Fibrosis and scarring of the myocardial tissue result in a reduced voltage of the electrical signals from the affected area. Electro-anatomical voltage mapping (EAVM) shows a good correlation between pathological findings and the electrophysiological characteristics of fibrosis [7]. Several studies have shown a strong correlation between left atrial fibrosis and AF recurrence using EAVM. Verma et al. [8] used EAVM to measure the extent of LA fibrosis (<0.5 mV) and its prognostic value for AF recurrence after pulmonary vein isolation (PVI) in 700 patients. LA scarring was the only predictor associated with higher AF recurrence [8]. Another study using EAVM prior to PVI provided strong evidence that LA fibrosis can be found in a significant proportion of patients with paroxysmal AF [9]. The presence and extent of atrial fibrosis did not correlate with any other potential risk factor (clinical type, LA diameter, age, duration of AF, etc.) in these studies. Kottkamp et al. suggested a new entity — fibrotic atrial cardiomyopathy (FACM) [10–12]. FACM is described as a specific preexisting chronic progressive disease of the LA predisposing patients to AF in the presence of modulators (inflammation, hypertension, obesity) and triggers (pulmonary vein foci). The authors suggested 5 stages of severity of fibrosis based on EAVM, defining normal voltage as above 1.5 mV: FACM 0 without any fibrosis; FACM I with moderate fibrosis (1.0–1.5 mV); FACM II with one area of severe fibrosis (<0.5 mV); FACM III with at least two areas of severe fibrosis, and FACM IV with severe diffuse fibrosis of the LA.

METHODS

Patients

All patients with a history of symptomatic paroxysmal or persistent AF who met the inclusion and exclusion criteria were consecutively included in this prospective observational study immediately before the planned PVI procedure. The exclusion criteria included: permanent AF, previous PVI, left ventricular ejection fraction <50% measured by echocardiography, history of myocardial infarction, acute or chronic inflammatory disease, history of malignant neoplasm, and hypo- or hyperthyroidism. In the period between March 2019 and June 2020, a total of 125 patients in whom PVI was performed were initially included. Subsequently, 11 patients who did not have an adequate electro-anatomical map of the LA according to the protocol described here were excluded from the study. The research was approved by the ethics committee at the University of Osijek Faculty of Medicine (KLASA: 602-04/19-08/04; URBROJ: 2158-61-07-19-09). All patients signed their informed consent.

Clinical parameters

The following clinical parameters were examined: age, sex, BMI, type of AF (paroxysmal, persistent), drugs and preparations taken (including antiarrhythmics and anticoagulants), as well as history of hypertension, coronary heart disease, diabetes, and cerebrovascular stroke. BMI was determined according to the definition of the World Health Organization: normal weight patients with BMI <25 kg/m2, overweight patients with BMI between 25 and 30 kg/m2, and obese patients with BMI ≥30 kg/m2. Left atrial volume measures were assessed by standard echocardiography.

Assessment of left atrial fibrosis

A detailed anatomical map (fast anatomical mapping [FAM]) and a voltage map (EAVM) of the LA were created using a three-dimensional mapping system and an ablation catheter with a Biosense Webster Smart Touch contact force sensor (CARTO 3, Biosense Webster) [10, 11]. The map was created exclusively in sinus rhythm, taking a minimum of 400 individual points evenly distributed throughout the LA. Points with catheter contact force of less than 5 g and more than 20 g were excluded. The program displays surfaces of the same voltage in the same color. Surfaces with a voltage criterion of ≤0.5 mV (low voltage zones [LVZ]) are shown in red, which indicates advanced tissue fibrosis. Areas with moderate fibrosis of voltage criterion between 0.5 and 1.5 mV are shown in blue and green. Healthy tissue of voltage criteria above 1.5 mV is shown in purple. The density of the points, defined by the distance between two points, is set at a minimum of 5 mm. The system is programmed to take only points at the end of expiration that meet the criteria for a minimum contact force of 5 g and adequate catheter stability. As a result of the EAVM, the following parameters were calculated using Biosense Webster CARTO 3 software: the volume of the LA, the total surface area of the LA, the number and total surface area of LVZs (described by voltage points ≤0.5 mV) and the share of LVZs in the surface of the LA.

