Abstract
Objectives: Migraine and epilepsy are 2 of the most common neurologic disorders in children. In this cross-sectional study we investigated a population of children with epilepsy to determine if children with a greater seizure burden or certain epilepsy syndromes had a higher risk of migraines. We also examined how often migraine is addressed and treated in a pediatric epilepsy cohort.
Methods: Between January 2010 and March 2011 we distributed questionnaires regarding headache symptoms and treatment to consecutive children with epilepsy seen in clinic at Johns Hopkins Hospital (400 children were studied). Records were subsequently reviewed for seizure type, age at onset, and treatment.
Results: The prevalence of migraine in our pediatric epilepsy population was 25%, which is greater than reported for children without epilepsy (3%–23%). Migraine was more prevalent in children ≥10 years (p = 0.0009), children with benign epilepsy with centrotemporal spikes (BECTS) (p = 0.003), and children with juvenile myoclonic epilepsy (JME) (p = 0.008). Migraine onset was more likely to have occurred after epilepsy was diagnosed (p = 0.0002), but was not more prevalent in those with intractable epilepsy. Only 50% of patients with weekly or greater migraines had documented discussions regarding headaches with their neurologist.
Conclusion: Migraine was comorbid in one-quarter of children with epilepsy in a tertiary care center. Children who were older or who had BECTS or JME were more likely to have migraines. Migraines were infrequently addressed within the neurology clinic. It is imperative to address comorbid migraine in treating children with epilepsy.
GLOSSARY
- BECTS=
- benign epilepsy with centrotemporal spikes;
- CSD=
- cortical spreading depression;
- ICHD-2=
- International Classification of Headache Disorders–2;
- JME=
- juvenile myoclonic epilepsy.
Migraine and epilepsy are 2 of the most common neurologic disorders seen in children. The prevalence of migraine in children is as high as 23% during adolescence.1 By contrast, 0.5%–1% of children have epilepsy.2 The latter represents a large proportion of children seeking care from a child neurologist.3 The co-occurrence of migraine and epilepsy has long been recognized in adults.4,–,6 In children, few studies have investigated this comorbidity and how it is addressed in an epilepsy clinic. Prior work includes only a few study subjects with varied criteria for defining migraine.7,8 Because of the large number of children with epilepsy in our clinics who anecdotally complain of frequent headaches, we sought to examine the prevalence of migraine in a population of children with an established diagnosis of epilepsy and how this was addressed by the provider. Children with more frequent and uncontrolled seizures seemed more likely to complain of headache. We generated several hypotheses, including that the prevalence of migraine in children with epilepsy would be greater than in historical controls without epilepsy, in those with greater seizure burden, and those with certain epilepsy syndromes such as benign epilepsy with centrotemporal spikes (BECTS) and juvenile myoclonic epilepsy (JME).9,10 Subsequently, we examined how migraines were addressed and treated by providers seeing the children for primarily a chief complaint of epilepsy.
METHODS
Participants.
This is a cross-sectional study of 400 children who were evaluated between January 2010 and March 2011 in either the pediatric neurology resident continuity clinic or pediatric epilepsy clinic at Johns Hopkins Hospital in Baltimore, Maryland. In resident continuity clinic all patients were seen by 1 of 8 pediatric neurology residents along with an attending child neurologist. Epilepsy clinic patients were seen by 1 of 4 pediatric epileptologists or a pediatric neurology nurse practitioner. New and return patients were included. Questionnaires were distributed by the clinic front desk or by the provider in attempts to include every child who came for an appointment for epilepsy during the period of the study.
Inclusion criteria, collection of data.
Parents of children (or the children themselves, if old enough, in the presence of their parents) ages 3 to 17 years who had a history of at least 2 afebrile seizures were asked to complete a brief questionnaire (appendix e-1 on the Neurology® Web site at www.neurology.org). Patients were excluded if they did not speak English or did not complete the questionnaire. The child's electronic patient record, up to and including the clinic visit during which the questionnaire was completed, was subsequently reviewed for pertinent history regarding the patient's seizure type, age at onset of first seizure, intractability, developmental stage, and treatment of both epilepsy and migraine.
Definitions.
