The individual risk factors and their relation to the seizure recurrence after the first unprovoked epileptic seizure in our group of 72 patients.
An epileptic seizure may be conceptualized as a paroxysmal pathological process in the brain of a heterogeneous etiology with heteromorphic clinical and electrophysiological manifestation.The cases of epileptic seizures are classified according to The International Classification of Epileptic Seizures (ICES) published for the first time by The International League Against Epilepsy (ILAE) in 1970 and revised in 1981 (Commission on Classification and Terminology of the International League Against Epilepsy, 1981). This classification is a clinical one related to semiology of the seizures not to their etiology. Therefore it is necessary to exclude an acutely occurring cause responsible for occurrence of the seizure. In such cases we talk about the so-called acute symptomatic seizures. The underlying cause may be structural (e.g. head trauma), metabolic, toxic (e.g. alcohol), or an acute CNS infection, etc.. The most frequent acute symptomatic seizures are the febrile seizures. In fact, the acute symptomatic seizures occur more frequently than epilepsy (“unprovoked” seizures). The risk of occurrence during one’s life is very high - approximately 5% in males and 2.5% in females. If the acutely occurring cause has been withdrawn or cured without a residuum in the form of a brain lesion, the seizures do not recur (Dasheiff, 1987; Fromm, 1987). The antiepileptic medication is necessary for suppressing the seizures in the acute stage but usually there is no need for treatment continuation after the complete cure of the underlying disease. If the acutely occurring cause was not responsible for epileptic seizure we talk about a so-called unprovoked seizure. If the unprovoked epileptic seizure occurs in relation to a preceding neurological insult, the disorders is regarded as secondary to this insult; we call it the late symptomatic epileptic seizure or late symptomatic epilepsy in case of seizure recurrence. A general principle of treatment for the symptomatic (secondary) epileptic seizures has been a primary effort for resolution of the underlying disease that is the etiological factor responsible for the seizures. Given that it is impossible, the antiepileptic treatment in accordance with the treatment guidelines for individual seizure types (together with adherence to right living, behavioural precautions and concomitant solutions of the social and psychological issues) is indicated (Hovorka et al. 2004a; Hovorka et al. 2004b; Ošlejšková, 2007). Approximately 5% of the population experiences one unprovoked epileptic seizure in the lifetime (Forsgren et al., 1996; Hauser et al. 1993). The febrile seizure before the age of 5 occurs in approximately 5% of population (Hauser et al. 1996). Only about 25% of people experiencing the first unprovoked seizure see the doctor and nearly always the seizure is a generalized tonic-clonic one. Most of the people have no risk factors for the onset of epilepsy, normal neurological examination as well as normal initial EEG (Pedley et al., 1995). The occurrence of the first unprovoked epileptic seizure requires always a thorough evaluation. The risk of misdiagnosis is high as non-epileptic seizures make 20-33% of newly-diagnosed cases.
EEG is an important non-invasive examination method that informs about electrical activity of the brain. It plays an important role in differential diagnostics of seizures. The greatest diagnostic benefit of EEG belongs to diagnosis of epilepsy. Finding of interictal epileptiform graphoelements supports the diagnosis of epilepsy with specificity of 96% (Vojtěch, 2008). In patients who experienced the first unprovoked epileptic seizure 30-40% catchment of specific epileptiform EEG abnormalities after the first EEG examination was most often reported in the literature (King et. al., 1998
Imaging examinations represent one of the basic methods in diagnostics of patients with epileptic seizures. Their development has significantly contributed to accurate diagnostics and classification of epileptic syndromes. It is necessary to realize that these methods can help reveal etiology of seizures and determine etiopathogenetic diagnosis. The fact that each patient after first epileptic seizure must undertake these examination is common and generally accepted. MRI is an advantageous imaging method for CNS. At present this method is the first choice method. MRI can reveal structural lesions and brain anomalies which CT examination, that is less sensitive, cannot (heterotopias, demyelinizations, anomalies of gyrifications, vascular malformations, etc.) (Bořuta et al., 2007). MRI is markedly more advantageous in patients with temporal epilepsy where it is able to express even very tiny structural changes and mesial temporal sclerosis (Carrilho et al., 1994). In addition, in MRI examination patients are not exposed, in contrast to CT, to radiation load. CT advantages involve better availability, relatively low price, possibility to examine non-cooperating patients because the examination takes only several seconds and it is less sensitive to movable artifacts. Moreover, CT has less contraindications comparing to MRI. MRI cannot be realized in patients with metal implants and clips, pacemaker, uncontrollable claustrophobia. Result can be adulterated if the patient does not cooperate.
In the past we also evaluated the findings of various modification of EEG examination and imaging methods in our patients who experienced solitary unprovoked epileptic seizure (Kollar et. al., 2009). We found that catchment of epileptiform manifestations in native EEG in patients who experienced solitary unprovoked epileptic seizure (14.29%) is lower than reported in literature (King et al., 1998; Shinnar et al., 1994; Vojtěch, 2008). It might be explained by accepted fact of transient incidence of abnormalities in EEG records. That is the reason that transient incidence of epileptiform EEG abnormalities in patients with epilepsy is considered the factor participating on different results of particular studies. High percentage of non-specific (non-epileptiform) abnormal EEG records in our cohort of patients who experienced solitary epileptic seizure was in agreement with literature data (Kollár et al., 2009). The results of our study – see Table 1.
It is very important to realize the limits of EEG examination. Firstly - normal EEG finding does not rule out clinical diagnosis of epilepsy and presence of epileptiform EEG abnormality does not confirm that the patient has epilepsy. Recurrent occurrence of abnormal interictal EEG findings in the group of non-epileptic seizures is also known (Kuba et al., 2001). That´s why EEG must be recognized as the method that plays a very important role in diagnostics of epilepsy or paroxysmal disorders, however, as the adjuvant examination method its role is limited. In clinically clear epileptic manifestations EEG can confirm, or in specific cases support, clinically clear diagnosis of epilepsy. In clinically absent typical epileptic manifestations high cautiousness is needed in evaluation of the diagnosis (incorrect evaluation or over-evaluation of EEG finding).
Statistical comparing of diagnostic benefits of CT and MRI examinations of the brain in our group of patients confirmed, as in the other works, that MRI examination of the brain in patients who experienced solitary unprovoked epileptic seizure is definitely the first choice method. The results of our evaluation – see Table 2,3.
Statistical evaluation of benefits of MRI examination comparing to CT examination by binomic test of proportions: p<0.0001 – high significant difference between proportions (Kollar et. al., 2009).
In the past we noticed, by the EEG finding evaluation in our group of outpatients with epilepsy, that only in a small amount of cases the EEG findings corresponded completely with the clinical image of epileptic seizure (Kollar et al., 2010). In relation to the imaging methods and their diagnostic agreement with clinical syptomatology of epileptic seizure and EEG findings, it is interesting that in the study of King et al. (1998), who evaluated the imaging of MRI abnormalities after the first epileptic seizure in 300 members of a group, consisting of both children and adults, they determined that, with patients having clinically diagnosed partial seizure, an epileptogenic lesion on the MRI was identified in 17% of cases. In 50 patients with clinical diagnosis of generalized seizure a structural lesion was identified in only one case and in the case of 49 patients with generalized epileptiform activity on EEG no structural lesion on MRI was identified. These facts covey to us the need to try and establish in our own group of patients the clinical typology of epileptic seizure, EEG findings and results of imaging methods. After this we determined the part of patients with complete diagnostic concordance between clinical image of epileptic seizure and results of auxiliary diagnostic methods (Kollar et al., 2010). The summary of all watched data in the group of patients after solitary epileptic seizure - see Table 4. The evaluation of clinical typology of epileptic seizures and results or realized examination - see Table 5.
The full diagnostic coincidence between the clinical picture of epileptic seizure, EEG examination (native interictal EEG, or EEG after SD or LTM-EEG after SD) and results of imaging methods (CT or MRI of the brain) we found only in 11 from 84 patients (13,1%) after solitary epileptic seizure. The receiving diagnosis of unclear seizure status was determined in 57 out of 116 patients (49,14%) dismissed, as mentioned in 10 years’ time period, with the diagnosis of solitary epileptic seizure. These percentages, together with a high part of unclear receiving diagnosis (the disturbance of consciousness of unclear etiology) in the patients, who were dismissed from our clinic with diagnosis of solitary epileptic seizure, suggests that the diagnosis of this group of patients is often problematic.
From unclear seizure status, which is accepted on the neurological departments, the more considerable part is made by unepileptic seizure status (Angus-Leppan, 2008; Perrig & Jallon, 2008). The correct diagnosis of seizure disorders require the strict observance of standard diagnostic proceeding (Martiniskova et al., 2009). The necessity are detailed anamnesis, adequate “erudition” of medical doctors working in this part of medicine, the right interpretation of auxiliary diagnostic methods results and in many cases the quality of cooperation between the doctors from other specializations (Bajaček et al., 2010; Hovorka et al., 2007; Kollar et al., 2010). Our results repeat the confirmation that diagnostic of seizure disorders with or without the disturbance of consciousness belong between the more difficult performances in the clinical praxis.
2. Solitary unprovoked epileptic seizure – the risk factors of probable seizure recurrence. To treat or not to treat the patient after the first unprovoked epileptic seizure?
At least 5% of the general population experience one unprovoked epileptic seizure during their life (Forsgren et al., 1996; Hauser et al., 1982; Hauser et al., 1993). This is in contrast with an approximately 3-4% cumulative incidence of epilepsy (at least two unprovoked epileptic seizures) and with an approximately 4% incidence of the acute symptomatic seizures (Hauser et al., 1996). The risk of seizure recurrence after the first epileptic seizure has been shown to be most frequently 30-40% (range 23-71%) (Annegers et al., 1986; Engel & Starkman, 1994; Hauser et al., 1990; Kollar et al., 2006; Mann, 2005). This figure oscillates notably depending on certain risk factors. There are multiple risk factors mentioned in the literature. Berg & Shinnar (1991), referring to already published studies and meta-analyses, stated that multiple factors influence the recurrence risk of epileptic seizures - see Table 6.
