Epilepsy Information

American Epilepsy Society (Abst. 2.057)

Diagnostic value of continuous EEG in the treatment of sub-arachnoid hemorrhage-induced vasospasm

Authors: J. M. Politsky, P. Rodgers, I. Ugorec

Failure to recognize and treat subarachnoid hemorrhage-induced vasospasm (SAH-V) results in significantly higher morbidity and mortality in this population of critically ill patients. We sought to determine if continuous EEG (CEEG) has predictive value for the early identification and treatment of SAH-V.


This was a retrospective review of all 108 patients admitted to our institution with a diagnosis of SAH in 2011. 56/108 patients underwent CEEG. Upon diagnosis, patients were classified as grade I-V using the Hunt Hess (HH) grading system. Patients underwent one (CEEG1) or two (CEEG2) phases of CEEG. CEEG was initiated within 24 hours of admission; CEEG2 was started at about day 6-7. In 46.4% of cases CEEG1 lasted < 72 hours, and in 53.6% the study lasted > 5 days. Abnormalities included mild background slowing (mbs), focal or diffuse slowing, polymorphic irregular sharply contoured waveforms (SAH-V pattern), spikes, electrographic seizures (Sz), periodic discharges (PD). Patient outcomes were tracked and categorized as discharge to home, rehab facility, long term care (LTC) facility, or death. Patients diagnosed with SAH-V were treated with triple H therapy (hypertension, hypervolemia, hemodilution).


Of the 56 patients who underwent CEEG, 15 underwent a second period of recording. CEEG2 was initiated because of clinical deterioration. Patients who underwent CEEG had the following HH breakdown: CEEG1: Grade I = 5; Grade II = 15; Grade III = 16; Grade IV = 15; Grade V = 5; CEEG2: Grade I = 0; Grade II = 3; Grade III = 4; Grade IV = 8; Grade V = 0. SAH-V was diagnosed in 34/56 (60.1%) who underwent CEEG. 19 patients had CEEG1 and 15 patients had CEEG2. 7/19 CEEG1 patients with SAH-V had prolonged CEEG studies. Of the 15 CEEG2 SAH-V patients 60% were discharged to home or rehab and 40% expired or went to long term care. The EEG was characterized as mbs in 100% of patients with no evidence of vasospasm and was characterized as abnormal due to SAH-V pattern in 86.7% of (13/15) CEEG2 SAH-V patients. Electrographic Sz and/or PDs occurred in 13.3% (2/15). Additionally, the SAH-V pattern was observed in 100% (11/11) of CEEG1 SAH-V patients in whom CEEG was prolonged. Electrographic sz and/or PDs occurred in 22.2% (2/11). Overall, this distinct SAH-V pattern occurred in 24/26 patients (92%) with SAH-V. The prevalence of sz and/or PDs was 4/26 (15%).

We identified a distinct electrographic pattern associated with vasospasm - the SAH-V pattern. Patient outcomes appear to be more favorable when HH grade is I-III and when SAH-V is identified and treated as early as possible. Our initial clinical experience in patients with SAH undergoing CEEG strongly suggests that CEEG can detect pathophysiologic changes associated with SAH-V as early or earlier than other accepted methods, such as clinical deterioration or angiogram. A prospective study comparing CEEG with CT or invasive angiogram is required to provide definitive evidence of the utility of this diagnostic tool.