To the best of our knowledge, this study is the first to document the significant influence of cardiac rehabilitation on serum apelin, myostatin and FSTL1 levels in patients after ACS. Moreover, despite some significant correlations, we have shown that serum apelin, myostatin, follistatin and FSTL1 levels are independent of the classical cardiovascular risk factors in patients after ACS undergoing cardiac rehabilitation.
Apelin was first discovered as the orphan G protein-coupled receptor APJ ligand. The gene that encodes apelin is located on chromosome Xq25-26.1. In humans, apelin expression is high in the heart, endothelial cells, chondrocytes, brain, skin, spleen, lungs and thymus; intermediate in the skeletal muscle; and relatively low in the pancreas, liver and kidney. There is high-quality published data on the beneficial role of apelin in cardiovascular system function. The following is a brief overview of the rich array of benefits that apelin provides for the circulatory system. First, apelin decreases blood pressure and this effect is mediated mainly through nitric-oxide synthase (eNOS) as well as regulation of the protein kinase B (Akt)/eNOS pathway. Elevated blood pressure is a well-proven cardiovascular risk factor leading to IHD, ACS, stroke and peripheral vascular diseases, and the cardioprotective role of apelin mediated by its ability to lower blood pressure is indisputable. Second, apelin plays an invaluable cardioprotective role in conditions of hypoxia, especially myocardial ischaemia. In response to hypoxia, the expression of apelin and its receptor is elevated, along with activation of the phosphoinositide 3-kinase (PI3K)/Akt and extracellular signal-regulated kinase (ERK) pathways, making apelin a crucial element of the hypoxia-counteracting system. During myocardial infarction, apelin decreases the infarct size and increases serum nitric oxide. What is important in the context of our study is that apelin also significantly improves cardiac function and repair after myocardial infarction by increasing angiogenesis, stromal cell-derived factor-1α, chemokine receptor 4 (CXCR-4) and homing of vascular progenitor cells. Apelin protects heart muscle against ischaemia/reperfusion injury by limiting the infarction size, ameliorating mechanical recovery or even reducing the ischaemia/reperfusion phenomenon itself. Third, apelin improves cardiac function in conditions of heart failure frequently caused by ACS by increasing cardiac output and decreasing blood pressure and peripheral vascular resistance. Finally, apelin acts against oxidative stress by diminishing ROS production and enhancing the activity of antioxidant enzymes.
We found that the serum apelin level almost doubled in group S after the intervention. Moreover, the serum apelin level was also significantly higher after the intervention in group S compared with group K. Hence, patients who do not undergo cardiac rehabilitation for many weeks after ACS have a low apelin level and they do not have a chance to benefit from the favourable effects of this myokine.
Myostatin is a member of the TGF-β superfamily; it is encoded by the MSTN gene located on chromosome 2q32.2. Skeletal muscle presents the highest expression of myostatin, although fat tissue and heart muscle also show high expression. Regarding the cardiovascular system, published data indicate that myostatin is an inhibitor of hyperplastic growth, cardiomyocyte proliferation and protein synthesis. Its increased expression in the heart is particularly a result of stress or developing pathology. In chronic heart failure, elevated myostatin production via the Smad-dependent pathway contributes to adverse cardiac remodelling. It leads to the reduction of cardiomyocyte hypertrophy and excessive synthesis of connective tissue. In a similar way, myostatin impacts the heart in the context of myocardial infarction, whereas in animal model its absence is connected with decreased fibrosis and a higher survival rate. Myostatin-mediated intensification of fibrosis also affects blood vessels, causing progressive vascular stiffness and the development of hypertension. Another negative influence on the vascular wall is the inhibition of nitric oxide production in endothelial cells. The suggested mechanism includes downregulation of eNOS expression via the aforementioned Smad pathway. Despite limited evidence, myostatin seems to have some crucial metabolic properties on the cardiovascular system. Tu et al. indicated that MSTN inactivation in mice has a protective impact on the development of insulin resistance, proatherogenic dyslipidaemia and atherosclerosis.
