Abstracts

Rationale: Vasospasm is a major complication of SAH and may be recurrent. Timely detection prior to the development of clinical signs facilitates early intervention that may prevent permanent neurological deficits. This is especially pertinent for ischemia involving the anterior brain regions that is clinically subtle in the acute stages. Prior studies predominantly reported the utility of quantitative CEEG in detecting vasospasm in patients with poor-grade SAH. We investigated the use of CEEG to predict the onset of clinical vasospasm and the response to treatment in an unselected group of SAH patients. Methods: Patients with aneurysmal SAH of all grades were prospectively studied. CEEG was performed using standard 10-20 international montage. Quantitative analysis of regional anterior alpha power was performed in this montage and in a selection of simplified montages involving 7-10 electrodes. Correlation with results using the 10-20 montages was determined for each day. We used the product of the standard deviation and mean alpha power as a composite measure of the variability and amplitude of alpha power. This was calculated over a 6-hour duration and repeated along a window sliding by 30 minutes. The measurements were graphically displayed to provide a continuous daily trend. A clinician predicted the clinical status of the patient for the ensuing day based on the clinical data available up to that morning. The reviewer was blinded to the time and treatment rendered for each episode of clinical vasospasm. A scoring system was used to denote if the status had improved, deteriorated or remained unchanged from the previous day. The daily CEEG trends prior to the morning in question were then provided and another prediction made. Scores were compared to the actual clinical state that was independently determined. Results: 8 patients were studied (Hunt-Hess 1-5). 1 patient was excluded due to technical difficulties with the CEEG recording. 4 patients developed clinical anterior vasospasm confirmed radiologically either by angiography, transcranial Doppler or CT evidence of new infarction. 3 of these patients developed recurrences of clinical vasospasm. Predictions of 42 days were made. The mean absolute difference between predicted and true scores was 1.57 with clinical data alone and 1.23 with additional information from the CEEG. The sensitivity of predicting a deterioration using clinical data was 30%, increasing to 70% with the addition of CEEG data (specificity 75% and 72% respectively). The ability to predict an improvement rose from 9% to 45.4% with CEEG data (specificity 64% and 68% respectively). For trending of the alpha power a simplified montage involving 10 electrodes had the best correlation coefficient with the full montage (more than 0.9 in 97% of recordings). Conclusions: CEEG can be used to predict the development of clinical vasospasm irrespective of the grade of SAH and is useful to predict recurrences that require further intervention. A simplified montage may be sufficient for this purpose.