3.1_Subjects included
A total of 240 patients were included in the study (male: female = 185:55, males = 77.1%), with a median age of 56 years [IQR: 51; 62] and a median BMI of 28.05 [IQR: 26.20; 31.34].
204 of them (85%) had undergone a revascularization procedure; of these, 174 (85.3%) had been treated with percutaneous transluminal coronary angioplasty (PTCA), 25 (12.3%) with aorto-coronary bypass and 5 (2.4%) with both. 50 patients (20.8%) were included in the program due to heart failure. 32 patients (13.3%) were included after having undergone surgery for valvular disease. e-Table 1 shows the demographic and baseline characteristics of the participants.
91 patients (37.9%) carried out 2 sessions/week, while 149 (62.1%) managed to complete 3/week, based on risk stratification, as previously explained. As an exception, one patient with heart failure only participated twice a week due to logistic problems in granting his presence to the hospital for more than 2/week, despite his proven motivation in taking part to the program.
3.2_Characteristics of the patients at baseline and at the end of follow-up
Before starting the MCRP, 159 patients (66.2%) had poor sleep quality (PSQI > 5). 120 patients (50%) had anxiety disorder (GAS ≥ 5), while 168 (70.0%) had a depressive disorder (GDS ≥ 2). At the end of the treatment, there were 135 patients with poor sleep quality (56.2%), those with an anxiety disorder were 85 (35.4%), and those with a depressive disorder were 143 (59.6%) (Figure 1).
Table 1 shows the baseline and post MCRP values for sleep quality (PSQI), quality of life (SF-36), anxiety and depression as well as functional capacity.
Table 1
Sleep quality, quality of life, anxiety, depression and functional capacity of subjects before and after the Multidisciplinary Cardiac Rehabilitation Program.
|
|
PRE
|
POST
|
|
|
|
n
|
Median
|
p25
|
p75
|
Median
|
p25
|
p75
|
change
|
p
|
PSQI total
|
240
|
7.00
|
5.00
|
11.00
|
6.00
|
4.00
|
11.00
|
-1.00
|
0.008
|
PSQI Sleep quality
|
240
|
1.00
|
1.00
|
2.00
|
1.00
|
1.00
|
2.00
|
0.00
|
0.023
|
PSQI Latency
|
240
|
1.00
|
0.00
|
2.00
|
1.00
|
0.00
|
2.00
|
0.00
|
0.023
|
PSQI Sleep duration
|
240
|
1.00
|
0.00
|
2.00
|
1.00
|
0.00
|
2.00
|
0.00
|
0.403
|
PSQI Sleep efficiency
|
240
|
0.00
|
0.00
|
2.00
|
0.00
|
0.00
|
1.75
|
0.00
|
0.070
|
PSQI Sleep disturbancies
|
240
|
1.00
|
1.00
|
2.00
|
1.00
|
1.00
|
2.00
|
0.00
|
0.033
|
PSQI Sleep drug
|
240
|
0.00
|
0.00
|
1.00
|
0.00
|
0.00
|
2.00
|
0.00
|
0.499
|
PSQI Sleep dysfunction
|
240
|
1.00
|
0.00
|
1.00
|
1.00
|
0.00
|
1.00
|
0.00
|
0.023
|
SF-36 total
|
240
|
50.75
|
35.62
|
66.66
|
57.55
|
40.17
|
74.06
|
+6.75
|
<0.001
|
SF36 Physical function
|
240
|
67.50
|
46.25
|
83.75
|
70.00
|
50.00
|
85.00
|
+2.50
|
0.074
|
SF36 Physical role
|
240
|
0.00
|
0.00
|
75.00
|
25.00
|
0.00
|
100.00
|
+25.00
|
0.003
|
SF36 Bodily pain
|
240
|
57.50
|
32.50
|
80.00
|
67.50
|
36.87
|
90.00
|
+10.00
|
0.008
|
SF36 General health
|
240
|
45.00
|
35.00
|
55.00
|
45.00
|
35.00
|
60.00
|
0.00
|
0.709
|
SF36 Vitality
|
240
|
45.00
|
30.00
|
60.00
|
50.00
|
35.00
|
70.00
|
+5.00
|
<0.001
|
SF36 Social function
|
240
|
67.50
|
42.50
|
90.00
|
77.50
|
55.00
|
100.00
|
+10.00
|
<0.001
|
SF36 Emotional role
|
240
|
66.66
|
0.00
|
100.00
|
100.00
|
0.00
|
100.00
|
+33.34
|
0.004
|
SF36 Mental health
|
240
|
60.00
|
44.00
|
76.00
|
64.00
|
48.00
|
80.00
|
+4.00
|
<0.001
|
SF36 Health transition
|
240
|
37.50
|
25.00
|
75.00
|
50.00
|
25.00
|
75.00
|
+12.50
|
0.004
|
GAS
|
240
|
4.50
|
0.00
|
7.00
|
3.00
|
0.00
|
6.00
|
-1.50
|
<0.001
|
GDS
|
240
|
3.00
|
1.00
|
5.00
|
2.00
|
0.00
|
4.75
|
