Exercise training – not a class effect: blood pressure more buoyant after swimming than walking

Few interventions to reduce cardiovascular risk are as effective and efficient as regular physical activity. In epidemiological studies, men [1,2] and women [3] who exercised regularly had lower rates of cardiovascular death. In randomized trials, regular aerobic exercise has been demonstrated to lower blood pressure by reducing regional sympathetic outflow, total peripheral resistance and heart rate. In addition exercise has been shown to augment vagal tone, endothelium-mediated vasodilation, insulin sensitivity and mood, and to lower cholesterol [4–9]. International and national scientific societies now consider regular aerobic exercise for at least 30 min per day, most days of the week, to be a key element of blood pressure management and cardiovascular risk reduction for both hypertensive and pre-hypertensive individuals. Swimming is placed on an equal footing with walking, jogging, or biking as a recommended aerobic exercise [10–15]. The implicit message in these guidelines is that aerobic exercise exerts a ‘class effect’ on blood pressure, with the actual formulation less important than its ‘dose’ or duration. But is this truly the case? Are all forms of regular exercise equally effective for the purpose of lowering blood pressure? A meta-analysis of 54 randomized controlled trials by Whelton et al. [16] concluded that all forms of aerobic exercise were equally effective in reducing blood pressure, regardless of type, frequency or intensity, but swimming exercise was not specifically evaluated. Are all forms of exercise equally effective in the young, the middle-aged and the elderly? The American College of Sports Medicine position statement on exercise and hypertension recognizes that the blood pressure-lowering effects of resistance exercise are of less magnitude than the effects of walking, cycling or jogging, but recommends such training as an adjunct to aerobic exercise as a means of maintaining muscle mass with advancing age [13]. Swimming may be an attractive form of exercise for older adults, and in particular for those immersed in water sports when young, as well as those with difficulty in weight bearing because of orthopedic conditions, or for the overweight. However, there are few data concerning the blood pressure-lowering effects of regular swimming. Current recommendations are based primarily on the results of a trial involving 18 men and women with hypertension of whom 12 were assigned to a 10-week swimming training program [17]. Before enrolment, most subjects were able to swim continuously for at least 10 min. Training lowered seated systolic blood pressure by 6 mmHg, on average, and resting heart rate by 10 beats/min, but diastolic blood pressure did not change significantly. In the current issue of the journal, Cox et al. [18] report the results of a randomized but unblinded controlled trial from Western Australia in which 116 healthy sedentary women, aged 50–70 years, were assigned randomly to a supervised 6-month program of either swimming or walking exercise of similar intended intensity. To be considered sedentary, these women reported performing less than 30 min a week of moderate activity over the prior 6 months. These women were non-smokers and modest alcohol consumers. Weight criteria were liberal (body mass index < 34 kg/m2). Although the authors describe their population as normotensive, it should be noted that women with blood pressures of up to 160/100 mmHg were accepted and, of the 116 subjects enrolled, 14 were receiving anti-hypertensive medication at entry. Exercise training was preceded by a 6-week run-in period. Towards the end of this period, blood pressure, fitness level and other physical characteristics were determined. The protocol consisted of three 30-min supervised sessions of exercise each week. All sessions were preceded by warm-up and followed by a cool-down period. Training intensity was estimated by calculating heart rate reserve, adjusting this value for the mode of exercise, and recording heart rate after 15 and 30 min of exercise. Walking was continuous from the outset, whereas swimming involved interval training interrupted by rest periods that were shortened as training progressed. Both indoor and outdoor pools, heated to 26.5°C, were used. Exercise intensity was initially at 50% of the heart rate reserve and increased to 60–70% after 2 months. The effectiveness of training was determined by field fitness tests performed at least 48 h after the last exercise session. Average supine and standing blood pressure were derived from a series of measurements obtained at each visit, using a validated automated device. Baseline blood pressure was calculated as the mean from three separate visits at the close of the run-in period. The investigators’ primary endpoint was a comparison of blood pressure prior to, and 6 months after, exercise training between the two groups. Training effect values were acquired on two distinct visits, with blood pressure on each occasion taken at least 48 h after the last bout of prescribed exercise. If both forms of exercise training were equally effective in lowering blood pressure, a similar hypotensive response would be anticipated. The key finding in this study is that both supine and standing systolic blood pressure, as measured 48 h after exercise were higher, relative to the walking exercise group, in those women who were randomized to the swimming program. A similar trend was noted for diastolic blood pressure. Embedded in the study design was further randomization to a behavioural intervention package intended to promote compliance with the exercise regimen, or usual care. This intervention was not shown to be effective, and is not discussed further. The average difference between the effect of the two training regimens was +4.4/1.4 mmHg for