Endurance events are rapidly growing in popularity, as part of a healthy lifestyle in many individuals. Although the benefits of regular, moderate exercise are undisputed in the vast majority of the population, the impact of prolonged exercise on the heart is less well characterized. Previous research has shown that long-distance running leads to reversible attenuation of global cardiac function, a phenomenon referred to as cardiac fatigue.

Moreover, detectable levels of cardiac troponin occur reproducibly in some participants in endurance exercise, a finding that correlates to the presence of wall motion abnormalities on echocardiography. This has contributed to concerns regarding possible detrimental effects of such events on the cardiovascular system. Although adverse cardiovascular events are uncommon in athletes, less trained individuals run a higher risk. Interestingly, it has been shown that cardiac fatigue and its associated biomarker release occur predominantly in subjects with less training. This suggests that appropriate preparation will afford protection against cardiac dysfunction associated with participation in endurance exercise events. Importantly, published reports in this field made this discovery post hoc, not in a prospective study design.

Intraventricular dyssynchrony is known to develop in the failing heart and has recently received considerable attention in the context of cardiac resynchronization therapy for heart failure. Although the benefit of CRT is limited to patients with severe disturbances of cardiac conduction and wide QRS complexes, mechanical dyssynchrony has also been described in patients with narrow QRS complexes. In fact, because dyssynchrony is seen in such a diverse range of cardiac disorders, it is increasingly felt to constitute an unspecific sign of myocardial failure. Several methods have been proposed to quantify intra–left ventricular (LV) mechanical dyssynchrony. Many studies have used one or more indices of the dispersion of cardiac mechanical activity on the basis of tissue velocities, typically the time from QRS onset to peak systolic velocity. One widely established technique was pioneered by Yu et al. and involves measuring the standard deviation across 12 myocardial segments. Synchrony may be altered following a brief bout of exercise, as shown by studies in patients with systolic heart failure and normal ejection fractions. However, previous research has not clearly shown whether participants in endurance exercise too may exhibit increased mechanical dyssynchrony. If so, this would suggest that athlete’s cardiac fatigue has phenomenologic similarities to clinical heart failure, albeit on a very mild scale.

We wished to study experienced long-distance race participants compared with new participants and control for age, and we therefore enrolled subjects into a prospective, age-matched design. Besides conventional biochemical analyses we studied the insertion/deletion (I/D) polymorphism of the angiotensin-converting enzyme (ACE) gene, on the basis of reports that this genetic trait plays a role for the heart’s response to prolonged exercise. This is important because D allele carriers have elevated levels of angiotensin II and exhibit adverse cardiac remodeling on activation of the renin-angiotensin system, which places them at an increased risk in the setting of myocardial disorders. We hypothesized that especially inexperienced long-distance runners exhibit increased dyssynchrony after racing, and we sought to establish the impact of carrying the ACE D allele in this setting.

Methods

Subjects

We approached male runners who were registered to participate in the Lidingöloppet 30-km cross-country race, held in September 2008 in Stockholm, Sweden. We recruited a group of new participants who had never taken part in this event before (new participants), as well as an age-matched group of runners with ≥8 previous Lidingöloppet 30-km race participations (repeat participants). Participants completed a questionnaire regarding training habits (hours of exercise per week). The runners’ weights and heights were recorded immediately before and after the race (body mass index was calculated as weight in kilograms divided by the square of height in meters), their blood pressures were recorded by cuff sphygmomanometry, and they underwent blood tests and echocardiography. We also recorded electrocardiograms at rest in a subgroup to ensure that the dyssynchrony measurements used were not influenced by possible alterations of depolarization in this group of trained athletes. Electrocardiography was not repeated after the race, as we have previously shown that QRS width does not increase after endurance running. Cardiovascular fitness was quantified post hoc as peak exercise capacity in a subgroup of participants who underwent standardized supine bicycle stress testing to volitional exhaustion. This study was approved by the institutional committee on human research, and all participants provided written informed consent.

Blood Tests

Blood specimens were taken 99 ± 41 min before the race started and repeated 44 ± 17 min after runners had crossed the finishing line. Blood samples were kept on ice and centrifuged within 4 hours of collection. Cardiac troponin T (TnT) was analyzed without prior freezing (Roche Diagnostics Scandinavia, Bromma, Sweden), and the cutoff for elevation was ≥0.03 μg/L (the clinical decision limit for myocardial necrosis), per the manufacturer’s advice.

