Acute and chronic effects on central hemodynamics and arterial stiffness in professional rowers

Arterial stiffness focusing on the aortic vascular bed is directly associated with the incidence of cardiovascular events (Boutouyrie and Vermeersch 2010). Pulse wave velocity (PWV) is accepted as a direct marker of arterial stiffness (Mitchell et al 2010) whereas the augmentation index (AIx) serves as an indirect marker (Laurent et al 2006, Baulmann et al 2010). Increasing arterial stiffness causes reduced buffering of pulsations from the heart and subsequently leads to a higher left ventricular afterload and altered coronary blood flow (Laurent et al 2006). Moreover, in population-based studies, a higher PWV is associated with a higher incidence of cardiovascular events and a higher all-cause and cardiovascular mortality (London and Cohn 2002, Williams 2004). Ageing as well as cardiovascular risk factors such as smoking and hypertension lead to an increase of arterial stiffness parameters and central pressures (McEniery et al 2005, Boutouyrie and Vermeersch 2010).

In general, sport is supposed to prevent cardiovascular diseases. Several studies demonstrated that recreational training, like running or cycling, may have a positive effect on arterial stiffness (Miyachi et al 2004, Edwards and Lang 2005, Sharman et al 2005). Beside recreational training, endurance exercise training showed a lower carotid (local) arterial stiffness as well as a higher cerebral perfusion (Tarumi et al 2013). Furthermore a combination of resistance training and endurance exercise training showed a positive effect on arterial stiffness and hemodynamics in postmenopausal women (Figueroa et al 2011). Contrary to the combination of recreational training and endurance exercise, resistance training alone causes a decrease in the arterial compliance. This was also shown during high intensity resistance training in young women (Cortez-Cooper et al 2005) as well as in a randomized interventional trial for men focusing on resistance training (Miyachi et al 2004). In an additional trial, Cortez-Cooper et al showed that resistance training did not show an effect on central compliance in middle-aged men and women (Cortez-Cooper et al 2008). Furthermore, Yoshizawa et al did not show the effects on PWV and blood pressure in middle-aged women in an interventional trial for resistance training (Cortez-Cooper et al 2008, Yoshizawa et al 2009). Resistance training, as well as data for marathon running on a professional or popular level, has not shown a beneficial or even a negative effect on PWV, and no significant effect on AIx (Vlachopoulos et al 2010). Potentially depending on the intensity of training, central as well as peripheral blood pressures were shown to be significantly higher in marathon runners (Vlachopoulos et al 2010). Rowing serves as a good model for chronic training effects on hemodynamics and arterial stiffness (Cook et al 2006). Professional rowing training includes both endurance (rowing, cycling and running) and strength training (Bell et al 1993, Steinacker 1993).

Only a small number of studies have dealt with acute effects. In contrast, published data focussing on effects of chronic exercise in different sports are discussed controversially (Edwards and Lang 2005, Vlachopoulos et al 2010). Three studies have been published addressing the chronic effects of rowing on arterial stiffness. All of them compared different groups of rowers, i.e. elite rowers via MRI, middle-aged rowers with a regular rowing exercise and old-aged rowers with a control group. They showed no or little positive chronic effects on arterial stiffness as well on peripheral hemodynamics (Cook et al 2006, Petersen et al 2006, Kawano et al 2012). Interestingly, none of these studies examined acute effects. However, despite the effects of chronic exercise on hemodynamics, data about acute effects on aortic pressures and arterial stiffness have not been addressed so far.

The aim of the present study was therefore to compare brachial (peripheral) and aortic hemodynamics as well as arterial stiffness in professional rowing athletes and matched controls at rest (chronic effects) and after an all-out test (acute effects).

