Journal of the Neurological Sciences 346 (2014) 307–309
Contents lists available at ScienceDirect
Journal of the Neurological Sciences journal homepage: www.elsevier.com/locate/jns
Short communication
Hypoglycemic seizures and epilepsy in type I diabetes mellitus Mercè Falip a,⁎, Júlia Miró a, Mar Carreño b, Sònia Jaraba a, Juan Luís Becerra c, Núria Cayuela a, Manuel Perez Maraver d, Francesc Graus e a
Epilepsy Unit, Neurology Service, Hospital Universitari de Bellvitge, Hospitalet de Llobregat, Spain Epilepsy Unit, Neurology Service, Hospital Clinic I Provincial de Barcelona, Barcelona, Spain Epilepsy Unit, Neurology Service, Hospital Germans Trias I Pujol, Badalona, Spain d Endocrinology Service, Hospital Universitari de Bellvitge, Hospitalet de Llobregat, Spain e Neuroimmunology Unit, Neurology Service, Hospital Clinic i Provincial de Barcelona, Barcelona, Spain b c
a r t i c l e
i n f o
Article history: Received 18 June 2014 Received in revised form 11 August 2014 Accepted 18 August 2014 Available online 27 August 2014 Keywords: Temporal lobe epilepsy Acute seizures Glutamic acid decarboxylase antibodies Hypoglycemia Diabetes mellitus type 1 Autoimmune epilepsy
a b s t r a c t Purpose: (1) To determine the characteristics of seizures and/or epilepsy among patients with adult onset type 1 diabetes mellitus (T1DM), and (2) to determine glutamic acid decarboxylase antibody (antiGADab) titres and other autoimmune characteristics of T1DM patients with seizure and/or epilepsy. Methods: Patients were recruited after reviewing the databases of one Diabetes Unit and two Epilepsy Units. Prevalence of suffering epilepsy and/or seizures was calculated in the group of adult onset T1DM patients seen consecutively in the Diabetes Unit. Serum antiGADab titres were determined in all patients. Results: Among the 229 T1DM patients, two had suffered hypoglycemic seizures (0.87%) and two unprovoked seizures (0.87%). We recruited 11 of these T1DM patients. Five reported only hypoglycemic seizures, and 6 temporal lobe epilepsy (TLE). Mean antiGADab titres in patients with T1DM and hypoglycemic seizures were lower (18.42 IU) than in those with T1DM and TLE (29.200 IU), p: 0.006. Patients with T1DM and TLE suffered more other autoimmune diseases than those with T1DM and hypoglycemic seizures, p: 0.015. Conclusions: Hypoglycemic seizures are rare in T1DM patients. AntiGADab titres do not differ from the general diabetic population. Patients with high titres of antiGADab are more likely to develop TLE. © 2014 Elsevier B.V. All rights reserved.
1. Introduction Patients with T1DM may experience hypoglycemic seizures or they may go on to develop epilepsy which is sometimes drug resistant. Hypoglycemia and hypoglycemic seizures used to be a common problem in diabetic patients treated with insulin [1]. Currently, rates of hypoglycemia and hypoglycemic seizures are lower, ranging between 4.5 and 14.4 per 100 patients depending principally on insulin treatment and age at diabetes onset [2]. The relationship between hypoglycemia and epileptic phenomena is complex. Experimental studies have shown that hypoglycemia increases cortical excitability, this hyperexcitability being more pronounced in the temporal lobe and hippocampus [3]. Several studies have pointed out an increased prevalence of epilepsy in T1DM respect to the normal population [4]. On the other hand, elevated levels of antiGADab, GAD being a major autoantigen in T1DM, seem to be involved in the pathogenesis of a number of neurological ⁎ Corresponding author at: Epilepsy Unit, Neurology Service, Hospital de Bellvitge, Feixa LLarga S/N, 08907 Hospitalet de Llobregat, Spain. Tel.: + 34 932607711; fax: + 34 932607864. E-mail address: [email protected] (M. Falip).
http://dx.doi.org/10.1016/j.jns.2014.08.024 0022-510X/© 2014 Elsevier B.V. All rights reserved.
