Association Between Migraine Comorbidity and Psychiatric Symptoms Among People With Newly Diagnosed Focal Epilepsy

Epilepsy and migraine are comorbid neurological conditions; people with epilepsy have almost double the prevalence of migraine compared with those without epilepsy (1). The conditions both consist of paroxysmal, short, and somewhat stereotyped events. They share triggers such as bright flashing lights, stress, and sleep deprivation. They might be preceded by transient neurological symptoms (aura). Both seizure and migraine attacks might be accompanied by some level of cognitive dysfunction. Seizures might be preceded, accompanied, or followed by non-migrainous headaches. Finally, comorbid psychiatric conditions are common among both populations (2, 3).

The neuropsychiatric conditions that are comorbid with both epilepsy (2) and migraine (3) have been well established. It is well known that the neuropsychiatric disorders common to these conditions can influence people’s prognosis and quality of life (2, 4–6). There is strong evidence that both the relationship between epilepsy and depression (7, 8) and the relationship between migraine and depression are bidirectional (9). For example, between attacks, patients experience fear of recurrent attack and may develop reactive symptoms of depression and anxiety (10). People with both conditions also have higher reported rates of suicidality (11, 12). However, most studies have examined the psychiatric comorbid conditions in relation to either epilepsy or migraine. Little is known about the association of the psychiatric symptoms and suicidality in patients with both conditions.

Given the paucity of knowledge about comorbid psychiatric conditions specifically among individuals with both epilepsy and migraine, the goal of this study was to compare the presence of symptoms of depression, anxiety, and suicidality among patients with newly diagnosed focal epilepsy with and without migraine. We hypothesized that psychiatric symptoms would be more common among people with focal epilepsy and migraine than among those without migraine because both migraine and epilepsy are comorbid with depression and anxiety. A higher burden of psychiatric symptoms would affect patient evaluation and management.

Methods

Participants

We used data collected for the Human Epilepsy Project (HEP) (13), a 6-year, prospective, observational, multicenter study to evaluate clinical characteristics of patients with newly treated focal epilepsy. The criteria for subjects to be included in the HEP were as follows: at least one confirmed spontaneous focal seizure and start of antiepileptic medication not more than 4 months before enrollment in the study. We included participants older than age 18. The patients were enrolled from July 18, 2012, through September 28, 2017. This study was approved by the New York University School of Medicine Institutional Review Board.

Questionnaires

All questionnaires were administered at the initial visit by study coordinators. The demographic questionnaire asked about ethnicity (Hispanic or Latino, not Hispanic nor Latino, unknown, or not reported) and race (American Indian or Alaska Native, Asian, Black or African American, Native Hawaiian or other Pacific Islander, White, unknown, or not reported). Those questions were asked independent of each other.

The Center for Epidemiologic Studies Depression Scale (CES-D) is a validated questionnaire developed by Lenore Sawyer Radloff in 1977 to assess depressive symptoms among the general population using 20 questions, all ranked on 4-point scales (14, 15). A cutoff score of at least 16 was used to screen for presence of significant depressive symptomatology.

The 7-item Generalized Anxiety Disorder (GAD-7) scale consists of seven questions, all ranked on a 4-point scale. It has demonstrated validity in patients with epilepsy and patients with migraine (16, 17). A cutoff score of at least 10 was used for the GAD-7 because it has good sensitivity and specificity for screening for generalized anxiety disorder (89% and 82%, respectively), panic disorder (74% and 81%, respectively), social anxiety disorder (72% and 80%, respectively), and posttraumatic stress disorder (86% and 84%, respectively) (18).

Suicidality was assessed using the validated Columbia-Suicide Severity Rating Scale (C-SSRS) (19). In this study, an answer of “yes” to at least one of the five suicidal ideation questions was scored as a 1 for C-SSRS ideation, and an answer of “yes” to at least one of the five attempt questions was scored as a 1 for C-SSRS behavior. An answer of “yes” to either at least one ideation question or at least one attempt question was scored as a 1 for C-SSRS ideation or behavior.

