Clinical characteristics as predictors of recurrent alcohol-related seizures

To determine whether clinical data available in the emergency department can accurately predict a subset of patients at low risk of developing recurrent seizures following one or more initial alcohol-related seizures in the out-of-hospital arena.

This was a retrospective secondary analysis of data obtained from the placebo arms of two prospective, randomized trials of drug treatments for the prevention of recurrent alcohol-related seizures. Subjects with and without one or more recurrent alcohol-related seizures during the study period were compared according to the following characteristics: 1) age, 2) gender, 3) daily ethanol consumption, 4) years of ethanol abuse, 5) previous alcohol-related seizure, 6) previous seizure of other etiology, 7) temperature, 8) heart rate, 9) systolic blood pressure, 10) diastolic blood pressure, 11) respiratory rate, and 12) ethanol level. Data were analyzed with t-tests and chi-square where appropriate.

One hundred five placebo-treated patients were analyzed and 31 (30%) developed recurrent alcohol-related seizures. None of the listed characteristics were statistically different between the two groups except for the initial ethanol level. Subjects with an ethanol level higher than 100 mg/dL were less likely (0%) to develop recurrent seizures than patients with a level equal to or below 100 mg/dL (36%) (p < 0.01).

An initial ethanol level higher than 100 mg/dL was significantly associated with a low risk for recurrent alcohol-related seizures during the observation period. No other low-risk clinical characteristics could be identified.