To report a standardized method of, and determine inter- and intratester reliability for, anthropometric assessment of limb circumference in dogs.
Prospective blinded study.
Labrador Retrievers (n = 20).
Unsedated dogs were manually restrained in lateral recumbency and triplicate measurements of limb circumference at the level of the proximal antebrachium, mid brachium, proximal crus, and mid-thigh were made using the Gulick II tape measure in the morning and afternoon of the same day. Observers were blinded to measurements made during each occasion and those made by co-observers. Estimates of inter- and intratester reliability were made for first and mean measurements using intra-class correlation coefficients (ICC).
Measurements of the proximal antebrachium were made with moderate to fair intratester reliability by all observers with ICC's ranging from 0.68-0.78 (1st measurement) and 0.67-0.78 (mean measurement), and moderate to fair intertester reliability with ICC's of 0.66-0.68 (1st measurement) and 0.70-0.72 (mean measurement). Measurements of the brachium, crus, and thigh typically had poor inter- and intratester reliability, ICC < 0.5.
Using the described method of muscle measurement in Labrador Retrievers only measurement of the proximal antebrachium was reliable; a single (1st) measurement was as reliable as using the mean of triplicate measurements.