We observed a higher serum myostatin level in group S after the intervention compared with baseline and with group K. Given that previous studies have shown circulating myostatin is diminished in response to training, our results require careful consideration. Saremi et al. evaluated healthy male subjects who received creatine supplementation and underwent resistance training, while Hittel et al. examined middle-aged men without cardiac disease and determined the myostatin level in muscles, but not in the blood. Therefore, our myostatin results may be the result of the myocardial pathology caused by ACS, which is well proven to upregulate myostatin. A study by Casterillo et al. indicates that elevated myostatin levels can persist for up to 2 months following an ACS. Another reason explaining the increase in myostatin is the positive correlation of its concentration with aerobic capacity, the improvement of which may be a result of rehabilitation. However, the effect of elevated myostatin on the heart is ambiguous: on the one hand, it supports blood circulation in stress conditions. It prevents metabolic alterations leading to the activation of glycolysis and glycogen accumulation, which may contribute to the development of heart failure. On the other hand, in chronic states increased myostatin leads to unfavourable cardiac remodelling and fibrosis. Nevertheless, considering that cardiac rehabilitation ameliorates cardiovascular system dysfunction after ACS, we hypothesise that the training implemented in our study represents an acute stressor on the heart, and elevated myostatin is a response to maintain proper circulation. The exact role of myostatin in cardiac rehabilitation requires further investigation.
Follistatin is a glycoprotein encoded by the FST gene located on chromosome 5q11.2 and expressed in almost all human tissues. Its general function is binding to and neutralisation of activin and other TGF-β family members. As an inhibitor of myostatin, follistatin stimulates cardiac muscle growth and reduces fibrotic remodelling. In a rat model of myocardial infarction, the development of heart failure is associated with lower follistatin levels, indicating a positive impact on heart regeneration after injury. Follistatin-dependent neutralisation of activin A specifically influences the vascular component of cardiac diseases. Activin A is a proinflammatory factor that promotes the expression of cell adhesion molecules, facilitating leucocyte adhesion and endothelial dysfunction. Impaired endothelial function leads to atherosclerotic lesions, whose building elements (macrophages, smooth muscle cells and endothelial cells) produce activin A to support disease progression. On the other hand, elevated follistatin correlates with the incidence of type 2 diabetes. The underlying mechanism involves adipose tissue insulin resistance and diminished suppression of insulin-dependent lipolysis.
We observed no significant changes in serum follistatin after cardiac rehabilitation. Considering that follistatin has antagonistic properties to myostatin and the fact that we observed increased serum myostatin in our patients after the intervention, we speculate that the effects of myostatin predominate over the effects of follistatin. As previously mentioned, increased myostatin expression in the heart is a result of a pathology such as ACS. The elevated level can persist for up to 2 months following the ischemic episode. The observed increase in myostatin alongside sustained follistatin concentration may be due to the stimulation of production via follistatin-independent pathways. Given the significant difference in the mean time since ACS in groups K and S (22 vs. 4 weeks), the elevated level of follistatin in group K may reflect the temporally delayed initiation of remodeling mechanisms that protect against the development of heart failure. This is supported by studies in animal models, which showed an increase in follistatin concentration at 4 and 8 weeks post-MI compared to the concentration noted after one week. The presented results and considerations argue for the existence of more complex mechanisms regulating the expression of myostatin and follistatin and their mutual interactions, which require further investigation.
FSTL1, produced by heart muscle cells, promotes proliferation of stem cell-derived cardiomyocytes in conditions of hypoxia, so this myokine is likely one of the major factors responsible for efficient cardiac self-repair after ACS. However, this still needs to be confirmed in human studies. In an animal model, FSTL1 is upregulated in the heart after myocardial infarction and ischaemia/reperfusion injury and protects cardiac myocytes against hypoxia/reoxygenation-induced apoptosis in a manner that is dependent on Akt and ERK. Recently elevated FSTL1 levels have been proposed as a reliable early biomarker of cardiovascular risk in humans, highlighting the increasingly recognised importance of this myokine in clinical cardiology and cardiological diagnostics.