2.2. Material and methods
We evaluated 116 patients (68 men, 48 women; age range 18-81 years) after a solitary epileptic seizure that had been hospitalized at our neurological department since January 1, 1997 to January 1, 2007. There were 84 patients having experienced an unprovoked seizure and 32 with an acute symptomatic epileptic seizure. The baseline information was obtained using a retrospective analysis of the medical records; eligible patients were contacted by telephone or by sending the questionnaire via the post. We evaluated the likelihood of seizure recurrence in 72 patients. Duration of the follow-up was 2 - 12 years. A certain portion of patients were followed prospectively at our outpatient department, others were monitored by their neurologists. We determined the number of patients in whom the seizure reoccurred, and time period between the first and the second epileptic seizures. We evaluated the following recurrence risk factors: incidence of the febrile convulsions; incidence of epilepsy in patients’ relatives; period of the day, when the seizure appeared; objective neurological examination; the clinical type of seizure; EEG findings; aetiology of the seizure and the influence of antiepileptic treatment initiation after the first seizure. The logistic regression was used for the statistical assessment of the data obtained.
The individual risk factors and their relation to the seizure recurrence after the first unprovoked epileptic seizure are shown in the Table 7.
The summary of the logistic regression:
Patients with the partial epileptic seizure had 5 times higher recurrence risk (OR = 5.12, 95 % CI: 0.79 – 32.89).
after the first
|Patients with seizure recurrence/|
patients with seizure recurrence + without
|Statistically significant factor in terms of the seizure recurrence?|
|Etiology||idiopathic + cryptogenic||12||18||12/30|
|Period of the day, when|
the seizure appears
|Clinical type of the|
|EEG findings||normal + non-epileptic abnormality||20||36||20/56|
of the febrile convulsions
|Antiepileptic treatment initiation after the first unprovoked|
|Seizure recurrence after the first epileptic seizure within …||3 months|
Patients with epileptiform EEG findings had 5 times higher recurrence risk (OR = 5.84, 95 % CI: 0.98 -34.62).
The antiepileptic treatment initiation after the first seizure seems to be the only statistically significant protective factor as the patients in our group had 7 times lower recurrence risk compared to the patients without medication (OR= 0.13; 95% CI: 0.03 – 0,6).
After the first unprovoked epileptic seizure we recorded the seizure recurrence in 26/72 patients (36.1%); in 24/26 patients (92.3%) the seizure recurred within 12 months after the first unprovoked seizure.
The antiepileptic treatment initiation after the first unprovoked seizure was the only significant factor decreasing the risk of seizure recurrence (7 times lower in our study) (OR = 0.13, 95% CI: 0.03-0.6) in our group of 72 patients. The influence of antiepileptic treatment initiation on the reduction of seizure recurrence has also been reported in earlier studies (Elwes et al., 1985; Kollar et al., 2006). On the contrary, the significance of this factor was not shown in other studies (Bora et al., 1995; Hopkins et al., 1988; Musicco et al., 1997). In our patients there were risk factors of the seizure recurrence showing a clinical, but not statistical, significance – the type of epileptic seizure (5 times higher recurrence risk in patients after the first partial epileptic seizure: OR = 5.12; 95%, CI: 0.79 – 32.89) and EEG findings (5 times higher recurrence risk in patients with epileptiform EEG findings: OR = 5.84; 95% CI: 0.98 – 34.62). In consistence with the conclusions of Berg & Shinnar (1991) we observed a decrease of the recurrence risk of seizure with time that elapsed since the first seizure. The recurrence risk after the first unprovoked seizure was the highest within 12 months (24/26 patients, 92.3%). The differences in results of the studies evaluating the risk factors of seizure recurrence after the first unprovoked epileptic seizure may be attributed to unequal methods and baseline criteria, as well as to diverse durations of the follow-up. The meta-analyses performed help us orient ourselves in this area (Berg & Shinnar, 1991). In case of the first unprovoked epileptic seizure appearance there is a vital need for a thorough evaluation (Kollar et.al., 2009; Martiniskova et al., 2009).
The antiepileptic treatment initiation in patients after the solitary unprovoked epileptic seizure was the only statistically significant factor decreasing the risk of seizure recurrence in our group of patients. Based on the recent knowledge and despite of this finding we propose an individual, rather than automatic, antiepileptic treatment initiation, considering all risks, likelihood of seizure recurrence, social and psychological factors, employment and the potential side effects of the treatment.
This work is partially supported by the Slovak Science Grant Agency (VEGA No 1/0755/09, VEGA No 1/4266/07).