Left atrial appendage closure versus direct oral anticoagulants after pulmonary vein isolation for atrial fibrillation: protocol for a multicentre, prospective, randomised, non-inferiority trial (PROMOTE study)

Abstract

Introduction

Atrial fibrillation (AF), with a prevalence of 1–2%, is the most common cardiac arrhythmia. AF is associated with a fivefold increased risk of cardioembolic events; approximately 20% of all strokes are caused by AF. Pulmonary vein isolation (PVI) has become the first-line treatment for AF. However, PVI cannot eliminate the residual stroke risk. Current guidelines recommend that anticoagulation be continued in this specific group of patients, regardless of the presence or absence of AF. In this large AF population post-PVI, who are considered to be in an earlier stage of AF, it is unknown whether left atrial appendage closure (LAAC) offers an alternative to direct oral anticoagulant (DOAC) therapy.

Methods and analysis

The trial will be a prospective, randomised, multicentre non-inferiority study comparing two treatment strategies in AF patients after atrial ablation. Patients will be randomly assigned to either percutaneous LAAC (group A) or DOAC treatment (group B) in a 1:1 ratio; both sequential and concomitant planned ablation with or without LAAC are accepted. Randomisation will be conducted using web-based randomisation software. A total of 1012 participants (506 patients per group) will be enrolled. The primary effectiveness measure will be the occurrence of any of the specified events within 24 months after randomisation: stroke/transient ischaemic attack/systemic thromboembolism, cerebral haemorrhage, other major haemorrhages (Bleeding Academic Research Consortium ≥2), cardiovascular mortality and all-cause mortality.

Ethics and dissemination

The study was approved by the Ethical Review Board of Shanghai Chest Hospital, China (KS(Y)20287). Written informed consent will be obtained from all participants. The trial will follow the Declaration of Helsinki and Good Clinical Practice. Confidentiality will be maintained with anonymised, securely stored data. Findings will be disseminated through peer-reviewed publications and conferences.

Trial registration number

ChiCTR2000036538.

STRENGTHS AND LIMITATIONS OF THIS STUDY.

• The study employs a rigorous multicentre, prospective, randomised controlled trial design with a large sample size of 1012 participants, which enhances internal validity and provides sufficient statistical power to detect clinically meaningful differences between treatment strategies.

• The inclusion of multiple tertiary cardiac centres across China increases the generalisability of the findings and ensures representation of diverse clinical settings and patient populations.

• The 24-month follow-up duration allows for comprehensive assessment of intermediate-term safety and efficacy outcomes; however, extended follow-up may be required to evaluate very long-term effects and rare adverse events related to both interventions.

Introduction and the rationale for the study

Atrial fibrillation (AF) is the most common sustained arrhythmia in middle-aged and elderly individuals. Both its incidence and prevalence increase with age.¹ Among those 65 years and older, the incidence reaches approximately 9%.² AF is closely associated with two major comorbidities—stroke and heart failure—both of which significantly affect its overall prognosis. While antiarrhythmic drugs have proven effective in managing heart failure secondary to AF with tachycardia, the prevention of stroke remains a persistent clinical challenge.

Stroke associated with AF carries nearly a fivefold higher risk compared with stroke from other causes. Approximately one in every five strokes can be attributed to AF.^{3 4} Moreover, AF-related strokes are typically more severe, leading to higher mortality, morbidity and recurrence rates, as well as more pronounced neurological deficits.⁵ Therefore, anticoagulation is the cornerstone of AF management. Current guidelines recommend four major approaches for AF treatment: (1) rhythm control—via pulmonary vein isolation (PVI), electrical cardioversion or antiarrhythmic drugs; (2) rate control—using antiarrhythmic drugs; (3) stroke and thromboembolic prevention—through anticoagulation therapy and (4) left atrial appendage closure (LAAC).^{6 7}

Despite the increasing use of PVI for symptom relief in AF, its effectiveness in preventing stroke remains controversial. Recurrence rates following ablation are as high as 50%,⁸,¹⁰ and the risk of stroke after PVI is not substantially reduced, highlighting the ongoing need for anticoagulation therapy even after successful ablation. ^{11 12}Consequently, anticoagulation remains the mainstay of stroke prevention post-PVI. However, anticoagulation carries inherent limitations, including bleeding risk and issues with long-term adherence, making it an imperfect preventive strategy and prompting exploration of alternative options under appropriate circumstances.