supine blood pressure, and +6.0/1.8 mmHg for standing blood pressure. A modest 3 beat/min fall in heart rate was observed in both groups. Because mean baseline blood pressure was 116/67 mmHg, these increases did not induce hypertension. Nonetheless, this finding may well dampen enthusiasm for the prescription of swimming exercise for older sedentary women. Before considering the potential implications of the present findings for the broader population of older sedentary subjects, both normotensive and hypertensive, we propose to review briefly the strengths and potential limitations of this trial and then consider potential mechanisms that might account for the different effects of these two training programs on blood pressure. This study is unique and important in that it addresses the question of swimming exercise in a randomized clinical trial, with walking as an active comparator. Its several methodological strengths bear emphasis. These include the number of women randomized, the explicit description of the randomization process, the use of an objective rather than a subjective measurement technique for blood pressure (i.e. to minimize bias in the absence of subject or investigator blinding), the acquisition of blood pressure on several dates to establish both baseline and final measurements (to attenuate its variation), supervision of these subjects while training, and monitoring of diet, salt and alcohol intake throughout the study. The investigators report that neither these latter variables, nor body weight, changed over the training period. The drop-out rate was only 14%, and similar in the two groups, as was adherence to the number of prescribed exercise sessions. By measuring blood pressure 48 h after the exercise bout, the investigators established the consequences of training free of any confounding by the acute after-effects of either swimming or walking exercise [19,20]. Finally, the duration of the study (6 months) was sufficient to provide an estimate of the effect of long-term compliance with an exercise program on blood pressure. Because there was no untrained group, the primary comparison was between walking and swimming, and thus the study cannot provide precise information as to whether swimming would have led to higher blood pressure compared to the untrained state. This is quite possible because, contrary to expectation [21], walking exercise did not lower supine blood pressure. Furthermore, Cox et al. [18] cite several publications reporting higher blood pressure in swimmers than in other endurance-trained athletes. A second important limitation is that these investigators did not recruit a homogenous group of subjects. Of the total, 14 were receiving anti-hypertensive therapy, but the actual number of hypertensives may have been greater. It is known that the effects of exercise on blood pressure are greater in magnitude in untreated hypertensive than in normotensive subjects [5,19]. However, it is not stated whether the response to exercise training in each group was in any way a function of baseline blood pressure. If the authors’ findings argue for caution in the prescription of exercise to older sedentary subjects, then it would be particularly important to know if the pressor effect of exercise was more marked in those subjects with treated hypertension than in those with normal blood pressure, who were untreated. Also unknown are the potential long-term interactions between water immersion, swimming exercise and either diuretics, which would affect central blood volume, or other anti-hypertensive therapies with effects on cutaneous and muscle vascular conductance. It was not possible to blind the participants, observers or intervention staff to the mode of exercise, but no bias towards outcome is presumed, and blood pressure was determined objectively throughout the trial. Unfortunately, because seated blood pressure was not reported, direct comparisons between the results of this and studies of modes of exercise are not straightforward. The absence of serial, or of ambulatory, blood pressure recordings in the immediate post-exercise period precludes any comparison of the magnitude or duration of post-exercise hypotension after walking and swimming. What might account for the differential effect of swimming and walking exercise on blood pressure? The authors suggest several mechanisms, including differences in heat exchange, stimulation of cutaneous afferents eliciting sympatho-excitatory responses, and greater oxygen consumption and pressor responses during arm than during leg exercise, but these factors should exert a greater effect on blood pressure during than after swimming. Studies comparing the after-effects of these two modes of exercise on skin sympathetic nerve activity and cutaneous vascular resistance might provide insight into the contribution of such mechanisms to the present observations. Another possibility is that hydrostatic pressures and cutaneous vasoconstriction divert more blood volume into the cardiopulmonary compartment during swimming. This, in turn, should stimulate a rise in blood pressure and cardiac output during swimming, an increase in natriuretic peptides during swimming, and a pressure and hormonal-induced natriuresis and diuresis following swimming. This would then be the analogous to administration of a diuretic. The result would be activation of the renin–angiotensin–aldosterone axis and, in all likelihood, attenuation or abolition of the fall in renal norepinephrine spillover related to the fall in blood pressure that occurs with upright exercise training involving stationary cycling [6]. None of these variables was determined in the present series, but this hypothesis could be tested in future investigations. Consistent with this concept is the documentation by Parker Jones et al. [22] of lower blood volume in post-menopausal female swimmers compared to runners, matched for training, competitive performance, and the use of hormonal replacement therapy. Their lower blood volume was also associated with a lower maximal aerobic capacity. These authors attributed the lower blood volume of swimmers to exercise training in the horizontal, versus the upright position, because they had previously found that endurance training had increased blood volume when this was performed upright, rather than supine. The current popularity of aquacize classes in North America suggests that many women, particularly those naïve to swimming, might actually prefer upright water exercise, which would eliminate any adverse pressor after-effects of facial immersion, difficulty establishing a breathing rhythm, or fear of asphyxia, and avoid the sprinting, or interval training aspect of the SWEAT 2 protocol. In sedentary normotensive post-menopausal women, an acute bout of upright treadmill exercise induces sustained increases in flow-mediated dilation in the forearm, and in calf and total peripheral conductance, with a significant positive correlation between the after-effects of such exercise on flow-mediated dilation and total peripheral conductance [20]. Upright exercise training has been shown to increase forearm nitric oxide production [23] and forearm endothelium-dependent vasodilation in both normotensive and hypertensive subjects [24]. It is not known whether either swimming or upright water exercise has effects on sheer stress or on blood viscosity similar to that of cycling or treadmill exercise, or whether swimming or exercise in water also induces these acute or chronic vasodilatory after-effects. These concepts could be explored in future experiments. Cox et al. [18] report a significantly higher training intensity for the group allocated swimming, but this did not translate into greater walking capacity, and blood pressure was higher, rather than lower, at the end of the training period. Whether there is a U-shaped relationship between exercise intensity and blood pressure remains a subject of debate. Whelton et al. [16] concluded that there is no such interaction. Importantly, Cox et al. [18] inform us that, in their earlier SWEAT trial [21], moderate exercise lowered the blood pressure of their older women but more vigorous exercise did not. Marceau et al. [25] randomized hypertensive patients to two exercise intensity training protocols; only low intensity exercise reduced daytime ambulatory systolic and diastolic blood pressure. The same group proposed that the lack of anti-hypertensive effect of higher intensity exercise training protocols may relate to a differential effect on the cardiopulmonary baroreflex control of vascular resistance [26]. A somewhat less rigorous swimming training program, in a more homogenous and untreated population, might have elicited a different blood pressure response than the present protocol. These women did not have well developed swimming skills. At the outset, 80% were anxious about putting their face in the water, and a substantial subgroup (25%) retained this anxiety to the very end of the study. Perhaps many of these women avoided swimming when younger, for a variety of reasons. In any event, a reasonable concern is that repetitive exposure to the anxiety of swimming may have raised blood pressure over time in some of these women through an adrenal maladaptive response. Indeed, if exercise training lowers blood pressure primarily by reducing sympathetic outflow to the kidney [6] and perhaps other hemodynamically important vascular beds [27] and, in addition, by augmenting baroreflex-mediated vagal drive to the sino-atrial node [28], it is conceivable that longstanding anxiety, stimulating sympathoadrenal or adrenocortical hormone release, could counter such neural adaptations and contribute to the rise in blood pressure observed in this cohort [29,30]. This hypothesis could be tested in future experiments by comparing the effects of swimming and walking training on adrenaline, cortisol, and the baroreflex regulation of heart rate. Post-menopausal women, who, as a population, are becoming increasingly sedentary and overweight, are encouraged to exercise to improve their cardiovascular risk profile. On the basis of their present finding, Cox et al. [18] suggest that any recommendation for sedentary older women to take up swimming should be made with caution, particularly if they are novice swimmers. The authors also propose frequent blood pressure monitoring. As with any unique and groundbreaking study, we are left with many important questions and avenues for future investigation. What advice should we offer women who enjoy swimming regularly? Continue? Abandon? Would such advice change if they were hypertensive? What should we advise men in this age group? In summary, the key message of the SWEAT 2 trial [18] is that when the primary objective is to lower blood pressure, exercise regimens are not interchangeable, at least in older sedentary women. For the post-menopausal woman seeking vascular health, land-based dynamic exercise has clear benefits. The provocative finding of higher blood pressure in women trained to swim is clearly of interest to physicians treating hypertension, to patients wishing to exercise to self-manage their blood pressure, and to the authors of exercise prescriptions for both primary and secondary prevention of cardiovascular events. Moreover, a series of experiments to define the mechanisms responsible for this differential effect of swimming versus walking exercise on blood pressure can be envisaged. In our view, the present finding should not dissuade habitual and efficient swimmers from pursuing this pleasure, but rather should stimulate closer monitoring of blood pressure, and an open discussion between the hypertensive patient and the physician of what we now know, and do not know.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.