Recent data have suggested a relation among long-term endurance sport practice, left atrial remodeling, and atrial fibrillation. We investigated the influence of an increased vagal tone, represented by the early repolarization (ER) pattern, on diastolic function and left atrial size in professional soccer players. Fifty-four consecutive athletes underwent electrocardiography, echocardiography, and exercise testing as part of their preparticipation screening. Athletes were divided into 2 groups according to presence or absence of an ER pattern, defined as a ST-segment elevation at the J-point (STE) ≥0.1 mm in 2 leads. For linear comparisons average STE was calculated. Mean age was 24 ± 4 years. Twenty-five athletes (46%) showed an ER pattern. Athletes with an ER pattern had a significant lower heart rate (54 ± 9 vs 62 ± 11 beats/min, p = 0.024), an increased E/e′ ratio (6.1 ± 1.2 vs 5.1 ± 1.0, p = 0.002), and larger volumes of the left atrium (25.6 ± 7.3 vs 21.8 ± 5.0 ml/m 2 , p = 0.031) compared to athletes without an ER pattern. There were no significant differences concerning maximum workload, left ventricular dimensions, and systolic function. Univariate regression analysis revealed significant correlations among age, STE, and left atrial volume. In a stepwise multivariate regression analysis age, STE and e′ contributed independently to left atrial size (r = 0.659, p <0.001). In conclusion, athletes with an ER pattern had an increased E/e′ ratio, reflecting a higher left atrial filling pressure, contributing to left atrial remodeling over time.
Recent data have documented a relation between long-term endurance sport practice and atrial fibrillation. In 1 study, 44% of athletes with paroxysmal atrial fibrillation at 40 years of age were formerly engaged in soccer. Because of acyclic nature and intensity, soccer is classified as a high-intensity intermittent team sport. Echocardiographic data have suggested that structural remodeling is often present in the left atrium of endurance athletes and a significant proportion (20%) showed enlarged atria according to established normal values. Underlying mechanisms for structural changes of the atrium and development of atrial fibrillation in athletes are not entirely clear. Increased parasympathetic tone, increased inflammatory markers, and remodeling of the left atrium have been discussed. Possibly, an increased left atrial pressure leads to atrial remodeling and dilatation. Endurance sport is associated with a high vagal tone and experimental data have suggested an increased left atrial pressure after vagal stimulation. The early repolarization (ER) pattern is one of the most common alterations in the athlete’s electrocardiogram. It is present especially in subjects engaged in basketball, volleyball, and soccer, and it is a surrogate for an increased vagal tone. We hypothesized that athletes with the ER pattern have higher left atrial filling pressures and larger left atria compared to athletes without this pattern.
Methods
From May to August 2008 consecutive professional soccer players underwent routine history, physical examination, electrocardiography, echocardiography, and exercise testing on a bicycle ergometer as part of their preparticipation screening for the national soccer league. Examinations were performed in the morning, ≥24 hours after the last strenuous training. All players had similar training methods, although there was some variability based on a player’s position. All players were engaged in professional soccer for >2 years. Athletes were divided into 2 groups according to presence or absence of the ER pattern. A follow-up visit was planned after 12 months. Written informed consent was obtained from each participant.
Twelve-lead electrocardiograms were obtained with a subject in supine position and recorded at 50 mm/s (Cardio Soft 5, GE Medical Systems, Freiburg, Germany). Heart rate at rest was determined after 1 minute in a supine position. Electrocardiographic patterns were evaluated according to commonly adopted clinical criteria. The ER pattern was defined as an upward ST-segment elevation ≥0.1 mV in 2 peripheral or precordial leads, beginning from an elevated J-point. For linear comparisons, average ST-segment elevation at the J-point (STE) was calculated from leads I, II, III, aVF, aVL, and V 2 to V 6 . The Sokolov index was calculated from the S- and R-wave voltages in leads V 1 and V 5 or V 6 . Average precordial T-wave amplitude was calculated from leads V 2 to V 6 .
Exercise testing was conducted according to recommendations of the American Heart Association. Cycle ergometry (Ergoline ERG 900, Bitz, Germany) was performed in the upright position. The exercise protocol included 3 minutes of cycling at 50 W and progressive loadings of 50 W every 3 minutes to exhaustion. Participants were encouraged to reach ≥85% of predicted maximal achievable heart rate (220 minus age in years). Midway through each stage of exercise, at peak exercise, and 1 minute to 3 minutes after cessation of exercise, data on symptoms, heart rate, blood pressure (as measured by indirect arm-cuff sphygmomanometry), and estimated workload (as determined from standard tables) in METs (1 MET equals 3.5 ml of oxygen uptake per kilogram of body weight per minute) were collected.
Standard transthoracic echocardiography was performed according to recommendations of the European Association of Echocardiography on a Vivid 7 Dimension machine (GE Vingmed, Horton, Norway; M3S 2.5-MHz transducer). Images were stored digitally and analyzed off-line (EchoPac PC, GE Vingmed). For all measurements 3 beats were stored and analyzed. Left atrial and ventricular volumes, left ventricular (LV) end-diastolic diameter, and LV mass were calculated according to current recommendations and indexed for body surface area. Pulse-wave Doppler was performed in the apical 4-chamber view to obtain peak early filling (E wave) and late diastolic filling (A wave) velocities, E/A ratio, deceleration time of the early filling wave, and the isovolumic relaxation time. Pulse-wave tissue Doppler imaging was performed in the apical 4-chamber view to acquire peak septal (e′ septal) and lateral (e′ lateral) mitral annular velocities and to calculate the average mitral annulus velocity (e′). The frame rate for tissue Doppler imaging measurements was >100 frames/s. Timings of aortic valve opening, peak systolic contraction, and aortic valve closure were measured in relation to the beginning of the QRS complex. Methods of image acquisition and postprocessing of strain measurement with 2-dimensional speckle tracking have been described previously. Global longitudinal strain and global systolic and diastolic strain rates were calculated for the entire U-shaped length of the LV myocardium (basal, mid, and apical segments of 2 opposite walls in each view). All images were obtained at a frame rate of 60 to 80 frames/s.
Data were analyzed with SPSS 17.0 Windows (SPSS, Inc., Chicago, Illinois). We evaluated numerical data for a normal distribution using the Kolmogorov-Smirnov test. Parametric data are presented as mean ± SD. Statistical comparisons of parametric data were made with Student’s t test for 2-group comparisons. Simple linear regression analysis was performed to determine the relation between pairs of continuous variables. In addition, a multivariate, stepwise linear regression analysis was performed to identify independent determinants of left atrial volume. Participants were dichotomized according to the median value of age in younger and older athletes for age comparisons. A 2-sided p value <0.05 was considered to indicate statistical significance.
Results
We included 54 athletes with a mean age of 24 ± 4 years in this study. All subjects were asymptomatic without a history of atrial fibrillation. Electrocardiogram showed mild abnormalities in 33 athletes (61%). Six (11%) showed a first-degree atrioventricular block, 15 (28%) showed isolated voltage criteria for LV hypertrophy, and 25 (46%) had an ER pattern. Eighteen athletes (33%) showed the ER pattern only in the precordial leads and 7 athletes (13%) showed an additional ER pattern in the inferior leads. Athletes with an ER pattern had a significant lower heart rate, a shorter QTc interval, and a larger Sokolov index compared to athletes without an ER pattern ( Table 1 ). The exercise test showed no significant differences concerning baseline and maximum systolic blood pressures, maximum heart rate, and maximum workload between groups ( Table 1 ).