Stroke in AF is well established to arise from left atrial thrombus formation.¹³,¹⁶ Approximately 90% of these thrombi originate from the left atrial appendage (LAA), a unique anatomical structure with reduced contractility that predisposes to blood stasis.¹⁷ This mechanistic understanding has driven the development of transcatheter LAAC. Recently, the LAA occlusion study III trial demonstrated that surgical LAA occlusion provides an additive reduction in stroke risk when combined with oral anticoagulation.¹⁸ Compared with stroke rates predicted by the CHA₂DS₂-VASc score (Congestive heart failure, Hypertension, Age ≥75 [2 points], Diabetes mellitus, prior Stroke/TIA/thromboembolism [2 points], Vascular disease, Age 65–74, Sex category [female]), LAAC has shown substantial reductions in total stroke—78% in Continued Access to PROTECT-AF and 69% in Continued Access to PREVAIL.¹⁹ Long-term results from the PROTECT-AF and PREVAIL trials further indicated that LAAC was superior to warfarin in preventing disabling or fatal strokes and in reducing bleeding complications.^{20 21}

In recent years, direct oral anticoagulants (DOACs) have become the first-line therapy for stroke prevention in non-valvular AF, as recommended by current guidelines.⁷ Within this context, the PRAGUE-17 trial demonstrated that LAAC was non-inferior to DOAC therapy.²² However, most comparative studies between DOACs and LAAC have enrolled patients with larger left atrial dimensions and higher CHA₂DS₂-VASc scores (blue circle, figure 1), representing more advanced stages of AF.²³ In contrast, patients undergoing ablation—particularly for paroxysmal AF—tend to be younger, with smaller atria and earlier disease stages. Notably, research specifically comparing LAAC and DOAC outcomes in this post-PVI population (red circle, figure 1) is lacking.

Figure 1. Study population of the current study (right circle) versus prior studies (blue circle). LA, left atrium; LAAC, left atrial appendage closure; PVI, pulmonary vein isolation; CHA₂DS₂-VASc score, Congestive heart failure, Hypertension, Age ≥75 [2 points], Diabetes mellitus, prior Stroke/TIA/thromboembolism [2 points], Vascular disease, Age 65–74, Sex category [female].

Regardless of PVI status, stroke risk in AF patients remains unmitigated.¹¹ Although PVI restores sinus rhythm, the AFFIRMAtrial Fibrillation Follow-Up Investigation of Rhythm Management (AFFIRM) trial demonstrated that rhythm restoration alone does not reduce stroke incidence.²⁴ Similarly, the CABANA (Catheter Ablation vs Antiarrhythmic Drug Therapy for Atrial Fibrillation) trial (2019) found no significant difference between catheter ablation and medical therapy in the composite endpoint of death, disabling stroke, serious bleeding or cardiac arrest.¹¹ Large national registries have also shown no difference in adjusted cardiovascular or all-cause mortality between patients treated with PVI and those managed with antiarrhythmic medications alone.²⁵ Consequently, long-term anticoagulation remains indispensable in patients following PVI.^{6 26 27} Due to endothelial injury during ablation, oral anticoagulation is advised for all patients post-PVI.⁶ Non-randomised long-term studies have shown lower stroke incidence among those maintaining a persistent sinus rhythm.²⁸ Therefore, both the 2020 European Society of Cardiology (ESC) guidelines and the 2023 American College of Cardiology (ACC) guidelines recommend continuing oral anticoagulation in all patients at moderate to high stroke risk, irrespective of PVI status.^{6 29}

In real-world practice, post-PVI patients represent a large and growing high-risk population for stroke.³⁰ PVI has become the primary treatment choice for AF, with procedure volumes increasing nearly 10-fold over the past decade.^{31 32} For stroke prevention, DOACs remain first-line pharmacologic therapy, whereas LAAC has emerged as a viable interventional alternative.³⁰ However, no dedicated study has specifically addressed this expanding post-PVI cohort. In the context of global population ageing and the growing cohort of patients undergoing PVI (red circle, figure 1), the comparative effectiveness of DOACs versus LAAC remains unresolved. High-quality scientific evidence is critically needed to inform clinical decision-making.

This study is a prospective, randomised, multicentre investigation to determine if LAAC is a reasonable alternative to oral anticoagulation in patients post-PVI. Both efficacy and safety endpoints are being systematically evaluated.

