Effects of exercise on platelet reactivity after myocardial infarction: a randomized clinical trial

Abstract Exercise training (ET) can lower platelet reactivity in patients with cardiovascular risk factors. However, the effects of ET on platelet reactivity in higher-risk patients is unknown. The aim of this study was to evaluate the effects of ET on platelet reactivity in patients with recent myocardial infarction (MI). Ninety patients were randomly assigned 1 month post-MI to the intervention (patients submitted to a supervised ET program) or control group. All patients were on dual antiplatelet therapy (DAPT). Platelet reactivity by VerifyNow-P2Y12 (measured by P2Y12 reaction units – PRUs) test was determined at baseline and at the end of 14 ± 2 weeks of follow-up at rest (primary endpoint), and multiplate electrode aggregometry (MEA) adenosine diphosphate (ADP) and aspirin (ASPI) tests were performed immediately before and after the maximal cardiopulmonary exercise test (CPET) at the same time points (secondary endpoints). Sixty-five patients (mean age 58.9 ± 10 years; 73.8% men; 60% ST elevation MI) completed follow-up (control group, n = 31; intervention group, n = 34). At the end of the follow-up, the mean platelet reactivity was 172.8 ± 68.9 PRUs and 166.9 ± 65.1 PRUs for the control and intervention groups, respectively (p = .72). Platelet reactivity was significantly increased after the CPET compared to rest at the beginning and at the end of the 14-week follow-up (among the intervention groups) by the MEA-ADP and MEA-ASPI tests (p < .01 for all analyses). In post-MI patients on DAPT, 14 weeks of supervised ET did not reduce platelet reactivity. Moreover, platelet reactivity was increased after high-intensity exercise (ClinicalTrials.gov: NCT02958657; https://clinicaltrials.gov/ct2/show/NCT02958657). Plain Language Summary What is the context? Platelet reactivity is reduced after exercise training in healthy individuals and patients with cardiovascular risk factors, but the effect in higher-risk patients is unknown. High-intensity exercise in untrained individuals increases platelet reactivity. The effect of dual antiplatelet therapy in inhibiting exercise-induced hyperreactivity is poorly understood. What’s new? Exercise training did not reduce platelet reactivity in post-myocardial infarction patients. High-intensity exercise increased platelet reactivity in post-myocardial infarction patients on dual antiplatelet therapy. Exercise training did not attenuate the exercise-induced increase in platelet reactivity. What’s the impact? The study suggests that strenuous exercise, if indicated, should be applied carefully to patients with high risk of recurrent ischemic events, even if on optimal medical therapy and after being trained.


Introduction
Despite the development of more effective antiplatelet drugs and improvements in invasive coronary interventions, the risk of recurrent ischemic events after an episode of acute coronary syndrome (ACS) is high, with large contemporary registries showing a mortality of 20% 12 months after acute myocardial infarction (MI) [1,2].
Cardiac rehabilitation based on exercise training (ET), which is still an underused therapy [3], plays an important role in reducing cardiovascular mortality in coronary artery disease (CAD) patients [4].Moreover, it is well established that ET leads to beneficial physiological effects regarding autonomic regulation [5], endothelial function [6] and inflammatory activity [7].In addition, ET reduces platelet reactivity among healthy individuals [8], as well as in overweight [9] and hypertensive patients [10].
On the other hand, acute and strenuous exercise has been shown to trigger ACS.Among patients who have suffered an MI, 13.6% reported having performed physical activity within 1 hour before the onset of ischemic symptoms, resulting in a 2.3-fold increase in MI risk after exercise compared to rest [11].Furthermore, exercise-related MI risk seems to be particularly important when exercise is performed by untrained individuals [12].It has already been demonstrated that a bout of highintensity exercise leads to an increase in platelet reactivity in both healthy and CAD patients [13][14][15], which may be one of the main pathophysiological mechanisms involved in exercise-related MI.
To the best of our knowledge, however, there is no information on the effects of ET on platelet function at rest and after strenuous exercise among higher cardiovascular risk populations, such as in post-MI patients.This trial aimed to evaluate the effects of ET and acute high-intensity exercise on platelet reactivity in post-MI patients on dual antiplatelet therapy (DAPT).

