Table 1.
Studies comparing the detection rates of atrial fibrillation between different cardiac monitoring techniques.
| Study | Number of Patients (n) | Study Design | Technique(s) and Timescale | Outcomes (Primarily Regarding AF Detection Rate) |
|---|---|---|---|---|
| Buck et al. [42] | 300 | RCT | 12-month ILR vs. 30-day ELR | In ischemic stroke patients monitored for 12 months using an ILR, AF was detected in 15.3% of patients compared to 4.7% of patients monitored with a 30-day ELR (RR 3.29 [95% CI 1.45–7.42], p = 0.03). |
| Koh et al. [43] | 203 | RCT | 30 days of KardiaMobile ECG (used for 30 s 3 times per day) vs. one additional round of 24-h Holter monitoring | In patients with a recent cryptogenic stroke or TIA, AF of duration ≥ 30 s was detected in 9.5% of patients in the KardiaMobile group compared to 2.0% in those receiving 24 h Holter monitoring (p = 0.024). |
| Liu et al. [21] | 158 | Prospective | 14-day ECG patch vs. 24-h Holter monitoring | AF and/or atrial flutter was detected in 9.5% of patients wearing the 14-day ECG patch compared to 3.8% in patients with 24 h Holter monitoring (p = 0.042). |
| Medic et al. [44] | 1000 | Retrospective (economic model) | 30-day MCOT followed by ILR vs. ILR alone over 12 months | In patients with cryptogenic stroke, 30-day MCOT followed by an ILR had an AF detection rate of 20.9% compared to a detection rate of 4.5% when using an ILR alone. Cost-effectiveness analysis revealed 7.72 times lower costs per AF patient detected when using MCOT initially (USD 29,598 per patient with detected AF) compared to when using an ILR only (USD 228,507 per patient with detected AF). |
| Chua et al. [20] | 32 | Prospective | 14-day ECG patch vs. 24-h Holter monitoring | Paroxysmal AF and/or atrial flutter was detected in 19% of patients wearing the ECG patch compared to 3% in patients with 24 h Holter monitoring (p < 0.05). |
| Perez et al. [45] | 419,297 | Prospective | 8 months of monitoring with an Apple Watch | Of Apple Watch wearers among the general population, 0.52% received a notification regarding an irregular pulse over 8 months. Of the patients who were notified to have an irregular pulse and subsequently wore and returned an ECG patch, AF was confirmed in 34%. Comparatively, in those who did not receive a notification regarding an irregular pulse, a diagnosis of AF was established in 1.0% of patients. |
| Derkac et al. [46] | 78,490 | Retrospective | MCOT vs. ILR analysed over 8 months | AF was diagnosed in 23.5% of patients with MCOT compared to 11.3% of patients with an ILR. It should be noted that the median prescription time for the MCOT group was 20 days compared to 30 days in those with the ILR group, despite the latter device having the potential to be used for a considerably longer duration. |
| Gladstone et al. [47] | 572 | RCT | 30 days of event-triggered loop recorder vs. 24 h ECG | AF was detected in 16.1% of patients with the event-triggered loop recorder compared to 3.2% of patients with 24 h ECG monitoring (p < 0.001). Episodes of AF spanning 150 s or longer were recorded in 9.9% of patients with event-triggered recorders compared to 2.5% of those with standard 24 h ECG monitoring (p < 0.001). These differences had clinical implications as 18.6% of patients in the loop recorder group had commenced anticoagulant therapy compared to 11.1% of those in the 24 h ECG monitoring group (p = 0.01). |
| Sanna et al. [48] | 441 | RCT | 6 months of ILR vs. conventional follow-up (ECG assessment at follow-up visits, with exact protocol at the discretion of each site) | AF was detected in 8.9% of patients with an ILR by 6 months compared to 1.4% in patients receiving conventional follow-up following a cryptogenic stroke (HR 6.4 [95% CI 1.9 to 21.7], p < 0.001). By 12 months, AF was detected in 12.4% of patients with an ILR compared to 2.0% in those with conventional follow-up (HR 7.3 [95% CI 2.6 to 20.8], p < 0.001). In the patients followed up for 36 months, these rates grew to 30% and 3.0% respectively (HR 8.8 [95% CI 3.5 to 22.2], p < 0.001). |
Abbreviations: AF = atrial fibrillation, HR = hazard ratio, RR = relative risk, CI = confidence interval, ECG = electrocardiogram, ILR = implantable loop recorder, ELR = external loop recorder, MCOT = mobile cardiac outpatient telemetry, RCT = randomised controlled trial, USD = United States dollar, TIA = transient ischemic attack.