Impact of Ligament of Marshall Resection on Atrial Fibrillation Occurrence in Patients Undergoing Coronary Artery Bypass Grafting: A Randomized Controlled Trial

Overview

Background: 20-40% of patients with ischemic heart disease managed with coronary artery bypass grafting (CABG) surgeries face the risk of postoperative atrial fibrillation (POAF), which is a known hazardous outcome. Being a well-known arrhythmogenic pathway, surgical excision of the ligament of Marshall may, theoretically, lower the incidence of POAF. Aim: In this study, we aimed to assess the safety and efficacy of prophylactic LoM resection in preventing POAF after on-pump CABG (ON-CAB.). Methods: 220 patients at elevated risk for POAF undergoing elective ON-CAB were randomized to receive either standard CABG alone (n = 110) or adjunctive LoM resection (n = 110) in this multicenter, randomized, assessor- and patient-blinded study with unblinded surgeons. ECG-verified POAF occurring within seven days was the primary outcome. Mortality, use of resources, and pharmaceutical therapy were secondary outcomes.

INTRODUCTION: Postoperative atrial fibrillation (POAF) is the most common arrhythmia after coronary artery bypass grafting (CABG), complicating 20-40% of cases and imposing a significant clinical and economic burden, including increased morbidity, stroke risk, prolonged hospitalization, and healthcare costs \[1,2\]. The pathophysiology of POAF is multifactorial, involving atrial structural remodeling, autonomic imbalance, inflammation, and perioperative stress \[3,4\]. Despite advances in surgical and medical management, including routine beta-blockade, prevention of POAF remains a persistent clinical challenge, particularly in high-risk patients. The ligament of Marshall (LoM) has emerged as a critical anatomical substrate in atrial arrhythmogenesis. This embryological remnant of the left superior vena cava contains myocardial fibers, sympathetic and parasympathetic innervation, and extensive connections to the left atrium \[5,6\]. Experimental and clinical evidence implicates the LoM as a source of focal triggers and a conduit for periatrial reentry, capable of initiating and sustaining atrial fibrillation (AF) \[7,8\]. The proven efficacy of catheter-based ablation targeting the LoM or adjacent vein of Marshall in reducing AF burden \[9,10\] underscores the translational potential of prophylactic surgical resection during CABG. Given this electrophysiological rationale, surgical resection of the LoM during CABG represents a feasible prophylactic intervention that could modify the atrial substrate at its source. The procedure is anatomically straightforward during median sternotomy and adds minimal operative time. However, robust evidence from multicenter randomized controlled trials is lacking, leaving surgeons without definitive guidance on its routine implementation. Preliminary single-center studies have reported promising but inconsistent results, limited by non-randomized designs, small sample sizes, and heterogeneous patient populations \[11,12\]. Consequently, the role of LoM resection as a preventive strategy remains undefined and has not been incorporated into surgical guidelines. Therefore, we conducted this multicenter, randomized controlled trial to evaluate the efficacy and safety of prophylactic LoM resection during elective CABG. We hypothesized that adjunctive LoM resection would significantly reduce the incidence of early postoperative atrial fibrillation in high-risk patients, without increasing major perioperative complications or prolonging ICU or overall hospital stay. METHODS: "Although conducted across three academic centers, all hospitals share a unified surgical, anesthetic, and postoperative care pathway, minimizing inter-center variability." \- Study Design and Oversight: Data on 220 patients who underwent elective ONCAB in Cairo, Sohag and Assiut university hospitals in the period between January 2021 and November 2025 was prospectively gathered and analyzed in accordance with the CONSORT 2010 guidelines \[13\] and aimed to evaluate the efficacy and safety of prophylactic ligament of Marshall (LoM) resection during elective coronary artery bypass grafting (CABG). \- Definitions and Criteria Ischemic Heart Disease Indications for surgical revascularization (ONCAB) were determined in accordance with contemporary guidelines issued by the American College of Cardiology (ACC), American Heart Association (AHA), and the Society for Cardiovascular Angiography and Interventions (SCAI) \[14\]. All enrolled patients had multivessel coronary artery disease or left main coronary artery disease for which coronary artery bypass grafting (CABG) was the recommended revascularization