There is increasing evidence that exercise training is beneficial in patients with congestive heart failure (CHF) (7,27,28). The decrease in muscular strength is an important disabling factor in these patients and leads to a considerable loss in quality of life (2,8,24). Although more recent publications have studied the effects of strength training (23), the majority of investigators have focused on cardiorespiratory endurance training programs, mainly cycle ergometer training (22,27). Recently, we reported a pilot study with strength training combined to endurance training in CHF. We observed that strength training, when progressively adapted, is feasible and safe in selected CHF patients. Moreover, maximal strength and muscular endurance of knee extensors and flexors and V̇O2peak improved significantly after 6 months (9). Therefore, we performed a randomized study to compare endurance training (ET) alone to combined training (CT), i.e., strength training combined with endurance training, in patients with CHF. The goals of this study were to compare the effects of cardiorespiratory endurance training to a similar program with the addition of resistance training.
Twenty clinically stable male CHF patients were enrolled. All patients had been recently hospitalized for heart failure and training was begun 6 wk after discharge. Patients were considered stable when severe rhythm disturbances had been ruled out by repetitive Holter monitoring and when drug therapy was well tolerated for 6 wk. During the study, drug therapy was not changed. Patients were randomly assigned either to bicycle ergometer training (ET) or to another group of 10 who performed a combined training consisting of cardiovascular endurance and strength training (CT). The protocol of the study was approved by the local Ethics Committee and written informed consent was obtained from all patients.
Cardiac function and endurance parameters.
All patients were evaluated before and after the training intervention with: 1) echocardiography (Hewlett-Packard Sonos 1500, 5 MHz transducer, Andover, MA), and the reader was completely unaware of the patient’s treatment group. Left ventricular end-diastolic diameter (LVED), fractional shortening (FS), and mitral insufficiency were measured by echocardiography. 2) Radionuclide ventriculography (RNV) (General Electric Optima camera, Milwaukee, WI), in all the studies left ventricular ejection fraction (LVEF) was determined semi-automatically without observer intervention. 3) Symptom-limited cardiopulmonary exercise test (CPX) bicycle test (ramp protocol 10 W·min−1) under medication. It was performed late in the morning under nonfasting conditions. Heart rate was measured on a 12-lead ECG. Blood pressure was taken by cuff. V̇O2 and V̇CO2 were measured breath by breath. Data were collected over 30 s and averaged every 10 s (MedGraphics CPX/D, St. Paul, MN).
Blood lactate was determined enzymatically using capillary blood from a hyperaemic earlobe, first at rest, then at 3-min intervals during stress, and at peak V̇O2 (Accusport Boehringer-Mannheim lactate test strips, Mannheim, Germany). Isokinetic testing was performed with an isokinetic dynamometer (Cybex 6000, Ronkonkoma, NY). The principle of the isokinetic dynamometer is to adapt its resistance to the muscle torque produced to keep the angular velocity constant, thus providing optimal resistance at every point of joint range of motion. It is a widely used, precise and reproducible method for assessing muscle strength. The isokinetic tests were performed by an experienced physical therapist, who was not involved in the rehabilitation program, and supervised by a cardiologist.
Knee extensors and flexors were evaluated at an angular speed of 180°·s−1. An endurance test was performed, and we analyzed the best repetition and the work output over 20 repetitions. The test was performed on the dominant side and in case of an injured knee on the healthy side.
Exercise Training Protocol
The training sessions took place in the cardiac rehabilitation department and were supervised by two physiotherapists and a cardiologist. Heart rates were checked with a frequency-meter (Polar, Kempele, Finland). At each session, heart rates were recorded during the different types of exercise. Blood pressure was measured and patients were checked for arrhythmias on a cardiac monitor.
The ET group performed a cardiovascular endurance training on a cycle ergometer. Interval training was used as training modality, with intervals of 2 min at 50% of V̇O2peak, followed by intervals of 2 min at 75% of V̇O2peak, respectively (23,25), for a total duration of 40 min.
