Listen to Your Heart

Premature atrial contractions (PACs) and premature ventricular contractions (PVCs) are common heart rhythm disorders that affect millions of people worldwide. According to the American Heart Association, PVCs are the most common type of arrhythmia, affecting up to 70% of individuals at some point in their lives. PACs are also prevalent, with studies suggesting that up to 50% of adults experience them.

Although most PVCs and PACs are not life-threatening and may not require treatment, research suggests that they may be an indication of underlying heart conditions that can lead to more severe health complications. Individuals with PVCs or PACs are more likely to develop other cardiovascular diseases, such as heart failure and stroke, compared to those without these conditions. The risk is highest for those with frequent PVCs or PACs, which are those that occur more than 30 times an hour.

Since premature heart contractions can present long before the onset of disease, it is important that they be detected early. This allows higher-risk cases to be more effectively treated through medication and lifestyle changes that may improve the individual’s long-term health outcomes.

Unfortunately, monitoring for these conditions can be very cumbersome, time consuming, and expensive. The standard test for PACs and PVCs is the electrocardiogram (ECG). These machines typically have a dozen electrodes that must be placed on the patient by a trained health practitioner. Out of necessity, these tests are limited to artificial, clinical settings, which prevents many, especially those with subclinical symptoms, from being tested.

Technological advances have made ECG recordings possible with affordable wearable and mobile devices in recent years, which could be the ideal means of increasing early detection. However, these devices usually only have a single electrode, and it is an open question as to how useful their data really is.

Recent research by a team at the University College London set out to answer this question. They performed a large-scale analysis of data collected by smartwatches, and other mobile devices, that have a built-in ECG function. Their goal was to determine if this data was of sufficient quality to accurately predict future health problems.

Data consisting of fifteen second samples from single-lead ECG devices was collected from 83,000 UK Biobank participants with a mean age of 58 years. This data was paired with future health outcomes for the same group of individuals. A machine learning algorithm was used to identify PVCs and PACs, which were then visually validated by medical professionals.

The team made corrections for confounding factors such as age and medication use, then performed a Cox regression analysis. This analysis revealed that PVCs detected in the single-lead ECG data could be linked to a twofold increase in later heart failure. Similarly, extra contractions attributed to PACs were associated with a twofold increase in cases of atrial fibrillation.

This work strongly suggests that the relatively simple ECG instrumentation that exists in many smartwatches and other wearable devices may be useful in self-screening for PACs and PVCs. This could provide an inexpensive and unobtrusive platform for early detection of individuals at a higher risk for future cardiac diseases. If these results are confirmed by further studies, the devices that many of us already use everyday may help to keep us healthier.