Predictors for recurrence of epileptic seizures in a general epilepsy population

The aim of our study was to identify predictors for recurrence of epileptic seizures in a large county hospital population. We identified 956 patients (18-67 years) with ICD 9 code 345 as primary diagnosis, seen at the Central Hospital of Akershus over 7 years (1987-1994). The diagnosis of epilepsy was confirmed for 696 of the patients. These were divided into two groups: (1) no seizures during the previous year (n = 485) and (2) seizures during the previous year (n = 184). To identify predictors for recurrence of seizures, we used neurologic deficit, number of AEDs used, CT-scan findings, EEG findings, aetiology, gender, age below and above 50 years and comorbidity as independent variables in a logistic regression model. In a univariate analysis, the strongest predictors for recurrence of seizures were: age above 50 years (OR = 5.2;P < 0.0001), known aetiology (OR = 1.4;P = 0.04) and use of two or more AEDs (OR = 5. 7;P < 0.0001). In the multivariate analysis age, more than 50 years (OR = 1.7;P = 0.0216) and use of two or more AEDs (OR = 5.6;P < 0.0001) were the only predictors for recurrence of epileptic seizures.