Atrial structure and function in middle‐aged, physically‐active males and females: A cardiac magnetic resonance study

Abstract Recent studies have reported on an association between endurance sport, atrial enlargement and the development of lone atrial fibrillation in younger, male cohorts. The atrial morphology and function of middle‐aged, physically‐active males and females have not been well studied. We hypothesized that middle‐aged males would demonstrate larger left atrium (LA) and right atrium (RA) volumes compared to females, but atrial function would not differ. LA and RA volume and function were evaluated at rest in healthy adults, using a standardized 3.0Tesla cardiac magnetic resonance protocol. Physical activity, medical history, and maximal oxygen consumption (V˙O2peak) were also assessed. Physically‐active, middle‐aged men (n = 60; 54 ± 5 years old) and women (n = 30; 54 ± 5 years old) completed this study. Males had a higher body mass index, systolic blood pressure, and V˙O2peak than females (p < .05 for all), despite similar reported physical activity levels. Absolute and BSA and height‐indexed LA and RA maximum volumes were higher in males relative to females, despite no differences in ejection fractions (p < .05 for all). In multivariable regression, male sex p < .001) and V˙O2peak (p = .004) were predictors of LA volume (model R 2 = 0.252), whereas V˙O2peak (p < .001), male sex (p = .03), and RV EF (p < .05) were predictors of RA volume (model R 2 = 0.377). While middle‐aged males exhibited larger atrial volumes relative to females, larger, prospective studies are needed to explore the magnitude of physiologic atrial remodeling and functional adaptations in relation to phenotypic factors.


| INTRODUCTION
Long-term vigorous physical activity and exercise have been associated with cardiac morphological adaptations known as the "Athlete's Heart". 1 The extent of this remodeling in the atria has been well characterized in younger male and female elite endurance athletes using echocardiography. 2,3 However, there is paucity of literature describing atrial structure and function in middle-aged physicallyactive adults, despite the high participation rates of recreational to sub-elite middle-aged endurance athletes. 4 This middle-aged cohort exhibits cardiac morphological adaption in response to years of an intensive exercise-training burden. We have recently characterized cardiac structure and function in this cohort using both electrocardiography 5 and echocardiography, 6 demonstrating significant and heterogeneous remodeling in the absence of increased atrial ectopy.
A recent echocardiography study has reported an association between larger left atrial size and higher cardiorespiratory fitness in physically-active adults across the lifespan. 7 While echocardiography is a common first tool to evaluate cardiac structure, it underestimates cardiac volumes in comparison to the gold standard of cardiac magnetic resonance (CMR) imaging. 8 While CMR has been applied to quantify atrial morphology in younger elite athletes, 9,10 only a single, small study has been completed in middle-aged males consisting of 10 former elite athletes and five controls. 11 The extent of atrial remodeling in this cohort may be particularly salient, as studies suggest a potential predilection for exercise induced cardiac remodeling and lone atrial fibrillation, 12 particularly in male athletes. 13 In fact, female athletes are less likely to develop atrial fibrillation at a younger age than men, and unlike men, there is no association between longstanding exercise dose and risk of atrial fibrillation. 13 Therefore, the primary objective of the current study was to characterize left atrium (LA) and right atrium (RA) morphology, as well as LA strain in middle-aged, physically-active adults. We hypothesized that males would demonstrate larger LA and RA volumes compared to females, but atrial function would not differ.

| General study procedures
This study was part of a larger investigation examining the cardiovascular phenotypic characteristics of healthy, physically-active males and females. 5,6 All participants were between 45 and 65 years of age with a long-standing history (>10 years) of physical activity participation. Exclusion criteria included a history of smoking, hypertension, diabetes mellitus, the use of cardio active drugs, or a history of chronic disease. Detailed medical and sport histories were obtained to confirm eligibility, and a physical examination was performed by a cardiologist. Participants were asked to abstain from caffeine (12 hours), alcohol (12 hours), and exercise (24 hours) prior to each study visit. The institutional research ethics board in accordance with the Declaration of Helsinki approved this study protocol. All participants provided written informed consent.

