Implementation and Early Evaluation of a Quantitative Electroencephalography Program for Seizure Detection in the PICU

Abstract

Objectives: To describe implementation and early evaluation of using quantitative electroencephalography for electrographic seizure detection by PICU clinician staff.

Design: Prospective observational study of electrographic seizure detection by PICU clinicians in patients monitored with quantitative electroencephalography. Quantitative electroencephalography program implementation included a continuous education and training package. Continuous quantitative electroencephalography monitoring consisted of two-channel amplitude-integrated electroencephalography, color density spectral array, and raw-electroencephalography.

Setting: PICU.

Patients: Children less than 18 years old admitted to the PICU during the 14-month study period and deemed at risk of electrographic seizure.

Interventions: None.

Measurements and main results: Real time electrographic seizure detection by a PICU team was analyzed for diagnostic accuracy and promptness, against electrographic seizure identification by a trained neurophysiologist, retrospectively reading the same quantitative electroencephalography and blinded to patient details. One-hundred one of 1,510 consecutive admissions (6.7%) during the study period underwent quantitative electroencephalography monitoring. Status epilepticus (35%) and suspected hypoxic-ischemic injury (32%) were the most common indications for quantitative electroencephalography. Electrographic seizure was diagnosed by the neurophysiologist in 12% (n = 12) of the cohort. PICU clinicians correctly diagnosed all 12 patients (100% sensitivity and negative predictive value). An additional eleven patients had a false-positive diagnosis of electrographic seizure (false-positive rate = 52% [31-73%]) leading to a specificity of 88% (79-94%). Median time to detect seizures was 25 minutes (5-218 min). Delayed recognition of electrographic seizure (> 1 hr from onset) occurred in five patients (5/12, 42%).

Conclusions: Early evaluation of quantitative electroencephalography program to detect electrographic seizure by PICU clinicians suggested good sensitivity for electrographic seizure detection. However, the high false-positive rate is a challenge. Ongoing work is needed to reduce the false positive diagnoses and avoid electrographic seizure detection delays. A comprehensive training program and regular refresher updates for clinical staff are key components of the program.