Trial design
This was a prospective, randomized, single-center, and open-label clinical trial (NCT 02958657).This study was conducted according to a protocol approved by the Scientific Committee of Instituto do Coracao (InCor) and by the Research Ethics Committee of HC/FMUSP.
Consecutive patients with type I MI according to the third universal definition of MI [16], admitted between December 2016 and December 2018, were assessed for eligibility.After obtaining written informed consent, eligible patients were randomized to an intervention group and submitted to a supervised ET program or to a control group (Supplemental Figure S1).All patients received information about healthy lifestyles, including advice to engage in physical activity, and were treated with DAPT consisting of aspirin plus a P2Y 12 inhibitor.The choice of P2Y 12 inhibitor was based on physician discretion.

Participants
Patients older than 18 years of age were included up to 30 days after suffering a type 1 MI.Exclusion criteria were regular ET prior to suffering an MI, a contraindication to exercise or to DAPT, a left ventricular ejection fraction <45%, a Killip class III or IV, a hematocrit level <33% or >52%, a glomerular filtration rate <30 ml/min/1.73m2 by the Modification of Diet in Renal Disease formula [17], an indication of coronary artery bypass graft due to the index event, oral anticoagulation, concomitant participation in another clinical trial or refusal to engage in an exercise program.

Interventions
Patients underwent a cardiopulmonary exercise test (CPET) at the first visit (30 ± 5 days after suffering an MI) and at the end of follow-up (14 ± 2 weeks after the first visit).A detailed description of the CPET method is available in the supplementary material.

Exercise training program
After the first study visit, patients randomized to the intervention group started a supervised cardiac rehabilitation program at the Cardiovascular Rehabilitation and Exercise Physiology Unit in the InCor.The exercise training program consisted of a minimum of two and a maximum of three sessions weekly (according to the patient's availability).Each exercise session included 5 min of warm-up, 40 min of aerobic exercise on a cycle ergometer, 10 min of local strengthening exercises, and 5 min of cool down with stretching exercises.The aerobic exercise intensity was established by the heart rate that corresponded to the anaerobic threshold and was 10% below the respiratory compensation point obtained in the initial CPET.

Platelet function tests and analysis
Blood samples were collected at both study visits (at enrollment and 14 weeks after the first visit) immediately before and after each CPET to analyze platelet reactivity at rest and after high-intensity exercise.Platelet reactivity to adenosine diphosphate (ADP) was analyzed by VerifyNow P2Y 12 test (Accriva Diagnostics, San Diego, CA, USA) measured by P2Y 12 reaction units (PRUs), and the multiplate electrode aggregometry (MEA) ADP test (Roche Diagnostics, Rotkreuz, Switzerland) measured by the area under the curve (AUC).Platelet reactivity to arachidonic acid (the thromboxane pathway) was analyzed with the MEA aspirin (ASPI) test (Roche Diagnostics, Rotkreuz, Switzerland) and measured by the AUC.We also evaluated the immature platelet count (IPC, 10 3 /ml) and fraction (IPF, %) by a fully automated flow cytometer assay (Sysmex 2100, Sysmex, Norderstedt, Germany).More detailed information about the blood sampling and platelet reactivity tests is provided in the supplementary material.

Trial endpoints
The primary endpoint of the study was to compare platelet reactivity by VerifyNow P2Y 12 tests in the intervention and control groups at the end of follow-up.The secondary endpoints were a) to analyze the primary endpoint by MEA-ADP and MEA-ASPI tests; b) to analyze platelet reactivity by MEA-ADP, MEA-ASPI, IPC, and IPF before and after the first CPET in the overall population; and c) to analyze platelet reactivity by MEA-ADP, MEA-ASPI, IPC, and IPF tests before and after the last CPET in the control and intervention groups.

Sample size
The sample size was calculated considering platelet reactivity by VerifyNow P2Y 12 test in post-MI patients on DAPT with 32.9 ± 16 PRUs [18].We choose a 35% reduction in platelet reactivity in the intervention group in comparison to the control group as the hypothesis to be tested in the study, as previous data demonstrated clinical relevance with a similar magnitude of platelet reactivity reduction [19].For a power of 80% and a two-tailed alpha of 0.05, the estimated sample was 62 patients (31 patients in each group).Assuming a percentage of losses of 40%, due to early interruptions of the training program, 90 patients were included in the study.

