Hydrotherapy, i.e. exercise in warm water, as a rehabilitation program has been considered potentially dangerous in patients with chronic heart failure (CHF) due to the increased venous return caused by the hydrostatic pressure. However, hydrotherapy has advantages compared to conventional training. We studied the applicability of an exercise programme in a temperature-controlled swimming pool, with specific reference to exercise capacity, muscle function, quality of life and safety.
Methods and results:
Twenty-five patients with CHF (NYHA II–III, age 72.1±6.1) were randomised into either 8 weeks of hydrotherapy (n=15), or into a control group (n=10). The training program was well tolerated with no adverse events. Patients in the hydrotherapy group improved their maximal exercise capacity (+6.5 vs.−5.9 W, P=0.001), isometric endurance in knee extension (+4 vs.−9 s, P=0.01) together with an improvement in the performance of heel-lift (+4 vs. −3 n.o., P=<0.01), shoulder abduction (+12 vs. −8 s, P=0.01) and shoulder flexion (+6 vs. +4, P=0.01) in comparison to patients in the control group.
Physical training in warm water was well tolerated and seems to improve exercise capacity as well as muscle function in small muscle groups in patients with CHF. This new approach broadens the variety of training regimes for older patients with CHF.
A major problem for patients with chronic heart failure (CHF) is their limited capacity to perform daily activities. Conventional physical training, such as cycle training or peripheral muscle training, has been found to improve physical function and well being in these patients . The prevalence of CHF increases with age and the majority of patients 65 years or older . However, most previous studies have investigated the effect of exercise in patients who are 65 years old or younger . Further, elderly patients often have other disabling diseases that diminish their ability to exercise. For example, it may be difficult or impossible for a patient with severe orthopaedic problems to cycle on an ergometer.
Hydrotherapy, exercise in warm water, is an alternative method of exercising since the buoyancy effect reduces loading. Exercises to improve mobility, strength, as well as cardiovascular fitness can easily be provided in water . Immersion in warm water has been used in bathing resorts in Europe in the beginning of the last century to reduce heart failure symptoms and enhance function and well being in patients with CHF. However, it has recently been advocated that patients with CHF should refrain from exercise in water due to the increased pre-load [1,5]. Immersion in water to the sternal notch causes an increased venous return resulting in both cardiovascular and renal effects [6,7]. In healthy people the thoracic blood volume, central venous pressure, cardiac output, and diureses increases during water immersion . Vagal tone measured by power spectrum analysis increases . Neurohormonal effects of water immersion, such as a decrease in renal sympathetic activity, a reduction in angiotensin II and aldosterone levels, and an increase in renin activity are also seen in patients with CHF as well as healthy people [7,9]. Water temperature, water depth and the posture of the body influence the physiological reactions to immersion . Tei et al.  assessed the effect of warm water immersion (41 °C swimming pool) and found an improvement in hemodynamics in patients CHF. Water immersion resulted in increased pre-load, ejection fraction, stroke volume, and cardiac output, and decreased peripheral resistance.
Patients with CHF are suffering from reduced quality of life due to several factors; these include a decreased physical function, dyspnea and fatigue . Previous studies have shown conflicting results, with improved exercise capacity not always being associated with improved quality of life .
The purpose of this study was to assess the applicability of an exercise programme in a temperate controlled swimming pool in older patients with CHF. The hypothesis was that training in warm water could be a feasible alternative of physical training for patients with CHF, resulting in enhanced physical performance, muscle function and quality of life.
Twenty-five patients (8 women) with stable CHF in NYHA functional class II–III, ejection fraction <45%, and 60 years of age or older were recruited. Medication for heart failure had to be stable for the previous 3 months. Exclusion criteria were diabetes, peripheral arterial disease, chronic pulmonary disease and status post stroke, or other disabling diseases that might interfere with the exercise protocol. After baseline testing, the patients were randomised, using a 2:1 ratio, in a stratified order to 8 weeks of hydrotherapy (n=15), or to a control period (n=10). The patients were stratified according to age over 70 years, NYHA-class III and female gender. Baseline characteristics of the study population are given in Table 1. The study complied with the declaration of Helsinki. The Ethics Committee of Göteborg University approved the research protocol and informed consent of the subjects was obtained.
|Training (n=15)||Control (n=10)||P-value|
|Duration of CHF (years)||5.3±2.6||6.0±5.2||ns|
|Etiology of CHF||12/2/1/0|
|NYHA class (II/III)||3/12||1/9|
|Beta blockers (n)||13||7|
2.2. Test procedures
2.2.1. Exercise tolerance
Ergometer exercise tests were conducted in an upright position. A ramp protocol was utilised with a 10-W increase every minute until exhaustion. Peak oxygen uptake (VO2 peak) and peak carbon dioxide production (VCO2 peak) was measured breath-by-breath using a V-max system (Sensor Medics, USA). Inspiratory flows and expiratory oxygen (O2) and carbon dioxide (CO2) concentrations were determined. All patients were familiarised with the test procedure through a submaximal exercise test (13 on RPE-scale) 1 week prior to the first test session.
2.2.2. Six-minute walking test
A standardised 6-min walking test  was used to assess exercise capacity related to activities of daily living. The patients were asked to walk as far as possible during 6 min on a pre-marked 30-m walkway. Heart rate was recorded using a Sport-tester (Polar Electro Oy, Kempele, Finland). Perceived exertion and rate of dyspnea was rated using the Borg scale . The test was carried out twice separated by 1 h of rest at the baseline test procedure. The value of the second test was used in the study.
2.2.3. Muscle strength and endurance
For measurement of the torque, a KINetic COMmunicator II (Kin-Com) (Chattanooga Group Inc., P.O. Box 489, Hixson, TN), was used. This equipment is a hydraulically driven and microcomputer-controlled device that operates in an isokinetic mode, i.e. an equal speed and a perfect adjusted resistance through the whole movement. The test was preceded by a 5-min warm-up on a test bicycle. The subjects sat with a hip angle of 90°, and the tested leg was attached to the lever arm of the dynamometer. The left and the right leg were randomly tested first. Knee extension and isometric strengths were measured at a 60°-knee angle. Isokinetic concentric and eccentric strength were measured at 60 and 180°/s for knee extensors for both legs. Isometric endurance was measured on the right leg as the time the person could keep 40% of his voluntary maximum strength at 60°-knee angle, with a follow-up of 5 min to evaluate the recovery process with a maximum voluntary isometric contraction every minute. On the left leg, isokinetic endurance was evaluated as the reduction of torque (in percent) between the first and the last three extensions in a series of 50 maximal contractions with an angle of 180°/s. Handgrip strength was recorded using Grippit® (AB Detector, Göteborg, Sweden). The maximum grip force and the mean value of the 10-s sustained grip was assessed. The test procedures have been described previously . Unilateral isotonic heel-lift , bilateral isometric shoulder abduction and unilateral isotonic shoulder flexion were also measured as described in Appendix A.
2.2.4. Assessment of quality of life
2.2.5. Training programme
The training programme comprised of 45-min sessions in a heated pool (33–34 °C), three times a week over an 8-week period.
The patients trained as a group following a low to moderate exercise level, i.e. 40–70% of maximal heart rate reserve. The basis posture was standing with water just below neck level. The exercise regime was designed to include muscles utilised in activities of daily living such as walking, dressing and household activities. The programme focused on peripheral muscle training but central circulatory exercises were also included. The purpose was to improve aerobic capacity, peripheral muscle strength and endurance. The physiotherapist used music to facilitate the correct pace of exercise. A heart rate recorder, Sport-tester (Polar Electro Oy, Kempele), was used to monitor the intensity. Details of the exercise programme are given in Appendix B. The control group was instructed to live their life as normal for 8 weeks and were not allowed to increase their habitual physical activity during this period.
The SPSS® 9.0 for Windows (Chicago, IL) was used to analyse the data.
Ratio and interval data are given as mean (±1 S.D. or 95% CI) and ordinal data as median and range. Wilcoxon's rank sum test was used for comparisons of paired observations within each study-group. The Mann–Whitney U-test was used to assess differences between groups. A sign rank test was used to compare ordinal data. A P-value ⩽0.01 was considered significant as a method to correct for multiple comparisons. An intension to treat design was used on all data, except for the compliance rate in the exercise group.
The training program was well tolerated, and there were no serious adverse events. Three patients reported mild fatigue after exercising during the first 2 weeks of the training period. One patient who had a history of paroxysmal artrial fibrillation had to finish the exercise programme due to a new episode of arrhythmia. The average adherence (total number of attended sessions) was 95%. Heart rate reserve during training is presented in Fig. 1.
3.1. Exercise tolerance
Patients in the hydrotherapy group showed a greater improvement in their maximal exercise capacity (+6.5 vs. −5.9 W, P=0.001), maximal oxygen uptake (+1.0 vs. −2.0 ml/kg/min, P=0.02) and 6-min walk test (+29.7 vs. +6.3 m, P=0.055) compared to the control group (Fig. 2 and Table 2).
|Before||After||P-value within the group||P-value vs. the control group|
|Exercise capacity (W)||T||84±23||91±24||0.01||<0.001|
|Oxygen uptake (ml/kg/min)||T||14.3±2.7||15.3±3.2||ns||0.02|
|Six-min walk (m)||T||421±115||450±94||0.02||0.055|
3.2. Muscle function
Patients in the hydrotherapy group significantly improved their ability to perform heel-lift, shoulder flexion and shoulder abduction (Fig. 3). There was also a significant improvement in isometric endurance in knee extension. There was no significant difference between the two groups in isokinetic peak torque at 60°/s and in isokinetic endurance in the right leg. There were no other significant differences regarding isokinetic and isometric strength in knee extension or in handgrip strength (Table 3).
|Knee extension in Kin-Com||Before||After||P-value within the group||P-value vs. the control|
|(60°/s N m) right leg||C||105±31||98±36||0.07|
|(60°/s N m) left leg||C||128±51||114±35||ns|
|(180°/s N m) right leg||C||76±25||73±26||ns|
|(180°/s N m) left leg||C||72±24||79±25||ns|
|Endurance decline in %||T||53±7||50±8||0.05||ns|
|of normal value left leg||C||50±10||46±9||0.1|
|60° (N) right leg||C||119±41||117±41||ns|
|60° (N) left leg||C||128±31||140±47||ns|
|Peak force (N)||T||315±96||328±92||0.1||ns|
|Peak force (N)||T||288±88||298±74||ns||ns|
|Clinical endurance tests|
|Shoulder flexion (n.o)||T||29±7||35±10||<0.01||0.01|
|Shoulder abduction (s)||T||78±21||90±23||0.02||0.01|
3.3. Quality of life
Quality of life improved significantly within the training group as measured by the total score and the physical dimension of the LHFQ (Table 4). In SF-36 there was no significant improvement after 8 weeks of training (Fig. 4). There were no statistically significant inter-group differences.
|LHFQ||Before||After||P-value within the group||P-value vs. the control group|
This is the first study to demonstrate that hydrotherapy can improve maximal performance as well as muscle function in small muscle groups in patients with CHF. Furthermore, the training programme was well tolerated and associated with an improvement in quality of life.
The adherence was high. One patient was excluded from the training programme because of an episode of paroxysmal atrial fibrillation. As the patient had a history of paroxysmal atrial fibrillation we do not believe the water training was the cause of the tachycardia. The patients enjoyed the training sessions and wished to continue exercising in water after the study period was over. Since exercise and physical activity improve function , it is of great importance that life long exercise adherence be maintained. Long-term adherence appears to be difficult to maintain without professional support . Therefore, it could be beneficial to offer patients a wider variety and more engaging training options. Hydrotherapy is often a very well appreciated way of exercising for older people .
4.2. Exercise capacity
In our study there was a 6% increase in peak VO2 in the training group and a 16% decrease in peak VO2 in the control group. Several other studies have described an increment of 10–30% in peak VO2 after aerobic exercise on a level of 40–80% of VO2 peak in patients with Ref. . The explanation for this might be that we included elderly patients in our study. A similar decrease in peak VO2 was seen in another study with elderly patients . The improvement in the 6-min walk test was comparable with several previous studies conducted on land in patients with CHF [21,23–25].
4.3. Muscle function
The lack of improvement in knee extensors might be explained by the difficulty to achieve enough resistance in water to improve quadriceps strength and endurance. The quadriceps muscle is a large muscle group and it can be difficult to construct sufficient resistance for such a large muscle group in water. This has also been seen in patients with polio exercising in water . There was a significant improvement in isometric endurance in knee extensors. This result is difficult to explain since very few of the exercises were focusing on isometric exercises. However, an increase in peripheral vasodilatation would increase oxygen supply to the skeletal muscles, which might facilitate this muscular improvement. Our training programme focused both on peripheral muscle training and on central circulatory exercises in order to improve poor muscle function as well as aerobic capacity. In a recent paper McKelvie  advocated the combination of aerobic exercise and peripheral muscle training to achieve optimal training effects in patients with CHF. The decrease in ergometer performance and muscle function tests in our control group may be explained as a consequence of normal disease progression in elderly patients and/or a random error in a small study population.
4.4. Quality of life
In our study we observed a significant improvement in the total score and in the physical dimension of LHFQ. According to the LHFQ, which is a disease specific instrument, our patient's physical function improved. On the contrary, Owen  could not show any improvement after 5 months of exercise in elderly patients. Others have shown an improvement in SF-36 scores after an exercise programme on land . These conflicting results are probably attributable to small study populations.
4.5. Effects of warm water immersion
It might beneficial for patients with CHF  that hot water immersion is also associated with other effects than increased pre-load, i.e. increased stroke volume, cardiac index, ejection fraction, reduced peripheral resistance, increased diuresis, as well as positive effects in neurohormonal secretion. We have preliminary results showing that water immersion at 33–34 °C induces an increase in stroke volume, ejection fraction and cardiac output in elderly patients with CHF  However, data on changes of stroke volume and cardiac output during water immersion in healthy older persons are conflicting. Small or absent increments have been reported [30,31]. One report has compared hot water immersion with sauna studying afterload reduction in patients with CHF. Similar effects as we found were recorded in both the healthy control group and the patients with CHF. However, the investigators used a very hot water temperature of 41 °C . A change in renal and neurohormone levels in both healthy subjects and in patients with CHF has also been reported, as a result of water immersion. The plasma concentration of antidiuretic hormone was decreased and diuresis was increased during water immersion to the sternal notch. Likewise, the concentration of renin, angiotensin II, aldosterone and the renal sympathetic outflow were decreased by water immersion [7,9].
In summary, the physiological reactions caused by warm water immersion, resemble those used in modern pharmacological treatment of CHF, except for the increased pre-load. The reduced peripheral resistance in combination with an increased cardiac function and positive effects on the neurohormonal profile might explain that water training in thermo-neutral water seems to be a safe and effective exercise alternative for patients with CHF.
4.6. Study limitations
Clinical muscle endurance testing has not been evaluated for reliability and validity in patients with CHF, but similar tests have been evaluated in healthy people and patients with rheumatic diseases . Due to the limited size of our study, the results could not be generalised for all patients with CHF. Although we randomised the patients and tried to stratify for age, NYHA-class, gender and baseline function.
Physical training in warm water was well tolerated and seems to be associated with significant improvement in exercise capacity and muscle function in small muscle groups in patients with CHF. Hydrotherapy may be a useful alternative for many elderly patients with CHF and simultaneously other disorders that impede mobility. Furthermore, our elderly patients might be more representative of the general CHF population than many previous studies.
This study was supported by Vårdalstiftelsen VD-98, Askers foundation, The Cardiac research found SU/Sahlgrenska, Renee Eanders foundation, and Hjalmar Svenssons foundation, LSR memorial foundation, and Gustav V memorial foundation. We wish to express our deep appreciation to Ms Marita Hedberg, BSc, for assistance with the muscle function test, Ms Bente Grüner Sweälv and Margareta Scharin Täng, BSc, for assistance with echocardiographic measurements.
Clinical muscle endurance test
Unilateral isotonic heel-lift. Touching the wall for balance with the fingertips and the arms elevated to shoulder height, the subjects performed a maximal heel-lift on a 10° tilted wedge, one lift every other second using a metronome (Taktell Piccolo, West Germany). The contra lateral foot was held slightly above the floor. The number of maximal heel-rises was counted for each leg.
Bilateral isometric shoulder abduction. The patients sat comfortably on a stool with their back touching the wall and with a 1-kg weight in each hand. They were asked to elevate both shoulders in 90° abduction and to keep this position as long as possible.
The time the patients could keep the shoulders in 90° angle of abduction was recorded.
Unilateral isotonic shoulder flexion. The patients sat comfortably on a stool with their back touching the wall holding a weight (2 kg for women and 3 kg for men) in the arm, which should be tested. The pace, 20 contractions per minute were held using a metronome. The patient were asked to elevate one shoulder from 0 to 90° flexion as many times as possible.
Appendix B. Exercise programme
The exercise programme performed in warm water for patients with heart failure. Music pace (beats per minute).
- Walking forwards, backwards and to the side with increasing speed (108 b/min).
- Walking and jogging in combination with arm movements in alternating directions (120 b/min).
Flexibility and endurance training
- Standing: bilateral arm movements in front of the body ‘drawing a lying eight.’ Arms stretched out at surface in front of the body, shoulder extension and flexion (60 b/min).
- Arms stretched out at surface in abduction, small fast abduction movements was performed with the arms bilaterally. Elbows and upper arms held towards thorax. Flexion movements with wrist flexors and lower arms and then the opposite in extension (60 b/min).
- Jumps in place: contra lateral knee towards contra lateral elbow. Body twisting with the arms working under surface in the opposite direction. Jog in place with high knees while swimming breaststrokes with the arms (130 b/min).
Endurance and strength training
- Eight exercises each performed for 2 min (60 b/min).
- Sitting: reciprocal knee flexion and extension using a chair with water level to the sternal notch.
- Standing: unilateral knee extension and flexion with a hip angle of approximately 45°.
- Standing: unilateral hip flexion and extension (fast small movements).
- Standing: reciprocal shoulder flexion and extension with paddles.
- Standing: bilateral shoulder abduction (90°) fast small movements with weights.
- Standing: reciprocal shoulder flexion and extension with floating weights.
- Standing: bilateral elbow extension (upper arms held towards thorax) with floating weights.
- Standing: unilateral heel-lift with water level to waist.
- Jumps in place: contra lateral knee towards contra lateral elbow. Bicycling forwards and backwards with the legs in a supine position (120 b/min).
- Jumps in place: ski jumps with reciprocal arm movements. Walking/Jumping side to side (120 b/min).
- Stretching exercises for leg and arm muscles.
- Supine position with floating devises and soft classical music.