A primary total knee arthroplasty for a knee with a valgus deformity is a
formidable surgical challenge. A valgus knee often has both bone and
soft-tissue abnormalities, including contracted lateral capsular and
ligamentous structures with or without medial laxity, contracted or lax
posterior soft tissues, osseous deficiency of the lateral femoral condyle
and/or tibial plateaus, external rotation deformity of the distal part of the
femur, secondary remodeling of the femoral and tibial metadiaphyseal region,
and patellar
maltracking1.
Despite advances in instrumentation for bone resection and alignment,
correcting a valgus deformity without relying on the use of a constrained
implant continues to be difficult for many
surgeons2.
After correct osseous alignment and positioning of the articular surfaces
have been achieved at the time of surgery, a strategy is necessary to ensure
correct soft-tissue balance throughout the range of motion. A surgical
technique should provide both immediate and long-term stability and cause no
notable increase in component loosening or wear rates. The structures most
commonly released in a valgus knee include the iliotibial band, the
posterolateral aspect of the capsule, the lateral collateral ligament, the
popliteus tendon, and the lateral head of the gastrocnemius muscle. In
addition, the medial collateral ligament may need to be shortened or advanced.
At present, there is no consensus regarding the sequence in which one or all
of these structures should be addressed.
In 1985, the senior one of us (C.S.R.) developed a new soft-tissue release
technique for valgus knees to address inherent instabilities that had been
noted with his earlier technique, originally described in
19793; the intent of
this new release was to avoid late-onset instability and the need for a
constrained
implant4. The
current report presents the five to fourteen-year follow-up results of total
knee arthroplasty performed with use of this updated technique.
We included all primary total knee replacements performed by the senior one
of us (C.S.R.), between January 1988 and December 1992, in knees with a
preoperative valgus angulation of =10° seen on standing anteroposterior
radiographs. Indications for surgery included pain and disability resulting
from knee arthritis as confirmed by radiographic evaluation. The goals of
surgery were to eliminate pain, correct the deformity, increase the range of
motion, and improve function. A total of 490 replacements were performed
during this time-period, and seventy-one patients (eighty-five knees) had a
preoperative valgus angulation of =10°. Thirty-two patients (thirty-six
knees) died, and four patients (seven knees) were lost to follow-up. The
remaining thirty-five patients (forty-two knees) formed the basis of this
study.
A bilateral procedure was performed in seven patients. Twenty-seven of the
patients were women, and eight were men. The average age of the patients at
the time of the surgery was sixty-seven years (range, twenty-seven to
eighty-two years). The preoperative diagnosis was osteoarthritis in
twenty-eight patients, rheumatoid arthritis in six, and posttraumatic
arthritis in one. Five patients had had a previous arthroscopy, two had had an
open meniscectomy, and one had had a supracondylar femoral osteotomy.
The PFC Modular total knee implant (DePuy Orthopaedics, a Johnson and
Johnson company, Warsaw, Indiana) was used in thirty-two knees, and the
Insall-Burstein-II implant (Zimmer, Warsaw, Indiana) was used in nine knees.
Both implants include a cam-post mechanism to substitute for the posterior
cruciate ligament and are of similar design. One knee with a severe valgus
deformity was treated with a constrained insert because mild midflexion
instability (defined as >5 mm of medial opening) was detected
intraoperatively.
Surgical Technique
If the medial joint space is >1 cm on anteroposterior weight-bearing
radiographs, less bone than is typically removed should be resected from both
the distal part of the femur and the proximal part of the tibia in order to
allow for soft-tissue balancing without elevation of the joint line or
creation of too large an extension gap. The tibial surface is cut at 90°
to its longitudinal axis, and the distal femoral resection is performed in
3° of valgus in relation to the anatomical axis as opposed to the typical
5° to 7° of valgus used for a varus knee; 3° of valgus is used in
order to protect against undercorrection of the underlying deformity. After
the proximal tibial and distal femoral bone cuts are made, the knee is
extended and is distracted with a lamina spreader, bringing the posterolateral
capsule complex under tension. Doing this should demonstrate a trapezoidal
extension gap. The tight soft-tissue capsular structures in the lateral
compartment are released intra-articularly, with use of electrocautery, at the
level of the tibial cut (Fig.
1-A). The release is performed transversely, from the lateral edge
of the resected posterior cruciate ligament to the posterior margin of the
iliotibial band, to create a rectangular extension gap. Electrocautery is used
to avoid injury to the peroneal nerve, which is usually located <1 cm from
the articular side. Both medial and lateral soft-tissue sleeves should have an
equal, 2 to 3 mm opening when a valgus or varus stress is applied with a
spacer block in place.
