Abstracts

Rationale: There is growing recognition of the value of capturing patient-reported outcomes (PROs), but while PROs have been widely incorporated into epilepsy clinical trials, little is known about their usefulness in routine clinical practice. The goals of this study were to report the experience of routine data collection of PROs in an adult epilepsy clinic and to understand how patient-reported cognitive side effects of antiepileptic drug (AED) treatment relate to number of AEDs and specific AED used.Methods: Adult patients presenting to the outpatient epilepsy clinic of a tertiary comprehensive epilepsy center from January to February 2015 who completed the 24-item Aldenkamp-Baker Neuropsychological Assessment Schedule (ABNAS) were included (N = 457). We analyzed the effect of number of AEDs prescribed and specific AED prescribed.Results: The study population (N=457) had the following demographic characteristics: 88% Caucasian, 8% African-American, 60% female, median age 42 years. In terms of number of AEDs prescribed, 14% were not prescribed any AEDs, 48% were prescribed one AED, 25% two AEDs, 10% three AEDs, and 3% four or more AEDs. Of those who started the ABNAS questionnaire, 95% completed it; 18% of patients required help completing the questionnaire. The overall rate of patient-reported cognitive impairment, defined as ABNAS score ≥ 15, was 44% (N=200). The overall ABNAS score was not significantly correlated with the number of AEDs prescribed when ABNAS was analyzed as a continuous variable, however when ABNAS score was dichotomized into low (< 15) and high (≥ 15) scores, high ABNAS scores did correlate with number of AEDs prescribed. Among patients not taking any AEDs, 31% reported cognitive impairment. Among patients prescribed one AED (N=220), 43% reported cognitive impairment, and the AEDs with the highest rate of patient-reported cognitive impairment were topiramate (TPM, 76%), lacosamide (LCM, 75%), and oxcarbazepine (58%). The AEDs with the lowest rates of subjective cognitive impairment were gabapentin (GBP, 22%), levetiracetam (LEV, 35%) and lamotrigine (LTG, 38%). Among patients prescribed two AEDs (N=114), the most common combinations were LEV and LCM (N=15) followed by LEV and LTG (N=13) with rates for patient-reported cognitive impairment of 40% and 23%, respectively.Conclusions: These results support the use of the ABNAS as a screening tool to detect AED-related cognitive impairment. As expected, TPM has among the highest rates of cognitive impairment and GBP and LEV have among the lowest. Interestingly the combination of LEV and LTG had a lower rate of cognitive impairment than LEV alone. Screening for patient-reported cognitive side effects in routine clinical practice provides valuable information for AED management.