Aging leads to structural and electrophysiological changes that increase the risk of postoperative atrial arrhythmias; however, noninvasive preoperative markers of atrial proarrhythmic conditions are still needed. This study is aimed at assessing whether interatrial dyssynchrony determined using two-dimensional speckle tracking echocardiography relates to proarrhythmic structural and functional remodeling. A cohort of 45 patients in sinus rhythm referred for cardiac surgery was evaluated by echocardiography and surface electrocardiogram the day before the intervention. Transmembrane potential, connexin, and potassium channel distribution, inflammatory, and nitrooxidative markers were measured from right atrial tissue obtained from patients. A difference greater than 40 milliseconds between right and left atrial free wall contraction confirmed the presence of interatrial dyssynchrony in 21 patients. No difference in relation with age, previous diseases, and 2-dimensional echocardiographic findings as well as average values of global longitudinal right and left atrial strain were found between synchronic and dyssynchronic patients. Postoperative atrial fibrillation incidence increased from 8.3% in the synchronic group to 33.3% in the dyssynchronic ones. P wave duration showed no difference between groups. Action potentials from dyssynchronous patients decreased in amplitude, maximal rate of depolarization, and hyperpolarized. Duration at 30% of repolarization increased, being markedly shorter at 90% of repolarization. Only the dyssynchronous group showed early and delayed afterdepolarizations. Atrial tissue of dyssynchronous patients displayed lateralization of connexin 40 and increased connexin 43 expression and accumulation of tumor necrosis factor-α in the intercalated disc. Tumor necrosis factor-α did not colocalize, however, with lateralized connexin 40. Nitroxidative marks and KATP channels increased perivascularly and in myocytes. Our results demonstrate that, as compared to a traditional surface electrocardiogram, the novel noninvasive echocardiographic evaluation of interatrial dyssynchrony provides a better identification of nonaged-related proarrhythmic atrial remodeling with increased susceptibility to postoperative atrial fibrillation.
1. Introduction
Arrhythmias usually complicate cardiovascular surgery, and aging is the main risk factor. Atrial fibrillation is the most frequent sustained arrhythmia with a peak of appearance between the second and fifth days of the postoperative stay [1]. Its incidence progressively increases from 18% in sexagenarians to 50% in octogenarians [2]. Arrhythmic events are usually self-limited, and treatment frequently restores sinus rhythm. However, postoperative atrial fibrillation (POAF) prolongs hospital stay and increases the risk of stroke and mortality [1].
Atrial aging is an elusive prooxidative and proarrhythmic condition. Oxidative stress, hyperadrenergic states, and inflammation contribute to age-related atrial remodeling, but these tissue alterations are challenging to identify noninvasively [3, 4]. In this context, atrial enlargement, stiffness, and conduction blockade are known risk factors, but they are present in only a few patients. Therefore, the mechanisms involved in the onset and perpetuation of POAF are difficult to foresee [5]. Additionally, surgery per se facilitates arrhythmias due to ischemia-reperfusion injury, which increases the preexisting oxidative stress state. The lack of tools to estimate the proarrhythmic substrate evidences the absence of preventive interventions.
Structure and function of the beating heart can reveal atrial hidden oxidation and inflammation. Signs of tissue remodeling, before dilatation, arise from atrial dynamic cyclic changes. New echocardiographic techniques like strain and strain rate represent the magnitude and rate of myocardial deformation. Strain can reflect distensibility and atrial contractility [6]. Nonmyocytic cells and extracellular matrix mainly affect distensibility [7]. Cardiomyocyte structure and intercellular communication determine contractile function. Each atrial segment follows a trajectory during the cardiac cycle representative of the tissue physiology [8]. Electrical remodeling involves alterations in both myocytes and nonmyocytic cells, which manifest as contractile dyssynchrony in the echocardiogram, in a similar way as reported for the ventricles [9].
Gap junctions in the heart provide low resistance pathways for propagating the action potential across the myocardium, contributing to electrical coupling and signal propagation [10]. Alteration of cardiomyocyte gap junctions and their main components, connexins (Cx), has been suggested to contribute to the formation of arrhythmias, including atrial fibrillation [11–13]. Accumulating evidence also suggests that inflammation and oxidative stress are involved in atrial remodeling. Detection of protein 3-nitrotyrosine is regarded as a marker of nitrooxidative stress and is observed especially in inflammatory processes. The reaction of peroxynitrite with tyrosine leads to the formation of 3-nitrotyrosine and promotes protein, lipid, and DNA damage [14, 15]. ATP-regulated potassium channels (KATP) are well-characterized metabolic and oxidative sensors in ischemia/reperfusion arrhythmias and here postulated as an interesting substrate of POAF [16].
