Background
Severe aortic valve stenosis (AS) is the most common valvular heart disease, and the aging population causes the prevalence of AS to increase each decade [1, 2]. Transcatheter aortic valve implantation (TAVI) is an emerging therapeutic approach for patients with this condition, primarily for those at high surgical risk, but it is increasingly used in a broader group of patients with AS [3, 4].
The TAVI procedure carries a risk of serious complications, such as stroke, life-threatening arrhythmias, and atrioventricular block that may require temporary or permanent pacing. Because of the anatomical relationship between the conduction system and the aortic valve complex, conduction disturbances are a common complication of TAVI [5]. One of the most frequent complications following this procedure is high-degree atrioventricular block (HAVB) [6].
This conduction disorder often necessitates permanent pacemaker implantation, which is why patients are typically monitored for 48 hours following TAVI. However, up to 40% of HAVB cases may develop later, leading to serious complications such as syncope or sudden cardiac death. Another common conduction disturbance associated with TAVI is new-onset left bundle-branch block (NO-LBBB), which occurs in approximately 40% of patients and tends to occur more frequently before hospital discharge.
Importantly, if LBBB becomes persistent, the risk of death increases [7].
New-onset atrial arrhythmias are a common complication after TAVI [8] and occur more frequently during the procedure and at a later stage when the patient is no longer monitored by ECG [9]. In usual clinical practice, monitoring is carried out for 24–48 hours after TAVI. During this period, many severe and life-threatening cardiac arrhythmias can be detected through such monitoring.
The risk of such arrhythmias decreases over time as the patient transitions out of the acute postoperative period. Hence, extended monitoring is considered not obligatory but may be warranted [5].
After TAVI, the risks of mortality and stroke are higher in patients with new-onset atrial fibrillation (NOAF) compared with those without atrial fibrillation (AF) [10]. Patients with NOAF are at increased risk of death and stroke, with the highest probability of occurrence in the subacute phase (1–30 days) after the procedure [11]. Because of the high incidence of NOAF among elderly patients undergoing TAVI and the fact that NOAF is often asymptomatic, episodes of AF may go undetected [12].
NOAF can be paroxysmal and is more likely to be detected after a longer monitoring period [9, 11, 13].
One helpful tool for identifying arrhythmias after a period of standard ECG monitoring is ambulatory electrocardiography (aECG). The use of continuous remote ECG monitoring to diagnose cardiac arrhythmias, including NOAF, can potentially lead to a significant reduction in complications such as stroke compared with standard ECG monitoring after TAVI.
The DRAGON-TAVI trial aims to compare a 30-day remote ECG monitoring strategy with standard care to improve arrhythmia diagnosis and risk assessment after TAVI in patients undergoing the procedure.
Methods
Study design
The DRAGON-TAVI study is designed as a multicenter, prospective, open-label, randomized trial. The study will enroll 250 patients, randomized 1:1 to the study group and the control group. Both groups will receive standard postoperative 24–48-hour ECG monitoring during hospitalization after TAVI and a regular follow-up visit with a standard ECG within 30 days.
Moreover, the study group will receive continuous 30-day ECG monitoring (standard 3-lead ECG Holter) with visits triggered by ECG findings. Only follow-up after hospitalization will be taken into account in the final analysis. The open trial design assumes that the investigator and the patient will be aware of the assigned randomization group.
Enrollment, participants, and the study flowchart
Inclusion and exclusion criteria are listed in Table 1. After the procedure, the patient is informed about the study and asked to provide consent. Randomization will be performed using a dedicated tool available at https://app.studyrandomizer.com.
After providing written informed consent and undergoing the procedure and randomization (after the TAVI procedure), patients randomized to the study group (30-day monitoring) will receive a monitoring device on the day of discharge (visit 1). They will subsequently have a 14-day telephone visit (visit 2) and a 30-day in-office visit (visit 3). All patients included in the study will receive information about the study protocol at discharge. https://app.studyrandomizer.com
The ECG will be monitored continuously for the entire 30-day period and reviewed daily by both the technician and the physician (investigator). If, during the monitoring period, the subject experiences a significant bradyarrhythmia (severe bradycardia, second- or third-degree atrioventricular block) or tachyarrhythmia (NOAF, AFl, supraventricular tachycardia, ventricular tachyarrhythmia, or other tachyarrhythmias), the investigator will be able to change the therapeutic strategy immediately, including therapy modification and a decision to admit the patient to the hospital. The decision on therapeutic management will be based on the type of arrhythmia and the patient’s clinical condition.
