Background: Systematic reviews (SRs) of randomized trials are the highest level of evidence since they summarize the best current research on specific questions. Therefore, it is crucial to keep them up-to-date, but even more important, is to assure they are including the most recent evidence at the time of publication. To our knowledge, little is known on how outdated SRs might be at publication. Objectives: To describe how updated SRs are at their time of publication. Methods: We analyzed all new SRs in Issue 1, 2008 of the Cochrane Database of Systematic Reviews (CDSR) and all SRs included in the ACP Journal Club during 2008. We excluded updates or analyses of previous SRs and SRs withdrawn from CDSR Issue 1, 2008. We extracted published data on: a) day/month/year of last search in MEDLINE, EMBASE and Cochrane Library; and b) publication date (day/month/year). For analysis we used the latest search date in any of the three databases. Results: We reviewed 217 SRs. Fourteen met the exclusion criteria and 203 SRs were included for analysis (77% Cochrane and 23% non-Cochrane SRs). The shortest period of time since the most recent search to the publication date was 22 days whereas the longest was 1606 days ( > 4 years). Mean time since last search to publication was 339 days, with no difference between Cochrane and non-Cochrane SRs (341 vs. 332 days). Twenty-five per cent of the SRs were published more recently than six months from last search, 45% between six months and one year and 30% were published later than one year from the last search. Conclusions: Close to one third of SRs are published after one year from latest search. Although there is no defined period of time, efforts should be made to avoid publishing SRs with searches over one year old.