Comparing heart rate variability: Smartwatch versus electrocardiogram
Heart rate variability (HRV) reflects the difference in time between heartbeats. It is a well-known and simple way to check how the heart is working without invasive procedures. HRV has been proposed to predict risk in patients with cardiovascular diseases, including myocardial infarction, chronic heart failure, or ischaemic stroke. HRV is usually measured using a 30-minute electrocardiogram (ECG) to identify slower heart rate changes. While this method works well, it is not practical for everyday use. On the other hand, HRV can be derived from smartwatch data, making it a more convenient and accessible option.
What you need to know
This study evaluated the accuracy of HRV measurements from the Garmin Vivoactive 4® smartwatch compared to ECG. It involved three groups of participants: 104 patients who suffered a myocardial infarction, 129 patients who experienced an ischaemic stroke, and 30 healthy participants as the control. For the patient groups, recordings were performed three days after myocardial infarction and five days after ischaemic stroke.
HRV and non-standard HRV-derived measurements were calculated from the 30-minute smartwatch and ECG recordings for comparison. Non-standard HRV-derived measurements included the average heart rate, heart rate changes over time, and the ability of the heart to slow down, among others. For more details, please look at Table 2 in the manuscript.
What the data revealed
HRV measurements from a regular smartwatch matched well with those from detailed ECGs in patients with heart or brain conditions and healthy participants. However, this is only true for measurements that show overall heart rate changes or those that look at slower heart rate variations. There is less agreement for heart rate measurements that track quick, short-term changes, especially in people with health issues.
Not so fast
Data recordings were limited to 30 minutes and did not reflect a real-life scenario. Indeed, all participants were lying on their backs while resting so that movement, contact errors, and outside influences were minimised. In addition, an experienced nurse checked the proper use of the smartwatches. Finally, the methods used in this study cannot be directly applied to real-life data, and only one type of smartwatch was tested.
Why does it matter?
Two previous studies with healthy participants compared HRV readings from different smartwatches and a fitness tracker heart rate sensor, but the results were inconsistent. Smartwatch and ECG-derived HRV markers had not been tested before in patients with heart or brain conditions who could benefit from continuous risk monitoring. Markers evaluating the variability in lower frequency ranges are robust and agree with the smartwatch and the ECG, suggesting that smartwatches may be valuable tools for continuous risk monitoring.
Take home messages
1. Clinical use of smartwatch-derived HRV should focus on low-frequency and global variability parameters, which correlate closely with ECG-derived HRV.
2. Caution should be exercised with high-frequency parameters, although it is not a limitation for risk prediction in cardiovascular patients.
3. Smartwatches appear to be promising tools for monitoring patients with known heart or brain conditions, although more research is needed.
Guest author: Solène Grosdidier, PharmD, PhD
This article was written as part of a series of ‘journal club’ summaries for Scientific Writers Ltd and is based on the following publication.
First Author: Theurl F, et al.
Journal: European Heart Journal – Digital Health
Date online: 23 March 2023
Other references:
