Our careers (at least partly) depend on our publications. The more, the better and to suit our needs we have a journal for any kind of publication. Sometimes, you read something and you may think “Hey, I have seen that before”. If the new study than confirms a previous finding, we apparently have a reproducible fact, which increases the likelihood that it is indeed true. Here is an example. Or not?
The problem: Vancomycin may be inferior to β-lactams for the empiric treatment of methicillin-susceptible Staphylococcus aureus (MSSA) bloodstream infection (BSI). The solution: A retrospective study in which empiric β-lactams are compared to vancomycin to assess clinical outcomes in patients with MSSA BSI. On April 26, 2016, Wong, Wong, Romney and Leung published their findings, see; between January 2007 and December 2014 they found in 2 hospitals 669 patients with MSSA BSI of which 255 were eligible for analysis (414 excluded). Conclusion: “Empiric β-lactams was associated with earlier clearance of bacteremia by a median of 1 day compared to vancomycin. Future prospective studies are needed to confirm our findings.” The journal had received the study on February 22, 2016.
Exactly 1 month later, on May 26, 2016, the exact same authors published a new study, see, “Prior studies suggested that vancomycin may be inferior to β-lactams for the empiric treatment of MSSA bacteremia. We assessed whether empiric therapy with β-lactams compared to vancomycin was associated with differences in clinical outcomes in patients with MSSA bacteremia.” Now they had performed a retrospective study (again between January 2007 and December 2014 in the exaxct same 2 hospitals) of 814 patients with MSSA BSI, of whom 400 met inclusion criteria. Apparently, they had somewhere found another 145 patients. This time “Empiric therapy with β-lactams was not associated with differences in all-cause mortality, recurrent infection, microbiological cure or hospital length-of-stay compared to vancomycin.” The journal had received this manuscript on November 26, 2015 (3 months before the “other” study had been submitted and 5 months before that one was published).
Isn’t that weird? The good thing is that the authors cannot be accused of self-citation. And what we need now is a meta-analysis, with “both” studies included as independent sources of evidence (meant as joke!). To confess, that’s how PhD student Denise van Hout, while working on a PICO for MSSA BSI, identified the studies.
She emailed the corresponding author of “both” studies twice, but did not get a reply. Then she contacted the editors of both journals (on January 25th 2017) and got a respons from 1: “I would like to introduce myself as the Editor of BMC Infectious Diseases. Many thanks for your email regarding the similarities between our article and the one published in Ann Clin Microbiol Antimicrob. We are aware of this issue and are currently following the Committee of Publication Ethics guidelines (http://publicationethics.org/) to resolve it. At present I cannot give you any further information but we hope that the situation will be resolved shortly.”
I have described what I saw/read, and I may have missed crucial information. If so, I sincerely apologize. If not, I think this is bad.
Denise van Hout does a PhD on the clinical epidemiology of antibiotic-resistant bacteria