Triggers and techniques in termination of partial seizures.

OBJECTIVE

Growing interest in seizure prediction exists as a means to deliver newer antiepileptic therapies, though patient self-termination of seizures has received little attention.

METHODS

Two hundred twenty-three patients able to recognize seizure onset were surveyed in an outpatient epilepsy clinic. A seven-question survey administered prospectively assessed self-reported seizure prediction and clinical techniques used for self-termination. Survey responses targeted percentage predictability of seizures, timing of clinical prediction, likelihood/timing of termination, frequency and effectiveness of methods used, and perspectives of patient and physician belief in self-termination.

RESULTS

Two hundred twenty-three patients (89 males) with a mean age of 42.7 years, average duration of epilepsy of 20.8 years and monthly mean seizure frequency of 4.1 comprised the study group. Thirty-eight percent completed >75% of the survey. Prior treatment included a mean of 6.0 AEDs (40/192 had surgery); 65% had ongoing seizures. Sixty percent of 223 patients reported a history of an aura, and 39% consistently noted auras for >75% of their current seizures. Of the patients with auras, seizure triggers were reported in 74%, with worry and stress (N=69), sleep deprivation (N=60), and missed medication (N=56) most frequently cited. Seventeen percent were positive/somewhat sure they could predict onset, with approximately 20% noting rapid onset in <15 seconds. Twenty-two of 82 noted that they had some ability to self-terminate their seizures, and 9% were positive that they could do so. Methods to self-terminate were effective (>75% certainty) in 35% (26/75). The primary methods were lying down/resting and taking extra medication.

CONCLUSION

The majority of patients with partial seizures recognize triggers of seizure onset. In addition, more than one-third believe they can effectively self-terminate their partial-onset seizures. Lying down, resting, and taking extra medication were the most common techniques instituted by patients. Correlating clinical symptoms at seizure onset with termination may help improve the sensitivity in seizure prediction.