis and fracture of the patella as well as contracture of the
patellar tendon. Impingement of the patella on the polyethylene tibial bearing
can also occur after distal femoral replacement. Although impingement of the
patella on the tibial bearing is occasionally seen after revision knee
replacement or primary knee replacement for the treatment of posttraumatic or
infectious arthritis, it is rare after primary knee replacement for the
treatment of osteoarthritis or rheumatoid arthritis and it has not been
mentioned after oncologic reconstruction, to our knowledge. Impingement can be
painful and can restrict knee flexion. We hypothesized that impingement has a
negative impact on knee function after distal femoral replacement.
Normal knees have a consistent relationship between the length of the
patella and the length of the patellar tendon
(LT)1. There is also
a consistent relationship between the length of the patellar tendon (LT) and
the "height" of the patellar tendon insertion (HI). Insall and
Salvati described the "height" of the patellar tendon insertion as
the perpendicular distance from a line drawn parallel to the tibial plateau to
the proximal-most insertion of the patellar tendon, and they reported that the
LT/HI ratio should approach
11. They described
an average LT/HI ratio of 1.85 with a standard deviation of 0.24. Patella baja
is defined as shortening of the patellar tendon by at least 20% relative to
the length of the patella, and patella alta is described as relative
lengthening of the patellar tendon by at least
20%1. Both of these
abnormalities are recognized complications of total knee arthroplasty and are
thought to be associated with anterior knee pain and a decreased range of
motion2,3.
The purpose of our study was to determine the prevalence of patellar
complications following distal femoral replacements after resections of bone
tumors.
Our institutional review board granted permission for this study. We
retrospectively reviewed information, from our prospectively collected
database, on a consecutive series of patients who had had a distal femoral
reconstruction after resection of a bone tumor between 1990 and 2000. The most
common diagnosis was osteogenic sarcoma. We included only patients who had
undergone intra-articular resection and reconstruction with use of a Finn
modular, segmental, rotating-hinge endoprosthesis (Biomet, Warsaw, Indiana)
after resection of a distal femoral bone tumor (forty-one patients) or after a
pathologic fracture secondary to radiation necrosis (two patients). The latter
two patients had been treated with wide resection followed by external beam
radiation therapy for a soft-tissue sarcoma in the distal part of the thigh. A
pathologic distal femoral fracture subsequently occurred and was treated with
open reduction and internal fixation. The hardware failed in both patients as
a result of nonunion. Distal femoral replacement was performed as a salvage
procedure. Patients requiring a custom prosthesis were excluded from the
study. We also excluded patients treated with a total femoral replacement, a
proximal tibial replacement, an allograft-prosthetic composite,
extra-articular resection, or an extensible prosthesis.
The cases of forty-three consecutive patients (nineteen men and twenty-four
women) who met our inclusion criteria were retrospectively reviewed. The
median age of the patients at the time of surgery was forty-one years. Six
patients died during the study period, and the average duration of follow-up
for those patients was forty months (range, ten to 101 months). The living
patients were followed for a minimum of forty-eight months and a mean of 7.5
years (range, four to 14.9 years).
International Society of Limb Salvage Score
Functional results were graded with use of the International Society of
Limb Salvage (ISOLS) scoring system, which is based on modifications of the
initial Musculoskeletal Tumor Society rating
scale4,5.
Five points is the highest possible score for each of the individual
categories, which include pain, emotional acceptance, supports, function,
walking ability, and gait, and a maximum of 30 points can be obtained. The
pain score is determined, in part, on the basis of whether the patient
requires pain medication and the type of medication (narcotic compared with
non-narcotic) that is needed. In addition, we specifically asked each patient
whether he or she had anterior knee pain.
Pain
Patients were categorized into two groups on the basis of the presence or
absence of anterior knee pain. Each patient was asked specifically whether he
or she had pain localized to the front of the knee.