The diagnosis of migraine was made based on the International Classification of Headache disorders, second edition (ICHD-2).11 See table 1 for migraine features and frequency within our population. The diagnosis of epilepsy was made by review of the electronic patient record. Intractable epilepsy was defined as failure of 2 or more anticonvulsants after adequate trials with an average of 1 or more seizures per month with no 3-month period of seizure freedom.12 Seizure syndrome was assigned if the patient's neurologist had documented a specific diagnosis in the medical record.
Frequency of International Classification of Headache Disorders–2 criteria for children with migraine in the study cohort
Standard protocol approvals, registrations, and patient consents.
The Institutional Review Board approved the study protocol and questionnaire. Completion of the questionnaire was considered to be consent for being included in the study.
Statistical analysis.
All data were analyzed using the Pearson χ2 test. Minitab 15 (State College, PA) was used to perform regression analyses. A multivariate regression analysis based on a stepwise regression was performed to determine if demographic factors were independent. Age at onset of epilepsy and migraine were correlated (Pearson). The level of significance was p < 0.05.
RESULTS
Patient characteristics.
Patient characteristics are detailed in table 2. Families of 453 children with epilepsy completed a questionnaire. Fifty-three children did not meet inclusion criteria and were not reviewed further (figure). The mean age was 9.7 years; median age was 10 years. A family history of migraine was present in 66% of those with migraine.
Compared prevalence of demographic groups
Prevalence and onset.
The overall prevalence of migraine in our cohort was 25%. In adolescents ages 12 to 17 years, it was 32%. The mean age at onset for migraine was 7.4 years. Onset of migraine was more likely after epilepsy onset, with 49% developing migraines after seizures began, compared to 23% before and 19% in the same year (p = 0.0002). The remainder had missing data for age at headache onset.
Predictive factors.
We examined predictive factors for migraine (table 2). A greater percentage of patients ≥10 years had migraines than those <10 years of age. Children with a diagnosed epilepsy syndrome were more likely to have migraines. This was especially true in those with BECTS and JME. In contrast, children with intractable epilepsy and those seen in epilepsy clinic did not have a higher prevalence of migraine.
Potential confounders affecting the likelihood of concomitant migraine and epilepsy were then examined. We were particularly interested in the effect of age as older children are known to be more likely to have migraine regardless of comorbid conditions.1 A stepwise regression analysis was used to identify the most relevant factors predicting migraine prevalence. Based on these results, a multivariate regression showed that patients were more likely to have migraine if they were ≥10 years or had BECTS or JME (table 3). Intractability and type of clinic setting were not identified as statistically significant in the stepwise regression. Because the latter factors were part of our original hypotheses, we also included them in a regression analysis. However, their inclusion did not change the results of the regression.
Migraine regression analysis (p < 0.001)
Potential confounders in these analyses include postictal migraine and anticonvulsant use. Three patients reported having migraine during seizures, 9 less than 3 hours after a seizure, and 14 less than 3 hours before. Five patients did not complete this question. Per the medical record, of these 31 patients, 19 had migraine more frequently than seizures. Of note, only 1 patient reported having migraines associated only with seizures; however, this was not specifically investigated in this study. Only 1 patient with migraine had a diagnosis of migralepsy. Lamotrigine may induce aseptic meningitis, which could potentially be confused with migraine, and valproate and topiramate could be potentially preventative treatments.13,14 Thus, we studied the prevalence of migraine in patients taking individual anticonvulsants. There was no difference in prevalence of migraine in children receiving carbamazepine, clonazepam, felbamate, lamotrigine, levetiracetam, oxcarbazepine, topiramate, or valproate.
Treatment and discussion in clinic.
Migraines typically are treated prophylactically based on their severity or frequency.15 Ibuprofen, acetaminophen, and sumatriptan are reported as effective abortive treatments.1 Our center's physicians typically consider prophylactic treatment for children with at least weekly migraines.16 Fifty children (51%) with migraine satisfied these criteria. Only 25 (50%) of these children had discussions regarding migraine clearly documented in the medical record. Parents treated children with at least weekly migraine frequency with over-the-counter medications (ibuprofen or acetaminophen) 88% of the time.
Upon review of the medical record, 28% of patients who had migraines more frequently than once a week were prescribed anticonvulsants that have been reported as additionally beneficial as headache prophylaxis (e.g., valproate, topiramate, or zonisamide). None of these families reported that their child was receiving prescription medication for acute migraine treatment or prophylaxis. No child with weekly or greater migraine frequency was prescribed a triptan. Only 2 (2%) children with migraine of any frequency were prescribed a triptan, and both had BECTS.
