Background– Acute diarrhea (AD) is a common, clinically self-limiting condition in dogs, often associated with increased abundance of pathobionts, but limited information is available on overall microbiome shifts.
Objective– Evaluate fecal microbiota in dogs with AD using qPCR and shotgun metagenomic sequencing. Animals– Multiple timepoints from 78 dogs with AD treated without antibiotics and one sample from 79 healthy controls (HC).
Methods– Retrospective analysis was performed on feces using qPCR (AD, n=78; HC, n=79) for specific bacterial taxa at baseline and up to 1 month. Shotgun metagenomic sequencing was performed on 17 AD and 31 HC. A deterministic machine learning (ML) model was trained to distinguish AD from HC on the qPCR data.
Results– On qPCR, AD showed increased E. coli, C. perfringens, Streptococcus and Fusobacterium (p < 0.05). Faecalibacterium, Bifidobacterium, and Bacteroides were decreased (p < 0.01). C. difficile was detected in 10.6% (5/47) of samples tested via qPCR. Only 4/78 (5%) dogs with AD had decreased P. hiranonis. Sequencing results were similar: C. perfringens and E. coli increased, while P. hiranonis remained unchanged (p = 0.6). The ML model accurately classified HC or AD 92% of the time (AUC: 0.97, 95%CI: 0.94-0.99), driven by increased C. perfringens and decreased Bifidobacterium. The abundance of C. perfringens decreased significantly over time (p < 0.01) in dogs with AD.
Conclusions and Clinical Importance – Acute diarrhea is characterized by transient, self-limiting bacterial shifts, including increases in C. perfringens, with C. difficile playing no significant role. P. hiranonis was not significantly reduced, indicating preservation of bile acid conversion.
