104 / 460
Clinical examination of the patellofemoral joint
103
anterior superior iliac spine. With the knee at
90º callipers are placed across the epicondyles
and a piece of string passed from the proximal
reference point over the knee at the midpoint of
the callipers and continues in a straight line
down the tibia. The transverse distance between
the centre of the tibial tubercle and the string is
the TTTG.
J-sign
The patient sits on the edge of the couch with
the knee in full extension. They then actively
flex the knee fully. The examiner observes the
tracking of the patella. A positive J-sign occurs
when the patella moves from lateral to medial
in the first 20º of knee flexion.
Smith
et al’s
[2] review suggested that patellar
instability can be confirmed with all of these
tests. However the sensitivity-specificity and
the reliability-validity of the test were open to
doubt. Surprisingly, despite the extent of the
literature on the Q-angle, and patellar medio
lateral orientation and position, little of this
pertained to patellar instability populations. It
was concluded that, despite examination being
a cornerstone in diagnosis, the evidence for the
utility of these tests of patellar instability was
methodologically flawed and there was
insufficient evidence to support their use. As a
result a study was set up to address this.
Intra- and inter-
observer reliability of
patellar instability
tests
[3]
Under the auspices of the International
Patellofemoral Study Group (IPSG), five
patients with known patellar instability were
recruited at the Norfolk & Norwich University
Hospital. They were all female with an average
age of 27 years (18 to 38 years) with a mean
duration of symptoms of 11 years (2 to 26 years)
and a mean body mass index of 22 (19 to 28).
Two had had previous patellar stabilisation
procedures, and all had bilateral symptoms.
Five experienced clinicians from the IPSG
were invited to take part in the study, which
was ethically approved. The clinicians were
instructed only to perform the tests that they
routinely used in their practiceswhen evaluating
a patient with patellar instability. Each patient
was randomly allocated to a clinician who then
performed and recorded the results of the tests,
screened from the other examiners.
After a break, the clinicians then repeated the
tests with further randomisation to the patient
order.
Assessment of inter- and intra-observer
reliability was made using Kappa analysis by an
independent statistician. These were interpreted
as showing no agreement when <0.00, slight
agreement at 0.00 to 0.20, fair at 0.21 to 0.40,
moderate at 0.41 to 0.60, substantial at 0.61 to
0.80, and almost perfect at 0.81 to 1.00.
Results
The inter-observer agreement was very poor for
most of the tests. Fair-to-moderate agreement
was found for the J-sign, patellofemoral
crepitus, and assessment of foot arch position.
The inter-observer reliability was greatest for
more generic lower limb examinations (foot
arch position, patellofemoral crepitus, and pain
on palpation of the medial retinaculum) rather
than more specialist tests (Q-angle, gravity
subluxation test, and tubercle-sulcus angle).
Intra-observer reliability showed moderate-to-
substantial agreement, particularly for
alignment, and observational & palpation tests
(tibial torsion, Bassett’s sign, Q-angle
performed at 30 degrees flexion).
Discussion
Inter-observer reliability was best for more
general examination tests and qualitative
assessments better than quantitative i.e. tibial
torsion “abnormal” as against “n degrees”.