Introduction

Some patients with epilepsy who have uncontrolled seizures experience bouts of increased seizure activity. The phenomenon of seizure clusters referred to here has also been termed acute repetitive seizures, seizure flurries, or serial, recurrent, or crescendo seizures. Seizure clusters may occur at any age and with any seizure type.^[1]

Definition of Seizure Clusters

There is no standard definition of a seizure cluster. In clinical trials, a common approach is to define seizure clusters as either ≥2 or ≥3 seizures in a 24-hour period. There is, however, no universally accepted definition for the number of seizures that constitutes a "cluster" or agreement on the number of seizures that minimally define a cluster. In part this is due to the variability of patient presentations.^[2,3]

A clinically useful definition is that a seizure cluster is a series of seizures closely grouped in time. There are typically short (or shorter than normal) interictal periods that often last only hours or even minutes. Another common clinical approach is to consider clustering as an increase over the patient's typical seizure frequency. Seizure clusters can be experienced over a time frame as short as several minutes or as long as 1 or 2 days, as seen in women with catamenial patterns of seizures that occur perimenstrually.^[2,3]

Prevalence of Seizure Clusters

The prevalence of seizure clusters ranges widely in studies, presumably due to the different definitions of clustering. In addition, data from tertiary care epilepsy clinics are skewed toward patients with intractable epilepsy, who are more likely to maintain prospective long-term seizure diaries.^[2,3]

In outpatient studies to date, the prevalence of seizure clusters varies from 13% to 76%. In inpatient studies, the range is 18% to 61%. The higher prevalence in inpatient studies probably reflects the effect of rapid withdrawal of antiepileptic medications in epilepsy monitoring units.^[2]

A population-based study was conducted using the UK general practice research database as of 2005.^[4] The researchers estimated the crude prevalence of acute repetitive seizures to be 2.3 per 10,000 and the age-adjusted prevalence to be 2.5 per 10,000. In this study, acute repetitive seizures were defined as ≥3 partial or generalized seizure episodes over a 24-hour period.

Recently, the Epilepsy Therapy Project conducted a prevalence study using My Epilepsy Diary.^[5] This anonymous digital seizure diary was made available via web and smartphone in 4 languages (English, French, Italian, and Spanish). The researchers obtained data from 28,697 unique users who kept a diary for ≥60 days. Seizure clusters were defined as ≥2 seizures per 24-hour period, 12:00 AM to 11:59 PM. Altogether, 5018 patients (17.5%) reported seizures occurring as clusters, and 1177 recorded >1 seizure in a 24-hour period. The patients ranged from age 1 year to >70 years, and most patients had seizure onset before age 30 years. In those who reported >1 seizure in a 24-hour period, the most common number of seizures on cluster days was 2 per 24 hours (Figure).

Figure. Seizure clusters were measured in the Epilepsy Therapy Project study by recording seizure clusters in a digital seizure diary. The majority had 5 or fewer seizures per day.

On days with ≥2 seizures, one-third of seizures occurred within 3 hours of each other and the rest occurred within 6 hours.

Risk Factors and Triggers for Seizure Clusters

Some risk factors for seizure clusters are extratemporal seizure onset, particularly frontal lobe onset; a history of head trauma with loss of consciousness; longer duration of epilepsy; history of status epilepticus; and poor seizure control.^[2,6] The latter risk factor, however, may be an artifact of high seizure frequency.^[3] For patients with focal seizures, the strongest risk factor for seizure clusters appears to be a previous experience of clustering.^[6] Seizure clustering related to the menstrual cycle, called catamenial epilepsy, has been described in up to 40% of women with epilepsy.^[2]

What precipitates seizure clusters is not well understood. In the study that used the digital seizure diary, patients were asked, "What was the primary trigger?" each time they reported a seizure or seizure cluster. Data about perceived precipitating factors were collected for 12,696 of the 29,341 seizures that occurred in clusters.^[5] Altogether, 28% of the seizure clusters were attributed to sleep, waking, or sleep deprivation; 12% to mood/stress; 4% to menstrual periods; 4% to missed medication or medication changes; 3% to medical illness; and 1% to use of alcohol or recreational drugs. Additionally, 11% of patients selected some other trigger, 30% said they didn't know what the trigger was, and 6% said there was no trigger.