disorders, including limbic encephalitis and pharmacoresistant TLE [5]. Many of these patients with antiGADab and drug resistant epilepsy also have adult onset T1DM. However, the significance of low titres of antiGADab in patients with T1DM and seizures or epilepsy remains unknown. The aims of the study are: to determine the prevalence and characteristics of patients suffering seizures and/or epilepsy among patients with adult onset T1DM, and to determine if there is any type of relationship between the presence and titres of anti GADab and seizures or epilepsy. 2. Material and methods 2.1. Study design Patients were recruited in the outpatient clinic of the Diabetes Unit of the Hospital Universitari de Bellvitge and in the Epilepsy Units of Hospital Universitari de Bellvitge and Hospital Clinic. Prevalence of suffering epilepsy and/or seizures was calculated in the group of adult onset T1DM patients seen consecutively in the Diabetes Unit. Due to the low prevalence of epilepsy in T1DM, to study the seizure and/or epilepsy features in this population of patients, we also used a
308
M. Falip et al. / Journal of the Neurological Sciences 346 (2014) 307–309
preexisting database containing patients (from the Bellvitge and the Clinic Hospitals) with seizures and positive antiGAD antibodies. 2.2. Definitions The following definitions were used to classify patients and seizures: - Hypoglycemic seizure Hypoglycemic seizure was a seizure temporally related to hypoglycemia (bthan 36 mg/dl) [6]. - Epilepsy Two or more unprovoked seizures (normal glucose level) separated by more than 24 h. 2.3. Immunological analysis A general immunological battery test was done, all antibodies being determined in EDTA plasma. AntiGADab was analyzed by immunohistochemistry and radioimmunoassay (RIA) in the Hospital Clinic of Barcelona, as described elsewhere [7], with values N 1 IU/ml considered positive. In patients with high titres of antiGADab (N 1000 IU), CSF antiGADab was also determined and intrathecal synthesis (IS) was calculated as described elsewhere [7].
hypoglycemic seizures were described as loss of awareness with oromandibular automatism without secondary generalization, often preceded by psychic or epigastric auras. EEG (routine) and MRI were normal in all patients with hypoglycemic seizures. EEG performed in the first 24 h after hypoglycemic seizures showed a continuous temporal slowing in 1/2 patients. Most of our patients suffering several hypoglycemic seizures 4/5 (80%) were erroneously diagnosed of epilepsy and were treated chronically with antiepileptic drugs. No changes in hypoglycemic seizure frequency or severity were observed; hypoglycemic seizures were only controlled after insulin treatment was adjusted. The usual treatment in these patients is a combination of long-acting insulin at night and rapid-acting insulin analog administered with main meals. After a severe hypoglycemic episode insulin doses are transiently reduced while the cause is evaluated and treated, particularly with regard to diet and exercise. Regarding patients with epilepsy, all of them had focal, temporal lobe epilepsy. Routine EEG showed interictal epileptiform discharges in one or both temporal lobes in 4/6 (66%) patients. Unprovoked seizures (normal glycemia) arising from one or both temporal lobe were recorded in 3/6 (50%) patients. MRI was normal in all patients with epilepsy, except one in whom a signal increase was observed in both hippocampi, without concomitant atrophy (see Table 1).
2.4. Other diagnostic tests Magnetic resonance imaging was performed in all patients with a 1.5-T unit (Philips Medical Systems, Best, The Netherlands). At least one routine EEG was performed in all patients (64 channel digital DeltaMed equipment). 2.5. Ethical approval The study was approved by the Ethical Committee of Hospital Universitari de Bellvitge. 2.6. Statistics Fisher's exact test and the Mann–Whitney U test were used for nominal and continuous data, respectively. 3. Results 3.1. Seizures and epilepsy prevalence
Table 1 Clinical and immunological characteristics of T1DM patients suffering either hypoglycemic seizures or TLE.
Gender Age (years) DM onset (years) Epilepsy onset (years) Age at first hypo Glycemia (mg/ml) Number of hypo. seizures Nocturnal seizures Aura GTCS CPS Interictal EEG EEG 24 h Ictal EEG
Among 229 consecutive patients with adult onset T1DM, whose onset of symptoms was between 2004 and 2012 and who were included in a diabetes database, 4 patients had suffered seizures. The seizures occurred in close temporal relationship with well documented hypoglycemic episodes in 2 patients. Two patients had epilepsy. Thus, the prevalence of epilepsy in our population was 2/229 (0.8%) and of hypoglycemic seizures of 2/229 (0.8%). From the two neurology databases 11 patients with seizure/epilepsy had positive antiGAD antibodies, 9 of them with T1DM. After excluding two patients who had been included in both the diabetes and neurology databases, we were left with a total of 11 patients with T1DM and seizures or epilepsy. Of these, 5 had only suffered hypoglycemic seizures. 4 had drug resistant TLE and 2 had well controlled TLE. In 4/6 (66%) patients with TLE + T1DM the onset of epilepsy was earlier than that of diabetes. On the other hand in all patients with hypoglycemic seizures the onset of diabetes was earlier. 3.2. Seizure and epilepsy characteristics Most hypoglycemic seizures were nocturnal, and seizure semiology was consistent with generalized tonic–clonic seizures. Diurnal
