Retraction Watch

Tracking retractions as a window into the scientific process

Another Hattori retraction over reused figure from cardiology pub

without comments

Another paper in Diabetologia by Yoshiyuki Hattori has been retracted for image duplication, marking the second of his articles in the journal to be pulled for that reason.

The notice for the article, “A glucagon-like peptide-1 (GLP-1) analogue, liraglutide, upregulates nitric oxide production and exerts anti-inflammatory action in endothelial cells,” states:

This article has been retracted by the Editor-in-Chief of Diabetologia following the discovery of redundant publication (part of Figure 2c was previously published in Cardiovascular Research; doi:10.1093/cvr/cvn226).

The original Cardiovascular Research article, “Cilostazol inhibits cytokine-induced nuclear factor-κB activation via AMP-activated protein kinase activation in vascular endothelial cells,” was published online in August 2008. It has been cited 17 times, according to Thomson Scientific’s Web of Knowledge, while the Diabetologia version has been cited eight.

Hattori, of Dokkyo University School of Medicine, now has at least five eight retractions stemming from the reuse of images, according to our count.

We’ll say this for Springer, which publishes Diabetologia: they make searching for retractions remarkably easy. In fact, they have a passive search function that does it for you: while you’re reading about one paper it calls up others and displays them in a window alongside the main paper. Bravo for that! Elsevier appears to have something similar, although it doesn’t appear to throw up retractions as often, at least in our limited sampling.

Update, 11:30 a.m. Eastern: A commenter points out that we missed one of Hattori’s retractions and failed to count a few from our other posts. (That’s why we love our readers, who keep us honest). The neglected article, “Statin blocks Rho/Rho-kinase signalling and disrupts the actin cytoskeleton: relationship to enhancement of LPS-mediated nitric oxide synthesis in vascular smooth muscle cells,” appeared in Biochimica et Biophysica Acta in 2004. Although we’d noted it in an earlier post as potentially problematic, at the time it had not been retracted. Here’s the notice:

The editors of the Journal of Cardiovascular Pharmacology and Biochimica et Biophysica Acta have discovered duplication of data in two manuscripts: “Disruption of the actin cytoskeleton up-regulates iNOS expression in vascular smooth muscle cells” (J. Cardiovasc. Pharmacol., 43 (2004) 209–213) and “Statin blocks Rho/Rho-kinase signaling and disrupts the actin cytoskeleton: relationship to enhancement of LPS-mediated nitric oxide synthesis in vascular smooth muscle cells” (Biochim. Biophys. Acta 1689 (2004) 267–272, doi:10.1016/j.bbadis.2004.04.006). We have requested an explanation from the authors, and an acceptable response has not been received. In the interests of scientific integrity, the editors of both journals have elected to retract both articles.

Hat tip: Commenter canoinve

Comments
  • WHY December 20, 2011 at 10:51 am

    Here, you can see what was wrong with the Figure. Keyword would be left-to-right inversion.
    http://blog.m3.com/Retraction/20111220/2

  • WHY December 20, 2011 at 11:18 am

    Actually this is No8.

    Diabetologia. 2005;48:1066-74.
    Metabolism. 2005;54:482-7.
    J Cardiovasc Pharmacol. 2004;43:209-13.
    Biochim Biophys Acta. 2004;1689:267-72.
    Cardiovasc Res. 2009;81:133-9.
    Cardiovasc Res. 2004;63:31-40.
    Cardiovasc Res. 2002;54:649-58.

  • V December 20, 2011 at 6:59 pm

    It’s not just that they re-used data, they basically made it up (by taking old pictures of unrelated gels and modifying them). The data was used in different experiments to “prove” different results about different substances.

    Some of the retraction notices make this seem like they just re-used graphs without proper copyright clearance. This seems like more than that.

  • conradseitz December 21, 2011 at 12:18 pm

    If I’m not mistaken, liraglutide is the brand name Victoza, an injection for type 2 diabetes, which is unfortunately not as popular as the earlier injection, Byetta (exenatide) (clinically a great drug for weight loss in obese diabetics) but is pushed by drug reps to practicioners as “just as good” and “only once a day (vs BID)”. How much more of the research for liraglutide is suspect?? This is really hitting close to home for medical practicioners. Like the research you depend on for treating and advising your patients is built on quicksand.

    • Marco December 21, 2011 at 1:25 pm

      You are not mistaken on the brand name, but you are rather fast on the trigger when you cast suspicion on liraglutide just because someone reused a figure without telling.

    • conradseitz December 22, 2011 at 12:42 pm

      Actually, I didn’t have as good results with Victoza as with Byetta, leading me to think it was a “me too” drug (both were approved before I retired.) (Me too, meaning a drug that capitalizes on the appeal of another drug.) Then I started reading this blog, and began to wonder how good was the science behind Victoza. True that the FDA approval of Victoza didn’t depend on the work of Hattori. But, also, he didn’t just reuse a figure; see the comment of V, above, who suggests he made it up.

      • Marco December 22, 2011 at 1:54 pm

        You are free to think what you want, but I still don’t see the connection. Hattori has had nothing to do with the official clinical studies that led to marketing approval.

  • conradseitz December 22, 2011 at 4:35 pm

    There is no connection. Purely coincidence that liraglutide hasn’t turned out to be such a great drug clinically. On the other hand…four fingers and a thumb.

  • Post a comment

    Threaded commenting powered by interconnect/it code.