The CT group performed a combined training of 20 min of endurance training on a cycle ergometer and 20 min of strength training, for a total training duration of 40 min as well. In the combined training group, endurance training was always performed first to guarantee cardiovascular and muscular activation before strength training. The training on the cycle ergometer was also performed with intervals of 2 min at 50% of V̇O2peak and 2 min at 75% of V̇O2peak. Before strength training began, 1RM (repetition maximal) was measured for each muscle group. Strength training was then performed in three series of 10 repetitions of 60% of 1 RM for each group trained. One repetition lasted 6 s (3-s concentric phase, 3-s eccentric phase), meaning that one series of 10 repetitions lasted 1 min. Between each series, the rest phase was 1 min. After 20 training sessions, 1RM was reevaluated and workload was adapted if needed to ensure training was still performed at 60% of 1RM.
The large muscle groups trained with weight machines were quadriceps (leg extension), hamstrings (leg curl), pectoris (seated arm press), latissimus dorsi (pull down), rhomboidus (rowing), and deltoidus (lateral arm abduction using dumbbells).
The ANCOVA method was used to test the impact of combined strength/endurance training versus endurance training alone. For each parameter of interest, a linear regression model was constructed, with the parameter value after 3 months as dependant variables. Independent variables (predictors) were the parameter value at baseline (continuous variable) and the randomization group (coded as a dummy variable). Statistical significance of randomization group was tested through examining changes in the regression sum of square when the variable “randomization group” was removed from the model. The P-value was derived from tables of F distribution.
To adjust for the imbalance in age at randomization, we introduced the variable age (as continuous variable) in an ANCOVA model having the ejection fraction at 3 months as dependant variable, and the baseline ejection fraction and the randomization group as independent variables. All statistical tests were two-sided.
Patients in the CT were slightly younger (P = 0.30) (see Table 1). Most patients were taking ACE-inhibitors, diuretics, and either beta-blockers or amiodarone. One patient had an internal defibrillator. During the training sessions, we observed no major problems, except one episode of severe hypotension in one patient of the ET group. This patient had to be hospitalized for 2 d, and after 1 wk he could resume training. All the patients accomplished 40 sessions during a period of 4 months.
Both groups did not differ significantly concerning baseline cardiopulmonary rest and stress parameters, except for V̇O2peak·kg−1 and peak lactate, which were significantly higher in the ET group (see Table 2). The baseline isokinetically measured strength parameters showed no significant difference.
After 3 months, LVEF assessed by RNV increased by 18% in CT patients, and it decreased by 11% in the ET patients. This result was corroborated by a significant increase of FS in the CT group and a significant decrease of FS in the ET group. The LV diameter mildly decreased in the CT group by 3% and increased in the ET group by 4%. Mitral insufficiency, which was determined semiquantitatively, was found in all patients. The degree of mitral insufficiency was moderate in four patients, two in each group, but it was only mild in the other patients. Peak V̇O2·kg−1 increased in the CT group by 8% and remained unchanged in the ET group. A significant 25% increase in peak lactate was observed in the CT group, whereas it slightly decreased in the ET group. In both groups, working capacity increased by 10%. Strength parameters, peak torque, and muscular endurance increased in both groups.
Because of the imbalance in age at baseline, we adjusted the results on the ejection fraction and the left ventricle total dimension using ANCOVA procedures. The age-adjusted P-values for the influence of the randomization group on LVEF and LVED were 0.0106 and 0.0024, respectively (Fig. 1).
By diminishing the risk of pulmonary congestion and serious arrhythmia’s, continuous improvement of medical treatment of CHF allows more intensive training interventions (4,18). Based on the results of a pilot study (9), we did a randomized study comparing the effects of endurance training (ET) versus combined endurance and strength training (CT).
The main result of our randomized comparative study is that CT improved LVEF, FS, and LVED. This improvement was found with two independent methods, scintigraphy and echocardiography.
