Type 1 diabetes and epilepsy: More than a casual association?

To the Editors: In 2006, McCorry et al. reported a strong epidemiologic association between type 1 diabetes mellitus (T1DM) and idiopathic generalized epilepsy (IGE) and claimed that this link would support an autoimmune pathogenesis for IGE (McCorry et al., 2006). However, O’Connel and coworkers did not confirm an increased risk of epilepsy among a series of patients with T1DM (O’Connell et al., 2008). We reviewed 249 T1DM patients (145 male; mean age 16.3 ± 7.02 years, range 1.7–34.5 years; mean age at disease onset 7.7 ± 6.9 years; range 0.8–18 years) attending the Diabetes Clinic at G. Gaslini Institute, Genoa, where a collaborative network provides epidemiologic data by an active surveillance model (Cotellessa et al., 2003). Extensive research was undertaken through the electronic database and subjects not followed up during the previous year were excluded. Epilepsy was diagnosed when ≥2 unprovoked euglycemic seizures (blood glucose >3.9 and <11.1 mmol/L) and syndromic classification (Commission on Classification and Terminology of the ILAE, 1989) was confirmed by reviewing clinical (family history, seizures type onset and frequency, response to treatment) and instrumental [electroencephalography (EEG), neuroimaging] data. Among 249 T1DM patients, 6 showed the electroclinical features of idiopathic epilepsy (4 IGE, 2 focal idiopathic epilepsy) (Table 1). In this group, mean age of T1DM onset was 4.6 ± 2.7 years; mean age at first seizure was 6.7 ± 4.4 years. None of these patients had a family history of epilepsy in first- and second-degree relatives. In all patients but one (case 4), T1DM preceded epilepsy onset, with a mean of 2.8 years (range 0.7–6.9 years). Patient 1 had concomitant celiac disease. Dosage of glutamic-acid decarboxylase antibodies (GADA) was available in two patients at T1DM diagnosis and in all cases at the last follow-up (Table 1). The high proportion of IGE observed in our series (4 of 249, 1.6%) (Hauser & Banerjee, 2008) confirms the data reported by McCorry et al. (McCorry et al., 2006). In all patients but one, T1DM manifested years before epilepsy. This may simply reflect different age of onset of the two conditions. Alternatively, it could support an autoimmune mechanism starting with T1DM and then affecting the central nervous system (McCorry et al., 2006). We did not perform GADA evaluation methodically at epilepsy onset in our patients, so that no definitive conclusion can be drawn about the existence of a pathogenetic link between GADA and epilepsy in T1DM patients (Striano et al., 2008). Interestingly, the patient with partial response to antiepileptic therapy (case 4) is the only one with recent GADA positivity. Because of the retrospective analysis, epilepsy could be underrecognized or overestimated in our T1DM population, but this is unlikely to explain the higher than expected degree of the association. Because both epilepsy and T1DM are serious worldwide problems with high medical and social management costs, this possible association deserves further investigation by prospective studies involving multiple centers. We confirm that we have read the Journal’s position on issues involved in ethical publication and affirm that this report is consistent with those guidelines. None of the authors has any conflict of interest to disclose.