Data analysis

Data was analyzed and visualized using the Python programming language (version 3.8.3) and Jupyter Notebooks [22]. Percentages of fibrosis were assigned first to two groups (NO group if fibrosis was less than 5%, and YES group if it was above 5%), then to four groups (NO for fibrosis <5%, I 5%–20%, II 20%–35%, and III above 35%). Areas of fibrosis were divided into four groups (NO if a patient was without fibrosis, I for 1 fibrotic area, II for 2 fibrotic areas, and III for 3 and more fibrotic areas). The dependent variable was tested for normality using the Shapiro–Wilk test of normality before performing ANCOVA to determine the differences in BMI between groups of patients with age, sex, hypertension, diabetes, and persistent AF as confounders. T-test and Mood’s median test, as appropriate, were used to assess differences between groups stratified by the presence of fibrosis in quantitative variables, while Fisher’s exact test was employed to examine respective differences in qualitative characteristics. All the tests were performed using SciPy, statsmodels, and pandas packages. Quantitative data are reported as mean (standard deviation [SD]) for normally distributed variables or median (interquartile range [IQR]) for not normally distributed data, while categorical variables are presented as the number of individuals (% of total).

RESULTS

The population of the study consisted of 114 patients (65% male and 35% female). The baseline characteristics of all patients, as well as the two groups of patients when stratified by the presence of significant LA fibrosis, are described in Table 1.

Association of LVZs and BMI

In the current cohort we assessed the association between BMI and groups of patients with and without significant (<5%) fibrosis (Figure 2).

Patients without significant fibrosis had a mean BMI (SD) of 29.3 (4.5) kg/m2, while patients with significant fibrosis had a mean BMI (SD) amounting to 28.9 (4.3) kg/m2. After adjusting for age, sex, hypertension, diabetes, and persistent AF, we found that no association remained statistically significant (P = 0.84). We further assessed the association between BMI and four groups of patients stratified by percentage of fibrotic area: NO for fibrosis <5%, I 5%–20%, II 20%–35%, and III above 35%. The mean (SD) BMI in these four groups was 29.2 (4.5) kg/m2, 29.3 (4.4) kg/m2, 28.2 (2.8) kg/m2, and 28.7 (5.2) kg/m2, respectively (Figure 3).

After adjusting for age, sex, hypertension, diabetes, and persistent AF, BMI was not significantly different between the groups (P = 0.57).

Similarly, no association was found between BMI and four groups of patients stratified by the number of fibrotic areas: NO if a patient was without fibrosis, I for 1, II for 2 and III for 3 and more (Figure 4)

.

Mean BMI (SD) in the four groups was 28.8 (4.7) kg/m2, 28.6 (4.3) kg/m2, 29.3 (4.4) kg/m2, and 29.5 (4.4) kg/m2, respectively (P = 0.64).

A subgroup of 21 (18%) patients with normal BMI was additionally analyzed. In this group, 14 (67%) patients had significant fibrosis of the LA. Patients with normal BMI and significant LA fibrosis had a median CHA2DS2-VASc score of 2.5 with IQR 2–3. Persistent AF was found in 4 (29%) patients, arterial hypertension in 9 (64%) patients, and 7 (50%) patients had diffuse LVZ covering ≥90% of the LA surface.

DISCUSSION

To our knowledge, this study is the first to investigate the association between excess body weight and LA fibrosis in humans. The main result of this study is that no association was found between the prevalence and the extent of LA fibrosis and BMI in patients undergoing PVI.

In our study, only 18% of patients were of normal weight, while 41% of patients were obese. The high prevalence of obesity in the investigated cohort could partly explain why 60% of patients had hypertension, 8% had diabetes, 12% had asymptomatic coronary artery disease, and 5% had a history of cerebral vascular incidents. There is no evidence of any association between these co-morbidities and atrial fibrosis. Weight loss results in a significant reduction of the AF burden, comparable with PVI results. However, we found no correlation between obesity and atrial fibrosis, which is the strongest risk factor for AF recurrence after PVI, suggesting other possible mechanisms for the higher AF burden in obese and overweight patients. There is strong evidence of a correlation between LA fibrosis and a worse prognosis for AF in terms of the progression of the disease and ineffective invasive therapy. Previous studies showed a strong correlation between AF and risk factors such as advanced age, hypertension, diabetes, obesity, and obstructive sleep apnoea. However, studies investigating some of these factors and the presence of left atrial fibrosis showed no correlation, suggesting a new idiopathic entity — fibrotic atrial cardiomyopathy.