Study population

Study design

Hemodynamic and stiffness evaluation

Statistical analysis

Baseline characteristics for both groups are shown in table 1. The groups did not differ significantly in age, height, weight and body mass index. However, both groups differed significantly in fasting glucose blood plasma concentration (rowers 73.2 $\pm$ 15.7 versus 89.3 $\pm$ 16.7 mg dl⁻¹ in controls; p  <  0.05). At baseline, HR was significantly lower in athletes (p  <  0.01, figure 2), whereas brachial DBP (p  <  0.05, figure 3) and aortic PP (p  <  0.05, figure 4) were significantly higher. As a prognostic parameter, AIx was significantly higher in athletes (9.1 $\pm$ 5.4) than in controls (7.0 $\pm$ 10.2; p  <  0.01, figure 6). There were no differences for aortic and brachial SBP and brachial PP (figures 5(a)–(c)). PWV was not significantly different (figure 7). The results of the multivariate analyses of AIx and PWV changes proofed to be blood pressure as well as HR unrelated.

Zoom In Zoom Out Reset image size

Zoom In Zoom Out Reset image size

Zoom In Zoom Out Reset image size

Zoom In Zoom Out Reset image size

Power was significantly higher in athletes (p  <  0.001, figure 1). Raw data and statistical analysis for hemodynamics and arterial stiffness recorded after exercise (17.5 $\pm$ 3.4 min) are shown in table 2. HR increased significantly in the rower's group 18 min after the test (49 $\pm$ 2 bpm versus 91 $\pm$ 3 bpm, p  <  0.001). Also in the control group HR increased significantly (70 $\pm$ 2 bpm versus 92 $\pm$ 4 bpm, p  <  0.001). For PWV we observed a significant increase during exercise in both groups (rowers: 6.6 $\pm$ 1.2 m s⁻¹ versus 7.8 $\pm$ 1.6 m s⁻¹, p  <  0.001; control group 6.0 $\pm$ 0.4 m s⁻¹ versus 8 $\pm$ 1.4 m s⁻¹, p  =  0.005). Brachial DBP (74 $\pm$ 2 mmHg versus 71 $\pm$ 2 mmHg, p  <  0.05) and aortic AIx (7 $\pm$ 10.2 versus 2 $\pm$ 6; p  <  0.01) decreased significantly in the control group after exercise, whereas AIx showed only a trend to decrease in the rowers'. In contrast to the brachial DBP, the brachial PP increased significantly (49 $\pm$ 7 mmHg versus 56 $\pm$ 9 mmHg, p  <  0.05) in the control group. Other parameters did not show significant changes in the control group as well as in the athletes group during exercise (table 2).

All chronic differences between both groups that were observed at rest diminished after exercise.

Our findings demonstrate that in competitive rowers central hemodynamics are, potentially as a result of chronic training, significantly higher than in age-matched controls. To the best of our knowledge, no data exist for acute changes of arterial stiffness in (young) professional rowers caused by physiological stress during or after an all-out test. Three publications investigating rowers showed a small benefit for the rowers compared to a control group (Cook et al 2006, Petersen et al 2006, Kawano et al 2012). These findings showed a discrepancy to the recently published review by Ashor et al showing that resistance and combined training had no beneficial effect on arterial stiffness and central hemodynamics (Ashor et al 2014). Endurance exercise on a regular basis has recently been shown to increase aortic compliance, whereas regular resistance training decreases it (Miyachi et al 2004). Therefore, competitive rowers, who undergo both, endurance and strength training, serve as an interesting model to evaluate their effects on hemodynamics and arterial stiffness. Previous studies reported that regular rowing in middle-aged and older subjects might be associated with favourable effects on aortic compliance (Cook et al 2006, Kawano et al 2012). A study investigated young elite rowers versus a control group measuring the proximal and distal PWV. Petersen and coworkers used MRI as the investigational method whereas we used an oscillometric arteriograph device (Petersen et al 2006) in our study. Due to its more widely accepted and validated approach, the arteriograph might be much more specific for PWV analysis also because it does not measure only a local PWV (Petersen et al 2006, Baulmann et al 2008). The authors concluded that simultaneously performed endurance training may compensate the stiffening effects of strength training. No data, however, are currently available for young, competitive rowers and also no data exist for the effects of an acute rowing exercise test. In summary, our data show that (1) as expected, rowers are much better adapted to physical strain. (2) Rowing at a competitive level leads to a lower resting blood glucose and heart rate at rest indicating differences in metabolism and significantly higher cardiac output. (3) Even at this young age, the chronic effects aortic pulse pressure and the AIx of the competitive rowers were significantly higher at rest than in the control group. This is not fully unexpected since a regulated cardiac output at a lower heart rate would be associated with a higher stroke volume, and so a higher pulse pressure, unless the aorta is much more distensible. Additionally, compensatory mechanisms, that have been postulated in young humans (Yu and Blumenthal 1963, O'Rourke 1976, McEniery et al 2005, Seals et al 2008), seem not to be present in our study population. (4) Interestingly, the above-mentioned differences between rowers and controls diminished after the acute all-out test (table 2). PWV of rowers and controls increased during the all-out test, although the difference in the rowers' group was not as high as in controls, which might be, at least in part, explained by a better adaption to physical strain.

