Endurance exercise in seniors: Tonic, toxin or neither?

Introduction: Cardiac adaptation to sustained exercise in the athletes is established. However, exercise-associated effect on the cardiac function of the elderly has to be elucidated. The aim of this study was to analyse left (LV) and right ventricular


| INTRODUC TI ON
The beneficial effects of physical activity on cardiac function and prevention of cardiovascular heart diseases are established (Arem et al.,;Fletcher et al., 1996).Regular and intensive exercise training is presumed to have a salutary impact on left ventricular (LV) and right (RV) ventricular morphology due to their response to volume and pressure stimulation by improving the cardiac performance (Chia et al., 2015;Morganroth, Maron, Henry, & Epstein, 1975).
Recent studies, however, have suggested the possibility of harmful effect of long-term high-intensity exercise.Myocardial fibrosis, RV dilation and dysfunction as well as arrhythmias have been detected, raising the question as to whether there is a level of exercise which is detrimental (La Gerche et al., 2015;Lindsay & Dunn, 2007;Siegel et al., 2001).However, these studies involved mostly young or middle-aged participants and the results were not uniform.
Few studies have addressed the effect of long-term regular training on cardiac physiology and structure in the elderly.Studies of this question reported variable results (Fleg et al., 1995;Gates, Tanaka, Graves, & Seals, 2003;Guirado et al., 2012;Tanaka et al., 2000).For example, it is known that as the heart ages, it undergoes changes which lead to increased wall stiffening and decreasing LV compliance (Cheitlin, 2003).It is, however, not clear whether the adaptations of the LV to exercise in an older population are associated with further modifications in diastology.Thus, the impact of sports on cardiac structure and function in the elderly warrants investigation.
The aim of this study was to analyse biventricular structure, systolic and diastolic function in senior participants in World Senior Games.

| Study Population
Participants in this study were recruited on site, at the Huntsman World Senior Games in St Georges, Utah, which is a yearly international senior sports competition for individuals aged 50 years and above.This yearly event involves up to 10,000 participants competing in a wide variety of sports from square dancing to mountain biking and long-distance races.The recruitment for this study spanned The participants were classified as low-intensity (square dance, bowling, softball, golf, hiking or leisure walkers [up to 30 min' walk at least once a week]), moderate-intensity (swimmer, volleyball, boot camp, basketball and tennis) and high-intensity group (triathlon, soccer, road race, cycling, sprint runners and long-distance [10 kilometres and above] runners) based on the level of dynamic or static intensity required to perform that sport (Mitchell, Haskell, Snell, & Van Camp, 2005).The participants were asked to report the number of years they were performing activities at their defined level.
After completing the questionnaires, each qualified participant underwent an examination during which blood pressure, heart rate, weight and echocardiography were obtained.Each participant gave written informed consent.

| Echocardiography
An echocardiogram was performed by experienced investigators on the site of the competition using commercially available echocardiographic equipment (iE33; Philips Medical Systems, Andover, MA).The examination was randomly performed before or at least 12 hr after a sporting event.Echocardiography examination was performed according to the guidelines of the American Society of Echocardiography (Lang et al., 2015).
Standard echocardiographic measurements of the left ventricular size and function were conducted as previously described (McManus et al., 2009).LV mass was calculated using truncated ellipse method.The analysis of diastolic function included the early and late diastolic trans-mitral inflow velocities (E and A), deceleration time (DT), the septal and lateral early diastolic velocities (E´) measured by PW Doppler tissue imaging (DTI) and the ratio of E/E`.
Left atrial volume was calculated by the biplane disc method and then indexed to the body surface area (LAVI).LA appendage and pulmonary veins were excluded in the measurement.
Analysis of right ventricular dimensions and function was conducted according to the current ASE guidelines (Rudski et al., 2010) including RV outflow tract (RVOT), inflow basal diameter (RVIT), tricuspid annular plane systolic excursion (TAPSE) and fractional area change (RVFAC).S´ of RV free wall was measured by DTI as an index of RV systolic function.Two-dimensional RV strain analysis was performed using offline Research-Arena program (TOMTEC Imaging Systems).
All analysis of data was performed by an experienced reader and blinded to the participant information (age and gender), the sport activity and the training level.The analysis of the LV and the analysis of the RV were performed by the same investigator.