-1.00
|
<0.001
|
METS
|
240
|
7.00
|
5.30
|
8.77
|
8.15
|
6.60
|
10.20
|
+1.15
|
<0.001
|
Baseline and final values and statistical significance are shown. |
Abbreviations - PSQI: Pittsburgh Sleep Quality Index; SF-36: 36-Item Short Form Survey; GAS: Goldberg Anxiety Scale; GDS: Goldberg Depression Scale; METS: Metabolic equivalents of task. |
3.3_Changes in sleep quality
At the end of the program, the median overall score on the PSQI scale was reduced from 7 points to 6 points (p = 0.008). In particular, the subscales that registered a change were sleep quality (p = 0.023), sleep latency (p = 0.023), sleep disturbance (p = 0.033) and sleep dysfunction (p = 0.023) (Table 1). These changes have been statistically significant, but the improvement could not be considered clinically significant, since < 3 points21. A quantitative analysis of the number of patients who showed a clinically significant improvement was performed: out of the 240, 60 patients registered a clinically significant improvement (25,0%), 163 did not show any clinically significant change (67.9%), while 17 presented a clinically significant worsening (7.1%).
3.4_Changes in quality of life, anxiety and depression levels
At the end of the program, the median overall score of the SF36 questionnaire increased from 50.75 points to 57.55 (p < 0.001). The subscales that registered a change were physical role (p = 0.003), bodily pain (p = 0.008), vitality (p < 0.001), social function (p < 0.001), emotional role (p = 0.004), mental health (p < 0.001 ) and health transition (p = 0.004) (Table 1).
To quantify the size of the changes in the different subscales, reference can be made to the hypothetical standard changes considered by Wyrwich and colleagues29. According to their work, the changes obtained in the physical role (+25 points: moderate effect) and emotional role (+33.33: moderate effect) subscales may be considered clinically significant.
At the end of the program, both the scores in the anxiety (-1.50, p < 0.001) and the depression subscale (-1.00, p < 0.001) were decreased (Table 1).
3.5_Changes in functional capacity
At the end of the program, the median functional capacity improved from 7.00 to 8.15 METS (p <0.001) (Table 1).
3.6_Relationship between quality of sleep, quality of life, anxiety, depression and functional capacity
After evaluating the correlation at baseline of quality of life, anxiety, depression and functional capacity with the PQSI score, we found a negative correlation with the quality of life score (p = 0.005) and a positive correlation with both Goldberg scales (p < 0.001). However, these correlations were weak (ρ < 0.55), as evidenced in Table 2.
Table 2
Correlation coefficients of the initial and final values of SF-36, Goldberg Anxiety Scale, Goldberg Depression Scale and functional capacity with the Pittsburgh Sleep Quality Index.
|
SF-36
|
GAS
|
GDS
|
METS
|
|
Spearman's correlation coefficient
|
p
|
Spearman's correlation coefficient
|
p
|
Spearman's correlation coefficient
|
p
|
Spearman's correlation coefficient
|
p
|
PRE
|
-0,485
|
0,005
|
0,529
|
<0,001
|
0,468
|
<0,001
|
-0,086
|
0,187
|
POST
|
-0,526
|
<0,001
|
0,512
|
<0,001
|
0,486
|
<0,001
|
-0,161
|
0,012
|
Abbreviations - SF-36: 36-Item Short Form Survey; GAS: Goldberg Anxiety Scale; GDS: Goldberg Depression Scale; METS: Metabolic equivalents of task. |
After applying the multivariate linear regression model, we saw that only the quality of life and the GAS independently correlated with the PSQI score (Table 3). These correlations were those maintained after the conclusion of the program (Tables 2 and 3).