This study is aimed at assessing whether interatrial dyssynchrony, determined by using two-dimensional speckle tracking echocardiography, relates to proarrhythmic structural and functional remodeling of the atria and whether this increases the susceptibility for POAF.
2. Materials and Methods
2.1. Subjects and Ethical Considerations
Patients with coronary artery disease, aortic stenosis, or the combination of both pathologies, scheduled for surgery at the Department of Cardiac Surgery (Clinic of Cuyo, Mendoza, Argentina), were prospectively enrolled between January 2018 and March 2020. Coronary disease and aortic valve stenosis severity were defined according to current ESC Guidelines to determine surgery indication [17, 18]. All subjects provided written informed consent under the research protocol approved by the Ethics Committee of the National University of Cuyo (Exp-Cuy: 22959/2017).
Clinical data including age, gender, and history of previous myocardial infarction and heart failure was collected. The presence of preoperative atrial fibrillation was determined according to previous electrocardiographic reports or diagnosis in medical history. Information regarding the following cardiovascular risk factors was collected: hypertension, dyslipidemia (low-density lipoprotein cholesterol above 100 mg/dL or the use of lipid-lowering drugs), smoking (current or any smoking habit in the past ten years), and diabetes mellitus (previous diagnosis of diabetes mellitus or glycated hemoglobin greater than 6.5%). The preoperative use of medications was also documented.
2.2. Inclusion and Exclusion Criteria
Patients over 18 years of age in sinus rhythm with an indication of cardiovascular surgery that gave written informed consent were included in the study. Exclusion criteria were as follows: indication of mitral or tricuspid valve repair or replacement, being older than 80 years, history of previous atrial fibrillation, presence of moderate valvular disease or valvular prosthesis, history of congenital cardiac abnormalities or cardiac tumors, emergency surgery, inability to provide informed consent, and a not entirely detectable left and right atrial profile from the apical four-chamber view during preoperative echocardiography.
2.3. Preoperative Electrocardiogram
Before surgery, patients underwent a 12-lead electrocardiogram (ECG) using the Synchronous ECG software V1.3.5. Measurement in milliseconds of the P wave and the PR segment was performed with the software caliper. According to the current classification, these patients were evaluated for the presence of an interatrial conduction disturbance called Bayes syndrome [19]. There are two major categories for this syndrome: complete and incomplete, both based on P wave duration and morphology in the 12-lead ECG.
2.4. Echocardiography and Atrial Strain
All patients were imaged in a left lateral decubitus position using ESAOTE ultrasound system equipment (MyLab30Gold Cardiovascular) with a 2-4 MHz/PA240 probe. Two-dimensional speckle tracking strain imaging was performed from the apical position by an experienced technician. The average frame rate for analysis was 60-80 frames/s. During a single breath-hold, three consecutive cardiac cycles were stored digitally for off-line analysis in the four-, two-, and three-chamber view. The entire right and left atriums were carefully visualized to prevent walls’ dropout.
Measurements focused on evaluating the indexed volume of the left atrium and the area of the right atrium. In the parasternal long-axis or short-axis view, M-mode of the left ventricle chamber was measured for diameter and wall thickness. In the apical 4-chamber view, the left ventricle ejection fraction was determined using the Simpson measurement. Mitral and tricuspid inflows were recorded at the tip of the valve leaflets. The peak velocities of early and late diastolic filling waves (E wave and A wave) and the E/A velocity ratio were measured. The e velocity was obtained by tissue Doppler averaging the lateral and septal mitral annulus values and the e wave of the tricuspid lateral wall. E/e’ values were obtained from both ventricles.
Strain and strain rate datasets were analyzed using a wall motion tracking software (ESAOTE MyLab). In apical views (4-chamber, 3-chamber, and 2-chamber), left atrial endocardial boundaries were manually measured at the end-diastolic phase. Right atria were only registered in the 4-chamber view. The values of the reservoir, conduit, and atrial contraction strain were recorded according to the EACVI/ASE/Industry Task Force to standardize deformation imaging [20]. The values of strain rate for the reservoir phase, the conduit, and contraction phases were also recorded (Figure 1). The same was also done in 4 chambers for the right atrium. The left atrium was subsequently divided into basal, medial, and roof atrial segments. This determined a total of 15 segments for the left atria when the 3 apical views were added. The right atrium was evaluated similarly, but only the lateral segments were taken into account, adding 3 more segments, thus making a total of 18 segments. Septal segments were considered as left atria.
(a)