- Email: [email protected]
Effects on quality of life, symptoms and daily activity 6 months after termination of an exercise training programme in heart failure patients
International Journal of Cardiology 77 (2001) 25–31 www.elsevier.com / locate / ijcard
Effects on quality of life, symptoms and daily activity 6 months after termination of an exercise training programme in heart failure patients Ronnie Willenheimer*, Erik Rydberg, Charles Cline, Kristian Broms, Birgitta Hillberger, ¨ Lena Oberg, Leif Erhardt Department of Cardiology, Malmo¨ University Hospital, Lund University, 5 -20502 Malmo¨ , Sweden Received 3 May 2000; received in revised form 25 August 2000; accepted 4 September 2000
Abstract Background: Exercise training in heart failure patients improves exercise capacity, physical function, and quality-of-life. Prior studies indicate a rapid loss of these effects following termination of the training. We wanted to assess any sustained post-training effects on patients global assessment of change in quality-of-life (PGACQoL) and physical function. Methods: Fifty-four stable heart failure patients were randomised to exercise or control. The 4-month exercise programme consisted of bicycle training at 80% of maximal intensity three times / week, and 49 patients completed the active study period. At 10 months (6 months post training) 37 patients were assessed regarding PGACQoL, habitual physical activity, and dyspnea-fatigue-index. Results: Both post-training patients (n517) and controls (n520) deteriorated PGACQoL during the 6-month extended follow-up, although insignificantly. However, post-training patients improved PGACQoL slightly but significantly from baseline to 10 months (P50.006), differing significantly (P50.023) from controls who were unchanged. Regarding dyspnea-fatigue-index, post-training patients were largely unchanged and controls deteriorated insignificantly, during the extended follow-up as well as from baseline to 10 months. Both groups decreased physical activity insignificantly during the extended follow-up, and from baseline to 10 months post-training patients tended to decrease whereas controls significantly (P50.007) decreased physical activity. Conclusion: There was no important sustained benefit 6 months after termination of an exercise training programme in heart failure patients. A small, probably clinically insignificant sustained improvement in PGACQoL was seen in post-training patients. Controls significantly decreased the habitual physical activity over 10 months and post-training patients showed a similar trend. Exercise training obviously has to be continuing to result in sustained benefit. 2001 Elsevier Science Ireland Ltd. All rights reserved. Keywords: Exercise training; Heart failure; Quality of life; Physical function; Long-term effects
1. Introduction Several studies have shown beneficial effects of physical endurance training in patients with depressed left ventricular function and / or clinical heart failure [1,2]. Benefit has been shown in terms of exercise capacity, oxygen uptake, ability to carry out daily *Corresponding author. Tel.: 146-40-331-000; fax: 146-40-336-209. E-mail address: [email protected] (R. Willenheimer).
activities, quality of life, and left ventricular function [1,2]. The effects have been found to be due mainly to improved peripheral muscular function and improved autonomic balance / reduced sympathetic activation, and are largely independent of central hemodynamics [1,2]. It seems, however, that there is quite rapid deterioration following discontinuation of the exercise training [1,2], indicating that the training must be continued in order to obtain sustained beneficial effects. It is not known if patients who have participated in a training programme will
0167-5273 / 01 / $ – see front matter 2001 Elsevier Science Ireland Ltd. All rights reserved. PII: S0167-5273( 00 )00383-1
26
R. Willenheimer et al. / International Journal of Cardiology 77 (2001) 25 – 31
spontaneously continue to carry out exercise training or increase their level of physical activity long-term (i.e. 6 months) after discontinuing the formal training programme. Neither is it known if controls will pick up the exercise training or change their physical activity after having participated in such a study. In a study of 49 patients with symptomatic heart failure we showed that a 16-week physical exercise training programme improved patients global assessment of change in quality of life (PGACQoL) and maximum exercise capacity [3]. After completion of the active study period the patients were followed for an additional 6 months, at the end of which they were interviewed regarding PGACQoL, habitual physical activity, and ability to carry out daily activities. Here we report the results of this extended follow-up.
2. Methods
2.1. Study design The study design has been described in part previously [3]. Eligible and willing patients were randomised 1:1 at baseline to complete an exercise training program or to serve as controls during a 4-month active study period. This period was followed by a 6-month extended follow-up at the end of which patients were interviewed. The total study duration was thus 10 months. No intervention was carried out during the extension period. The medication was kept largely unchanged throughout the study duration. Control patients were asked not to change their degree of physical activity during the active study period, but neither training patients nor controls were instructed regarding physical activity during the 6-month extended follow-up. The interviews were done by two investigators. The scores from previous interviews were not available to the investigators at the time of the second and third assessments, and all data were blindly evaluated. The variables used to assess response to the exercise training (effect variables) were prospectively defined as changes from baseline to study end in peak oxygen uptake (peak-VO 2 ), maximum exercise capacity (EC max ), dyspnea-fatigue index (DFI), and PGACQoL. The variables assessed at the 10-month interview were DFI and PGACQoL.
The study was approved by the local Ethics Committee and conforms with the principles outlined in the Declaration of Helsinki.