Consequences of Seizure Clusters

Morbidity and Mortality

Seizure clusters are often treated at home with benzodiazepines. These drugs are generally well tolerated but are not free of adverse effects, notably sedation and risk of respiratory suppression when used in high doses or repeatedly.^[7,8] In the absence of prompt treatment, seizure clusters can evolve into status epilepticus.^[9,10] Even when seizure clusters don't progress, they often lead to emergency department visits.^[11] Seizure clusters that occur in the epilepsy monitoring unit may negatively affect the ability to locate the epileptogenic zone, thus interfering with presurgical evaluation of patients who have refractory epilepsy.^[3]

Seizure clustering can also be associated with postictal psychosis.^[12-14] Sequelae include higher risk for recurrence of postictal psychosis, the need for prophylactic or standing psychotropic medication, or even psychiatric ward admission.^[2]

Most worrisome of all, seizure clustering might have a direct effect on mortality. In a long-term observational study of childhood-onset epilepsy in Finland (average follow-up, 37 years), the mortality rate for patients who had seizure clusters during treatment with antiepileptics was 42% vs 14% for patients who did not experience seizure clusters.^[15] This finding may be confounded by greater severity of epilepsy or higher risk of status epilepticus in patients who experience clustering.^[2] The risk of death was not significantly increased among patients who had clusters only prior to drug treatment.^[15]

Quality of Life

Until recently, data were lacking on the quality of life (QoL) and emotional well-being of affected patients and their caregivers. The Seizure Cluster Burden of Illness US Survey confirmed that seizure clusters have substantial adverse effects on patient and caregiver QoL, including emotional, financial, and social components.^[16] The survey assessed attitudes and opinions of patients, caregivers, and physicians. Most respondents in all groups felt that seizure clusters have a moderate or major negative impact on patient QoL. In fact, physicians (80%) were significantly more likely than patients (70%) or caregivers (66%) to have this opinion.^[16] Most patients and caregivers reported that seizure clusters negatively affect the patient's independence, mood, and ability to engage in normal activities (Table 1).

Table 1. Negative Impacts of Seizure Clusters on Patients' Lives

Patients and caregivers reported information about their perceptions of negative QoL in several discrete categories. There tended to be a large level of agreement on the impact of each aspect of a patient's QoL.^[16]

In particular, more than two-thirds of patients indicated that their work had been affected by seizure clusters. More than half of this group reported that they had experienced a seizure at work, had to stop working altogether for a period of time, and/or had to disclose their condition to their employer.^[16] Patients reported that seizure clusters made them feel exhausted (76%), confused/slow-thinking (70%), stressed (63%), depressed (62%), helpless (56%), scared (51%), and/or discouraged (49%).^[16] Most patients, caregivers, and physicians agreed that seizure clusters add significant financial burden to the patient's life.^[16]

More than half of caregivers reported a substantially negative or moderately negative effect of seizure clusters on their own QoL.^[16] Table 2

Table 2. Negative Impacts of Seizure Clusters on Caregivers' Lives

In addition to an impact on patients, caregivers also feel a significant impact of seizures on their quality of life. Many factors contributed to a negative quality of life, particularly mood, work and hobbies, and travel.^[16]

Half of caregivers reported that the patient's seizure clusters had a somewhat negative or very negative effect on their work. Of this group, almost half (49%) said they sometimes had to reduce work hours or take whole days off and 43% experienced having to stop working altogether for a period of time.^[16] When asked to describe how they feel when the patient experiences seizure clusters, caregivers most frequently reported feeling stressed (67%), helpless (64%), scared (59%), and/or overwhelmed (52%).^[16]

In certain important respects, the responses of physicians were too optimistic when compared with those of patients or caregivers. For example, nearly three-quarters of physicians, compared with only about half of patients and caregivers, agreed or strongly agreed that patients with seizure clusters can achieve a fulfilling life.^[17] Additionally, significantly fewer physicians than patients strongly agreed that seizure clusters impede daily life. When asked about emotional and financial burdens, significantly fewer physicians than patients strongly agreed that there was a significant increase on emotional and financial burden associated with seizure clusters.^[17]

Conclusion

Seizure clusters are associated with negative impacts on QoL and health. Increased physician education about managing seizure clusters may improve communication and optimize patient care by establishing an early diagnosis of seizure clusters and then employing appropriate management techniques. As research continues to evolve, patients will have greater options to manage their seizure clusters. Physicians can improve seizure control and decrease emergency facility use and thus medical costs by proactively discussing seizure clusters with their patients and appropriately diagnosing the clusters.