Abstracts

Rationale: Traumatic brain injury (TBI) is a major cause of acquired epilepsy (Annegers et al., 1998). Epidemiological studies (Guekht et al., 2010) revealed that TBI was the main identified cause of epilepsy in Russia. The purpose of the present study was to evaluate incidence, risk factors for early and late seizures and outcomes after TBI in the prospective study of the consecutive cohort of patients.  Methods: A prospective study of consecutive cohort of adult patients hospitalized with TBI in 2 centers in Moscow. Persons with preexisting cognitive decline and/or epilepsy were excluded. Patients were followed from admission (within 24 hours after TBI) for 2 years or until death. Clinical evaluation, CT or MRI, Glasgow coma scale (GCS) assessment were performed in the hospital; clinical, MMSE and HADs assessment -in the follow- up. Severity of TBI has been categorized by the GCS values into mild (13-15), moderate (9-12) and severe (low GCS - 8 or less) Results: 237 patients (178 men, 59 women, age 42.2 (SD 15.2) were included; 49 (20.7%) experienced seizures. 43 (18.1%) had early seizures (ES, up to 7 days after TBI), 15 (6.3%) - late seizures (LS). 9 out of 43 (20%) with ES also experienced LS compared to 0.03% of those who had no ES (p=0.0000). ES was the significant predictor of LS (OR 8.3, 95%CI 2.76 -24.9, p=0.0002). There was significantly higher proportion of patients with low GCS in the ES group (34.9% vs 13% without seizures). In univariate logistic regression low GSC values on admission were significant (p< 0.00001) risk factor of seizures. OR for any seizures was 6.6 (95%CI 2.6- 16.7), for ES - 8.4 (95%CI 3.2- 21.5), ES were associated with bilateral lesions or lesion in the left hemisphere (Kruskal-Wallis test, p =0.04). Any seizures and ES were significantly associated with contusions and the volume of hematoma. In multivariate logistic regression, the following parameters were significant predictors of any seizures: depressed skull fracture (OR 25.4, 95%CI 6.1 - 105.7, p=0.0000), midline shift (OR 19.3, 95%CI 4.5 - 82.0, p=0.0000), alcohol at admission (OR 12.2, 95%CI 3.8-39.5, p=0.0000). Penetrating injury was borderline significant (OR 3.3, 95%CI 0.9 - 11.2, p=0.06). For ES only, significant risk factors were: depressed skull fracture (OR 8.4, 95%CI 2.3- 29.9, p=0.0009), midline shift (OR 8, 95%CI 2.1-30.1, p=0.002), alcohol at admission (OR 8.8, 95%CI 2.9 - 26.1, p=0.0000). ES (but not LS) were the important predictor of mortality. 18/43 (41.9%) of patients with ES died vs 8/194 (4.1%) of patients without ES (p=0.0000). MMSE values were significantly lower and HADs anxiety and depression scale values were higher (p<0.01) in the groups of patients with seizures. Conclusions: The occurrence of seizures is a medically and functionally important complication of TBI. Seizures are frequent in patients with TBI. Early seizures are associated with increased risk of mortality. Seizures after TBI are associated with cognitive decline and depression/anxiety. References: Annegers et al, N Engl J Med 1998;338:20-4. Guekht et al., Epilepsy Res. 2010;92(2-3): 209-18