2.2. Patients Patients were recruited at Malmo University Hospital, with a primary catchment area of 250 000 inhabitants. Inclusion criteria were: 1) 8 points according to the Boston heart failure criteria [4]; 2) left ventricular ejection fraction 0.45 at the most recent radionuclide or echocardiographic examination (not older than one year at inclusion); and 3) 75 years of age. Exclusion criteria were: 1) change of clinical status and / or medication within 4 weeks prior to inclusion; 2) myocardial infarction, heart surgery, or coronary angioplasty within three months prior to inclusion; 3) inability to perform a bicycle test; 4) exercise-terminating angina pectoris, ST-depressions (.2 mm in .1 lead), blood pressure fall (.10 mm Hg), or arrhythmia (e.g. ventricular tachycardia / fibrillation, increasing frequency of ventricular extrasytolies, supraventricular tachycardia .170 beats / min) at the most recent maximal exercise test (including the baseline test); 5) pulmonary disease judged to be the main exercise-limiting factor and / or peak expiratory flow rate ,50% of the age- and sex-adjusted reference value; 6) New York Heart Association class IV; and 7) clinically significant aortic stenosis. Eligible patients were identified from hospital records and invited to participate in the study. Of the invited patients 40% did not wish to participate. Of the 61 patients who accepted seven were found to meet with at least one exclusion criterion. Thus, 54 patients were included and all gave informed consent to participate. Four patients randomised to training were withdrawn before study start due to withdrawn consent (n52), myocardial infarction (n5l) and back pain (n51) between the baseline examinations and study start. Fifty patients (27 controls, 23 training patients) entered the first 4-month active study period. One patient randomised to training developed rheumatoid arthritis during training and, thus, 49 patients completed the active study period. Five patients died during the 6-month extended follow-up, two of whom were controls. Due to administrative reasons 5 of the remaining 42 patients were not
R. Willenheimer et al. / International Journal of Cardiology 77 (2001) 25 – 31
interviewed at 10 months. Thus, 37 patients were assessed at the 10-month interview.
2.3. Exercise testing Each patient had performed at least one exercise test prior to this study. The exercise tests were at both baseline and 4 months performed in the morning, fasting or at least 3 h after a light breakfast, using upright bicycle ergometry, and were always limited by exhaustion. The initial work load of 30 W was increased stepwise by 10 W every minute. Using a MedGraphics Cardiopulmonary Gas Exchange System CPX / MAX model 762014-102, a breath-bybreath analysis of total body oxygen consumption and carbondioxide production was performed. Calibration against gases of known concentrations, as well as adjustments for air humidity, temperature, and air pressure were made. Continuous 12-lead ECG-registration with computerised ST-analysis was performed, as well as blood pressure measurement, registration of effort, chest pain, and dyspnea according to the Borg scale every second minute.
2.4. Dyspnea fatigue index, global quality of life, and physical activity score Symptoms from daily activities were assessed by DFI [5]. DFI addresses how much daily activity the patient can tolerate (part 1), how large a work load the patient can tolerate before symptoms arise (part 2), and how fast the patient can work (part 3). For each part the patient can receive 0 (most severe symptoms) to 4 (no symptoms) points. Thus, patients were assigned between 0 (very pronounced symptoms from daily activities) and 12 points (no symptoms). At four (end of active study period) and 10 months patients were asked to rate PGACQoL in comparison with baseline. No change was assigned the figure 0, deterioration was rated from 21 (mild) to 23 (severe), and improvement from 1 (little) to 3 (much). This quality of life assessment is similar to the commonly used and validated patients global assessment of quality of life [6,7]. Since we were specifically interested in change in global quality of life we applied the PGACQoL The degree of habitual physical activity was assessed in each patient. Average time (min / week)
27
spent on each physical activity was noted, and the intensity by which the activity was performed was classified as either low (1 point), moderate (2 points), or high (3 points). An activity score was calculated for each activity according to the formula: time3 intensity 2 / 100. For each patient the scores for the various activities were then added, resulting in a total activity score. At the 4-month interview (end of the active study period), activity due to the exercise training program was not included in the score.
2.5. Exercise training protocol The exercise training protocol has been described previously [3]. The patients carried out cycle ergometer interval training at a heart rate corresponding to 80% of peak-VO 2 (at the baseline exercise test)65 beats / min, for as long as possible during each interval. In patients with atrial fibrillation, exercise intensity corresponded to exhaustion grade 15 according to the Borg scale (approximately 80% of maximum). The training was performed in group, supervised by a physiotherapist who ensured that the protocol was complied with. The exercise time was gradually increased from 15 min twice a week to 45 min three times a week from week 7, for a total of 16 weeks.
2.6. Statistics The t-test was used to assess differences regarding changes in continuous variables. Within group changes were examined by analysis of variance for repeated measures with the Bonferroni-Dunn test. To assess between-group differences regarding nominal variables a X 2 or Fisher’s exact test were performed. Data are expressed as mean6SD. Two-tailed P-values ,0.05 were considered significant.
3. Results
3.1. Baseline variables The baseline data of the 37 patients who were interviewed at 10 months are shown in Table 1. These 37 patients showed similar baseline variables compared to the 49 patients who completed the 4-
R. Willenheimer et al. / International Journal of Cardiology 77 (2001) 25 – 31
28 Table 1 Baseline data Variable a
Training group (n5l 7)
Control group (n520)
Age (years) Women (%) IHD (%) Heart failure duration (months) LVEF LVEDD (mm / m 2 ) NYHA class Diuretic treatment (%) ACEi treatment (%) Digitalis treatment (%)
6465 29 85 28617 0.3560.12 3164 2.160.7 94 100 53
6468 30 78 24615 0.3860.10 3164 2.460.7 90 100 30
a ACEi, angiotensin-converting enzyme inhibitor; DFI, dyspnea fatigue index; IHD, ischemic heart disease; LVEF, left ventricular ejection fraction; LVEDD, left ventricular end-diastolic diameter adjusted for body surface; NYHA, New York Heart Association.
month active study period. The training (n517) and control (n520) groups were well balanced regarding all baseline variables. All but two patients had an ejection fraction less than 50% at inclusion. Six (16%) patients had chronic atrial fibrillation. The aetiology for heart failure was ischemic (documented myocardial infarction or angiographically verified coronary artery disease) in 29 patients (78%).
