Exercise-induced arterial hypertension - an independent factor for hypertrophy and a ticking clock for cardiac fatigue or atrial fibrillation in athletes?

Background : Exercise-induced arterial hypertension (EIAH) leads to myocardial hypertrophy and is associated with a poor prognosis. EIAH might be related to the “cardiac fatigue” caused by endurance training. The goal of this study was to examine whether there is any relationship between EIAH and left ventricular hypertrophy in Ironman-triathletes. Methods: We used echocardiography and spiroergometry to determine the left ventricular mass (LVM), the aerobic/anaerobic thresholds and the steady-state blood pressure of 51 healthy male triathletes. The main inclusion criterion was the participation in at least one middle or long distance triathlon. Results: When comparing triathletes with LVM <220g and athletes with LVM >220g there was a significant difference between blood pressure values (BP) at the anaerobic threshold (185.2± 21.5 mmHg vs. 198.8 ±22.3 mmHg, p=0.037). The spiroergometric results were: maximum oxygen uptake (relative VO 2max) 57.3 ±7.5ml/min/kg vs. 59.8±9.5ml/min/kg (p=ns). Cut-point analysis for the relationship of BP >170 mmHg at the aerobic threshold and the probability of LVM >220g showed a sensitivity of 95.8%, a specificity of 33.3%, with a positive predictive value of 56.8 %, a good negative predictive value of 90%. The probability of LVM >220g increased with higher BP during exercise (OR: 1.027, 95% CI 1.002-1.052, p= 0.034) or with higher training volume (OR: 1.23, 95% CI 1.04 -1.47, p = 0.019). Echocardiography showed predominantly concentric remodelling, followed by concentric hypertrophy. Conclusion: Significant left ventricular hypertrophy with LVM >220g is associated with higher arterial blood pressure at the aerobic or anaerobic threshold. The endurance athletes with EIAH may require a therapeutic intervention to at least prevent extensive stiffening of the heart muscle and exercise-induced cardiac fatigue.


Introduction
Myocardial hypertrophy in hypertensive patients has a negative influence on long-term prognosis 1 , cardiac arrhythmias and mortality 2,3 . Myocardial hypertrophy in otherwise healthy, non-hypertensive individuals can be caused by exercise-induced arterial hypertension (EIAH) 4-6 and may also result in poor prognosis 7 . Moreover, exercise-induced hypertrophy may cause sudden cardiac death in athletes 8 . However, myocardial hypertrophy induced by extensive exercise presents so called "normal diastolic" function, which might be a result of "physiological" adaptation 9,10 . EIAH or elevated blood pressure values during exercise might have a "negative" impact on cardiac function in athletes and might be one of the important factors causing "exercise-induced" cardiac fatigue (Figure 1). Our hypothesis is provocative, but this suggestion might become important for many professional and leisure athletes ( Figure 2).
Physical activity in the general population is of fundamental importance 11,12 . The role of EIAH in normotensive adult athletes 13 or healthy men is currently under discussion 6,14 . It is unclear how far endurance sport can influence a "negative remodelling" of the athlete's heart 15 . The dosage of exercise bouts which causes cardiac injury 18,19 , and the "true pathologic values" of EIAH are unknown or under debate 6 .
Endurance sport is linked to cardiac injury 20 . In individual cases, long term training might lead to arrhythmias 21 , atrial fibrillation 22,23 or myocardial fibrosis 16,24 and early sudden cardiac death 24-26 , female athletes are less commonly affected 26,27 .
The type of sport discipline has also an influence on the type of hypertrophy. Some authors distinguish the strength-trained heart and an endurance-trained heart 10,28 . Further factors that might influence exercise-induced hypertrophy are genetic factors 29,30 , gender 31 , environmental factors 32 , endocrine factors 33 and arterial hypertension 34 .
In this study, we examined the impact of EIAH on cardiac hypertrophy in 51 normotensive (at rest) healthy Ironman athletes with long daily training times.

Materials and methods
The influence of EIAH on cardiac hypertrophy was examined in 51 male triathletes (mean age 37.

