The article, “Incident atrial fibrillation hazard in hypertensive population: a risk function from and for clinical practice” by Alves-Cabratosa et al (Alves-Cabratosa L, García-Gil M, Comas-Cufí M, Ponjoan A, Martí R, Parramon D, Blanch J, Ramos R), has been retracted. The article was published in Hypertension (2015;65:1180–1186. doi: 10.1161/HYPERTENSIONAHA.115.05198), online on April 6, 2015 and in print in the June 2015 issue of the journal.
The authors detected a mistake in the population definition as follows:
The authors agreed with one of the manuscript reviewer’s in further excluding participants under treatment with antiarrhythmics (class I and III) and selective calcium-channel blockers with direct cardiac effects, as surrogates of known atrial fibrillation at entry. But instead of eliminating those participants under such surrogate treatment, at entry, previously to the baseline, they eliminated those under surrogate treatment from the baseline onwards, and regardless of whether they had new-onset atrial fibrillation during follow-up or not.
As a result:
The new-onset atrial fibrillation incidences were reported inadvertently lower than they really are.
The model has been slightly modified. A model with 6 variables was originally presented: age, weight, total cholesterol, heart failure, valvular heart disease, and antihypertensive drugs other than agents acting on the renin-angiotensin system. After correction of the population definition the model contains 5 variables: age, weight, chronic obstructive pulmonary disease, heart failure, and valvular heart disease. That is, chronic obstructive pulmonary disease enters the model, whereas total cholesterol and antihypertensive drugs abandon it.
Unfortunately, since the confusion was at the level of population definition, data values in figures and tables have suffered slight modifications as well. The results have not suffered clinically relevant modifications, and the main message also remains the same. However, due to the number of changes required, this article is retracted.
The authors regret the mistake in population definition and the possible confusion that this has caused.
- © 2016 American Heart Association, Inc.