Cardiac rehabilitation (CR) has long been considered a cornerstone in secondary prevention after acute coronary syndrome (ACS). It is a coordinated multidisciplinary intervention that aims to improve the cardiac patient's physical, psychological and social functioning. CR programs include exercise training and strategies to reduce modifiable cardiovascular risk factors such as hypertension, dyslipidemia, diabetes and smoking, and to improve adherence to pharmacologic and non-pharmacologic therapy.1,2

CR has been shown to reduce reinfarction, cardiovascular and non-cardiovascular readmission and death3,4,2,5 and to improve health-related quality of life, and to be cost-effective.1,6

Exercise training programs increase patients’ cardiorespiratory fitness, reducing symptoms and improving physiologic responses to physical effort.7 As a consequence, heart rate and blood pressure decrease during physical activity, reducing myocardial oxygen demand.1

The positive effects of CR in cardiovascular disease include stabilization of the progression of atherosclerosis and improvement in endothelial function, resulting in increased coronary compliance, elasticity and flow.1,8–11

Exercise testing is essential before CR to assess exercise capacity, to prescribe exercise training programs, to stratify cardiovascular risk and to estimate prognosis. After CR, it provides an objective measure of the program's effect on functional capacity.

Achievement of higher functional capacity is clinically significant because it not only impacts on health-related quality of life but it also is prognostically valuable, strongly predicting the risk of death. Assessing fitness at baseline and after CR is important since an improvement in functional capacity lowers the risk of death.12

Cardiopulmonary exercise testing (CPET) is the best tool for this purpose, providing a detailed assessment of functional capacity through the analysis of respiratory gas exchange during exercise. Among important CPET parameters is peak oxygen uptake (pVO2), which is an accurate measure of exercise capacity and an important predictor of outcome in patients with coronary artery disease (CAD).13–15 Peak VO2 reflects tissue oxygen uptake in exercise, and depends on cardiac output, arterial oxygen content and oxygen extraction ability of muscle tissue. A supervised exercise training program can increase pVO2 by 11-36%, with the greatest improvement in the most deconditioned individuals.3 Peak VO2 varies according to age and gender, so it is also adjusted in relation to the predicted value for age and gender (%pVO2).15

Another useful parameter, considered by some authors to be a better prognostic parameter than pVO2, is the slope of the ventilation/CO2 ratio (VE/VCO2 slope), which expresses ventilatory efficiency and has been shown to be an independent predictor of outcome.14,16,17 The ratio (VE/VCO2 slope)/pVO2 integrates the two above parameters and also has prognostic value.18

Another form of assessment is calculation of peak circulatory power (PCP), defined as pVO2 times systolic pressure at peak exercise.

Despite its advantages, CPET is not routinely used after ACS.

The aim of the present study was, using CPET parameters, to identify which patients would benefit most in terms of functional capacity after a CR program. Patients were divided according to pVO2 and left ventricular ejection fraction (LVEF) at baseline, and the benefit of CR was evaluated by comparing CPET performance before and after the program.

During the study period 4724 patients were admitted with ACS diagnosis. Of these, around 50% were transferred from non-tertiary centers without a catheterization laboratory and were followed in those hospitals after discharge. Of the 2363 (50%) followed in our center, only 12% agreed to be included in the hospital's CR program and completed the program.

The study included the 129 patients who underwent CPET before and after CR, 111 (86.0%) male, mean age 56.3±9.8 years, 88.1% of whom had suffered STEMI. The clinical characteristics of the overall population are presented in Table 1.

Compared to G1, G2 patients (with lower baseline functional capacity) were older, included fewer males and had a higher prevalence of diabetes and dyslipidemia (Table 1).

CR improved functional capacity in both groups. In G1, pVO2 increased from 17.7±1.8 to 22.1±7.2 ml/kg/min (p<0.001). PCP and exercise test duration also increased significantly (p<0.001 for both). VE/VCO2 slope/pVO2 decreased from 1.7±0.4 to 1.4±0.6 (p<0.001). There was no significant change in VE/VCO2 slope in this group (Table 2).

In G2 all CPET parameters improved after CR, pVO2 rising from 28.1±5.6 to 29.7±6.2 ml/kg/min (p<0.001) and %pVO2 from 96.3% to 100.3% (p=0.001). PCP and exercise test duration also increased significantly. VE/VCO2 slope and VE/VCO2 slope/pVO2 decreased significantly (p=0.032 and p<0.001, respectively) (Table 2).

As expected, CPET parameters were substantially different between groups before and after CR, with a better performance in G2. However, when range of improvement is compared, G1 achieved a greater benefit after CR. The improvement in pVO2 was greater in G1, in absolute value (ΔpVO2 4.4±7.3 vs. 1.6±5.4; p=0.018) and in percentage of predicted value (Δ%pVO2 17.9±35.4% vs. 4.0±19.9%; p=0.009). In G1 improvement in pVO2 was more than 10% in 71% of patients, contrasting with only 40.8% in G2. The increase in PCP was also significantly greater in G1 (883.3±1217.3 vs. 238.5±1250.5 mmHg.ml/kg/min; p=0.015). Exercise test duration increased more in G1, although not significantly (Table 3).

