Objective To determine whether phenobarbital treatment of epileptic dogs alters serum thyroxine (T4) and thyroid-stimulating hormone (TSH) concentrations. Design Cross-sectional study. Animals 78 epileptic dogs receiving phenobarbital (group 1) and 48 untreated epileptic dogs (group 2). Procedure Serum biochemical analyses, including T4 and TSH concentrations, were performed for all dogs. Additional in vitro analyses were performed on serum from healthy dogs to determine whether phenobarbital in serum interferes with T4 assays or alters free T, (fT4) concentrations. Results Mean serum T4 concentration was significantly lower, and mean serum TSH concentration significantly higher, in dogs in group 1, compared with those in group 2. Thirty-one (40%) dogs in group 1 had serum T4 concentrations less than the reference range, compared with 4 (8%) dogs in group 2. All dogs in group 2 with low serum T4 concentrations had recently had seizure activity. Five (7%) dogs in group 1, but none of the dogs in group 2, had serum TSH concentrations greater than the reference range. Associations were not detected between serum T4 concentration and TSH concentration, age, phenobarbital dosage, duration of treatment, serum phenobarbital concentration, or degree of seizure control. Signs of overt hypothyroidism were not evident in dogs with low T4 concentrations. Addition of phenobarbital in vitro to serum did not affect determination of T4 concentration and only minimally affected fT4 concentration. Conclusions and Clinical Relevance Clinicians should be aware of the potential for phenobarbital treatment to decrease serum T4 and increase TSH concentrations and should use caution when interpreting results of thyroid tests in dogs receiving phenobarbital. (J Am Vet Med Assoc 1999:215:489–496)