If the extension gap remains unbalanced after the intra-articular release,
the iliotibial band is lengthened in a controlled manner as necessary from
inside with use of the so-called pie-crusting technique, which consists of
multiple oblique stab incisions 1 cm above the joint line (Figs.
1-B and
2). This process continues
until a balanced extension gap has been achieved.
Attention is then turned to the flexion gap. No soft-tissue releases are
performed with the knee in flexion; rather, femoral bone cuts are made to
attain the correct soft-tissue balance in flexion. The rotational alignment of
the femoral component is determined by placing an anteroposterior cutting
block parallel to the tibial cut surface while a lamina spreader separates the
posterior edge of the cutting block from the tibial cut surface
(Fig. 3). Prior to cutting the
posterior femoral condylar bone parallel to the tibial cut surface, it is
necessary to verify that the tibial cut is in fact 90° to the long axis of
the tibia and that the soft tissues are balanced in extension. A varus tibial
cut or over-release of the medial side will lead to internal rotation of the
femoral component and possibly to patellar tracking problems. If rotational
malalignment is suspected, alignment can be checked by referencing the cutting
block with respect to the anteroposterior axis of Whiteside or the
transepicondylar
axis5.
Forty of the forty-two knees underwent the extensive lateral soft-tissue
release as described above. The two remaining knees were relatively well
balanced after routine exposure and the bone cuts and did not require
extensive releases.
Method of Evaluation
Preoperatively, each involved knee was evaluated for weight-bearing
alignment, flexion contracture, and ligamentous instability. Preoperative
radiographic analysis included standing anteroposterior, lateral, and sunrise
views of the affected knee as well as an anteroposterior view of the pelvis.
Full-length radiographs were made when the knee had complex triplanar
deformities. Radiographs were evaluated for osseous deformity, patellar
thickness and position, and alignment of the ipsilateral hip. (A varus hip
necessitates cutting the distal part of the femur into more valgus.)
Additionally, anteroposterior radiographs were scrutinized for soft-tissue
laxity such as medial-lateral opening and/or tibial subluxation.
The Knee Society clinical rating system was used for preoperative and
postoperative clinical evaluation, with a slight modification of the knee
alignment scoring as originally proposed in our previous
paper4. With the
Knee Society score, points are deducted when the anatomic alignment of the
knee is <5° or >10° of valgus. However, because the goal in this
series of valgus knees was to obtain an alignment of 3° to 5°, the
scoring system was changed so that deductions were made for an alignment of
<2° or >7° of valgus. Clinical and functional scores of =85
points were categorized as excellent; 70 to 84 points, as good; 60 to 69
points, as fair; and <60 points, as poor.
At the time of the latest follow-up, the tibial and femoral components were
evaluated radiographically with use of the Knee Society roentgenographic
evaluation system6.
Lateral and skyline radiographs were used to assess the patella for tilt,
displacement, residual bone thickness, coverage ratio, and radiolucency. In
addition to component positioning, each radiograph was assessed for the
presence of osteolysis, which was defined as an expanding area of focal
radiolucency measuring =1 cm in diameter. Any component with a
circumferential radiolucency at the bone-cement or component-cement interface
was considered to be loose.
Kaplan-Meier survivorship analysis was performed with use of revision for
any reason as the end point. A second survivorship analysis was done with
mechanical failure (aseptic loosening or instability) as the end point.
Survivorship curves were created with use of commercially available software
(GraphPad InStat; GraphPad Software, San Diego, California).
Clinical Results
The thirty-five patients (forty-two knees) were followed for an average of
nine years (range, five to fourteen years). The mean preoperative pain score
was 8 points, which improved to 48 points at the end of the first
postoperative year, 48 points at five years, and 47 points at ten years. The
mean preoperative and latest postoperative ranges of motion were both
110°. The mean score for stability of the knee improved from 17 points
preoperatively to 24 points at one year, 24 points at five years, and 24
points at ten years. There were no cases of late-onset instability. The mean
score for walking ability improved from 20 points preoperatively to 44 points
at one year, 44 points at five years, and 42 points at ten years. The score
for stair-climbing improved from a mean of 17 points preoperatively to 43
points at one year, 43 points at five years, and 41 points at ten years. The
clinical score improved from a mean of 30 points preoperatively to 94 points
at one year, 94 points at five years, and 93 points at ten years. The
functional score improved from a mean of 34 points preoperatively to 85 points
at one year, 85 points at five years, and 81 points at ten years. Of the nine
patients for whom the score for function was <80 points, all but one had
either severe spinal stenosis or a minimum of two other lower-extremity joint
arthroplasties.
