Success and Failure after Stifle Stabilization Surgery 319because of low statistical power. Inconsistencies
in delivery and interpretation of the question-
naire results in sizeable variabilities that can
only be overcome by employing large study
populations. In addition, it is necessary to be
cognizant of the caregiver placebo effect that
undoubtedly influences responses from owners
and other caregivers (Conzemius & Evans
2012).
Veterinary examination
Before surgery, a clinician may assign a subjec-
tive score to limb function based on assessment
of dog’s level of pain, lameness, and posture
while sitting. While each clinician differs to
some degree, observations when the patient
is walking and/or trotting may include a dip
in the patient’s hip or head, degree of limb
carriage, and stride length. During physical
examination the clinician may note the joint’s
range of motion, the amount of muscle mass,
and even if the leg is positioned ‘normally’
when sitting. All of these can contribute to a
determination of the patient’s quality of limb
function. Documentation of findings are gener-
ally recorded via a numeric rating score (NRS),
where numbers are assigned to each finding,
or via a visual analog scale (VAS) where limb
function is assigned by marking a line with one
end of the line representing ‘clinically normal’
(sound) and the other end representing ‘could
not be more lame’ (i.e., non weight-bearing)
(Waxman et al. 2008). The NRS is limited
because there are a restricted number of groups
(four or five) within a classification, and dogs
within a group can have appreciable differences
in their lameness severity (Waxmanet al. 2008).
A VAS does not have this limitation because it
provides a continuous scale that may be easier
to handle statistically. Regardless, both schemes
provide only the opinion of the observer. Sev-
eral manuscripts have tried to offset these
limitations by including multiple, trained
observers. Unfortunately, in one publication
which tried to validate clinician observation of
gait, neither trained nor untrained observers
could reliably identify lameness. There were
large disagreements between individuals.
Untrained observers (first-semester veterinary
students) had the same visual acuity for dog
lameness as boarded surgeons (Waxmanet al.
2008). The only saving grace was a finding that
trained clinicians provided repeatable data;
that is, they consistently made the same mis-
take. This would allow a clinician to compare
groups over time, but not necessarily to com-
ment on the success or failure of an individual
dog.Pedometers and accelerometers
The greatest limitation to data collected at a vet-
erinary hospital is that it measures a moment
in time, and not the day-to-day activity of the
dog at home. A pedometer or accelerometer can
measure patient activity level at home over an
extended period of time. In one study, pedome-
ters were successfully used to measure physical
activity in dogs over a 14-day period (Chanet al.
2005). The pedometer accuracy varied depend-
ing on the patient’s size (it overestimated walk-
ing in large dogs and underestimated walking
in small dogs), but correlated well with over-
all reports of the dog’s activity level at home
and the dog’s condition body score. Accelerom-
eters are more sophisticated in that some can
measure changes in acceleration in the x-, y-,
and z- axes. Thus, body movement in any direc-
tion is measured. In one study that determined
variability in accelerometer data in companion
dogs, large day-to-day, and even week-to-week,
variations occurred in dogs, but within dogs, a
full 7-day comparison of total activity counts
from one week to the next provided the least
variable estimate of the dogs’ activity (Dowet al.
2009). It was also reported that accelerometers
might be most useful for documenting changes
in the dog’s activity over time.
While pedometers and accelerometers esti-
mate patient activity, to date, there is little infor-
mation available as to whether such data can be
translated to something clinically relevant, such
as distance traveled and/or activity intensity.
For example, it may currently be found that a
patient had an increase in accelerometer counts
of 20%, but what exactly does that mean? In
the present authors’ opinion, if this meant that
the dog was willing to travel 20% more dis-
tance or performed 20% more intense activities
(e.g., stairs, jumping), it would much more
beneficial.