Abstracts

Rationale: There are no multi-drug studies comparing psychiatric side effects (PSE’s) of anti-epileptic drugs (AEDs) in children. Methods: As part of the Columbia AED Database, we reviewed medical/psychiatric history, AED use, efficacy and psychiatric side effects (PSEs) in patients with epilepsy between 2-18 years of age seen at the Columbia Comprehensive Epilepsy Center between January 1, 2000 to January 1, 2005. PSEs recorded included anxiety, behavioral change, depression, hyperactivity, irritability/moodiness, and psychosis. AED-related intolerability was defined as PSEs resulting in dose change or discontinuation. Attribution of PSEs to a given AED was determined by physician notes. Non-AED predictors of PSEs were investigated using a multivariate logistic regression, and rates of AED-related PSEs (incidence and intolerability) were compared by chi-square comparisons in 166 patients newly started on any of the following AEDs at our center: acetazolamide (AZM), carbamazepine (CBZ), clobazam (CLB), clonazepam (CZP), ethosuxamide (ESX), felbamate (FBM), gabapentin (GBP), levetiracetam (LEV), lamotrigine (LTG), oxcarbazepine (OXC), phenobarbital (PB), phenytoin (PHT), topiramate (TPM), valproic acid (VPA), and zonisamide (ZNS. Past psychiatric history included the following pre-existing psychological conditions: ADHD, aggression/behavioral disturbances, anxiety, panic attacks, psychosis, suicide attempt, depression, autism and Asperger’s Disease. Results: The average incidence rate of AED-related PSEs was 17.1% and the average rate of AED-related intolerable PSEs was 15%. Non-AED predictors of PSE’s significant in multivariate analysis included presence of a past psychiatric history (25.7% with psychiatric history versus 10.7% without psychiatric history; O.R. 6.1; 95% C.I. 3.6-10.3; p< 0.0001); age of onset of epilepsy before the age of 1 year (31.6% before 1 year of age versus 14.4% older than 1 year of age; O.R. 4.1; 95% C.I. 1.7-9.7; p=0.0010) and age of onset between 1-3 years (20.5% between 1-3 years of age versus 16.4% of all others; O.R. 4.7; 95% C.I. 2.2-10.0; p=0.0001); and being on AED polytherapy (17.8% on polytherapy versus 12.5% on monotherapy; OR 2.7; 95% C.I. 1.6-4.7; p=0.0002). In patients newly started on any AED, there was a trend towards a higher than average incidence of PSEs attributed to LEV (25.5%, p<0.1). Table 1 shows incidence and intolerability rates for PSEs related to all AEDs. LEV related PSEs and rate of intolerability were significantly higher than average in patients without psychiatric history (incidence 27.3%, p<0.001; intolerability 21.2%, p<0.05), but not in those with known psychiatric disease. No other results reached statistical significance. Conclusions: Children who have an early age of onset, prior psychiatric history, or are receiving polytherapy appear to be at highest risk for developing psychiatric side effects from AEDs. LEV was associated with a higher risk of PSEs than other AEDs, but only in children without a prior psychiatric history.