Prosthesis
The Finn modular, segmental, rotating-hinge endoprosthesis (Biomet), made
of cobalt-chromium, was designed to optimize weight-bearing forces and improve
patellofemoral kinematics. The sagittal dimension of the femoral component
mimics the normal anatomy of the knee. This provides patellar offset that
helps to optimize the lever arm of the quadriceps mechanism. The prosthesis
allows flexion to 135°, extension to 0°, external rotation to 20°,
and internal rotation to
20°6,7.
Operative Technique
A medial parapatellar incision was used. A modified intraarticular excision
including an en bloc excision of the distal part of the femur along with the
synovium and capsule was performed in all
patients8. The
patella, patellar tendon, and rectus femoris tendon were preserved, whereas
the fat pad was resected. The popliteal vessels were dissected to their
trifurcation, identified, and preserved, and the geniculate and capsular
vessels were ligated. Fifteen millimeters of the tibial plateau were resected
to accommodate the 3-mm metal tibial plate and the =12-mm polyethylene
tibial bearing.
The decision to resurface the patella was made intraoperatively and was
based on two major parameters. The first was the quality of the patellar
articular cartilage. Pristine cartilage tended to mitigate against
resurfacing. The second consideration was the patellar morphology with respect
to the patellofemoral articulation. If there was poor congruence of the
patella and the distal femoral patellar flange, the patella was resurfaced.
Patellar replacements were all polyethylene and were fixed with
methylmethacrylate cement.
Rehabilitation
Continuous passive motion was started on the second postoperative day if
fascia had been retained and could be closed and the skin edges appeared
healthy. Mobilization was delayed until five days postoperatively if the
fascia was deficient. Motion was delayed until any marginal necrosis was
demarcated as partial thickness. Full-thickness necrosis prompted wound
revision.
Radiographic Evaluation
The patellar position was evaluated with use of the Insall-Salvati ratio.
The sagittal length of the patella was compared with the length of the
patellar tendon on a lateral radiograph of the knee. The length of the patella
was considered to be the greatest sagittal distance from the proximal to the
distal pole (Fig. 1). The
ligament was measured on its posterior surface from the distal patellar pole
to its proximal-most insertion on the tibial tubercle. The measurements from
three consecutive postoperative visits, made with use of digital images and
the manufacturer's software to a refinement of a tenth of a millimeter (PACS
system 8.1; GE Medical Systems, Chicago, Illinois), were averaged. According
to the Insall-Salvati method, patella baja is considered to be present when
the ratio is <0.8, and patella alta is considered to be present when the
ratio is >1.2.
We used a variation of the method described by Insall and Salvati to
measure the height of the joint line above the patellar tendon insertion (HI).
Patellar height was determined by drawing a horizontal line parallel to the
tibial bearing surface. A perpendicular line was then drawn from the proximal
aspect of the patellar tendon insertion to the horizontal line running
parallel to the tibial bearing (Fig.
1). We measured the patellar tendon length (LT) as described
above. We calculated the ratio of the patellar tendon length to the height of
the patellar tendon insertion (LT/HI ratio).
We examined several parameters, including the thickness of the tibial
bearing and the amount of distal femoral resection, to determine whether they
were associated with the final patellar position. The arcs of motion in each
group were compared with those in each of the other patellar position
categories (i.e., patella baja compared with normal compared with patella
alta) to determine whether the patellar position influenced the range of
motion. The range of motion of the knee following surgery was measured with a
standard goniometer and was expressed as a degree of motion.
Statistical Analysis
All data were analyzed with use of SPSS software (SPSS 11.5, Chicago,
Illinois). Continuous data were evaluated with use of paired t tests.
Categorical data were evaluated with use of chisquare tests. We considered a p
value of <0.05 to be significant.
The mean range of motion at the time of the last follow-up was 94°
(range, 45° to 120°). The mean Insall-Salvati ratio was 0.97 (range,
0.52 to 2.04). The mean LT/HI ratio was 1.3 (range, 0.8 to 2.8). The mean
ISOLS score was 23.5 points (range, 16 to 30 points), which is 78% of the
maximum score of 30 points.
