Abstracts

Rationale: Repeat EEGs are a problem the inpatient setting. Repeat EEG suggests a critical event is missed in the interval between the two recordings, an event that if captured would have aided in the diagnosis of patients. The continuous disconnection and reconnection of EEG recording devices creates opportunity for increased cost and wasted time. The aim of this study is to examine if those with short-interval repeat cEEGs can be identified, with the goal of limiting these events.

Methods: Retrospective chart review comprising adult patients admitted to a hospital setting with at least one EEG occurring during course of care. This study was approved by the IRB as a QI project. All patients were referred to UW Health University Hospital for evaluation and treatment. During inpatient stay course of care notes were obtained. All imaging orders, and results were noted. Additionally, patient length of stay and discharge status was retained. These medical records were retrieved and then reviewed for this study. Patient outcomes, defined as length of stay and discharge status, were assessed.
All patients analyzed were adults ( >18 years or older) identified by a search of electronic medical records at the University of Wisconsin Health University Hospital. Selection criteria included any patient who had an EEG recording from date to date. No other attempts were made to limit or exclude subjects, and the patient's medical records with at least one EEG recording were reviewed. The medical records of identified patients were checked for additional EEG orders or recordings (within 48 hours of original), notable EEG activity, length of stay, and discharge status.


Results: A total of 61 patients were identified with an EEG study beginning within 48 hours of termination of a prior EEG during their hospital admission. A total of 100 control patients with a single EEG study and 16 patients with an EEG study ordered within 48 hours of termination of a prior EEG but not performed were also analyzed.

Neither age nor sex differed across the groups. Repeat EEGs were more likely to be admitted to the neurosurgery service (27/61 vs 12/100) and more likely to involve an intracranial hemorrhage (17/61 vs 7/100). Patients with repeat EEGs were more likely to die during their admission (15/61 vs 19/100) and less likely to be discharged home (11/61 vs 32/100).


Conclusions: Short-interval (< 48 hours) repeat EEGs in the inpatient setting are clinically problematic. Not only is a clinically-relevant event missed from cEEG monitoring, but additional time is expended for clinical decision making and EEG technicians expend their valuable time removing and reattaching electrodes. Identification factors associated with short-interval repeat EEGs allows and the development of a risk score provides additional information to aid the determination of duration of cEEG monitoring. Additional work to identify additional factors and validate the findings is underway.

Funding: N/A