For CHF patients, 3 months of strength training, at a frequency of three sessions per week, is a rather intensive program. Such a strength-training program induces at least three important adaptations. First, an increase in motor unit recruitment and in motor unit firing frequency (16); second, preferential recruitment of fast-twitch muscle fibers; and third, fiber hypertrophy (1). None of these mechanisms has been directly measured in our study, but some findings indicate an increase of muscle mass and a preferential recruitment of fast-twitch fibers. Indeed, the significant increase in peak torque, i.e., of maximal muscular strength, depends on the cross-sectional muscular area and muscular mass (21). The gain of body weight in the CT group with a loss of body weight in the ET group might also reflect an increase in muscle mass. Finally, the significant increase in peak lactate in the CT group (Table 2), with a lower peak RER than in the ET group, is consistent with an increase of muscular mass and a preferential recruitment of fast-twitch fibers, known to produce more lactate than slow-twitch fibers (17) rather than an improvement of anaerobic metabolism.
The improvement of muscular strength appears to be related to the beneficial effects of CT on left ventricular function and peak V̇O2, confirming the key role of skeletal muscle abnormalities in the syndrome of CHF. The “muscle” hypothesis of this syndrome states that alterations in skeletal muscles not only generate symptoms but contribute, via reflex mechanisms, to further neurohormonal activation and progression of the syndrome (8). It is important to note that strength parameters, peak torque, and muscular endurance also increased in the ET group. This finding confirms that this training modality also improves peripheral skeletal muscles in CHF. Nevertheless, it appears that CT constitutes a stronger training stimulus for the altered musculature than ET. An increase in muscular strength may contribute to attenuate neurohormonal activation and progression of CHF.
Recently, muscular mass has been described as an important predictor of peak V̇O2 and a major determinant of exercise capacity in noncachectic CHF patients (6). In the CT group, both V̇O2peak and exercise capacity increased. Strength training proved to be useful in the physical treatment of type 2 diabetes mellitus by improving the altered metabolic function also by increasing muscular mass (11). For healthy elderly people improvement of walking endurance by strength training has been described (3).
Muscle fiber hypertrophy requires 6–7 wk of regular strength training (16,26), and our training intervention lasted 12–16 wk (40 sessions). The positive results on both cardiovascular and strength parameters confirm that major training effects in CHF patients already occur after 3 months, and this finding has been reported by other authors (2,5,16,28).
The ET group improved functional class and working capacity. Similar results, i.e., improvement of NYHA score, working capacity, anaerobic threshold, and exercise time, have been described in a number of trials using endurance programs (27). Peak V̇O2 remained unchanged in the ET group. Two recent trials, with a comparable training intensity and duration, showed similar results (19,29). As in our study, mean age of the patients was relatively high 62 yr (29) and 67 yr (19). It should be noted that in our study patients performed an interval training method (23,25) and that each training session lasted 40 min in order to be comparable with the total training time of the CT group. In contrast, a German study showed a 31% increase in peak V̇O2 after ET. In this study, mean age of the patients was 50 yr and training modality was ET alone (40 min 5–6 times per week), with adaptation of the target training heart frequency and a total duration of 6 months (13). Therefore, the fact that that peak V̇O2 did not change might be explained by the relative short total duration (3 months), the frequency (only 3 times per week), and the relative high mean age of the patients.
In the ET group, LVED increased while LVEF and FS slightly decreased during the training period. Although most investigators used peak V̇O2 to assess the effects of exercise training in CHF patients, there is little information in literature about direct assessment of LV function. In a comparable Italian study, 50 patients (mean age 56) underwent ET 3 times per week, and LVED was assessed after 3 months. This study also found a slight increase of LVED but, after exercise training, in contrast to our results, a mild increase of LVEF, which was determined by echocardiography. All the patients in our study had mitral regurgitation, which was, however, only mild or moderate. It can be speculated that ET is associated with a more pronounced stress-induced mitral regurgitation than CT, thus aggravating ventricular remodeling (12). It would be consecutive to address this important aspect of CHF training therapy in a further study.
Peak lactate did not change in the ET group. In the context of increased working capacity and exercise time, this probably reflects a different training adaptation than in the CT group where peak lactate increased. After prolonged endurance training, the lactate performance curve shifts to the right reflecting an improved aerobic endurance, i.e., a higher anaerobic threshold (21). This shifting to the right indicates that aerobic metabolism can be maintained for a longer period with less lactate (19).
