Researchers at the University of Tennessee College of Veterinary Medicine, working with the Gastrointestinal Laboratory at Texas A&M University, enrolled twelve healthy research colony Beagles with no history of gastrointestinal signs and no antimicrobial exposure for at least six months. One dog was excluded early after developing acute intervertebral disc disease unrelated to the trial, leaving eleven animals for analysis. Each dog received azithromycin at 2 mg/kg by mouth every eight hours for eight days, a regimen corresponding to roughly one-half to one-fifth of the dose used for antimicrobial purposes. Fecal samples were collected at baseline, during treatment, and at multiple points across a five-week recovery window, with daily fecal scoring throughout the 45-day study. All dogs acted as their own controls, a design considered appropriate given the well-established temporal stability of the canine fecal microbiome.

The analytical approach combined quantitative PCR measurement of sixteen core bacterial taxa with a validated Dysbiosis Index, the only analytically validated assay for fecal dysbiosis in dogs and cats. In parallel, twenty-nine conjugated, unconjugated, and oxo-isoforms of primary and secondary bile acids were quantified using liquid chromatography-tandem mass spectrometry. The choice of bile acids was not arbitrary. Bile acid metabolism is one of the most sensitive functional readouts of gut microbial health, because a single bacterial species, Peptacetobacter hiranonis, performs the critical conversion of primary bile acids into secondary forms through enzymes encoded by the bai operon. When that conversion falters, it signals more than a shift in microbial census: it signals a loss of metabolic function with potential consequences for the entire intestinal ecosystem.

The results confirmed the researchers’ hypothesis with striking clarity. Even at the reduced prokinetic dose, azithromycin triggered measurable dysbiosis. By day six, the abundances of seven of sixteen core taxa had fallen significantly, including P. hiranonis, Faecalibacterium, and Fusobacterium, three of the most functionally important members of the canine gut community. Faecalibacterium produces short-chain fatty acids that fuel the cells lining the intestine and modulate mucosal immunity, while Fusobacterium is a major butyrate producer that ferments amino acids rather than carbohydrates. P. hiranonis dropped dramatically, with median abundance falling from roughly 6.65 to 1.77 log DNA per gram of feces during treatment. Fecal scores also rose significantly during administration, though they remained within the normal range, suggesting at most a mild, transient loosening of stool.

The metabolic consequences followed directly from the microbial losses. By day eleven, the percentage of primary unconjugated bile acids had risen significantly, with a median of about 94 percent in the most affected animals, while secondary bile acids, lithocholic acid and deoxycholic acid, declined sharply. The strength of the relationship was remarkable: the increase in primary bile acids correlated negatively with P. hiranonis abundance at a coefficient of -0.936, a near-perfect inverse association. This tight coupling makes physiological sense. With the bacterium that performs 7-alpha-dehydroxylation suppressed, primary bile acids accumulated unconverted. Because elevated primary bile acids can stimulate colonic water secretion and alter mucosal permeability, the authors speculate that this metabolic shift may even have contributed to the mild increase in fecal scores observed during treatment.

The good news is that the damage was largely temporary. By day twenty-six, P. hiranonis had normalized in all but one dog, and every animal had recovered by day thirty-six. Fecal bile acid proportions returned to baseline by day forty-five. Of the nine taxa significantly altered by the drug, eight bounced back within the study window. This recovery pattern contrasts favorably with previously published work on tylosin and metronidazole, where sustained elevations in primary bile acids have been documented in individual dogs for up to eight weeks. The authors note, however, that their recovery measurements may have been flattered by the sampling schedule, since bile acids were only measured on days eleven and forty-five, leaving the exact timing of metabolic recovery unknown.

One finding stood out as genuinely atypical. Streptococcus, a group of gram-positive commensals that usually flourishes during dysbiosis thanks to its tolerance of the acidic, substrate-rich conditions antibiotics create, instead declined significantly by day eleven and never recovered for the duration of the study. The pattern defies the standard ecological script. The authors ruled out simple pharmacokinetic explanations: azithromycin’s reported half-life in dogs ranges from 51 to 90 hours, meaning the drug should have been fully eliminated within about nineteen days, and post-antibiotic effects last only hours. The most plausible explanation, they argue, is a lasting ecological disruption within the fecal microbiome, one whose clinical significance in healthy dogs remains uncertain but which deserves attention in future work.

Equally unexpected was the absence of Escherichia coli overgrowth, a hallmark response to metronidazole, tylosin, and amoxicillin in dogs. The researchers attribute this to azithromycin’s distinctive macrolide structure, which confers higher basicity and greater intracellular uptake, enhancing its activity against Enterobacteriaceae. In human medicine, the drug demonstrates excellent activity against common enteropathogenic strains of E. coli. This pharmacological idiosyncrasy also helps explain why the Dysbiosis Index, which normally rises sharply after antibiotic exposure, increased only modestly in this study. Because the two taxa that typically drive the index upward, E. coli and Streptococcus, both decreased instead, the index understated the true extent of dysbiosis. The authors emphasize that the tool must always be interpreted alongside the abundances of all core bacteria, since clear functional change was evident even when the composite score appeared reassuring.

Clinically, the study offers a measured but important message. Short-term low-dose azithromycin was well tolerated in healthy dogs, with no gastrointestinal adverse effects, no changes in body weight or body condition, and normalization of nearly every disturbed parameter within five weeks. Yet the authors caution against extrapolating too readily to the clinic. Healthy research Beagles do not represent the dogs that would actually receive the drug as a prokinetic, and animals with pre-existing dysbiosis, altered bile acid metabolism, or chronic enteropathy could respond very differently. Antimicrobial resistance, which was not evaluated here, remains an additional consideration whenever an antibiotic is deployed for a non-antimicrobial purpose. The prokinetic efficacy of azithromycin in dogs, moreover, has never been rigorously proven, so future dosing strategies may differ from the one tested. What the study establishes is a proof of principle: even at doses one-fifth of those used against infection, azithromycin is not microbiologically silent. It remodels the gut ecosystem and disrupts one of its most essential metabolic functions, and it does so in a pattern that is mostly, though not entirely, reversible. For veterinarians weighing the drug against alternatives, and for researchers designing the next generation of prokinetic trials, that evidence now exists.

Subject of Research: Effects of low-dose azithromycin on the fecal microbiome and fecal bile acids in healthy dogs

Article Title: Effect of low dose azithromycin on the fecal microbiome and fecal bile acids in healthy dogs

Article References: Effect of low dose azithromycin on the fecal microbiome and fecal bile acids in healthy dogs. (n.d.). https://doi.org/10.1093/jvimsj/aalag220

Image Credits: AI Generated

DOI: 10.1093/jvimsj/aalag220

Keywords: azithromycin, dogs, microbiome, bile acids, dysbiosis, prokinetics, Peptacetobacter hiranonis, veterinary medicine, qPCR, gut health, macrolides, Streptococcus