A total of thirty-five patellar complications occurred in twenty-seven
patients (63%). These complications included patellar impingement on the
tibial polyethylene, patella baja, patella alta, osteonecrosis, fracture,
dislocation, loosening of the patellar component, patellar pain requiring
revision, and avulsion of the patellar tendon. The nine patients with patella
baja had a total of eighteen patella-related complications.
Eleven patients (26%) had impingement of the patella on the tibial
polyethylene bearing, which was evident radiographically. In one patient, the
patella was revised specifically because of this problem
(Fig. 2). On the basis of the
sample size, there were no significant differences in the rate of anterior
knee pain, the Insall-Salvati ratio, or the range of motion between the group
with impingement and the group without impingement. The mean LT/HI ratio was
0.9 in the group with impingement and 1.4 in the group without impingement (p
= 0.07). The mean ISOLS score was 21.2 points in the group with impingement
compared with 24.2 points in the group without impingement (p = 0.01). With
this sample size, we found no association between tibial polyethylene
thickness and impingement.
Nine (21%) of the forty-three patients had patella baja. Their average
ISOLS score (and standard deviation) was 20.1 ± 4.4 points compared
with 24.8 ± 3.9 points in the group with a normal patellar position (p
= 0.004). The average range of motion in the group with patella baja was
80.6° ± 24° compared with 97.0° ± 16.5° in the
group with a normal patellar position (p = 0.02). On the basis of the sample
size, there was no difference in the length of the distal femoral resection
between these two groups. We also found no association between patella baja
and anterior knee pain, or between the thickness of the tibial polyethylene
and patella baja.
Five (12%) of the forty-three patients had patella alta. On the basis of
the sample size, they did not have a worse mean ISOLS score than the patients
with a normal Insall-Salvati ratio. Furthermore, with this small sample size,
we found no association between patella alta and the thickness of the tibial
polyethylene, the length of the distal femoral resection, or anterior knee
pain.
Twenty-three patients had an arc of motion of =100° at the time of
follow-up. Nineteen of these twenty-three patients had a normal Insall-Salvati
ratio. Ten of the nineteen patients with <100° of motion had an
abnormal Insall-Salvati ratio (p = 0.03)
(Fig. 3).
There were two cases of clinically relevant osteonecrosis of the patella
(Fig. 4). The diagnosis was
based on a finding of sclerosis and/or evidence of collapse on the lateral
radiograph9. Both
patients were treated with observation and protected weight-bearing and had
radiographic evidence of stabilization without fracture or fragmentation.
Two patellae fractured. Both fractures occurred during normal walking with
seemingly trivial trauma. Both fractures healed with immobilization.
There was one case of patellar dislocation. It was treated with open
reduction, lateral release, and medial plication.
One patellar component loosened. It was replaced with a cemented
polyethylene patellar component.
Fourteen (33%) of the forty-three patients reported anterior knee symptoms
in the form of pain and/or a clicking sensation. Eight (19%) of the
forty-three patients reported a clicking sensation that they found
disconcerting but not painful or progressive. Three patients reported
intermittent anterior knee pain that resolved with rest. Two patients stated
that the anterior knee symptoms were constant but responded to rest and
anti-inflammatory medication. One patient had pain that required patellar
revision. This patient also had patella baja and patellar impingement. She was
treated with débridement, revision of the patellar component, and
exchange of the tibial polyethylene bearing to a smaller-sized component, and
the symptoms resolved.
A patellar tendon avulsion occurred in one patient postoperatively. It was
treated with immobilization. At the time of the last follow-up, the patient
had a painless knee but a 40° extension lag.
We compared the fifteen patients who had undergone patellar resurfacing
with the twenty-eight who had not had the patella resurfaced. On the basis of
this sample size, no significant difference was found between the two groups
with regard to ISOLS score, range of motion, or anterior knee symptoms.