MRI Seizure free on AED Serum antiGAD (IU/ml), RIA CSF antiGAD (IU/ml), RIA AntiGAD IS AntiIA2 Antithyroid ab Antigliadin ab ANAS AntiDNA AntiCLP Other autoimmune disease
T1DM + hypoglycemic seizures
T1DM + TLE
N=5
N=6
4 males/1 females 36.6 (34–42) 23.4 (15–31) –
2 males/4 females 48.2 (33–65) 38.6 (27–55) 35.7 (22–51)
29.4 (23–36) 37.4 (34–42) 6 (2–15)
– – –
3/5 (60%) 2/5 (40%) 5/5 (100%) 5/5 (100%) 5/5 Normal 2/6 unilateral temporal lobe slowing Not done
0/6 (0%) 4/6 (66%) 1/6 (16%) 2/6 (33%) 2/6 normal –
5/5 Normal 0/5
Unilateral TLS 1/6 Bilateral TLS 2/6 5/6 normal 2/6
18.4
29.200
Not done
48.4 (N = 4)
– 5/5 negative 0/5 (0%) 0/5 (0%) 0/5 (0%) 0/5 (0%) 0/5 (0%) 0/5 (0%)
4/4 (100%) 5/6 negative 4/6 (66%) 3/6 (50%) 1/6 (16%) 1/6 (16%) 1/6 (16%) 4/6 (66%)
Significance
NS NS p: 0.053
p: 0.061 NS p: 0.061 NS NS
NS
p: 0.006
NS p: 0.015 NS NS NS p: 0.015
DM: diabetes mellitus, Hypo: hypoglycemia, EEG: electroencephalogram, MRI: magnetic resonance image, GTCS: generalized tonic–clonic seizure, CPS: complex partial seizure, IS: intrathecal synthesis, CSF: cerebrospinal fluid, RIA: radioimmunoprecipitation assay, AntiIA2: Antiinsulinoma-associated antigen-2 autoantibodies, AED: antiepileptic drug, EEG 24 h: electroencephalogram done during the first 24 h after a seizure, TLS: temporal lobe seizure, ANAS: antinuclear antibodies, antiCLP: anticardiolipin antibodies, NS: not significant.
M. Falip et al. / Journal of the Neurological Sciences 346 (2014) 307–309
3.3. AntiGADab titres and other autoimmune characteristics Mean serum antiGADab titre in patients with hypoglycemic seizures was 18.42 IU (2–32 IU) which was similar to the mean serum antiGADab titres of the general T1DM population (30 IU) (p: not significant). Mean serum antiGADab titre in patients with well controlled TLE was 43.442 IU and in drug resistant TLE was 22.079 IU. These titres were higher than in T1DM patients with hypoglycemic seizures (p: 0.006). Intrathecal synthesis of antiGADab was positive in 4/4 (100%) patients in whom it was determined. None of the patients with T1DM and hypoglycemic seizures suffered from other autoimmune diseases; in contrast, 4/6 (66%) patients with epilepsy and T1DM reported one or more other autoimmune diseases; celiac disease (3), myasthenia gravis (1), systemic lupus erythematosus (1), vitiligus (1), hypothyroidism (1) (p: 0.015). In addition, other autoantibodies were observed in this group, the most frequent being antithyroid antibodies, found in 4/6 (66%) patients (p: 0.015). No other autoantibodies were found in patients with hypoglycemic seizures. 4. Discussion In our cohort of consecutive adult T1DM patients with seizures and/ or epilepsy, hypoglycemic seizures were rare (0.87%), as was epilepsy (0.87%); this result does not support the hypothesis suggested by other studies that epilepsy is more frequent in T1DM patients [4], but is in agreement with the study by O'Connell et al. [8]. Although many hypoglycemic seizures are generalized tonic–clonic, some studies, in which hypoglycemic seizures had been recorded, suggest that they arise in the temporal lobe and have secondary generalization [9]. This may have happened in our patients, as suggested by seizure semiology and EEG performed in the first 24 h. The mechanisms underlying epileptic phenomena during hypoglycemia remain unknown. In one study using transcraneal magnetic stimulation it was observed that cortical excitability changed with fluctuations in blood glucose levels [10]. Moreover, hypoglycemia may induce acute symptomatic seizures and also predispose to the development of epileptogenic processes. It is generally accepted that recurrent episodes of hypoglycemia can result in epilepsy due to the possible permanent damage in the hippocampus [11]. Our results do not support this idea, as the subjects in our study suffered several episodes (in one case more than 10) of hypoglycemic seizures without becoming epileptic. Animal studies have suggested that the severity of hypoglycemia is directly related to hippocampal damage and not to seizure severity [12]. It is possible that repeated severe episodes of hypoglycemia may cause permanent damage to the hippocampus and in consequence, the potential to develop epileptogenesis. On the other hand, all our patients with hypoglycemic seizures were only diagnosed after suffering a generalized tonic–clonic seizure. We hypothesize that some of the autonomic symptoms of hypoglycemia may be simple partial seizures arising in the temporal lobe. A correct diagnosis of hypoglycemic seizures is very important in these patients, who may be treated with antiepileptic drugs without noticeable benefit, but exposed to deleterious side effects like the metabolic ones (weight gain and insulin resistance with valproate, for example). Previous studies have shown that antiGADab produce an imbalance between excitatory and inhibitory neurotransmitters [13]. However, not all patients with high antiGADab will be epileptic. In our series of 229 T1DM patients, 3 patients had high antiGADab titres without epilepsy, but all of them had other neurological diseases (Stiff-person syndrome, multiple sclerosis or ataxia). Interestingly the antiGADab titres of our patients with hypoglycemic seizures were lower than the levels observed in those with epilepsy + T1DM. None of our patients with very low antiGADab (b50 IU) had epilepsy and all of our drug resistant patients had high titres of antiGADab, although high titres may also be observed in patients with well controlled epilepsy. Mean antiGAD ab titres in patients with well controlled TLE was higher than in drug
