404 / 460
Magnetic Resonance Imaging in Patellofemoral Instability
403
condylar measurement is ideal for evaluating
patellar instability is still muddled. While our
standard measurement of the lateral condylar
supports its biomechanical role in resisting
lateral displacement of the patella, the
significance of the central and medial condyles
may reflect a lack of condylar development or
failure to form trochlear groove.
TTTG has been well researched in patients
with patellar instability since Judet
et al.
first
established the measurement [1, 5, 8, 19, 22].
Investigators have had mixed results regarding
the validity or reliability of MRI-based TTTG
measurements compared to CT [19, 22]. Our
mean values (Controls 10.96mm±0.39; PFJDs
18.69mm±0.81) reflect the cutoff of 20mm as
recommended by Dejour,
et al.
[5]. TTTG is a
measurement that is classically based in axial
radiographs and CT scans, but as Schoettle
et
al.
demonstrated, this measurement can reliably
be applied to MRI [22] with similar cutoffs.
Our research further corroborates this.
Axial radiographs and CT imaging of the
patellofemoral joint is well documented and is
the basis for many of our current diagnostic
measurements. MRI imaging has the distinct
advantage of revealing chondral morphology,
which more accurately depicts the patellar-
trochlear relationship in comparison to
subchondral bone morphology (as would be
seen by CT scan and radiography) [15-17].
Recent MR-based studies have begun to show
that although MRI is accurate (in that the
measured values are reproducible), the
preciseness between CT and MRI varies with
respect to previously established pathological
patellofemoral measures, such as sulcus angle
for example [17, 18]. In contrast, other
measures, such as Trochlear Groove Depth,
have remained similar to CT based measures
[9, 18, 21]. Given these mixed results, one
could question the practice of applying cut-off
values obtained from CT based studies to MRI
measurements. Our research shows that some
of these classic CT based measurements change
while others remain the same when working
with MRI imaging.
Conclusions and future
directions
By detecting significant morphological dif
ferences between the two groups, our findings
justify the use of MRI to obtain many of the
measures of patellofemoral instability histo
rically obtained with CT scan and plain
radiographs. In fact, measures of all the four
recognized factors of patellar instability were
found significant. This demonstrates that
patellofemoral instability is a result of multiple
factors, with instances of some small changes
over many measurements or a large change in a
few key measurements. Patellar tilt measures,
such as Angle of Fulkerson, proved to be an
excellent group ofmeasurements for delineating
between Controls and those with instability.
Patella alta ratios, such as Insall-Salvati and
Caton-Deschamps, demonstrated statistically
significant difference between controls and
recurrent dislocators that coincided with
established CT cutoff. Trochlear morphology
measures such as Sulcus Angle, trochlear
groove depth, and lateral trochlear inclination
demonstrated statistical significance, though
Sulcus Angle and Lateral Trochlear Inclination
did differ from established values. The next
logical step for research is to pursue statistical
analysis of our data to create established cutoffs
for MRI as previous groups have with CT and
XR and apply them to prospective trials in
order to establish which measurements remain
a reliable delineators between normal knees
and those with patellofemoral instability. By
replacing the need for CT, the recurrent
patellofemoralpatientisexposedtosignificantly
less radiation. As such, MRI is an appropriate
tool to aid the clinician in obtaining the
radiographic information that would have been
obtained by CT scan in patients with recurrent
patellofemoral instability.