Thousands of people with heart failure live with implanted defibrillators, small devices that can shock the heart back into rhythm. Those defibrillators also send data to hospitals. Researchers in Spain say that data could serve as an early warning.
A study led by Spain’s National Center for Cardiovascular Research (CNIC) has built a model that reads those remote signals. It aims to spot when a patient enters a period of higher risk, the center said on November 19. The target is ventricular fibrillation, which the center calls the most dangerous cardiac arrhythmia.
Devices that shock after the fact
The patients in question carry an implantable cardioverter-defibrillator, or ICD. These devices detect and treat severe arrhythmias, sometimes with an electrical shock, in people at risk of sudden death. The study focused on heart failure patients whose left ventricle pumps less well than normal.
They save lives, the CNIC says, but the shocks are not without consequence. Knowing when a patient enters a more vulnerable period remains an unresolved need.
The research was published in Circulation: Arrhythmia and Electrophysiology, a journal of the American Heart Association. It suggests the device could help anticipate an arrhythmia rather than simply react once one appears.
Starting from a simple premise
The team began with a hypothesis about how the heart changes as the disease advances. As heart failure worsens, the idea goes, pressure inside the heart rises and its walls stretch further. That would disrupt the spread of the electrical impulse.
The researchers first tested this in experimental animal models fitted with an implantable defibrillator. As the disease progressed, the local electrical signal weakened and activation of the ventricles, the heart’s lower chambers, slowed.
In the animals, those changes were linked by the end of follow-up to a far higher risk of ventricular fibrillation than in a disease-free control group, the CNIC says.

Thousands of patients, 50 hospitals
The team then used those findings to train a neural network, a type of artificial intelligence that recognizes patterns. It learned from remote recordings of cardiac electrical activity from 5,089 patients across 50 Spanish hospitals.
The model was then tested on a separate group of 2,040 patients at seven centers in Spain. Those patients carried devices from a different manufacturer.
The tool tracks how the electrical signal changes over time. It also follows premature ventricular complexes, extra beats that start early in the heart’s lower chambers. It produces a risk score from 0 to 100 that updates with every transmission.
A score that rose before trouble
In patients who went on to have ventricular fibrillation, the score rose in the months before the event. In the other patients, it stayed stable.
The tool predicted arrhythmic events better than clinical and demographic factors such as age or medical history, the CNIC says.
“Unlike other algorithms, our model is grounded in the pathophysiology of heart failure, so its predictions carry a clinical meaning that physicians can interpret,” said David Filgueiras-Rama. Filgueiras-Rama is a cardiologist at Hospital Clínico San Carlos in Madrid, heads a CNIC research group and coordinated the study.
Early-warning systems are used in other areas of health too. In an unrelated example, Switzerland checks wastewater for flu, RSV and COVID, which could help spot new viruses early.
No extra device needed
Jesús Diz Díaz, a CNIC researcher, is joint first author with José María Lillo Castellano of the Fundación Interhospitalaria para la Investigación Cardiovascular. Diz Díaz noted that the data already exist.
“[T]he tool uses information that defibrillators already send to hospital, so it could be incorporated into remote monitoring platforms without the need to implant another device,” Diz Díaz said. In the same remark, the researcher said this could help bring forward check-ups or adjust treatment during periods of higher risk.
In the future, specialists could get automatic alerts when a patient’s risk rises sharply, according to the authors. They see the work as a step toward more personalized, preventive cardiology.
Heart health has also been studied over longer timescales. In an unrelated study, a Finnish study that followed 2,690 men for decades tied good heart health in their 40s to living to 90.
What comes next
The release does not describe the score in routine patient care. It presents it as a tool that could be added to remote monitoring platforms. According to the CNIC, the information needed to anticipate some potentially fatal arrhythmias already sits in the millions of remote transmissions ICDs send each day.
Partners included Queen Mary University of London and Istituti Clinici Scientifici Maugeri in Pavia, Italy. Many Spanish hospitals and research centers also took part, according to the center.
Other health research looks at what environmental exposures leave inside the body. In an unrelated study, scientists in Mexico found microplastics in the blood and urine of people in one Oaxaca community.
Photo: CNIC