There are some indications of a possible pro-inflammatory, thus a profibrotic effect of obesity on left atrial tissue, but no correlation has been established. Important findings from the ovine model have been published by Abed et al. [23]. In this study, animals were on a high-calorie diet that resulted in weight gain. The animals underwent electrophysiological studies, cardiac magnetic resonance imaging, and histology at baseline, 4- and 8-monthly intervals. Weight gain resulted in progressive remodeling characterized by conduction heterogeneity, bi-atrial enlargement, and increased pericardial fat volumes. Histology showed increased atrial interstitial fibrosis and inflammation [23].

Apart from the role of cardio-metabolic factors associated with obesity in promoting AF, several other factors have been observed and investigated. Diastolic function impairment has a strong correlation with the risk for AF onset [24, 25]. Possible mechanisms include enlargement of the LA area with consequent changes in the electrical properties of the tissue that promotes re-entry. Cardiac fat deposits overlying the LA also seem to be an important factor in promoting AF in obese patients. Investigators from the Framingham Heart Study Offspring and Third Generation Cohorts analyzed 2317 patients and showed an association of total pericardial fat volume measured by computed tomography and prevalence of AF [26]. Wong et al. [27] showed that pericardial fat, measured by magnetic resonance in this study, is also a predictor of AF recurrence after PVI. Possible mechanisms are local inflammation mediated by paracrine mediators and the arrhythmogenicity of interpolated fat tissue [28].

Strengths and limitations of the study

A high number of patients with excess body weight was included in this study, giving strength to evidence on LA fibrosis within this population. However, the proportion of these patients is not consistent with other studies investigating the extent of LA fibrosis. Body mass index is strongly correlated with the development of AF [19], progression from paroxysmal to persistent AF, [20] and with AF burden, [21] and was the only parameter of general obesity used in this study. Epicardial adipose tissue (EAT) emerged as an important factor in AF pathogenesis in recent studies. Metaanalysis showed a highly significant correlation between BMI and EAT, but the correlation was even greater between EAT and waist circumference and between EAT and visceral adipose tissue [29]. These indices of obesity should also be included in future research on left atrial fibrosis. This study investigated the extent of LA fibrosis only in patients with AF, therefore the results could not be extrapolated to the healthy population. Although a correlation between low voltage zones in the LA and histological findings of fibrosis has been confirmed, exact voltage cut-off values have not been established yet. Commonly used voltage cut-off values in most of the studies for normal tissue, moderate fibrosis, and severe fibrosis are ≥1.5 mV, 0.5–1.5 mV, and ≤0.5 mV, respectively. In this study, we observed only areas of severe fibrosis (≤0.5 mV). Areas of moderate fibrosis were not analyzed in this study but might also play an important role in the development of FACM.

Electro-anatomical voltage mapping can give inconsistent results. Various catheters are used in studies to collect voltage points within the body of the LA. Multi-polar catheters can collect a greater number of voltage points over the same period of time than bipolar catheters, resulting in more detailed EAVM. Multipolar catheters use tissue impedance for contact assurance, while new generation bipolar catheters use contact force criteria (≥5 g/mm2).

Article information

Acknowledgements: This research was supported by the project “GLYCARD: Glycosylation in Cardiovascular Diseases” (UIP-2019-04-5692), funded by the Croatian Science Foundation. We would like to express our immense appreciation to our dear friend Maroje Sorić, PhD, assistant professor at the Faculty of Kinesiology of the University Zagreb, whose advice was crucial for this research. We are also particularly grateful for the assistance provided by our friend and colleague Nikola Jutriša from the electrophysiology lab at the Magdalena Clinic.

Conflict of interest: None declared.

Open access: This article is available in open access under Creative Common Attribution-Non-Commercial-No Derivatives 4.0 International (CC BY-NC-ND 4.0) license, allowing to download articles and share them with others as long as they credit the authors and the publisher, but without permission to change them in any way or use them commercially. For commercial use, please contact the journal office at kardiologiapolska@ptkardio.pl.

How to cite: Szavits Nossan J, Šesto I, Štambuk K, et al. Is the extent of left atrial fibrosis associated with body mass index in patients undergoing pulmonary vein isolation for atrial fibrillation? Kardiol Pol. 2021; 79(5): 531–538, doi: 10.33963/KP.15910.

REFERENCES