Adverse effects of antiepileptic drugs and quality of life in pediatric epilepsy
Correspondence Address: Source of Support: None, Conflict of Interest: None DOI: 10.4103/0028-3886.158203
Source of Support: None, Conflict of Interest: None
Background: In pediatric epilepsy, health-related quality of life (HRQOL) may be affected across the physical, psychological, social, and school domains. Studies have shown that antiepileptic drugs (AEDs) could have a significant negative impact on HRQOL, but these findings are scarce and inconsistent.
Keywords: Antiepileptic drugs; pediatric epilepsy; quality of life; side effects
The internationally accepted definition of epilepsy is that "it is a disorder of the brain characterized by an enduring predisposition to generate epileptic seizures and by the neurobiologic, cognitive, psychological, and social consequences of this condition."  The cornerstone of its successful management is the use of antiepileptic drugs (AEDs). However, the use of AEDs requires a continuous risk-benefit analysis that attempts to maximize seizure control, optimize cognitive, psychological, and social functioning, and minimize the adverse effects of AEDs. Alongside its significant beneficial effects, evidence also indicates that AEDs may have varied adverse effects that may predominantly affect the cognitive functions, emotional well-being, as well as the motor and somatic functions.  Depending on the population investigated or the methods used for analysis (e.g., unstructured interviews, screening, or spontaneous reporting), up to 90% of adults with epilepsy have reported some adverse effects of AEDs. ,,
There is a prevailing opinion that numerous developmental, psychological, behavioral, educational, and social disabilities among children and adolescents, who are being treated for epilepsy, could be partially explained by the adverse effects of AEDs. , This opinion garners particular credence when one considers that adverse events occur in up to 32% of the treated pediatric patients according to some studies.  The age of onset of epilepsy, the severity of seizures, the worry/concerns of the family and the psychological impact on them, as well as the cognitive impairment prevailing in children who are suffering from epilepsy, may all influence the quality of life in pediatric epilepsy. A recent review summarizing various studies related to this field has shown that AEDs may also have a negative impact on the child's well-being and daily functioning. 
Over the past two decades, the impact of pediatric epilepsy and its co-morbidities on the lives of children has been occasionally evaluated using patient-reported outcomes (PRO) such as the health-related quality of life (HRQOL) index.  These studies have provided important data regarding the day-to-day living and functioning status of these children suffering from epilepsy.  A systematic review of studies on pediatric epilepsy published in 2011 showed that pediatric epilepsy has a significant influence on the HRQOL issues, particularly across the physical, psychological, social, and school-related domains.  Several studies have also investigated the effects of AEDs on HRQOL together with other epilepsy-related clinical and socio-demographic factors. The prevailing opinion is conflicting, with some studies showing that the numbers of AEDs (monotherapy vs. polytherapy) or their side effects are not related to the HRQOL; ,,,, while, several other studies have demonstrated that polypharmacy ,,,, and adverse effects of AEDs and the symptoms of their neurotoxicity, could considerably affect the HRQOL domains. ,,, At least two well-designed studies demonstrated that there could be a two-fold negative effect of AEDs on the HRQOL scores; and, that the level of well-being and functioning may be predicted by the actual number of AEDs being given as well as by the AED-related adverse effects. ,
This inconsistency in data regarding the AEDs and their effects in impacting the HRQOL index may be influenced by the characteristics of the sample population included; it may, however, also be significantly altered by the methodology by which the HRQOL index and the adverse events related to the AEDs were assessed. In the previous studies, epilepsy-specific HRQOL questionnaires were used more frequently than the generic ones. The generated data, investigating the influence of AEDs on HRQOL, could be biased by the increased sensitivity of the epilepsy-specific questionnaires when compared to the general HRQOL ones.  In addition, most of the studies do not consider reports about the HRQOL together from the children as well as their parents. This is an important issue because studies suggest that parents/proxies and children disagree with each other while providing the HRQOL data, and that it is always necessary to include both these raters in the assessment. , Furthermore, the previous studies included mainly adolescents with epilepsy, while young children as well as those patients with intellectual disabilities were rarely included. Finally, a great majority of the studies have solely evaluated how the number of AEDs impacted the HRQOL scores and but did not take into account the influence of their adverse effects. This is a crucial issue, because adverse effects are frequently reported in pediatric epilepsy; and, the higher the number of AEDs taken, more is the likelihood that adverse effects would occur. 
The aim of this study was to evaluate the influence of the adverse effects of AEDs on the general (physical and psychosocial) domains of everyday functioning among children and adolescents with epilepsy using the rating provided by the pediatric patients themselves and their parents.