Study objective

To determine whether LAAC is non-inferior to DOAC in patients with AF at moderate-to-high risk of thromboembolism, as defined by a CHA₂DS₂-VASc score of ≥2 in men or ≥3 in women.

Trial design

This is an investigator-initiated, prospective, multicentre, randomised, non-inferiority study comparing two treatment strategies, PVI+LAAC and PVI+DOAC. Eligible patients will be randomised in a 1:1 ratio to receive either LAAC or DOAC therapy. Both sequential and concomitant ablation ± LAAC procedures are permitted. Patient recruitment began in May 2021 and was completed in May 2024, with the final 24-month follow-up visits anticipated by May 2026. The study flowchart is shown in figure 2.

Figure 2. Patients’ flowchart. AF, atrial fibrillation; LAAC, left atrial appendage closure; PVI, pulmonary vein isolation; NOACs, non-vitamin K antagonist oral anticoagulants; CHA₂DS₂-VASc score, Congestive heart failure, Hypertension, Age ≥75 [2 points], Diabetes mellitus, prior Stroke/TIA/thromboembolism [2 points], Vascular disease, Age 65–74, Sex category [female]).

Study population—inclusion criteria

The study will include patients diagnosed with AF who are recommended to receive anticoagulation therapy. The inclusion criteria for this study will encompass individuals who exhibit non-valvular AF (including paroxysmal, persistent or permanent forms), have provided a signed informed consent form and meet the following requirements:

1. The presence of non-valvular AF (paroxysmal, persistent or permanent).

2. CHA2DS2-VASC score: male ≥2 points, female ≥3 points.

3. Patients aged 18 years or older, but younger than 80 years old.

4. Diameter of the left atrium (LA) <55 mm.

5. Glomerular filtration rate >30 mL/min.

6. Those who can tolerate short-term anticoagulation/antiplatelet therapy.

Study population—exclusion criteria

Exclusion criteria will be as follows:

1. Diameter of the LA ≥55 mm.

2. Preoperative transoesophageal echocardiography (TEE) or cardiac CT imaging (Cardiac CT angiography, CCTA) detected thrombosis or suspected thrombus in the LA or LAA.

3. The preoperative TEE examination shows that the LAA’s anatomy is not suitable for the LAAO under the existing technology and equipment.

4. Transthoracic echocardiography (TTE) examinations suggest that the left ventricular ejection fraction is less than 30%.

5. TTE examination reveals that there is more than 10 mm of pericardial effusion with unknown causes.

6. Comorbidities other than AF that require long-term anticoagulation therapy (such as post mechanical valve replacement, spontaneous or recurrent venous thromboembolism, etc).

7. Moderate or severe mitral valve stenosis (valve area <1.5 cm²) or mechanical valve prosthesis.

8. Patients with severe valvular heart disease or abnormal heart structure (such as major atrial septal defect, ventricular septal defect) requiring surgical treatment, or severe coronary heart disease requiring surgical bypass.

9. New-onset ischaemic stroke/transient ischaemic attack (TIA) without haemorrhagic conversion, yet is not suitable for initiation of anticoagulation therapy based on the National Institute of Health Stroke Scale score and neurologist evaluation.

10. Acute ischaemic stroke with haemorrhagic conversion or intracranial haemorrhage triggered by oral anticoagulation therapy, for whom anticoagulation is not appropriate after multidisciplinary assessment.

11. Active peptic ulcer with bleeding <3 months.

12. Coagulation dysfunction or coagulation factor deficiency.

13. The participant is capable of bearing children and is currently pregnant or intends to become pregnant during the study period.

14. The estimated survival period is <1 year.

15. Uncontrolled heart failure, New York Heart Association Functional Classification IV.

16. Creatinine clearance less than 30 mL/min.

Randomisation protocol

Participants will be randomly assigned in a 1:1 ratio to either the interventional LAAC group (group A) or the DOAC group (group B) using a web-based randomisation system. Both sequential and concomitant PVI procedures are permitted, with radiofrequency ablation as the standard PVI technique. The interval between a prior PVI and randomisation will not exceed 90 days.

The randomisation software is designed to maintain balance in baseline CHA₂DS₂-VASc characteristics between groups, ensuring comparable thromboembolic risk profiles. The randomisation process will be conducted externally and independently from all participating centres to prevent allocation bias. If a patient must be excluded after randomisation (eg, due to the detection of a thrombus on postrandomisation TEE), this decision will be made solely by an unmasked statistician to preserve study integrity.

