Sample characteristics
Of the 1,385 participants who provided informed consent for the follow-up and were alive in May 2019, 881 (63.6%) participated in the third follow-up of RHESA and filled out the corresponding questionnaire after up to two reminders. Additionally, there were 95 people who participated in the follow-up two years after AMI but did not participate in the third follow-up. Still, because they provided the relevant information for the current analysis in the earlier questionnaire they were also included in the analysis. Furthermore, we received information about 118 patients who died before their second follow-up through a questionnaire filled out by their respective GPs. The median duration of follow-up for all patients was 24 months. For patients, who experienced an event, the median duration of follow-up was 8 months.
Of the 1,094 participants that were included in our study, about one quarter (24.9%) had been enrolled in a DMP while 58.5% took part in CR after treatment of the initial AMI, 189 patients (17.3%) participated both in DMPs and CR. Of all patients, 33.9% did not participate in either CR or a DMP (Table 1). CR participants were more likely to be also enrolled in DMP and vice versa. Of all DMP participants, 18.3 % were enrolled in a DMP before the registered AMI. The remaining enrolled in median in the second month after hospital discharge.
Those who participated in CR were younger and more often smokers at the time of AMI than those who did not participate (Table 1). In contrast, those who participated in a DMP were less often smokers than those who did not. STEMI was most common among CR participants.
Table 1. Characteristics of the study population
|
|
DMP only
|
CR only
|
Both
|
None
|
Total
|
|
|
|
|
|
|
N=83
|
N=451
|
N=189
|
N=371
|
n=1094
|
|
|
Age (mean)
|
years
|
70.9
|
64.9
|
64.5
|
69.9
|
67.0
|
|
|
(95% CI)
|
(68.5-73.2)
|
(63.8-66.0)
|
(62.9-66.1)
|
(68.6-71.2)
|
(66.2-67.7)
|
|
|
Age groups
|
pct
|
|
|
|
|
|
|
|
25 – 49
|
|
3.6
|
11.5
|
12.2
|
6.2
|
9.2
|
|
|
50 – 59
|
|
16.9
|
25.1
|
22.2
|
18.9
|
21.9
|
|
|
60 – 69
|
|
19.3
|
28.2
|
27.5
|
20.5
|
24.8
|
|
|
70 – 79
|
|
42.2
|
23.7
|
33.3
|
31.0
|
29.3
|
|
|
80 +
|
|
18.1
|
11.5
|
4.8
|
23.5
|
14.9
|
|
|
Male sex
|
pct
|
77.1
|
71.8
|
68.3
|
69.5
|
70.8
|
|
|
|
(95% CI)
|
(66.6-85.6)
|
(67.4-76.0)
|
(61.1-74.8)
|
(64.6-74.2)
|
(68.1-73.5)
|
|
|
Diabetes
|
pct
|
27.7
|
18.6
|
9.5
|
14.8
|
16.5
|
|
|
(95% CI)
|
(18.5-38.6)
|
(15.1-22.5)
|
(5.7-14.6)
|
(11.4-18.9)
|
(14.3-18.8)
|
|
|
Smoker
|
pct
|
19.3
|
38.1
|
30.2
|
28.0
|
31.9
|
|
|
(95% CI)
|
(11.4-29.4)
|
(33.6-42.8)
|
(23.7-37.2)
|
(23.5-32.9)
|
(29.2-34.8)
|
|
|
Hypertension
|
pct
|
91.6
|
83.2
|
79.4
|
83.8
|
83.4
|
|
|
(95% CI)
|
(85.6-97.4)
|
(79.4-86.5)
|
(72.9-84.9)
|
(79.7-87.4)
|
(81.0-85.5)
|
|
|
Obesity
|
pct
|
20.5
|
25.9
|
17.5
|
18.9
|
21.7
|
|
|
(95% CI)
|
(12.4-30.8)
|
(22.0-30.3)
|
(12.3-23.6)
|
(15.0-23.2)
|
(19.3-24.2)
|
|
|
STEMI
|
Pct
|
28.9
|
51.0
|
48.7
|
34.0
|
43.1
|
|
|
(95% CI)
|
(19.5-39.9)
|
(46.3-55.7)
|
(41.4-56.0)
|
(29.2-39.0)
|
(40.2-46.1)
|
|
|
DMP = disease management program, CR = cardiac rehabilitation; smoker and diabetes only relevant if being current at the time of the initial myocardial infarction registered in the data base
age/age groups = age in years at the time of the initial acute myocardial infarction
About one third of all participants experienced MACE within two years of follow up and 9% experienced a reinfarction (Table 2). Those who participated in DMP had experienced more MACE than those who participated in CR, as evidenced by the deaths in the group of 83 DMP participants that did not take part in CR. The mean age of this subgroup was 71 years and therefore much higher than the average age of all DMP participants.
