Intraobserver and interobserver agreement of 8 segmental reflexes in healthy dogs

Background: No available literature supports the claim that the patellar and withdrawal (flexor) reflexes are the only reliable segmental reflexes in dogs.

Objective: Measure intra- and interobserver agreement of 8 segmental reflexes in dogs without clinical evidence of orthopedic or neurologic disease.

Animals: One-hundred and one client- or staff-owned dogs between 1 and 10 years of age with no clinical evidence of orthopedic disease, myelopathy, or neuromuscular disease.

Methods: Descriptive study. The intraobserver proportion of agreement (%) of responses to selected segmental reflexes in right versus left limbs by 3 observers was calculated and reported. The interobserver agreement of 2 observers of responses to selected reflexes was estimated by calculating proportions of agreement, kappa values, and 95% confidence intervals. A segmental reflex with an acceptable agreement was defined as that with a proportion of agreement ≥90% and a Kappa value ≥0.61 in both limbs.

Results: The intraobserver proportion of agreement for all 3 observers was high (≥95%) for the extensor carpi radialis, withdrawal, patellar, and cranial tibial reflexes. Between observers 1 and 3 and observers 2 and 3, the interobserver proportion of agreement was high (≥ 92%) for the extensor carpi radialis (κ 0.66, not determined [ND]), withdrawal (both limbs, κ ND), patellar (κ ND), and cranial tibial reflexes (κ ND).

Conclusions and clinical importance: The extensor carpi radialis, withdrawal, patellar, and cranial tibial reflexes had a higher proportion of agreement and kappa values between 2 observers.