Seminal NEJM Ambulatory BP Study Retracted:
Now What?

January 30, 2020

After a major study linking ambulatory blood pressure monitoring to all-cause mortality was retracted by the New England Journal of Medicine , more perspective is emerging on what the clinical fallout of the retraction should be.

The study, as reported by Medscape at the time, was said to be the largest ever conducted of ambulatory blood pressure monitoring and showed that measuring blood pressure over a 24-hour period predicts all-cause and cardiovascular mortality much better than does office or clinic measurement. 

The retraction, in the form of a letter to the NEJM yesterday from the study authors, consists of just one sentence.

"Because we have identified inaccuracies in the analytic database and data analyses underlying 'Relationship between Clinic and Ambulatory Blood-Pressure Measurements and Mortality,' which was published in the April 19, 2018, issue of the Journal, we wish to retract the article," it states.

The authors are a Spanish group led by José R. Banegas, MD, and Luis M. Ruilope, MD, Universidad Autónoma de Madrid in Spain, along with Bryan Williams, MD, from University College London, United Kingdom.

Williams provided more information on the retraction to Medscape Medical News.

"This study involved linkage of the world's largest ambulatory blood pressure monitoring (ABPM) data set from the Spanish registry with clinical outcome data to examine the relationship between ambulatory BP and mortality," he said.

"The database development and statistical analysis for this study was conducted by Dr Banegas (the first and corresponding author for the paper) and his statistician, at the Autónoma University in Madrid," Williams noted.

"While evaluating the data when responding to readers' correspondence about the study, some errors in the original data tables were identified, which obviously concerned some of us," he said, "so we decided to undertake a review of the core database and repeat the entire analysis with an independent statistical team. Subsequent evaluation by this team, working with some of us, identified errors in the core database linking the ABPM data to mortality and this had led to errors in the resultant data tables.

"We immediately informed all authors and the NEJM of our concerns and concluded that the original data analysis was unreliable and notified the journal that the original paper should be retracted," Williams added.

The concerns relate to the errors in linkage of mortality data to the ABPM records in the core database developed by the statistical team, he said, "and we have subsequently asked for an investigation into how the errors in the core database created by Dr Banegas and his statistician have arisen, and this is in its early stages and ongoing."

In the meantime, working with the new independent statistical team, they are in the process of validating the new data linkage of ABPM records and mortality data from the core data, "and plan to repeat the analyses and reissue this important data as promptly as possible," Williams added.

"I have been directly involved with a small core team of the original authors and the independent statistical team in leading this review of the data, and this is an up-to-date summary of events," Williams said.

Seminal Paper

Commenting for Medscape Medical News, Raymond Townsend, MD, author of an editorial accompanying the original NEJM paper, said there had been some "rumblings" about the data since its publication.

Townsend, who is director of the hypertension program and professor of medicine at Perelman School of Medicine at the University of Pennsylvania in Philadelphia, said, "This was a seminal paper — a huge study with data from 67,000 individuals — from a highly respected group that has been much cited since its publication."

"There was a lot of correspondence after its publication that made some interesting points, particularly about possible methodological issues that may have limited the generalizability of the findings," he said. "I have also heard from several US experts who felt something was not quite right when the data was scrutinized, so I was half expecting something to happen, but thought it may be in the form of a correction to a table or a revision of some sort. A retraction of the whole paper is more than I expected."

Still, Townsend said, "Bryan Williams is one the top names in hypertension in Europe. I have every confidence in him and I believe if he is involved, it will be sorted out with credibility."

Further, Townsend does not think the retraction will have any major effects on clinical practice.

"ABPM is a very solid tool," he said. "There is nothing better to determine the presence or absence of hypertension, however it is defined. This will not change. It will, however, be important to get this data corrected and know what the actual findings are. It is too important an issue to leave up in the air."

Also commenting for Medscape Medical News, John William McEvoy, MB BCh, professor of preventive cardiology at the National University of Ireland, Galway, said, "This is a surprising development and disappointing I am sure for the authors. Any retraction from NEJM is both unusual and also noteworthy because the paper will likely have been reviewed intensely prior to publication. However, mistakes can happen and the respected group of authors appear to have promptly taken the appropriate corrective action."

"Thankfully, in the particular case of this paper, the body of evidence in favor of ABPM is also so strong that the retraction (and hopefully revision) of this one paper is unlikely to lead to changes in clinical guidelines or practice," McEvoy noted. "Nonetheless, providing accurate results in a revision will be important so that our overall understanding of ABPM is accurate."

"This unfortunate episode also highlights the critical importance of postpublication peer review and letters to the editor, which in this case resulted in concerns that some of the results reported in the original paper were unanticipated (specifically the risk of patients treated to a normal BP was not higher than the risk of persons with naturally normal BP — it should have been higher). These concerns led to this retraction, which hopefully will be followed by a revised publication," he added.

"A Very Big Deal"

Another hypertension expert, William White, MD, told Medscape Medical News that "a retraction of a paper from the NEJM is a very big deal. It is embarrassing for the both the authors and the journal."

White is professor emeritus at the Calhoun Cardiology Center, University of Connecticut School of Medicine in Farmington, and a past president of the American Society of Hypertension.

"The NEJM are remarkable in their review process — the statistical editors really are very tough. They usually go through all the data meticulously." he said.  

"This particular paper was of great interest to everyone in the field. These investigators have a very good reputation, having developed this registry database and published many papers from it. However, I did notice some curiosities in some of the tables in the paper — some of the data didn't seem statistically possible and I wrote to the investigators about it. I think several others did too," White commented.  

"I don't think this will undermine the field of ABPM," he added. "There are quite a number of prospective cohort studies that have shown this technology to be better at predicting cardiovascular outcomes than office or home measurements. I would continue to promote this as the best method of measuring blood pressure."

These sentiments were further echoed by Michael Weber, MD, professor of medicine at the SUNY Downstate College of Medicine in New York City, and editor-in-chief of the Journal of Clinical Hypertension.

"I know several of these authors personally and by reputation and have the very highest regard for the integrity and quality of their work," Weber said. "Therefore, I must conclude that the inaccuracies now discovered in their article resulted from inadvertent errors of analysis. Furthermore, I believe that the action of these scholars in now retracting the paper reflects most positively on their uncompromising commitment to the highest standards of scientific rigor.

"There is ample evidence beyond this article that ambulatory blood pressure monitoring has certain advantages over conventional office blood pressure measurements in the diagnosis of hypertension, and I expect that current guidelines advocating increased use of ambulatory monitoring for accurately classifying patients will not be changed," he added. 

N Engl J Med. Published online January 29, 2020. Correspondence | Retraction

For more Medscape Neurology news, join us on Facebook and Twitter


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.