Treatment procedures

LAAO arm (group A, LAAO group)

Participants in group A will undergo percutaneous LAA closure. The choice of occlusion device—either the Amulet (St Jude Medical) or the Watchman (Boston Scientific, Natick, MA)—will be determined by the implanting centre. Procedural details, including imaging guidance and intraoperative management, will follow each centre’s standard protocols and the operator’s discretion, using either TEE or angiographic visualisation as appropriate. Preprocedural anticoagulation management will also be decided by the treating physician.

Non-vitamin K antagonist OAC arm (group B, DOAC arm)

Participants in group B will receive DOAC therapy according to current clinical guidelines. The specific agent (eg, rivaroxaban or apixaban) will be selected through shared decision-making between the physician and patient, considering factors such as clinical profile, insurance coverage and drug availability. Patients with contraindications to DOACs are excluded from this study. Antiplatelet therapy may be prescribed as deemed appropriate by the investigator, and adjustments to the antithrombotic regimen will be based on clinical judgement.

Crossovers and procedure timing

Patients randomised to one group may cross over to the other if medically indicated (eg, increased bleeding risk requiring discontinuation of DOAC therapy) or on patient request. All crossover events will be documented and reported. Participants may receive either sequential or concomitant PVI and LAA closure. Sequential treatment will apply to patients with prior PVI, while concomitant treatment will involve those undergoing combined ablation and LAA closure during the same session.

As this is an open-label study, neither participants nor investigators are blinded to treatment allocation.

Allocation generation, mechanism and implementation

Participant allocation will be performed using a computer-generated randomisation sequence within a validated clinical trial management system. The system employs a permuted block randomisation algorithm with variable block sizes to ensure balanced assignment throughout the study. Stratification factors, including study site and key baseline characteristics, are incorporated to maintain comparability across subgroups.

On confirmation of eligibility, participants will be automatically assigned to the next available treatment allocation in real time, with each assignment documented with a date and time stamp for audit purposes. The randomisation sequence will be securely stored and accessible only to authorised personnel who are not involved in patient recruitment or assessment. This computer-based approach ensures accurate, unbiased and concealed treatment allocation while preserving the integrity of the randomisation process. The system will also enable ongoing monitoring of allocation balance and generate reports for interim analyses.

Postimplant index procedure medication regimen

Postprocedure management will be tailored according to the patient’s bleeding risk profile. For patients with HAS-BLED (Hypertension, Abnormal renal/liver function, Stroke, Bleeding history or predisposition, Labile international normalized ratio, Elderly, Drugs/alcohol) <3, dual therapy with an OAC and either aspirin 100 mg once per day or clopidogrel 75 mg once per day will be prescribed for 3 months. After a 3-month TEE evaluation, if no thrombus, significant peridevice leak (≤5 mm) or device malposition is detected, OAC will be discontinued, and dual antiplatelet therapy with aspirin and clopidogrel will continue for another 3 months, followed by lifelong aspirin monotherapy (100 mg once per day).

For patients with HAS-BLED ≥3, monotherapy with OAC will be given for 3 months. After satisfactory TEE reassessment at 3 months, OAC will be discontinued and replaced with aspirin 100 mg once per day or clopidogrel 75 mg once per day for 3 months, followed by lifelong aspirin monotherapy (100 mg once per day).

Experience with the LAAC procedure

To minimise the influence of operator variability on outcomes, all participating centres are high-volume institutions with established expertise, each performing more than 10 LAAC procedures annually. Likewise, all surgeons involved in the study have personally performed at least 10 LAAC procedures per year. This focus on procedural experience and technical consistency ensures high-quality interventions, thereby enhancing the reliability and validity of study results.

Anticoagulation treatment arm (group B, DOAC group)

Participants assigned to group B will receive therapy with an approved DOAC—rivaroxaban, apixaban or dabigatran—at manufacturer-recommended dosages (for rivaroxaban, 15 mg or 20 mg once per day). Dosing will be individualised according to clinical factors such as comorbidities, body weight, age and renal function. Treatment adherence will be assessed at each follow-up visit through patient self-report and medication review to ensure compliance.

