Abstracts

Rationale: The utility of continuous video electroencephalography (cVEEG) in detecting seizures has been well documented in the literature. The majority of these studies have focused on seizures in critically ill patients, a population in which 8-48% of patients may experience a seizure. The aim of this study was to determine the rationale and the yield of cVEEG in detecting seizures in patients admitted to general neurology and medical units in addition to critical care units. Methods: The Henry Ford EEG laboratory database was retrospectively searched to identify patients who underwent cVEEGs from 10/1/2011 through 12/31/2011 and then from 02/11/2012 through 04/14/2012 as part of IRB approved quality implementation study. All cVEEGs were initially reviewed by epileptologists at the Henry Ford Comprehensive Epilepsy Program. Demographic data, rationale for cVEEG, requesting hospital unit and the presence of seizures were collected on each patient. Each cVEEG report was reviewed to determine any abnormal finding defined as periodic epileptiform discharges (PEDs), interictal epileptiform activity (IEA), triphasic waves (TW), intermittent rhythmic delta activity (IRDA), focal slowing, background slowing, excessive beta activity, or normal. Results: A total of 135 patients who had cVEEG were included (65 males, 70 females) with ages ranging 16-94 years-old (mean 56.5). Twenty-three patients (17%) had either one clinical seizure (12 patients), one subtle seizure (7 patients), or one clinical and one subtle seizure (4 patients). Patients with a referring diagnosis of seizure were more likely to experience a seizure when compared to patients with other referring diagnoses (34.8% vs 7.9%, p<0.001, Fisher's exact test) (See table 1). In addition, patients with PEDs on cVEEG had significantly higher rate of seizures when compared to the other primary diagnoses (52.6% vs 11.2%, p<0.001, Fisher's exact test) (Table 2). Neurology units were most likely to order cVEEG with 74.1% (n=100), with 11.1% (n=15) from other ICUs and 14.8% (n=20) from other units. Overall, the neurology units accounted for 19 of the 23 seizure patients. Although many of these patients with seizures were in critical care units (7 in the NICU, 5 in the stroke unit, and 2 in other ICUs), 9 patients were in general neurology and other general medical units. The difference in seizure rates among the units (neurology 19%, other ICUs 13.3%, other units 10%) was not significant (p=0.69, Fisher's exact test). Conclusions: This study demonstrates that seizures are common in hospitalized patients and illustrates high utility for cVEEG in patients with suspected seizures in critical care units as well as in general neurology and medical units. Seventeen percent of patients receiving cVEEG had at least one seizure. Patients with PEDs had almost 5 times the risk of having a seizure when compared to patients with other diagnoses and patients with a referring diagnosis of seizure were over 4 times as likely to experience a seizure during monitoring. cVEEGs remains a helpful diagnostic tool in detecting seizures in different hospital unit settings.