We found a significant increase in the serum FSTL1 level in group S after the intervention, the first evidence that this myokine increases in response to cardiac rehabilitation in patients after ACS. On the other hand, we documented a higher serum FSTL1 level in group K compared with group S at baseline. Considering that FSTL1 may be a cardiovascular risk biomarker, we hypothesise that cardiovascular risk in patients after ACS increases with time spent without cardiac rehabilitation and is notably high.
Despite the undoubtedly advantageous effect of cardiac rehabilitation on the cardiovascular system in patients after ACS - rendering cardiac rehabilitation a highly recommendable procedure in this population - the pathways that link cardiac rehabilitation with cardiovascular benefits are not completely clear. Our study provides new details for this linkage.
Group S showed a reduction in systolic and diastolic blood pressure due to cardiac rehabilitation, consistent with previous studies documenting that even short-term cardiac rehabilitation significantly reduces blood pressure. Apelin significantly decreases blood pressure, and the serum apelin level increased after the intervention in our study. Thus, we presume that the hypotensive effect of cardiac rehabilitation in group S was partially mediated by apelin. Apelin also ameliorates cardiac function and repair after ACS via intensified angiogenesis and upregulation of stromal cell-derived factor-1α, CXCR-4 and homing of vascular progenitor cells, and presents antioxidative properties. Thus, we hypothesise that the beneficial effect of cardiac rehabilitation in patients after ACS is driven by an elevated serum apelin level.
Myostatin seems to exert rather unfavourable effects on the cardiovascular system. However, some studies have shown that myostatin stabilises the metabolic status and energy homeostasis of the heart and counteracts cardiac hypertrophy. Therefore, we presume that this was the major role of elevated serum myostatin we observed after cardiac rehabilitation: heart muscle was not healthy and was negatively affected by ACS.
Although serum follistatin did not increase significantly after the intervention, we observed some negative correlations between the serum follistatin level and blood pressure in group S, which confirms the unquestionable cardioprotective effect of this myokine. Considering the fact that the blood follistatin level increases during exercise, we hypothesise that follistatin is also responsible for the benefits of cardiovascular system resulting from cardiac rehabilitation.
In a rat model of myocardial infarction, the blood FSTL1 level and FSTL1 expression in cardiac and skeletal muscles increase after aerobic training. Recently, in an animal model of myocardial infarction, researchers reported that FSTL1 is a molecular link between 4 weeks of training and training-derived cardiovascular benefits. Specifically, FSTL1 expression is induced; elevated blood FSTL1 correlates positively with FSTL1 expression in skeletal muscles and correlates negatively with reduced cardiac fibrosis; functional performance improves; and angiogenesis is induced in the myocardium. In group S, we documented a strong negative correlation between the serum FSTL1 level and maximum systolic blood pressure for the CPX according to Bruce's protocol. Note that the maximum systolic blood pressure for the CPX is a well-proven cardiovascular risk factor for MACE. Thus, we hypothesise that our 2-week cardiac rehabilitation protocol for patients after ACS increases serum FSTL1 and thus reduces the risk of MACE. This hypothesis needs to be tested in additional studies.
Our regression analysis of the relationship between serum myokine levels and cardiovascular risk parameters showed only a marginal association in case of FSTL1 and single associations for the other three myokines. This means that the favourable effect of myokines on the cardiovascular system in patients after ACS undergoing cardiac rehabilitation is not mediated by classical cardiovascular risk parameters (i.e. blood pressure, heart rate, body mass, BMI, FTC, MM and the result of the exercise test). Rather, these myokines, especially apelin, myostatin and follistatin, are independent factors linking cardiac rehabilitation with the resulting cardiovascular benefits. Our results support our hypothesis that myokines are not just another agent regulating cardiovascular system performance; they represent an independent pathway linking physical exercise with amelioration of cardiovascular system function of high cardioprotective potency in patients after ACS. While this hypothesis needs to be tested in future studies, if it is confirmed, it may bring numerous clinical benefits for patients with cardiovascular diseases. Nevertheless, based on our results and the literature, serum myokine levels should be interpreted as a new generation of biomarkers of cardiac rehabilitation efficiency in patients after ACS.