| Exercise training history
All participants completed a two-week exercise diary detailing the mode, duration, and intensity of each work out. A Likert scale was completed to determine the consistency of exercise training during the preceding 10 years and the training burden was determined by calculating both the h/week and the modality-specific metabolic equivalent of a task (MET)Áh/week of vigorous exercise. 14 2.3 | Resting blood pressure and maximal exercise testing

| CMR assessment and analysis
On a separate day, resting CMR in the supine position was performed to evaluate cardiac morphology. Images were acquired using a 3.0 Tesla magnet (Siemens MAGNETOM Skyra 3.0T with TIM and DOT technology) using a phased-array cardiac coil and retrospective electrocardiographic gating, completed by one of two operators blinded to group and clinical status. Steady-state free precession images were obtained during breath hold at end-expiration. CINE images were acquired to obtain a contiguous short axis stack (slice thickness of 8 mm, no gaps), as well as long-axis 4-chamber and 2-chamber images, as previously reported by our group. 16  software (Circle Cardiovascular Imaging, Inc, Calgary, Canada) was used to perform LA analyses. An automated tracking algorithm was applied, and manual adjustments were performed as needed to attain optimal wall tracking. We computed longitudinal atrial strain as (L 1 À L 0 )/L 0 , where L 1 is the change of atrial myocardial length throughout the atrial cycles and L 0 is the resting (or reference) length in a relaxed state at diastasis (end of atrial diastole). body mass index (p = .005), and BSA (p < .001) compared to females.

| Statistical analysis
Males and females did not differ with respect to exercise training burden (both p = .113), but males did have a higher relative _ VO 2peak (p = .024). Males had a higher resting SBP (p = .002) and DBP (p = .003) compared to females.

| CMR atrial morphology and sex
Absolute atrial volumes as well as volumes indexed to both BSA and height are reported (Tables 2-3

| CMR Assessment of right versus left atrial remodeling
The relative magnitude of atrial remodeling in the LA versus RA

| Predictors of left and right atrial remodeling
In univariate analysis, _ VO 2peak (r = 0.37, p < .001), resting HR (r = À0.31, p = 0.003) were independent predictors of LA max volume indexed to height. Point bi-serial correlation demonstrated a positive association between male sex and LA max volume (r = 0.42, p < .001, Figure S1). Notably, SBP was not significantly associated with LA max volume (r = 0.11, p = .281). In multivariable regression

| DISCUSSION
In the current study, we characterized atrial structure and function in healthy, middle-aged adults. Our study cohort was unique in that

| Atrial volumes and sex
Although the magnitude of cardiac remodeling varies considerably across individual athletes, male sex was the most significant predictor of LA maximum volume (indexed to height) in our multivariable regression analyses. Our findings support previous work where males have larger atrial volumes than females. 19 In particular, Mosen et al. 9 observed less atrial remodeling in female versus male endurance athletes when analyzing the ratios of LA and RA to total heart volume. In contrast, Letnes et al. 7 also observed negligible differences in atrial volumes when adjusting for cardiorespiratory fitness.
The observation of larger atria in male athletes may be due to a number of underlying mechanisms. In the current study, males had higher systolic BP than females, which may contribute to atrial remodeling. 20 In addition, a larger atria may be in part attributed to androgenic hormones that influence cardiac protein synthesis. 21 Skeletal muscle mass, training volume, and plasma volume expansion may also influence the cardiovascular adaptations associated with training. 22 Importantly, a recent randomized controlled trial demonstrated no sex differences in the magnitude of left atrial remodeling in previously sedentary adults who completed 24 months of high-intensity training. 23 Future investigation into the functional parameters of the LA and RA both at rest and during exercise may help to distinguish the influence of biological sex on atrial morphology and response to pathological conditions.