Migraine was assessed using the American Migraine Prevalence and Prevention migraine assessment questionnaire (20, 21). The 13 questions for this assessment tool inquire about migraine criteria as described in the International Classification of Headache Disorders–Third Edition, such as at least five attacks following criteria 2–4; migraine lasting at least 4 hours; nausea, vomiting, or both; photophobia; and at least two of the following four characteristics: unilateral migraine, throbbing migraine, migraine of at least moderate pain intensity, and migraine aggravated by activities such as climbing stairs (22). The questionnaire also asks participants to rate the intensity of the head pain they have during their most severe headache on a scale ranging from 1 (very mild) to 10 (the worst pain).

The Migraine Disability Assessment Test (MIDAS) is a validated measure of the impact of migraine on patients’ quality of life. MIDAS scores are classified in four categories: little or no disability for scores ranging from 0 to 5, mild disability for scores ranging from 6 to 10, moderate disability for scores ranging from 11 to 20, and severe disability for scores of at least 21 (23).

Statistical Analyses

The main predictor was migraine comorbidity. The main outcomes were CES-D score ≥16, GAD-7 score ≥10, and at least one positive answer on the C-SSRS. First, data distribution was assessed for normality using the Kolmogorov-Smirnov test. Because the data set was not normally distributed, the analyses were performed using the Mann-Whitney U test, chi-square test, and linear regression. Given that age and gender are associated with migraine, anxiety, and depression, we used age and gender as covariates in the logistic regressions. There was no need for analysis of missing data because less than 10% of patients were missing some data. The participants with missing data were omitted from statistical analyses. Statistical analyses were conducted with Excel version 15.0.4833.1001 and IBM SPSS Statistics, version 22.0.0.

Results

Demographic Characteristics

Of 349 patients with new-onset focal epilepsy, 74 (21.2%) had migraine (Table 1). The proportion of women in the comorbid migraine group was markedly higher than among the group of patients without migraine (75.7% vs. 55.6%; χ²=9.75, df=1, p<0.001) (Table 1). The median age of the total sample was 36 years (mean=37.4 years [SD=12]). In the group without migraine, the median age was 36 years (range, 19–64 years; mean=37.6 years [SD=12]). In the group with migraine, the median age was 34 years (range, 19–60 years; mean=36.3 years [SD=11.9]). No difference in age was found between the patients without migraine and the patients with migraine (Mann-Whitney U test=0.36). No statistically significant differences were found between the patients without migraine and the patients with migraine in terms of race and highest level of education, (Table 1). The proportion of patients with comorbid migraine working a full-time job was lower than the group without migraine (66.2% vs. 50%; χ²=6.5, df=1, p=0.011) (Table 1). Among the 29 participants who identified as Hispanic or Latino, one identified as Black, 20 identified as White, and the rest checked the “unknown” box for the race question on the demographics questionnaire.

Headache Characteristics

Among the patients with migraine, the mean age at headache onset was 17.9 years (SD=8.5). The mean headache pain intensity score was 8.1 (SD=1.8). The mean MIDAS score was 10.3 (SD=19.1). More than a quarter (29.8%) of patients with migraine reported moderate to severe disability from their migraine.

Primary Outcomes

Using the chi-square test and a CES-D score threshold of at least 16, the proportion of patients with focal epilepsy who screened positive for depressive symptoms was higher in the group with migraine than in the group without migraine (35.2% vs. 22.7%, p=0.031) (Table 2, Figure 1). Using linear regression with gender, age, and migraine as predictors and CES-D score as the dependent variable, only migraine was significant (standardized β=0.13, p=0.018), not age or gender. Using the chi-square test and a GAD-7 threshold of at least 10, the proportion of patients with focal epilepsy who screened positive for anxiety symptoms was higher in the group with migraine than in the group without migraine (16.9% vs. 4.8%, p<0.001) (Table 2, Figure 1). In a linear regression with gender, age, and migraine as predictors and GAD-7 as the dependent variable, only age was significant (standardized β=−0.11, p=0.048). Comorbid migraine was not associated with the presence or absence of CSSR-S suicidal ideation or behavior (Table 2).