Table 3
Linear correlation of the initial and final values of Pittsburgh Sleep Quality Index with respect to the SF-36 and Goldberg scales
|
Baseline
|
Post MCRP
|
|
β
|
95% CI
|
p
|
β
|
95% CI
|
p
|
SF-36
|
-0.055
|
-0.085/-0.025
|
<0.001
|
-0.068
|
-0.098/-0.037
|
<0.001
|
GAS
|
0.457
|
0.233/0.682
|
<0.001
|
0.342
|
0.100/0.584
|
0.006
|
GDS
|
0.190
|
-0.109/0.490
|
0.212
|
0.141
|
-0.165/0.448
|
0.364
|
constant
|
8.351
|
6.104/10.59
|
<0.001
|
9.759
|
7.378/12.141
|
<0.001
|
R2
|
0.333
|
0.340
|
Abbreviations - CI: confidence interval; MCRP: Multidisciplinary Cardiac Rehabilitation Program; SF-36: 36-Item Short Form Survey; GAS: Goldberg Anxiety Scale; GDS: Goldberg Depression Scale; |
3.7_Differences in response to treatment between patients with and without heart failure
To assess the baseline status of patients with and without heart failure, a bivariate analysis of all baseline characteristics was performed between the two groups of patients. Patients with heart failure had a higher prevalence of moderate to severe systolic dysfunction and a lower baseline functional capacity (5.27 vs 7.10, p < 0.001); compared to those without heart failure. On the contrary, no statistically significant differences were found in sleep quality (p = 0.909), quality of life (p = 0.135), anxiety (p = 0.092) or depression (p = 0.281) (Table 4).
Table 4
Comparison of baseline data based on heart failure
|
without HF
|
with HF
|
p
|
|
N = 190
|
79.2%
|
N = 50
|
20.8%
|
|
Demographics
|
|
Male N | %
|
148
|
77.9
|
37
|
74.0
|
0.560
|
Age median | IQR
|
56.00
|
10.00
|
58.00
|
13.00
|
0.394
|
BMI median | IQR
|
27.80
|
5.13
|
28.50
|
5.55
|
0.438
|
Pittsburgh Sleep Quality Index
|
|
PSQI median | IQR
|
7.00
|
6.25
|
7.00
|
6.25
|
0.909
|
Short Form-36 questionnaire
|
|
SF-36 total score median| IQR
|
49.96
|
33.93
|
54.10
|
23.52
|
0.135
|
Goldberg scales Median | IQR
|
|
GAS median | IQR
|
5.00
|
5.00
|
3.00
|
7.00
|
0.092
|
GDS median | IQR
|
3.00
|
4.00
|
3.00
|
3.25
|
0.281
|
METS median | IQR
|
7.10
|
3.30
|
5.27
|
2.04
|
< 0.001
|
Systolic Function*
|
|
Normal function n | %
|
110
|
57.9
|
7
|
14.0
|
< 0.001
|
Mild dysfunction n | %
|
56
|
29.5
|
6
|
12.0
|
Moderate dysfunction n | %
|
17
|
8.9
|
16
|
32.0
|
Severe dysfunction n | %
|
7
|
3.7
|
21
|
42.0
|
Abbreviations - HF: heart failure; IQR: interquartile range; BMI: body mass index; PSQI: Pittsburgh Sleep Quality Index; SF-36: Short Form-36 questionnaire; GAS: Goldberg Anxiety Scale; GDS: Goldberg Depression Scale; METS: metabolic equivalents of task |
*Systolic Function classification according to the Left Ventricular Ejection Fraction (LVEF), as reported in Lang RM et al., 2015 – Normal: LVEF 50–70% (midpoint 60%); Mild dysfunction = LVEF 40–49% (midpoint 45%); Moderate dysfunction = LVEF 30–39% (midpoint 35%); Severe dysfunction = LVEF less than 30% |
Comparing the characteristics of the two groups at the end of program, patients with heart failure finished the program with a better score on the SF-36 questionnaire for quality of life (66.72 vs 54.94, p = 0.015). As for METS values, patients with heart failure had a lower score at the end of the program than patients without heart failure (5.68 vs 9.10, p < 0.001) (Table 5).
