Electrocardiogram (ECG) patch monitors for the detection of cardiac rhythm abnormalities

Key points from the evidence

1. Medical technology guidance (MTG) published in 2020 by the National Institute for Health and Care Excellence (NICE) recommends “Zio XT (an ECG patch monitor) as an option for people with suspected cardiac arrhythmias who would benefit from ambulatory ECG monitoring for longer than 24 hours only if NHS organisations collect information on:

• resource use associated with use of Zio XT®
• longer-term clinical consequences for people who have monitoring with Zio XT® (such as incidences of further stroke, TIA and other thromboembolisms, arrhythmia-related hospitalisations, mortality, uptake of anticoagulants or other changes in medication related to the monitoring result)”.

2. The NICE recommendation is based on four comparative studies (n=357 participants), including one randomised controlled trial (RCT) carried out in the United Kingdom (UK) that compared the Zio XT®’s 14-day monitoring period with a 24-hour Holter monitor and an external loop recorder (Novacor R.Test 4), which is a type of cardiac event recorder (CER).

• Zio XT® provided a higher diagnostic yield for detecting arrhythmias over total wear time compared with the 24-hour Holter monitor and Novacor R.Test 4.
• The relative diagnostic accuracy of the Zio XT® patch is uncertain. One study found agreement between the patch and Holter devices over the same 24-hour period. Another study found that the Holter monitor detected 11 arrhythmias, which were not detected by the Zio XT® patch, over a simultaneous 24-hour monitoring period. Nine of these ‘missed’ by the patch were attributed to algorithm and human reviewer errors when interpreting the data from the Zio XT® patch.

3. Evidence from a 2021 NICE medtech innovation briefing (MIB), included findings from two studies comparing an ECG patch monitor (the Carnation Ambulatory Monitor (CAM®) patch) with Holter monitors and the Novacor R.Test 4 for detecting AF.

•    A small UK-based randomised study (n=21 adults) reported that over 2 weeks, the CAM® device was more effective than Novacor R.Test 4 in detecting AF (odds ratio 5.8, 95% confidence interval (CI) 1.1 to 32.1, p=0.042).
• The CAM® patch detected more clinically significant rhythm abnormalities in patients (n=23, 46%) compared with Holter monitors (n=6, 12%) over 24 hours, in a prospective study in the United States (US) and New Zealand (n=50 adults). The majority of patients (n=48, 96%) preferred the wearability and comfort of the CAM® patch.

4. Results from two systematic reviews published after the NICE guidance reinforced the findings from the NICE MTG and MIB.

•     There was no statistically significant difference between ECG patch monitors and Holter monitors in detecting AF. ECG patch monitors were associated with shorter application times and better usability than Holter monitors.
• The pooled AF detection rate for ECG patch monitors, from three studies (n=221), was 9.1% (95% CI 3.3% to 22.6%, I2=6.4%, p=0.34).

5. One prospective comparative study (n=151 patients) that evaluated the diagnostic accuracy and usability of an ECG patch monitoring device (ECG247 Smart Heart Sensor) found that it demonstrated a diagnostic accuracy of 95% (95% CI 91 to 98) for automatic AF detection, while the Holter monitor had an accuracy of 81% (95% CI 74 to 87). False positives AF diagnoses occurred in 4% (6 of 137) of recordings analysed by the ECG247 algorithm compared with 16% (26 of 142) analysed by the Holter monitor’s algorithm. The study found no difference in adverse events between the patch and the Holter monitor.

6. Only minor adverse events have been associated with ECG patch monitors, limited to mild skin reactions like contact dermatitis and occasional adhesive failure. No serious adverse events were reported in any of the studies identified. The main safety consideration of ECG patch monitors relates to the potential for misclassification or missed arrhythmias because of detection algorithm limitations.

7. Evidence indicates that ECG patch monitors show good usability and are well received by patients compared with traditional Holter monitors. A higher proportion of participants reported the Zio XT® (93.7%) as being comfortable to wear compared with the Holter monitor (51.7%).

8. Our resource impact analysis for NHSScotland found that ECG patch monitors were resource saving compared with Holter monitoring in each of the populations we considered: a post-stroke and TIA population; and a population who would receive Holter monitoring in current practice because of symptoms (the ‘cardiology population’). The net resource savings were driven by fewer people having a recurrent stroke. In the cardiology population additional cost savings were as a result of fewer people having a myocardial infarction (MI).

•      In the post-stroke and TIA population, our model estimated cumulative resource savings of £14.6 million over 5-years, based on 689 fewer people having a recurrent stroke.
• In the cardiology population, our model estimated cumulative resource savings of £12.1 million, based on 541 fewer people having a stroke and 301 fewer people having an MI.
• In both models, ECG patch monitors were resource saving in the year of implementation and became increasingly resource saving in each subsequent year modelled.