DISCUSSION
This study is one of the few to look at a large number of children with epilepsy to evaluate for the prevalence of migraine and how it is treated within this cohort. Previous studies of children with epilepsy have used headache criteria other than the ICHD-2 criteria for determining pediatric migraine.7,8 Our study used the ICHD-2 criteria and looked specifically at children with intractable seizures. The largest study to look at children with epilepsy was a retrospective chart review of 475 children with any form of epilepsy in which 14.7% had headache with migrainous features.7 One other looked at 50 children to evaluate for headache.8 Previous data looking at children with comorbid headache and epilepsy also found that headache was more likely to occur after epilepsy onset.8,17 One recent study looked at children with headache to investigate if they were more likely to have epilepsy. The authors retrospectively studied almost 1,800 children with all types of headache (including migraine) to evaluate their risk of epilepsy. They determined that the prevalence of epilepsy was 3.1% (56 children).17
Our first hypothesis, that children with epilepsy would be more likely to have migraine, was supported in this cohort. The prevalence of migraine within our population of children with epilepsy was surprisingly high compared to previous studies7,8,18 and the age at onset quite young.17 Even though many of the children with migraines and epilepsy were young, there was still a higher prevalence in adolescents overall in our study, as reported previously.19 The onset of migraine after epilepsy onset may be explained by the theory of cortical spreading depression (CSD). CSD is thought to cause migraine and epileptic discharges cause epilepsy. These 2 changes in brain activity may affect one another. Epileptic discharges may lead to lowering of the CSD threshold and subsequent onset of migraine.20 Alternatively, activation of the trigemino-vascular system by epileptic discharges also has the potential to lead to migraine.21
Previous work has considered a common mechanism for seizure and headache.20,21 Additionally, we observed clinically that children with epilepsy appear to have more headache. Based on this we predicted that those with a higher seizure burden (intractable epilepsy or in epilepsy clinic and therefore more likely to have refractory epilepsy) would have a higher prevalence of migraine. This was not confirmed in this study. This may be partially explained by the higher likelihood of migraine in the BECTS and JME populations, both of which are not typically intractable.
Our study identified 2 contrasting epilepsy syndromes with increased migraine frequency. Both BECTS and JME are highly prevalent syndromes; however, children with BECTS have a defined time period during which they experience both clinical and electrographic seizures while spontaneous remission is low in JME. Children with BECTS often have continuing electrographic activity even when their seizures are well-controlled with medication, while children with well-controlled JME may have normal EEGs.
The likelihood of increased headache frequency in children with BECTS has been somewhat unclear in the literature to date. Some older studies did not find a higher prevalence of migraine in those with BECTS.22 A more recent study looking at 72 children with BECTS compared to children without epilepsy determined that the risk of migraine in BECTS was twice as high.23 Another study using the ICHD-2 criteria for diagnosing migraine in children found that children with BECTS and partial epilepsy had higher rates of migraine than those with other types of epilepsy.9 Until recently the relationship between JME and headache has not been examined. One recent retrospective study looked at the relationship between JME and headache and found a significantly higher risk for headache in JME patients.24 The results of our study also noted a high correlation. How JME as well as BECTS increase the risk for migraine is still unclear. It may be that the abnormal electrical activity specific to these syndromes lends itself to the development of migraine or that genetic abnormalities leading to these disorders also cause changes in the brain that are associated with migraine. Regardless of cause, these children should be identified as high risk for migraine and are then potential candidates for treatment for both disorders.
Our findings of low rates of migraine discussion in clinic as well as apparent treatment, while concerning, are consistent with other studies of pediatric migraine in children without epilepsy.25,26 One study of adolescents reported that only 16% were prescribed medications to treat their migraines.26 A study in adults with epilepsy and migraine reported that 4% of patients with both conditions were treated with a prescription medication for the headaches even though 33% had been diagnosed with severe headaches.5 Several possible explanations exist for these findings in our cohort. Epilepsy is often the primary focus when these children come to clinic. With limited clinic time headache may not always be addressed. Some child neurologists who focus their practice on epilepsy may be less comfortable treating headache. There also may be a concern for polypharmacy as many parents and providers do not want to prescribe multiple medications to children. Of note, no information was obtained about the success of abortive treatments, so this cohort may have been successfully treated with over-the-counter medications and triptans therefore would have been unnecessary.