Control group subjects will undergo a standard 12-lead ECG at discharge and a 14-day in-office visit, which will include a standard physical examination and another 12-lead ECG. Baseline characteristics will be recorded and analyzed, including demographics, comorbidities, baseline ECG, echocardiographic findings, laboratory test results, TAVI procedure details, and medications at discharge. Clinical outcomes will be followed for 12 months ± 1 month.
After 12 months, the final in-office visit (visit 4), with a 12-lead ECG, will be performed. The study flowchart is presented in Figure 1.
All arrhythmias and other pathological conditions will be assessed over a 30-day period using Holter ECG. NOAF will be defined as any arrhythmia with the electrocardiographic features of AF that lasts at least 30 s on a Holter rhythm strip [14]. Atrial flutter will be diagnosed as any arrhythmia with the ECG characteristics of atrial flutter that lasts at least 30 s on a rhythm strip.
Atrioventricular block will be defined as the presence of second-degree atrioventricular block (Mobitz I or Mobitz II) or third-degree atrioventricular block, regardless of duration, on a Holter rhythm strip.
If the study ends earlier than planned (for example, if the participant withdraws consent, removes the Holter device, or does not follow the study rules), a final visit will be scheduled. During this visit, the reason for leaving the study will be recorded. The study team will explain the study procedures in detail and schedule all visits accordingly.
Patients will provide verbal and written informed consent to participate in the study after receiving detailed instructions describing the study protocol and its associated risks and benefits. Patients can withdraw at any time without endangering their medical care. The Medical University of Silesia Ethics Committee in Katowice, Poland, has approved the study protocol (bioethics committee approval number BNW/NWN/0052/KB1/78/I/23).
The study was registered at www.clinicaltrials.gov (NCT07521800).
ECG monitoring device and platform
Patients in the study group will be monitored using PocketECG III (MEDICALgorithmics, USA) and a dedicated platform. The ECG device is applied to the patient’s chest using three electrodes. Once inserted, the device automatically and continuously records and transfers the ECG to a dedicated platform via GSM.
Additionally, the device allows a subject to mark symptomatic events. The analysis will be performed as described above.
Study outcomes
The primary endpoints will be the assessment of:
Secondary endpoints will include all-cause mortality; major adverse cardiovascular events (MACE, i.e., cardiovascular death, stroke, myocardial infarction, and hospitalization for heart failure); episodes of syncope; and cardiac arrhythmias (AF/AFl duration >30 s, sustained ventricular tachycardia [sVT], non-sustained ventricular tachycardia [nsVT], and other atrioventricular conduction disorders) during 12 months of follow-up.
Statistical analysis and sample size calculation
Primary and secondary endpoints will be compared between the study and control groups. Data will be analyzed on an intention-to-treat basis. Besides the incidence of endpoints, we plan to analyze the following: (i) time to the occurrence of the endpoints in the compared groups; (ii) the relationship between the endpoints and baseline characteristics; and (iii) the relationship between the endpoints and procedure-related details.
Logistic regression analysis will be performed to identify factors associated with AF recorded after discharge. A p-value of <0.05 will be considered statistically significant throughout the analyses.
We assumed a control incidence of 5–10% for newly detected AF/AFl in non-continuously monitored TAVI patients. Based on contemporary reports that ~30-day extended ECG monitoring increases 30-day AF/AFl detection into the mid-teens and, in selected series, above 20%, we powered the trial to detect an absolute difference of ~12.5 percentage points (e.g., ~20% vs. 7.5%) [15–19].
Using a two-sample test of proportions (two-sided α=0.05; 80% power; 1:1 allocation), the minimum required sample size is N ≈ 236 (118 per arm) under the normal approximation. To accommodate uncertainty and potential loss to follow-up, we set the target at N = 250 (125 per arm). Statistical calculations will be performed using MedCalc® Statistical Software version 22.026 (MedCalc Software Ltd, Ostend, Belgium).
