The authors identify several important limitations.
Small sample
Only 20 dogs were studied, and dividing them between stress paradigms reduced statistical power further.
Small microbiome changes may therefore have been missed.
Colony dogs
The dogs lived within a controlled research colony.
They may not represent the wider pet-dog population.
The dogs appeared to habituate
Repeated exposure may have caused some dogs to become accustomed to the stress events.
Different results might occur with more severe, frequent or prolonged stress.
Acute rather than chronic stress
The study investigated short-term stressful experiences.
It does not tell us what happens in dogs experiencing chronic anxiety or persistent environmental stress.
Microbiota composition is only one layer
Although shotgun sequencing provides detailed taxonomic and functional information, the study did not comprehensively measure the metabolome.
It is possible for microbial activity or gut-brain signalling to change without major changes in microbial composition.
Timing of sampling
Faecal samples were collected within the first 48 hours after the events.
Changes occurring outside that sampling window would not necessarily have been detected.
For these reasons, the results should not be interpreted as showing that stress can never influence the canine microbiome.
They show that these particular acute stress events did not produce detectable microbiome changes in this particular population.
