Ultrasonographic evaluation of the adrenal glands in healthy dogs: repeatability, reproducibility, observer-dependent variability, and the effect of bodyweight, age and sex.

Adrenal length and width were determined from two-dimensional ultrasound longitudinal images. In study 1, 540 measurements of adrenal glands were attempted from five healthy beagle dogs by three different observers with different levels of expertise in ultrasonography, to determine the variability of adrenal gland measurements. Of these, 484 measurements were included in the statistical analysis, since 16 measurements of the left adrenal gland and 40 for the right could not be visualised by the observer. In study 2, a single measurement of both adrenal glands was taken from each of 146 dogs by the most trained observer from study 1, and the effects of different health status (healthy dogs v dogs with non-adrenal diseases), bodyweight, age and sex were assessed. A total of 267 measurements were included in the statistical analysis. The lowest intra- and inter-day coefficient of variation values were observed for the left adrenal gland and by the most trained observer. The health status had no statistically significant effect on adrenal gland length or width, whereas age had a significant effect only for the left adrenal gland (the greater the age, the greater the width or length) and sex had a significant effect only for the right adrenal gland (the width was larger in males and the length larger in females). The bodyweight had a significant effect for the length of both adrenal glands (the greater the bodyweight, the greater the length), but not the width. The differences between sd and coefficient of variation values for the width of the left adrenal gland were not statistically significant between the three observers, whereas they were statistically significant for the right adrenal gland.