In our study, 80% of the patients were over 55 yr old, and this is representative of the mean age of CHF patients addressed to our physiotherapy department for ambulatory cardiac rehabilitation. Although age dependence of trainability obviously concerns ET (Fig. 1), a regression model showed that the positive association between CT and LVEF was only marginally affected. The results of our study suggest that age 60 is a threshold above which training treatment is less effective to improve LV function. As the incidence of CHF raises substantially after the age of 65, and meanwhile 50% of the potential candidates for cardiac rehabilitation belong to this age group, considerations about training modalities for elderly CHF patients will become very important (2).
Data regarding the effects of strength training in CHF patients are rare compared with the amount of data available for the domain of endurance training. Numerous publications have shown beneficial effects of ET: improvement of peak V̇O2, cardiac output, ventilatory abnormalities, histological, and biochemical characteristics of skeletal muscle and neurohormonal overactivity (8,10,13,14). An improvement of endothelial function of limb arteries has also been found (20). For patients with coronary insufficiency, a positive training effect on the endothelial function of the coronary arteries has recently been described (15). At the present time, the specific mechanisms of CT can only be speculated about. Mechanisms may be similar to ET, but CT may constitute a stronger stimulus.
The small number of patients and the lack of a control group are obvious limitations, especially because the training effects are age dependent and the different age groups are not adequately represented. Moreover, it should be noted that baseline parameters were worse in the CT group, indicating more severe CHF.
In this small randomized study, the positive effects of CT on LV function were significantly better than ET. It appears that for stable CHF patients a greater benefit was derived from combined endurance/strength training than from endurance training alone. The improvement of muscular strength seems to be the major determinant for the observed positive effects on LV function.
The results of this study corroborate the results of our pilot study (9). Larger trials, with possibly concomitant invasive measurements, should confirm these results and try to elucidate the differential effects of CT compared to ET.
Although the organization and the infrastructure of CT are expensive, the feasibility and the promising preliminary results should encourage the community of CHF rehabilitation centers to address this training modality in further studies. Apart from positive effects on LV function and peak V̇O2, a nearly normalized strength has a favorable impact in the daily life of CHF patients and may help to prevent progression of the CHF syndrome.
A future task will concern the development of in hospital programs where stable CHF patients can continue strength training during prolonged periods and maintain the positive training effects over the long term. The key issue will be to demonstrate whether improvement in muscular and cardiac parameters induced by CT do translate into a prolonged survival.
We would like to thank Marianne Kayser for her precious help in organizing this study and Dr. D. R. Wagner for reviewing the manuscript.
The study was supported by the Société Luxembourgeoise de la Recherche sur les Maladies Cardiovasculaires.
1. Abernethy, P. J., J. Jurimae, P. A. Logan, A. W. Taylor, and R. E. Thayer. Acute and chronic response of skeletal muscle to resistance exercise. Sports Med. 17: 22–38, 1994.
2. Ades, P. A. Cardiac rehabilitation
and secondary prevention of coronary heart disease. N. Engl. J. Med. 345: 892–902, 2001.
3. Ades, P. A., D. L. Douglas, T. Ashikaga, J. L. Utton, and K. S. Nair. Weight training improves walking endurance in healthy elderly persons. Ann. Intern. Med. 124: 568–572, 1996.
4. ATMA Amiodarone Trials Meta-Analysis Investigators. Effect of prophylactic amiodarone on mortality after acute myocardial infarction and in congestive heart failure: meta-analysis of individual data from 6500 patients in randomised trials. Lancet 350: 1417–1424, 1997.
5. Bellardinelli, D., G. Georgiou, A. Cianci, and A. Purcaro. Randomized, controlled trial of long-term moderate exercise training in chronic heart failure: effects on functional capacity, quality of life and clinical outcome. Circulation 99: 1173–1182, 1999.
6. Cicoira, M., L. Zanolla, L. Franceschini, et al. Skeletal muscle mass independently predicts peak oxygen consumption and ventilatory response during exercise in noncachectic patients with chronic heart failure. J. Am. Coll. Cardiol. 37: 2080–2085, 2001.
7. Coats, A. J. S. Exercise training for heart failure coming of age. Circulation 99: 1138–1140, 1999.
8. Coats, A. J. S. What causes the symptoms of heart failure? Heart 86: 574–578, 2001.
9. Delagardelle, C., P. Feiereisen, R. Krecké, B. Essamri, and J. Beissel. Objective effects of a 6 months’ endurance and strength training program in outpatients with congestive heart failure. Med. Sci. Sports Exerc. 31: 1102–1110, 1999.
