Atrial fibrillation (AF) screening is important because AF can be paroxysmal and asymptomatic. Although AF is closely linked to certain abnormal echocardiographic parameters, these parameters have never been incorporated as criteria for AF screening. The SAFE-ECHO study is a prospective, randomized, open-label, multicenter trial to evaluate AF detection using extended electrocardiogram (ECG) monitoring in patients without AF who meet echocardiographic criteria, including left atrial enlargement (LAE), left ventricular hypertrophy (LVH), E/e’ > 14, grade II/III diastolic dysfunction, valvular heart disease (VHD), or heart failure (HF). Patients will be randomized in a 1:1 ratio to contemporary care (control arm) versus proactive staged screening (study arm). In the study arm, patients will undergo a 30-second single strip ECG and 7-day continuous ECG monitoring at enrollment and every 3 months for 1 year. The total study duration is 2 years, comprising a 1-year screening period and an additional year of follow-up. The primary endpoint is detection of new-onset AF within the 12-month screening period. The secondary endpoint is any arrhythmia of clinical significance or those requiring clinical interventions or a change of treatments. The tertiary endpoint is a composite of ischemic stroke, systemic thromboembolism, myocardial infarction, HF hospitalization, or emergent visit or mortality after the screening period, which will be evaluated for exploratory purposes only. In conclusion, the SAFE-ECHO study aims to evaluate the feasibility of specific echocardiographic parameters as criteria for AF screening and to determine the AF detection yield of scheduled extended ECG monitoring, which may help inform future AF screening strategies.
Trial registration : URL: https://clinicaltrials.gov ; Unique identifier: NCT07278089.
Atrial fibrillation (AF) is the most common arrhythmia in the world and timely diagnosis is crucial for preventing adverse events. The diagnosis of AF is usually straightforward, with typical symptoms and 12-lead electrocardiogram (ECG) presenting characteristic features of AF. However, detecting AF can be challenging in asymptomatic patients and those with paroxysmal AF, underscoring the need for effective screening strategies. AF screening typically entails the proactive invitation of a predefined population to undergo a structured screening program, mostly commonly based on age criteria. Besides, monitoring duration is a key determinant of AF detection, as demonstrated in the Crystal AF trial which reported significantly increased AF yield using long-term monitoring with an insertable cardiac monitor (ICM) in patients with cryptogenic stroke. Therefore, an effective AF screening strategy must consider the target population, along with the frequency and duration of cardiac rhythm monitoring, while balancing cost and benefit. , To date, an optimal screening strategy remains undetermined.
Structural and functional remodeling of the left atrium (LA) is common in AF, typically manifesting on echocardiography as LA dilatation, impaired LA function, and elevated left ventricular (LV) filling pressure. Advanced LA structural remodeling and worsening atrial cardiomyopathy may further exacerbate AF severity and increase the risk of adverse events. Meanwhile, certain diseases such as valvular heart disease (VHD), hypertrophy cardiomyopathy, or cardiac amyloidosis may predispose to the development of AF. In summary, AF is closely associated with structural and functional cardiac abnormalities, and echocardiography is a valuable tool for its assessment and risk stratification.
Therefore, we hypothesized that certain echocardiographic parameters may predispose to an elevated likelihood of AF occurrence and could serve as potential criteria for AF screening.
Pilot study: single-arm AF screening study in patients with abnormal echocardiographic parameters (SAFE-ECHO single arm)
In our previous single-arm pilot study, 118 patients with abnormal echocardiographic parameters underwent scheduled 7-day continuous ECG monitoring every 3 months for 1 year. AF was detected in 14 patients (11.9%), with 12 of them detected within 6 months. ( Table 1 and Figure 1 ). All AF detection was solely through 7-day ECG monitoring, not a single-time ECG. Anticoagulants were administered to 12 patients accordingly. This AF detection rate was higher than that based solely on routine clinical diagnosis estimated using a validated AF prediction scheme, the modified Taiwan AF score (11.9% vs 0.62%, p = 0.002). The E/e’ criterion had the highest AF rate (23.4%), followed by diastolic dysfunction (20.6%), LA enlargement (LAE) (20.4%), LV hypertrophy (LVH) (7.7%), VHD (6.7%), and heart failure (HF) (5.3%) ( Figure 2 ). The AF detection rate of patients who met combinations of 2 different criteria was also analyzed and the ‘LAE + E/e’‘ group had the highest rate at 32%, whereas the ‘HF + LVH‘ and ‘HF + VHD‘ subgroups showed no detection of AF ( Figure 3 ). These pilot data were intended to demonstrate feasibility and to inform key design assumptions, including event rate estimates and monitoring protocol, for the subsequent randomized trial, and were not designed to establish definitive evidence of clinical efficacy.
