 |
developing dangerous heart rhythm problems during exercise. This new research, led by the University of Surrey, highlights how short electrocardiogram (ECG) recordings could be used as a practical screening tool in everyday training.
Cardiac arrhythmias (abnormal heart rhythms) are very common in athletic horses. In many cases they are harmless, but some can reduce performance or, in rare cases, lead to collapse or sudden death during exercise. One of the biggest challenges for veterinarians is identifying which horses are at risk before serious problems occur.
Traditionally, detecting these issues requires ECG monitoring during high-intensity exercise. However, this can be time-consuming, expensive, and not always practical. The new study suggests that useful information can instead be obtained from short ECG recordings taken at rest or during light exercise, such as a warm-up trot.
The research team analysed ECG data from 110 Thoroughbred and Standardbred racehorses in the United States during routine training. These recordings were collected using portable (ambulatory) ECG devices, allowing horses to be monitored while moving freely. From these data, the researchers selected 60-second segments where the signal quality was good and the heart rate was stable, typically between 60 and 100 beats per minute.
Rather than simply looking for obvious irregular beats, the researchers used advanced mathematical techniques to assess the “disorderliness” of the ECG signal. This is known as non-linear analysis and includes methods such as entropy and complexity measurements (for example, Lempel–Ziv complexity and Shannon entropy). These techniques examine how predictable or irregular the heart’s electrical activity is, even when the rhythm appears normal (sinus rhythm).
The idea behind this approach is that subtle changes in the ECG signal may indicate an underlying tendency to develop arrhythmias later, especially during intense exercise. In other words, the heart may show early warning signs before any abnormal beats are visible.
To test this, the team applied six different algorithms to the ECG data and compared their ability to distinguish between horses that did and did not develop arrhythmias during exercise. The best-performing method achieved an “area under the curve” (AUC) of 0.86. For context, an AUC of 0.5 indicates no better than chance, while 1.0 represents perfect accuracy. Therefore, a value of 0.86 suggests good diagnostic performance.
An important finding was that the test was particularly effective at ruling out horses that are unlikely to be at risk. This means it could be used as a first-line screening tool: horses with a negative result could continue normal training, while those that test positive could be referred for more detailed testing, such as a full exercising ECG.
This approach builds on earlier work showing that similar ECG analysis methods can detect atrial fibrillation, another common arrhythmia in horses. Because these methods work at different heart rate ranges, it may be possible in the future to screen for multiple heart conditions during a single, low-intensity session.
Overall, this study demonstrates that short, simple ECG recordings—combined with advanced data analysis—could help identify at-risk horses earlier. This has important implications for improving both performance monitoring and the safety and welfare of racehorses.
For more details, see:
Vadim Alexeenko, Hamid Tavanaeimanesh, Freya Stein, Jenifer Gold, Lauren Hughes, Molly McCue, Celia Marr, Sian Durward-Akhurst & Kamalan Jeevaratnam
Detection of exercising ectopic atrial and ventricular beats using non-linear analysis of clinically normal racehorse electrocardiograms at rest or low-intensity exercise.
Sci Rep (2026).