Table 5
Comparison of final data based on heart failure
|
without HF
|
with HF
|
|
|
N = 190
|
79.2%
|
N = 50
|
20.8%
|
|
|
Median
|
IQR
|
Median
|
IQR
|
p
|
PSQI
|
6.00
|
7.00
|
6.00
|
4.25
|
0.288
|
SF 36 total score
|
54.94
|
32.43
|
66.72
|
24.24
|
0.015
|
SF36 Physical function
|
70.00
|
35.00
|
60.00
|
26.25
|
0.032
|
SF36 Physical role
|
0.00
|
75.00
|
50.00
|
100.00
|
0.035
|
SF36 Bodily pain
|
67.50
|
55.00
|
78.75
|
55.00
|
0.072
|
SF36 General health
|
47.50
|
25.00
|
45.00
|
20.00
|
0.518
|
SF36 Vitality
|
45.00
|
35.00
|
52.50
|
26.25
|
0.065
|
SF36 Social function
|
72.50
|
45.00
|
87.50
|
45.00
|
0.311
|
SF36 Emotional role
|
66.66
|
100.00
|
100.00
|
36.30
|
0.002
|
SF36 Mental health
|
64.00
|
32.00
|
76.00
|
37.00
|
0.026
|
SF36 Health transition
|
25.00
|
50.00
|
75.00
|
50.00
|
<0.001
|
GAS
|
3.00
|
6.00
|
2.50
|
4.00
|
0.085
|
GDS
|
2.00
|
5.00
|
1.50
|
3.25
|
0.061
|
METS
|
9.10
|
3.23
|
5.68
|
2.43
|
<0.001
|
Abbreviations - HF: heart failure; IQR: interquartile range; PSQI: Pittsburgh Sleep Quality Index; SF-36: Short Form-36 questionnaire; GAS: Goldberg Anxiety Scale; GDS: Goldberg Depression Scale; METS: metabolic equivalents of task |
Evaluating the response to treatment between patients with and without heart failure heart failure, differences in the improvement in funcional capacity could also be observed, greater in patients without heart failure (1.00 vs 0.25, p = 0.005). As for the median changes in sleep quality, quality of life, anxiety and depression, no statistically significant differences could be observed (Table 6).
Table 6
Comparison of differences between baseline and final values based on heart failure
|
Without HF
|
With HF
|
|
|
N = 190
|
79.2%
|
N = 50
|
20.8%
|
|
|
Median
|
IQR
|
Median
|
IQR
|
p
|
PSQI
|
0.00
|
4.00
|
-1.00
|
4.00
|
0.183
|
SF 36 total
|
3.39
|
15.07
|
4.84
|
20.02
|
0.427
|
GAS
|
0.00
|
2.00
|
0.00
|
2.25
|
0.430
|
GDS
|
0.00
|
1.00
|
0.00
|
2.00
|
0.986
|
METS
|
1.00
|
2.60
|
0.25
|
1.97
|
0.005
|
Abbreviations - HF: heart failure; IQR: interquartile range; PSQI: Pittsburgh Sleep Quality Index; SF-36: Short Form-36 questionnaire; GAS: Goldberg Anxiety Scale; GDS: Goldberg Depression Scale; METS: metabolic equivalents of task |
3.8_Differences in response to treatment between patients who have undergone 2/week sessions and those who have undergone 3/week sessions
In our study, low-risk patients underwent 2 sessions/week, while medium and high-risk patients had 3 sessions/week. The only exception was represented by a patient from the heart failure group who followed the 2/week program due to problems getting to the hospital. To verify whether this may bias the analysis, the results in median changes obtained for all the variables have been compared, showing no differences when performing 2 or 3 weekly treatments (p > 0.05).
3.9_Monitoring for complications
During the treatment sessions, no major complications of traumatic, orthopedic or cardiovascular nature have been observed, that would imply a suspension of the training program. 13 patients experienced an episode of angina pectoris and 1 an episode of supraventricular tachycardia. These complications were resolved after medical evaluation, without the need of suspending the cardiac rehabilitation program.