Anterior knee symptoms were reported by four of the fifteen patients who had
had resurfacing compared with ten of the twenty-eight who had not had
resurfacing.
With regard to the patella, distal femoral replacement should be considered
as a separate entity from primary or revision total knee arthroplasty. First
of all, during distal femoral resection the vascular supply to the patella is
necessarily compromised to remove the tumor. This may lead to osteonecrosis
and fracture of the
patella10.
Devascularization of the patellar tendon may also cause scar formation and
patellar tendon tethering, leading to patella
baja11. Second,
distal femoral resection is often performed in patients who are younger than
those who typically undergo total knee arthroplasty. This has important
implications regarding patellar resurfacing. Specifically, one would expect
the patellar articular cartilage to be in good condition in young patients and
in those who do not have arthrosis. This would mitigate against resurfacing in
the majority of cases. However, in a study of the results of distal femoral
replacement with use of the Kotz prosthesis, patellar resurfacing was shown to
improve quadriceps strength compared with that associated with an unresurfaced
patella9. Finally,
the position of the joint line deserves special attention in a distal femoral
replacement. It is desirable to reconstitute the normal position of the joint
line in order to maintain optimal patellofemoral biomechanics. We believe that
devascularization of the patella and difficulties with reconstituting the
joint line contributed to the higher rate of patellar complications in our
series.
It is possible that the patellar complications found in our study were
related to the prosthesis, which was designed to optimize patellar offset in
order to maximize quadriceps function. This is particularly important after
distal femoral resection in which part of the extensor musculature has been
removed. Increased offset has the negative effect of increasing stress on the
patella, which may lead to patellar complications. However, in a review of
twenty-four cases in which the Finn prosthesis was utilized for non-oncologic
reasons, Westrich et al. reported one case of patella alta with an extensor
lag in a patient who had had a previous patellar tendon rupture but they did
not report any other abnormalities associated with the
patella7.
In our series, the two most common patellar complications after distal
femoral replacement were patella baja and patellar impingement. One cause of
patella baja has been reported to be tethering of the patellar tendon by
fibrous bands11.
Formation of fibrous bands may have been facilitated in our series by
inadequate physical therapy postoperatively. All of our patients began
supervised active and passive range-of-motion exercises and gait training on
the day after the surgery. However, lapses in physical therapy were common
during postoperative chemotherapy, and this may have compromised the ultimate
functional result.
Scarring of the patellar tendon may also be related to the relative
devascularization that occurs after distal femoral resection. Of the six
arteries supplying the patella and/or patellar tendon, the anterior tibial
recurrent artery is the only vessel that can be reliably spared during the
performance of a medial parapatellar arthrotomy followed by distal femoral
replacement. This devascularization may predispose the patellar tendon to scar
formation and eventually to shortening. It is interesting to note that Koshino
et al. reported gradual "low riding of the patella" after primary
total knee
arthroplasty12. A
lateral release was performed in all of the patients in their series, which
may have led to devascularization of the patella and/or patellar tendon. A
lateral release may also predispose the patella to osteonecrosis; we found two
cases of osteonecrosis after distal femoral replacement. It is possible that
there were more cases of subclinical osteonecrosis, but we did not assess
asymptomatic patients with bone-scanning. Neither case of osteonecrosis was
associated with a fracture. There were two patellar fractures in our series,
which occurred after relatively trivial trauma. It is unclear whether the
fractures occurred in necrotic bone that we were unable to detect on plain
radiographs.
There were eleven cases of patellar impingement in our series. Six occurred
in conjunction with patella baja. The impingement may have been caused by the
relatively low-riding patella, but five cases occurred in patients with a
normal Insall-Salvati ratio. In those patients, the joint line was elevated
relative to the patella and patellar tendon. Four of the five patients with
impingement and a normal Insall-Salvati ratio had an LT/HI ratio of <1.
