39

Success and Failure after Stifle

Stabilization Surgery

Michael G. Conzemiusand Richard B. Evans

Introduction

Balanced and concurrent consideration of suc-
cess and failure after stifle surgery should be
used to define and explain how ‘successful’
a procedure is. To accomplish this, investiga-
tors must: (i) understand how to calculate and
report the successes and failures; (ii) carefully
choose the outcome measure(s) used to mea-
sure success (e.g., owner survey, gait analysis)
and failure (e.g., second surgery); and (iii) eval-
uate and report data in a clinically meaningful
manner. All of this discussion is relative to time
after the intervention. Other factors, such as age,
breed, surgeon experience, owner’s input, can
also influence outcomes.

Balancing success and failure

Since surgery is a comparatively invasive inter-
vention that has inherent risks to the patient,
it is most reasonable to consider the advan-
tages and disadvantages concurrently when
discussing and reporting techniques. Although
there may be disagreement regarding the exact
mathematical method that should be used to
report advantages and disadvantages of an
intervention, the concept remains the same;

benefits and risks associated with an interven-

tion need to be evaluated together. TheNumber

Needed to Treat(NNT) is the number of subjects

that need to be treated to prevent one additional

bad outcome (e.g., limping). If an intervention

has a NNT of 6, it means that six patients

must be treated with that intervention before

one additional bad outcome (e.g., lameness) is

prevented relative to the control group. A NNT

of 1 would suggest that every patient treated

with an intervention improves, and none in the

control group improve. A NNT of 10 would

suggest a high placebo effect. Absolute Risk

Reduction(ARR; arithmetic difference between

groups) is the change in the risk of an outcome,

and is the inverse of NNT (Hutton 2009).

Alternatively, theRelative Risk Reduction(RRR;

proportional difference between groups) can be

calculated.

The likelihood of an adverse event or treat-

ment failure is just as important to the patient

and client as the probability of success. The

Number Needed to Harm(NNH) is the number of

patients that had an adverse event (e.g., implant

removal) because of the intervention that would

not have been exposed to the adverse event if

there were in the control group. A NNH of 1

suggests that 100% of treated patients had an

adverse event, and they would not have that

Advances in the Canine Cranial Cruciate Ligament, Second Edition. Edited by Peter Muir. © 2018 ACVS Foundation.
This Work is a co-publication between the American College of Veterinary Surgeons Foundation and Wiley-Blackwell.

317