In the article appearing on page 1220 of the November 2010 issue, the meta-analysis reported included four research articles that were subsequently retracted. After removing the results from these articles and re-performing the meta-analysis, the revised results, originally in table 2, are as follows:
Two subanalyses in the Discussion also required revision. For studies published over the last 5 yr, the revised pooled weighted percentage error for pulse contour (13 studies) is 49.1%, and for transthoracic electrical bioimpedance (TEB) (5 studies) it is 50.2%. For the older software versions of the Vigileo/FloTrac device (Edwards Lifesciences, Irvine, CA), the revised percentage error is 46.5%, and for versions 1.07 and later the revised percentage error is 50.9%.
