Skip to main content
NCBI home page
Journal of Veterinary Internal Medicine logoLink to Journal of Veterinary Internal Medicine
. 2026 Sep 27;40(5):aalag220. doi: 10.1093/jvimsj/aalag220

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

Colin Smith 1, Roman Husnik 2,✉, Frederic Gaschen 3, Claudia Kirk 4, Xiaojuan Zhu 5, Jan Suchodolski 6
PMCID: PMC13616115  PMID: 42801332

Abstract

Background

Azithromycin has prokinetic effects at low doses in people, cats, and anecdotally, in dogs; however, antibiotics alter the fecal microbiome and metabolome.

Hypothesis/Objectives

To prospectively evaluate the effects of low-dose azithromycin, hypothesizing that it would alter the fecal microbiome and fecal bile acids in healthy dogs.

Animals

Eleven healthy research dogs.

Methods

Prospective study. Dogs received azithromycin (2 mg/kg PO q8h) for 8 days. Fecal samples were collected before (day 1), during (day 6), and after azithromycin administration (days 11, 26, 36, 45). Fecal consistency was assessed daily. Microbiome analysis included qPCR quantification of 16 core bacterial taxa and the Dysbiosis Index. Fecal concentrations of conjugated and unconjugated bile acids (UBAs) were measured on days 1, 11, and 45.

Results

Fecal scores increased during azithromycin administration (P = .009). Abundance of 7 of 16 taxa decreased significantly on day 6, including Peptacetobacter hiranonis (P. hiranonis; P < .001), Faecalibacterium (P < .001), and Fusobacterium (P = .031). Abundance of Streptococcus significantly decreased on day 11 (P = .003). Percentage of primary UBAs increased (P = .017; median, 94.06%; range, 0.73%-96.08%), returned to baseline by day 45, and negatively correlated with abundance of P. hiranonis (r = -0.936 [95% CI, -0.973, -0.852], P < .001). Abundance of 8 of 9 significantly altered taxa recovered, with abundance of Streptococcus remaining decreased.

Conclusions and clinical importance

Low-dose azithromycin induces a mostly reversible dysbiosis and reversible decrease in secondary fecal UBAs. Atypical decreases in Streptococcus abundance can also occur.

Keywords: dysbiosis, metabolome, motility, Peptacetobacter hiranonis, prokinetics

Introduction

Gastrointestinal motility disorders are an underestimated cause of morbidity in dogs and can be primary or, more commonly, secondary to systemic or primary gastrointestinal disease.1,2 Treatment involves addressing the underlying cause and often concurrent use of prokinetics.1,2

Common prokinetics include serotonergic drugs and motilin agonists. Motilin agonists mimic endogenous motilin, which initiates phase III of interdigestive motility.2,3 Macrolides such as azithromycin and erythromycin bind motilin receptors and exert prokinetic effects at low doses. Erythromycin is well-established as a prokinetic in people and has efficacy in dogs and cats.4–8 However, the use of erythromycin in people is limited by cytochrome P450 inhibition and arrhythmogenic potential.9,10 Thus, azithromycin has been explored as an alternative, with limited studies demonstrating comparable prokinetic effects and a better safety profile.11–13 Low-dose azithromycin has established evidence in cats and clinical support in dogs, with prokinetic doses approximately 1/2-1/5 those used for antimicrobial purposes.2,14,15

Despite lower prokinetic dosing, alterations to the gut microbiome with azithromycin remain possible, as seen with metronidazole, tylosin, amoxicillin, and the combination of enrofloxacin and metronidazole, administered at antimicrobial doses in dogs.16–26 Reported effects include reduced species richness and diversity, increased Dysbiosis Index (DI), and decreased Peptacetobacter hiranonis, resulting in reduced conversion of unconjugated primary to secondary bile acids (BAs), a major functional indicator of microbial activity and homeostasis. Furthermore, in some dogs these changes can persist beyond administration and, with select antibiotics, lead to incomplete recovery.16,17,21–23,26

In people, low-dose azithromycin is used in the treatment of gastroparesis and chronic pulmonary disease, with doses approximately half that used for acute infection.27–30 Even at these lower doses, it reduces bacterial richness and diversity and has functional consequences including decreased short-chain fatty acid (SCFA) production, suppression of beneficial metabolic pathways, and reduced anti-inflammatory cytokines.31

Despite clinical use, no studies have assessed the effects of low-dose azithromycin on the fecal microbiome or metabolome in dogs. Given its use as a prokinetic and importance as an antimicrobial, understanding these effects is clinically relevant. Therefore, the purpose of this study was to evaluate the effects of low-dose azithromycin on the fecal microbiome and fecal BAs in healthy dogs. We hypothesized that its administration would induce alterations in fecal microbial composition and BA metabolism.

Materials and methods

This study was approved by the Institutional Animal Care and Use Committee at the University of Tennessee. Twelve healthy research Beagles with no gastrointestinal signs or antimicrobial exposure within 6 months were enrolled. Dogs were individually housed, fed a standardized diet, and received prophylactic deworming before study initiation.

A 45-day prospective study was performed with dogs receiving low-dose azithromycin (2 mg/kg PO q8h) for 8 days (day 3-10). Fecal samples were collected at baseline (day 1), during (day 6), and post-azithromycin administration (day 11, 26, 36, and 45), and fecal scores performed daily. Fecal samples were stored at -80 °C and submitted to the Texas A&M Gastrointestinal Laboratory, where all analyses were performed. Schematic representation of the study timeline is noted in Figure 1. Body weight and body condition score (BCS) were collected at baseline and at the end of the study.

Figure 1.

Schematic representation of the study timeline. Azithromycin (2 mg/kg PO q8h) was administered from days 3--10. Fecal samples were collected on days 1, 6, 11, 26, 36, and 45. Quantitative PCR analysis of 16 core bacterial taxa and calculation of the DI were performed on all sample days. Fecal primary and secondary UBA concentrations were assessed on days 1, 11, and 45. Abbreviations: UBA, unconjugated bile acid; DI, dysbiosis index.

Open in a new tab

Schematic representation of the study timeline. Azithromycin (2 mg/kg PO q8h) was administered from days 3-10. Fecal samples were collected on days 1, 6, 11, 26, 36, and 45. Quantitative PCR analysis of 16 core bacterial taxa and calculation of the DI were performed on all sample days. Fecal primary and secondary UBA concentrations were assessed on days 1, 11, and 45. Abbreviations: UBA = unconjugated bile acid; DI = dysbiosis index.

Quantitative PCR (qPCR) was used to measure the abundance of 16 core bacterial taxa using methods previously described.32 The DI was then derived using select taxa, following established methods.33 Quantification was performed on select days (day 1, 6, 11, 26, 36, and 45).

A total of 29 conjugated, unconjugated, and oxo-isoforms of primary and secondary BAs were measured using a validated liquid chromatography–tandem mass spectrometry method.34 Quantification was performed on select days (day 1, 11, and 45).

Data were assessed for normality and reported as mean ± standard deviation for normal data or median and range for non-normal data. Changes over time were analyzed with Friedman tests and adjusted using Benjamini-Hochberg’s False Discovery Rate. Post hoc pairwise comparisons were performed using Bonferroni correction method. Related-Samples Wilcoxon Signed Rank Test evaluated pre and post-values in weight and BCS. Analyses were performed in IBM SPSS v28, and figures generated in GraphPad Prism v7. Correlation analysis used repeated measures correlation by R version 4.5.1 rmcorr package version 0.7.0.35 Significance was set at P < .05.

Comprehensive description of the materials and methods can be found in the Appendix.

Results

Study sample

A total of 12 dogs were enrolled; however, 1 dog was excluded due to the development of acute thoracolumbar intervertebral disc disease unrelated to the study.

