PROGNOSTIC SIGNIFICANCE OF SYNDROMIC CLASSIFICATION IN NEWLY DIAGNOSED EPILEPSY PATIENTS

Purpose: To assess the frequency, timing, patterns and prognostic significance of change of the syndromic diagnosis in a cohort of newly diagnosed adults with epilepsy.Method: 180 newly diagnosed adults from an epilepsy centre were included and followed for 852.4 person years. For each patient, the syndromic diagnosis was made at entry according to simplified categories. Changes of the diagnostic categories were then recorded retrospectively during follow-up. A change of the syndromic diagnosis was made and dated, based on new data on seizure types and/or EEG findings and/or neuroimaging results. Using survival analysis, the cumulative time-dependent probability of change of syndromic category and of 2-year remission was measured. Differences were tested with the logrank test. A multivariate analysis was performed using Cox\u2019s proportional hazard function. Results: The sample included 104 men and 80women aged 15 through 84 years. At entry, 61 patients (33.9%) had partial epilepsies, 42 (23.3%) had generalised epilepsies, 19 (10.6%) had undetermined epilepsies, and 58 (32.2%) had isolated seizures. The syndromic diagnosis was changed during follow-up in 54 cases (30%). The cumulative probability of change was 10% at six months, 16%, 19%, and 25% at 12, 24, and 36 months. 83% of changes were in patients with isolated seizures who relapsed (45 cases). A total of 105 patients (58.3%) achieved 2-year remission. The cumulative probability of remission was similar when comparing patients with and without change of the syndromic diagnosis (log-rank 0.11; p = ns), after excluding those with isolated seizures at entry, and when adjusting for age, sex, disease duration at entry, and treatment in the multivariate analysis model. Conclusion: Except for isolated seizures, the syndromic diagnosis at entry changed infrequently (mostly during the first year) in adults and did not affect the chance of long-term remission of epilepsy