Table 2 Proportion of patients experiencing negative relevant outcomes within two years after AMI
|
DMP only
n=83
|
CR only
n=451
|
Both
n=189
|
None
n=371
|
Total
N=1094
|
MACE1
|
45.8
|
14.6
|
19.1
|
27.5
|
22.1
|
35.0-57.1
|
11.5-18.2
|
13.7-25.4
|
23.0-32.3
|
19.7-24.7
|
MACE2
|
51.8
|
24.6
|
31.2
|
35.0
|
31.4
|
40.6-62.9
|
20.7-28.9
|
24.6-37.8
|
30.2-40.1
|
28.6-34.2
|
Reinfarction
|
10.8
|
8.9
|
7.4
|
10.2
|
9.2
|
5.1-19.6
|
6.4-11.9
|
4.1-12.1
|
7.4-13.8
|
7.6-11.1
|
Stroke
|
1.2
|
3.3
|
2.7
|
3.0
|
2.9
|
0.0-6.5
|
1.9-5.4
|
0.9-6.1
|
1.5-5.2
|
2.0-4.1
|
PCI
|
9.6
|
8.7
|
11.1
|
10.0
|
9.6
|
4.3-18.1
|
6.1-11.2
|
7.0-16.5
|
7.1-13.5
|
7.9-11.5
|
CABG
|
0.0
|
5.8
|
5.8
|
3.0
|
4.4
|
0.0-4.3
|
3.8-8.3
|
2.5-9.2
|
1.5-5.2
|
3.3-5.8
|
Cardiac Death
|
13.3
|
0.7
|
2.7
|
5.7
|
3.7
|
6.8-22.5
|
0.1-1.9
|
0.9-6.1
|
3.5-8.5
|
2.6-5.0
|
Death (other)
|
25.3
|
2.9
|
9.0
|
11.3
|
8.5
|
16.4-36.0
|
1.5-4.9
|
5.3-14.0
|
8.3-15.0
|
6.9-10.3
|
MACE 1= composite endpoint including Reinfarction, Stroke and Death (cardiac / other),
MACE 2= composite endpoint including MACE 1 plus percutaneous coronary intervention (PCI) and coronary artery bypass graft (CABG)
DMP = Disease Management Program; CR = cardiac rehabilitation
Determinants of DMP enrollment and participating in CR
In the multivariable model, smoking at the time point of AMI was associated with lower participation in DMP, but not with lower participation in CR (Fig. 2 and 3, supplemental table 1). In contrast, higher age was associated with lower participation in CR but not in DMP. STEMI was also associated with increased participation in CR.
Fig. 2 Relative risk for DMP enrollment
Sex: reference = male; STEMI: reference = NSTEMI; Age per 10 years;
Remaining variables are binary (yes vs. no)
Fig. 3 Relative risk for participating in cardiac rehabilitation
Sex reference = male; STEMI = ST-elevation myocardial infarction with reference = non-STEMI;
Age = age in years continuously in 10 year steps;
Remaining variables are binary (yes vs. no)
Association between participation in DMP or CR after AMI and outcomes during follow up
The comparison of MACE1 and MACE2 showed higher absolute numbers of events and narrower CI for MACE2 without dilution effects. Thus, MACE2 was used as the primary endpoint in all cox regression analyses.
Participation in DMP was not associated with improved outcomes (crude hazard ratio = 0.93; 95% CI 0.65-1.33), while participation in CR was associated with risk reduction of about 50% (0.52; 0.41-0.65). These results were virtually unchanged after adjustment for age, sex, several diseases and a mutual adjustment for DMP and cardiac rehabilitation (Figure 4).
Overall, the effects of stratification for the considered subgroups were small indicating that selection of participants according to these variables did not strongly affect the impact of either CR or DMP.
Age, sex and obesity did not show an association with change in survival time in our two-year observation. Smokers showed a lower hazard rate with the confidence interval still containing the null effect (HR=0.76; 95% CI=0.55-1.05), similarly there was a slightly increased risk of MACE in participants with diabetes, but the confidence interval included 1 (HR=1.21; 95% CI 0.88-1.67).
Sensitivity analysis with 77.6% of all DMP participants who began their program after the AMI did not change the outcome noticeably. The HRs for DMP and CR were 0.98 (0.62-1.57) and 0.55 (0.43-0.71), respectively.
When stratified by the time point of smoking cessation, the effect was somewhat stronger in those CR-participants, who stopped smoking before the CR, when compared to those who did not stop smoking before CR or did not smoke (Table 3). However, a strong association between smoking status and CR effect was not found.
Table 3: Effects of participation in cardiac rehabilitation stratified by smoking status
Model
Effects of rehabilitation in those who
|
hazard ratio* (95% CI)
|
…did not smoke at time point of AMI
|
0.59 (0.45 – 0.78)
|
… smoked at time point of AMI
|
0.45 (0.27 – 0.74)
|
… smoked at time point of AMI, but stopped before rehabilitation
|
0.39 (0.18 – 0.83)
|
… smoked at time point of AMI, but did not stop smoking before rehabilitation
|
0.51 (0.26 – 1.01)
|
*participating in cardiac rehabilitation vs. no participation regarding MACE1 occurrence
Fig. 4 Association between secondary prevention or patient’s risk factors and MACE
Sex: reference = male; STEMI: reference = NSTEMI; Age = age in years continuously
Remaining variables are binary (yes vs. no); * = adjusted model