When patients were grouped according to LV systolic function, both groups presented improved functional capacity after exercise training. In GA, pVO2 increased from 24.7±6.1 to 27.7±6.9 ml/kg/min (p=0.002), %pVO2 from 84.9±19.4 to 95.4±24.0% (p=0.001) and exercise test duration from 13.4±2.6 to 15.0±3.3 min (p<0.001) (Table 2). GB patients also achieved a significant improvement, as demonstrated by increased pVO2, %pVO2, PCP and exercise test duration and decreased VE/VCO2 slope and (VE/VCO2 slope)/pVO2 ratio (Table 2).

There were no significant differences in CPET parameters between these two groups before or after CR. GA presented greater increases in pVO2 (3.1±5.6 vs. 2.0±6.1 ml/kg/min; p=0.382), %pVO2 (10.6±19.0 vs. 6.2±26.9%; p=0.291) and exercise test duration (1.6±2.5 vs. 2.0±2.6; p=0.920), although they were not significant.

In patients with initial VE/VCO2 slope >30, a significant improvement in this parameter was achieved after CR (34.4±5.8 vs. 31.1±5.1; p=0.02).

In a follow-up of 5.17±2.04 years, there were no significant differences in death, cardiac events, lifestyle or adherence to therapy between the study groups (Table 4). The lower mortality and ACS recurrence in all groups is noteworthy. Regarding lifestyle, more than 90% of the overall population maintained therapeutic adherence and more than half of patients performed regular physical exercise and kept to a healthy diet.

The main result of our study was that the CPET parameter pVO2 is useful in the selection of ACS patients who will have a better functional response to exercise training.

Our data indicate that the inclusion of ACS patients in a supervised 36-session exercise training program improved functional capacity independently of baseline performance, and that patients with pVO2 <20 ml/kg/min before CR derived greater benefit. The improvement in pVO2 was seen in absolute terms, as well as adjusted to predicted value for age and gender. As demonstrated in previous papers,1,21 our data showed that the most deconditioned individuals had the greatest improvement in functional capacity after the program, with 71% of patients in G1 achieving an improvement of >10% in pVO2 compared to 40.8% in G2.

This greater benefit was also reflected in VE/VCO2 slope/pVO2, PCP and exercise test duration, which have important prognostic implications. Kavanagh et al. demonstrated that the main predictor of outcome in CAD patients is exercise tolerance, as objectively measured by pVO2. A 10% decrease in mortality was observed for each increase of 1 ml/kg/min in pVO2.13

Comparing patients according to baseline LVEF, our research demonstrates that CR is beneficial regardless of LV systolic function. More than half of the patients (55.9%) in GA saw an increase in pVO2 of >10%, compared to 45.3% in GB. This finding, although not statistically significant, suggests that patients with LV dysfunction may derive more benefit from an exercise training program. Previous studies have shown that patients with LV systolic dysfunction benefit more from CR.22

The number of ACS patients undergoing CR was low for various reasons, as analyzed in previous studies, including financial constraints, difficulties in transportation, conflicts with work, low motivation and geographic distance. There are multiple barriers to CR, three of which have been identified as patient-related, physician-related and health system-related factors.23

Cardiac rehabilitation, though widely recommended in ACS guidelines, is poorly implemented in several countries, including Portugal, where previous studies have shown that only 3% of ACS patients discharged from hospital are admitted to a CR program.24–26 Although an increasing number of ACS patients have been included in CR programs in Portugal in recent years, recruitment in the country remains far below the European average of 30%.27

Including more patients in CR programs will require modifications in the health system, including increases in the number of CR centers and improvements in specialized training of health professionals. Failure by cardiologists and surgeons to refer patients for CR is particularly evident in some cases, including the elderly, women, obese patients, smokers and those of low socioeconomic status or suffering from depression.28 Only with a more suitable health structure and better trained professionals will be possible to recruit and motivate more ACS patients for CR.

Although in our center there is presently no limitation on the number of ACS patients who can be referred for CR, it should be borne in mind that many hospitals do not have facilities available to provide CR for all patients. Knowing who will derive the greatest benefit will help in the selection of the best responders to CR.

This is a single-center retrospective study and thus suffers from inherent limitations, including the small number of patients. The size of our sample was due to the fact that of the patients followed after discharge in our center, only a minority agreed to be included in CR at the hospital and completed the program, for the reasons described.

Furthermore, the present study only included patients who performed CPET before and after CR, and since CPET is not a routine procedure, this excluded patients who underwent standard exercise testing, which does not include gas exchange analysis.

In conclusion, of the heterogeneous group of patients with ACS, those with the worst functional capacity identified by CPET, and those with lower LVEF, obtained the greatest benefit from CR. These individuals should be the main target for referral for comprehensive CR, and CPET may be in fact a valuable tool for their identification.

The authors have no conflicts of interest to declare.