For group S, when we compared the STEMI and NSTEMI subgroups, we saw no differences in myokine levels at baseline or after the intervention. Furthermore, pre-post comparisons for the serum follistatin levels were the same for the STEMI and NSTEMI subgroups. After performing the Bonferroni correction for apelin, statistical significance in the pre-post comparison was maintained exclusively in the STEMI group. The observed difference may be the result of its cardioprotective effect when exposed to hypoxia and the reduction of the ischemia/reperfusion phenomenon, the severity of which is greater in STEMI patients. However, the serum myostatin level increased only in the NSTEMI subgroup, while the serum FSTL1 level increased only in the STEMI subgroup, in which it more than doubled. These results strongly suggest that these two types of ACS differ significantly not only in the pathophysiology, but also in the response to cardiac rehabilitation. As myostatin improves the metabolic status, energy homeostasis and proper circulation in the heart and prevents cardiac hypertrophy, our results indicate that only patients with NSTEMI experience these benefits of myostatin in response to cardiac rehabilitation. The lack of such a myostatin response in patients with STEMI emphasises the extreme cardiac pathology of STEMI. Conversely, the serum FSTL1 level increased in response to cardiac rehabilitation only in patients with STEMI. This suggests that this group of patients could benefit more from the effect of this myokine. However, the serum FSTL1 level did not differ between the STEMI and NSTEMI subgroups after the intervention. This suggests that patients with STEMI had a relatively low serum FSTL1 level at baseline, again highlighting the more serious cardiac pathology of STEMI compared with NSTEMI.
The issue of myokine response to cardiac rehabilitation in patients after ACS has not been well investigated. Thus, many issues require further clarification. Currently, the influence of different training modalities on myokine response remains unknown. Moreover, it should be clarified how the duration of each training session and the entire training programme alter serum myokine levels. Finally, it would be reasonable to investigate the issue of myokines at the molecular level, especially how physical training modifies the expression of different myokines in cardiac muscle. Future research on this topic should be multi-center and focus on a long-term assessment that includes changes in myokine concentrations, the factors influencing them, but also clinical aspects such as the effect of cardiac rehabilitation on exercise tolerance and survival. Further investigation is also warranted for the gaps in foundational knowledge concerning myokines demonstrated in this study - namely, the impact of the superimposition of ACS and physical exercise on myostatin concentration changes, and the precise action of the follistatin-myostatin axis in this specific context.
Our results have numerous clinical implications. First, we showed that a 2-week cardiac rehabilitation protocol is long enough to induce beneficial myokine response and that it is not too short even for clinical use. Second, based on ours and others' outcomes, in the future cardiac rehabilitation training protocols may be modified to yield the most favourable myokine response for the patient after ACS. Third, we demonstrated that patients with STEMI and NSTEMI differ in their myokine response after cardiac rehabilitation. Thus, it seems that future cardiac rehabilitation protocols should be tailored for different types of patients after ACS and individualised regarding the training modality and duration to maximise the clinical cardiovascular benefits for the patient. Fourth, determination of serum myokine levels should be considered for clinical use to stratify a patient's cardiovascular risk and to predict a patient's response to cardiac rehabilitation. The results regarding changes in myokine concentrations and clinical cardiovascular parameters in the study group showed parallel trends in both sets of measures -- specifically, an increase in apelin, myostatin, and FSTL1 accompanied by a decrease in SBP and DBP. However, subsequent logistic regression analysis revealed no significant correlation between these variables. Therefore, the identified myokines should be regarded as independent indicators of favorable physiological adaptation to exercise. Finally, based on ours and others' studies, in the near future supplementation with exogenous myokines may be investigated and implemented for patients after ACS to increase the advantageous effect of cardiac rehabilitation or even to enable to obtain such an effect in patients who may be unable to perform physical exercise.
