Liver spots: Hepatology journal yanks research letter on caffeine-fibrosis link, saying authors falsified data

Work from a prolific father-son team of liver researchers in Germany has come under scrutiny after accusations that they falsified data in a 2009 letter to the editor that appeared in the Journal of Hepatology.

The letter, retracted in the September issue (after an online notice in June), referred to a 2008 article in the journal by Axel Gressner, his son Olav, and their colleagues at University Hospital in Aachen in which the authors reported that doses of caffeine might be an effective treatment for liver fibrosis, scarring of the organ that results from chronic ailments such as cirrhosis or hepatitis.

Epidemiologic evidence has suggested that people who drink coffee are somewhat protected from liver fibrosis—although some experts dispute the purported connection—and the German group claimed to have been among the first to find a plausible molecular mechanism for the link. Their November 2008 paper on the subject has been cited 16 times, according to Thomson Scientific’s Web of Science, a hefty number for just 22 months.

In their follow-up letter, they went a step further, stating that injecting rats with caffeine blocked the expression of a key protein associated with growth of connective tissue necessary for the formation of liver scars.

But the evidence backing the letter appears to be far weaker than the researchers initially let on. According to the retraction notice, the authors doctored their manuscript—which had been accepted without any modifications, the smoothest sailing a scientific paper can hope for—after receiving the okay for publication. They

made major modifications at the stage of the galley proofs without informing the Editor or the Associate Editor.

Said changes?

1. The number of rats per group was reduced from 5 to 1; therefore, instead of a total of 20 rats being studied, a total of only 4 rats were studied.

2. Throughout the manuscript sentences have been changed from “rats” to “rat”.

3. The standard deviation values were removed at the proof stage and more importantly some results were completely altered.

4. A full paragraph was added at the end of the paper, which was not present in the original version submitted to the journal.

That left the editors with the inescapable conclusion that

the authors have presented falsified results which misled the Editors. As a consequence to the aforementioned breaches, we have decided to retract this Letter to the Editor.

Word of the manipulated manuscript evidently circulated through the liver research community long before the retraction appeared. Keith Lindor, a Mayo Clinic physician and editor of Hepatology, says his publication launched its own investigation into studies by the Gressners in early 2010. However, the journal determined that no retractions were necessary.

Scott Friedman, a leading liver specialist at Mount Sinai Medical Center in New York tells that the retracted study was “minor” and would not affect patient care. What the fallout will be for the Gressners remains unclear, adds Friedman. Whether the incident involved intentional misconduct, sloppiness or naivete isn’t certain. And while Olav Gressner is a relative newcomer to  liver research, his father has been a “highly respected” member of the field for three decades whose integrity as a scientist has not been impugned, Friedman notes.

Neither the journal’s editor nor the Gressners responded immediately to Retraction Watch’s requests for comment.

0 thoughts on “Liver spots: Hepatology journal yanks research letter on caffeine-fibrosis link, saying authors falsified data”

  1. Reminds me of the old joke:

    “An experiment was conducted on rats. Half of the rats were given treatment and the other half were controls. The treatment group survived. The other rat died.”

Leave a Reply to Neuroskeptic Cancel 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.