Abstracts

Rationale: Temporal lobe epilepsy (TLE) is the most common partial epilepsy in adults and is an important target for genetic studies. Because TLE is prevalent, a large number of patients might be expected to be eligible for and participate in studies; this may not be the case. Here we describe recruitment strategies and enrollment results in a genetic study of TLE, discuss issues that may arise during recruitment, and provide suggestions for future studies. Methods: Potentially eligible participants were ascertained from the Columbia University Comprehensive Epilepsy Center (CEC) and The Neurological Institute Department of General Neurology (NI) through chart review of incoming patients and consultation with treating physicians. Over 1,400 adult and 302 pediatric patients have been seen at the CEC, and 384 at the NI in the past two years. In addition, over this time period, more than 900 patients were seen by another NI neurologist with a large proportion of epilepsy patients and separate records. To be eligible for the study, patients had to have non-symptomatic TLE (Mesial Temporal Sclerosis and Focal Cortical Dysplasia were not excluded), good quality MRI and EEG, age of onset ? 35, and no known disease-associated mutation. Potentially eligible patients were approached by their treating physician who obtained permission for study researchers to initiate contact. Results: 1,840 patients were screened over 8 months; 772 from the CEC, 161 from NI, and 907 from the additional neurologist. From the CEC, only 73/772 (9%) were eligible; from NI, 6/161 (4%) were eligible, and from the additional NI neurologist 130/907 (14%) were eligible to participate. From all sources, only 209 (11%) were found to be eligible. Exclusions included: not idiopathic/cryptogenic (970), wrong localization/syndrome (518), age of onset over 35 (289), non-epileptic (179), and no clear diagnosis (98). 388 patients were excluded for more than one reason. Of eligible patients, 31 (less than 2% of all screened) participated. More than half of eligible patients did not participate because they had not yet been contacted by their physician (166). Other reasons included refusal (9), living out of catchment area (2), and requesting compensation beyond protocol (1). Conclusions: Recruitment for TLE genetic studies can be time consuming and difficult. Even at a major academic surgical epilepsy center, a small fraction of patients screened participate. In our study, overall only 11% of screened patients were eligible to participate, and the most common reason for exclusion of patients was known or presumed symptomatic etiology. Stringent inclusion criteria are important in genetic studies for creating a group of well-characterized subjects with reliable phenotypes, but this may slow recruitment. Waiting for physicians to contact their patients prior to contact by research staff created the greatest obstacle to recruitment. Obtaining permission from treating physicians to contact patients directly can streamline enrollment, but is not acceptable for all physicians and patients. Broadening recruitment to additional populations may also be helpful.