Football training improves cardiovascular health profile in sedentary, premenopausal hypertensive women
Sport and Health Sciences, College of Life and Environmental Sciences, St. Luke's Campus, University of Exeter, Exeter, UK
Department of Food and Nutrition, and Sport Sciences, University of Gothenburg, Gothenburg, Sweden
Faculty of Natural and Health Sciences, University of the Faroe Islands, Torshavn, Faroe Islands
Corresponding author: Magni Mohr, College of Life and Environmental Sciences, St. Luke's Campus, Heavitree Road, Exeter EX1 2LU, UK. Tel: +44 1392724759, Fax: +44 1392 264726, E-mail: email@example.com
The present study examined the effects of short-term recreational football training on blood pressure (BP), fat mass, and fitness in sedentary, 35–50-year-old premenopausal women with mild hypertension. Forty-one untrained, hypertensive women were randomized into a football training group (n = 21; FTG) and a control group (n = 20; CON). FTG performed 45 ± 1 1-h small-sided football training sessions during the 15-week intervention period. BP, body composition (dual-energy x-ray absorptiometry), blood lipid profile, and fitness level were determined pre- and post-intervention. After 15 weeks, systolic and diastolic BP, respectively, were lowered more (P < 0.05) in FTG (−12 ± 3 and −6 ± 2 mmHg) than in CON (−1 ± 1 and 1 ± 2 mmHg). Total body fat mass decreased more (P < 0.05) in FTG than in CON during the 15-week intervention period (−2.3 ± 0.5 kg vs 0.4 ± 0.3 kg). After 15 weeks, both total cholesterol (−0.4 ± 0.1 mmol/L vs 0.1 ± 0.2 mmol/L) and triglyceride (−0.2 ± 0.1 mmol/L vs 0.3 ± 0.2 mmol/L) were lowered more (P < 0.05) in FTG than in CON. Yo-Yo intermittent endurance level 1 test performance increased more (P < 0.05) in FTG than in CON (111 ± 18% vs 1 ± 3%) during the 15-week intervention period. In conclusion, short-term football training resulted in a marked reduction in BP and induced multiple improvements in fitness and cardiovascular health profile of untrained, premenopausal women with mild hypertension.
Arterial hypertension is associated with cardiovascular morbidity and mortality, and it is well known that the risk of arterial hypertension is markedly elevated by an inactive lifestyle and poor dietary habits (Lawes et al., 2002; Manson et al., 2004; Blair, 2009). There is strong evidence that physical training lowers arterial blood pressure (BP), improves aerobic fitness, and counteracts several other cardiovascular risk factors in patients with mild-to-moderate hypertension (Pescatello et al., 2004; Fagard & Cornelissen, 2007), but it is still debated whether the magnitude of response is related to the type of training performed.
Several recent studies suggest that high-intensity intermittent training is highly effective in improving aerobic fitness and other physiological adaptations of importance for cardiovascular health status (Tjønna et al., 2008; Bangsbo et al., 2010; Krustrup et al., 2010a; Nybo et al., 2010; Moholdt et al., 2012). Recreational football can be considered to be an intense and variable type of interval training as average heart rates are above 80–85% of maximal heart rate and 10–50% of the time is spent in the highest aerobic training zone for untrained men (Krustrup et al., 2007, 2009; Coutts et al., 2009; Randers et al., 2010a) and women (Krustrup et al., 2010b; Randers et al., 2010b), even for individuals with no prior experience of football. Several studies have shown that short-term recreational football does indeed lower BP for normotensive men (Krustrup et al., 2009) and facilitates a wide range of physiological adaptations with importance for cardiovascular health profile, such as maximal oxygen uptake, heart function, peripheral arterial function, body fat mass, and low-density lipoprotein (LDL)-cholesterol (Krustrup et al., 2010a, b; Randers et al., 2012). In addition, it has been shown that 3 and 6 months of recreational football training results in a substantial reduction in mean arterial pressure (MAP) and arterial stiffness in middle-aged men with mild-to-moderate hypertension (Andersen et al., 2010a, b; Knoepfli-Lenzin et al., 2010; Krustrup et al., 2013; Schmidt et al., 2013). Specifically, it has been demonstrated that 2 × 60-min small-sided football sessions per week over 6 months resulted in a 13 mmHg reduction in systolic blood pressure (SBP) and an 8 mmHg reduction in diastolic blood pressure (DBP) for 30- to 55-year-old men with mild-to-moderate hypertension, in combination with improvements in VO2max and endothelial function as well as a lowering of fat mass, resting heart rate, and heart rate during fixed submaximal exercise. These results suggest that recreational football can indeed be used as an effective broad-spectrum training intervention for young and middle-aged men. However, it is yet to be investigated whether football can be used as part of the non-pharmacological treatment of women with mild-to-moderate hypertension.
Gender differences have been shown to be present within a range of physiological adaptations to exercise training (Pedersen & Saltin, 2006; Fagard & Cornelissen, 2007). For example, women appear to display smaller reductions in BP after exercise training interventions in comparison with their male counterparts. This observation has also been confirmed after a 3- to 4-month period with recreational football training in normotensive individuals, where the women revealed positive but less pronounced effects on BP and fat mass reductions compared with the age-matched men (Krustrup et al., 2009, 2010b; Randers et al., 2010b). It is therefore of interest to investigate the effect of football training on cardiovascular disease risk profile in sedentary women with mild-to-moderate hypertension.
Thus, the objective of the present study was to examine the adaptive response to 15 weeks of recreational football training on BP and other physiological variables associated with cardiovascular health in untrained, premenopausal, hypertensive women.
Materials and methods
Forty-one sedentary, premenopausal women with mild hypertension were recruited as subjects in the study. The subjects were selected from 262 volunteers based on training history, medication, BP and body mass index. A total of 85 participants were recruited, and 44 were randomly assigned to two different swimming groups as part of another study, whereas 41 took part in the present study. The 41 participants were randomized into a football training group (FTG; n = 21; Table 1) and a control group (CON; n = 20; Table 1a). The sample size was calculated from expected changes in the primary outcomes (BP and fat mass) based on previous studies on football for hypertensive and normotensive men and normotensive women (Krustrup et al., 2009, 2010b, c, 2013; Andersen et al., 2010a, b). The study was approved by the ethical committee of the Faroe Islands as well as the Sport and Health Sciences Research Ethics Committee at the University of Exeter, Exeter, UK, and conducted in accordance with the Declaration of Helsinki. After being informed verbally and in writing of the experimental procedures and associated risks, all participants gave their written consent to take part in the study.
Table 1a. Baseline anthropometrical measures and exercise performance for untrained premenopausal hypertensive women in the football training group (FTG) and the inactive control group (CON)
Yo-Yo IE1 (m)
Submax HR (%)
Data are means ± SE.
LBM, lean body mass; Submax HR, heart rate after 200 m of Yo-Yo IE1 test; TFM, total fat mass; Yo-Yo IE1, Yo-Yo intermittent endurance level 1 test performance.
45 ± 3
165 ± 1
79.8 ± 2.8
36.6 ± 2.0
42.7 ± 1.4
420 ± 45
93.0 ± 1.0
43 ± 3
166 ± 1
77.3 ± 2.3
33.0 ± 1.6
43.7 ± 1.2
458 ± 43
92.3 ± 1.0
Table 1b. Baseline blood, resting heart rate, and blood lipids for untrained premenopausal hypertensive women in the football training group (FTG) and the inactive control group (CON)
Data are means ± SE.
SBP, systolic blood pressure; DBP, diastolic blood pressure; MAP, mean arterial pressure; RHR, resting heart rate; TPC, total plasma cholesterol, plasma high-density lipoprotein (HDL)- and low-density lipoprotein (LDL)-cholesterol, as well as plasma triglyceride concentration.
139 ± 2
86 ± 2
104 ± 2
73 ± 2
5.8 ± 0.1
1.4 ± 0.1
3.6 ± 0.2
1.3 ± 0.1
134 ± 4
82 ± 3
99 ± 2
77 ± 2
5.3 ± 0.2
1.4 ± 0.1
3.5 ± 0.2
1.0 ± 0.1
The study was designed as a randomized controlled trial. After initial testing of 262 subjects, 41 participants with mild hypertension were enrolled in the study. The selection criteria were a sedentary lifestyle for the last 2 years (not taking part in regular training or physical activity), mild-to-moderate hypertension (MAP > 100 mmHg), and body mass index > 25 kg/m2. Participants using beta-blockers were excluded due to the heart rate-lowering properties of these agents, whereas participant using other antihypertensive drugs were not excluded from the study. Of the 41 participants that took part in the intervention, a total of 3 participants from FTG and 2 from CON were using antihypertensive drugs such as thiazides and angiotensin receptor blockers. The 41 subjects were randomized into two groups, i.e., a football training group (FTG) and an inactive control group (CON). FTG took part in recreational football training (1-h sessions 3 times per week for 15 weeks, with a targeted attendance of 3 times per week), whereas CON had no training or lifestyle changes during the same period. One additional subject in the FTG suffered an Achilles tendon rupture during the first week of training and was omitted from the group. Beside this incident, no major injuries occurred. Some subjects had to skip certain training sessions due to muscle soreness, but no one was absent for more than three consecutive sessions. All subjects performed an intermittent endurance test with heart rate recordings and had their BP, body fat content, lean body mass, and blood lipid profile measured before and after the intervention. The primary endpoints were arterial BP and whole body fat mass, whereas the secondary endpoints were lean body mass plasma cholesterol and triglycerides, resting and submaximal heart rates, as well as endurance performance.
FTG completed a total of 45 ± 1 (range: 40–52) training sessions over the 15 weeks intervention period, corresponding to 3.0 ± 0.1 (range: 2.7–3.5) sessions per week. Every session lasted 1 h and consisted of small-sided football games (from 4v4 to 10v10), as previously described (Krustrup et al., 2010a; Randers et al., 2010a). A total of four 1-h sessions were organized per week with a targeted attendance of 3 times per week. A trained football coach was present during all training sessions in order to control the duration of the training and to ensure competitive games. Heart rate was measured during a training session in the 1st and 15th week of training.
The participants reported at the hospital at 08:00 h after an overnight fast and were resting in a supine position for a 2-h period. SBP and DBP were measured according to standard procedures (see Krustrup et al., 2013) using an automatic BP monitor (HEM-709; OMRON, Schaumburg, Illinois, USA) once every 30 min over the 2-h resting period. The average of the four measurements was used as the test result. MAP was calculated as 1/3 SBP + 2/3 DBP. Resting HR was measured during the same time intervals as the BP recordings.
Resting blood sampling and scaling
A resting blood sample was collected under standardized conditions from an antecubital vein between 07:00 and 08:00 h after an overnight fast using the venipuncture technique. The blood was rapidly centrifuged for 30 s and analyzed by an automatic analyzer (Cobas Fara, Roche, France) using enzymatic kits (Roche Diagnostics, Mannheim, Germany) for determination of total cholesterol, LDL-cholesterol, high-density lipoprotein (HDL)-cholesterol, and triglyceride levels. After blood sampling, whole-body mass was determined using a platform scale (OHAUS1-10, Pinebrook, NJ, US).
Dual-energy x-ray absorptiometry (DXA) scanning
Whole-body fat mass and lean body mass were evaluated by total body DXA scanning (Norland XR-800, Norland Corporation). The body was segmented in accordance with standard procedures to evaluate regional fat distribution and all analyses were performed using Illuminatus DXA software (Norland Corporation, Torshavn, Faroe Island). The effective radiation dose was < 0.2 mSv per scan.
Exercise performance testing
The participants completed the Yo-Yo intermittent endurance test, level 1 (Yo-Yo IE1) before and after the training period. Yo-Yo IE1 consists of 2 × 20-m shuttle runs interspersed with a 5-s recovery period consisting of 2 × 2.5-m jogs (see Bangsbo & Mohr, 2012). There is a gradual speed progression during the test, which is controlled by a CD player (Bradley et al., 2011). The participants run until the point of exhaustion, defined as the second time they are unable to complete the 2 × 20-m runs at the required pace (Bangsbo & Mohr, 2012). The pre-intervention test was performed within 10 days of the first training session and the post-intervention test 4 days after the last training session. The tests were conducted indoors on a wooden surface at environmental temperatures of 18–20 °C. The tests were preceded by a short warm-up consisting of the first three of the 2 × 20-m shuttle runs, followed by a 2-min recovery period before the real test. Heart rate was measured continuously during the tests using a Polar Vantage NV chest belt monitor weighing ∼100 g (Polar Electro Oy, Kempele, Finland), and HRmax was determined as previously described (Bradley et al., 2011). The pre- and post-tests were conducted at the same time of the day. All participants were familiarized with the test procedure prior to the experiment as per guidelines presented in Krustrup et al. (2013). The participants were instructed to avoid exercise training and intake of alcohol the day prior to testing and nutritional items rich in caffeine on the day of testing. In addition, the participants were also instructed to follow similar nutritional guidelines during the last 24 h before both test periods.
Data are presented as means ± SEM. All data were tested for normality using the Shapiro–Wilk test. Differences in baseline values between FTG and CON and differences in pre- vs post-intervention values within FTG and CON were evaluated by analyses of variance (ANOVA). Between-group differences in the intervention-induced delta values (0 week vs 15 weeks) were tested using one-way ANOVA. The significance level was set to P < 0.05.
Cardiovascular loading during training
Mean and peak HR in FTG were 145 ± 8 and 180 ± 2 bpm during training in the first and last weeks of the intervention, corresponding to 80.5 ± 1.1 and 98.9 ± 1.4%HRmax, respectively. No differences were observed in heart rate loading in the first and last weeks of training. Time spent per session in the heart rate zones 70–79, 80–89, and > 90%HRmax was 16.4 ± 1.4, 24.4 ± 1.8, and 12.7 ± 1.1 min, respectively, corresponding to 27.3 ± 2.3%, 40.7 ± 3.0%, and 21.2 ± 1.8% of the training time.
Prior to the intervention period, SBP and DBP were not different between groups (Table 1b). After 15 weeks, FTG displayed a greater decrease (P < 0.05) than CON in SBP (−12 ± 3 mmHg vs 1 ± 1 mmHg) and DBP (−6 ± 2 mmHg vs 1 ± 2 mmHg) (Fig. 1). MAP decreased more (P < 0.05) in FTG than CON during the 15-week intervention period (−8 ± 2 mmHg vs 0 ± 1 mmHg), with a decrease in absolute values from 104 ± 2 to 96 ± 3 mmHg in FTG. Nineteen of the 21 subjects in FTG had a decline in MAP during the intervention period.
Body fat and lean body mass
No baseline differences were observed in body composition between FTG and CON (Table 1a). After 15 weeks, total body fat percentage had decreased more (P < 0.05) in FTG than in CON (2.1 ± 0.7% vs 0.5 ± 0.4%), with a decrease from 44.1 ± 1.3% to 42.0 ± 1.0% in FTG. Accordingly, total body fat mass decreased more (P < 0.05) in FTG than in CON during the 15-week intervention period (Fig. 2), with a decrease of 2.3 ± 0.5 kg in FTG. Lean body mass increased (P < 0.05) from 42.7 ± 1.4 to 43.9 ± 1.2 kg in FTG during the intervention period, but this change was not significantly different from CON (43.7 ± 1.2 kg vs 44.1 ± 1.1 kg) (Fig. 2). The change in total body weight (−1.4 ± 0.5 kg vs 1.0 ± 1.4 kg) was not significantly different between FTG and CON during the 15-week intervention period (Fig. 2).
Plasma cholesterol and triglycerides
No baseline differences were observed in blood lipid profile between FTG and CON (Table 1b). After 15 weeks, total plasma cholesterol was lowered more (P < 0.05) in FTG than in CON (−0.4 ± 0.1 mmol/L vs 0.1 ± 0.2 mmol/L), whereas the intervention-induced changes in HDL- and LDL-cholesterol were not significantly different between FTG and CON. After 15 weeks, plasma triglyceride was reduced more (P < 0.05) in FTG than CON (−0.2 ± 0.1 mmol/L vs 0.3 ± 0.2 mmol/L).
Yo-Yo IE1 performance pre-intervention was not different between FTG and CON (Table 1a). After 15 weeks, the Yo-Yo IE1 performance was elevated more (P < 0.05) in FTG than in CON (454 ± 81 m vs −4 ± 17 m, corresponding to 111 ± 18% vs 1 ± 3%) (Fig. 3(a)).
Resting and submaximal heart rate
Resting HR was not different between groups pre-intervention (Table 1b). Resting HR decreased (P < 0.05) in FTG from 73 ± 2 to 66 ± 2 bpm during the intervention period, but the intervention-induced change was not significantly different between FTG and CON (−7 bpm vs −3 bpm).
At baseline, relative HR after 2 min (200 m) of the Yo-Yo IE1 was not different between groups. After 15 weeks, the relative HR after 2 min of the Yo-Yo IE1 was lowered more (P < 0.05) in FTG than in CON (−7.2 ± 0.5%HRmax vs 0.9 ± 0.3%HRmax) (Fig. 3(b)).
The present study is the first to show a marked reduction in BP after 15 weeks of small-sided football training in sedentary premenopausal, women with mild hypertension. Moreover, we observed broad spectrum improvements in aerobic fitness and other markers of cardiovascular health profile after the football training, including improved intermittent exercise performance, marked lowering of heart rate during submaximal exercise, and a reduction in fat mass and plasma cholesterol.
A major finding of the present study was that SBP and DBP were reduced by as much as 12 and 6 mmHg after 15 weeks of football training for premenopausal hypertensive women, whereas no effects were observed for the control group. This magnitude of response is twofold higher than has previously been observed after training interventions involving predominantly aerobic training for hypertensive women averaging reductions of 5 and 3 mmHg, respectively (Kelley & Kelley, 1999; Cornelissen & Fagard, 2005). This finding provides further support for the notion that football training is an effective type of training, as indicated by a series of previous studies in men with mild-to-moderate hypertension showing decreases of 12/8, 8/8, and 11/9 mmHg after 3 months of training (Andersen et al., 2010a, b; Knoepfli-Lenzin et al., 2010; Schmidt et al., 2013) and 13/8 mmHg after 6 months of training (Krustrup et al., 2013), with changes being more pronounced than average changes of 7/5 mmHg after short-term interval training (Kelley & Kelley, 1999; Cornelissen & Fagard, 2005). Previous studies using football as a model to reduce BP have suggested that the reductions were related to a combination of mechanisms, including a reduction in sympathetic tone and an increase in parasympathetic activity as evaluated by heart rate variability measurements and resting heart rate measurements (Andersen et al., 2010a, b; Knoepfli-Lenzin et al., 2010), an improved flexibility of the heart and vascular system as evaluated by echocardiography and endothelial function tests (Andersen et al., 2010b; Krustrup et al., 2013; Schmidt et al., 2013) and a reduced resistance due to higher muscular vascularization as evaluated by histochemical analyses of muscle capillarization (Krustrup et al., 2009, 2010a, b, c). The present finding of a reduction in resting heart rate of 7 bpm indicates a reduction in the sympathetic outflow and is of similar magnitude to that of previous observations after 16 weeks of football training in untrained, normotensive women (7 bpm; Krustrup et al., 2010a, b, c) and in normotensive and hypertensive men after 12 weeks of training (5–12 bpm’; Krustrup et al., 2009, 2013; Schmidt et al., 2013). However, it should be emphasized that the observed change in resting heart rate for the football groups of 7 bpm was not significantly different from the 3 bpm drop in the control group. Furthermore, none of the other possible mechanisms behind a change in BP were examined in the present study and the relative contribution of this and other mechanisms for hypertensive women after a period of football training requires further investigation.
Obesity, high cholesterol, and low aerobic fitness are other variables with a marked influence on the risk of cardiovascular morbidity and mortality (Hu et al., 1998; Wiklund et al., 2008). A vast majority of the participants in the present study were obese or heavily obese, with an average fat percentage of 44%, and had poor aerobic fitness as evaluated by a Yo-Yo IE1 intermittent exercise performance as low as ∼450 m on average. Moreover, more than half of the participants had total cholesterol values above 5.5 mM and LDL-cholesterol values above 2.5 mM, with average values of 3.5 mM. After 15 weeks of football training, the fat mass had declined by as much as 2.3 kg, more than previously reported after football training over 16 weeks for premenopausal untrained women (1.4 kg; Krustrup et al., 2010b) and over 12 weeks for 25- to 65-year-old female hospital employers (1.0 kg; Barene et al., 2013) and similar to football studies with untrained men training two to three times 1-h per week for 12 weeks (2.0 kg; Krustrup et al., 2009). As seen in many other football studies, the total body weight only decreased slightly as the present football intervention resulted in an increase in lean body mass of 1.2 kg, which is similar to the positive adaptations previously observed for untrained premenopausal women after 16 weeks of training (1.4 kg; Krustrup et al., 2010b). Moreover, the football training group had a reduction in total cholesterol and LDL/HDL ratio that was similar to or more pronounced than previous reports on 3–4 months of football training for untrained women and men, and the present study also revealed a decrease in plasma triglyceride that has not been seen in previous studies using short-term recreational football training (Krustrup et al., 2009, 2010a, 2013; Andersen et al., 2010a, b; Knoepfli-Lenzin et al., 2010; Schmidt et al., 2013). Actually, as many as 77% of the participants in the football group had normalized plasma triglyceride values below 1.5 mM after the training period, compared to 43% at baseline.
Last but not least, it was observed that 15 weeks of football training resulted in a marked reduction in aerobic loading during submaximal exercise, indicating an increase in cardiovascular fitness. Thus, the heart rate after 2 min of moderate-intensity intermittent exercise in Yo-Yo IE1 was 14 bpm lower after than before training, whereas no change was observed for the control participants. A drop in heart rate of 8–20 bpm during standardized submaximal running or cycling exercise has also been observed in previous studies with short-term football training for women and men (Krustrup et al., 2009, 2010a, 2013). Specifically, heart rate was lowered by 8 bpm during submaximal cycling after 12 weeks of twice-weekly 1-h football training sessions for hypertensive untrained men, in correspondence with an increase in maximal oxygen uptake of 8% (Krustrup et al., 2013), and by 14–18 bpm during submaximal walking and running after 16 weeks of training for untrained normotensive women, in correspondence with an increase in maximal oxygen uptake of 15%. Maximal oxygen uptake was not measured in the present study, but a big improvement of 111% (454 m) was observed in Yo-Yo IE1, in contrast to the control group, which had unaltered performance (CON: −4 m). Considering that intermittent Yo-Yo shuttle-running test performance is largely correlated with maximal oxygen uptake in untrained and habitually active women and men (Krustrup et al., 2003, 2010b; Bradley et al., 2011; Randers et al., 2012), this finding clearly supports the notion of markedly improved cardiovascular fitness.
In conclusion, short-term recreational football training resulted in a marked reduction in BP in untrained, premenopausal women with mild hypertension. Additionally, the football training caused a series of changes in body composition, lipid profile, and cardiovascular fitness, contributing to a large overall reduction in risk factors for cardiovascular disease.
The present study indicates that football training can be used as part of the non-pharmacological treatment of women with mild hypertension, even for those with no previous experience of football. The high training attendance and the surprisingly limited dropouts in the present study also demonstrates that recreational football is a health promoting activity with a great potential. In addition, the improved aerobic fitness may also result in an increase in the everyday life activity as it has become easier to cycle, walk the stairs, and to do shopping and gardening.
We would like to thank the participants for their great efforts and positive attitude. In addition, the technical assistance provided by Susanna Holm, Heðin Joensen, Ivy Hansen, Gunnrið Jóannesarson, Ann Ostero, Ebba Andreassen, Maud av Fløtum, Hildigunn Steinholm, Liljan a Flotum Petersen, Marjun Thomsen, Guðrið Andórsdóttir, David Childs, Sarah Jackman, and Jens Jung Nielsen is greatly appreciated. The study was supported by a grant from the Faroese Research Council, as well as by The Faroese Confederation of Sports and Olympic Committee (Itrottarsamband Foroya), The Faroese Football Association (FSF) and the Danish Sports Confederation (Danmarks Idrætsforbund). In addition, financial support was obtained from Eik Bank.