Carbon and nitrogen isotopic composition of commercial dog food in Brazil

Abstract

Background Brazil is a low- to medium-income country and has the second largest pet food market in the world with 8% of world pet food consumption. The lowest-income social class spends around 17% of their domestic budget on pet food and other items related to pets. Consumers are frequently misled by advertising as there is no precise information about the main sources of protein, carbohydrates and fat in the labels, and the Brazilian pet food industry can legally claim that their products contain certain items like salmon or beef even if they use just a flavoring compound. Methods The stable isotope methodology compares the stable isotope ratios of carbon (13C/12C) and nitrogen (15N/14N) between source and product. The 13C/12C ratio of a specific product (e.g., dog food) reveals the proportions of C4 (maize) and C3 (soybean, rice and wheat) plants in that product and the 15N/14N ratio reveals the proportion of the compounds derived from animals. With this isotopic data, we used MixSIAR, a Bayesian stable isotope-mixing model, to estimate the proportion of maize, grains, poultry and beef in dog food. Results The δ13C values of dry dog food ranged from −24.2‰ to −12.8‰, with an average (± standard-deviation) of −17.1‰ ± 2.8‰. The δ13C values of wet pet food ranged from −25.4‰ to −16.9‰, with an average (± standard-deviation) of −21.2‰ ± 2.4‰, which was significantly lower (p < 0.01). The δ15N values of the dry and wet food ranged from 1.7‰ to 4.2‰, and from 0.5‰ to 5.5‰, respectively. The average δ15N values of dry food (2.9‰ ± 0.5‰) was not higher than the wet food (2.6‰ ± 1.3‰) (p > 0.01). The output of the MixSIAR showed a low proportion of bovine products in dry dog food samples. On the other hand, poultry was obviously the dominant ingredient present in most of the samples. Maize was the second dominant ingredient. Wet and dry dog food showed similar isotopic analysis results. The only difference was a lower proportion of maize and higher proportion of grains in wet dog food. Discussion The main finding is that dog food in Brazil is mostly made of approximately 60% (ranging from 32% to 86%) animal-based and 40% (ranging from 14% to 67%) plant-based products. Poultry and maize are the main ingredients. Poultry is added as a by-product or meal, which avoids competition between dogs and humans for meat products, while they can compete for maize. On the other hand, a large proportion of plant-based products in dog food decreases the energy and environmental footprint, since plant-based food products tend to be less harmful compared to animal-based products. Labels can mislead consumers by showing pictures of items that are not necessarily part of the product composition and by not showing the detailed information on the proportion of each ingredient. This information would allow customers to make their own choices considering their pet’s nutrition, the competition between animals and humans for resources and environmental sustainability.

High variability within pet foods prevents the identification of native species in pet cats’ diets using isotopic evaluation

“…In an animal that eats only one or a select few food sources, such as a grazing Merino sheep, the δ 13 C value range is 0.84 (Männel, Auerswald & Schnyder, 2007). The ranges in food isotope values we obtained are very similar to those obtained in a recent Brazilian study looking at the contents of food for domestic dogs (Galera et al, 2019). The cats in our study, from both the US and UK, had δ 13 C value ranges of around 6 .…”

“…Because most pet cats are consistently fed one kind of food at home, and because many pet foods are thought to include corn or corn-fed livestock (Schnepf, 2011;Galera et al, 2019), we expected that a stable isotope approach would have a high power to discern pets consuming native prey from those eating only pet food. However, we found that the high variability of isotope values for pet foods, across and within brands, not to mention across countries, makes it extremely difficult to determine the diet of individual cats.…”

“…We received most of our samples of cat food directly from owners and thus only had broad descriptions of the flavors from the volunteers who sent them. Having access to the labels with the nutrition information and ingredients may have helped to understand the variability between foods, however many pet food labels lack specific information about the main sources of carbohydrates, protein, and fat provided and can be misleading (Galera et al, 2019). Among dry food alone there was a range of nearly 10 in δ 13 C values.…”