We performed a meta-analysis of studies on benign epilepsy of childhood with centrotemporal spikes (BECT) to ascertain whether clinical characteristics and outcome can be stated unequivocally. Using the Index Medicus and Medline CD+, we identified 525 publications. After applying the criteria of the International League against Epilepsy (ILAE) for BECT, 32 publications on 2,561 patients remained. After correction for inclusion bias and multiple publications on the same patient groups, 13 cohorts, comprising a total of 794 patients, were included. The aggregate proportional remission was 0.977; hence, no factors influencing outcome could be identified. Age at onset ranged from 3 months to 14 years, age at last seizure ranged from 3 to 18 years. A Kurtzke survival analysis of proportions of children in remission by age was performed; at an older age, the proportion of patients in remission was 0.9997. Publications had highly heterogeneous methodologies and population characteristics; we conclude that current knowledge on BECT has been determined mainly by retrospective studies of biased cohorts, and that the uniformity per se of BECT as an epileptic syndrome may be, at least in part, a result of selection bias. We conclude that early prediction of seizure outcome in a new patient with BECT can not be given with certainty. Prospective, population-based studies are needed to delineate the clinical and EEG characteristics of this syndrome.