Five dogs were spayed females and 6 dogs neutered males. The mean age was 3.3 ± 0.8 years and initial mean body weight 12.1 ± 1.6 kg. Median initial BCS was 5/9 (range 4-6). Pre-study physical examination and lab work (complete blood count, biochemistry, and urinalysis) were performed in all dogs and yielded no clinically relevant abnormalities. All dogs had a negative fecal flotation and normal fecal scores (Purina Fecal Score [PFS]: 2/7; range 2-3) before the study.

Study period

No gastrointestinal adverse effects were noted throughout the study period, and post-study physical examinations were unchanged from baseline. The mean body weight at the end of the study was 12.0 ± 1.7 kg, and median BCS 4/9 (range 4-6). There were no significant differences in body weight or BCS at the end of the study compared to baseline.

A significant increase in median fecal score occurred on day 6 (PFS: 3/7; range 2-3; P = .009) but returned to baseline on day 11 (PFS: 2/7; range 2-3). Subsequent median fecal scores remained unchanged for the remainder of the study.

Canine microbiota DI

All dogs had a normal DI at baseline (DI < 0), as shown in Figure 2. The median DI was significantly lower (P = .002) on day 26 compared to baseline, but otherwise was not significantly different at any other time point.

Figure 2.

Dysbiosis index at baseline (day 1), during (day 6), and post-azithromycin administration (day 11, 26, 36, and 45) in eleven healthy dogs. DI values less than 0: normal DI indicating no shifts in the overall diversity of the intestinal microbiota; however, individual bacterial groups outside the reference interval are suggestive of mild dysbiosis; 0-2: mild to moderate shift in the overall diversity of the intestinal microbiota; and greater than 2: significant dysbiosis. Azithromycin (2 mg/kg PO q8h) was administered on days 3--10. Thick horizontal bars represent median values at each time point (day 1, 6, 11, 26, 36, and 45) and thin horizontal bars with an asterisk (*) represent a significant difference (p less than 0.05) compared to baseline (day 1). The gray-shaded region denotes the reference range (DI less than 0). Abbreviations: DI, dysbiosis index.

Open in a new tab

Dysbiosis index at baseline (day 1), during (day 6), and post-azithromycin administration (day 11, 26, 36, and 45) in eleven healthy dogs. DI values <0: Normal DI indicating no shifts in the overall diversity of the intestinal microbiota; however, individual bacterial groups outside the reference interval are suggestive of mild dysbiosis; 0-2: Mild to moderate shift in the overall diversity of the intestinal microbiota; and >2: Significant dysbiosis. Azithromycin (2 mg/kg PO q8h) was administered on days 3-10. Thick horizontal bars represent median values at each time point (day 1, 6, 11, 26, 36, and 45) and thin horizontal bars with an asterisk (*) represent a significant difference (P < .05) compared to baseline (day 1). The gray-shaded region denotes the reference range (DI < 0). Abbreviations: DI = Dysbiosis Index.

Dysbiosis index bacterial taxa

The abundances of 7 bacterial taxa included in the DI on days 1, 6, 11, 26, 36, and 45 are presented in Table 1. Abundance over time for P. hiranonis is shown in Figure 3 and the remaining DI taxa in Figure 4. P. hiranonis abundance significantly decreased on day 6 (P < .001) and remained decreased on day 11 (P = .01). By day 26, all but 1 dog had normalized, and all dogs recovered by day 36. Faecalibacterium abundance significantly decreased on day 6 (P < .001), with most dogs falling below the reference range (Figure 4A). It remained significantly decreased on days 11 and 26 (P < .001) but had returned within the reference range, and increased by day 36, though remained lower relative to baseline. Fusobacterium abundance significantly decreased on day 6 (P = .03), falling below the reference range, and increased on day 11 but was lower than baseline (Figure 4B). Fusobacterium abundance returned to baseline levels on day 26. Streptococcus abundance became significantly lower on day 11 (P = .003), and remained significantly reduced for the duration of the study without recovery (Figure 4F). Escherichia coli abundance did not differ significantly at any time point (Figure 4E). Turicibacter abundance remained statistically unchanged throughout the study, though at multiple time points exceeded the upper limit of the reference range (Figure 4C). Blautia abundance did not change significantly at any point (Figure 4D).

Table 1.

Fecal abundances of DI bacterial taxa.

P. hiranonis Blautia Faecalibacterium Fusobacterium Turicibacter E. coli Streptococcus
Reference interval 5.1-7.1 9.5-11.0 3.4-8.0 7.0-11.3 4.6-8.1 0.9-8.0 1.9-8.0
Day 1 6.65
(6.34-7.08)		 10.51
(9.98-10.74)		 5.79
(5.19-6.96)		 8.45
(7.78-9.24)		 8.34
(7.35-8.81)		 4.09
(2.08-5.56)		 7.19
(5.42-8.22)
Day 6 1.77*
(0.1-5.38)		 9.39
(9.16-10.19)		 3.16*
(1.39-3.42)		 6.79*
(6.08-8.98)		 8.63
(8.17-9.07)		 1.8
(1.39-3.83)		 4.55
(3.59-6.9)
Day 11 1.44*
(1.22-6.35)		 9.68
(9.44-10.94)		 3.82*
(2.59-4.4)		 7.21
(6.82-7.42)		 7.9
(6.00-8.8)		 2.06
(1.59-3.3)		 3.92*
(2.6-5.3)
Day 26 6.66
(3.9-6.78)		 10.77
(10.48-11.06)		 3.73*
(3.44-4.51)		 8.59
(7.43-9.89)		 8.17
(6.28-9.13)		 4.29
(2.54-6.63)		 2.95*
(2.72-3.86)
Day 36 6.64
(5.56-7.01)		 10.79
(9.68-11.12)		 4.6
(4.21-5.8)		 8.75
(7.82-9.71)		 8.49
(6.49-9.22)		 7.09
(4.58-8.11)		 3.98*
(3.37-4.74)
Day 45 6.51
(6.24-6.85)		 10.69
(10.22-10.97)		 4.24
(3.82-5.32)		 9.03
(8.08-9.45)		 8.08
(6.94-9.13)		 6.41
(5.15-7.19)		 3.75*
(2.95- 5.04)
Open in a new tab

Abbreviations: DI = Dysbiosis Index; E. coli = Escherichia coli; P. hiranonis = Peptacetobacter hiranonis; qPCR = quantitative polymerase chain reaction.

Measured fecal abundances using qPCR of 7 bacterial taxa included in the DI at baseline (day 1), during (day 6), and post-azithromycin administration (day 11, 26, 36, and 45) in eleven healthy dogs. Data is expressed as median (minimum—maximum) log DNA/gram of feces. An asterisk (*) indicates significantly different (P < .05) fecal abundances compared to baseline (day 1).

Figure 3.

P. hiranonis fecal abundance (log DNA/gram feces) at baseline (day 1), during (day 6), and postazithromycin administration (day 11, 26, 36, and 45) in eleven healthy dogs. Azithromycin (2 mg/kg PO q8h) was administered on days 3--10. Thick horizontal bars represent median values at each time point (day 1, 6, 11, 26, 36, and 45) and thin horizontal bars with an asterisk (*) represent a significant difference (p less than 0.05) compared to baseline (day 1). The gray-shaded region denotes the reference range. Abbreviations: P. hiranonis, Peptacetobacter hiranonis

Open in a new tab

Peptacetobacter hiranonis fecal abundance (log DNA/gram feces) at baseline (day 1), during (day 6), and post-azithromycin administration (day 11, 26, 36, and 45) in eleven healthy dogs. Azithromycin (2 mg/kg PO q8h) was administered on days 3-10. Thick horizontal bars represent median values at each time point (day 1, 6, 11, 26, 36, and 45), and thin horizontal bars with an asterisk (*) represent a significant difference (P < .05) compared to baseline (day 1). The gray-shaded region denotes the reference range. Abbreviations: P. hiranonis = Peptacetobacter hiranonis.