| Statistical analysis
All analyses were performed using StataIC 13 (StataCorp LP).All continuous variables are expressed as mean ± SD.The Shapiro-Wilk test was used to assess the distribution of continuous variables.Comparisons between continuous variables were performed using either t tests, Wilcoxon's signed-rank test, one-way ANOVA or Kruskal-Wallis test, as appropriate.Categorical variables were compared using chi-square tests or Fisher's exact tests, as appropriate.Multiple linear regression analysis was used to examine the effects of demographic characteristics, medical history of HTN or diabetes, and the measurements of systolic and diastolic function.
The intraclass correlation coefficients (ICCs) with 95% confidence intervals on the basis of two-way mixed average measures were used to assess the reliability of results in volume and Doppler measurements performed by 2 observers in a sample of 10 randomly selected examinations.

| RE SULTS
A total of 178 subjects were enrolled in the study (Figure 1) of which 86 (48%) were men.Average age was 68 ± 8 years with 83 of 178 participants (47%) in the age range between 70 and 80 years.With respect to the intensity of the exercise regimen, 51 (29%) were in the low-intensity group, 52 (29%) in the moderate-intensity group and 75 (42%) in the high-intensity group.Of 82 in whom information about exercise duration was available, 59 had more than 20 years of consistent exercise regimen three or more days of the week.
Continuous exercise exceeded 10 years in 25 of 30 subjects (83%) in the moderate-intensity level group and 41 of 44 subjects (93%) in the high-intensity level group.Furthermore, 36 (82%) of the highintensity level of exercise group had sustained this level of exercise for more than 20 years.Among the three groups, there were no significant differences in age, prevalence of hypertension, diabetes mellitus, active smoking and hyperlipidaemia.However, the highintensity exercise group had a statistically significant lower BMI and BSA than the two other groups (Table 1).
Left ventricular size was larger in high-intensity level group (Table 2).In comparing parameters of diastolic function, the high-intensity exercise group had lower deceleration time of inflow through the mitral valve (242 ± 54 milliseconds in low-intensity, 221 ± 52 ms in moderate-intensity and 215 ± 58 ms in high-intensity exercise group, p = .03).When the high-intensity group was compared to a combined group of moderate-and low-intensity groups, there was lower mitral valve deceleration time in the high-intensity group but the difference was not statically significant (215/-58 ms versus.231 ± 55 ms; p = .07).The differences among the three groups remained statistically significant after adjusting for age, sex, hypertension and diabetes.There was a consistent trend of increasing values in E/A ratio with increasing intensity of exercise although this was not statistically significant.E´ showed higher values with higher exercise intensity (Table 2).
Left atrial volume index was larger in the high-intensity group compared to low-and moderate-intensity groups (p = .001).It remains significantly larger when adjusting for age, gender, heart rate, HTN and DM (p = .002).
The RV basal internal diameter (mm) was larger in proportion to exercise intensity 35.8 ± 4 in the low-intensity group, 38.6 ± 5 in the moderate-intensity group and 39.2 ± 5 in the high-intensity group (p = .05).RV size was also larger in the high-intensity group when compared with moderate-and low-intensity groups combined (p = .04).The RV systolic function expressed as TAPSE and S´ in the lateral RV wall showed an increasing trend with higher intensity of exercise although the difference was not statistically significant.
No difference was seen in RV fractional area change (FAC) among the three groups.The absolute value of global RV strain (%) was unchanged between groups (−22.5 ± 4.5 in low-intensity, −21.2 ± 3.1 in moderate-intensity and −22.8 ± 3.9 in high-intensity exercise group, p = .15).
The analysis of high-intensity athletes with over 20-year exercise training more than 3 days a week showed significantly higher E prime in the LV septal and lateral wall compared to the other participants.
When comparing the high-intensity group who have exercised regularly for more than 20 years with the rest in the same group, there was a tendency towards higher E´ and lower E/E´ although this finding did not achieve the statistical significance (Table 3).