| Atrial volumes and training
A recent meta-analysis involving 16 echocardiographic studies of athletes (primarily males in high dynamic sports) reported that indexed LA volumes were unrelated to the type of sport performed. 24 Our atrial morphology data in middle-aged, physically-active adults (ranging from recreationally-active to sub-elite) advances our understanding of atrial volumes among males and females who have engaged in long-standing physical activity participation, as previous work often compared younger, male elite athletes to sedentary adults. 9,11,[25][26][27] Despite being the largest CMR study assessing atrial volumes in active middle aged persons to date, our sample size was not sufficient to rigorously assess the association between atrial volumes and physical activity modality.

| Atrial strain, endurance training, and sex
Our current CMR study findings revealed atrial enlargement and strain values within normal physiological range in middle-aged, physically-active adults. While our middle-aged males had greater atrial enlargement than females, females exhibited superior atrial strain which may be secondary to their attenuated atrial remodeling response. Our findings expand upon previous echocardiographic studies focused primarily on younger, elite male athletes. In particular, a meta-analysis of nine two-dimensional speckle tracking echocardiographic studies observed that global LA longitudinal strain data was marginally lower in elite athletes relative to untrained athletes (mean ages ranging from 22 to 43 years of age). 30 Notably, these studies primarily examined male athletes (7 of 9 studies had 100% male participants and 2 of 9 studies has 100% female participants), and therefore, were unable to evaluate the influence of endurance training and sex on atrial function. Recent work has also shown that worse atrial strain with higher atrial volumes in middleaged male athletes may be associated with atrial fibrillation. 31 Whether this occurs as a result of structural, functional or electrical remodeling could not be determined with the current study.

| Allometric scaling of the atria in athletes
The scaling of atrial volume warrants consideration. We reported various allometric scaling data in the present paper (Table S1)

| Clinical implications
Whilst we demonstrated significantly more absolute and relative atrial enlargement in middle-aged males relative to females, the clinical significance, such as a propensity to atrial arrhythmias in the future, remains unclear. A body of literature has emerged linking cardiac remodeling to lone AF, particularly in male relative to female athletes. 13 Female athletes are less likely to develop atrial fibrillation at a younger age than men, and unlike men, there is no association between longstanding exercise dose and risk of atrial fibrillation. 13 This finding supports the emerging concept that absolute atrial size, as opposed to "indexed" atrial size, may be the prime determinant of the propensity toward AF.
Furthermore, recent CMR studies utilizing advanced atrial tissue characterization techniques to quantify scar tissue demonstrate a correlation between atrial fibrosis and AF risk. 34 We were unable to perform high resolution atrial isotropic scans to assess the presence of atrial fibrosis. More studies are required to better characterize the atrial remodeling response at the tissue level in physically-active cohorts, and to establish the association between "pathologic" remodeling of the atria, which includes absolute atrial size, markers of "excessive" or pathologic atrial fibrosis and the risk for lone atrial fibrillation in athletic cohorts. 12

| Limitations
Our study findings should be considered in light of the following limitations. We utilized a cross-sectional design, which precludes any attribution of causality between atrial volumes and long-term endurance exercise training. The self-reporting of physical activity history may have introduced recall and response bias. We had a smaller cohort of physically-active female participants; yet, we still observed that males have larger atrial volumes relative to females who per-

| CONCLUSIONS
The current study used gold standard CMR methods to measure atrial volumes and strain obtained from middle-aged, physically-active male and females. Although males had larger LA and RA volumes compared to the females, females exhibited superior atrial strain and strain rate values. Nonetheless, we observed significant heterogeneity in atrial structure among physically-active, middle-aged adults. Larger, prospective studies are needed to explore the magnitude of physiologic cardiac remodeling and functional adaptations in relation to atrial fibrosis, exercise training volume, and other phenotypic factors.

CONFLICT OF INTEREST
The author declares that there is no conflict of interest that could be perceived as prejudicing the impartiality of the research reported.

DATA AVAILABILITY STATEMENT
Author elects to not share data.