Endpoints

Primary efficacy endpoint

The primary efficacy endpoint of the study will be a composite of four components within 24 months following randomisation:

1. Stroke/TIA/systemic thromboembolism.

2. Cerebral haemorrhage and other major haemorrhages (Bleeding Academic Research Consortium (BARC) ≥2).

3. Cardiovascular mortality.

4. All-cause mortality events.

For patients who experience more than one event, only the first qualifying event will contribute to the primary endpoint analysis (time-to-first-event approach). Cardiovascular death is adjudicated as a subset of all-cause mortality and will be counted only once in the composite. A stroke is characterised by the sudden onset of focal or global neurological impairment, presenting with signs or symptoms consistent with stroke and categorised as either ischaemic or haemorrhagic. Diagnosis confirmation is necessary, typically performed by a neurology or neurosurgical specialist, through neuroimaging procedures, or via lumbar puncture. A TIA is defined as any neurological deficit not meeting the aforementioned criteria for stroke, specifically lasting less than 24 hours without evidence of new haemorrhage or infarction on imaging. Systemic embolism refers to acute vascular insufficiency or occlusion in the extremities or other non-central nervous system organs, supported by clinical, imaging or surgical/autopsy findings indicating arterial occlusion, excluding other probable mechanisms.

Secondary efficacy endpoint

1. Recurrence rate of AF.

2. Rehospitalisation rate related to AF.

3. Quality of life score (Atrial Fibrillation Effect on Quality-of-Life score).

Primary safety endpoint

7-day perioperative operation-related complications rate, including mortality, operation-related pericardial tamponade/pericardial effusion, ischaemic stroke/TIA/systemic thromboembolic events, cerebral haemorrhage and other major haemorrhages (BARC ≥grade 2), occluder dislocation, vascular access complications that require intervention or surgical treatment (such as arteriovenous fistulas and pseudoaneurysms, etc).

Monitoring of events and outpatient follow-up

Outpatient follow-up visits will be conducted at 1, 3, 6, 12, 18 and 24 months after randomisation. Cardiac echocardiography will be performed at the 1-month and 12-month visits to assess cardiac structure. At 12 and 24 months, a 24-hour Holter monitor will be used to evaluate cardiac rhythm. During each follow-up, medication use, treatment adherence and study endpoints will be assessed to ensure compliance with the prescribed regimen and to identify any factors requiring adjustment of the treatment strategy. Participants will be followed for 24 months as prespecified in the study protocol. Extended follow-up beyond this period will continue to evaluate the durability of treatment effects and monitor long-term safety outcomes.

All suspected endpoint events will be reviewed and adjudicated by an independent clinical events committee whose members are blinded to treatment assignment.

Sample size calculation/assumption

The sample size was calculated using PASS V.11 software (NCSS, LLC, Kaysville, Utah, USA). The calculation was based on the following assumptions: the power of the test=0.9, and the statistical significance level=0.05. Based on data from previous large-scale DOAC trials, the annual incidence of the primary endpoint was estimated to be 8% in the DOAC group and 7% in the interventional group.³³,³⁶ A non-inferiority margin of 5% was applied. Allowing for an anticipated 20% dropout rate, a total of 1012 participants (506 per group) will be required to test the non-inferiority hypothesis. Data analyses will be performed according to both the intention-to-treat and as-treated principles. To account for multiplicity, a hierarchical testing strategy will be applied: secondary endpoints will be formally tested only if the primary efficacy and safety endpoints achieve their prespecified criteria. If these are not met, secondary endpoints will be reported descriptively and considered exploratory. Missing data will be handled by multiple imputations.

Standard descriptive statistics will be used to summarise the data. Continuous variables will be expressed as medians with IQRs and compared using the Wilcoxon rank-sum test. Categorical variables will be presented as frequencies and percentages and compared using the Mantel-Haenszel χ² test. Time-to-event outcomes will be estimated using Kaplan-Meier curves, and differences between the LAAC and DOAC groups will be evaluated with log-rank tests. Logistic regression and Cox proportional hazards models will be applied, as appropriate, to assess the influence of baseline characteristics on clinical endpoints. All analyses will be conducted using SAS V.9.4 (SAS Institute, Cary, NC, USA).

Trial organisation

This investigator-initiated academic trial is supported by the Clinical Research Plan of the Shanghai Hospital Development Center, China (grant No SHDC2020CR1039B) and Shanghai Arrhythmia Research Center Project (grant No 2022ZZ01008). The Department of Cardiology at Shanghai Chest Hospital serves as the coordinating centre, responsible for the overall organisation and execution of the study.