309
resistant TLE (43.442 vs 22.079 IU) because one well controlled TLE patient had very high titres (86.300 IU) and probably, due to small sample size, false differences have been observed. Patients with high antiGADab titres, in addition to epilepsy tend to suffer from other autoimmune disorders and, had a higher number of other autoantibodies, a finding also reported by other authors [14]. Interestingly epilepsy tends to debut before T1DM. Intrathecal synthesis was observed in 4/4 patients (100%), indicating a clonal B cell activation in the CNS as observed in patients with other neurological diseases associated to antiGADab [15]. Our study has some limitations: the lack of ictal recording in patients suffering hypoglycemic seizures, the lack of capill.lar glycemia determination in the majority of seizures considered unprovoked, the small sample size, all patients had suffered adult onset T1DM and no long term follow up of the patients. Hence the conclusions must be taken with caution. In summary, our study suggests that hypoglycemic seizures are nowadays rare in T1DM patients. In these patients, antiGADab titres do not differ from the general diabetic population. Patients suffering several hypoglycemic seizures may not necessarily become epileptic. Patients with high titres of antiGADab are more likely to develop epilepsy, generally temporal and often drug resistant. They are also at risk of developing other autoimmune disorders with time. Conflict of interest Dr Falip, Dr Miró, Dr Carreño, Dr Jaraba, Dr Becerra, Dr Cayuela, and Dr Perez Maraver have no conflicts of interest. Dr Graus serves on the editorial board of Lancet Neurology and has received research support from the Fondo Investigaciones Sanitarias. We confirm that we have read the Journal position on issues involved in the ethical publication and affirm that this report is consistent with those guidelines. Acknowledgments The authors thank all the patients for their collaboration and Mr Dave McFarlane for the technical assistance. This study was supported in part by a grant from the Spanish National Institute of Health (PI10/ 00738). References [1] Varghese P, Gleason V, Sorokin R, et al. Hypoglycaemia in hospitalized patients treated with antihyperglycaemic agents. J Hosp Med 2007;4:234–40. [2] Katz ML, Volkening LK, Anderson BJ, et al. Contemporary rates of severe hypoglycaemia in youth with type 1 diabetes: variability by insulin regimen. Diabet Med 2012;29:926–32. [3] Auer RN. Hypoglycemic brain damage. Metab Brain Dis 2004;19:169–75. [4] Shober E, Otto KP, Dost A, et al. Association of epilepsy and type 1 diabetes mellitus in children and adolescents: is there an increased risk for diabetic ketoacidosis? J Pediatr 2012;160:662–6. [5] Falip M, Carreño M, Miró J, et al. Prevalence and immunological spectrum of temporal lobe epilepsy with glutamic acid decarboxylase antibodies. Eur J Neurol 2012;19: 817–33. [6] Beghi E, Carpio A, Forsgren L, et al. Recommendation for a definition of acute symptomatic seizure. Epilepsia 2010;51:671–5. [7] Saiz A, Arpa J, Sagasta A, et al. Autoantibodies to glutamic acid decarboxylase in three patients with cerebellar ataxia, late-onset insulin-dependent diabetes mellitus, and polyendocrine autoimmunity. Neurology 1997;49:1026–30. [8] O'Connell NM, Harvey A, Mackay M, et al. Does epilepsy occur more frequently in children with type 1 diabetes? J Paediatr Child Health 2008;44:586–9. [9] Lapenta L, Di Bonaventura C, Fattouch J, et al. Focal epileptic seizure induced by transient hypoglycaemia in insulin-treated diabetes. Epileptic Disord 2010;12:84–7. [10] Badawy RA, Vogrin SJ, Lai A, et al. Cortical excitability changes correlate with fluctuations in glucose levels in patients with epilepsy. Epilepsy Behav 2013;27:455–60. [11] Sherin A, Anu J, Peeyush KT, et al. Cholinergic and GABAergic receptor functional deficit in the hippocampus of insulin-induced hypoglycemic and streptozotozininduced diabetic rats. Neuroscience 2012;202:69–76. [12] Schauwecker PE. The effects of glycemic control on seizures and seizure-induced exitotoxic cell death. BMC Neurosci 2012;13:94. [13] Grimaldi G, Monto M. Brain H-MR spectroscopy in anti-GAD antibodies cerebellar ataxia. J Neurol 2010;37(5):303–5. [14] Liimatainen S, Peltola M, Sabater L, et al. Clinical significance of glutamic acid decarboxylase antibodies in patients with epilepsy. Epilepsia 2010;51:760–7. [15] Dalakas M, Li M, Fujjii M, et al. Stiff person syndrome. Quantification, specificity, and intrathecal synthesis of GAD65 antibodies. Neurology 2001;57:780–4.