Randomization
Randomization was performed utilizing a computer-generated random number table at a rate of 1:1 (1 control: 1 intervention) from the 1 st to 40 th patient, and at a rate of 1:2 (1 control: 2 interventions) from the 41 st to 90 th patient to adjust for more frequent early follow-up interruptions in the intervention group.

Statistical methods
The categorical variables are described as absolute numbers or percentages and were compared using the chi-square test or Fisher's exact test, as appropriate.The continuous variables are described as the mean and standard deviation (those with a Gaussian distribution) or the median and 25th and 75th percentiles (those with a non-Gaussian distribution).The Shapiro-Wilks test was used for normality evaluation.Paired Student's t-test (Gaussian distributions) or Wilcoxon (non-Gaussian distributions) tests were used for comparisons between the groups.A binary logistic regression model was performed in post hoc exploratory analysis.All tests were two-tailed, and a value of p < .05 was considered statistically significant.The statistical software package used for statistical analysis was IBM SPSS 25.0 Statistics.

Results
As shown in Figure 1, a total of 1,028 patients were assessed for eligibility; 938 patients were excluded for different reasons.Ultimately, 90 patients were randomized, and 65 patients (34 in the intervention group and 31 in the control group) completed follow-up and were analyzed for the primary endpoint.
The baseline characteristics of the population are presented in Table 1.The intervention and control groups were well balanced.Most patients were male, with a mean age of 58.9 years, and presented with ST elevation MI (60%).About 10% of the patients had not undergone percutaneous coronary interventions for the index event and were treated with optimal medical therapy only.Compared to patients submitted to percutaneous coronary interventions (PCI), this cohort was similar regarding demographics characteristics and comorbidities (p=NS for age, sex, hypertension, diabetes), but had more residual coronary stenosis (≥70% of coronary lumen obstruction) at discharge (80.0% vs. 45.6%,p = .04).Despite that, no statistical difference was observed in ischemic findings by CPETs (p = .10for the first and p = 1.0 for the second CPET) between patients treated with or without PCI.Approximately one-third of the population had diabetes mellitus and were previous smokers, and 41.5% had coronary obstruction in 3 or more vessels.Despite that the groups were similar, the Intervention group had a numerically higher proportion of STEMI and higher median troponin peak, with a numerically lower incidence of diabetes and a median glycated protein.All patients were on DAPT (aspirin [100%] plus clopidogrel [93,8%] or ticagrelor [6,2%]), and 84.6% were taking beta blockers at baseline.

Effects of exercise training on platelet reactivity
Visit 1 was performed 31.9 (±4.0) days after the patients had suffered an MI, and the mean follow-up period (interval between visits 1 and 2) was 13.8 (±1.4) weeks.Patients allocated to the intervention group attended 26.2 (±9.2) training sessions, which corresponds to a 72.5% adherence rate to the planned sessions.In the control group, only one patient (3.2%) reported regular unsupervised physical activity (≥150 min of moderate-intensity physical activity per week) during the follow-up.
Patients in the intervention group showed improvement in cardiorespiratory capacity at the end of the follow-up compared to baseline, characterized by a statistically significant increase in maximal oxygen consumption (VO 2 ) during the CPET at Visit 2 (24.2 ± 6.8 ml/kg/min) compared to the value obtained at Visit 1 (21.8 ± 5.2 ml/kg/min), p = .005.In contrast, in the control group, the maximal VO 2 value did not significantly change during follow-up (20.8 ± 4.6 ml/kg/min at Visit 1 and 21.7 ± 5.0 ml/kg/min at Visit 2, p = .18)(Supplemental Figure S2).
Moreover, there were no significant differences in platelet function by MEA-ADP (p = .16)and MEA-ASPI (p = .28)in the control and intervention groups at the end of follow-up.The results are shown in Figure 2A.
LVEF: left ventricle ejection fraction; CABG: coronary artery bypass graft; DAPT: dual antiplatelet therapy; ACS: acute coronary syndrome.Exercise and platelet reactivity after myocardial infarction 3 In a prespecified analysis in which only patients randomized to the intervention group who had adherence rates to the ET program greater than 50% (19 or more training sessions, n = 28) and 75% (28 or more training sessions, n = 16), were included; the results were similar to the results we found in the overall population for the primary endpoint (p = .64and p = .25,considering the VerifyNow P2Y 12 tests at the end of follow-up in the control and intervention groups with 50% and 75% adherence rates, respectively).
The primary endpoint was also consistent within the prespecified subgroups: there was no significant interaction between the primary endpoint and any of the subgroups analyzed (Supplemental Figure S3).We performed a post-hoc sensitivity analysis and observed similar results for the primary endpoint when we excluded patients not submitted to PCI for the index event (p = .99),patients with ischemic results in any CPET performed (p = .20)or patients with residual coronary stenosis (p = .83).We also performed a post-hoc sensitivity analysis excluding participants in the ticagrelor (n = 4) and the primary result was maintained: 178.5 ± 12.8PRU vs. 170.5 ± 10.7PRU, for control and intervention group, respectively (p = .62).

