What Caught Our Attention: When the New England Journal of Medicine (NEJM) publishes a correction that is more than a misspelling of a name, we take a look. When NEJM publishes a 500-word correction to the data in a highly cited article, we take notice. This study tested the effects of a drug to prevent blood loss in patients undergoing heart surgery; it’s been the subject of correspondence between the authors and outside experts. The correction involved tweaks — lots of tweaks — to the text and tables, which did not change the outcomes.
Journal: New England Journal of Medicine
Authors: Paul S. Myles, Julian A. Smith, Andrew Forbes, Brendan Silbert, Mohandas Jayarajah, Thomas Painter, D. James Cooper, Silvana Marasco, John McNeil, Jean S. Bussières, Shay McGuinness, Kelly Byrne, T.V. Chan, Giovanni Landoni, and Sophie Wallace (ATACAS Investigators of the ANZCA Clinical Trials Network)
Affiliations: Alfred Hospital and Monash University, Australia; St. Vincent’s Hospital, Australia; Royal Adelaide Hospital, Australia; Derriford Hospital, United Kingdom; Institut Universitaire de Cardiologie et de Pneumologie de Québec, Canada; Auckland City Hospital, New Zealand; Waikato Hospital, New Zealand; Chinese University of Hong Kong, Hong Kong; Istituto di Ricovero e Cura a Carattere Scientifico San Raffaele, Italy; Vita-Salute San Raffaele University, Italy;
Tranexamic Acid in Patients Undergoing Coronary-Artery Surgery (Original Article, N Engl J Med 2017;376:136-148). In the final sentence of the Trial Drug subsection of Methods (page 137), the number of patients should have been 1526, rather than 1392. … In Table 2 (page 142), in the “Surgery type” list, “Nonelective” should have been “Elective.” In the same table, footnotes regarding missing data have been added, the footnote symbols have been reordered to reflect the changes, and the data in those rows have been updated … In the first paragraph of the Bleeding-Related Outcomes subsection in Results (page 143), the final parenthetical P value should have been P=0.003, rather than P <0.001.
In the Postoperative Seizures and Other Adverse Events subsection of Results (page 143), the first sentence should have begun, “Postoperative seizures occurred in 15 patients who received tranexamic acid and in 2 patients who received placebo (0.7% vs. 0,1%; relative risk, …,” rather than “… in 15 patients (0.7%) in the tranexamic acid group and in 2 patients (0.1%) in the placebo group (relative risk, ….” In the final sentence of the same subsection, the parenthetical should have read, “8 hours [95% CI, 5 to 14] vs. 9 hours [95% CI, 6 to 16],” rather than “5 hours [95% CI, 8.0 to 14.0] vs. 6.0 hours [95% CI, 9.0 to 16.0].”
Date of Article: October 2016
Times Cited, according to Clarivate Analytics’ Web of Science: 32 (a “highly cited” paper, meaning it fell in the top 1 percent of papers in its field for its publication year)
Date of Notice: February 22, 2018
Like Retraction Watch? You can make a tax-deductible contribution to support our growth, follow us on Twitter, like us on Facebook, add us to your RSS reader, sign up on our homepage for an email every time there’s a new post, or subscribe to our daily digest. If you find a retraction that’s not in our database, you can let us know here. For comments or feedback, email us at email@example.com.