3.2. Compliance and safety Compliance with the exercise training (% sessions attended) among the 17 patients in the training group was 70%, ranging from 54.5 to 97.7, with the exception of one patient with only 27.3%. No adverse reactions to the training were reported. None of the training patients experienced any worsening of heart failure during the study, and none of them were hospitalised, whereas three of the control patients were. Five patients died during the extended followup (from 4 to 10 months), three of whom were
originally in the training group. Since these patients were not interviewed at 10 months they are not otherwise accounted for in this report.
3.3. Changes in effect variables during the active study period The results of the 4-month initial study period for the 37 patients interviewed at 10 months are shown in Table 2. They were similar to the results of all the 49 patients who completed the initial study period [3]. With training (n517) there was an improvement in all the four effect variables (peak-VO 2 , EC max , DFI, and PGACQoL) compared to controls (n520), although the between-group differences were not statistically significant (Table 2). Over the 4-month initial study period the within-group changes in peakVO 2 , EC max and DFI were not statistically significant among the training patients (Table 2), whereas PGACQoL improved significantly within-group (Table 2).
Table 2 Effect variables and PA score at baseline, 4 months, and 10 months Variable a
Peak-VO 2 (ml / kg / min) EC max (W) DFI (units) QoL (units) PA (units) a
Training group (n517)
Control group (n520)
Betweengroup P 0–4 m
Betweengroup P 0–10 m
Baseline
4m
10 m
P 0–4 m
P 0–10 m
Baseline
4m
10 m
P 0–4 m
P 0–10 m
16.263.2
17.364.4
–
0.133
–
16.163.3
16.663.6
–
0.232
–
0.454
–
113626 7.861.7 060 70697
120633 8.162.1 1.361.3 60685
– 8.262.2 0.760.9 48641
0.062 0.131 ,0.0001 0.914
– 0.269 0.006 0.240
112630 6.961.8 060 71668
112634 7.061.8 0.461.4 42655
– 6.462.0 0.061.0 32641
0.841 0.852 0.223 0.014
– 0.154 0.825 0.007
0.126 0.608 0.064 0.507
– 0.073 0.023 0.481
PA, physical activity score; m, months; peak-VO 2 , peak oxygen consumption; EC max , maximal exercise capacity; DFI, dyspnea fatigue index; QoL, patients global assessment of change in quality of life.
R. Willenheimer et al. / International Journal of Cardiology 77 (2001) 25 – 31
3.4. Quality of life, physical activity, and dyspneafatigue-index at the 10 -month assessment Both groups deteriorated PGACQoL during the 6-month extended follow-up, and the changes were of similar magnitude in the two groups (Table 2). Within- and between-group differences were not significant. The change in PGACQoL from baseline to 10 months differed slightly but significantly between the groups (Table 2). Training patients showed a small, statistically significant improvement, whereas control patients showed no change. Training and control patients did not differ significantly regarding the physical activity score at baseline, 4 months, or 10 months (Table 2). Both groups continuously decreased the physical activity score, from baseline to 4 months, during the extended follow-up, and from baseline to 10 months. Over the entire 10-month period this decrease was significant in the control group. There were however no significant between-group differences. The training patients were largely unchanged regarding DFI over the 6-month extended follow-up and during the entire 10 months (Table 2), whereas the control patients deteriorated somewhat. Withinand between-group differences were not significant.
4. Discussion We have previously reported that a 4-month exercise training programme improved exercise capacity and quality of life in 22 training patients compared to 27 controls [3]. Our findings were in good agreement with the results of other studies [1,2]. Six months after termination of the initial active study period 37 patients were interviewed regarding PGACQoL, habitual physical activity, and ability to carry out daily activities (DEL). These 37 patients were comparable to all the 49 patients regarding baseline data and the results of the initial 4-month intervention. Thus, although the changes in the effect variables at 4 months did not significantly differ between the 17 training patients and 20 controls (who were interviewed at 10 months), they were similar to those for all the original 22 training patients and 27 controls. Since the original 22 training patients significantly improved exercise capacity and quality of life vs. the
29
original 27 controls at 4 months, the lack of statistical significance between the 17 training patients and 20 controls (who were interviewed at 10 months) most likely was due to insufficient power as a result of the reduced number of patients. Thus, we feel that the patients interviewed at 10 months were representative to the entire study group completing the initial 4month active study period. Among the 17 training patients there was only a small (1.1 ml / kg / mm) and insignificant improvement in peak-V0 2 during the active training period. This was similar to the change found among all the 22 training patients (0.9 ml / kg / mm) [3]. In a prior publication [3], we reported that men with ischemic aetiology showed an increase in peak-VO 2 comparable to that found in other studies [1,2], including mainly this category of patients. The smaller overall increase in peak-VO 2 in our study may be due to the inclusion of other patient categories. Among the training patients the improvement in PGACQoL achieved during the initial 4-month training period was reduced considerably during the 6month extended follow-up. Despite this there was a slight but significant improvement at 10 months compared to baseline. This is however to be expected in patients who at the time of the interview knew that they had been training and who would have expected some sustained benefit from that effort. Therefore, this small sustained benefit in PGACQoL at 10 months is probably of no clinical significance. The training patients showed a trend towards a decreased physical activity score during the training programme. This is not surprising considering that the score did not take into account the activity during the training sessions. Interestingly, training patients did not spontaneously increase their habitual physical activity after participating in the 16-week exercise training programme. On the contrary, they decreased their physical activity, although statistically insignificantly. When the study was planned there was a concern that the control patients might start to exercise on their own during the active study period. At baseline they were therefore specifically asked not to. As shown by the physical activity score they did not increase their physical activity during the active study period. Nor did they increase physical activity during the 6-month extended follow-up, despite the fact that it was allowed. Instead, they decreased their
