Background: Although the use of clinical reasoning has been evaluated for several neurological presentations, this approach has not yet been investigated for dogs with vestibular syndrome.
Methods: Two hundred and thirty‐nine dogs presenting with vestibular syndrome were included in this retrospective study. Univariate analysis of variables (clinical history, signalment, clinical presentation and neurological examination findings) was performed. Variables with p < 0.3 were selected for logistic regression.
Results: Ninety‐five percent of dogs were represented by eight conditions: idiopathic vestibular disease (n = 78 dogs), otitis media interna (n = 54), meningoencephalitis of unknown origin (n = 35), brain neoplasia (n = 26), ischaemic infarct (n = 25), intracranial empyema (n = 4), metronidazole toxicity (n = 3) and neoplasia affecting the middle ear (n = 3). Idiopathic vestibular disease was associated with higher age, higher bodyweight, improving clinical signs, pathological nystagmus, facial nerve paresis, absence of Horner's syndrome and a peripheral localisation. Otitis media interna was associated with younger age, male gender, Horner's syndrome, a peripheral localisation and a history of otitis externa. Ischaemic infarct was associated with older age, peracute onset of signs, absence of strabismus and a central localisation.
Conclusions: Discrete clinical features can be used to identify the most likely diagnosis in dogs with vestibular syndrome.