The Accuracy of Smartwatches Compared to Holter Monitors for Heart Rate Monitoring in Atrial Fibrillation: A Pilot Study

Abstract

Introduction: Smartwatches (SW) with photoplethysmographic heart rate (HR) monitoring capability are increasingly popular. We present a pilot study to assess the accuracy of two SW (Fitbit Charge HR [FB] and Apple Watch Series 3 [AW]) compared with Holter monitoring in ambulatory patients in atrial fibrillation (AF). Methods: Patients with persistent AF referred for 24-hour Holter monitoring were prospectively recruited. Six patients with sinus rhythm (SR) were recruited as a comparator. The Holter monitor was the criterion measure. Each patient was then allocated to wear either a FB or AW with their Holter for 24 hours. Statistical analysis: Pearson (r) correlation coefficients and Bland-Altman comparison with 95% limits of agreement (LoA) were evaluated to assess agreement between SW and Holter HR. Bias was then calculated as the mean difference between the SW and Holter HR. A ± 10-beat difference between Holter-HR and SW-HR was used as the metric for accuracy of SW derived HR. Results: Over 53,000 heart rate values were recorded from 32 patients. Both SW performed well in SR (Bias <1 beat, LoA −11 to 11 beats, r = 0.87, p ≤ 0.001) with accuracy 92 - 95%. However, in AF, both devices underestimated HR with wide LoA (bias -9 beats, LoA -41 to 23 beats, r = 0.60, p < 0.001). The degree of underestimation was more pronounced when HR >80. In AF, accuracy was reduced: 77% for AW and 56% for FB. Conclusion: In ambulatory patients, smartwatches accurately measure HR in SR. However, they have reduced accuracy and underestimate HR in AF.