Discussion
The DRAGON-TAVI trial assesses whether a remote ECG monitoring device, compared with standard patient monitoring, could improve the outcomes of TAVI treatment. This device can record atrioventricular conduction blocks and arrhythmias in real time and thus provide a more precise diagnosis, with a significant reduction in the risk of ischemia and other adverse events associated with this procedure. To our knowledge, this is the first and largest randomized controlled trial (RCT) to evaluate remote ECG monitoring as a diagnostic tool for therapeutic strategies after TAVI, focusing on significant cardiac arrhythmias.
Unlike previously published studies evaluating post-TAVI rhythm surveillance, the DRAGON-TAVI trial has several distinguishing features. Prior investigations have largely been single-center or focused on selected high-risk populations, often assessing isolated endpoints such as delayed high-grade atrioventricular block or NOAF detected by event monitors or patch devices. In contrast, DRAGON-TAVI is a multicenter, prospective, randomized trial comparing a standardized 30-day continuous remote ECG monitoring strategy with standard post-discharge care.
The study evaluates a broad spectrum of clinically relevant arrhythmias, including atrial fibrillation/flutter, supraventricular and ventricular tachyarrhythmias, atrioventricular conduction disturbances, and pauses, and links their detection to immediate therapeutic consequences. Continuous daily ECG review with predefined triggers for clinical intervention allows assessment not only of diagnostic yield but also of the impact of early arrhythmia detection on patient management. By integrating prolonged monitoring into a structured follow-up pathway and comparing it against routine care, DRAGON-TAVI addresses a key gap in current evidence and may provide randomized data to inform future recommendations on post-TAVI rhythm monitoring strategies.
The DRAGON-TAVI study is one of the few multicenter, prospective, open-label, randomized (1:1) trials evaluating an ECG device for prolonged remote monitoring in patients undergoing TAVI. Our results could provide convincing evidence for the use of remote monitoring after TAVI and could suggest that this approach is effective as a diagnostic tool for recognizing the presence of AV block and NOAF. Recently, several studies have evaluated post-TAVI monitoring, providing a helpful approach for diagnosing arrhythmias.
One post-TAVI study by Ream et al. [20] involved patients with no history of implantable devices who did not require pacing device implantation before hospital discharge after the procedure. The patients were discharged from the hospital with a 30-day ambulatory event monitor (AEM) in place, and the study found that 10% of patients with the device developed high-degree atrioventricular block (H-AVB) between three and 24 days after TAVI, although better clarification of the incidence, risk factors, and selection of patients for outpatient monitoring was needed [20]. In another prospective study, Tian et al. [21] used a remote monitoring system for 30 days in patients without an implantable device after TAVI.
The primary endpoint was Mobitz II atrioventricular block or third-degree atrioventricular block (II/III AVB) at 30-day follow-up, and 8.6% of subjects required device implantation within 30 days after TAVI [22]. In a single-center, prospective study, Skaf et al. [23] used a 30-day rhythm monitor in patients after TAVI and found that new-onset grade II or III atrioventricular block occurred in 7.1% of monitored patients, and NOAF occurred in 6%. Hence, this 30-day monitoring strategy should be considered for patients at risk of high-grade atrioventricular block [15].
The use of this device for prolonged monitoring has the benefit of faster diagnosis of NOAF, which, in many patients undergoing TAVI, is only detected at the onset of stroke [11, 23]. In addition, it is worth noting that, particularly in the elderly patient population, extensive strokes are judged by expert consensus to be severe clinical events at the level of death, with a significant reduction in quality of life and family involvement in the burdensome therapeutic and treatment process [24]. Also, an ischemic stroke in the course of AF may become hemorrhagic in the next phase, which has even more drastic consequences, especially as patients after TAVI will be receiving intensive antiplatelet therapy at a time of increased NOAF risk [21, 25].
As evident from recent studies on remote ECG monitoring after TAVI, further research is needed, along with a formal evaluation comparing a strategy of remote monitoring with usual care. Prior studies have focused on NOAF or atrioventricular conduction block, and limited data are available on high-risk patients, including the elderly. More prolonged monitoring, using more sophisticated methods, would be expected to improve the detection of arrhythmias, including NOAF, as well as identify patients at risk of AV block necessitating pacemaker therapy.
Conclusions
In conclusion, the DRAGON-TAVI trial will provide data on a 30-day remote ECG monitoring strategy compared with standard care to improve diagnosis and risk assessment of arrhythmias in patients undergoing TAVI.