Thus, our findings raise the question whether and how chronic competitive rowing can lead to favourable long-term effects on aortic compliance, when at a younger age arterial stiffness is significantly deteriorated. One might speculate that the quality and/or quantity of training might influence the hemodynamic effects. Rowers perform a higher percentage of resistance training than other sportsmen, especially runners (Leveritt et al 1999, Fiskerstrand and Seiler 2004). Thus, in the rowers group, strength training components may have overweight endurance training components, potentially leading to unfavourable chronic effects on arterial stiffness (Miyachi et al 2004). One might assume a reversal point at which the positive influence of endurance training turns to become a negative one (Vlachopoulos et al 2010). One might additionally speculate that the low heart rate, generally assumed as protective against cardiovascular diseases (Greenland et al 1999, Reil and Bohm 2007), might significantly influence arterial stiffness positively during lifetime. A long-term follow up study of these competitive subjects should be established.

Our data support the hypothesis that professional/competitive sports might have negative impact on central and peripheral vasculature. Thus, our data may encourage studies that compare different sports with respect to changes of vascular hemodynamics in acute and long-term settings. We hypothesise that the amount of specific training components might decide whether sport is beneficial or not. Our data suggest that chronic strength training might have unfavourable effects and should as a specific training component be reduced in favour of the athletes' health. More specifically, further studies should address the question at which point of time of different training strategies training itself becomes harmfully on central hemodynamics and aortic stiffness. In addition, these studies should focus on parameters of endothelial dysfunction that might function as mediators of harmful effects.

Limitations of our trial: 1. The all-out test was different in the study groups. Nevertheless, both groups were stressed up to their individual physical limit. 2. An additional limitation is that the number of participants in our study was relatively small. 3. Specific biochemical parameters, such as lipids, have not been measured in our study although they may have, at least a small, impact on stiffness parameters even at this young age. 4. Important acute physiological changes during the all-out test including beside others peripheral vasodilatation and endothelial dysfunction have not been determined. 5. Additional markers for a maximum physical stress during the acute tests such as lactate threshold and maximum oxygen intake were not determined. 6. Potential effects of sex on our results can not be distinguished from our data due to the limited study size. 7. Finally chronic effects of the individual nutrition habits e.g. carbohydrate intake which affects stiffness were not assessed.

Practical implications:

• Professional rowers exhibit chronic deteriorated aortic pulse pressure and arterial stiffness compared to controls.
• Cardiovascular prognosis and outcome might therefore worsen in professional rowers.
• Long-term follow-ups and investigations in terms of training effects on chronic arterial stiffness parameters are urgently needed to clarify our findings.
• Acute training effects on arterial stiffness in professional rowers need to be further investigated.

For this project there has been no financial assistance through any grants. The 'Deutscher Ruderverband' (German rowing association) supported this study by providing the option to measure the athletes during a training camp. We have no conflicts of interest to declare.