Limitations in this study included the use of a questionnaire which did not allow for additional questioning or clarification regarding specific answers. This included questioning regarding peri-ictal headaches. A number of the subjects had at least a partial correlated timing of their migraine with their seizures leading to a potential pathophysiologic interaction. We relied on information recorded correctly by the families. Misinterpretation of questions could have led to recording of inaccurate information and may explain some of the low numbers seen on knowledge of prescription treatment. For example, families may be aware that their child is on a prescription medication for seizures, but not realize that it treats migraine as well. In this questionnaire, we only collected data from children who came to our clinics during the duration of the study. While we surveyed a wide range of patient demographics it is not known if our results are generalizable to patients outside of a tertiary care setting. Patients who chose to come to our center for care may have had more comorbidities, specifically migraine, than those who went to community-based centers for their care. Our population was compared to historical controls in order to compare our series to the overall prevalence of migraine alone, which does not allow for a direct cohort comparison.25,26 Although it would have been ideal to have a group of patients with migraine alone to compare other demographics to, the primary purpose was to evaluate risk factors of seizure disorders. Parents generally filled out the questionnaires so it is possible that they did not know the extent of their children's headaches and the children's answers to the questions may have varied from their parents' if asked themselves or alone. Additionally, including family history in further data collection for those with and without migraine would likely add to determining causative factors for comorbidity of the 2 disorders.
The comorbidity of epilepsy and migraine has been the source of discussion and investigation for over 100 years. In 1873, Liveing27 wrote in his treatise on migraine that he would describe “the intimate relations of megrim with the whole family of neurosal disorders of which Epilepsy is the type” Our study serves as the background needed for further investigation into the reason for this comorbidity in children and how best to treat it. Knowing the risk factors for the co-occurrence of migraine in a patient with epilepsy is the first step in knowing who to target for treatment. Child neurologists must be alert to comorbidities in children with epilepsy, and based on these results, adolescents and children with both BECTS and JME should be screened for migraine at each visit.
Understanding the relationship of these disorders is critical to furthering our knowledge regarding reasons for their co-occurrence and genetic links which could subsequently lead to an improvement in treatment options. Treatments that work for both migraine and epilepsy could then be considered in children with this comorbidity and trials of new therapies may be focused on these children. Further studies regarding the relationship, genetic or otherwise, of these 2 disorders will be helpful to understanding and learning how to treat them.
AUTHOR CONTRIBUTIONS
Sarah A. Kelley, MD: conceptualized and designed the study, analyzed and interpreted the data, drafted and revised the manuscript. Adam L. Hartman, MD: analyzed and interpreted the data, revised the manuscript. Eric H. Kossoff, MD: conceptualized and designed the study, analyzed and interpreted the data, revised the manuscript. The statistical analysis was conducted by all authors.
DISCLOSURE
S. Kelley reports no disclosures. A. Hartman receives research support from NIH (NINDS) and Johns Hopkins University School of Medicine; receives income from his clinical practice and reading EEGs (21% effort); and has provided an expert opinion in two medicolegal cases. E. Kossoff is a consultant for Eisai Inc. and Atkins Nutritionals, Inc. Go to Neurology.org for full disclosures.
Footnotes
-
Supplemental data at www.neurology.org
- Received December 14, 2011.
- Accepted March 28, 2012.
- Copyright © 2012 by AAN Enterprises, Inc.
REFERENCES
Disputes & Debates: Rapid online correspondence
- Response to Dr. Kasteleijn- Nolst Trenite et al
- Comorbidity of migraine and epilepsy is especially prevalent in children
REQUIREMENTS
If you are uploading a letter concerning an article:
You must have updated your disclosures within six months: http://submit.neurology.org
Your co-authors must send a completed Publishing Agreement Form to Neurology Staff (not necessary for the lead/corresponding author as the form below will suffice) before you upload your comment.
If you are responding to a comment that was written about an article you originally authored:
You (and co-authors) do not need to fill out forms or check disclosures as author forms are still valid
and apply to letter.
Submission specifications:
- Submissions must be < 200 words with < 5 references. Reference 1 must be the article on which you are commenting.
- Submissions should not have more than 5 authors. (Exception: original author replies can include all original authors of the article)
- Submit only on articles published within 6 months of issue date.
- Do not be redundant. Read any comments already posted on the article prior to submission.
- Submitted comments are subject to editing and editor review prior to posting.