Abstracts

Rationale: Sudden unexplained death in epilepsy patients (SUDEP) is a major concern for both parents and neurology practitioners. Specific risk factors for sudden death have been evaluated and the most frequently reported risk factors include: nocturnal seizures, being in bed, and a history of generalized tonic-clonic seizures. There is some evidence that SUDEP may be preventable if another person is able to give assistance. We know of only one published study in adults evaluating a nocturnal bed seizure alarm. We evaluated the same system (Med Page seizure alarm) to determine its efficacy in detecting nocturnal seizures in children and adolescents.Methods: 29 children were enrolled in the study. Patients were recruited from a population being admitted to the Epilepsy Monitoring Unit for evaluation of ongoing seizure disorders. Patient ages ranged between 12 months to 22 years, with multiple seizure types including generalized tonic-clonic, generalized myoclonic, simple and complex partial, and generalized tonic. The patients' weights varied between 9.08-119.5 kg. Patients had standard video electroencephalography (VEEG), cardiopulmonary monitors and nursing monitoring. The Med page seizure alarm was placed in the bed according to manufacturer instructions. The VEEG record was reviewed to detect any seizure events missed by the alarm, parents, or nursing staff. If the Med page alarm system did not detect a seizure event, adjustment to the sensitivity was made in an attempt to document the best parameteres for each age/weight of the patient. All VEEG events were reviewed retrospectively in 24 hour increments by board certified epileptologists to determine if events were related to electrographic seizure activity. Patient care decisions were not affected by the study. Data collected included all seizures occuring during monitoring, if the bed alarm was triggered, the type of seizure, and any alarm triggers not associated with seizure activity.Results: Of the 29 patients monitored, 69 seizures were recorded by video EEG: 11 GTCS, 42 complex partial seizures, 6 generalized tonic, 8 generalized myoclonic, and 2 simple partial. The bed alarm sounded a total of 39 times, 35 of which were false alarms. Of the four times the alarm sounded at the time of an electrographically confirmed seizure, three were generalized tonic clonic seizures (in a female weighing 75kg), and one was a complex partial seizure with motor phenomenon (in a male weighing 79kg).Conclusions: The Med page seizure alarm is ineffective in detecting generalized tonic-clonic, complex partial, generalized tonic, generalized myoclonic, and simple partial seizures in children and adolescents. The best detection rate was for generalized tonic-clonic seizures at 27.3%, and this only occurred in patients with weights consistent with adult size. Other, more effective seizure alarm systems are needed to help in detection of nocturnal events. Research funded by The Shainberg Neuroscience Research Fund.