30
R. Willenheimer et al. / International Journal of Cardiology 77 (2001) 25 – 31
habitual physical activity during both study periods, and the decrease over the entire 10-month period was significant. There was essentially no change in the ability to carry out daily activities among the training patients, during the extended follow-up as well as during the entire 10 months. There was a trend towards a decline in the control group. There were however no significant between- group differences. Some of the prior randomised and controlled studies examining the effects of exercise training in heart failure patients had a cross-over design [8–12]. In these studies patients were specifically instructed to avoid physical training during the resting period. Thus, it has been possible to assess the effects of rest during 3–8 weeks after the end of the training period. Generally, the training effects have been lost during the resting period. Since patients in our study did not increase their habitual physical activity after the training period they are comparable to the patients in the prior studies in that respect. Our results seem to agree well with these prior results. In a recent exercise training study in elderly patients with heart failure [13], the cross-over design was applied in a different manner. In order to examine the effects of discontinued attendance at the exercise sessions patients were not instructed to reduce exercise, other than the formal training, during the 12-week resting period following the training period. Also in that study the effects of the exercise training were lost during the resting period. Our results and the results of that study differ somewhat with respect to quality of life. In the Owen study training had no beneficial effect on quality of life. Consequently, there was no effect to sustain after termination of training. In our study there was a significantly beneficial effect on PGACQoL from the training and this was to some extent sustained after termination of the training programme. In the Owen study quality of life was assessed by the Minnesota Living with Heart Failure Questionnaire, possibly not well suited for quality of life assessments in elderly heart failure patients [13]. No formal presentation of self reported global quality of life was made in that paper. Thus, since the differing results may be due to different methodology the results may not be in disagreement. However, it cannot be precluded that there is an actual difference between the two studies,
which might be explained by the fact that patients in our study were younger and trained more intensely, more often, and for a longer period of time. In the present study we did not assess exercise capacity or oxygen uptake at 10 months, 6 months after the active study period. Therefore, we do not know if there was a sustained effect on these effect variables. Consequently, we cannot compare our results with those of previous studies in that respect, and this is a limitation to our study.
5. Conclusions Four months exercise training in patients with heart failure carried no important sustained beneficial effects 6 months after termination of the training programme. A small sustained improvement in PGACQoL was seen in post-training patients. This was to be expected given that patients obviously were not blinded to the training and was, therefore, probably clinically insignificant. Controls significantly decreased the habitual physical activity over 10 months and post-training patients showed a similar trend. There were no sustained beneficial effects on the ability to carry out daily activities. Our results indicate that exercise training has to be continuing to result in any sustained benefit.
Acknowledgements We thank our co-workers, Evy Hallgren, Ingrid Ohlsson, Marie Holmberg, Katty Reuterskidld, and Anneli Iwarson, without the help of whom this study could not have been completed. This study was supported by a grant from the Swedish Society for Patients with Heart and Lung Diseases.
References [1] Piepoli ME, Flather M, Coats MS. Overview of studies of exercise training in chronic heart failure: the need for a prospective randomized multicentre European trial. Eur Heart J 1998;19:830–41. [2] The European Heart Failure Training Group. Experience from controlled trials of physical training in chronic heart failure. Protocol and patient factors in effectiveness in the improvement in exercise tolerance. Eur Heart J 1998;19:466–75.
R. Willenheimer et al. / International Journal of Cardiology 77 (2001) 25 – 31 [3] Willenheimer R, Erhardt L, Chine C, Rydberg E, Israelsson B. Exercise training in heart failure improves quality of life and exercise capacity. Eur Heart J 1998;19:774–81. ¨ ¨ ¨ K. Validity of clinical [4] Remes J, Miettinen H, Reunanen A, Pyorala diagnosis of heart failure in primary health care. Eur Heart J 1991;12:315–21. [5] Feinstein AR, Fisher MB, Pigeon JG. Changes in dyspnea-fatigue ratings as indicators of quality of life in the treatment of congestive heart failure. Am J Cardiol 1989;64:50–5. [6] Cline CMJ, Willenheimer RB, Erhardt LR, Wikiund I, Israelsson BYA. Health-related quality of life in elderly patients with heart failure. Scand Cardiovasc J 1999;33:278–85. [7] Rector TS, Kubo SH, Cohn IN. Patients self-assessment of their congestive heart failure Part 2: Content, reliability and validity of a new measure, the Minnesota Living with Heart failure Questionnaire. Heart Failure 1987;1:198–209. [8] Coats AJS, Adamopoulos S, Radaelli A et al. Controlled trial of physical training in chronic heart failure. Circulation 1992;85:2119– 31.
31
[9] Meyer TE, Casadei B, Coats AJS, Conway J, Davey PP, Sleight P. Angiotensin-converting enzyme inhibition and physical training in heart failure. J Intern Med 1991;230:407–13. [10] Adamopoulos S, Coats AJS, Brunotte F et al. Physical training improves skeletal muscle metabolism in patients with chronic heart failure. J Am Coll Cardiol 1993;21:1101–6. [11] Piepoli M, Volterrani M, Clark AL, Adamopoulos S, Sleight P, Coats AJS. Contribution of muscle afferents to the hemodynamic, autonomic, and ventilatory responses to exercise in patients with chronic heart failure. Effects of physical training. Circulation 1996;93:940–52. [12] Meyer K, Schwaibold M, Westbrook S et al. Effects of exercise training and activity restriction on 6-minute walking test performance in patients with chronic heart failure. Am Heart J 1997;133:447–53. [13] Owen A, Croucher L. Effect of an exercise programme for elderly patients with heart failure. Eur J Heart Failure 2000;2:65–70.