Contents lists available at ScienceDirect
Journal of the Neurological Sciences journal homepage: www.elsevier.com/locate/jns
Short communication
Hypoglycemic seizures and epilepsy in type I diabetes mellitus Mercè Falip a,⁎, Júlia Miró a, Mar Carreño b, Sònia Jaraba a, Juan Luís Becerra c, Núria Cayuela a, Manuel Perez Maraver d, Francesc Graus e a
Epilepsy Unit, Neurology Service, Hospital Universitari de Bellvitge, Hospitalet de Llobregat, Spain Epilepsy Unit, Neurology Service, Hospital Clinic I Provincial de Barcelona, Barcelona, Spain Epilepsy Unit, Neurology Service, Hospital Germans Trias I Pujol, Badalona, Spain d Endocrinology Service, Hospital Universitari de Bellvitge, Hospitalet de Llobregat, Spain e Neuroimmunology Unit, Neurology Service, Hospital Clinic i Provincial de Barcelona, Barcelona, Spain b c
a r t i c l e
i n f o
Article history: Received 18 June 2014 Received in revised form 11 August 2014 Accepted 18 August 2014 Available online 27 August 2014 Keywords: Temporal lobe epilepsy Acute seizures Glutamic acid decarboxylase antibodies Hypoglycemia Diabetes mellitus type 1 Autoimmune epilepsy
a b s t r a c t Purpose: (1) To determine the characteristics of seizures and/or epilepsy among patients with adult onset type 1 diabetes mellitus (T1DM), and (2) to determine glutamic acid decarboxylase antibody (antiGADab) titres and other autoimmune characteristics of T1DM patients with seizure and/or epilepsy. Methods: Patients were recruited after reviewing the databases of one Diabetes Unit and two Epilepsy Units. Prevalence of suffering epilepsy and/or seizures was calculated in the group of adult onset T1DM patients seen consecutively in the Diabetes Unit. Serum antiGADab titres were determined in all patients. Results: Among the 229 T1DM patients, two had suffered hypoglycemic seizures (0.87%) and two unprovoked seizures (0.87%). We recruited 11 of these T1DM patients. Five reported only hypoglycemic seizures, and 6 temporal lobe epilepsy (TLE). Mean antiGADab titres in patients with T1DM and hypoglycemic seizures were lower (18.42 IU) than in those with T1DM and TLE (29.200 IU), p: 0.006. Patients with T1DM and TLE suffered more other autoimmune diseases than those with T1DM and hypoglycemic seizures, p: 0.015. Conclusions: Hypoglycemic seizures are rare in T1DM patients. AntiGADab titres do not differ from the general diabetic population. Patients with high titres of antiGADab are more likely to develop TLE. © 2014 Elsevier B.V. All rights reserved.
1. Introduction Patients with T1DM may experience hypoglycemic seizures or they may go on to develop epilepsy which is sometimes drug resistant. Hypoglycemia and hypoglycemic seizures used to be a common problem in diabetic patients treated with insulin [1]. Currently, rates of hypoglycemia and hypoglycemic seizures are lower, ranging between 4.5 and 14.4 per 100 patients depending principally on insulin treatment and age at diabetes onset [2]. The relationship between hypoglycemia and epileptic phenomena is complex. Experimental studies have shown that hypoglycemia increases cortical excitability, this hyperexcitability being more pronounced in the temporal lobe and hippocampus [3]. Several studies have pointed out an increased prevalence of epilepsy in T1DM respect to the normal population [4]. On the other hand, elevated levels of antiGADab, GAD being a major autoantigen in T1DM, seem to be involved in the pathogenesis of a number of neurological ⁎ Corresponding author at: Epilepsy Unit, Neurology Service, Hospital de Bellvitge, Feixa LLarga S/N, 08907 Hospitalet de Llobregat, Spain. Tel.: + 34 932607711; fax: + 34 932607864. E-mail address: [email protected] (M. Falip).
http://dx.doi.org/10.1016/j.jns.2014.08.024 0022-510X/© 2014 Elsevier B.V. All rights reserved.