Table 1
Baseline characteristics and echocardiographic parameters of the pilot single-arm AF screening cohort (pilot data only)
| All cohort (n = 118) | E/e’ >14, (n = 47) | diastolic dysfunction ≥II (n = 34) | VHD (n = 15) | HF (n = 57) | LA enlargement (n = 54) | LVH (n = 26) | |
|---|---|---|---|---|---|---|---|
| Age, years | 66.1 ± 15.3 | 69.4 ± 13.7 | 71.4 ± 13.5 | 66.5 ± 15.0 | 64.5 ± 16.3 | 69.2 ± 14.4 | 64.8 ± 15.0 |
| Male sex, n (%) | 64 (54.2%) | 24 (51.1%) | 17 (50%) | 7 (46.7%) | 35 (61.4%) | 26 (48.1%) | 16 (61.5%) |
| Underlying history | |||||||
| HTN, n (%) | 65 (55.1%) | 30 (63.8%) | 22 (34.7%) | 8 (53.3%) | 26 (45.6%) | 35 (64.8%) | 16 (61.5%) |
| DM, n (%) | 21 (17.8%) | 10 (21.3%) | 7 (20.6%) | 3 (20%) | 10 (17.5%) | 10 (8.5%) | 8 (30.8%) |
| HF, n (%) | 46 (39%) | 16 (34%) | 13 (38.2%) | 4 (26.7%) | – | 18 (33.3%) | 7 (26.9%) |
| CAD, n (%) | 10 (8.5%) | 4 (8.7%) | 3 (8.8%) | 1 (6.7%) | 8 (14%) | 3 (5.6%) | 1 (3.8%) |
| Stroke, n (%) | 1 (0.8%) | 1 (2.1%) | 1 (2.9%) | 1 (6.7%) | 0 | 1 (1.9%) | 1 (3.8%) |
| CKD, n (%) | 48 (40.7%) | 24 (51.1%) | 19 (55.9%) | 8 (53.3%) | 25 (43.9%) | 21 (38.9%) | 10 (38.5%) |
| Laboratory data | |||||||
| NTproBNP, (IQR), pg/mL | 920 ± 2091 (125-752) | 1430 ± 3137 (151-933) |
1621 ± 3528
(152-920) |
1630 ± 5656
(228-2255) |
1069 ± 2460
(148-983) |
1040 ± 2579 (105-857) | 1129 ± 2194 (169–823) |
| Hgb, g/dL | 13.4 ± 3.9 | 12.7 ± 2.7 | 12.3 ± 2.7 | 12.3 ± 2.6 | 13.5 ± 2.3 | 12.7 ± 2.5 | 14.8 ± 6.8 |
| A1C, % | 6.1 ± 0.8 | 6.2 ± 1.0 | 6.2 ± 0.8 | 5.8 ± 0.6 | 6.1 ± 1.0 | 6.0 ± 0.7 | 6.0 ± 0.7 |
| Echocardiographic parameters | |||||||
| LA diameter, mm | 40.0 ± 6.3 | 40.9 ± 5.4 | 43.1 ± 6.0 | 44.0 ± 7.8 | 39.0 ± 6.3 | 44.1 ± 6.0 | 41.6 ± 6.8 |
| LAVI, ml/m2 | 36.3 ± 12.6 | 37.4 ± 12.2 | 40.8 ± 10.8 | 41.1 ± 15.6 | 32.1 ± 13.0 | 43.4 ± 10.2 | 36.7 ± 12.6 |
| LV end-diastolic diameter, mm | 48.0 ± 7.8 | 46.9 ± 7.4 | 48.3 ± 7.2 | 50.8 ± 11.4 | 50.5 ± 8.1 | 48.4 ± 8.0 | 44.8 ± 6.6 |
| LV end-systolic diameter, mm | 33.0 ± 9.5 | 31.3 ± 9.0 | 32.4 ± 8.3 | 35.8 ± 13.7 | 37.9 ± 9.9 | 32.1 ± 10.1 | 28.6 ± 7.0 |
| IVSd, mm | 11.9 ± 3.6 | 12.2 ± 3.3 | 12.6 ± 3.5 | 12 ± 3.3 | 10.9 ± 2.7 | 12.4 ± 3.3 | 17.5 ± 3.1 |
| PWd, mm | 10.6 ± 2.9 | 10.9 ± 3.0 | 11.1 ± 3.3 | 10.7 ± 2.0 | 10.0 ± 2.4 | 11.1 ± 3.1 | 14.3 ± 3.3 |
| MV E, cm/s | 74.8 ± 25.9 | 87.7 ± 25.6 | 88.8 ± 29.2 | 96.0 ± 34.9 | 67.4 ± 24.9 | 82.3 ± 27.3 | 70.6 ± 24.8 |
| MV A, cm/s | 87.4 ± 28.0 | 100.0 ± 31.3 | 99.3 ± 31.1 | 97.0 ± 31.4 | 81.3 ± 24.4 | 93.2 ± 30.0 | 84.1 ± 30.7 |
| E/e’ | 13.3 ± 5.2 | 17.8 ± 4.3 | 17.7 ± 4.7 | 16.1 ± 6.0 | 11.2 ± 3.9 | 14.3 ± 5.6 | 16.0 ± 6.7 |
| LVEF, % | 54.0 ± 10.7 | 55.3 ± 11.0 | 55.6 ± 10.4 | 52.7 ± 11.3 | 48.4 ± 11.2 | 56.4 ± 10.8 | 58.5 ± 8.9 |
| RVSP, mmHg | 31.0 ± 10.6 | 33.0 ± 10.8 | 37.2 ± 12.8 | 41.0 ± 12.5 | 31.3 ± 11.5 | 33.1 ± 11.0 | 31.0 ± 10.4 |
CAD = coronary artery disease; CKD = chronic kidney disease; DM = diabetes mellitus; e’ lateral = average of the motion velocity of the septal and lateral mitral annulus in the early phase of diastole; HF = heart failure; HTN = hypertension; IVSd = end-diastolic interventricular septum thickness; LA = left atrium; LAVI = LA volume index; LV = left ventricle; LVEF = LV ejection fraction; LVH = LV hypertrophy; MV A = late diastolic trans-mitral inflow velocity; MV E = early diastolic trans-mitral inflow velocity; NT-proBNP = N-terminal pro-B type natriuretic peptide; PWd = end-diastolic posterior wall thickness; RVSP = right ventricular systolic pressure; VHD = valvular heart disease.
Cumulative AF detection during follow-up from a single-arm pilot study . These findings are presented to support feasibility and inform the design of the randomized trial.