Iatrogenic elevation of the joint line is caused either by oversizing of the
tibial polyethylene (usually to correct a limb-length discrepancy) or by
implanting a distal femoral prosthesis of inadequate length. Once the
constrained prosthesis is locked into position intraoperatively, it may be
difficult to detect subtle abnormalities in the joint line position.
The complications that we have reported may be at least in part under the
control of the surgeon. Restoration of the normal anatomic joint line may help
to prevent abnormalities in the patella and patellar tendon. Careful
measurement of the resected specimen and the tibial plateau resection are
necessary first steps. The calculation of the length of the prosthetic
reconstruction must include the length of the tibial tray and polyethylene in
addition to the distal femoral prosthesis. Patellar devascularization is
unavoidable when the distal part of the femur is resected to remove an osseous
tumor. However, intensive physical therapy and surveillance of the patellar
tendon length may help to prevent patellar tendon contracture postoperatively.
?
Insall J, Salvati E. Patella position in
the normal knee joint. Radiology. 1971;
101: 101-4.101101
1971
[PubMed]
Blackburne JS, Peel TE. A new method of
measuring patellar height. J Bone Joint Surg Br.
1977;59:
241-2.59241
1977
[PubMed]
Aglietti P, Buzzi R, Gaudenzi A.
Patellofemoral functional results and complications with the posterior
stabilized total condylar knee prosthesis. J Arthroplasty.
1988;3:
17-25.317
1988
[PubMed][CrossRef]
Enneking WF. Modification of the system
for functional evaluation of surgical management of musculoskeletal tumors.
In: Enneking WF, editor. Limb salvage in musculoskeletal
oncology. Bristol-Meyers/Zimmer Orthopaedic Symposium. New York:
Churchill Livingstone; 1987. p
626-39.626
1987
Enneking WF, Dunham W, Gebhardt MC,
Malawar M, Pritchard DJ. A system for the functional evaluation of
reconstructive procedures after surgical treatment of tumors of the
musculoskeletal system. Clin Orthop Relat Res.
1993;286:
241-6.286241
1993
[PubMed]
Finn HA, Golden D, Kneisl JA, Simon MA.
The Finn knee: rotating hinge replacement of the knee. In: Brown KLB, editor.
Complications of limb salvage: prevention, management and
outcome. 6th International Symposium. Montreal. Montreal:
International Society of Limb Salvage; 1991. p
413-5.413
1991
Westrich GH, Mollano AV, Sculco TP, Buly
RL, Laskin RS, Windsor R. Rotating hinge total knee arthroplasty in severely
affected knees. Clin Orthop Relat Res.
2000;379:
195-208.379195
2000
[PubMed][CrossRef]
Healey J. Bone and soft tissue tumors
around the knee. In: Insall JN, Scott WN, editors. Surgery of the
knee. Vol 2, section XI, chapter 95.
Philadelphia: Churchill Livingstone; 2001. p
1997-2028.21997
2001
Tsuboyama T, Windhager R, Campanacci D,
Galletti S, Catani F, Capanna R, Campanacci M, Yamamuro T, Kotz R. Patellar
resurfacing in tumor endoprosthesis: a study in regard to power transmission.
Chir Organi Mov. 1994; 79:
243-50.79243
1994
[PubMed]
Kayler DE, Lyttle D. Surgical
interruption of patellar blood supply by total knee arthroplasty. Clin
Orthop Relat Res. 1988;229:
221-7.229221
1988
Thorpe CD, Bocell JR, Tullos HS.
Intra-articular fibrous bands. Patellar complications after total knee
replacement. J Bone Joint Surg Am. 1990;
72: 811-4.72811
1990
[PubMed]
Koshino T, Ejima M, Okamoto R, Morii T.
Gradual low riding of the patella during postoperative course after total knee
arthroplasty in osteoarthritis and rheumatoid arthritis. J
Arthroplasty. 1990;5:
323-7.5323
1990
[CrossRef]