204 / 460
Recession wedge trochleoplasty for major trochlear dysplasia
203
to begin the cut with a rigid osteotome and to
complete it with the saw. The end of the cut
ended approximately 5mm from the sulcus
terminalis, in order to respect the distal
osteochondral hinge and facilitate wedge
closure. Then the anterior slanting cut was
made so as to join the first two cuts. Osteotomy
was completed and correction obtained after
removal of the proximally based wedge by
progressively applying digital pressure to the
trochlea. The amount of bone removed was just
enough to allow the trochlea to settle into a
deeper position, without modifying the
trochlear groove. Fixation used 3.5mm
cancellous screws, positioned just laterally to
the cartilage surface with heads which could be
countersunk into the bone. Initially, 3 cancellous
screws (2 lateral, 1 medial) were used; this was
later changed to just 2 lateral screws, without
any problem to date. Patients were managed in
an extension brace for the first 6 weeks
postoperatively. Theywere allowed tomobilize,
bearing full weight on crutches. The protocol
included restricting knee flexion to 0-60° for
the first 3 postoperative weeks, and then slowly
increasing the range of motion by daily
increments up to 90° at the 6
th
week.
Postoperative assessment
Postoperative assessment included clinical
examination, assessment of symptoms (pain)
and functional scores. Patients were followed
up with plain radiographs (standing AP, lateral
and Merchant views). A physical exam was
performed by an independent observer; KOOS
[12], IKDC and Kujala [13] scores were used.
Different subgroups of patients were compared
to identify predictive factors for success of
trochleoplasty: with or without prior surgery;
patellofemoral chondropathy observed during
surgery; degenerative change on preoperative
radiographs. Patellofemoral cartilage damage
was assessed using the ICRS classification.
Patellofemoral chondropathy at the time of the
index operation was defined as an abnormal
(grade 2: lesions extending down to <50% of
cartilage depth) or severely abnormal aspect of
the cartilage (grade 3: lesions extending down
to >50% of cartilage depth or to calcified layer
or blisters; or grade 4: osteochondral injury,
lesions extending just through the subchondral
bone-plate, or deeper defects down into
trabecular bone) [14].
Patellar height and trochlear dysplasia were
assessed on the lateral view. Measurements
were performed by the same independent
observer on a digitized version of the lateral
X-rays, according to the technique described
by Dejour [7] (fig. 2). PTA was assessed on the
Merchant view [8].
Results
Seven of the 19 knees had had previous surgery
for instability (Table 1). Symptom duration
prior to the index operation ranged from 1 to
22 years (mean, 11 years). Trochlear dysplasia
was classified as grade A in 1 case, grade B in
7, grade C in 5 and grade D in 6. The
patellofemoral articular surface was considered
macroscopically normal at the time of operation
in 14 knees and abnormal in 5 on the ICRS
classification; lesions were grade 3 on the
patella (n=3), grade 2 on the patella and lateral
trochlea (n=1), and grade 2 on the patella (n=1).
At the time of operation, all knees required
additional procedures to correct instability
factors (Table 2). Thirteen knees underwent TT
medialization and distalization, 1 distalization
only, 2 medialization only and 2 underwent TT
lateralization to correct negative TT-TG caused
by a previous surgery. Eight knees underwent
MPFLreconstruction according to the technique
described by Fithian
et al.
[15]. Minimum
follow-up was 12 months, with a mean of 34
months (range, 12 to 71).
Complications
There were no intra- or peri-operative adverse
events such as nerve injury, vascular problems
or deep venous thrombosis. No patients showed
non-union of the osteochondral block. One
patient required arthrolysis for knee stiffness
1 year after the index operation: knee flexion
before the second operation was 100°, and 140°
after arthroscopic arthrolysis associated to