Are Epilepsy and Type 1 Diabetes Linked?

October 29, 2015

Recent findings could explain underlying pathophysiology of these disorders and have therapeutic implications.

A new study may add pharmacoresistant epilepsy to the list of comorbidities associated with type 1 diabetes (T1DM). The UK study found that the prevalence of T1DM was twice as high among patients with pharmacoresistant epilepsy compared to the general population. The findings could shed new light on the underlying pathophysiology of these disorders, and have therapeutic implications.

The study is also important because it reports on the largest group of individuals with comorbid epilepsy and T1DM identified to date. Past studies looking at this issue have suffered from failure to differentiate between T1DM and T2DM.

In the study, researchers did a retrospective review of people with pharmacoresistant epilepsy who had been admitted to a tertiary care institution in the UK between January 2001 and January 2011. Researchers defined pharmacoresistant epilepsy as lack of seizure control during the past 12 months, despite trials of at least two different antiepileptic medications. They diagnosed epilepsy using clinical history, EEG, and MRI. T1DM was self-reported.

Key Results:

• The analysis included 2016 people with pharmacoresistant epilepsy, of which 20 had T1DM (point prevalence: 9.9 cases per 100 individuals).

• Estimates of the prevalence of T1DM in the general population range from 3.4-4.4 per 1000.

• In 80% of patients, T1DM preceded the development of epilepsy by a median of 1.5 years.

• Cryptogenic/unknown epilepsy was significantly more likely with T1DM, compared to T2DM or no diabetes (85% vs 35% and 40%, P=0.045).

• All patients with T1DM and epilepsy had focal epilepsy, of which 45% was temporal lobe epilepsy.

The increased prevalence of epilepsy among those with T1DM may have to do with anti-glutamic acid decarboxylase (GAD-A) antibodies, which could be the potential target of new therapies, the authors conjectured.

“A particular association between T1DM and cryptogenic/unknown epilepsy would be in keeping with the proposed pathophysiological link between T1DM and epilepsy, an autoimmune process associated with antibodies to glutamic acid decarboxylase [GAD-A],” wrote first author Mark Keezer of University College, London, UK, and colleagues in Switzerland and The Netherlands.

About 80% of people with T1DM have GAD-A, which may also be elevated in about 6% of people with epilepsy, according to background information in the article. Past studies have suggested that GAD-A may be more common in people with temporal lobe and pharmacoresistant focal epilepsy than generalized epilepsy. High GAD-A serum levels have also been linked to more severe epilepsy.

Potential therapeutic implications could include immunosuppressive therapy and antiepileptic drugs directed towards GABA, which may be more effective in patients with GAD-A, according to the authors.

“The prevalence of T1DM appears to be increased in people with pharmacoresistant epilepsy and is associated with cryptogenic/unknown epilepsy,” the authors concluded, “…Our findings may have pathophysiological as well as therapeutic implications, especially in the context of previous evidence regarding GAD-A.”

Take Home Points

• A retrospective cohort study in the UK found that the prevalence of pharmacoresistant epilepsy is two times higher among patients with T1DM compared to estimates for the general population.

• In 80% of patients, T1DM preceded the development of epilepsy, and the vast majority had cryptogenic/unknown epilepsy.

• Many people with T1DM have anti-glutamic acid decarboxylase antibodies (GAD-A), which are also found in people with epilepsy.

• GAD-A may be involved in the underlying pathophysiology of T1DM and epilepsy, which could have therapeutic implications.

Reference: Keezer MR, et al. Type 1 diabetes mellitus in people with pharmacoresistant epilepsy: Prevalence and clinical characteristics. Epilepsy Res. 2015 Sep;115:55-57.