Anthropometry and echocardiography
Anthropometric baseline data of triathletes are listed in Table 1. The blood pressure values of the two groups (LVM <220g and LVM >220g) are visualized in Figure 3 (for exact values see Table 2). In Figure 3 one can see that triathletes with LVM >220g have higher blood pressure values at ventilatory aerobic threshold (VAT) and anaerobic threshold (RCP) and at maximum achieved Watt-level (Wattmax). The results showed myocardial hypertrophy in most participants and were classified as according to Lang et al. 36 . Normal morphology was found in three triathletes, eccentric hypertrophy was shown in one athlete, concentric remodelling was observed in 26 triathletes and concentric hypertrophy in 21. Right ventricular remodelling or other pathological findings of the right ventricle were not found in any of the athletes. Left ventricular function was good in all triathletes (EF >55%). All relevant echocardiographic values are shown in Table 1. All further parameters are shown in Table 2, sorted according to the p-value.
Spiroergometry/physiological performance/blood pressure values Oxygen uptake, ergometer performance and heart rate with VAT, RCP and at peak capacity are shown in Table 3. Participants with LVM >220g achieved at all thresholds and at maximum level higher power output values. Relative oxygen uptake values were slightly higher in the group with LVM >220g but not significant different at the maximum stage of loading. Spiroergometric maximum oxygen uptake (relVO 2 max) was 57.3±7.5ml/min/kg vs. 59.8±9.5ml/min/ kg (p=n.s.) for LVM <220g vs. >220g, respectively. Table 4 shows the cut-point analysis for blood pressure values and the probability of development of LVM >220g. BP values over 180mmHg at the aerobic threshold might define the athletes at risk of developing LVM >220g and a possible further cardiac fatigue.

Left ventricular hypertrophy
According to the values reported by Devereux et al. 40 and Bove et al. 41 , we divided the triathletes in two groups: group 1 (LVM >220g) and group 2 (LVM <220g) to assess the possible reasons for left ventricular hypertrophy. The significant differences between the two groups are shown in Table 1 and The probability of dependent factors for LV-hypertrophy was calculated by odds ratios (Table 5). Odds ratios analysis showed a significant relationship between the arterial pressure values during exercise (significant p-values at the aerobic and anaerobic threshold in Table 5). The significant values are bold in Table 5. A further relationship was found between bike training times and overall training times and LVM >220g ( Figure 4). Values above 1.0 show this significant relationship. absolute values primarily were not defined. Odds ratios analysis was used to calculate the probability of elevated blood pressure and hypertrophy. The estimation of sensitivity/specificity to detect the boundaries of the EIAH and the positive or negative predictive values for the blood pressure boundaries should be performed during the study. Our interest was directed to the aerobic and anaerobic threshold, because this is a constant level of blood pressure maintained during training or competitions. A Vivid 7 model echocardiograph manufactured by general Electric was used for the examinations. The Ergobike 8I manufactured by Daum and the Metalizer 3B produced by Cortex were used for the spiroergometric examination 35 .
The assessment of each triathlete was performed in 2011 and 2012 on the same day with the echocardiography first followed by spiroergometry. The spiroergometry was performed as follows: the stress test (exercise bike) was conducted in stages after successful gas and volume calibration: 50W for 3 minutes, 100W for further 3 minutes and thereafter increased by another 30W for 3 minutes (ramp-test).
The test ended when the subject could no longer maintain the predefined rpm of 90 or if the subject was exhausted.
The echocardiographic analysis was conducted according to general recommendations 36,37 . The formula recommended by the American Society of Echocardiography (ASE) was used for calculate the muscle mass. Enddiastolic LV-volume (EDV) and Endsystolic LVvolume (ESV) were determined monoplane after the modified Simpson method 36 .
The spiroergometric analyses were conducted according to previously published protocols 38,39 : VAT (ventilatory aerobic threshold) was determined as the first non-linear increase of the ventilatory equivalent for oxygen without simultaneous increase of the ventilatory equivalent for CO 2 , and RCP (respiratory compensation point: anaerobic threshold) was determined as simultaneous nonlinear increase of both ventilatory equivalents according to previous recommendations 38,39 .
VO 2 max was registered as the highest average value of oxygen absorption over 30 seconds.