The data for this study was drawn from 75 children and adolescents with epilepsy (hereinafter being referred to as "children") and at least one parent who participated as a proxy-rater [Table 1]. These children were consecutive patients from the Department of Neurology, Pediatric Clinic, Clinic Center, Nis, Serbia. They were recruited from September 2010 to September 2011. All the participants were assessed during their regular clinic visits (after an informed consent was obtained prior to the study). The patients included in the study were of the age range 8-16 years, having an active epilepsy (a history of at least one seizure during the past year despite the constant use of AEDs), and the ability to read/comprehend Serbian. Pediatric patients with a progressive neuropsychiatric disorder, severe neurological impairment, or intellectual disability were not considered. The intellectual abilities were assessed prior to the initiation of the study. The children were categorized as those with normal (intelligence quotient [IQ] range 90-110) or borderline intellectual functioning (IQ range 71-89). The works described in the article has been carried out in accordance with The Code of Ethics of the World Medical Association (Declaration of Helsinki) for experiments involving humans and it was approved by the Ethics Committee of Pediatric Clinic, Clinic Center, Nis, Serbia.
The Pediatric Quality of Life Inventory (PedsQL) was used for self- and proxy-rating of the HRQOL index.  Children and their parents completed the Serbian version.  The PedsQL instrument has 24 items allocated into four scales: Physical (8 items), Emotional (5 items), Social (5 items), and School (5 items). Close inspection of the PedsQL revealed that five items of this scale were similar to items in the scale that was used to measure the adverse effects of the AEDs. These included the following: "having hurts and aches," "low energy," "trouble sleeping," "paying attention in class," and "forgetting things." Therefore, PedsQL's total and two scale scores were computed after omitting these items. The physical functioning score is computed by dividing the sum of the total items by the number of items answered in the physical scale (6 items). The psychosocial functioning score is calculated by computing the mean as the sum of the items divided by the number of items answered in the emotional, social, and school scales (12 items). Finally, the PedsQL total score was computed as the sum of all items divided by the number of items answered. Higher scores indicate better levels of functioning. In the present study, Cronbach's alpha coefficient for internal consistency reliability of the PedsQL scores ranged from 0.70-0.91.
The Adverse Event Profile (AEP) was used to assess the presence and severity of adverse effects. , The AEP is a 19-item instrument that assesses the presence and severity of the most frequent AED-related adverse effects during the previous 4 weeks. The severity of each symptom is rated on a 4-point response scale (1 = never a problem to 4 = always a problem). The sum of all items is the total scale score, where higher scores indicate more frequent and severe problems with AED-related adverse effects. Four of 19 items of the AEP can also be symptoms of depressive or anxiety disorders (i.e. "restlessness," "feelings of anger or aggression towards others," "nervousness and/or agitation," and "depression").  These items may raise the suspicion that higher AEP scores may actually be due to symptoms of depression/anxiety and not because of adverse effects. Therefore, these four items were omitted from the total AEP score (that had a possible range of 15-60). The AEP was determined by conducting an interview with a child and/or parent by a trained medical staff. In the present study, Cronbach's alpha coefficient for the internal consistency reliability of the AEP was 0.89.
The Hague Seizure Severity Scale (HSSS) was used to assess the seizure severity.  The HSSS instrument is a proxy-rated, 13-item scale, which enquires about the child's responses during and after the seizures. There is a 4-point response scale and the sum of all items is the total scale score, where higher scores indicate greater seizure severity (possible range: 13-53). In the present study, Cronbach's alpha coefficient for the internal consistency reliability of the HSSS was 0.90.
The mean (and standard deviation) values were calculated for the PedsQL scores [Table 2]. [Table 1] shows all the variables recorded that have a potential to influence the studied aspects of functioning. Univariate analysis was performed initially (Pearson correlation coefficient, t-test, or analysis of variance) and only the variables significantly related to the PedsQL scores (P < 0.05, two-tailed) were chosen as independent variables for modeling in the subsequent multivariate analysis. The data was analyzed by multivariate analysis in order to determine the unique main and interactive AED-related adverse effects. The dependent variables in the models were the three PedsQL scores. The effect size was estimated using partial ç2 and it was interpreted as small = 0.01; medium = 0.06; and, large = 0.14.  Separated models were tested for children and parents.
In the univariate analysis, besides the AEP score, the patient's IQ significantly correlated with the self-rated PedsQL scores, while the IQ, the HSSS score, and the seizure frequency were significantly correlated to the parent-rated PedsQL scores. There were statistically significant, inverse correlations between all PedsQL scores and the AEP score for both raters (r coefficient ranged 0.28-0.59, P < 0.02).
In the multivariate models of the children's ratings, the AEP score, the IQ, and the interaction between them accounted for a significant proportion of variation in all three PedsQL scores [Table 3]. However, the AEP score was the only factor that emerged as being significantly associated with the PedsQL total (P < 0.01; partial ç2 = 0.08) and psychosocial functioning scores (P < 0.02; partial ç2 = 0.07).
In the multivariate models of the parents' ratings, the AEP score, the IQ, the HSSS score, the seizure frequency, and the interaction among them accounted for a significant proportion of variation in all three PedsQL scores [Table 4]. The AEP score was the only factor that emerged as being significantly associated with the PedsQL total score (P < 0.001; partial ç2 = 0.26) and of the physical functioning score (P < 0.02; partial ç2 = 0.07). In the model with the psychosocial functioning, the IQ and the HSSS score accounted for an additional small proportion of variation (P = 0.05; partial ç2 = 0.05), while the AEP score accounted for a high proportion of variation (P < 0.001, partial ç2 = 0.30).