strategy. Postoperative Atrial Fibrillation (POAF) The primary outcome, postoperative atrial fibrillation, was defined as any episode of atrial fibrillation or atrial flutter lasting ≥30 seconds and occurring within the first 7 postoperative days. Diagnosis required documentation by either: (1) continuous 5-lead telemetry monitoring with automated arrhythmia detection and daily review by a blinded cardiologist, or (2) a standard 12-lead electrocardiogram. Both symptomatic and asymptomatic episodes were included, irrespective of the need for therapeutic intervention. This definition is consistent with consensus criteria from contemporary international guidelines \[15-17\]. High Risk for POAF Patients were classified as high risk for POAF using a validated composite risk definition. Eligibility required the presence of at least two of the following established preoperative risk predictors \[18,19\]: \- Age ≥65 years. * Left atrial enlargement, defined as an anteroposterior diameter \>40 mm on preoperative transthoracic echocardiography. * Documented history of paroxysmal atrial fibrillation not requiring long-term anticoagulation. * Chronic obstructive pulmonary disease (COPD) requiring regular pharmacological treatment. Participants: Patients were eligible for inclusion if they were aged 18 years or older and scheduled for elective coronary artery bypass grafting (CABG). All enrolled patients were considered at high risk for postoperative atrial fibrillation. Written informed consent was obtained from all participants prior to enrollment. Patients were excluded if they had preoperative atrial fibrillation requiring active treatment, a history of previous cardiac surgery, severe left ventricular systolic dysfunction (ejection fraction \<30%), contraindications to anticoagulation or antiarrhythmic therapy, or if they required emergency CABG. \- Ethical approval: The ethical committee of Assiut University hospitals approved the study protocol, Assiut University, Egypt. Ethical Approval Number (IRB registration number): 04-2025-300780. Every patient signed an informed written consent for acceptance of the operation. This work has been carried out in accordance with The Code of Ethics of the World Medical Association (Declaration of Helsinki) for studies involving humans \[20\]. \- Randomization and Blinding: Eligible patients were randomized in a 1:1 ratio to either the intervention group (coronary artery bypass grafting with ligament of Marshall resection) or the control group (standard coronary artery bypass grafting). Randomization was performed using a computer-generated, stratified block randomization sequence, with variable block sizes of 4 and 6, and stratification according to participating center and age (\<65 vs. ≥65 years) to ensure balance across treatment arms. Allocation concealment was achieved using sequentially numbered, opaque, sealed envelopes, prepared by an independent statistician. Envelopes were opened in the operating room after induction of anesthesia and prior to sternotomy. Due to the nature of the surgical intervention, blinding of the operating surgeons was not feasible. However, intensive care unit staff, postoperative care teams, cardiac anesthesiologists involved in postoperative monitoring, electrophysiologists adjudicating arrhythmia events, and outcome assessors were all blinded to group allocation. Continuous ECG telemetry recordings were independently reviewed by two cardiologists blinded to treatment assignment, with discrepancies resolved by consensus. \- Surgical Protocol: Preoperative Management: All patients received standardized preoperative care including beta-blocker continuation (unless contraindicated), statin therapy, and glycemic control for diabetic patients. Anesthetic Technique: A standardized general anesthetic protocol was used across all centers: induction with midazolam (0.03-0.05 mg/kg), fentanyl (5-10 μg/kg), and propofol (1-2 mg/kg); maintenance with sevoflurane (1-2%) and continuous infusion of remifentanil (0.1-0.3 μg/kg/min). Neuromuscular blockade was achieved with rocuronium (0.6 mg/kg). In both groups, A standard on-pump coronary artery bypass grafting (ON-CAB) technique was used. All operations were performed through a median sternotomy. The left internal mammary artery was harvested (commonly pedicled unless skeletonization was indicated as in diabetic patients), and additional conduits, including the radial artery or saphenous vein, were obtained when required. Systemic anticoagulation was achieved using intravenous heparin to maintain an activated clotting time exceeding 480 seconds. Cardiopulmonary bypass (CPB) was established via ascending aortic cannulation and two-stage venous cannulation of the right