Statistical analysis
The entire statistical analysis plan was designed as follows: Stata/IC 13.1 for Windows was used for data preparation and statistical analysis. The Mann-Whitney-U-Test was used to compare the groups with LVM >220g and LVM <220g. Odds Ratios were calculated to measure the association between blood pressure, training habits and the probability of LVM >220g. Since these exposure variables are quantitative variables, an approximate estimate of the log oddsratio for a one-unit increase in exposure and a 1-degree-of-freedom test for trend were calculated. All statistical tests were two-sided with a signficance level of 0.05.
In addition, sensitivity, specificity, positive and negative predictive values as well as the proportion of correctly classified participants were calculated for each possible cut-point of blood pressure to describe the performance of blood pressure as a "diagnostic test" for LVM >220g.
EIAH on cardiac structures in triathletes is shown in Figure 2. A specific case of cardiac remodelling is shown in Figure 5. In this Figure  are shown normal heart cavities of a triathlete with EIAH in 2011 and massive atrial enlargement in 2014. Exercise-induced hypertension was often discussed in the 1990s 4,5,7 reflecting the results of the Framingham Study 5 . The negative role of EIAH in non-athletic men is relatively clear 6 , but the impact on athletes needs to be discussed and the "pathological range" of EIAH should be evaluated. Exercise-induced hypertension promotes myocardial hypertrophy 4 and increases cardiovascular risks 7 in normotensive men. Athletes with EIAH are in similar way "persons at risk" and may develop a pathological cardiac chamber enlargement and atrial fibrillation, but have less "cardiovascular risk" because of the healthier life style and the positive impact of sport in the development of arteriosclerotic complications.
Cardiac adaptation to exercise, left ventricular hypertrophy and sudden cardiac death The specific endurance training of triathletes leads to physiological changes of performance parameters 45 and also results in changes in cardiac function or heart structures 46 . This adaptation is linked to the nature and magnitude of the physical exercise 31 . The physiological adaptation is a "harmonic increase in size" of a healthy heart caused by physical activity 47 . The term "athlete's heart" 9,31 has been known since 1899 48 . Cardiovascular adaptations to exercise have been systematically defined according to the type of endurance training.

Discussion
The most interesting finding of this study is that myocardial hypertrophy depends on exercise-induced arterial hypertension. This confirms the results described by Douglas et al. 13 and Longas-Tejero et al. 42 , who found a hypertensive response to exercise in eight of 37 healthy athletes (18 soccer players, 12 mountain climbers and seven canoeists). In this cited study, athletes with EIAH showed higher LVM (205g/m 2 ) compared to those without exaggerated blood pressure response to exercise (143g/m 2 ). There is no consensus about the value of systolic blood pressure that constitutes EIAH 6 . According to our study, it seems that a systolic BP value >180mmHg at the aerobic threshold indicates exercise-induced arterial hypertension. So far, the boundaries for EIAH have never been estimated. In this study, we have chosen the aerobic threshold as the measuring point because the majority of the triathlete's training is carried out at this level. When the hypertensive BP value is reached, we should analyse whether a careful low dosage treatment might be beneficial (for example with ACE inhibitors or AT 1 -blockers). The goal of such therapy would be to cut the blood pressure peaks (bouts) during training or competitions and avoid an increase of stiffness of the aorta 43 or LV-hypertrophy in people at risk. Raised BP bouts can lead to pathological enlargement of atrial dimensions in athletes ( Figure 2 and Figure 5). Enlargement of the left atrium may lead to atrial fibrillation and higher activity of electric circuits. There are no clear statements or guidelines regarding the role of EIAH in the daily practice of sports medicine 44 . This manuscript may encourage a discussion about this important issue. The possible impact of       Table 5. Values above 1.0 show a significant relationship between bike training times and overall training times and LVM >220g.
Concentric hypertrophy in triathletes has already been described 49,50 . Douglas et al. 50 suggested that athletes develop hypertrophy possibly due to the systolic blood pressure increase under exercise, which could be explained by the frequency of the training. Diastolic function was shown to be normal under those conditions.
In the present study, odds ratio analysis showed a significant relationship of myocardial thickening to exercise-induced blood pressure. It can be assumed that training over an extended period with exercise-induced blood pressure elevation can lead to hypertrophy in a similar way to that found in pathological conditions with arterial The difference between triathletes and racing cyclists is that the thriathlete's training does not only take place under strength/endurance conditions, but also under running conditions. Modified strength training also results in different changes in the cardiac structures of triathletes in comparison to professional racing cyclists 51 . In a study including 14 professional cycle racers it was shown that the left ventricular muscle mass resulted in eccentric hypertrophy compared to concentric hypertrophy as also shown in our study. Therefore, the functional changes found in the cardiac structures for triathletes resemble the changes in runners 52 . Sudden cardiac death of athletes is more common in male athletes 27, 53 . The different causes of sudden cardiac death are silent coronary disease 54 , hypertrophic cardiomyopathy 55 and arrhythmogenic right ventricular cardiomyopathy 56 (Table 6). Maron et al. 55 described hypertrophic cardiomyopathy as common cause of sudden cardiac death (36%) in young athletes and 8% were presented with indeterminate LV-hypertrophy (possible HCM). The probability of the negative role of hypertrophy in athletes 10 and the problem of qualitative and quantitative relevance are under discussion 57,58 .
Left ventricular "fatigue" Some papers have reported that excessive endurance training may cause negative remodelling of cardiac structures 15,59 . Predominantly marathons and Ironman-distance triathlons can cause a transient   overload of the right ventricle 59 . Fibrosis of the left ventricle in older runners was described as a possible cause of death 24,25 . Numerous investigations regarding the increase in bio-markers (mainly Troponin cTnI and NTproBnP) in runners of marathons 60 as well as triathlon 61 competitions have been conducted. A significant increase in bio-markers after the race was found in all those studies. Uniformly, this was considered as a proof of possible injuries to the heart muscle 62 . Overall, the increase in bio-markers in athletes with intensive muscle work should not necessarily be interpreted as heart specific 63 , because it also depends on the athlete's weight 64 and may be associated with the myolysis (creatine kinase up to 10000 U/l after long-term running) 65 . The discussion on this issue is ongoing 17,20,66 .