Effects of high-intensity exercise on platelet reactivity
Platelet reactivity was significantly increased by both the MEA-ADP and MEA-ASPI tests after high-intensity exercise (maximal CPET) compared to rest values at Visit 1 in the overall population (Figure 2B).Similarly, we observed higher total and immature platelet counts after the CPET compared to the rest (p < .01 for both).However, despite a numerical increase in median IPF, the difference for this parameter did not reach significance (Table 2).
At the end of follow-up, patients were submitted to the second maximal CPET, and even after the ET period, an exercise-induced increase in platelet reactivity by MEA-ADP and MEA-ASPI tests was observed in the intervention group (Figure 2C).There were also increases in total platelet count (p < .01);regarding IPF and IPC, despite a numerical increase, there were no significant changes (p = .83and p = .18,respectively) after the CPET compared to the rest in the intervention group.The control group patients showed similar results, despite not achieving statistical significance for the MEA-ADP test (Supplemental Table SI).

Effects of exercise training on platelet reactivity
We found that in patients with recent MIs on DAPT, cardiac rehabilitation based on supervised ET with moderate-intensity exercise for 14 weeks did not reduce platelet reactivity, either by the ADP pathway with VerifyNow P2Y 12 and MEA-ADP tests or by thromboxane pathway with MEA-ASPI test.Despite the fact that HTPR rates by VerifyNow P2Y 12 were also similar between groups, the HTPR rate by MEA-ADP was lower in patients submitted to ET at the end-of-follow-up when compared to the control group, which may suggest a possible benefit of ET in this population.However, the HTPR analysis should be interpreted with caution, mainly for the fact that there were few patients (n = 3) with HTPR by MEA-ADP at the end-of-follow-up, and the results were not consistent in both utilized methods.Despite some conflicting results, most of the evidence to date suggests a reduction in platelet reactivity with ET [8][9][10]20,21], but these findings were not reproduced in the present publication.However, to the best of our knowledge, our study is the first in the literature to analyze this issue in a high-risk post-MI population on DAPT, which may explain, at least partially, the controversy between studies.Moreover, the platelet function tests chosen (point-of-care, with whole blood) may also have influenced the observed differences.
Platelet activation remains high even in the later stages after acute coronary syndromes [22].In this context, the presence of cardiovascular disease, particularly after an acute event associated with thrombosis and ischemia, may lead to a high basal level of platelet activation, which may have attenuated the beneficial ET effect on platelet reactivity observed in our population.Most previous studies showing a reduction in platelet reactivity induced by ET included healthy patients or patients with cardiovascular risk factors but did not include patients with previous cardiovascular events [8][9][10]20,21].Similar to our findings, Arosio et al. did not find a reduction in platelet reactivity after ET in patients with symptomatic peripheral artery disease (PAD) [23].Furthermore, in addition to a high basal platelet activation status, PAD and post-ACS patients may not have achieved great improvement in cardiorespiratory capacity, which is likely due to their peripheral and/or cardiorespiratory limitations.Although the increase in maximal VO 2 in the ET group in our study was similar to that found in other cohorts of CAD patients [24], it was numerically lower than those observed in patients without cardiovascular disease [8,21] submitted to similar training volumes.Exercise and platelet reactivity after myocardial infarction 5 It should be noted that the platelet function tests used in our trial were point-of-care tests that utilized whole blood and hence were more physiological than those used in previous studies with platelet-rich plasma [8][9][10]20,21].The VerifyNow and MEA tests are validated, reproducible, and recommended by expert consensus for use in clinical practice [16] and are correlated with clinical outcomes such as stent thrombosis, bleeding, and mortality [25].
Another important issue was the concomitant use of DAPT.P2Y 12 receptor antagonists promote the important inhibition of platelet reactivity by the ADP pathway [26].Thus, it is possible that ET was not sufficient to overcome the antiaggregating effects of these drugs in our population.Moreover, there is considerable interindividual variability in the antiplatelet effect, particularly with clopidogrel, which was used by more than 90% of our population.This important variability may occur not only due to genetic characteristics (e.g.polymorphisms related to cytochrome P450 [27,28] but also due to interactions between concomitant therapies, such as proton pump inhibitors, which were used by approximately 30% of our patients [29,30]. It is possible that a more intense ET, such as high-intensity interval training (HIIT), that is associated with a more pronounced increase in maximal VO 2 among CAD patients [24,31] could have caused greater changes in platelet reactivity in this high-risk population.
Although we could not demonstrate a physiological benefit of exercise training, it is well established that cardiac rehabilitation based on exercise improves clinical outcomes [32][33][34].In a recent metanalysis of 85 randomized clinical trials, exercise-based cardiac rehabilitation was associated with less MI, hospitalization, and cardiovascular mortality among CAD patients.Additionally, cardiac rehabilitation may increase quality-of-life [33].Considering the metanalysis results and other known beneficial physiological effects of exercise training [5,6,34], our findings should not preclude the indication of supervised exercise to post-MI patients.

