Variation in Macronutrient Composition of Popular, Premium, and Clinical Canine Diets Fed to Client-Owned Ostearthritic Dogs

The objective of this study was to compare the guaranteed analysis (GA) and analyzed proximate analysis measurements of commercial diets fed to a cohort of client-owned osteoarthritic dogs. These data were derived from a larger study designed to evaluate the demographics, dietary and medical history, and exercise regimen of dogs with osteoarthritis. Owners also were asked to provide a sample and label of the diet fed to their dog. A total of 51 different diets were collected and subsequently categorized into three main categories: Popular (n=11), Premium (n=32), and Clinical (n=8). Our hypothesis was that the difference between GA and analyzed values would be greatest in the ‘Popular’ brand category due to the larger variability in ingredient sources based on minimum-cost formulations. All diets were analyzed for dry matter (DM) and ash (AOAC, 2006), crude fiber (AOAC, 2002), acid-hydrolyzed fat (AACC, 1983 and Budde, 1952). Crude protein (CP) was calculated from Leco total nitrogen values (AOAC, 2006). Analyzed DM concentrations conformed to the GA for all 51 diets. That is, the average difference (mean ± SD) between analyzed and GA values were below the maximum guaranteed on the labels: Popular dry (-3.34% ± 1.20) and wet (-5.35% ± 0.82), Premium dry (-3.77% ± 1.36) and wet (-2.70%), and Clinical (-2.45% ± 1.13). With the exception of one Popular diet (-0.72% less than GA), CP value differences (DM basis) of Popular (4.26 ± 2.03), Premium (4.08 ± 2.28), and Clinical (3.84 ± 1.74) diets all conformed to the guaranteed minimum values. Crude fiber differences on a DM basis were the most variable, with 9%, 44%, and 75% of Popular, Premium, and Clinical diets respectively, failing to conform to the guaranteed maximum values. Similarly, differences between analyzed acid hydrolyzed fat vs. crude fat GA on a DM basis were also highly variable, with 18%, 31%, and 25% of Popular, Premium, and Clinical diets respectively, failing to conform to the guaranteed minimum values. From this information, it cannot be concluded that ‘Popular’ brand diets have the greatest overall deviations from the GA; however, it can be concluded that measurements for crude fiber are the most variable macronutrient profile. The results of this study indicate that crude fiber is too variable of a measurement to be used as the industry standard for pet food diets.