Limitations and future directions
The cross-sectional design of this study does not allow a causality regarding the negative role of EIAH in athletes to be established. Although our data suggest that left ventricular hypertrophy might be related to EIAH beyond the normal exercise-induced adaptation, confirmatory longitudinal work is necessary.
The results of this study support the authors' subjective impression of daily practice and engagement in sports medicine over 15 years. We observe rhythm disorders in many cyclists and triathletes around the age of 50, and many of them have elevated blood pressure values during exercise. The probability of increasing stiffness of the aorta as an aging process supported by EIAH remains to be discussed. The present study attempts to analyse the probability of LVM and EIAH and should stimulate further follow-up investigations. It is a very important aim to prevent a potential fibrosis of the left atrium 67 or left ventricular myocardium in athletes 16 in order to avoid "negative cardiac remodelling" induced by exercise and to preserve the positive effects of physical activity 12 . Approximately two million people participate in long-distance races in the United States annually 68 and there are only limited data regarding their exercise-induced blood pressure, which might be one of the main factors triggering cardiac events 69,70 .

Conclusions
The relationship between myocardial hypertrophy and arterial blood pressure during exercise remains an open issue. The literature 13,42 seems to suggest a clear relationship. The relevance of EIAH has to be examined in the future in consideration of serious reports 8,57,58 . The cited authors suggested the isolated (without EIAH) exerciseinduced hypertrophy as a substrate for sudden cardiac death or rhythm disorders. EIAH may enhance the "physiological" exerciseinduced hypertrophy in a pathological way. Accordingly, the blood pressure values or EIAH should be thoroughly examined during routine or pre-event check-up.
The long training-times for Ironman-distances of triathletes with EIAH can lead to additional enlargement of the heart cavities ( Figure 2) and may trigger possible sudden cardiac death during triathlon competitions 71 .
There is strong evidence that athletes have higher incidence of atrial fibrillation and bradyarrhythmias increasing with age 21-23 . We don't know the definitive reasons for this, but EIAH and LVM might be one of the factors. Cases of early death in individual cases due by myocardial fibrosis are possible 24,25 . However, the general prevalence or incidence of EIAH in athletes is unknown. The problem of EIAH seems to be linked more to competitive athletes with vigorous training and mainly to males. It is known that low-intensity training 72 and aerobic exercise have a positive impact on blood pressure lowering [73][74][75] . The hypertensive or non-hypertensive response to exercise seems to be related to hereditary factors 76 , to aging or to the individual arterial stiffness 43 . It is crucial to define the people at risk and possibly start therapy 77 . In our daily practice we treat the athletes at risk with low-dose ACE-inhibitors or AT 1 -blockers before training or competition. The dosage should be tested using an exercise test. Possible therapies for the prevention of fibrosis or atrial fibrillation have already been discussed 23,78 .
Further international, prospective, longitudinal studies on possible negative cardiac remodelling caused by EIAH and sport should be conducted. These studies could help to avoid the adverse effects of sport in people at risk. The overlap of EIAH and exercise-induced hypertrophy has the potential for increased QT-dispersion 79 and is a ticking clock for cardiac fatigue especially for middle aged men. Independent of all the competitive sporting activities with an enormous importance for hobby-athletes, media and industry, physical activity in general population is of fundamental importance 11,12 .

