Processing of corn-based dog foods through pelleting, baking and extrusion and their effect on apparent total tract digestibility and colonic health of adult dogs

Different food processing parameters may alter starch granule structure and its cooking degree. With lower thermomechanical energy, more resistant starch (RS) is retained in the food, which may benefit gastrointestinal (GI) health. The objective of this study was to determine the effect of food processing on dietary utilization and dog gut health. Experimental diets containing 56% corn as the sole starch source were produced through pelleting, baking, and extrusion and compared to a baked control diet in which the corn was replaced with dextrose. The extruded diet resulted in the highest level (P < 0.05) of in vitro starch cook and lowest RS, while baked was intermediate and pelleted had the lowest starch cook and highest RS. To evaluate the in vivo effects of these treatments, 12 dogs were adapted to foods for 9 d, and feces were collected for 5 d in a replicated 4 × 4 Latin square design. Feces were scored for consistency using an ordinal scale, and parametric data included apparent digestibility (ATTD), parameters indicative of gut health, and the microbial composition, which was centered log-ratio transformed before operational taxonomic unit (OTU) analyses. Fecal scores were analyzed by ordinal logistic regression, and parametric data were analyzed as mixed models. Overall ATTD was greater (P < 0.05) in extruded, followed by baked and pelleted. Dogs fed the control had osmotic diarrhea, whereas dogs fed the other treatments had mostly acceptable fecal scores, with extrusion leading to the best fecal quality. The control also led to high fecal pH and low SCFAs, indicating dysbiosis. All corn foods had similar (P > 0.05) fecal SCFAs and extruded tended (P = 0.055) to promote higher fecal butyrate than baked and pelleted. The microbiome of dogs fed the corn foods had similar α diversity indices, and OTUs at the species and phyla levels were mostly alike and different from the control. In conclusion, the higher levels of in vitro RS did not translate into a better in vivo fermentation profile, and extruded kibble performed best regarding fecal quality, ATTD, and fecal SCFAs.