To explore the hypothesis that raised anticardiolipin antibodies, glutamic acid decarboxylase, and antinuclear antibodies may be associated with epilepsy and/or pharmacoresistance, we studied titers in 74 epileptic patients and 50 controls. Epileptic patients were divided into two groups according to their response to anticonvulsant therapy. Group I included 52 children (30 females and 22 males with a mean age+/-SD of 7.0+/-2.4 years) suffering from different types of epilepsy who were treated with various anticonvulsants. Group II included 22 children (10 females and 12 males with a mean age of 6.2+/-3.6 years) suffering from therapy resistant epilepsy. We found that the prevalence of anticardiolipin antibodies was significantly higher in epileptic patients than in controls, while there was no significant difference between patients who were seizure free and those with uncontrolled epilepsy. No significant difference was found in glutamic acid decarboxylase antibodies between epileptic children and controls, and between patients who were seizure free and those with uncontrolled epilepsy. A significant difference in the incidence of antinuclear antibodies was found between epileptic children and controls, while no difference was found between well-controlled and drug-resistant epilepsy. In conclusion, the prevalence of anticardiolipin and antinuclear antibodies was higher in patients with epilepsy than in controls. There was no significant difference in serum glutamic acid decarboxylase antibodies between epileptic children and controls, and between patients who were seizure free and those with uncontrolled epilepsy.