disorders, including limbic encephalitis and pharmacoresistant TLE [5]. Many of these patients with antiGADab and drug resistant epilepsy also have adult onset T1DM. However, the significance of low titres of antiGADab in patients with T1DM and seizures or epilepsy remains unknown. The aims of the study are: to determine the prevalence and characteristics of patients suffering seizures and/or epilepsy among patients with adult onset T1DM, and to determine if there is any type of relationship between the presence and titres of anti GADab and seizures or epilepsy. 2. Material and methods 2.1. Study design Patients were recruited in the outpatient clinic of the Diabetes Unit of the Hospital Universitari de Bellvitge and in the Epilepsy Units of Hospital Universitari de Bellvitge and Hospital Clinic. Prevalence of suffering epilepsy and/or seizures was calculated in the group of adult onset T1DM patients seen consecutively in the Diabetes Unit. Due to the low prevalence of epilepsy in T1DM, to study the seizure and/or epilepsy features in this population of patients, we also used a
308
M. Falip et al. / Journal of the Neurological Sciences 346 (2014) 307–309
preexisting database containing patients (from the Bellvitge and the Clinic Hospitals) with seizures and positive antiGAD antibodies. 2.2. Definitions The following definitions were used to classify patients and seizures: - Hypoglycemic seizure Hypoglycemic seizure was a seizure temporally related to hypoglycemia (bthan 36 mg/dl) [6]. - Epilepsy Two or more unprovoked seizures (normal glucose level) separated by more than 24 h. 2.3. Immunological analysis A general immunological battery test was done, all antibodies being determined in EDTA plasma. AntiGADab was analyzed by immunohistochemistry and radioimmunoassay (RIA) in the Hospital Clinic of Barcelona, as described elsewhere [7], with values N 1 IU/ml considered positive. In patients with high titres of antiGADab (N 1000 IU), CSF antiGADab was also determined and intrathecal synthesis (IS) was calculated as described elsewhere [7].
hypoglycemic seizures were described as loss of awareness with oromandibular automatism without secondary generalization, often preceded by psychic or epigastric auras. EEG (routine) and MRI were normal in all patients with hypoglycemic seizures. EEG performed in the first 24 h after hypoglycemic seizures showed a continuous temporal slowing in 1/2 patients. Most of our patients suffering several hypoglycemic seizures 4/5 (80%) were erroneously diagnosed of epilepsy and were treated chronically with antiepileptic drugs. No changes in hypoglycemic seizure frequency or severity were observed; hypoglycemic seizures were only controlled after insulin treatment was adjusted. The usual treatment in these patients is a combination of long-acting insulin at night and rapid-acting insulin analog administered with main meals. After a severe hypoglycemic episode insulin doses are transiently reduced while the cause is evaluated and treated, particularly with regard to diet and exercise. Regarding patients with epilepsy, all of them had focal, temporal lobe epilepsy. Routine EEG showed interictal epileptiform discharges in one or both temporal lobes in 4/6 (66%) patients. Unprovoked seizures (normal glycemia) arising from one or both temporal lobe were recorded in 3/6 (50%) patients. MRI was normal in all patients with epilepsy, except one in whom a signal increase was observed in both hippocampi, without concomitant atrophy (see Table 1).
2.4. Other diagnostic tests Magnetic resonance imaging was performed in all patients with a 1.5-T unit (Philips Medical Systems, Best, The Netherlands). At least one routine EEG was performed in all patients (64 channel digital DeltaMed equipment). 2.5. Ethical approval The study was approved by the Ethical Committee of Hospital Universitari de Bellvitge. 2.6. Statistics Fisher's exact test and the Mann–Whitney U test were used for nominal and continuous data, respectively. 3. Results 3.1. Seizures and epilepsy prevalence
Table 1 Clinical and immunological characteristics of T1DM patients suffering either hypoglycemic seizures or TLE.
Gender Age (years) DM onset (years) Epilepsy onset (years) Age at first hypo Glycemia (mg/ml) Number of hypo. seizures Nocturnal seizures Aura GTCS CPS Interictal EEG EEG 24 h Ictal EEG
Among 229 consecutive patients with adult onset T1DM, whose onset of symptoms was between 2004 and 2012 and who were included in a diabetes database, 4 patients had suffered seizures. The seizures occurred in close temporal relationship with well documented hypoglycemic episodes in 2 patients. Two patients had epilepsy. Thus, the prevalence of epilepsy in our population was 2/229 (0.8%) and of hypoglycemic seizures of 2/229 (0.8%). From the two neurology databases 11 patients with seizure/epilepsy had positive antiGAD antibodies, 9 of them with T1DM. After excluding two patients who had been included in both the diabetes and neurology databases, we were left with a total of 11 patients with T1DM and seizures or epilepsy. Of these, 5 had only suffered hypoglycemic seizures. 4 had drug resistant TLE and 2 had well controlled TLE. In 4/6 (66%) patients with TLE + T1DM the onset of epilepsy was earlier than that of diabetes. On the other hand in all patients with hypoglycemic seizures the onset of diabetes was earlier. 3.2. Seizure and epilepsy characteristics Most hypoglycemic seizures were nocturnal, and seizure semiology was consistent with generalized tonic–clonic seizures. Diurnal
MRI Seizure free on AED Serum antiGAD (IU/ml), RIA CSF antiGAD (IU/ml), RIA AntiGAD IS AntiIA2 Antithyroid ab Antigliadin ab ANAS AntiDNA AntiCLP Other autoimmune disease
T1DM + hypoglycemic seizures
T1DM + TLE
N=5
N=6
4 males/1 females 36.6 (34–42) 23.4 (15–31) –
2 males/4 females 48.2 (33–65) 38.6 (27–55) 35.7 (22–51)
29.4 (23–36) 37.4 (34–42) 6 (2–15)
– – –
3/5 (60%) 2/5 (40%) 5/5 (100%) 5/5 (100%) 5/5 Normal 2/6 unilateral temporal lobe slowing Not done
0/6 (0%) 4/6 (66%) 1/6 (16%) 2/6 (33%) 2/6 normal –
5/5 Normal 0/5
Unilateral TLS 1/6 Bilateral TLS 2/6 5/6 normal 2/6
18.4
29.200
Not done
48.4 (N = 4)
– 5/5 negative 0/5 (0%) 0/5 (0%) 0/5 (0%) 0/5 (0%) 0/5 (0%) 0/5 (0%)
4/4 (100%) 5/6 negative 4/6 (66%) 3/6 (50%) 1/6 (16%) 1/6 (16%) 1/6 (16%) 4/6 (66%)
Significance
NS NS p: 0.053
p: 0.061 NS p: 0.061 NS NS
NS
p: 0.006
NS p: 0.015 NS NS NS p: 0.015
DM: diabetes mellitus, Hypo: hypoglycemia, EEG: electroencephalogram, MRI: magnetic resonance image, GTCS: generalized tonic–clonic seizure, CPS: complex partial seizure, IS: intrathecal synthesis, CSF: cerebrospinal fluid, RIA: radioimmunoprecipitation assay, AntiIA2: Antiinsulinoma-associated antigen-2 autoantibodies, AED: antiepileptic drug, EEG 24 h: electroencephalogram done during the first 24 h after a seizure, TLS: temporal lobe seizure, ANAS: antinuclear antibodies, antiCLP: anticardiolipin antibodies, NS: not significant.