Figure 4.

Fecal abundances (log DNA/gram feces) of core bacterial taxa included in the DI: (A) Faecalibacterium, (B) Fusobacterium, (C) Turicibacter, (D) Blautia, (E) E. coli, and (F) Streptococcus, at baseline (day 1), during (day 6), and after azithromycin administration (days 11, 26, 36, and 45) in eleven healthy dogs. Azithromycin (2 mg/kg PO q8h) was administered on days 3--10. Thick horizontal bars represent median values for each time point, and thin horizontal bars marked with an asterisk (*) indicate a significant difference compared to baseline (p less than 0.05). Gray-shaded regions denote the lab's reference range for each taxon if available. Abbreviations: DI, dysbiosis index; E. coli, Escherichia coli

Open in a new tab

Fecal abundances (log DNA/gram feces) of core bacterial taxa included in the DI: (A) Faecalibacterium, (B) Fusobacterium, (C) Turicibacter, (D) Blautia, (E) E. coli, and (F) streptococcus, at baseline (day 1), during (day 6), and after azithromycin administration (days 11, 26, 36, and 45) in eleven healthy dogs. Azithromycin (2 mg/kg PO q8h) was administered on days 3-10. Thick horizontal bars represent median values for each time point, and thin horizontal bars marked with an asterisk (*) indicate a significant difference compared to baseline (P < .05). Gray-shaded regions denote the lab’s reference range for each taxon if available. Abbreviations: DI = Dysbiosis Index; E. coli = Escherichia coli.

Additional core bacterial taxa

The abundances of 9 bacterial taxa not included in the DI on days 1, 6, 11, 26, 36, and 45 are presented in Table 2. Akkermansia abundance significantly decreased on day 6 (P = .001), remained significantly decreased on days 11 and 26 (P = .03), and increased on day 36. Bacteroides abundance significantly decreased on day 6 (P = .001) and remained significantly decreased on day 11 (P = .001) but recovered on day 26. Bifidobacterium abundance significantly decreased on day 6 (P < .001) and achieved statistical recovery on day 11; however, remained decreased relative to baseline for the rest of the study. Enterococcus abundance significantly increased on day 6 (P = .01) and remained significantly increased on days 11 (P < .001) and 26 (P = .046). Recovery occurred on day 36. Prevotella copri abundance significantly decreased on day 6 (P = .003) and remained significantly decreased on day 11 (P = .002) and 26 (P = .001). Recovery occurred on day 36. Ruminococcaceae, Ruminococcus gnavus, Clostridium perfringens, and Lactobacillus abundances did not experience any significant changes throughout the study.

Table 2.

Fecal abundances of non-DI bacterial taxa.

Akkermansia Bacteroides Bifidobacterium C. perfringens Enterococcus Lactobacillus P. copri Ruminococcaceae R. gnavus
Day 1 2.59
(1.89-3.50)		 6.20
(5.38 -7.12)		 5.61
(3.05-7.47)		 3.16
(1.13-5.48)		 3.82
(2.88-5.41)		 7.29
(5.20-7.95)		 14.48
(13.14-16.12)		 5.92
(5.33-6.19)		 11.28
(10.82-11.81)
Day 6 1.92*
(1.18-2.57)		 4.38*
(3.38-6.44)		 2.96*
(1.77-4.67)		 3.17
(1.13-3.73)		 6.25*
(4.90-6.89)		 5.82
(4.66-8.35)		 9.52*
(9.52-10.71)		 4.76
(4.38-5.11)		 10.73
(9.49-11.74)
Day 11 2.12*
(1.55-3.20)		 4.58*
(3.28-5.85)		 3.93
(3.54-4.20)		 3.85
(2.27-4.20)		 6.18*
(5.73-6.74)		 5.58
(4.45-7.84)		 9.52*
(9.52-11.53)		 5.04
(4.89-5.82)		 11.79
(10.99-12.03)
Day 26 2.10*
(1.91-2.47)		 5.98
(4.84-7.27)		 3.76
(3.24-3.93)		 2.47
(1.13-3.25)		 5.77*
(3.51-6.54)		 5.93
(3.55-8.20)		 9.52*
(9.52-12.20)		 6.07
(5.89-6.23)		 11.50
(11.28-11.76)
Day 36 2.24
(1.73-2.60)		 6.22
(4.47-7.03)		 4.19
(3.86-4.45)		 1.98
(1.13-3.31)		 3.69
(2.55-5.28)		 6.47
(4.02-8.09)		 13.50
(9.52-15.35)		 6.19
(4.81-6.51)		 11.33
(10.14-11.73)
Day 45 2.41
(1.75-2.75)		 6.54
(4.49-7.09)		 3.76
(3.15-4.08)		 2.68
(1.13-3.38)		 3.12
(2.11-5.56)		 4.74
(3.35-8.22)		 13.99
(9.52-16.28)		 6.15
(5.98-6.44)		 11.48
(11.12-11.76)
Open in a new tab

Abbreviations: C. perfringens = Clostridium perfringens; DI = Dysbiosis Index; P. copri = Prevotella copri; qPCR = quantitative polymerase chain reaction; R. gnavus = Ruminococcus gnavus.

Measured fecal abundances using qPCR of 9 bacterial taxa not included in the DI at baseline (day 1), during (day 6), and post-azithromycin administration (day 11, 26, 36, and 45) in eleven healthy dogs. Data is expressed as median (minimum—maximum) log DNA/gram of feces. An asterisk (*) indicates significantly different (P < .05) fecal abundances compared to baseline (day 1).

Fecal BA concentrations

Low-dose azithromycin significantly affected the proportions of fecal BAs. The percentage of primary and secondary unconjugated bile acids (UBAs) was significantly altered on day 11 compared to baseline, as demonstrated in Figure 5. Specifically, the percentage of primary UBAs increased and secondary UBAs decreased (P = .017). Both returned to their respective baseline by day 45. This shift in primary UBAs showed a strong negative correlation with P. hiranonis (r = -0.936 [95% CI, -0.973, -0.852], P < .001), as demonstrated in Figure 6.

Figure 5.

Measured percentage of primary (A) and secondary (B) unconjugated bile acids at baseline (day 1) and after azithromycin administration (day 11 and 45) in eleven healthy dogs. Azithromycin (2 mg/kg PO q8h) was administered on days 3--10. Thick horizontal bars represent median values for each time point, and thin horizontal bars marked with an asterisk (*) indicate a significant difference compared to baseline (p less than 0.05). Abbreviations: BA, bile acid; UBA, unconjugated bile acid

Open in a new tab

Measured percentage of primary (A) and secondary (B) unconjugated bile acids at baseline (day 1) and after azithromycin administration (day 11 and 45) in eleven healthy dogs. Azithromycin (2 mg/kg PO q8h) was administered on days 3-10. Thick horizontal bars represent median values for each time point, and thin horizontal bars marked with an asterisk (*) indicate a significant difference compared to baseline (P < .05). Abbreviations: BA = bile acid; UBA = unconjugated bile acid.

Figure 6.

Strong negative correlation between percentage of primary unconjugated bile acids (% of total UBAs) and P. hiranonis (log DNA/g feces) across all days (1, 11, and 45) in eleven healthy dogs. Azithromycin (2 mg/kg PO q8h) was administered on days 3--10. The gray-shaded region denotes the reference range of P. hiranonis. Decrease in P. hiranonis below the reference range is associated with abnormal bile acid conversion and an increase in primary UBAs greater than 50%. Abbreviations: BA, bile acid; UBA, unconjugated bile acid; P. hiranonis, Peptacetobacter hiranonis.

Open in a new tab

Strong negative correlation between percentage of primary unconjugated bile acids (% of total UBAs) and P. hiranonis (log DNA/g feces) across all days 1, 11, and 45, and in eleven healthy dogs. Azithromycin (2 mg/kg PO q8h) was administered on days 3-10. The gray-shaded region denotes the reference range of P. hiranonis. Decrease in P. hiranonis below the reference range is associated with abnormal bile acid conversion and an increase in primary UBAs >50%. Abbreviations: BA = bile acid; P. hiranonis = Peptacetobacter hiranonis; UBA, unconjugated bile acid.

The primary BA, cholic acid, significantly increased on day 11 (P = .03), whereas chenodeoxycholic acid remained non-significantly altered at this time point. Both secondary UBAs, lithocholic acid (P = .003) and deoxycholic acid (P = .001), were significantly decreased on day 11. All returned to baseline at the conclusion of the study.

Two dogs on day 11 maintained low primary UBA and high secondary UBA concentrations. P. hiranonis was minimally affected compared to baseline in these dogs.

Discussion

The study aim was to describe the effects of low-dose azithromycin on the fecal microbiome and fecal BA concentrations in healthy dogs, with the hypothesis that its administration would alter the fecal microbiome and fecal BA metabolism. Our findings support this hypothesis. Although administered at a lower dose than used for antimicrobial purposes, azithromycin altered the fecal microbiome and BA metabolism. Dysbiosis was evident and supported by significant decreases in several core bacterial taxa, including P. hiranonis, Faecalibacterium, and Fusobacterium, which fell below reference intervals in most dogs, and reduced conversion of primary to secondary UBAs. These effects were mostly reversible upon discontinuation, with normalization of fecal BAs and recovery of all but 1 taxon by the end of the study.

Evidence of dysbiosis in this study was supported by significant decreases in several core and functionally important bacterial taxa, including P. hiranonis, Faecalibacterium, and Fusobacterium. Faecalibacterium produces SCFAs, which provide energy for enterocytes and modulate mucosal immunity.36 Additionally, SCFAs maintain an acidic luminal environment that suppresses pH-sensitive pathobionts and regulate microbial virulence factors.36,37 Decreased Faecalibacterium occurs in dogs with gastrointestinal disease and after antibiotic exposure.19,21,22,26,37 Fusobacterium also contributes to SCFA production, most notably butyrate, and primarily ferments amino acids (eg, glutamate, lysine) rather than carbohydrates.38 Additionally, Fusobacterium abundance appears to be associated with dietary protein, as it is present in relatively higher amounts in carnivorous species and dogs fed a raw diet.39,40 This taxon is typically reduced in dysbiosis and similarly decreases with antibiotic use or in chronic enteropathies.17,19,21–23,26,41  P. hiranonis produces 7-α-dehydroxylation enzymes encoded by the bai operon and is the primary converter of primary to secondary UBAs in dogs.34,36 Production of secondary UBAs is essential for intestinal health, as their depletion can allow for overgrowth of enteropathogens and loss of beneficial commensals. High concentrations of primary BAs that can occur due to reduced conversion can also potentially lead to secretory diarrhea in a small subset of animals.36,42 Decreases in P. hiranonis occur after broad-spectrum antibiotic use and in a subset of dogs with chronic inflammatory enteropathies, while a decrease is rarely observed in dogs with acute diarrhea.42–45

Azithromycin use was associated with a significant increase in primary UBAs, alongside decreases in both secondary UBAs and P. hiranonis abundance, consistent with reduced conversion. Primary UBAs were negatively correlated with P. hiranonis, in agreement with previous data.21 Though similar shifts in BAs are reported with metronidazole and tylosin at antimicrobial doses, recovery patterns differ compared to azithromycin. In this study, BAs normalized within 5 weeks, whereas tylosin and metronidazole show sustained elevations in primary UBAs up to 8 weeks in individual dogs.21,22 Furthermore, it is speculated that recovery occurred earlier in this study given normalization of P. hiranonis at day 26; however, BAs were only measured on day 11 and day 45. Thus, recovery of fecal BAs possibly occurs more rapidly after low-dose azithromycin; although, the tylosin and metronidazole studies were conducted in pet dogs, and housing environment (eg, pet vs research colony) has been shown to possibly influence the fecal microbiome of dogs.46

Together, the significant decreases in several core taxa, including P. hiranonis, Faecalibacterium, and Fusobacterium, and the significant decrease in secondary UBAs demonstrate that low-dose azithromycin caused dysbiosis. These changes highlight not only shifts in composition but also functional consequences that could have negative effects on host physiology. Additionally, though not measured in this study, it is probable that SCFAs were reduced in these dogs based upon decreased abundance of key SCFA producers, an outcome observed with metronidazole use in cats.47 SCFAs decrease in dogs with chronic enteropathy (CE), where dysbiosis is also commonly present.37 Furthermore, in these dogs, Faecalibacterium positively correlates with propionate production. It is also conceivable that the decrease in secondary UBAs, in part, influenced the transient increase in fecal scores during azithromycin administration, as increased primary UBAs can potentially contribute to secretory diarrhea due to increased colonic fluid secretion and altered mucosal permeability.36,48 However, cautious interpretation is warranted, as fecal scores remained within the normal range (PFS: 2-3) and their assessment is based on a subjective grading scheme. Thus, the mild increase in fecal scores might not have been clinically relevant or reflected true change. Nevertheless, it is clear that low-dose azithromycin caused dysbiosis on the basis of both compositional and functional changes; however, the net result of these changes remains to be determined given the complexity of the fecal microbiome and possible influence on distant organ systems. Fortunately, the documented changes appear to be reversible given secondary UBAs and all but 1 taxon recovered by the end of the study.

E. coli is a facultative anaerobe and normal commensal of the intestinal tract in dogs. Increased abundance occurs in dogs with CE and after administration of metronidazole, tylosin, or amoxicillin.17,19,20,23,24,26,41 The absence of E. coli overgrowth in this study compared to other antibiotics could reflect differences in antimicrobial activity. E. coli commonly resists aminopenicillins, metronidazole, and inhibits tylosin uptake.18,49 In contrast, azithromycin’s unique macrolide structure confers higher basicity and allows greater intracellular uptake, enhancing activity against Enterobacteriaceae.50 In people, azithromycin demonstrates excellent activity against common enteropathogens, including multiple strains of E. coli.51–53 Thus, the lack of overgrowth in this study, and numerical decrease, could reflect a direct antimicrobial effect of azithromycin.

Streptococcus are gram-positive commensals that often increase during dysbiosis due to their growth advantage as lactic-acid-producing bacteria when luminal pH decreases with excess undigested substrate.36,54,55 Despite evidence of dysbiosis, low-dose azithromycin led to a sustained and significant decrease in Streptococcus, the only taxon to not recover by the end of the study. This contrasts with findings for other antibiotics. Both tylosin and metronidazole increase Streptococcus abundance, whereas aminopenicillins have yielded mixed results.17–21,23,24

Direct suppression of growth is therefore a plausible explanation for the lack of overgrowth. More notable, however, was the sustained decrease throughout the 5-week recovery period. The sustained decrease is unlikely to be explained by azithromycin’s half-life or post-antibiotic effects. Reported half-lives (51-90 hours) in dogs indicate drug elimination within ~19 days, and post-antibiotic effects limited to hours.56,57 The most likely explanation is ecological disruption of the fecal microbiome.

The DI is the only current analytically validated assay for measuring fecal dysbiosis in dogs and cats.33 It incorporates 7 important core bacterial taxa to generate a numerical value to reflect the degree of dysbiosis, and is interpreted together with changes in all included taxa. An increase in the DI correlates with total shifts observed by 16S rRNA and DNA shotgun sequencing, with a higher DI correlating with larger shifts on sequencing in these studies.21,45,58,59 The DI is used to document dysbiosis after antibiotic exposure and shows significant increases after administration of metronidazole, tylosin, and a combination of enrofloxacin and metronidazole.19,21–23,26 In this study, while the DI increased numerically at day 6 and was significantly higher compared to the following days, the total increase was smaller than would be expected based on previous antibiotics studies. This is likely due to the fact that E. coli and Streptococcus, which typically show a strong increase in dysbiosis after antibiotic use, instead decreased with low-dose azithromycin. This emphasizes that the DI should be interpreted together with the abundances of all core bacteria, as the decrease in functionally important core taxa (ie, P. hiranonis, Faecalibacterium, and Fusobacterium) indicated clear change after low-dose azithromycin administration.

Aside from a possible transient, mild decrease in fecal consistency, no other clinically relevant abnormalities were observed in this study. These findings suggest that short-term use of low-dose azithromycin is well tolerated in healthy dogs and unlikely to cause clinically relevant adverse effects. However, healthy dogs do not represent the clinical population in which low-dose azithromycin would be used. Clinical dogs could have pre-existing dysbiosis, altered BA metabolism, and other gastrointestinal disturbances that could increase their susceptibility to microbiome and metabolomic alterations and associated clinical consequences. Alternatively, effects could be similar to those seen in this study. Therefore, results should be interpreted cautiously when considering direct applicability and highlight the need for future research into clinical populations. Furthermore, although antimicrobial resistance was not evaluated in this study, the potential for resistance selection represents an additional consideration when deciding on its use as a prokinetic.

This study had several limitations. First, no control group was included; however, this was considered appropriate given evidence supporting stability of the fecal microbiome in dogs and long-term stability noted in people.21,22,41,60 Second, full sequencing of the fecal microbiome was not performed and metabolomic assessment was restricted to BAs. However, the DI was selected for its established correlation with metagenomic sequencing to mitigate this limitation, and BAs were considered a relevant metabolite given their essential functions and comparative value to established literature. Furthermore, though full microbiome sequencing provides additional data, interpretation and comparison of this information is limited given interlaboratory variability and compositionality of the data, as taxonomy is expressed as relative proportions (ie, data is expressed as a percentage of total) in contrast to measuring total abundance by qPCR.61 Third, as discussed, this study used healthy dogs, which does not represent clinical populations that would be receiving low-dose azithromycin. Therefore, results should be interpreted cautiously when considering direct applicability to such a population. Fourth, though the sample size was sufficient per the power analysis, the number of dogs remained limited. Finally, only 1 dosing regimen was used, meaning results might not apply to alternative protocols. This is of particular importance as prokinetic efficacy of azithromycin in dogs remains to be proven, which could ultimately yield different dosing strategies than used in this study.

This study demonstrated that low-dose azithromycin altered the fecal microbiome and fecal BAs in healthy dogs. The absence of Streptococcus and E. coli overgrowth also indicates a potentially unique response compared to other antibiotics. Therefore, when used short-term at a lower dose for prokinetic purposes, azithromycin appears to induce compositional and functional changes that are mostly reversible.

Acknowledgments

The authors thank the University of Tennessee Office of Laboratory Animal Care staff for their dedicated care of the dogs during this study. The authors also acknowledge the Purdue University College of Veterinary Medicine as the site of the pilot study that generated preliminary data for the power analysis. The results of this work were presented in part as an abstract at the 2025 American College of Veterinary Internal Medicine (ACIM) Forum in Louisville, Kentucky.

Abbreviations

BA(s)

bile acid(s)

BCS

body condition score

CE

chronic enteropathy

DI

Dysbiosis Index

PFS

purina Fecal Score

qPCR

quantitative polymerase chain reaction

SCFA(s)

short-chain fatty acid(s)

UBA(s)

unconjugated bile acid(s)

Appendix

Animals

This study was approved by the Institutional Animal Care and Use Committee at the University of Tennessee. Twelve healthy research colony Beagles were included in this study. Dogs were included if they had no history of gastrointestinal signs or antimicrobial exposure within the preceding 6 months. They were deemed healthy based on the absence of clinical disease, physical examination, complete blood count, biochemistry, urinalysis, and negative fecal screening (zinc and sugar flotation). Dogs were housed individually, had a stable daily routine, and were cared for by trained, experienced laboratory staff. The only variability to their routine was multiple supervised group interactions per week at the discretion of the laboratory staff. Long-term diet in these dogs was Purina ONE® Lamb & Rice Formula Dry Dog Food fed once per day in the morning and daily treats of Purina ONE® Plus Vibrant Maturity Adult 7+ Classic Ground Turkey & Barley Entrée Wet Dog Food. The only modification to their dietary schedule as part of this study was restricting daily treat intake to once per day. This was implemented 4 weeks before the study’s onset. Lastly, dogs received prophylactic deworming (fenbendazole 50 mg/kg PO q24h for 3 days) and a bath within 1 week of starting the study.

Study design

A prospective study with twelve healthy research colony Beagles was performed over a 45-day period. Dogs were administered azithromycin (2 mg/kg PO q8h) over 8 days, corresponding to days 3-10. Dogs were then monitored for 35 days for adverse effects, changes to fecal consistency, and select fecal samples collected (day 1, 6, 11, 26, 36, and 45). Body weight and BCS were recorded at baseline (day 1) and at the end of the study (day 45) by a single investigator using the same scale and performed in the evening. No control group was included given the proven stability of the canine gut microbiome.21,22,41,60 Instead, subjects acted as their own control with comparison to their baseline sample (day 1).

Dogs remained individually housed and fed the diet combination as described, which was unchanged. Dogs were monitored daily by laboratory staff for evidence of gastroenteritis (vomiting, diarrhea, anorexia, or abdominal discomfort), drug-related nausea (changes in appetite, hypersalivation), or other signs of illness. Daily fecal scoring was performed by a combination of trained, experienced laboratory staff and 2 investigators using the PFS.62

Fresh fecal samples after spontaneous defecation were acquired within 12 hours, a time at which the gut microbiome has proven to remain stable in dogs.63 Days of collection included baseline (day 1), during (day 6), and post-azithromycin (day 11, 26, 36, and 45). Each fecal sample was weighed, divided into 3 aliquots (minimum 5 grams), and immediately stored at -80°C. All samples were submitted as a single batch at the end of the study to the Gastrointestinal Laboratory at Texas A&M University, where all analyses were performed.

Quantitative PCR analysis and calculation of the DI

Quantitative PCR was used to measure the abundance of multiple bacterial taxa (Blautia, P. hiranonis, E. coli, Faecalibacterium, Fusobacterium, Streptococcus, Turicibacter, Bifidobacterium, Bacteroides, Akkermansia, C. perfringens, Enterococcus, Lactobacillus, P. copri, Ruminococcaceae, and R. gnavus) on fecal samples from days 1, 6, 11, 26, 36, and 45 as previously described.32 Results were expressed as the log amount of DNA for each bacterial taxon per gram of feces. The DI was then derived using select taxa following established methods.33 DI values <0: normal DI indicating no shifts in the overall diversity of the intestinal microbiota; 0-2: mild to moderate shift in the overall diversity of the intestinal microbiota; and >2: significant dysbiosis.

Fecal BA concentrations

A total of 29 conjugated, unconjugated, and oxo-isoforms of primary and secondary BAs were measured using a validated liquid chromatography–tandem mass spectrometry method.34 Cholic acid and chenodeoxycholic acid were defined as primary UBAs, and deoxycholic acid and lithocholic acid as secondary UBAs formed via P. hiranonis-mediated 7-α-dehydroxylation.34,64,65 Quantification was performed on select days (day 1, 11, and 45). Results were reported as nanograms per milligram of dry fecal matter and percentage of total UBAs.

Statistical analysis

The sample size was determined based on data from a previously performed pilot study to ensure adequate power and detection of significance. This pilot study was composed of 5 healthy research dogs who were administered an identical dosing schedule of azithromycin (2 mg/kg PO q8h for 8 days); however, the study duration was shorter. Sample size was estimated using a single factor repeated-measures design and a sample of 12 subjects, measured at 5 time points, achieving 84% power to detect differences in the DI among the means over 5 time points (-4.41, -3.81, -1.21, -4.13, and -3.94) at a P < .05 significance level. The standard deviation across subjects at each time point was assumed to be 2.3.

All data sets were tested for normality using the Shapiro–Wilk W normality test and QQ plots. Descriptive statistics were calculated for each variable. Data are reported as mean and standard deviation for normally distributed data; otherwise, median and range (minimum-maximum) were reported.

For assessment of fecal score (PFS), the DI, qPCR of individual bacterial taxa, and fecal BAs (grouped and ungrouped), Friedman tests were performed for data over time (IBM Statistical Package for the Social Sciences (SPSS) version 28) and adjusted for multiple tests using Benjamini-Hochberg’s False Discovery Rate (SAS version 9.4, proc multtest, SAS Institute Inc., Cary, NC, United States). Post hoc pairwise comparisons were performed using Bonferroni correction method. Related-Samples Wilcoxon Signed-Rank Test was used to test the difference between pre- and post-values in weight and BCS. IBM SPSS version 28 was used for the statistical analysis (IBM Corp. Released 2021. IBM SPSS Statistics for Windows, Version 28.0. Armonk, NY: IBM Corp) and GraphPad Prism to generate relevant figures (GraphPad Prism 7.0, 180 GraphPad Software Inc). The correlation analysis was conducted using repeated-measures correlation by R version 4.5.1 rmcorr package version 0.7.0.35 Significance was set at P < .05.

Contributor Information

Colin Smith, Department of Small Animal Clinical Sciences, University of Tennessee College of Veterinary Medicine, Knoxville, Tennessee 37996, United States.

Roman Husnik, Department of Small Animal Clinical Sciences, University of Tennessee College of Veterinary Medicine, Knoxville, Tennessee 37996, United States.

Frederic Gaschen, Department of Veterinary Clinical Sciences, School of Veterinary Medicine, Louisiana State University, Baton Rouge, Louisiana 70803, United States.

Claudia Kirk, Department of Small Animal Clinical Sciences, University of Tennessee College of Veterinary Medicine, Knoxville, Tennessee 37996, United States.

Xiaojuan Zhu, Research Computing Support, Office of Innovative Technologies, University of Tennessee College of Veterinary Medicine, Knoxville, Tennessee 37996, United States.

Jan Suchodolski, Gastrointestinal Laboratory, Department of Small Animal Clinical Sciences, Texas A&M University, College Station, Texas 77843-4474, United States.

Author contributions

Colin Smith (Data curation, Formal analysis, Investigation, Project administration, Visualization, Writing—original draft, Writing—review & editing), Roman Husnik (Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Project administration, Resources, Supervision, Visualization, Writing—original draft, Writing—review & editing), Frederic P Gaschen (Writing—review & editing), Claudia Kirk (Resources, Writing—review & editing), Xiaojuan Zhu (Formal analysis), and Jan S Suchodolski (Conceptualization, Formal analysis, Methodology, Resources, Writing—review & editing)

Conflicts of interest

Authors declare no conflicts of interest.

Funding

Supported with funding from the Department of Small Animal Clinical Sciences of the University of Tennessee.

Off-label antimicrobial declaration

Azithromycin was administered off-label for prokinetic effects at a lower dose than is recommended for treatment of infection.

Institutional animal care and use committee or other approval declaration

All animal use was approved by the University of Tennessee Institutional Animal Care and Use Committee and the study was performed in an accredited, by the Association for the Assessment and Accreditation of Laboratory Animal Care, facility.

Human ethics approval declaration

Authors declare human ethics approval was not needed.

