Initial EEG in status epilepticus is helpful in predicting seizure recurrence.

OBJECTIVES: There is paucity of prospective studies evaluating the role of EEG in the prognosis of status epilepticus (SE). The present study aims at evaluating the role of clinical, EEG and radiological changes in predicting the outcome of SE.

METHODS: Consecutive patients with SE were prospectively evaluated. Status epilepticus was classified into convulsive and nonconvulsive. Consciousness at admission was assessed by Glasgow coma scale (GCS) and presence of focal neurological signs was noted. CT or MRI abnormalities were recorded. A 10 or 21 channel EEG record wcs obtained 1 hour after therapy. The patients were treated with IV sodium valproate or phenytoin according to a fixed protocol. The EEG changes were categorized into discrete, merging, continuous, periodic pattern, slowing and normal. Outcome was assessed by seizure control at 1 hour, sustained seizure control for 24 hours and death.

RESULTS: There were 70 patients with SE whose age ranged between 1 and 85 years, 26 were females and 11 below 15 years. The underlying etiology was CNS (Central nervous system) infection in 35, stroke in 11, metabolic encephalopathy in 17, drug default in 2 and no apparent cause was evident in 5 patients. Convulsive SE was present in 65 and nonconvulsive in 5. Cranial CT and MRI were abnormal in 36, and MRI revealed additional abnormality in 3 patients. The EEG at 1 hour was abnormal in 51 out of 53 patients and revealed discrete pattern in 4, merging in 5, continuous in 5, periodic in 6, multiple patterns in 2 and slowing in 29 patients. In 7patients with convulsive SE, clinical seizures were controlled but EEG showed ictal pattern. Clinical seizure recurred in 38 patients within 24 hours and 15 of them had ictal discharges at one hour EEG. Response to antiepileptic drugs correlated with underlying etiology, duration of SE and associated medical illness. 19 patients died but not due to SE per se. Death did not correlate with EEG abnormality, however, 1-hour EEG abnormality predicted seizure recurrence.

CONCLUSION: EEG is useful in monitoring SE and its abnormality at 1 hour predicts seizure recurrence within 24-hour.