Conventional Echocardiography

Baseline echocardiography was performed following a prolonged rest period 1 to 2 days before the race in 35 runners and approximately 3 months after the race in eight subjects (four new and four repeat participants). All subjects underwent postrace echocardiography exactly 1 day after finishing the race. The protocol was designed on the basis of our recent finding that cardiac fatigue is demonstrable for up to 6 days after a race in senior endurance runners. Examinations were performed using commercially available equipment (Vivid 7; GE Vingmed Ultrasound AS, Horten, Norway). Digitized three-cycle loops were analyzed offline at a dedicated workstation (EchoPAC version 6.00; GE Vingmed Ultrasound AS) by a single blinded expert (A.S.). LV mass (LVM) was calculated according to the Penn convention. Left atrial volume was calculated using the prolate ellipse method. LV end-diastolic volume (LVEDV), LV end-systolic volume, and LV ejection fraction (LVEF) were measured using Simpson’s biplane method. Right ventricular (RV) systolic function was assessed by calculating fractional area change (end-diastolic area − end-systolic area). Dimensions were indexed to body surface area.

Doppler Tissue Imaging

Color-coded loops of three beats were recorded onto digital media. Temporal resolution was prioritized by using frame rates > 280 frames/sec to achieve high sensitivity in the analysis of cardiac mechanical events. Tissue velocities were recorded offline as the median of three cardiac cycles. The basal segments of the septum, left lateral wall, and RV free wall were interrogated using a circular sample area (5 × 5 mm). Peak systolic velocity (S _m ), early diastolic relaxation (E _m ), and late diastolic lengthening velocity (A _m ) were recorded. The time from QRS onset to peak systolic velocity (T _s ) was recorded ( Figure 1 ) in six segments at the cardiac base and six at the midventricular level (septal, anteroseptal, anterior, lateral, posterior, and inferior). The standard deviation for T _s across these 12 segments was computed (T _s -SD) as previously described. The maximum difference in T _s between any two segments was also recorded (T _s -Diff). The difference in T _s between basal septum and basal lateral wall (septal-to-lateral [S-L] delay) was computed and defined as positive when septal peak preceded lateral peak. Aortic valve opening was identified by standard pulsed-wave Doppler such that the analysis of T _s was based on tissue velocities during systole and not the isovolumetric phase. The coefficient of variation (standard deviation of differences divided by the mean) was 13% for T _s and 11.9% for T _s -SD (analyzed in 15 echocardiograms randomly selected from the present study population) and 6% to 15% for longitudinal myocardial velocities (analyzed previously in our laboratory). These figures are very similar to those reported using color-coded tissue Doppler to analyze basal segments in a multicenter setting.

Echocardiographic analysis of tissue velocities using color-coded tissue Doppler.

Genetic Analysis of the ACE Gene

Deoxyribonucleic acid was isolated from whole blood using the Qiagen Blood and Cell Culture DNA Midi kit (Qiagen GmbH, Hilden, Germany), and the individuals were genotyped for the I/D polymorphism in the ACE gene by polymerase chain reaction, as reported elsewhere. The polymorphism consists of an insertion of a 287–base pair Alu repeat (EMBL-SVA X62855 ; NC_000017.10 ).

Statistical Analyses

Data are presented as mean ± SD or as median (interquartile range) as appropriate. All statistical analyses were performed using SPSS version 15.0 (SPSS, Inc., Chicago, IL). Variables were analyzed for normality using the Kolmogorov-Smirnov test. Correlations were assessed using Pearson’s r or Spearman’s ρ as appropriate. Differences between groups were analyzed using a Mann-Whitney U test or an unpaired t test as appropriate. Categorical variables were analyzed using χ ² tests (e.g., elevated troponin vs experience, ACE genotype vs experience). Analysis of variance or Kruskal-Wallis analysis was applied for the analysis of continuous variables between more than two groups as appropriate (e.g., in the case of ACE genotypes). Variables that correlated with T _s -SD in bivariate analysis were included in a linear regression model. Residuals were examined for normality and homoscedasticity and also used to identify outliers. For analysis of predictors of dyssynchrony, ACE genotype was treated as an ordinal variable on the basis of the number of copies of the D allele. In all statistical analyses, a two-tailed P value < .05 was considered significant. Power analysis was used to establish the sample size required to detect a difference in peak workload for α = 0.05 and β = 0.80.