Creation of a predictive equation to estimate fat-free mass and the ratio of fat-free mass to skeletal size using morphometry in lean working farm dogs

Creation of a predictive equation to estimate fat-free mass and the ratio of fat-free mass to skeletal size using morphometry in lean working farm dogs
Peer reviewed



To develop an equation that accurately estimates fat-free mass (FFM) and the ratio of FFM to skeletal size or mass, using morphometric measurements in lean working farm dogs, and to examine the association between FFM derived from body condition score (BCS) and FFM measured using isotope dilution.


Thirteen Huntaway and seven Heading working dogs from sheep and beef farms in the Waikato region of New Zealand were recruited based on BCS (BCS < 3, 3–4, > 4) using a nine-point scale. Bodyweight, BCS, and morphometric measurements (head length and circumference, body length, thoracic girth, and fore and hind limb length) were recorded for each dog, and body composition was measured using an isotopic dilution technique. A new variable using morphometric measurements, termed skeletal size, was created using principal component analysis. Models for predicting FFM, leanST (FFM minus skeletal mass) and ratios of FFM and leanST to skeletal size or mass were generated using multiple linear regression analysis.


Mean FFM of the 20 dogs, measured by isotope dilution, was 22.1 (SD 4.4) kg and the percentage FFM of bodyweight was 87.0 (SD 5.0)%. Median BCS was 3.0 (min 1, max 6). Bodyweight, breed, age and skeletal size or mass were associated with measured FFM (p<0.001). There was a good correlation between predicted FFM and measured FFM (R2=0.96), and for the ratio of predicted FFM to skeletal size and measured values (R2=0.99). Correlation coefficients were higher for the ratio FFM and leanST to skeletal size than for ratios using skeletal mass. There was a positive correlation between BCS-derived fat mass as a percentage of bodyweight and fat mass percentage determined using isotope dilution (R2=0.65).


As expected, the predictive equation was accurate in estimating FFM when tested on the same group of dogs used to develop the equation. The significance of breed, independent of skeletal size, in predicting FFM indicates that individual breed formulae may be required. Future studies that apply these equations on a greater population of working Huntaway and Heading dogs are needed to establish the utility of these equations on a large scale. Such studies could ascertain if there is a ratio for lean mass to skeletal size below which the risk of injury or disease increases. If these equations prove useful they would provide an objective and non-invasive measure to determine when welfare in individual dogs is compromised by underfeeding.

Access to the full text of this article is available:

through another providers website:

If you're a member or subscriber and believe you should have access:



Register for an account

Request new password

The whole of the literary matter of the New Zealand Veterinary Journal is copyright Taylor and Francis, Downloading this article signifies agreement with the terms and conditions of electronic access.