Consent
All athletes provided written informed consent to voluntary testing of the performance and using the data for the study. Triathletes underwent their annual medical check-up or examination for planning their training, which would have been carried out in clinical routine in any case. A special approval by an ethics committee was not mandatory because of the study independent character of the examinations. The examinations were a part of clinical routine support of the triathletes. Pharmaceutical interventions in the triathletes were not affected by the study.

Grant information
The author(s) declared that no grants were involved in supporting this work. 1.

2.
3. The study examined the potential relationship between exercise-induced arterial hypertension (EIAH) and left ventricular hypertrophy in 51 normotensive healthy male triathletes. For that purpose, they divided the study population into two groups according to left ventricular mass (LVM), choosing as cut-off point of significance left ventricular hypertrophy: 220g and analyzed the systolic blood pressure (BPs) in a cardiopulmonary exercise test on a cycle-ergometer. Triathletes with LVM> 220g showed significantly higher BPs at the anaerobic threshold and a trend to higher BPs values at the aerobic threshold and at maximum exercise level.

Open Peer Review
They concluded that significant left ventricular hypertrophy is associated with higher blood pressure at the anaerobic and aerobic threshold and suggest that endurance athletes with EIAH could benefit from preventive antihypertensive treatment.

Comments:
The assessment of blood pressure response to exercise has been typically limited to systolic values. Classic studies such as the Framingham study and preliminary studies from our group have identified the diastolic pressure response to exercise as the best predictor of late development of hypertension. It would have been interesting to perform a complete analysis of the blood pressure response during exercise, including both systolic and diastolic values.
The authors relate EIAH to the development of atrial fibrillation and in fact, this association is used in the title. The study analysis though, did not demonstrate this association, as clinical or electric data related to atrial fibrillation are not analyzed. Athletes with LV mass >220g showed a significantly higher left atrial diameter and a tendency to a higher left atrial volume but all values are within normal limits. Moreover, the author connects EIAH with aorta stiffness but no differences between groups were documented in LVOT and all values are also within normal limits.
The authors claim that they did not find any particular right ventricular remodeling, but the only considering that more and more evidence suggest that intense exercise training can lead to harmful cardiac remodelling in athletes who are otherwise healthy.
In support of a negative role for EIAH, findings reported in this cross-sectional study indicate that left ventricular hypertrophy is associated with higher arterial blood pressure in the athletes examined. This paper provided interesting information; however, as acknowledged by the authors further research is required to clearly establish the causal relationship between EIAH and left ventricular hypertrophy. Moreover, based on the results presented in this study, it seems precipitate to suggest that endurance athletes with EIAH may require a therapeutic intervention such as low-dose ACE-inhibitors or AT1-blockers before training or competition to prevent exercise-induced cardiac hypertrophy.
The authors report that the problem of EIAH is mostly encountered in men. The authors may want to consider that this apparent sex-specific regulation of EIAH is presumably a "gender"-related differences rather than a "sex"-related difference; considering that it is likely that less women will be inclined to enrol in vigorous training such as Ironman-distance triathlon, suggesting that biological and/or physiological differences are probably not involved in the male prevalence of EIAH.
The title of the paper is somewhat misleading as the data reported in this study is not directly related to atrial fibrillation. Similarly, the study does not address the relationship between EIAH and bradycardia or QT dispersion. Furthermore, given that no data directly addressed the link between EIAH and AF, bradycardia and QT dispersion, the authors should consider revising their text to not over extrapolate their findings.

I have read this submission. I believe that I have an appropriate level of expertise to confirm that it is of an acceptable scientific standard.
No competing interests were disclosed. Competing Interests: