Ashwagandha for Dogs: Evidence, Mechanisms, and Limitations

What ashwagandha is and how it is classified

Ashwagandha (Withania somnifera) is a plant in the nightshade family (Solanaceae) used for centuries in Ayurvedic practice. The root and root extract — referred to in the veterinary literature as ARE (ashwagandha root extract) — are the forms most commonly studied in dogs. As a botanical, ashwagandha belongs to the broader category of adaptogens: plant compounds proposed to support the body's ability to regulate physiological stress responses.

A narrative review of adaptogens in humans and dogs notes that plant substances in this class have been described as supporting the regulation of cortisol levels, neurotransmission, and neuroplasticity, though the review acknowledges that research in dogs specifically remains limited and that further study is needed to confirm effectiveness and safety in animal populations (Kępińska-Pacelik and Biel, 2025; DOI: 10.3390/app15105402). The same review observes that dogs, like humans, respond to plant compounds, which frames the scientific interest in extrapolating findings from human adaptogen research to veterinary practice, while also highlighting the evidence gap.

Ashwagandha's proposed mechanisms in the stress-axis context center on modulating cortisol, a glucocorticoid that rises in response to physiological and psychological stressors. Controlled dog trials have tested whether oral ARE supplementation shifts measurable cortisol and related biomarkers — work summarized in the sections below.

What the behavioral trial found

A small study has measured anxiety-related behavior directly. In a randomized, double-blind, placebo-controlled trial, 24 dogs experiencing stress or anxiety received ashwagandha root extract or placebo for four weeks. The published abstract reports an improvement in the owner-rated fear and anxiety domain compared with placebo (Kaur et al., 2022; DOI: 10.1016/j.jveb.2022.03.002).

This guide's assessment of that study is limited to its publisher abstract; the full text was not available for appraisal. A short trial in a small group is an early signal. It does not establish that different extracts, combination supplements or long-term use will produce the same result.

Cortisol and stress-axis findings in dogs

Evidence about cortisol comes from a randomized, double-blind, placebo-controlled trial (n=20 healthy geriatric dogs) by Bharani et al. (2024; PMCID: PMC11288135), which administered ARE orally at 15 mg/kg daily for 60 days. The trial found that serum cortisol levels were significantly lower in the ARE-treated group at day 60 compared to placebo, a directionally meaningful result given cortisol's role as a primary biochemical marker of physiological stress (Bharani et al., 2024; PMCID: PMC11288135).

Several contextual limitations apply to this finding. The study enrolled healthy geriatric dogs, not dogs with documented anxiety or stress disorders. The sample was small (n=20) and the population was geriatric, which limits generalizability to younger or clinically anxious dogs. No behavioral outcomes — such as observed anxiety signs or behavioral rating scales — were reported alongside the cortisol measurement.

The same trial reported that leukocyte count decreased significantly after 60 days of ARE treatment (p < 0.05), which the authors interpreted as potentially indicating reduced systemic inflammation; however, leukocyte count is an indirect proxy and the interpretation is speculative without clinical confirmation (Bharani et al., 2024; PMCID: PMC11288135). Serum glucose levels were also significantly reduced in ARE-treated dogs at both 30 and 60 days compared to placebo (Bharani et al., 2024; PMCID: PMC11288135), a finding the authors noted but did not attribute to a specific mechanism within the bounds of the study.

The cortisol result in this geriatric study does not show whether the dogs felt or behaved less anxiously. Kaur et al. (2022; DOI: 10.1016/j.jveb.2022.03.002) separately measured owner-rated behavior; the two kinds of outcome should not be treated as interchangeable.

Antioxidant and inflammation markers

The Bharani et al. (2024; PMCID: PMC11288135) RCT measured a panel of oxidative stress and inflammatory biomarkers alongside the cortisol data. Taken together, the antioxidant findings were consistently directional:

Gut health and microbiome findings

A second RCT from the same research group (Bharani et al., 2025; PMCID: PMC11794502) investigated ARE's effects on gut parameters in a separate cohort of 12 healthy geriatric beagle dogs over 60 days, using a randomized, double-blind, placebo-controlled design. The primary focus was gastrointestinal physiology.

Key findings from this trial include:

• Short-chain fatty acids (SCFAs): Total faecal SCFA levels were significantly higher in ARE-treated dogs after 60 days compared to placebo (Bharani et al., 2025; PMCID: PMC11794502). Propionic acid — one SCFA constituent — was specifically higher in the ARE group at day 60 (p < 0.01) (Bharani et al., 2025; PMCID: PMC11794502). SCFAs are produced by gut bacteria during fermentation and are associated with intestinal health, though their direct relevance to anxiety or stress is not established in this trial.
• Faecal score: The faecal score — a measure of stool quality — was significantly lower (improved) in ARE-treated dogs at day 30 and day 60 compared to placebo (Bharani et al., 2025; PMCID: PMC11794502).
• Plasma L-citrulline: Plasma L-citrulline was significantly higher in ARE-treated dogs after 60 days (Bharani et al., 2025; PMCID: PMC11794502). L-citrulline is a marker of intestinal function and enterocyte health.
• Haematological endpoints: RBC counts were significantly higher in ARE-treated dogs at day 60 (p < 0.01), and haemoglobin levels were higher at both 30 and 60 days (p < 0.001 vs. placebo) (Bharani et al., 2025; PMCID: PMC11794502). ALT and AST were significantly lower in the ARE-treated group at day 30 and day 60 (Bharani et al., 2025; PMCID: PMC11794502), replicating the liver-enzyme direction observed in the 2024 trial.
• Null findings: ARE did not significantly alter faecal pH, plasma lactate, or intestinal-type alkaline phosphatase (I-ALP) levels. Glucose, urea, and creatinine were also unaltered, suggesting no negative renal effects from ARE in this cohort (Bharani et al., 2025; PMCID: PMC11794502).

The 2025 trial enrolled healthy research beagles only (n=12), which limits generalizability to mixed-breed pet dog populations and to dogs with gastrointestinal disease.

Multi-herb blends containing Withania somnifera

Beyond isolated ARE studies, one small trial has examined a multi-ingredient supplement containing ashwagandha alongside other botanicals. Ciarcia et al. (2025; PMCID: PMC12696704) conducted an uncontrolled case series of 7 senior dogs receiving a blend of Passiflora incarnata, Withania somnifera (ashwagandha), and Taraxacum officinale (dandelion root). Key findings included:

• Inflammatory markers: C-reactive protein (CRP) and serum IL-6 were statistically lower at 20 and 40 days of supplementation compared to baseline (Ciarcia et al., 2025; PMCID: PMC12696704). IL-10 also decreased significantly at both time points, though the clinical direction of this change is ambiguous without a control group (Ciarcia et al., 2025; PMCID: PMC12696704).
• Oxidative stress: Total antioxidant capacity (TAC) increased significantly after 40 days, while MDA — a lipid peroxidation marker — was significantly reduced by 40 days (Ciarcia et al., 2025; PMCID: PMC12696704). A marker of DNA oxidative damage (8OHdG) also decreased significantly by day 40 (Ciarcia et al., 2025; PMCID: PMC12696704).
• Gut microbiota: No statistically significant changes in alpha or beta diversity of the gut microbiota were detected over the 40-day period (Ciarcia et al., 2025; PMCID: PMC12696704).
• Safety markers: Glucose, creatinine, BUN, SDMA, haemoglobin reticulocyte, and albumin remained within normal ranges throughout the trial, with no significant pre-post differences (Ciarcia et al., 2025; PMCID: PMC12696704).

The uncontrolled design means that causal attribution to any ingredient — including ashwagandha — is not possible from this study. The blend's effects cannot be disentangled across its three botanical components. The sample of 7 dogs is substantially underpowered for reliable effect estimation.

Evidence gaps and limitations

The studies reviewed here include a small behavioral trial, two geriatric biomarker trials, an uncontrolled multi-herb case series and a narrative review. They answer different questions. The following limits matter when interpreting their findings:

Frequently asked questions