M. Falip et al. / Journal of the Neurological Sciences 346 (2014) 307–309
3.3. AntiGADab titres and other autoimmune characteristics Mean serum antiGADab titre in patients with hypoglycemic seizures was 18.42 IU (2–32 IU) which was similar to the mean serum antiGADab titres of the general T1DM population (30 IU) (p: not significant). Mean serum antiGADab titre in patients with well controlled TLE was 43.442 IU and in drug resistant TLE was 22.079 IU. These titres were higher than in T1DM patients with hypoglycemic seizures (p: 0.006). Intrathecal synthesis of antiGADab was positive in 4/4 (100%) patients in whom it was determined. None of the patients with T1DM and hypoglycemic seizures suffered from other autoimmune diseases; in contrast, 4/6 (66%) patients with epilepsy and T1DM reported one or more other autoimmune diseases; celiac disease (3), myasthenia gravis (1), systemic lupus erythematosus (1), vitiligus (1), hypothyroidism (1) (p: 0.015). In addition, other autoantibodies were observed in this group, the most frequent being antithyroid antibodies, found in 4/6 (66%) patients (p: 0.015). No other autoantibodies were found in patients with hypoglycemic seizures. 4. Discussion In our cohort of consecutive adult T1DM patients with seizures and/ or epilepsy, hypoglycemic seizures were rare (0.87%), as was epilepsy (0.87%); this result does not support the hypothesis suggested by other studies that epilepsy is more frequent in T1DM patients [4], but is in agreement with the study by O'Connell et al. [8]. Although many hypoglycemic seizures are generalized tonic–clonic, some studies, in which hypoglycemic seizures had been recorded, suggest that they arise in the temporal lobe and have secondary generalization [9]. This may have happened in our patients, as suggested by seizure semiology and EEG performed in the first 24 h. The mechanisms underlying epileptic phenomena during hypoglycemia remain unknown. In one study using transcraneal magnetic stimulation it was observed that cortical excitability changed with fluctuations in blood glucose levels [10]. Moreover, hypoglycemia may induce acute symptomatic seizures and also predispose to the development of epileptogenic processes. It is generally accepted that recurrent episodes of hypoglycemia can result in epilepsy due to the possible permanent damage in the hippocampus [11]. Our results do not support this idea, as the subjects in our study suffered several episodes (in one case more than 10) of hypoglycemic seizures without becoming epileptic. Animal studies have suggested that the severity of hypoglycemia is directly related to hippocampal damage and not to seizure severity [12]. It is possible that repeated severe episodes of hypoglycemia may cause permanent damage to the hippocampus and in consequence, the potential to develop epileptogenesis. On the other hand, all our patients with hypoglycemic seizures were only diagnosed after suffering a generalized tonic–clonic seizure. We hypothesize that some of the autonomic symptoms of hypoglycemia may be simple partial seizures arising in the temporal lobe. A correct diagnosis of hypoglycemic seizures is very important in these patients, who may be treated with antiepileptic drugs without noticeable benefit, but exposed to deleterious side effects like the metabolic ones (weight gain and insulin resistance with valproate, for example). Previous studies have shown that antiGADab produce an imbalance between excitatory and inhibitory neurotransmitters [13]. However, not all patients with high antiGADab will be epileptic. In our series of 229 T1DM patients, 3 patients had high antiGADab titres without epilepsy, but all of them had other neurological diseases (Stiff-person syndrome, multiple sclerosis or ataxia). Interestingly the antiGADab titres of our patients with hypoglycemic seizures were lower than the levels observed in those with epilepsy + T1DM. None of our patients with very low antiGADab (b50 IU) had epilepsy and all of our drug resistant patients had high titres of antiGADab, although high titres may also be observed in patients with well controlled epilepsy. Mean antiGAD ab titres in patients with well controlled TLE was higher than in drug
309
