“Major error” forces retraction of ghrelin study

Try as we may, we can’t cover every retraction in real time. But on the principle that late is better than later, here’s one from 2012 that we’ve been meaning to get to.

neuroreportThe journal NeuroReport has retracted a 2011 article by a group of researchers who evidently discovered a fatal flaw in one of their figures.

The article, “Ghrelin prevents neuronal apoptosis and cognitive impairments in sepsis-associated encephalopathy,” by a team of intensivists from Zhejiang University in Hangzhou, China, purported to find that ghrelin, which stimulates appetite, appeared to have something of a protective effect against the ravages of sepsis in rat brains. It has been cited three times, according to Thomson Scientific’s Web of Knowledge, including once by the retraction.

As the abstract stated:

The present study explored the effect of ghrelin in protecting neurons from apoptosis in sepsis-associated encephalopathy. Ghrelin (100 nM) increased the cell viability treated with lipopolysaccharide (1.0 μg/ml, 24 h). The expression of p-Akt and Bcl-2 were decreased and caspase-3 increased both in lipopolysaccharide-treated primary hippocampal cultures and in the cecal ligation and perforation model, which were alleviated in the presence of ghrelin. In vitro, the protecting effect of ghrelin was almost abolished by the Akt inhibitor, SH-5. In vivo, the cecal ligation and perforation rats exhibited emotional, learning, and memory deficits. Administration of ghrelin attenuated the cognitive deficits significantly. These results indicate that ghrelin alleviates neuronal apoptosis and subsequent cognitive impairments in sepsis-associated encephalopathy through the Akt pathway.

Not so fast, according to the notice:

The authors have retracted this article.

The authors state: A major error was found according to recent data, which was repeated in a larger number of subjects. The error occurred in behavior study performance in Fig. 4b. The authors’ latest data revealed that application of ghrelin did not rescue the emotional memory evaluated by the inhibitory avoidance test. The previous protocol of behavior study led to a misleading conclusion. In the previous work, the number of rats did not seem to be large enough to evaluate the cognitive influence or the inhibitory avoidance test was not applied to evaluate the cognitive deficits in rats who had undergone cecal ligation and perforation surgery. The decision has been approved by all authors and the medical center where the experiment was carried out.

The authors are planning to conduct the behavior study using several protocols to reassess the cognitive defects.

The original study used 110 rats. We’re not sure how many more would have been necessary for a robust finding. And, wouldn’t that kind of power calculation have been something identified in the review process? We’ve contacted the journal and will update this post if we learn more.

0 thoughts on ““Major error” forces retraction of ghrelin study”

  1. Reblogged this on lab ant and commented:
    N=110 the effect must be extremely small to be not reliably detected…I think they had some major flaw in their set-up whatever that might have been. But at least they recognized it and did the right thing!

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.