atrium. Moderate systemic hypothermia (32-34°C) was maintained throughout CPB, with alpha-stat pH management. Myocardial protection was provided using antegrade cold blood cardioplegia with a 4:1 blood-to-crystalloid ratio, administered through the aortic root, with repeated doses every 20 minutes or following completion of each distal anastomosis. Distal coronary anastomoses were performed during a single aortic cross-clamp period using 7-0 or 8-0 polypropylene sutures. Proximal anastomoses were constructed on the ascending aorta using a partial occlusion clamp and 6-0 polypropylene sutures. After completion of all anastomoses, patients were rewarmed to 36.5°C and successfully weaned from CPB, followed by reversal of systemic heparinization with protamine sulfate administered at a 1:1 ratio. Intervention-Specific Technique: LoM Resection Group The LoM resection was performed after pericardiotomy but before initiation of cardiopulmonary bypass, with the heart in situ: Step-by-Step Surgical Technique: 1. Exposure: The heart was elevated with a moist laparotomy pad placed posteriorly, tilting the left ventricle anteriorly and rightward. This exposed the left lateral pericardium and the area between the left pulmonary artery superiorly and the left pulmonary veins inferiorly. 2. Identification of Anatomical Landmarks; Figure 1: o Superior boundary: Left pulmonary artery. o Inferior boundary: Superior pulmonary vein. * Medial boundary: Left atrial appendage. * Lateral boundary: Left phrenic nerve (which was carefully identified and preserved). 3. Identification of LoM: The LoM appears as a whitish, cord-like structure running obliquely in the epicardial fat pad between the left superior pulmonary vein and the left atrial appendage (Figure 1). It was palpated as a tense band when gentle traction was applied to the atrial appendage. Figure 1; Intraoperative Identification of LoM LOM: Ligament of Marshall, Blue Asterisk: left pulmonary artery, Red Triangle: Left upper pulmonary vein, white Arrow: Left atrial appendage, white dotted line: course of the left phrenic nerve 4. Dissection Technique: The overlying epicardium was incised longitudinally using Metzenbaum scissors to expose the ligament of Marshall (LoM). The LoM was then carefully dissected circumferentially from the surrounding adipose tissue using a combination of sharp and blunt dissection. A right-angled clamp was subsequently passed beneath the LoM to develop a clear plane between the structure and the underlying left atrial wall. 5. Resection: After gentle mobilization of the left atrial appendage, the left atrial ridge was exposed, allowing clear visualization of the ligament of Marshall (LoM) as a fibrofatty bundle containing myocardial fibers, autonomic nerve fibers, and small vascular structures. Careful sharp and blunt dissection was performed to free the LoM from surrounding tissues, with meticulous attention to avoid injury to adjacent structures, particularly the left circumflex coronary artery and the left pulmonary veins. Following complete isolation, the proximal and distal ends of the LoM were secured either by ligation with 5-0 polypropylene sutures or by application of two medium hemoclips placed proximally near its insertion into the coronary sinus region and distally near the left atrial appendage. The intervening segment was then excised completely, ensuring no residual LoM tissue remained. 6. Hemostasis: Meticulous hemostasis was achieved using bipolar cautery. Any small venous tributaries (potential connections to the vein of Marshall) were clipped or coagulated. 7. Operative Time: The entire LoM resection procedure added a mean of 6.2 ± 2.1 minutes to the operative time (measured from identification to completion of hemostasis). Quality Control: All participating surgeons underwent a standardized training session including video demonstration and supervised practice on two cases before independent performance. Intraoperative photographs of the dissected LoM were required for the first 5 cases at each center for central review. Control Group: Patients underwent identical CABG procedure without any manipulation or dissection in the region of the LoM. The left lateral pericardial reflection was left undisturbed. \- Postoperative Management Protocol A standardized postoperative protocol was implemented across all centers: Monitoring: All patients were monitored with continuous 5-lead telemetry for 7 days postoperatively. A 12-lead ECG was obtained immediately postoperatively, daily for 7 days; The study focused on early POAF, which represents \>90% of AF events after CABG and is the most clinically relevant window, and whenever symptoms occurred or telemetry suggested arrhythmia. POAF Definition and