Effects of high-intensity exercise on platelet reactivity
We found higher platelet reactivity in patients after high-intensity exercise than at rest.This effect has already been documented in healthy individuals [8,13,15,35,36] and stable CAD patients [14,37,38].Moreover, among patients with CAD, even a bout of moderate-intensity exercise may increase platelet reactivity [39].These findings were similar to those observed in the present study but different from what has been previously demonstrated [8]; in our study, high-intensity exercise-induced platelet reactivity persisted, even after patients were submitted to supervised ET.Two studies [14,40] have evaluated the effects of DAPT on platelet reactivity induced by high-intensity exercise with conflicting results.Although all patients were on DAPT in our study, we found higher platelet reactivity after CPET compared to rest, before and after 14 weeks of training.
The present study observed, for the first time, an increase in IPC after physical exertion; this response was proportional to the observed increase in total platelet count, as the fraction of immature platelets remained stable, suggesting that the population of platelets possibly stored in the spleen and released after adrenergic stimulation in the circulatory system has a similar proportion of immature platelets when compared to the proportion of circulating platelets.Chamberlain et al. reported that the adrenergic stimulation of the spleen by exercise did not increase the concentration of indirect markers of immature platelets [41].In our study, this mechanism was assessed by an automated device that directly identified immature (or "reticulated") platelets by staining ribonucleic acids.This finding raises the hypothesis that an even more potent antiplatelet therapy would not reach the necessary threshold to inhibit exercise-induced platelet hyperreactivity, as it is associated with higher platelet turnover and more immature platelets being released in the circulatory system.
As platelet reactivity [42,43] and IPC [44] are associated with major cardiovascular events, our findings may be of clinical importance and suggest that post-MI patients may be at higher risk after high-intensity exercise, even when properly treated (100% of the patients on DAPT and statin, more than 80% received inhibitors of the renin-angiotensin system and more than 90% received beta-blockers).It should be noted, however, that the absolute increase in platelet reactivity induced by highintensity exercise was small, and we did not analyze sequential platelet tests in order to investigate how long platelet reactivity would take to return to baseline levels.So, it is speculative whether the exercise-induced platelet hyperreactivity may lead to any clinical implication in this population.

