Characteristics of patients initiated on the new antiepileptic drugs: a PADS study

Whereas randomized controlled trials remain a standard for evaluating and comparing efficacy and safety of the new antiepileptic drugs (AEDs), postmarketing drug research offers a useful means of comparing efficacy and safety of new AEDs. However, differences in baseline characteristics of patients in different drug groups create the potential for bias in drug comparison studies. In this study, baseline demographic characteristics of 1,386 patients initiating lamotrigine (LTG), tiagabine (TGB), or topiramate (TPM) were compared to identify patient characteristics that may influence AED use in epilepsy patients. Data were collected at 14 epilepsy centers and included medications, seizure types and syndromes, and prior adverse events. There were 402 patients in the LTG group, 725 TPM, and 259 TGB. The groups differed both in their number of concurrent AEDs (p<0.001) and in their number of prior AEDs (p<0.01). There was no difference in proportion with partial versus generalized epilepsy syndromes. The groups differed in the proportions of patients with complex partial seizures (p=0.049), primary generalized tonic-clonic seizures (p=0.01), and myoclonic seizures (p=0.03). Baseline behavioral adverse event rate was lowest in patients initiating TPM (p<0.01); LTG patients had the lowest rate of prior AED-related rash (p=0.02). There was no relationship between AED assignment and patient age, age of epilepsy onset, epilepsy duration, institutionalization status, gender, or psychiatric history. Numerous epidemiological differences were identified among patients placed on the new AEDs, including current and prior AED profiles, seizure types, and prior adverse event history. Accounting for these differences is of crucial importance because they may bias conclusions of nonrandomized post-marketing trials comparing the drugs.