The publisher of a European Heart Rhythm Association journal is investigating a paper that lists the incorrect registration number for a clinical study of a widely used but controversial heart implant.
The paper, published in EP Europace in May, described a prospective study of the Watchman FLX, from Boston Scientific, for a procedure called left atrial appendage occlusion (LAAO) in 400 people with abnormal heart rhythms. The authors concluded the device was “feasible, safe, and resource efficient, achieving high procedural success and low complication rates.”
Cardiologists have warned of caveats with the device and the LAAO procedure itself, mainly that early studies found LAAO caused a “significantly higher rate of a primary composite endpoint that included stroke, systemic embolism, cardiovascular/unexplained death, and major bleeding.” A study earlier this year found LAAO with a Watchman device increased rates of stroke and other complications when compared to oral medications.
The clinical trial registration in the Chinese Clinical Trial Registry — ChiCTR2400087516 — published in the paper corresponds to a randomized controlled trial on high-power atrial fibrillation radiofrequency ablation, which doesn’t involve LAAO or the Watchman device. While the Europace paper claims the trial was prospective, the entry in the cited registration states the study was retrospective.
A separate clinical trial registration in the Chinese Clinical Trial Registry, ChiCTR2400085158, seems to describe the methods reported in the study.
Authors Qian Tang and Wei Li, both of Guizhou Medical University, are listed on the clinical trial record included in the study. Neither responded to our request for comment regarding whether the trial was included in error and if there is a corresponding registration for the described Europace study. Co-corresponding author Runze Huang also did not respond to our email. According to a 2023 news article, lead author Wei Zhou is the second surgeon in China to become a “Watchman FLX Global Training Expert,” a role that is not included in the study disclosures.
The article acknowledged funding from Guizhou Medical University, but the clinical trial registration likely corresponding to the study stated the funding came from Boston Scientific.
John Mandrola, a cardiac electrophysiologist at Baptist Health in Louisville, Kentucky, told Retraction Watch the fact neither the editors or the peer reviewers picked up on the incorrect registration number “speaks to the poor attention to detail on the part of the authors and reviewers.”
“Most critically however, the study itself isn’t really a study. It’s more of a report on ‘look what we did,’” said Mandrola, who has been critical of the evidence for the device. (Disclosure: Mandrola is a frequent contributor to Medscape, where one of our cofounders works.) “It hardly needs nor likely has a real registration.”
A sleuth who did not provide their name flagged the trial to us earlier this month. They said they notified the journal of their concerns on June 11 and said Europace “subsequently indicated that the matter would be investigated.”
Giuseppe Boriani, editor-in-chief of the journal, told us the ethics committee at Oxford University Press, which publishes the journal, “is evaluating this issue and will provide a [sic] feedback soon.” Boriani, of the University of Modena and Reggio Emilia in Italy, also forwarded us a September 24 email he sent to the authors requesting clarification from the authors following our request for information but did not respond when we asked whether the journal knew about the incorrect trial before we contacted him.
The wrong registration number might be “a simple mix-up,” Mandrola said. “I struggle to understand why Europace, a pretty decent journal in our field, published the study.”
Like Retraction Watch? You can make a tax-deductible contribution to support our work, follow us on X or Bluesky, like us on Facebook, follow us on LinkedIn, add us to your RSS reader, 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 protected].
Even if ChiCTR2400085158 is considered the intended registration for the ROSE-FLX study, there appear to be several important discrepancies between the registry entry and the published article.
First, ChiCTR2400085158 describes a single-center study, whereas the published ROSE-FLX article describes the study as a prospective, multicenter registry.
Second, the eligibility criteria in the registration do not appear to match those reported in the published article. The inclusion criteria and study population described in the registry differ from those presented in the manuscript.
Third, the planned sample size in the registration is also different from the number of patients reported in the publication. These discrepancies raise the question of whether ChiCTR2400085158 can actually be regarded as the prospective registration for the study reported in the article.
This distinction is important because simply replacing the currently cited registration number (ChiCTR2400087516) with ChiCTR2400085158 may not resolve the underlying registration issue. If the registered protocol differs materially in study design, recruitment setting, eligibility criteria, and sample size from the published study, clarification would still be needed regarding whether the published ROSE-FLX cohort was prospectively registered as reported.