References

  • 1. Whitehead  K, Cortes  Y, Eirmann  L. Gastrointestinal dysmotility disorders in critically ill dogs and cats. J Vet Emerg Crit Care (San Antonio). 2016;26:234-253. 10.1111/vec.12449 [DOI] [PubMed] [Google Scholar]
  • 2. Husnik  R, Gaschen  F. Gastric motility disorders in dogs and cats. Vet Clin North Am Small Anim Pract. 2021;51:43-59. 10.1016/j.cvsm.2020.09.002 [DOI] [PubMed] [Google Scholar]
  • 3. Liu  N, Abell  T. Gastroparesis updates on pathogenesis and management. Gut Liver. 2017;11:579-589. 10.5009/gnl16336 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 4. Carlson  RG, Hocking  MP, Courington  KR, Sninsky  CA, Vogel  SB. Erythromycin enhances delayed gastric emptying in dogs after roux-Y antrectomy. Am J Surg. 1991;161:31-35. 10.1016/0002-9610(91)90357-j [DOI] [PubMed] [Google Scholar]
  • 5. Itoh  Z, Nakaya  M, Suzuki  T, Arai  H, Wakabayashi  K. Erythromycin mimics exogenous motilin in gastrointestinal contractile activity in the dog. Am J Physiol. 1984;247:G688-G694. 10.1152/ajpgi.1984.247.6.g688 [DOI] [PubMed] [Google Scholar]
  • 6. Diamond  SJ, Omer  E, Kiraly  L. In search of the ideal promotility agent: optimal use of currently available promotility agents for nutrition therapy of the critically ill patient. Curr Gastroenterol Rep. 2017;19:63. 10.1007/s11894-017-0604-7 [DOI] [PubMed] [Google Scholar]
  • 7. Mougey  EB, Saunders  M, Franciosi  JP, Gomez-Suarez  RA. Comparative effectiveness of intravenous azithromycin versus erythromycin stimulating antroduodenal motility in children. J Pediatr Gastroenterol Nutr. 2022;74:25-32. 10.1097/mpg.0000000000003271 [DOI] [PubMed] [Google Scholar]
  • 8. Husnik  R, Gaschen  FP, Fletcher  JM, Gaschen  L. Ultrasonographic assessment of the effect of metoclopramide, erythromycin, and exenatide on solid-phase gastric emptying in healthy cats. J Vet Intern Med. 2020;34:1440-1446. 10.1111/jvim.15787 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 9. Ray  WA, Murray  KT, Meredith  S, Narasimhulu  SS, Hall  K, Stein  CM. Oral erythromycin and the risk of sudden death from cardiac causes. N Engl J Med. 2004;351:1089-1096. 10.1056/nejmoa040582 [DOI] [PubMed] [Google Scholar]
  • 10. Haefeli  WE, Schoenenberger  RA, Weiss  P, Ritz  R. Possible risk for cardiac arrhythmia related to intravenous erythromycin. Intensive Care Med. 1992;18:469-473. 10.1007/bf01708583 [DOI] [PubMed] [Google Scholar]
  • 11. Larson  JM, Tavakkoli  A, Drane  WE, Toskes  PP, Moshiree  B. Advantages of azithromycin over erythromycin in improving the gastric emptying half-time in adult patients with gastroparesis. J Neurogastroenterol Motil. 2010;16:407-413. 10.5056/jnm.2010.16.4.407 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12. Moshiree  B, McDonald  R, Hou  W, Toskes  PP. Comparison of the effect of azithromycin versus erythromycin on antroduodenal pressure profiles of patients with chronic functional gastrointestinal pain and gastroparesis. Dig Dis Sci. 2010;55:675-683. 10.1007/s10620-009-1038-3 [DOI] [PubMed] [Google Scholar]
  • 13. Shakir  AK, Altaf  MA. Azithromycin induces migrating motor complexes in pediatric patients undergoing antroduodenal motility studies. J Pediatr Pharmacol Ther. 2018;23:390-394. 10.5863/1551-6776-23.5.390 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 14. Rutherford  S, Gaschen  F, Husnik  R, Fletcher  J, Gaschen  L. Ultrasonographic evaluation of the effects of azithromycin on antral motility and gastric emptying in healthy cats. J Vet Intern Med. 2022;36:508-514. 10.1111/jvim.16385 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15. Lappin  MR, Blondeau  J, Boothe  D, et al.  Antimicrobial use guidelines for treatment of respiratory tract disease in dogs and cats: antimicrobial guidelines working group of the international society for companion animal infectious diseases. J Vet Intern Med. 2017;31:279-294. 10.1111/jvim.14627 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16. Marshall-Jones  ZV, Patel  KV, Castillo-Fernandez  J, et al.  Conserved signatures of the canine faecal microbiome are associated with metronidazole treatment and recovery. Sci Rep. 2024;14:5277. 10.1038/s41598-024-51338-7 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17. Belchik  SE, Oba  PM, Wyss  R, et al.  Effects of a milk oligosaccharide biosimilar on fecal characteristics, microbiota, and bile acid, calprotectin, and immunoglobulin concentrations of healthy adult dogs treated with metronidazole. J Anim Sci. 2023;101:skad011. 10.1093/jas/skad011 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18. Werner  M, Suchodolski  JS, Straubinger  RK, et al.  Effect of amoxicillin-clavulanic acid on clinical scores, intestinal microbiome, and amoxicillin-resistant Escherichia coli in dogs with uncomplicated acute diarrhea. J Vet Intern Med. 2020;34:1166-1176. 10.1111/jvim.15775 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 19. Marclay  M, Dwyer  E, Suchodolski  JS, Lidbury  JA, Steiner  JM, Gaschen  FP. Recovery of fecal microbiome and bile acids in healthy dogs after tylosin administration with and without fecal microbiota transplantation. Vet Sci. 2022;9:324. 10.3390/vetsci9070324 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20. Espinosa-Gongora  C, Jessen  LR, Kieler  IN, et al.  Impact of oral amoxicillin and amoxicillin/clavulanic acid treatment on bacterial diversity and β-lactam resistance in the canine faecal microbiota. J Antimicrob Chemother. 2020;75:351-361. 10.1093/jac/dkz458 [DOI] [PubMed] [Google Scholar]
  • 21. Pilla  R, Gaschen  FP, Barr  JW, et al.  Effects of metronidazole on the fecal microbiome and metabolome in healthy dogs. J Vet Intern Med. 2020;34:1853-1866. 10.1111/jvim.15871 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 22. Manchester  AC, Webb  CB, Blake  AB, et al.  Long-term impact of tylosin on fecal microbiota and fecal bile acids of healthy dogs. J Vet Intern Med. 2019;33:2605-2617. 10.1111/jvim.15635 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 23. Igarashi  H, Maeda  S, Ohno  K, Horigome  A, Odamaki  T, Tsujimoto  H. Effect of oral administration of metronidazole or prednisolone on fecal microbiota in dogs. PloS One. 2014;9:e107909. 10.1371/journal.pone.0107909 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24. Grønvold  AMR, L’Abée-Lund  TM, Sørum  H, et al.  Changes in fecal microbiota of healthy dogs administered amoxicillin. FEMS Microbiol Ecol. 2010;71:313-326. 10.1111/j.1574-6941.2009.00808.x [DOI] [PubMed] [Google Scholar]
  • 25. Suchodolski  JS, Dowd  SE, Westermarck  E, et al.  The effect of the macrolide antibiotic tylosin on microbial diversity in the canine small intestine as demonstrated by massive parallel 16S rRNA gene sequencing. BMC Microbiol. 2009;9:210. 10.1186/1471-2180-9-210 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26. Whittemore  JC, Moyers  TD, Price  JM. Randomized, controlled, crossover trial of prevention of antibiotic-induced gastrointestinal signs using a synbiotic mixture in healthy research dogs. J Vet Intern Med. 2019;33:1619-1626. 10.1111/jvim.15553 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 27. Tillman  EM, Smetana  KS, Bantu  L, Buckley  MG. Pharmacologic treatment for pediatric gastroparesis: a review of the literature. J Pediatr Pharmacol Ther. 2016;21:120-132. 10.5863/1551-6776-21.2.120 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28. Wedzicha  JA, Calverley  PMA, Albert  RK, et al.  Prevention of COPD exacerbations: a European respiratory society/American thoracic society guideline. Eur Respir J. 2017;50:1602265. 10.1183/13993003.02265-2016 [DOI] [PubMed] [Google Scholar]
  • 29. Cruz  RLD, Que  FAD, Bacud  LNC, Rosario  MCO. Efficacy of azithromycin in preventing pulmonary exacerbations among patients with bronchiectasis: a systematic review and meta-analysis. Cureus.  2026;18:e101574. 10.7759/cureus.101574 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 30. Sandman  Z, Iqbal  OA. Azithromycin. M. Lindsay Grayson, Sara Cosgrove, Suzanne Crowe, William Hope (eds.), Kucers the use of antibiotics: a clinical review of antibacterial, antifungal, antiparasitic, and antiviral drugs, seventh edition. CRC Press, 2024:1122-1149. 10.1201/9781315152110 [DOI] [Google Scholar]