| D ISCUSS I ON
The present study showed that there were no detrimental effects on LV and RV function in the cardiac function of the elderly at varying level of activity.Furthermore, there were indications that the more intense the effort, the better the function.We found a significant inverse correlation of exercise level with the important diastolic parameter mitral inflow deceleration time.There was also a tendency towards a positive correlation of high-intensity exercise and Doppler tissue E wave.High-level athletes had significantly larger indexed LA volumes when adjusted for age, gender, HTN and DM, albeit these LAVI values were only slightly increased above the upper normal limit (34 ml/m 2 ).
Moreover, the results showed that in the high-intensity athletes, the RV size was larger and had a significantly higher longitudinal strain in the free wall.In the remaining groups, RV size and function showed no significant differences.
With continuous exercise training, the ageing heart undergoes some hemodynamic changes as a part of a physiological adaptation to higher cardiac output.Exercise minimizes the reduction of maximal oxygen consumption that usually occurs with ageing and has an impact on morphological LV changes (Fujimoto et al., 2010;Maufrais et al., 2017).The LV structural adaptation was mainly found in the increase LV dimension and mass, while LV systolic function remained rather unchanged by long-term exercise.This is consistent with our The reported effects of continuous exercise on the diastolic function are, however, not unanimous in the literature (Baggish et al., 1985;Case lli, Di Paolo, Pisicchio, Pandian, & Pelliccia, 2015;Fleg et al., 1995;Popovic et al., 2006).In the normal ageing heart, reports have shown that the diastolic filling undergoes age-related changes (Caballero et al.,;Cheng et al., 2010;Munagala et al., 2003).
The early diastolic filling decreases and the atrial contribution increases to achieve the adequate diastolic LV filling required to maintain adequate stroke volume.The mechanism underlying theses structural changes are not well known.The slowing of relaxation has been attributed to a reduction in the elastic function of the myocyte (Carrick-Ranson et al., 2012;Hirota, 1980).
Other parameters have been proposed to explain the diastolic changes in the elderly.Increase in afterload due to more stiff vascular system and elevated blood pressure with age, a reduced LV compliance reducing LV-LA gradient in early diastole independently of the relaxation and elasticity of the LV chamber have been described (Arbab-Zadeh et al., 2004; Lakatta & Levy, 2003;   Weisfeldt, 1998 ).
The diastolic changes are expressed on Doppler echocardiography as a lower E/A ratio, a lower E´ that has been considered to be a non-invasive measure of LV relaxation and a higher E/E`, while LV compliance was more represented in the deceleration time of the E wave and isovolumetric relaxation time (Arbab-Zadeh et al., 2004;Garcia, 2008;Nagueh, Middleton, Kopelen, Zoghbi, & Quinones, 1997;Thomas et al., 1992).
Our finding is consistent with other authors who showed that E/A ratio was similar in fit and sedentary seniors, while DT was lower.This suggests that the changes in the diastolic filling in the trained heart are related to more compliant LV chambers while the age-related ventricular relaxation properties are unchanged.

