Abstracts

Rationale: Previous studies suggested that patients with epilepsy tend to have poorer cognitive function, personal or social behavior and quality of life (QOL) than normal subjects. But there are limited data presented how much recurrent seizures have influence on cognition, behavior, and quality of life in newly diagnosed epilepsy. We investigated seizure effect to cognitive function, behavioral changes, and quality of life during 48 weeks treatment with lamotrigine (LTG) or carbamazepine (CBZ) monotherapy in patients with newly diagnosed or untreated partial epilepsy. Methods: Newly diagnosed and untreated partial epilepsy patients randomized to receive LTG or CBZ monotherapy. Study duration consisted of 8-week dose titration period and 40-week maintenance period. Patients categorized in seizure-free or not-seizure-free subgroup during 40-week maintenance period. Neuropsychological test, symptom check list-90, and QOLIE-31 assessed at baseline, 16 weeks and final 48 weeks. Primary outcome measured by group-by-time interaction to evaluate seizure effect between subgroups at baseline and 48-week endpoint. Secondary outcome measured by changes of cognitive and QOL scores to seizure effect between subgroups. Linear-mixed model analysis applied to estimate time-variable change of scores combined with seizure effect. Results: A total of 110 patients were eligible and 73 completed the 48-week study (LTG, n=39, 68.4%; CBZ, n=34, 64.2%). Our primary outcome revealed with group-by-time interaction in two subscales of California Verbal Learning Test (CVLT) [serial clustering index (p<0.001) and recognition (p=0.020)], overall QOL score (p=0.020) and two subscales of QOLIE [seizure worry (p<0.001), social score (p<0.001)]. Seizure effect became more significant in performance with serial clustering, and narrowed differences between subgroups in performance with recognition of CVLT. Seizure-free group revealed significant improvement of overall QOL score, seizure worry and social score in about of group-by-time interaction. Secondary outcome showed intergroup differences measured by Rey Complex Figure Test (RCFT) copy (p=0.035), correct response of Stroop color (p=0.029), Stroop Word/Color Interference Task (p<0.001), two subscales of Controlled Oral Word Association Task (COWAT) [semantic fluency task (p<0.001), phonemic fluency task (p=0.042)], three subscales of CVLT [short delay free recall (p=0.017), long delay free recall (p=0.026), long delay cued recall (p=0.023)], and one subscale of QOLIE (energy, p=0.021) with better performance in seizure-free group.