Platelets and exercise prescription post-MI
In addition to the increased platelet reactivity demonstrated in the present manuscript, high-intensity exercises are also associated with increased sympathetic tone, myocardial oxygen consumption, shear stress, and endothelial injury, besides activation of the coagulation cascade and inflammatory activity [6,7].These physiological exercise-induced changes may explain, at least partially, the increased risk of MI after high-intensity exercises [11].Current guidelines recommend that post-MI patients should start in low-to moderate-intensity exercise, be referred to an exercisebased cardiac rehabilitation program soon after discharge post-ACS, and contraindicates competitive sports in patients with CAD and high risk of exercise-induced events [45].
On the other hand, the HIIT strategy has been associated with greater cardiorespiratory fitness compared to moderate intensity exercise training [24,34] and, in spite of being cardiorespiratory fitness inversely associated with all-cause mortality, the HIIT effect in clinical outcomes is still not totally clarified.
Additionally, a metanalysis of 69 randomized clinical trials with CAD patients found that cardiovascular mortality was reduced by structured exercise training irrespective of exercise intensity, frequency, or session time and suggested that high levels of adherence to the training program were associated with lower mortality [32].
As our findings suggests that high-intensity exercise may not be totally safe in post-MI patients, it highlights the importance of implementing large randomized clinical trials focused in hard endpoints to compare the safety and efficacy of different exercise prescriptions in this population.

Limitations
Our study has some limitations.First, our study had an open-label design.Although the intervention (exercise training) made it impossible to blind the research participants, the investigators were also aware of the randomization group.The CPET, however, was performed by blinded investigators, as well as platelet function tests and IPC assays.
Second, it is possible that our findings could be due to lack of statistical power.We planned our sample size based on a hypothesized relative reduction of 35% in platelet reactivity with exercise training.Of note, novel P2Y 12 inhibitors, such as ticagrelor, can reduce platelet reactivity in around 60% compared with a less potent P2Y 12 inhibitor such as clopidogrel [19] and this reduction translates into a relative reduction in clinical hard endpoints of 16% [46].Therefore, we understand that the effect size of physical training over platelet reactivity that was hypothesized in our study was modest and is in accordance with the reduction in clinical events observed with physical training among CAD patients [33].Moreover, lower platelet reactivity reduction, as what was observed when high-dose clopidogrel (150 mg daily) was compared to standard-dose (75 mg daily) over 6 months and lead to 15% relative reduction in median platelet reactivity (211 vs. 250 PRU, respectively) [47] was not linked to hard endpoints.Third, we observed a lower than ideal adherence rate to the rehabilitation program and a high rate of early follow-up interruption in the intervention group: 35.8% of the patients randomized to the intervention group did not complete the follow-up, and among those who did, 27.5% abstention was observed.Despite being a limitation of the study, an adherence below the desired rate reflects what is observed in clinical practice and is inherent to intervention studies that involve lifestyle changes.Moreover, a sensitivity analysis that included only patients with ideal adherence to the training program did not change our main results.New strategies, such as home-based cardiac rehabilitation with remote or semisupervised monitoring that has already been shown to be equivalent to ET performed in rehabilitation centers in terms of reducing cardiovascular outcomes and increasing functional capacity [48,49] or more intense ET, should be conducted in future studies.
Fourth, despite finding an important increase in the platelet reactivity and in the IPC after exercise in our study, both among trained and untrained patients, these findings should be regarded as nominally significant and not adjusted for multiplicity.
Finally, all the participants received information about lifestyle changes and the importance of physical activity as routine care after suffering an MI.So, there is no guarantee that patients in the control group did not perform physical activity during follow-up.

Conclusion
Supervised exercise training for 14 weeks did not reduce platelet reactivity in patients on DAPT after suffering an MI.In addition, platelet reactivity was higher after acute and intense exercise than at rest, even after patients had completed the exercise training period and used DAPT.Our findings suggest a potential mechanism for the association between strenuous exercise and recurrence of ischemic cardiovascular events in post-MI patients.

Figure 2 .
Figure 2. Platelet reactivity at the end of follow-up in the control and intervention groups by VerifyNow P2Y 12 (primary endpoint), MEA-ADP and MEA-ASPI tests (A), before and after the first CPET in total population (B) and in control and intervention groups after second CPET (C) by MEA-ADP and MEA-ASPI tests.

Table II .
Baseline characteristics of the population.