10. Drexler, H., U. Riede, T. Münzel, H. König, E. Funke, and H. Just. Alterations of skeletal muscle in chronic heart failure. Circulation 85: 1751–1759, 1992.
11. Eriksson, J. G. Exercise and the treatment of type 2 diabetes mellitus: an update. Sports Med. 27: 381–391, 1999.
12. Grigioni, F., M. Enriquez-Sarano, K. J. Zehr, K. R. Bailey, and A. J. Tajik. Ischemic mitral regurgitation: long-term outcome and prognostic implications with quantitative Doppler assessment. Circulation 103: 1759–1764, 2001.
13. Hambrecht, R., J. Niebauer, E. Fiehn, B. Kälberer, et al. Physical training in patients with stable chronic heart failure: effects on cardiorespiratory fitness and ultrastructural abnormalities of leg muscles. J. Am. Coll. Cardiol. 25: 1239–1249, 1995.
14. Hambrecht, R., E. Fiehn, C. Weigl, et al. Regular physical exercise corrects endothelial function and improves exercise capacity in patients with chronic heart failure. Circulation 98: 2709–2715, 1998.
15. Hambrecht, R., A. Wolf, S. Gielen, et al. Effect of exercise on coronary function in patients with coronary artery disease. N. Engl. J. Med. 342: 454–460, 2000.
16. Hurley, B. F., and S. M. Roth. Strength training in the elderly. Sports Med. 30: 249–268, 2000.
17. Komi, P. V. (Ed) Strength and Power in Sport.
. (German translation) Köln: Deutscher Ärzte-Verlag (International Olympic Committee 1993), 1993, pp. 62–64.
18. Kühlkamp, V. C. Mewis J. Mermi, R. F. Bosch, and L. Seipel. Suppression of sustained ventricular tachyarrhythmias: a comparison of d,l,-Sotalol with no antiarrhythmic drug treatment. J. Am. Coll. Cardiol.
19. Larsen, A. I., T. Aarsland, M. Kristiansen, A. Haugland, and K. Dickstein. Assessing the effect of exercise training in men with heart failure. Eur. Heart J. 22: 684–692, 2001.
20. Linke, A., N. Schoene, S. Gielen, et al. Endothelial dysfunction in patients with chronic heart failure: systemic effects of lower-limb exercise training. J. Am. Coll. Cardiol. 37: 392–397, 2001.
21. McArdle, W. D., F. I. Katch, and V. L. Katch. Essentials of Exercise Physiology, 2nd Ed.
Baltimore: Lippincott Williams & Wilkins, 2000, pp. 126–127, 412.
22. Mckelvie, R. S., K. T. Koon, N. Mccartney, D. Humen, T. Montague, and S. Yussuf. Effects of exercise training in patients with congestive heart failure: a critical review. J. Am. Coll. Cardiol. 25: 789–796, 1995.
23. Meyer, K. Exercise training in heart failure: recommendations based on current research. Med. Sci. Sports Exerc. 33: 525–531, 2001.
24. Minotti, J. R., I. Christoph, R. Oka, M. W. Weiner, L. Wells, and B. M. Massie. Impaired muscle function in patients with congestive heart failure. J. Clin. Invest. 88: 2077–2082, 1991.
25. Mookerjee, S. The application of interval training for exercise prescription in cardiac rehabilitation
. J. Cardiopulm. Rehabil. 18: 233–235, 1998.
26. Phillips, S. M. Short-term training: when do repeated bouts of resistance exercise become training? Can. J. Appl. Physiol. 25185–193, 2000.
27. Shephard, R. J. Exercise for patients with congestive heart failure. Sports Med. 23: 75–92, 1996.
28. Shephard, R. J., and G. J. Balady. Exercise as cardiovascular therapy. Circulation 99: 963–972, 1999.
29. Wielenga, R. P., I. A. Huisveld, E. Bol, et al. Safety and effects of physical training in chronic heart failure. Eur. Heart J. 20: 872–879, 1999.