Effects on quality of life, symptoms and daily activity 6 months after termination of an exercise training programme in heart failure patients Ronnie Willenheimer*, Erik Rydberg, Charles Cline, Kristian Broms, Birgitta Hillberger, ¨ Lena Oberg, Leif Erhardt Department of Cardiology, Malmo¨ University Hospital, Lund University, 5 -20502 Malmo¨ , Sweden Received 3 May 2000; received in revised form 25 August 2000; accepted 4 September 2000
Abstract Background: Exercise training in heart failure patients improves exercise capacity, physical function, and quality-of-life. Prior studies indicate a rapid loss of these effects following termination of the training. We wanted to assess any sustained post-training effects on patients global assessment of change in quality-of-life (PGACQoL) and physical function. Methods: Fifty-four stable heart failure patients were randomised to exercise or control. The 4-month exercise programme consisted of bicycle training at 80% of maximal intensity three times / week, and 49 patients completed the active study period. At 10 months (6 months post training) 37 patients were assessed regarding PGACQoL, habitual physical activity, and dyspnea-fatigue-index. Results: Both post-training patients (n517) and controls (n520) deteriorated PGACQoL during the 6-month extended follow-up, although insignificantly. However, post-training patients improved PGACQoL slightly but significantly from baseline to 10 months (P50.006), differing significantly (P50.023) from controls who were unchanged. Regarding dyspnea-fatigue-index, post-training patients were largely unchanged and controls deteriorated insignificantly, during the extended follow-up as well as from baseline to 10 months. Both groups decreased physical activity insignificantly during the extended follow-up, and from baseline to 10 months post-training patients tended to decrease whereas controls significantly (P50.007) decreased physical activity. Conclusion: There was no important sustained benefit 6 months after termination of an exercise training programme in heart failure patients. A small, probably clinically insignificant sustained improvement in PGACQoL was seen in post-training patients. Controls significantly decreased the habitual physical activity over 10 months and post-training patients showed a similar trend. Exercise training obviously has to be continuing to result in sustained benefit. 2001 Elsevier Science Ireland Ltd. All rights reserved. Keywords: Exercise training; Heart failure; Quality of life; Physical function; Long-term effects
1. Introduction Several studies have shown beneficial effects of physical endurance training in patients with depressed left ventricular function and / or clinical heart failure [1,2]. Benefit has been shown in terms of exercise capacity, oxygen uptake, ability to carry out daily *Corresponding author. Tel.: 146-40-331-000; fax: 146-40-336-209. E-mail address: [email protected] (R. Willenheimer).
activities, quality of life, and left ventricular function [1,2]. The effects have been found to be due mainly to improved peripheral muscular function and improved autonomic balance / reduced sympathetic activation, and are largely independent of central hemodynamics [1,2]. It seems, however, that there is quite rapid deterioration following discontinuation of the exercise training [1,2], indicating that the training must be continued in order to obtain sustained beneficial effects. It is not known if patients who have participated in a training programme will
0167-5273 / 01 / $ – see front matter 2001 Elsevier Science Ireland Ltd. All rights reserved. PII: S0167-5273( 00 )00383-1
26
R. Willenheimer et al. / International Journal of Cardiology 77 (2001) 25 – 31
spontaneously continue to carry out exercise training or increase their level of physical activity long-term (i.e. 6 months) after discontinuing the formal training programme. Neither is it known if controls will pick up the exercise training or change their physical activity after having participated in such a study. In a study of 49 patients with symptomatic heart failure we showed that a 16-week physical exercise training programme improved patients global assessment of change in quality of life (PGACQoL) and maximum exercise capacity [3]. After completion of the active study period the patients were followed for an additional 6 months, at the end of which they were interviewed regarding PGACQoL, habitual physical activity, and ability to carry out daily activities. Here we report the results of this extended follow-up.
2. Methods
2.1. Study design The study design has been described in part previously [3]. Eligible and willing patients were randomised 1:1 at baseline to complete an exercise training program or to serve as controls during a 4-month active study period. This period was followed by a 6-month extended follow-up at the end of which patients were interviewed. The total study duration was thus 10 months. No intervention was carried out during the extension period. The medication was kept largely unchanged throughout the study duration. Control patients were asked not to change their degree of physical activity during the active study period, but neither training patients nor controls were instructed regarding physical activity during the 6-month extended follow-up. The interviews were done by two investigators. The scores from previous interviews were not available to the investigators at the time of the second and third assessments, and all data were blindly evaluated. The variables used to assess response to the exercise training (effect variables) were prospectively defined as changes from baseline to study end in peak oxygen uptake (peak-VO 2 ), maximum exercise capacity (EC max ), dyspnea-fatigue index (DFI), and PGACQoL. The variables assessed at the 10-month interview were DFI and PGACQoL.
The study was approved by the local Ethics Committee and conforms with the principles outlined in the Declaration of Helsinki.
2.2. Patients Patients were recruited at Malmo University Hospital, with a primary catchment area of 250 000 inhabitants. Inclusion criteria were: 1) 8 points according to the Boston heart failure criteria [4]; 2) left ventricular ejection fraction 0.45 at the most recent radionuclide or echocardiographic examination (not older than one year at inclusion); and 3) 75 years of age. Exclusion criteria were: 1) change of clinical status and / or medication within 4 weeks prior to inclusion; 2) myocardial infarction, heart surgery, or coronary angioplasty within three months prior to inclusion; 3) inability to perform a bicycle test; 4) exercise-terminating angina pectoris, ST-depressions (.2 mm in .1 lead), blood pressure fall (.10 mm Hg), or arrhythmia (e.g. ventricular tachycardia / fibrillation, increasing frequency of ventricular extrasytolies, supraventricular tachycardia .170 beats / min) at the most recent maximal exercise test (including the baseline test); 5) pulmonary disease judged to be the main exercise-limiting factor and / or peak expiratory flow rate ,50% of the age- and sex-adjusted reference value; 6) New York Heart Association class IV; and 7) clinically significant aortic stenosis. Eligible patients were identified from hospital records and invited to participate in the study. Of the invited patients 40% did not wish to participate. Of the 61 patients who accepted seven were found to meet with at least one exclusion criterion. Thus, 54 patients were included and all gave informed consent to participate. Four patients randomised to training were withdrawn before study start due to withdrawn consent (n52), myocardial infarction (n5l) and back pain (n51) between the baseline examinations and study start. Fifty patients (27 controls, 23 training patients) entered the first 4-month active study period. One patient randomised to training developed rheumatoid arthritis during training and, thus, 49 patients completed the active study period. Five patients died during the 6-month extended follow-up, two of whom were controls. Due to administrative reasons 5 of the remaining 42 patients were not
R. Willenheimer et al. / International Journal of Cardiology 77 (2001) 25 – 31
interviewed at 10 months. Thus, 37 patients were assessed at the 10-month interview.
2.3. Exercise testing Each patient had performed at least one exercise test prior to this study. The exercise tests were at both baseline and 4 months performed in the morning, fasting or at least 3 h after a light breakfast, using upright bicycle ergometry, and were always limited by exhaustion. The initial work load of 30 W was increased stepwise by 10 W every minute. Using a MedGraphics Cardiopulmonary Gas Exchange System CPX / MAX model 762014-102, a breath-bybreath analysis of total body oxygen consumption and carbondioxide production was performed. Calibration against gases of known concentrations, as well as adjustments for air humidity, temperature, and air pressure were made. Continuous 12-lead ECG-registration with computerised ST-analysis was performed, as well as blood pressure measurement, registration of effort, chest pain, and dyspnea according to the Borg scale every second minute.
2.4. Dyspnea fatigue index, global quality of life, and physical activity score Symptoms from daily activities were assessed by DFI [5]. DFI addresses how much daily activity the patient can tolerate (part 1), how large a work load the patient can tolerate before symptoms arise (part 2), and how fast the patient can work (part 3). For each part the patient can receive 0 (most severe symptoms) to 4 (no symptoms) points. Thus, patients were assigned between 0 (very pronounced symptoms from daily activities) and 12 points (no symptoms). At four (end of active study period) and 10 months patients were asked to rate PGACQoL in comparison with baseline. No change was assigned the figure 0, deterioration was rated from 21 (mild) to 23 (severe), and improvement from 1 (little) to 3 (much). This quality of life assessment is similar to the commonly used and validated patients global assessment of quality of life [6,7]. Since we were specifically interested in change in global quality of life we applied the PGACQoL The degree of habitual physical activity was assessed in each patient. Average time (min / week)
27
spent on each physical activity was noted, and the intensity by which the activity was performed was classified as either low (1 point), moderate (2 points), or high (3 points). An activity score was calculated for each activity according to the formula: time3 intensity 2 / 100. For each patient the scores for the various activities were then added, resulting in a total activity score. At the 4-month interview (end of the active study period), activity due to the exercise training program was not included in the score.
2.5. Exercise training protocol The exercise training protocol has been described previously [3]. The patients carried out cycle ergometer interval training at a heart rate corresponding to 80% of peak-VO 2 (at the baseline exercise test)65 beats / min, for as long as possible during each interval. In patients with atrial fibrillation, exercise intensity corresponded to exhaustion grade 15 according to the Borg scale (approximately 80% of maximum). The training was performed in group, supervised by a physiotherapist who ensured that the protocol was complied with. The exercise time was gradually increased from 15 min twice a week to 45 min three times a week from week 7, for a total of 16 weeks.
2.6. Statistics The t-test was used to assess differences regarding changes in continuous variables. Within group changes were examined by analysis of variance for repeated measures with the Bonferroni-Dunn test. To assess between-group differences regarding nominal variables a X 2 or Fisher’s exact test were performed. Data are expressed as mean6SD. Two-tailed P-values ,0.05 were considered significant.
3. Results
3.1. Baseline variables The baseline data of the 37 patients who were interviewed at 10 months are shown in Table 1. These 37 patients showed similar baseline variables compared to the 49 patients who completed the 4-
R. Willenheimer et al. / International Journal of Cardiology 77 (2001) 25 – 31
28 Table 1 Baseline data Variable a
Training group (n5l 7)
Control group (n520)
Age (years) Women (%) IHD (%) Heart failure duration (months) LVEF LVEDD (mm / m 2 ) NYHA class Diuretic treatment (%) ACEi treatment (%) Digitalis treatment (%)
6465 29 85 28617 0.3560.12 3164 2.160.7 94 100 53
6468 30 78 24615 0.3860.10 3164 2.460.7 90 100 30
a ACEi, angiotensin-converting enzyme inhibitor; DFI, dyspnea fatigue index; IHD, ischemic heart disease; LVEF, left ventricular ejection fraction; LVEDD, left ventricular end-diastolic diameter adjusted for body surface; NYHA, New York Heart Association.
month active study period. The training (n517) and control (n520) groups were well balanced regarding all baseline variables. All but two patients had an ejection fraction less than 50% at inclusion. Six (16%) patients had chronic atrial fibrillation. The aetiology for heart failure was ischemic (documented myocardial infarction or angiographically verified coronary artery disease) in 29 patients (78%).
3.2. Compliance and safety Compliance with the exercise training (% sessions attended) among the 17 patients in the training group was 70%, ranging from 54.5 to 97.7, with the exception of one patient with only 27.3%. No adverse reactions to the training were reported. None of the training patients experienced any worsening of heart failure during the study, and none of them were hospitalised, whereas three of the control patients were. Five patients died during the extended followup (from 4 to 10 months), three of whom were
originally in the training group. Since these patients were not interviewed at 10 months they are not otherwise accounted for in this report.
3.3. Changes in effect variables during the active study period The results of the 4-month initial study period for the 37 patients interviewed at 10 months are shown in Table 2. They were similar to the results of all the 49 patients who completed the initial study period [3]. With training (n517) there was an improvement in all the four effect variables (peak-VO 2 , EC max , DFI, and PGACQoL) compared to controls (n520), although the between-group differences were not statistically significant (Table 2). Over the 4-month initial study period the within-group changes in peakVO 2 , EC max and DFI were not statistically significant among the training patients (Table 2), whereas PGACQoL improved significantly within-group (Table 2).
Table 2 Effect variables and PA score at baseline, 4 months, and 10 months Variable a
Peak-VO 2 (ml / kg / min) EC max (W) DFI (units) QoL (units) PA (units) a
Training group (n517)
Control group (n520)
Betweengroup P 0–4 m
Betweengroup P 0–10 m
Baseline
4m
10 m
P 0–4 m
P 0–10 m
Baseline
4m
10 m
P 0–4 m
P 0–10 m
16.263.2
17.364.4
–
0.133
–
16.163.3
16.663.6
–
0.232
–
0.454
–
113626 7.861.7 060 70697
120633 8.162.1 1.361.3 60685
– 8.262.2 0.760.9 48641
0.062 0.131 ,0.0001 0.914
– 0.269 0.006 0.240
112630 6.961.8 060 71668
112634 7.061.8 0.461.4 42655
– 6.462.0 0.061.0 32641
0.841 0.852 0.223 0.014
– 0.154 0.825 0.007
0.126 0.608 0.064 0.507
– 0.073 0.023 0.481
PA, physical activity score; m, months; peak-VO 2 , peak oxygen consumption; EC max , maximal exercise capacity; DFI, dyspnea fatigue index; QoL, patients global assessment of change in quality of life.
R. Willenheimer et al. / International Journal of Cardiology 77 (2001) 25 – 31
3.4. Quality of life, physical activity, and dyspneafatigue-index at the 10 -month assessment Both groups deteriorated PGACQoL during the 6-month extended follow-up, and the changes were of similar magnitude in the two groups (Table 2). Within- and between-group differences were not significant. The change in PGACQoL from baseline to 10 months differed slightly but significantly between the groups (Table 2). Training patients showed a small, statistically significant improvement, whereas control patients showed no change. Training and control patients did not differ significantly regarding the physical activity score at baseline, 4 months, or 10 months (Table 2). Both groups continuously decreased the physical activity score, from baseline to 4 months, during the extended follow-up, and from baseline to 10 months. Over the entire 10-month period this decrease was significant in the control group. There were however no significant between-group differences. The training patients were largely unchanged regarding DFI over the 6-month extended follow-up and during the entire 10 months (Table 2), whereas the control patients deteriorated somewhat. Withinand between-group differences were not significant.
4. Discussion We have previously reported that a 4-month exercise training programme improved exercise capacity and quality of life in 22 training patients compared to 27 controls [3]. Our findings were in good agreement with the results of other studies [1,2]. Six months after termination of the initial active study period 37 patients were interviewed regarding PGACQoL, habitual physical activity, and ability to carry out daily activities (DEL). These 37 patients were comparable to all the 49 patients regarding baseline data and the results of the initial 4-month intervention. Thus, although the changes in the effect variables at 4 months did not significantly differ between the 17 training patients and 20 controls (who were interviewed at 10 months), they were similar to those for all the original 22 training patients and 27 controls. Since the original 22 training patients significantly improved exercise capacity and quality of life vs. the
29
original 27 controls at 4 months, the lack of statistical significance between the 17 training patients and 20 controls (who were interviewed at 10 months) most likely was due to insufficient power as a result of the reduced number of patients. Thus, we feel that the patients interviewed at 10 months were representative to the entire study group completing the initial 4month active study period. Among the 17 training patients there was only a small (1.1 ml / kg / mm) and insignificant improvement in peak-V0 2 during the active training period. This was similar to the change found among all the 22 training patients (0.9 ml / kg / mm) [3]. In a prior publication [3], we reported that men with ischemic aetiology showed an increase in peak-VO 2 comparable to that found in other studies [1,2], including mainly this category of patients. The smaller overall increase in peak-VO 2 in our study may be due to the inclusion of other patient categories. Among the training patients the improvement in PGACQoL achieved during the initial 4-month training period was reduced considerably during the 6month extended follow-up. Despite this there was a slight but significant improvement at 10 months compared to baseline. This is however to be expected in patients who at the time of the interview knew that they had been training and who would have expected some sustained benefit from that effort. Therefore, this small sustained benefit in PGACQoL at 10 months is probably of no clinical significance. The training patients showed a trend towards a decreased physical activity score during the training programme. This is not surprising considering that the score did not take into account the activity during the training sessions. Interestingly, training patients did not spontaneously increase their habitual physical activity after participating in the 16-week exercise training programme. On the contrary, they decreased their physical activity, although statistically insignificantly. When the study was planned there was a concern that the control patients might start to exercise on their own during the active study period. At baseline they were therefore specifically asked not to. As shown by the physical activity score they did not increase their physical activity during the active study period. Nor did they increase physical activity during the 6-month extended follow-up, despite the fact that it was allowed. Instead, they decreased their