Discussion

There were more women in the group with migraine, which was expected because the prevalence of migraine is higher among women (24). However, this difference did not play a role in the findings. Among our sample of people with newly diagnosed focal epilepsy and migraine, the age at onset of headache is consistent with the age at onset of migraine in the general population (24).

This cross-sectional study of select psychiatric symptoms among patients with new-onset focal epilepsy who have or do not have comorbid migraine had two key findings: First, comorbid migraine among patients with focal epilepsy is associated with depression symptoms. Second, the two groups had no difference in terms of anxiety symptoms or suicidality.

The prevalence of depression among adults in the United States in 2017 was 7.1%. Depression is the most common psychiatric condition comorbid with epilepsy and one of the most common comorbid with migraine (3, 25, 26). Although the reported prevalence of depression among people with migraine is highly variable, ranging from 8.6% to 47.9%, several studies have shown that patients with migraine are 2.5 times more likely to have depression than people without migraine (3). Although the reported prevalence of depression among people with epilepsy is also highly variable, ranging from 11.2% to 39.0%, a large population-based study showed an odds ratio of 1.8 (95% CI=1.0, 3.1) for a lifetime episode of major depressive disorder among people with epilepsy (2). In this study, the proportion of patients with focal epilepsy without migraine who screened positive for depressive symptoms is higher than expected in the general population of people without migraine and without epilepsy. The proportion of patients with newly diagnosed focal epilepsy with comorbid migraine who screened positive for depressive symptoms is within the range observed in cohorts of people with migraine and without epilepsy, which suggests that the increased risks of developing depressive symptoms independently associated with migraine and epilepsy do not directly add up among the population of people with both migraine and epilepsy.

Potential theories for the association between comorbid migraine and depression among people with focal epilepsy include dysfunction in the serotonergic neural activities, shared pathophysiology (27, 28) with affective disorders, pain, socioeconomic status difficulties (29, 30), and shared genetics (31).

Limitations

Our study has several limitations. The study is cross-sectional, and therefore we cannot determine temporality (e.g., whether the depression might have predated the epilepsy). However, this is a population with focal epilepsy newly diagnosed within 4 months of enrollment, so the participants are known to not have a high seizure burden or long exposure to antiepileptic medications. Second, for this specific analysis of the data, we used validated screening instruments for depression and anxiety rather than clinical diagnoses. Third, we had no data on headache frequency, headache medications, seizure frequency and localization, and seizure medications. Medication use might be a confounder in the association between comorbid depression and migraine among people with new-onset focal epilepsy. For example, providers might be more inclined to prescribe topiramate than lamotrigine for new-onset focal epilepsy to patients with comorbid migraine. However, the participants were assessed within 4 months of focal epilepsy diagnosis, so as noted earlier, they did not have long-term exposure to any antiepileptic medications for seizure treatment. Although to our knowledge the literature has no guidelines for the treatment of epilepsy with comorbid migraine, therapeutic independence when treating two comorbid conditions presents many advantages, such as optimal treatment of both conditions and lower risk of unacceptable side effects. Focal epilepsy localization is another potential confounder for which we do not have data. For example, temporal lobe epilepsy is thought to be most likely to be associated with depression.

Future Studies

Future work might address the potential benefit of systematically screening people with focal epilepsy and migraine for depression and study pharmacological and nonpharmacological management of depression in people with focal epilepsy and migraine. For example, progressive muscle relaxation therapy has been shown to be beneficial to people with epilepsy and to people with migraine (32, 33), and it might be helpful in treating the neuropsychiatric symptoms of patients with epilepsy and comorbid migraine. Third, future studies might assess whether there is an association between postictal headaches and psychiatric symptoms. Fourth, another direction would be to look at clinical outcomes to assess whether comorbid migraine and a potentially higher risk of depressive symptoms are associated with a worse prognosis in people with focal epilepsy.

Conclusions

Among a sample of patients with newly diagnosed focal epilepsy, 21.2% had migraine. The patients with epilepsy and comorbid migraine reported more symptoms of depression than those without migraine. Clinicians should be alert for signs of depression among patients with epilepsy and migraine. Given the paucity of studies evaluating the relationship among epilepsy, migraine, and other associated psychiatric conditions, further studies should be performed to better assess interactions between conditions and the treatment implications.