Radiographic Results
The mean preoperative anatomic valgus angle was 15° (range, 10° to
32°), which was corrected to a mean of 5° of valgus (range, 0° to
10° of valgus) at the time of the latest follow-up. Correction to between
2° and 7° of valgus was achieved in thirty-seven of the forty-two
knees. None had a varus alignment, and two patients had a valgus alignment of
>8°. The mean femoral (a) angle was 4°, and the mean tibial
(ß) angle was 1°. The mean flexion (?) angle was 7°, and the
mean lateral tibial (d) angle was 5°. Postoperative residual
patellar thickness averaged 19 mm, with a mean difference of 2 mm between the
medial and lateral sides. The mean amount of patellar component tilt in
relation to the femoral component was 5°, and the mean postoperative
displacement of the patella from the center of the trochlea was 7 mm.
No radiolucencies were noted adjacent to any of the forty-two femoral or
tibial components at the time of the latest follow-up. No tibial or femoral
component was associated with osteolysis or had radiographic evidence of
loosening. Patellar loosening with displacement was noted in one knee five
years after the surgery. This patient became symptomatic eleven years after
the index surgery and underwent revision.
Complications
An early superficial infection that required débridement and
irrigation developed in one patient, who then had an uneventful postoperative
course. One patient with polyarticular rheumatoid arthritis had a small area
of skin necrosis over the patella. Although the final outcome was unaltered,
her rehabilitation was delayed. Deep venous thromboses in the calf developed
in two patients, and one patient had a nonfatal pulmonary embolus. There were
no peroneal nerve palsies or patellar dislocations.
Three of the thirty-five patients required revision surgery. One revision
was done to treat loosening of the patellar component as mentioned above. The
second revision was performed because of excessive wear of the articular
insert associated with synovitis but not osteolysis in a heavy, active
patient. This patient was treated with exchange of the tibial polyethylene
insert and has since done well. The third revision was performed to treat a
deep infection one year after the index operation. This patient underwent
staged revision, and had good stability and range of motion (0° to
95°) at ten years postoperatively.
Kaplan-Meier survivorship analysis revealed a survival rate (and 95%
confidence interval) of 83% ± 9.6% at fifteen years with revision
surgery for any reason as the end point. When only mechanical failure was the
end point, the survival rate was 85% ± 9.6% at fifteen years
(Fig. 4).
The most challenging aspect of primary total knee arthroplasty in a valgus
knee is achieving soft-tissue balance. Over the last twenty years, numerous
approaches and soft-tissue procedures have been
advocated5,7-12.
Whiteside recommended sequential releases of the iliotibial band, popliteus,
lateral collateral ligament, and lateral head of the
gastrocnemius5. He
also performed a tibial tubercle transfer when the Q angle (the angle
subtended by the quadriceps and patellar tendons) was >20°.
Buechel7, Fiddian et
al.8, and
Keblish10 suggested
using a lateral capsular approach for the treatment of valgus deformity. Healy
et al.9 and Krackow
et
al.11,13
recommended medial soft-tissue advancement combined with lateral soft-tissue
releases. Stern et al. advised that constrained femoral components be used in
severely valgus knees in which the ligamentous balancing is tenuous, to allow
for easy conversion to a constrained insert if lateonset instability
occurs12.
In 1979, Insall et al. described their technique of soft-tissue
balancing3. In this
technique, the iliotibial band is divided transversely above the joint line
while the lateral aspect of the capsule, the lateral collateral ligament, and
the popliteus tendon are detached from the lateral femoral condyle. The
lateral retinaculum is routinely released longitudinally as well. The senior
one of us (C.S.R.) thought that this technique led to an unacceptably high
rate of late-onset instability, which prompted him to develop a less extensive
soft-tissue release, thereby potentially reducing the need for a constrained
prosthesis4. He
discourages the routine use of a constrained prosthesis, which he believes
should be utilized for only the most complex valgus deformities.
The revised technique, in use since 1985, involves an intra-articular
release, in extension, of the contracted posterolateral aspect of the capsule
in a graduated, stepwise fashion along with so-called pie-crusting of the
iliotibial band4. We
believe that the technique is reproducible and is less technically demanding
than many other procedures, such as a lateral approach or medial soft-tissue
imbrication. In our study, pain relief, joint stability, and correction of the
alignment did not decline with time. In series of total knee replacements
ranging in size from twenty-five to 134, the rates of peroneal palsy and
patellar dislocation have ranged between 1% and 4% for knees treated for
valgus
deformity3,12,14-17.
There were no peroneal palsies in our study.
In conclusion, total knee arthroplasty for the treatment of valgus
deformity requires correction of both osseous and ligamentous abnormalities.
Recent advances in instrumentation have made bone resection and alignment
easier but do not address ligamentous balancing. The soft-tissue release
described herein for valgus deformity is not technically demanding and has
consistently produced excellent long-term clinical and radiographic results.
We recommend the technique for the management of valgus knees undergoing total
knee arthroplasty. ?