Initially, eight large-volume tertiary cardiac centres from academic hospitals across China—most equipped with cardiac surgery facilities—have consented to participate. Additional academic centres are expected to join as the study progresses. Oversight is provided by the principal investigators and the PROMOTE Research Committee, which convenes bimonthly to fulfil three key responsibilities: (1) protocol development and oversight: maintaining a scientifically rigorous research design and ensuring participant safety; (2) ethical review and compliance: verifying adherence to ethical principles and regulatory requirements; (3) data monitoring and safety assessment: conducting regular safety reviews and recommending protocol adjustments or study termination when necessary.

Data collection will be web-based. The database and randomisation software were developed by an independent third party, Hangzhou Tigermed Consulting Co, which also manages data capture, database maintenance and statistical analysis. No investigators have direct access to the randomisation system or database. Furthermore, no external organisations—such as device manufacturers or pharmaceutical companies—were involved in the design, protocol development or conduct of this study.

The Data Monitoring Committee (DMC), consisting of three independent experts in clinical research, biostatistics and the relevant medical field, will perform interim analyses every 12 months to evaluate study safety, efficacy and overall progress. The DMC will review adverse events, assess protocol adherence and modifications, monitor patient recruitment and provide recommendations regarding study continuation. All reviews will be conducted with strict confidentiality and full independence. Blinded interim reports will be prepared by the statistical team and shared only with the principal investigator and the institutional review board (IRB).

Participant information will be collected, managed and protected in accordance with rigorous confidentiality and data-security standards. These include anonymisation of all datasets, secure electronic storage with encrypted access, restricted data access to authorised personnel and full compliance with applicable data-protection regulations. Access to the final trial dataset will be limited to approved research personnel and designated analysts under institutional data-use agreements to ensure confidentiality and research integrity.

The study results will be disseminated through peer-reviewed publications, conference presentations and clinical trial registries, with authorship based on substantial intellectual contribution and full protocol, participant-level data and statistical code made publicly accessible through a designated repository within 12 months of study completion.

Ethics approval

The study protocol has been reviewed and ethics approval obtained from the Shanghai Chest Hospital Ethical Review Board, Shanghai, China (approval reference number: KS(Y)20287). The study physician will obtain written informed consent from all study participants (consent form provided as online supplemental file 1). Protocol modifications will be promptly communicated to all key stakeholders, including the Research Ethics Committee/IRB, regulatory authorities and investigators, within 15 days of approval. Comprehensive documentation will detail the rationale, specific changes and potential study impacts. All modifications will be systematically logged and tracked to ensure transparency and regulatory compliance.

The study was registered with the Chinese Clinical Trial Registry (ChiCTR2000036538) at https://www.chictr.org.cn/showproj.html?proj=59270.

Discussion

The PROMOTE study is a prospective, randomised, multicentre trial designed to evaluate whether LAAC is a reasonable alternative to oral anticoagulation in patients following AF ablation. The incidence and prevalence of AF increase with age, and although PVI has advanced considerably in recent decades, the recurrence rate after ablation remains as high as 50%. Moreover, the risk of stroke after PVI persists, necessitating continued anticoagulation therapy. However, a growing number of patients require anticoagulation but have contraindications to long-term therapy. Previous studies have demonstrated that LAAC is non-inferior to both warfarin and DOACs for stroke prevention, yet no trial has specifically examined these strategies in the post-PVI population. Given the large and expanding number of patients undergoing PVI and the widespread use of DOACs, a randomised trial directly comparing LAAC and DOAC therapy in this setting is both timely and clinically significant.

This study has several potential limitations. The 24-month follow-up period may not capture very long-term outcomes or rare adverse events associated with either intervention. As an open-label, multicentre trial conducted exclusively in China, variations in operator experience and institutional practice may influence outcomes and limit generalisability to other populations. Furthermore, recruitment and follow-up were partially affected by the COVID-19 pandemic, which may have introduced minor delays or inconsistencies.

Ethics and dissemination

The study was approved by the Ethical Review Board of Shanghai Chest Hospital, China (KS(Y)20287). Written informed consent will be obtained from all participants. The trial will follow the Declaration of Helsinki and Good Clinical Practice. Confidentiality will be maintained with anonymised, securely stored data. Findings will be disseminated through peer-reviewed publications and conferences.