Detection: POAF was defined as an irregular rhythm without discernible P waves lasting ≥30 seconds, confirmed by 12-lead ECG or two independent cardiologists reviewing telemetry strips. Asymptomatic episodes detected only on telemetry were included. POAF Management Protocol: * First episode: Rate control with beta-blockers (metoprolol) or calcium channel blockers (diltiazem) if hemodynamically stable. * Persistent AF \> 24 hours or hemodynamic instability: Chemical cardioversion with intravenous amiodarone (150 mg bolus over 10 minutes, then 1 mg/min for 6 hours, then 0.5 mg/min for 18 hours). * Electrical cardioversion for hemodynamic compromise refractory to medication. Anticoagulation: Initiated for AF lasting \>48 hours (therapeutic enoxaparin or unfractionated heparin bridging to warfarin, target INR 2.0-3.0). Standard Care: Uniform protocols for analgesia, fluid management, ventilator weaning, transfusion triggers (hemoglobin \< 8.0 g/dL or \< 9.0 g/dL with cardiac ischemia), and chest tube removal (output \< 200 mL/24 hours). \- Outcome Measures Primary Outcome: Incidence of postoperative atrial fibrillation within 7 days after surgery. Secondary Outcomes: 1. Safety Outcomes: * In-hospital all-cause mortality * Reoperation for bleeding/tamponade * Stroke (confirmed by neurology consultation and imaging) * Renal failure requiring dialysis * Deep sternal wound infection * Perioperative myocardial infarction (Universal Definition, 4th Edition) 2. Resource Utilization: * Total duration of inotropic support (hours) * Total dose of inotropic drugs (inotropic equivalent score) * Volume of chest tube drainage in first 48 hours (mL) * Number of units of blood products transfused * Need for and duration of amiodarone infusion (as a downstream clinical consequence rather than an independent endpoint). * Length of ICU stay (hours) * Total postoperative hospital stay (days) 3. Procedure-Specific Outcomes: * Operative time (skin-to-skin) * Cross-clamp time * Cardiopulmonary bypass time * Conversion to other arrhythmia surgery - Data Collection and Management: Data were collected prospectively using standardized case report forms. Key timepoints included: * Preoperative: Demographics, medical history, medications, laboratory values, echocardiographic parameters. * Intraoperative: Surgical times, technical details, complications, transfusion requirements * Postoperative (hourly for 24h, then daily): Vital signs, rhythm status, medication administration, drainage output, laboratory results. * Discharge: Summarized outcomes, complications, discharge medications. Data were entered into a secure, web-based electronic data capture system (REDCap) with automated range and consistency checks. Source data verification was performed on 20% of randomly selected cases by an independent monitor. Sample Size Calculation Based on previous single-center studies \[11,12\] and institutional data, we estimated a POAF incidence of 35% in the control group. To detect an absolute risk reduction of 15% (to 20% in the intervention group) with 80% power and a two-sided alpha of 0.05, 100 patients per group were required. Accounting for a 10% dropout/loss-to-follow-up rate, the total target sample size was 220 patients (110 per group). Statistical Analysis Analyses followed the intention-to-treat principle. Continuous variables were presented as mean ± standard deviation for normally distributed data or median with interquartile range for skewed data, and compared using Student's t-test or Mann-Whitney U test as appropriate. Categorical variables were presented as frequencies and percentages and compared using chi-square or Fisher's exact test. The primary outcome (POAF incidence) was analyzed using a chi-square test. A multivariable logistic regression model was constructed to adjust for potential confounders identified a priori: age, left atrial diameter, history of paroxysmal AF, and CPB time. Adjusted odds ratios with 95% confidence intervals were reported. Time-to-event data (time to first AF episode) were analyzed using Kaplan-Meier curves and log-rank test. All statistical tests were two-sided, with P \< 0.05 considered statistically significant. No interim efficacy analyses were planned. All analyses were performed using SPSS version 27.0 (IBM Corp.) and R version 4.2.1. Ethical Considerations The study posed minimal additional risk as LoM resection was performed during standard CABG without prolonging ischemic time. Potential risks included bleeding from the dissection site and injury to adjacent structures (phrenic nerve, left atrial wall), which were minimized by standardized technique and surgeon training. Patients were informed they could withdraw at any time without affecting their clinical care.