  • 31. Choo  JM, Martin  AM, Taylor  SL, et al.  The impact of long-term macrolide exposure on the gut microbiome and its implications for metabolic control. Microbiol Spectr. 2023;11:e00831-e00823. 10.1128/spectrum.00831-23 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 32. Sung  CH, Pilla  R, Chen  CC, et al.  Correlation between targeted qPCR assays and untargeted DNA shotgun metagenomic sequencing for assessing the fecal microbiota in dogs. Animals (Basel). 2023;13:2-16. 10.3390/ani13162597 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 33. Al Shawaqfeh  MK, Wajid  B, Minamoto  Y, et al.  A dysbiosis index to assess microbial changes in fecal samples of dogs with chronic inflammatory enteropathy. FEMS Microbiol Ecol. 2017;93:1-8. 10.1093/femsec/fix136 [DOI] [PubMed] [Google Scholar]
  • 34. Blake  AB, Toresson  LC, Chen  CC, et al.  Altered iso- and oxo-fecal bile acid concentrations in dogs with chronic enteropathy. Pets.  2025;2:18. 10.3390/pets2020018 [DOI] [Google Scholar]
  • 35. Bakdash  JZ, Marusich  LR. Repeated measures correlation. Front Psychol. 2017;8:1-13. 10.3389/fpsyg.2017.00456 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 36. Suchodolski  JS. Analysis of the gut microbiome in dogs and cats. Vet Clin Pathol. 2022;50:6-17. 10.1111/vcp.13031 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 37. Minamoto  Y, Minamoto  T, Isaiah  A, et al.  Fecal short-chain fatty acid concentrations and dysbiosis in dogs with chronic enteropathy. J Vet Intern Med. 2019;33:1608-1618. 10.1111/jvim.15520 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 38. Anand  S, Kaur  H, Mande  SS. Comparative in silico analysis of butyrate production pathways in gut commensals and pathogens. Front Microbiol. 2016;7:1945. 10.3389/fmicb.2016.01945/full [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 39. Hanifeh  M, Huhtinen  M, Sclivagnotis  YS, Lyhs  U, Grönthal  T, Spillmann  T. Adhesion of Bacteroides vulgatus and Fusobacterium varium to the colonic mucosa of healthy beagles. Vet Sci. 2024;11:1-11. 10.3390/vetsci11070319 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 40. Kim  J, An  JU, Kim  W, Lee  S, Cho  S. Differences in the gut microbiota of dogs (Canis lupus familiaris) fed a natural diet or a commercial feed revealed by the Illumina MiSeq platform. Gut Pathog. 2017;9:68. 10.1186/s13099-017-0218-5 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 41. Pilla  R, Suchodolski  JS. The role of the canine gut microbiome and metabolome in health and gastrointestinal disease. Front Vet Sci. 2020;6:1-12. 10.3389/fvets.2019.00498 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 42. Stavroulaki  EM, Suchodolski  JS, Xenoulis  PG. Effects of antimicrobials on the gastrointestinal microbiota of dogs and cats. Vet J. 2023;291:105929. 10.1016/j.tvjl.2022.105929 [DOI] [PubMed] [Google Scholar]
  • 43. Reisinger  A, Stübing  H, Suchodolski  JS, Pilla  R, Unterer  S, Busch  K. Comparing treatment effects on dogs with acute hemorrhagic diarrhea syndrome: fecal microbiota transplantation, symptomatic therapy, or antibiotic treatment. J Am Vet Med Assoc. 2024;262:1657-1665. 10.2460/javma.24.03.0153 [DOI] [PubMed] [Google Scholar]
  • 44. Stübing  H, Suchodolski  JS, Reisinger  A, et al.  The effect of metronidazole versus a synbiotic on clinical course and core intestinal microbiota in dogs with acute diarrhea. Vet Sci.  2024;11:1-13. 10.3390/vetsci11050197 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 45. Chen  CC, Pilla  R, Toresson  L, et al.  Microbial gene profiling and targeted metabolomics in fecal samples of dogs with chronic enteropathy with or without increased dysbiosis index. J Vet Intern Med. 2025;39:e70199. 10.1111/jvim.70199 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 46. Tanprasertsuk  J, Jones  R, Tate  D, et al.  Beagle dogs living in household or controlled environments show different gut microbiome profiles when fed an identical diet or diet type. Res Sq. 2023:1-21. 10.21203/rs.3.rs-3167499/v1 [DOI] [Google Scholar]
  • 47. Martini  SE, Schmidt  T, Huang  W, et al.  Effects of metronidazole on the fecal microbiota, fecal metabolites, and serum metabolites of healthy adult cats. Pets.  2025;2:19. 10.3390/pets2020019 [DOI] [Google Scholar]
  • 48. Mekjian  HS, Phillips  SF, Hofmann  AF. Colonic secretion of water and electrolytes induced by bile acids: perfusion studies in man. J Clin Invest. 1971;50:1569-1577. 10.1172/jci106644 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 49. Ma  Y, Pirolo  M, Jana  B, Mebus  VH, Guardabassi  L. The intrinsic macrolide resistome of Escherichia coli. Antimicrob Agents Chemother. 2024;68:e0045224. 10.1128/aac.00452-24 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 50. Gomes  C, Martínez-Puchol  S, Palma  N, et al.  Macrolide resistance mechanisms in Enterobacteriaceae: focus on azithromycin. Crit Rev Microbiol. 2017;43:1-30. 10.3109/1040841x.2015.1136261 [DOI] [PubMed] [Google Scholar]
  • 51. Denham  JD, Nanjappa  S, Greene  JN. Treatment of enteropathogenic Escherichia coli diarrhea in cancer patients: a series of three cases. Case Rep Infect Dis. 2018;2018:8438701. 10.1155/2018/8438701 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 52. Nitschke  M, Sayk  F, Härtel  C, et al.  Association between azithromycin therapy and duration of bacterial shedding among patients with Shiga toxin-producing enteroaggregative Escherichia coli O104:H4. JAMA.  2012;307:1046-1052. 10.1001/jama.2012.264 [DOI] [PubMed] [Google Scholar]
  • 53. Gomes  C, Ruiz-Roldán  L, Mateu  J, Ochoa  TJ, Ruiz  J. Azithromycin resistance levels and mechanisms in Escherichia coli. Sci Rep. 2019;9:6089. 10.1038/s41598-019-42423-3 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 54. Rojas  CA, Park  B, Scarsella  E, et al.  Species-level characterization of the core microbiome in healthy dogs using full-length 16S rRNA gene sequencing. Front Vet Sci. 2024;11:1-17. 10.3389/fvets.2024.1405470 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 55. Suchodolski  JS, Camacho  J, Steiner  JM. Analysis of bacterial diversity in the canine duodenum, jejunum, ileum, and colon by comparative 16S rRNA gene analysis. FEMS Microbiol Ecol. 2008;66:567-578. 10.1111/j.1574-6941.2008.00521.x [DOI] [PubMed] [Google Scholar]
  • 56. Zur  G, Soback  S, Weiss  Y, Perry  E, Lavy  E, Britzi  M. Azithromycin pharmacokinetics in the serum and its distribution to the skin in healthy dogs and dogs with pyoderma. Vet J. 2014;200:122-126. 10.1016/j.tvjl.2013.12.022 [DOI] [PubMed] [Google Scholar]
  • 57. Shepard  RM, Falkner  FC. Pharmacokinetics of azithromycin in rats and dogs. J Antimicrob Chemother. 1990;25:49-60. 10.1093/jac/25.suppl_a.49 [DOI] [PubMed] [Google Scholar]
  • 58. Vecchiato  CG, Sabetti  MC, Sung  CH, et al.  Effect of faecal microbial transplantation on clinical outcome, faecal microbiota and metabolome in dogs with chronic enteropathy refractory to diet. Sci Rep. 2025;15:11957. 10.1038/s41598-025-96906-7 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 59. Li  Q, Larouche-Lebel  É, Loughran  KA, Huh  TP, Suchodolski  JS, Oyama  MA. Gut dysbiosis and its associations with gut microbiota–derived metabolites in dogs with myxomatous mitral valve disease. mSystems.  2021;6:1-16. 10.1128/msystems.00111-21 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 60. Garcia-Mazcorro  JF, Dowd  SE, Poulsen  J, Steiner  JM, Suchodolski  JS. Abundance and short-term temporal variability of fecal microbiota in healthy dogs. MicrobiologyOpen.  2012;1:340-347. 10.1002/mbo3.36 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 61. Forry  SP, Servetas  SL, Kralj  JG, et al.  Variability and bias in microbiome metagenomic sequencing: an interlaboratory study comparing experimental protocols. Sci Rep. 2024;14:1-14. 10.1038/s41598-024-57981-4 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 62. Purina Institute . Purina fecal scoring chart: nutritional and clinical assessment. Purina Institute, Accessed October 7, 2025. https://www.purinainstitute.com/sites/default/files/2024-02/fecal-chart.pdf.
  • 63. Langon  X. Validation of method for faecal sampling in cats and dogs for faecal microbiome analysis. BMC Vet Res. 2023;19:274. 10.1186/s12917-023-03842-7 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 64. Toresson  L, Ludvigsson  U, Olmedal  G, et al.  Repeated fecal microbiota transplantation in dogs with chronic enteropathy can decrease disease activity and corticosteroid usage. J Am Vet Med Assoc. 2025;264:571-580. 10.2460/javma.25.08.0563 [DOI] [PubMed] [Google Scholar]
  • 65. Correa, Lopes B, Chen  CC, Sung  CH, et al.  Correlation between Peptacetobacter hiranonis, the baiCD gene, and secondary bile acids in dogs. Animals (Basel). 2024;14:1-15. 10.3390/ani14020216 [DOI] [PMC free article] [PubMed] [Google Scholar]

Articles from Journal of Veterinary Internal Medicine are provided here courtesy of Oxford University Press

RESOURCES