The major strength of our study is its novelty: to the best of our knowledge, it is the first trial investigating the myokine response to cardiac rehabilitation in patients after ACS. Furthermore, our results should be translatable into future clinical guidance for cardiac rehabilitation. We investigated several outcomes - the serum levels of four myokines as well as anthropometric and cardiovascular parameters - which enabled us to compare the patients' responses to cardiac rehabilitation and to situate our results in the broader clinical context. We also compared the myokine response between patients with STEMI and NSTEMI to provide new knowledge on the differences between these two ACS types. Finally, we enrolled a relatively high number of patients in our trial, which strongly increased the credibility of our outcomes. This study is a part of the CARDIO-REH randomised study and represents an extension and continuation of our previous research.
The greatest limitation of our study is its relatively short duration. However, even a 2-week cardiac rehabilitation protocol yielded significant changes in the myokine response, and these findings have clinical implications for the future treatments of patients recovering from ACS. A further limitation of the short observation period is the limited ability to determine the clinical impact of cardiac rehabilitation, specifically through a comparison of exercise tolerance, readmission rates, and long-term survival nevertheless, it is important to emphasize that the primary objective of the study was to evaluate only the short-term effect of cardiac rehabilitation. Due to the aforementioned short observation period, the long-term effect on myokine concentration is unclear and provides a basis for planning further research in this direction. Second, the study population was recruited in a single center, thus a study group is relatively homogenous. The participants were similar in age, clinical status, and pharmacological therapy, which strengthens the internal validity of our observations but may compromise their external validity. This homogeneity also extended to a general lack of advanced comorbidities or severe functional limitations. For this reason, our results may not be representative of outcomes in older, frailer patients or those with significant conditions like chronic heart failure, renal disease, or morbid obesity. Although women constituted 36-40% of the study group - the sample size was insufficient to draw robust conclusions regarding sex-specific responses to CR. On the other hand, the study was conducted in an inpatient cardiac rehabilitation medical center rather than in an outpatient setting, which ensured appropriate adherence to the rehabilitation program and measurements standardization. Third, we did not examine myokine gene expression; we only used ELISA to measure serum levels. In a human study it would be necessary to obtain tissue biopsies to measure gene expression, a procedure that would have been quite harmful for our patients, especially considering that the great majority of them, in accordance with medical guidance, used antiplatelet therapy and thus have an increased risk of bleeding. Moreover, the use of ELISA allowed us to include a relatively large number of patients, which enhanced the reliability and value of our results. The baseline difference between the study and control groups in the post-ACS period also constitutes a limitation of our study. However, this difference arises from the fact that patients in group S were recruited during the recommended period for cardiac rehabilitation implementation, whereas patients in group K did not undergo cardiac rehabilitation. Therefore, this limitation should be taken into account and the results interpreted with caution.
Although the sample size was determined a priori based on expected differences in apelin levels, the relatively small number of participants still limits the generalizability of the findings. Moreover, the study may have been underpowered to detect smaller effect sizes for other biochemical outcomes, which increases the risk of type II error. We also acknowledge that the sample size was calculated based on the primary outcome and is not sufficient to conduct adequately powered subgroup (STEMI/NSTEMI) analyses. Thus, subgroup analyses were exploratory and should be interpreted with caution. Another limitation concerns the issue of multiple comparisons. While Bonferroni correction was applied to comparative statistics, the correlation and regression analyses were exploratory and thus not corrected for multiple testing. This approach increases the risk of type I error, and the results should be considered as preliminary, requiring further replication.