The Prevalence of Drug-Resistant-epilepsy and its Associated Factors in Patients with Epilepsy

Background: Drug-resistant-epilepsy (DRE) is a major challenge in adult patients with epilepsy. The majority of previous studies evaluating the risk factors of DRE have been conducted in children. Therefore, this study aimed to investigate the prevalence of DRE and its associated factors in adults. Methods: All patients aged over 12 years with an established diagnosis of epilepsy since at least one year before the admission who were admitted with seizure to the neurology ward of a tertiary care hospital were consecutively included from 20th March 2014 to 19th March 2020. Patients were classi�ed into two groups of DRE and non-DRE groups. The archived-les of the patients were retrospectively reviewed and the data were extracted and recorded in a pre-prepared checklist. Results: A total of 436 patients were investigated. The most common causes of epilepsy were idiopathic (53.9%), vascular (22.5%), and CP/developmental disorders (8.3%). There was no signi�cant difference between DRE and non-DRE patients in terms of age, sex, seizure type (generalized/partial). Vascular causes were more prevalent in the non-DRE group, and idiopathic, post-traumatic/surgery, MS/degenerative, poor compliance, CP/developmental disorders, and space-occupying lesions were more prevalent in the DRE group. In multivariate regression analysis only the presence of CP/developmental disorders was independently associated with a higher probability of DRE (adjusted OR= 2.170, 95% CI = 1.017 to 4.633, p = 0.045). Conclusion: The prevalence of DRE is still considerably high. Therefore, considering its serious consequences, more investigations should be carried out to determine proper strategies for reducing its incidence. We found the history of CP/developmental disorders to be independently associated with DRE.


Introduction
Epilepsy affects 65 million people worldwide, almost 80% of whom live in developing countries [1,2].
Meta-analysis studies on the incidence of epilepsy indicate that the incidence in developing countries and developed countries is on average 67.8 / 100,000 and 43.4 / 100,000, respectively [3][4][5].Recurrent seizures can bring about a variety of physical, psychological, and social morbidities [6].On the other hand, by complete control of the seizures, all these consequences are preventable which can be achieved by an optimal dose of antiepileptic medications [7].However, a considerable proportion of patients with epilepsy suffer from drug-resistant-epilepsy (DRE) which affects 30% to 40% of adult epileptic patients [8][9][10].
A variety of de nitions has been proposed for DRE in different periods which varied in different aspects including the number of failed antiepileptic drugs, seizure frequency, and temporal criteria (time duration for the occurrence of the determined number of seizures) [11].Therefore, the International League Against Epilepsy (ILAE) in a broad consensus de ned DRE as "failure of adequate trials of at least two tolerated, appropriately chosen and used antiepileptic drug schedules (whether as monotherapies or in combination) to achieve sustained seizure freedom" [12].
The prediction of the future development of DRE is still a great challenge in the setting of clinical practice.Previous studies have postulated some possible risk factors for developing DRE including younger onset age, symptomatic etiology, abnormal electroencephalography (EEG) ndings, and neurological impairment or mental retardation at the time of diagnosis, high-frequency seizures [11].However, the majority of these studies have been conducted in children [13][14][15][16].Therefore, this study aimed to investigate the prevalence of DRE and its associated factors in adults with epilepsy.

Methods
In this case-control study, all patients aged over 12 years with an established diagnosis of epilepsy since at least one year before the admission who were admitted with seizure to the neurology ward of a tertiary care hospital were consecutively included from 20th March 2014 to 19th March 2020.Patients admitted with seizure due to secondary causes such as medication (three-cyclic antidepressants, tramadol, etc.), metabolic disorders, electrolytes disturbances, and those with pseudo-seizures were excluded from our study.Patients were classi ed into two groups of DRE and non-DRE groups based on the de nition of DRE proposed by the ILAE consensus mentioned above [12].Accordingly, the patients were classi ed as DRE group if they had seizures despite using at least two well-selected antiepileptic medications with appropriate doses.The non-DRE group were consisted of those who did not meet the de nition of DRE but had seizures despite using one antiepileptic medication and were seizure-free for at least 3 months before their admission.The archived-les of the patients were retrospectively reviewed and the following data were extracted and recorded in a pre-prepared checklist.Demographic data including age and sex, history of diabetes, cerebrovascular accidents (CVA), and hypertension (HTN), seizure type, and cause.
EEG of the patients was also retrospectively analyzed by an attending neurologist who was blind to the clinical data of patients and classi ed into two categories of normal/abnormal EEGs.Classi cation of the seizure type into generalized and partial and determination of seizure cause was conducted based on the history, physical examination, and EEG ndings.
The study protocol was approved by the medical ethics committee of the (hided for peer review).
Statistical analyses were conducted by using IBM SPSS Statistics version 22 (SPSS Inc., Chicago, Illinois).The normal distribution of the variables was tested using the Kolmogorov-Smirnov test.Age was compared between groups using the independent t-test.Categorical data were compared using the Pearson chi-square test.Multivariable logistic regression was used to detect independently associated factors with the development of DRE.A p-value of less than 0.05 was considered statistically signi cant.

Discussion
Fair and poor seizure control are associated with a signi cantly lower quality of life and a higher prevalence of depression, anxiety, and stress in patients with epilepsy [17].Also, DRE is reported to have serious physical and emotional consequences and it poses a great burden on the family as well as on social, educational, and health services [18].Furthermore, these patients are at higher risk of some potentially life-threatening events, arising as a result of uncontrolled seizures including aspiration, electrolyte imbalance, brain edema, refractory status epilepticus, renal failure, cardiac arrhythmias, and unexplained sudden death [15].It should be noted that the prevalence of DRE is considerably high.In the current study, 22% of adults with epilepsy admitted during six years had DRE.According to a systematic review and meta-analysis on 19 studies, the pooled incidence of DRE in epilepsy patients was 20% (95% CI 14-27%).Though there was a great heterogeneity among the included studies in this review (ranging from a prevalence of 6-51%) [11].Another systematic review and meta-analysis on 15 studies estimated that prevalence of DRE is 25% (95% CI 17-32%) [19].The heterogeneity among studies is partially due to the different DRE de nitions.In the current study, we applied the most updated de nition of DRE which was established by the consensus of ILAE [12].In the systematic review by Kalilani et al. of 15 included studies, only 4 studies (10.5%) used a case de nition of DRE that complied with the ILAE de nition of DRE [19].Some investigators de ned DRE with higher the number of failed antiepileptic medications [15,[20][21][22][23].While other studies only required failure of one antiepileptic medication [24][25][26].
Furthermore, the discrepancy among different studies is attributable to the differences in the genetic/ethnicity predisposition [27].As far as we investigated, our study is the rst study on the Iranian population, evaluating the prevalence and associated factors of DRE in adults.
In the current study in a relatively large number of epileptic patients, we identi ed a positive history of abnormalities of these diseases.However, due to the intellectual disabilities of these patients, there is a possibility of a lack of adequate adherence to the prescribed antiepileptic medications.Therefore, rst, perinatal care should receive more attention to reduce the possibility of developing these anomalies.Second, these patients should be encouraged and helped out to use their medications accurately.
Some studies postulated that the rate of partial epilepsy is signi cantly higher in DRE patients [26, 30,31].However, we did not nd any signi cant association between seizure type and DRE.Similarly, Kalilani et al. by a meta-analysis of 15 studies data, failed to detect any signi cant relationship between seizure type and DRE [19].However, it should be noted that the number of patients with partial seizure was relatively low in our study.This is because we conducted this study on the admitted patients and patients with partial seizures are rarely get admitted to the wards.Therefore, the results of our study on the relationship between seizure type and DRE may hardly generalizable for the whole population.Thus, it is warranted that this issue gets investigated in future population-based studies.
As like other retrospective studies, the main limitation of our study was the lack of recorded data ascertainment, which meant that the data on the cause of epilepsy could not be ascertained.Nonetheless, the patients were admitted to a teaching hospital and were under observation by attending neurologists.Moreover, the number of used antiepileptic medications and characteristics of the seizures was recorded based on the patient's/components' claims.

Conclusion
The of DRE is still considerably high.Therefore, considering its serious consequences, more investigations should be carried out to determine proper strategies for reducing its incidence.We found the history of CP/developmental disorders to be independently associated with DRE.Therefore, regarding perinatal care to reduce the incidence of CP/developmental disorders and encouraging them to use their medications accurately can be helpful.

Figures
Figure 1 Age distribution in patients admitted with epilepsy in two groups of drug-resistant epilepsy (DRE) and non-DRE The proportion of drug-resistant epilepsy (DRE) in different causes of seizure

Table 1
the DRE group and 58.4 percent of non-DRE group had impaired EEG (p-value = 0.001).The two groups of patients had a signi cant difference in the type of seizure causes.Vascular causes were more prevalent in the non-DRE group and idiopathic, post-traumatic/surgery, MS/degenerative, poor compliance, CP/developmental disorders, and space-occupying lesions were more prevalent in the DRE group (Figure MS: Multiple Sclerosis, CP: Cerebral palsy, CVA: Cerebrovascular accidentThere was no signi cant difference between DRE and non-DRE patients in terms of age, sex, seizure type (generalized/partial), and positive history of DM and CVA (Table1).However, patients in the non-DRE group more frequently had HTN than the DRE group (13.3% vs 5.2% p-value = 0.027).76.3 percent of the patients in

Table 2
MS: Multiple Sclerosis, CP: Cerebral palsy, CVA: Cerebrovascular accident CP/d29]lopmental disorders as an independently associated factor with DRE.However, age and sex were not associated with DRE.Likewise, Farghaly et al. by evaluation of 437 patients with epilepsy also reported that age and sex were not associated with DRE but mental retardation signi cantly increased the risk of DRE[26].These results were in concordance with the results of Chawla et al. and Eriksson and Koivicco, who reported that perinatal problems were the leading cause of DRE[28,29].The association of CP/developmental disorders with DRE may be attributed to the pathophysiology and brain structural