The key to accurate decision-making is to use the best available evidence. Systematic reviews aim to identify and combine evidence using systematic methods to minimize bias to provide reliable data for patient care. While systematic reviews can address different clinical questions, the methodology is most developed for systematic reviews of randomized controlled trials. Such systematic reviews include all available trial evidence, enhance the precision of the estimates of treatment effects, and identify where evidence is lacking or where sufficient evidence is already available. However the term "systematic review" does not guarantee that a review covers all the available data, that the validity of included studies has been appropriately assessed, or that data have been combined appropriately in meta-analyses. Biases in systematic review include those related to identifying data (publication bias, language bias, selective reporting of outcomes) and those due to the design and conduct of trials (selection bias, performance bias, detection bias, attrition bias). Thus, readers should read a systematic review carefully before accepting its results and conclusions. This review examines the information that can be provided by systematic reviews of randomized controlled trials together with the biases that can potentially jeopardize the results and conclusions.