Atrial fibrillation (AF) is one of the most common complications following coronary artery bypass grafting or valve surgery. Many groups have tried to understand the underlying mechanism and have formulated different hypotheses to explain its origin. An imbalance of the autonomic nervous system after surgical intervention has been accepted as one of the determinants for the onset of postoperative AF.1–3
The intrapericardial and epicardiac innervation network of the heart has been studied extensively.4 However, some aspects of its physiology and role still remain unclear in particular in the context of cardiac surgery.
The surgical procedure of a cardiac denervation was described by Arnulff in 19395 with the goal to increase coronary perfusion. Melo et al, first addressed the problem of postoperative AF by surgical denervation. This group reported a reduction of postoperative AF after coronary bypass grafting in a series of patients.6 However, surgical denervation lead to permanent destruction of the nerval network and the sympathetic and parasympathetic balance. The long term implications of permanent denervation have not yet been fully determined.
By contrast, a temporary denervation may be achieved with a new method of blockade of the major ganglions which are located in the so called fat pads of the heart. We have assessed the role of FS injection as a method to modify conduction in the fad pads, mimic denervation and hereby reduce the incidence of postoperative AF in patients undergoing coronary artery bypass grafting or valve surgery.
Mazgalev and coworkers reported a Temporary Cardiac Denervation with Human FS in a proof of principle study in canines.7 An immediate modification of the conduction of electrical impulses through the cardiac ganglia was achieved following injection of 1 mL of the fibrin sealant (FS) Omrix.
These animal data support the hypothesis that the injection of the sinoatrial (SA) and atrioventricular (AV) node fat pad with surgical sealant may provide a temporary, protective parasympathectomy as a means to reduce postoperative AF. The effects on cardiac conduction through the fad pads were found to return to baseline at 4 weeks post injection. Hence, the effect was temporary and reversible. Histological assessment of the cardiac fat pad tissue was also observed to normalize at 4 weeks post injection.
According to the experimental data of Mazgalev and coworkers, the present pilot—study is aimed at creating a temporary conduction block through the SA and AV parasympathetic nodes by performing a biopolymer injection (human surgical sealant) into the cardiac fat pads and creating a biologic, non-ablative cardiac denervation to reduce the incidence and severity of AF after routine coronary artery bypass grafting or aortic valve surgery.
PATIENTS AND METHODS
Twenty eligible patients who underwent coronary bypass grafting or valve surgery in our Institution were included in this pilot—study. Patients with a previous history of atrial arrhythmias were excluded. Prophylactic medications for the prevention of AF were not permitted either before or after surgical intervention. Patients who had been receiving beta blockers before surgery, however, were eligible for enrolment and did not interrupt the treatment, which was also resumed postoperatively (Table 1).
SURGICAL TECHNIQUE AND DESCRIPTION OF PROCEDURE
After median sternotomy and opening of the pericardium, the nerves surrounding the large vessels at the base of the heart and that run from the right side of the superior vena cava and end at the level of the midportion of the anterior pulmonary artery were identified. The surgeon also identified the SA node fat pad that is located at the level of the right pulmonary vein–atrial junction, and the AV node fat pad that is located at the junction of inferior vena cava (IVC) and the left atrium and the corresponding cardiac ganglia using an electrostimulating detector (Fig. 1).
The detection of the ganglion was defined as positive if the heart rate could be reduced to 20% of base level under stimulation and returned to normal immediately after discontinuing the stimulation.
FS was injected into the fat pad around the identified spot using a 2-channel injector (containing 1 mL of thrombin and 1 mL of fibrinogen) and the 25 or 27 gauge needle. An amount of 1 to 2 mL CM-1 (Omrix) sealant was applied 1 to 2 mm under the epicardial surface of each of the 2 fat pads. CM-1 injection was performed slowly (0.1 mL/s) in one uninterrupted motion as the needle was slowly withdrawn. The injection was made in such a manner to create a visible “button” disk of CM-1 that approximates the visible size of the fat pad boundaries, injecting 1 to 2 mL into each fat pad (Fig. 2).
A total of up to 4 mL fibrin glue was delivered (2 mL to each fat pad). After this procedure, conventional coronary artery bypass grafting or valve surgery was performed either with or without cardiopulmonary bypass under normothermia or moderate hypothermia. Myocardial protection was achieved by using warm blood cardioplegia as described by Calafiore et al.8
Detection of AF after surgical intervention was based on daily 24 hour monitoring during the first 4 postoperative days in all patients. Telemetry was used in all patients between day 4 and 7. A 24-hour holter electrocardiogram was performed 1 day before discharge.
In all patients the major parasympathetic ganglion within the SA-fat pad could be identified by electric stimulation. This was defined either by a reduction of the basic heart rate of 20% (n = 4) or by a complete temporary onset of a SA block (n = 16). The second ganglion within the AV-fat pad could not be identified with certainty in respect to the above defined indicators. Injection of CM-1 was performed easily and safely in all patients. Post injection stimulation of the target area could not reproduce a lowering of the heart rate or a SA-block.
No major postoperative complications were observed, including no in-hospital or 30-day mortality. However, 1 patient (5%) died on the 33rd postoperative day. This 1 patient, who was discharged with permanent AF, suffered pneumonia and respiratory failure during his stay in a rehabilitation unit and finally died in septic shock in another hospital. This event was apparently not related to the particular CM-1-sealant treatment. There were no infectious events during the postoperative course, and no need for permanent pacemaker placement.
Twenty percentage of the studied population (n = 4, patients) developed postoperative AF. One patient (5%) showed persistent AF at the time of discharge.
In 1 patient, temporary pacemaker stimulation was required due to sinus bradycardia (Tables 1 and 2).
Postoperative AF is a frequent complication after major cardiac surgery that may be associated with other serious adverse events such as cerebral emboli. It increases postoperative health care resource utilization and prolongs the hospital stay.
The onset of AF after surgery varies with a maximum between the second and fourth postoperative day. However, a later onset up to 14 days after surgery is possible. Some of these events are asymptomatic or paroxysmal and only telemetry allows for an exact diagnosis.9
Medical treatment focuses on conversion and limitation of AF frequency. Beta blockers and Amiodarone are the agents used most. If pharmacologic therapy is not successful within a period of a few days, electric cardioversion is usually performed to reestablish a normal sinus rhythm.10
The literature is rich with evidence of the anatomic and physiologic cardiac innervation in humans.4,11 Apart from the intercellular mechanism of the conducting system, the balance between sympathetic and parasympathetic innervation may play an important role as a source for the development of postoperative AF. Many authors hypothesized that re-entrant phenomena are one of the primary mechanisms for AF.12–14 A parasympathetic dominance prolongs the refractory time for cellular electric stimulation and transduction, hence it may increase the re-entry capability of the atrial myocardium.15
The parasympathetic ganglions play an important role for the parasympathetic innervations of the heart. The anatomic pathways are multiple and mostly discrete, however, some major ganglions are located within the aortic fat pad and the adjacent aorta and pulmonary artery tissue as well as within the interatrial groove and within a fat pad, which is located in a triangle between the IVC and the coronary sinus (see Fig. 3). All those fat pads can be accessed surgically. Melo et al6 assessed the feasibility and efficacy of permanent operative ventral cardiac denervation by dividing these anatomic structures and the nervous network between the ganglia. According to Melo et al, this fast and low risk procedure achieved a significant reduction of both the incidence and severity of AF after routine coronary artery bypass surgery. Patients younger than 70 years of age are expected to have a higher success rate than those older than 70 years.6
In contrast to a permanent destruction of the parasympatic net surrounding the heart, Mazgalev and coworkers reported a temporary cardiac denervation with human FS in a proof of principle study in canines.7 An immediate modification of the conduction of electrical impulses through the cardiac ganglia was achieved following injection of 1 mL of CM-1 (Omrix) human surgical sealant.
The above listed data supported the hypothesis that injection of the SA and AV node fat pad with human surgical sealant may provide a temporary, protective parasympathectomy as a means to reduce postoperative AF following cardiac surgery in the same matter as described by Melo et al. In contrast to the permanent blockage, the effects on cardiac conduction through the fad pads were observed to return to “normal” (baseline) at 4 weeks post injection. Hence, the effect was temporary and reversible. Histologic assessment of the cardiac fat pad tissue was also observed to be “normal” at 4 weeks post injection, without the development of scared tissue or inflammatory reaction.7
The aim of this pilot—study was to evaluate the feasibility and safety of a temporary fat pad ablation and its vagal control of the SA node, AV node and the onset and severity of AF after cardiac (mostly coronary bypass) surgery in human beings.
The surgeon identified the SA node fat pad that is located at the level of the right pulmonary vein–atrial junction, and the AV node fat pad that is located at the junction of IVC and the left atrium with an electric stimulation device. The identification of the major ganglion within the interatrial groove was easy and 100% successful (in 80% with a total arrest of the conduction). The smaller ganglion, located close to the IVC, was usually not clearly identified by an abrupt lowering of the heart rate. A single injection of CM-1 was performed under the epicardial surface of each of the 2 fat pads.
Surgery was prolonged for at mean of 1 minute for detecting the ganglia and injecting the sealant. There was no surgical event. In the first postoperative period (in hospital), there was no major adverse event in respect to death, myocardial infarction, cerebral event, infection, or SA-/AV-block. There was 1 patient who needed temporary pace maker stimulation for 48 hours due to sinus bradycardia. No patient needed permanent pacemaker stimulation.
Four patients (20%) developed postoperative AF within the first 10 days after surgery. Three of the AF events were temporary and persisted less than 48 hours. Only in 1 patient (5%) the arrhythmia could not be converted into sinus rhythm during the 10 day postoperative period. This patient was the only one, who needed specific antiarrhythmic medication after discharge. In follow-up, 1 patient died during rehabilitation, due to a severe pneumonia and respiratory failure. This event, although within the first 33 days after surgery was obviously not related to the particular CM-1 sealant treatment.
The results of this first feasibility protocol showed, that the injection of a polymer, such as CM-1 sealant is safe and does not affect normal atrial conduction in humans. According to these first results, the effect on the development and duration of postoperative AF cannot be determined due to the small patient sample and lack of a control group. In face of the observed residual postoperative AF rate of 20% in our study population, CM-1 sealant treatment of the 2 major ganglia within the interatrial groove and between the IVC and CS does not appear to be sufficient to prevent all postoperative AF. Further clinical trials that include larger patient populations and a more careful investigation and interpretation of the electrocardiogram variability under rest and exercise are needed going forward. In addition, randomized controlled trials to confirm the long term benefit of temporary autonomic denervation of parasympathetic ganglia as a prophylactic adjunct to prevent postoperative AF in humans are needed.
Post cardiac surgery AF is a frequent complication which increases health care resource utilization and may be associated with serious adverse events. The rationale of a temporary reduction of the parasympathetic influence on the atrial conduction follows that of prior animal experiments. Two major ganglion structures within 2 fad pats have been treated in the presented feasibility study. This report represents the first human experience with FS injected into the SA and AV fat pad in patients undergoing low to moderate risk open heart surgery and showed it to be safe and feasible. However, further research is needed to clearly document the benefit in respect to preventing or alleviating AF following cardiac surgery.
1. Hogue C, Dimitrovich P, Stein P, et al. RR interval dynamics before atrial fibrillation in patients after coronary bypass graft surgery. Circulation
2. Kalman J, Munawar M, Howes LG, et al. Atrial fibrillation after coronary artery bypass grafting is associated with sympathetic activation. Ann Thorac Surg
3. Hogue C, Hyder M. Atrial fibrillation after cardiac operations: risks, mechanism and treatment. Ann Thorac Surg
4. Bargmann W, Doerr W. Das Herz des Menschen
. Stuttgart: Georg Thieme Verlag; 1963.
5. Arnulff G. De la section du plexus pré–aortique: justification et technique. Presse Med
6. Melo J, Voigt P, Sonmez B, et al. Ventral cardiac denervation reduces the incidence of atrial fibrillation after coronary artery bypass grafting. J Thorac Cardiovasc Surg
7. Seil O, Zhang Y, Bibevski S, et al. Vagal denervation and atrial fibrillation inducibility: epicardial fat pad ablation does not have long term effects. Heart Rhythm
8. Calafiore AM, Teodori G, Mezzetti A, et al. Intermittent antegrade warm blood cardioplegia. Ann Thorac Surg
9. Creswell LL, Schuessler RB, Rosenblom M, Cox JL. Hazards of postoperative atrial arrhythmias. Ann Thorac Surg
10. Aerra V, Kuduvalli M, Moloto AN, et al. Does prophylactic sotalol and magnesium decrease the incidence of atrial fibrillation following coronary artery bypass surgery: a propensity—matched analysis. J Cardiothorac Surg
11. Brobeck JR. Best & Tailor’s Physiological Basis of Medical Practice
. 10th ed. Baltimore: Williams & Wilkins; 1979.
12. Cox JL, Canavan TE, Schuessler RB, et al. The surgical treatment of atrial fibrillation. II. Intraoperative electrophysiologic mapping and description of the electrophysiologic basis of atrial flutter and atrial fibrillation. J Thorac Cardiovasc Surg
13. Cox JL. A perspective on postoperative atrial fibrillation in cardiac operations. Ann Thorac Surg
14. Konings KT, Kirchhof CJ, Smeets JR, et al. High-density mapping of electrically induced atrial fibrillation in humans. Circulation
15. Davis Z, Jacobs HK, Bonilla J, et al. Retaining the aortic fat pad during cardiac surgery decreases postoperative atrial fibrillation. Heart Surg Forum