resistant TLE (43.442 vs 22.079 IU) because one well controlled TLE patient had very high titres (86.300 IU) and probably, due to small sample size, false differences have been observed. Patients with high antiGADab titres, in addition to epilepsy tend to suffer from other autoimmune disorders and, had a higher number of other autoantibodies, a finding also reported by other authors [14]. Interestingly epilepsy tends to debut before T1DM. Intrathecal synthesis was observed in 4/4 patients (100%), indicating a clonal B cell activation in the CNS as observed in patients with other neurological diseases associated to antiGADab [15]. Our study has some limitations: the lack of ictal recording in patients suffering hypoglycemic seizures, the lack of capill.lar glycemia determination in the majority of seizures considered unprovoked, the small sample size, all patients had suffered adult onset T1DM and no long term follow up of the patients. Hence the conclusions must be taken with caution. In summary, our study suggests that hypoglycemic seizures are nowadays rare in T1DM patients. In these patients, antiGADab titres do not differ from the general diabetic population. Patients suffering several hypoglycemic seizures may not necessarily become epileptic. Patients with high titres of antiGADab are more likely to develop epilepsy, generally temporal and often drug resistant. They are also at risk of developing other autoimmune disorders with time. Conflict of interest Dr Falip, Dr Miró, Dr Carreño, Dr Jaraba, Dr Becerra, Dr Cayuela, and Dr Perez Maraver have no conflicts of interest. Dr Graus serves on the editorial board of Lancet Neurology and has received research support from the Fondo Investigaciones Sanitarias. We confirm that we have read the Journal position on issues involved in the ethical publication and affirm that this report is consistent with those guidelines. Acknowledgments The authors thank all the patients for their collaboration and Mr Dave McFarlane for the technical assistance. This study was supported in part by a grant from the Spanish National Institute of Health (PI10/ 00738). References [1] Varghese P, Gleason V, Sorokin R, et al. Hypoglycaemia in hospitalized patients treated with antihyperglycaemic agents. J Hosp Med 2007;4:234–40. [2] Katz ML, Volkening LK, Anderson BJ, et al. Contemporary rates of severe hypoglycaemia in youth with type 1 diabetes: variability by insulin regimen. Diabet Med 2012;29:926–32. [3] Auer RN. Hypoglycemic brain damage. Metab Brain Dis 2004;19:169–75. [4] Shober E, Otto KP, Dost A, et al. Association of epilepsy and type 1 diabetes mellitus in children and adolescents: is there an increased risk for diabetic ketoacidosis? J Pediatr 2012;160:662–6. [5] Falip M, Carreño M, Miró J, et al. Prevalence and immunological spectrum of temporal lobe epilepsy with glutamic acid decarboxylase antibodies. Eur J Neurol 2012;19: 817–33. [6] Beghi E, Carpio A, Forsgren L, et al. Recommendation for a definition of acute symptomatic seizure. Epilepsia 2010;51:671–5. [7] Saiz A, Arpa J, Sagasta A, et al. Autoantibodies to glutamic acid decarboxylase in three patients with cerebellar ataxia, late-onset insulin-dependent diabetes mellitus, and polyendocrine autoimmunity. Neurology 1997;49:1026–30. [8] O'Connell NM, Harvey A, Mackay M, et al. Does epilepsy occur more frequently in children with type 1 diabetes? J Paediatr Child Health 2008;44:586–9. [9] Lapenta L, Di Bonaventura C, Fattouch J, et al. Focal epileptic seizure induced by transient hypoglycaemia in insulin-treated diabetes. Epileptic Disord 2010;12:84–7. [10] Badawy RA, Vogrin SJ, Lai A, et al. Cortical excitability changes correlate with fluctuations in glucose levels in patients with epilepsy. Epilepsy Behav 2013;27:455–60. [11] Sherin A, Anu J, Peeyush KT, et al. Cholinergic and GABAergic receptor functional deficit in the hippocampus of insulin-induced hypoglycemic and streptozotozininduced diabetic rats. Neuroscience 2012;202:69–76. [12] Schauwecker PE. The effects of glycemic control on seizures and seizure-induced exitotoxic cell death. BMC Neurosci 2012;13:94. [13] Grimaldi G, Monto M. Brain H-MR spectroscopy in anti-GAD antibodies cerebellar ataxia. J Neurol 2010;37(5):303–5. [14] Liimatainen S, Peltola M, Sabater L, et al. Clinical significance of glutamic acid decarboxylase antibodies in patients with epilepsy. Epilepsia 2010;51:760–7. [15] Dalakas M, Li M, Fujjii M, et al. Stiff person syndrome. Quantification, specificity, and intrathecal synthesis of GAD65 antibodies. Neurology 2001;57:780–4.