The present study evaluated the relationship between frequency and severity of the adverse effects of AEDs and the HRQOL among children with epilepsy. Therefore, it was different from the majority of the previously published studies that predominantly evaluated the relationship between the number of AEDs taken, the presence of other risk factors, and the HRQOL. Our findings showed that the frequency and severity of adverse effects could substantially affect the general levels of HRQOL.
Parents, as proxy-raters, reported a moderate impact of the present AED-related adverse effects on the levels of physical functioning of their children. The children on their own did not perceive any role of the AED-related adverse effects in restricting their physical activities such as walking, bathing, or running. However, both raters reported a large impact of the AED-related adverse effects upon the levels of psychosocial functioning in children, including the important aspects of emotional, social, and school functioning. The finding that the levels of psychosocial functioning were affected much more than the levels of physical functioning may be explained by the fact that AEDs affect higher cognitive functions and emotions responsible for psychosocial functioning to a much greater extent than motor functioning, such as mobility of the patient. 
Several studies, that mainly recruited adolescents with epilepsy, have demonstrated that adverse events of AEDs significantly influence the well-being and functioning of the patients across various epilepsy specific domains, such as the social and emotional impact of epilepsy, the present and future worries, the need for concealment, or the quest for normality. ,,,, Modi and her coworkers showed that the side-effects of AED were associated with changes in the HRQOL in children with newly diagnosed epilepsy.  Extending the previous findings, our data shows that AED-related adverse effects substantially impact the overall functioning in children and affect the general aspects of the HRQOL in the physical and psychosocial domains. This is an important consideration for clinical practice in deciding appropriate actions to reduce the adverse effects of AEDs like lowering the dosage, changing the dosage schedule, or changing its type. Thus, it may be possible to develop specific interventions for reducing the burden of these adverse events while targeting specific domains like the social and school functioning. The latter are mainly related to the adverse events of the AEDs being administered and are not necessarily related to the problems specifically arising due to the prevalence of epilepsy in the patient (such as memory problems or disturbed sleep). For example, it might be possible to develop structured physical activity programs to improve the physical functioning; to specifically educate the child and the parents regarding the adverse effects of the AEDs and the appropriate ways to combat them; or, to develop specific programs to enhance learning and training if the deleterious side-effects of AED are responsible for lowering the psycho-social achievements of the child. Considering the importance of schooling in a pediatric patient suffering from epilepsy, it would be important to develop interventions for reducing the burden of adverse effects of AEDs as this would have a significant psychosocial impact on the child's condition.
This study has major strengths that include considering both children and parents as raters, including different epilepsy types, including children with borderline intellectual functioning, and studying various variables responsible for influencing the pediatric quality of life issues. The results, however, should be interpreted with caution. Firstly, the number of subjects is not large enough to obtain a significant impact in the specified models. The cross-sectional study design does not lend itself to studying the temporal relationship between the adverse events of AEDs and its impact on the pediatric population over a period of time. It also does not take into account the natural history of the etiology of seizures while studying the impact of the adverse effects of AEDs. These limitations could only be overcome by conduction of the study in a longitudinal fashion over a period of time.  Additionally, children and adolescents with lower intellectual functioning were not evaluated, a fact that could bias the results and limit the extrapolation of the results of this study to that of the general pediatric population (that may often include children with impaired intellectual faculty who are having seizures).
Evaluating the impact of AED-related adverse effects on the HRQOL scale is extremely challenging. In the first place, all symptoms covered by a questionnaire are not necessarily due to the adverse effects of AEDs (such as a poor concentration or memory problems), but may in fact be genuine neuropsychological or neuropsychiatric problems prevalent due to the presence of epilepsy.  In addition, other factors could also be relevant in influencing the quality of life. It is worth mentioning that of all the variables considered in the regression models in our study, only the IQ and seizure severity (the latter was rated by the subject using the HSSS) were found to be responsible for the proportion of the variance in the HRQOL indices. However, other variables have been previously reported to be significant risk factors affecting the HRQOL scale.  The type of medication administered, the epilepsy control over time, the severity of the seizure, and the underlying brain pathology, could all be important factors to consider in models studying the association between the adverse effects of AEDs and the HRQOL. Studies with a longer follow-up conducted on patients with newly diagnosed pediatric epilepsy (such as the one conducted Modi et al.),  studies conducted on children on a changing AED regimen or on children with active epilepsy who were medicated for prolonged periods, are needed.
In conclusion, a significant relationship was found between the frequency and severity of AED-related adverse effects and the levels of HRQOL in children with epilepsy. In particular, it was discovered that AEDs placed a huge impact on the psychosocial functioning of children. As additional data becomes available from follow-up studies, it would be possible to develop specific interventions for reducing the burden of AEDs and their adverse events by targeting specific domains.
We would like to thank to Eric Ganz for helping us with the English language editing.
[Table 1], [Table 2], [Table 3], [Table 4]