| Variable | Early Repolarization Pattern | p Value | |
|---|---|---|---|
| No | Yes | ||
| (n = 29) | (n = 25) | ||
| Age (years) | 24 ± 4 | 25 ± 5 | 0.470 |
| Body mass index (kg/m 2 ) | 23.4 ± 1.2 | 23.8 ± 1.0 | 0.239 |
| Body surface area (m 2 ) | 2.03 ± 0.09 | 2.02 ± 0.12 | 0.923 |
| Electrocardiogram | |||
| Heart rate at rest (beats/min) | 62 ± 11 | 54 ± 9 | 0.024 |
| P duration (ms) | 104 ± 12 | 107 ± 9 | 0.270 |
| PQ duration (ms) | 156 ± 24 | 161 ± 31 | 0.525 |
| QRS duration (ms) | 98 ± 8 | 101 ± 7 | 0.314 |
| QRS degree (°) | 75 ± 24 | 76 ± 32 | 0.953 |
| QTc duration | 415 ± 22 | 399 ± 24 | 0.017 |
| Sokolov index (mV) | 2.8 ± 0.7 | 3.4 ± 0.8 | 0.009 |
| Mean T-wave amplitude (μV) | 641 ± 158 | 703 ± 268 | 0.325 |
| Exercise test | |||
| Systolic Riva Rocci blood pressure at rest (mm Hg) | 115 ± 10 | 117 ± 9 | 0.543 |
| Peak systolic Riva Rocci blood pressure (mm Hg) | 182 ± 24 | 192 ± 30 | 0.214 |
| Peak heart rate (beats/min) | 167 ± 16 | 168 ± 14 | 0.880 |
| Peak heart rate (percent maximum) | 85 ± 8 | 86 ± 6 | 0.705 |
| Maximum workload (MET) | 13.1 ± 1.6 | 13.5 ± 1.4 | 0.335 |
| Echocardiogram | |||
| Left atrial volume (ml/m 2 ) | 21.8 ± 5.0 | 25.6 ± 7.3 | 0.031 |
| Ventricular septum (mm) | 10.5 ± 0.8 | 10.8 ± 1.0 | 0.427 |
| Left ventricular end-diastolic diameter (mm/m 2 ) | 25.3 ± 1.8 | 25.2 ± 2.0 | 0.974 |
| Left ventricular mass (g/m 2 ) | 99.8 ± 16.0 | 102.0 ± 16.0 | 0.646 |
| Left ventricular ejection fraction (%) | 57.8 ± 3.4 | 57.1 ± 3.2 | 0.447 |
| Peak E wave (cm/s) | 74 ± 12 | 74 ± 11 | 0.915 |
| Peak A wave (cm/s) | 45 ± 0.11 | 45 ± 0.11 | 0.877 |
| E/A | 1.73 ± 0.43 | 1.73 ± 0.45 | 0.999 |
| Peak e′ septal (cm/s) | 12.8 ± 2.3 | 11.3 ± 1.3 | 0.008 |
| Peak e′ lateral (cm/s) | 16.7 ± 3.2 | 13.6 ± 2.9 | 0.009 |
| Peak e′ average (cm/s) | 14.7 ± 2.4 | 12.4 ± 1.8 | <0.001 |
| E/e′ | 5.11 ± 0.96 | 6.10 ± 1.20 | 0.002 |
| Isovolumic relaxation time (ms) | 75 ± 10 | 76 ± 10 | 0.621 |
| E-wave deceleration time (ms) | 140 ± 15 | 150 ± 17 | 0.044 |
| Time to aortic valve opening (ms) | 85 ± 17 | 88 ± 16 | 0.606 |
| Time to peak contraction (ms) | 338 ± 33 | 351 ± 37 | 0.208 |
| Time to aortic valve closure (ms) | 384 ± 20 | 395 ± 22 | 0.072 |
| Global strain (%) | −17.9 ± 1.8 | −17.7 ± 2.3 | 0.716 |
| Systolic strain rate (1/s) | −0.98 ± 0.13 | −0.95 ± 0.15 | 0.444 |
| Diastolic strain rate (1/s) | 1.40 ± 0.21 | 1.35 ± 0.25 | 0.471 |
Left atrial volume was mildly abnormal (>29 ml/m 2 ) in 6 athletes (11%) and moderately abnormal (>34 ml/m 2 ) in 2 athletes (4%). Mild LV hypertrophy (LV mass index >115 g/m 2 ) was present in 7 athletes (13%) and moderate LV hypertrophy (LV mass index >131 g/m 2 ) was present in 2 athletes (4%). No athlete showed an abnormal LV end-diastolic diameter index. There was no correlation between left atrial volume and LV mass or LV end-diastolic diameter. Five athletes (9%) had a mildly abnormal ejection fraction (EF; <55%), no EF was <50%. We found a significant correlation between EF and global longitudinal strain (r = 0.70, p <0.001). Concerning diastolic function, the E/e′ ratio was in normal limits (<9) in all athletes. Lateral peak e′ was higher than septal peak e′. The ER pattern was associated with larger volumes of the left atrium, an increased E/e′ ratio, a lower septal, lateral and average peak e′, and a longer deceleration time of the E wave. There were no significant differences between groups concerning LV dimensions, LV mass, and systolic function, expressed as EF, global longitudinal strain, and strain rate ( Table 1 ).
Univariate regression analysis revealed significant correlations between age and left atrial volume, between STE and left atrial volume, and between STE and E/e′ ( Figure 1 ). We found no relevant correlation among E/A ratio, isovolumic relaxation time, deceleration time, E/e′, and left atrial volume. In a stepwise multivariate regression analysis age, STE, and e′ contributed independently to left atrial size (r = 0.659, p <0.001). Colinearity was low (tolerance >0.73).