TA B L E 2 Echocardiographic parameters of study participants
In the higher intensity exercise level group, there was a tendency to higher E´ and lower E/E´ in those with long-time (>20 years) exercise history, suggesting that relaxation rate is spared in this group.
The increase in LA size has been shown in multiple studies as a common occurrence in trained athletes and reflects physiological adaptation to exercise conditioning (D'Andrea et al., 2010;Iskandar, Mujtaba, & Thompson, 2015).Furthermore, an increased risk of atrial fibrillation has been suggested though the underlying mechanism is still not well known (Guasch, Mont, & Sitges, 2018).In our study, atrial fibrillation was an exclusion criterion.
The effect of exercise training on the right ventricle has gained in interest the last decades (Krumm et al., 2018;La Gerche et al., 2015).
The adaptation of the RV to physical activity has been reported in the most studies as an increase in RV size.Our study shows that there was a positive correlation between the level of training and RV dimension which is consistent with these studies (Bohm et al., 2016).
The dilation of RV was limited to the inflow tract, while the outflow dimension was unchanged.The outflow tract has been reported not to participate in the hemodynamic adaptation of the RV to exercise, and when changed, it is more an expression of structural disease such as in arrhythmogenic right ventricular cardiomyopathy (ARVC) (D'Ascenzi et al., 2016) et al., 2016;Schoor et al., 2016).We did not observe such a tendency in this study population where RV function was preserved in longterm exercise.

| Limitations
The limitations in our study are primarily due to its inherent nature as an observational study.There may be concern for selection bias limiting the generalization of the findings of this study since the participants were not randomized into different exercise groups.
However, there were no significant differences with respect to the patient characteristics between these groups.The distribution of the participants in three groups based on the intensity level required to perform the activity rather than the kind of sport could mask differences in function between distinct sports.However, the small sample size of athletes would have limited the ability to detect such changes.Another limitation is that history of exercise regimen was reported which may lead to potential errors in recall.
Participants with hypertension were included in the study as long as hypertension is controlled which was defined as blood pressure of 140/80 mmHg.Long-standing hypertension has been shown to lead to LV remodelling including diastolic dysfunction.However, the results of this study remained unchanged after adjusting for hypertension, confirming that it had no significant influence on the outcome of the study.Participants underwent the scans before any vigorous activity, at least 12 hr before or after their athletic events.It is unclear if this could have influenced the findings.The echocardiograms were obtained at the physiological baseline as much as possible.
The heart twist has been suggested as another echocardiographic-based metric to further explore the LV diastolic function as there is increase in LV torsion with both age and physical endurance exercise.However, this was not been addressed in our study protocol.

| CON CLUS ION
LV diastolic filling is not only preserved, but may also be enhanced in older athletes, especially those who participate in high-intensity exercise level over long period of time, while systolic left and right ventricular function remain unchanged.These data support the theory that the influence of endurance exercise training in senior hearts is beneficial.

ACK N OWLED G M ENTS
We would like to thank the administration of The Huntsman World Senior Games in Utah for their support during the recruitment of the participants in this study.We also thank Liticia Blazquez Arroyo, three years-2014, 2015 and 2016.Participation was voluntary and self-selected.The participants in the study were informed about the study by electronic information on the event website, distribution of flyers and the presence of a representative from the study group from the University of California, San Francisco (UCSF), at the site of the enrolment.All subjects who agreed to participate in the study were required to answer a questionnaire.Exclusion criteria were a history of cardiopulmonary disease, known coronary artery disease (CAD), atrial fibrillation (AF), more than mild valvular heart disease (VHD) or uncontrolled hypertension (HTA) and age less than 50 years.No one had a history of chemotherapy or radiation therapy.
. The age-related functional changes reported in the normal ageing RV include lower longitudinal displacement expressed by lower TAPSE and RVS' in the elderly.There were no differences in these values between exercise level and duration in our participants.In a group of young athletes, D'Ascenzi et al. showed that the RV dilation occurs during the competition period compared to pre-and postcompetition while the RV function remains mainly unchanged.Other investigators have suggested a negative effect of exercise training on systolic RV function, both close to the exercise performance and in long term, a possible cause being myocyte necrosis generating myocardial fibrosis in the thin RV wall (D'Ascenzi TA B L E 3 Participants with high-intensity level of exercise; diastolic parameters in those who sustained this level of exercise for more and those for less than 20 years Abbreviations: DT, deceleration time; MV A, inflow velocity during late diastolic filling; MV E', mitral annular velocity on tissue Doppler imaging on lateral (lat) and medial (med) wall; MV E, mitral valve inflow velocity during early diastolic filling.