To evaluate the sensitivity of quantitative electroencephalogram (QEEG) to detect electrographic seizure in critical traumatic brain injury patients in the intensive care unit (ICU) and further compare it to the raw electroencephalogram (REEG).

Six-hour QEEG epochs with 8 electrodes with international 10-20 system were performed in 18 critical traumatic brain injury patients with suspect seizure. REEG was performed during the same time. Each epoch was reviewed by two groups of physician independently. Two neurophysiologists reviewed REEGs and QEEG to identify seizures to serve as the gold standard. Analysis time and seizure detected frequency of REEG and QEEG were compared to identify the sensitivity and error rate for seizure identification.

A significant different sensitivity for seizure identification was found between QEEG and REEG [82.75% (50.00%-91.67%) vs 59.83% (0-83.33%), P=0.002]. Analysis time was significant shorter in QEGG relative REEG [0.09 (0-0.67) h vs 0.06 (0~0.34) h,P<0.001]. No significant false rate was found between two groups [0.09 per hour (0-0.67) vs 0.06 per hour (0-0.34), P=0.131].

QEEG can be used by doctors to shorten EEG review time with better sensitivity for seizure identification in critical traumatic brain injury patients with seizure in ICU.

Critical traumatic brain injury patients; Quantitative electroencephalogram; Raw electroencephalogram; Sensitivity; False rate