Frequency of Alterations in Apheresis-Related Abstracts Prior to Publications as Peer-Reviewed Articles.

Abstract

High-quality evidence supporting clinical practice is lacking in apheresis. A potential source of evidence is provided by abstracts submitted to the Annual Meetings of the American Association of Blood Banks (AABB) and the American Society for Apheresis (ASFA). However, there is potential for study conclusions to be altered significantly following abstract presentations prior to publications in peer-reviewed journals. Therefore, we evaluated the discordance rate between apheresis-related meeting abstracts and their corresponding published articles. Abstracts accepted to either AABB or ASFA Annual Meetings from 2005 to 2012 and corresponding PubMed-indexed peer-reviewed articles' abstracts published prior to 9/2014 were reviewed for altered methods, results, and conclusions. When present, changes were evaluated for clinical significance. During the 8-year period, 198 out of 1152 abstracts were published as peer-reviewed articles. Of these, 36 (18.2%) presented discordant results, six of which (16.7%) were potentially clinically significant. An alteration in results (58.3%) was the leading reason for discordance. The discordance rate for ASFA abstracts was significantly higher (HR = 4.69, P = 0.0028) than the AABB ones. However, clinically significant alterations occurred more frequently among AABB abstracts (P = 0.025). Approximately 18% of meeting abstracts demonstrated alterations prior to publication in peer-reviewed journals. Given that approximately one in six changes represented clinically significant alterations, potentially affecting clinical practice, we recommend caution when modifying one's clinical practice based on abstract presentations at Annual Meetings. Future studies involving abstracts from both the International Society for Apheresis and the World Apheresis Association should also be performed.