30
R. Willenheimer et al. / International Journal of Cardiology 77 (2001) 25 – 31
habitual physical activity during both study periods, and the decrease over the entire 10-month period was significant. There was essentially no change in the ability to carry out daily activities among the training patients, during the extended follow-up as well as during the entire 10 months. There was a trend towards a decline in the control group. There were however no significant between- group differences. Some of the prior randomised and controlled studies examining the effects of exercise training in heart failure patients had a cross-over design [8–12]. In these studies patients were specifically instructed to avoid physical training during the resting period. Thus, it has been possible to assess the effects of rest during 3–8 weeks after the end of the training period. Generally, the training effects have been lost during the resting period. Since patients in our study did not increase their habitual physical activity after the training period they are comparable to the patients in the prior studies in that respect. Our results seem to agree well with these prior results. In a recent exercise training study in elderly patients with heart failure [13], the cross-over design was applied in a different manner. In order to examine the effects of discontinued attendance at the exercise sessions patients were not instructed to reduce exercise, other than the formal training, during the 12-week resting period following the training period. Also in that study the effects of the exercise training were lost during the resting period. Our results and the results of that study differ somewhat with respect to quality of life. In the Owen study training had no beneficial effect on quality of life. Consequently, there was no effect to sustain after termination of training. In our study there was a significantly beneficial effect on PGACQoL from the training and this was to some extent sustained after termination of the training programme. In the Owen study quality of life was assessed by the Minnesota Living with Heart Failure Questionnaire, possibly not well suited for quality of life assessments in elderly heart failure patients [13]. No formal presentation of self reported global quality of life was made in that paper. Thus, since the differing results may be due to different methodology the results may not be in disagreement. However, it cannot be precluded that there is an actual difference between the two studies,
which might be explained by the fact that patients in our study were younger and trained more intensely, more often, and for a longer period of time. In the present study we did not assess exercise capacity or oxygen uptake at 10 months, 6 months after the active study period. Therefore, we do not know if there was a sustained effect on these effect variables. Consequently, we cannot compare our results with those of previous studies in that respect, and this is a limitation to our study.
5. Conclusions Four months exercise training in patients with heart failure carried no important sustained beneficial effects 6 months after termination of the training programme. A small sustained improvement in PGACQoL was seen in post-training patients. This was to be expected given that patients obviously were not blinded to the training and was, therefore, probably clinically insignificant. Controls significantly decreased the habitual physical activity over 10 months and post-training patients showed a similar trend. There were no sustained beneficial effects on the ability to carry out daily activities. Our results indicate that exercise training has to be continuing to result in any sustained benefit.
Acknowledgements We thank our co-workers, Evy Hallgren, Ingrid Ohlsson, Marie Holmberg, Katty Reuterskidld, and Anneli Iwarson, without the help of whom this study could not have been completed. This study was supported by a grant from the Swedish Society for Patients with Heart and Lung Diseases.
References [1] Piepoli ME, Flather M, Coats MS. Overview of studies of exercise training in chronic heart failure: the need for a prospective randomized multicentre European trial. Eur Heart J 1998;19:830–41. [2] The European Heart Failure Training Group. Experience from controlled trials of physical training in chronic heart failure. Protocol and patient factors in effectiveness in the improvement in exercise tolerance. Eur Heart J 1998;19:466–75.
R. Willenheimer et al. / International Journal of Cardiology 77 (2001) 25 – 31 [3] Willenheimer R, Erhardt L, Chine C, Rydberg E, Israelsson B. Exercise training in heart failure improves quality of life and exercise capacity. Eur Heart J 1998;19:774–81. ¨ ¨ ¨ K. Validity of clinical [4] Remes J, Miettinen H, Reunanen A, Pyorala diagnosis of heart failure in primary health care. Eur Heart J 1991;12:315–21. [5] Feinstein AR, Fisher MB, Pigeon JG. Changes in dyspnea-fatigue ratings as indicators of quality of life in the treatment of congestive heart failure. Am J Cardiol 1989;64:50–5. [6] Cline CMJ, Willenheimer RB, Erhardt LR, Wikiund I, Israelsson BYA. Health-related quality of life in elderly patients with heart failure. Scand Cardiovasc J 1999;33:278–85. [7] Rector TS, Kubo SH, Cohn IN. Patients self-assessment of their congestive heart failure Part 2: Content, reliability and validity of a new measure, the Minnesota Living with Heart failure Questionnaire. Heart Failure 1987;1:198–209. [8] Coats AJS, Adamopoulos S, Radaelli A et al. Controlled trial of physical training in chronic heart failure. Circulation 1992;85:2119– 31.
31
[9] Meyer TE, Casadei B, Coats AJS, Conway J, Davey PP, Sleight P. Angiotensin-converting enzyme inhibition and physical training in heart failure. J Intern Med 1991;230:407–13. [10] Adamopoulos S, Coats AJS, Brunotte F et al. Physical training improves skeletal muscle metabolism in patients with chronic heart failure. J Am Coll Cardiol 1993;21:1101–6. [11] Piepoli M, Volterrani M, Clark AL, Adamopoulos S, Sleight P, Coats AJS. Contribution of muscle afferents to the hemodynamic, autonomic, and ventilatory responses to exercise in patients with chronic heart failure. Effects of physical training. Circulation 1996;93:940–52. [12] Meyer K, Schwaibold M, Westbrook S et al. Effects of exercise training and activity restriction on 6-minute walking test performance in patients with chronic heart failure. Am Heart J 1997;133:447–53. [13] Owen A, Croucher L. Effect of an exercise programme for elderly patients with heart failure. Eur J Heart Failure 2000;2:65–70.