Aerobic exercise, but not isometric handgrip exercise, improves endothelial function and arterial stiffness in patients with myocardial infarction undergoing coronary intervention: a randomized pilot study

Cardiovascular diseases including myocardial infarction (MI) are a leading cause of death worldwide. They accounted for 31% of all deaths globally in 2016 [1], and 85% of these deaths are due to acute MI and stroke [1].

Endothelial dysfunction is a characteristic feature that precedes the development of atherosclerosis [2] and asymptomatic structural vascular changes [3] as well as clinical manifestations of acute MI. On the other hand, changes in the middle layer of arteries are associated with arterial stiffness and major cardiovascular outcomes such ST-elevation MI [4]. Thus, the assessment of both endothelial function and arterial stiffness may help prevent MI and/or recurrent infarctions.

Flow-mediated dilation (FMD) is a method that assesses endothelial function by measuring changes in arterial diameter in response to hyperemia [5] and it is an indirect measure of the risk for cardiovascular events, including MI [6, 7]. Pulse wave velocity (PWV) is a method used to detect changes in the middle layer of arteries and PWV measures are associated with vascular stiffness [8].

Aerobic exercise (AE) has become an integral part of cardiac rehabilitation programs [9‐12] while more evidence of vascular and hemodynamic benefits of other exercise modalities such as strength training is needed. According to the Physical Activity Guidelines for Americans, older adults should do multicomponent physical activity combining aerobic, muscle-strengthening and balance training [13]. Also, strength training is extremely important because it helps improving the ability to perform activities of daily living. As for its effects on the artery, low- to moderate-intensity strength training apparently improves endothelium-dependent vasodilation and decreases arterial stiffness [14‐16] while high-intensity strength training seems to induce greater arterial stiffness [17]. Yet, these effects have been associated with long-term adaptations in healthy populations.

As for isometric exercise (IE) in particular, studies have shown that a single IE session was able to induce an hypotensive effect in individuals with hypertension [18, 19] but this benefit has not been evidenced in other studies [20]. Regarding endothelial function, a single exercise session decreased FMD [21], and although this impairment may be interpreted as harmful to individuals with hypertension, these effects appear transient since eight weeks of IE improved resting FMD in this population [22]. Regarding arterial stiffness, there is evidence showing increased PWV during isometric handgrip exercise [23] and after one exercise session [24]. One randomized clinical trial that assessed the effects of a single IE session in patients with coronary artery disease reported increased PWV [25].

Given that (1) strength training have health benefits [26], (2) isometric handgrip is a form of exercise easily applicable for muscle strengthening, but (3) cardiovascular benefits of isometric handgrip exercise are not entirely understood in post-acute MI individuals (who have several risk factors for cardiovascular diseases and reduced cardiac function) and (4) evidence on this exercise modality is scarce, in particular in individuals post-acute MI, we conducted a randomized pilot study to compare the effects of one session of isometric handgrip exercise and one session of AE on FMD and PWV in volunteers undergoing percutaneous coronary intervention (PCI) after acute MI. Our hypothesis was that one single IE session is enough to improve FMD and PWV, and that one single session of either IE or AE have similar effects in this population.

We conducted a randomized, evaluator-blind, parallel‐group controlled pilot study following the Consolidated Standards of Reporting Trials (CONSORT) guidelines [27] and the principles of the Declaration of Helsinki. The research project was approved by the research ethics committee at Instituto de Cardiologia do Rio Grande do Sul/Fundação Universitária de Cardiologia (ICFUC) (protocol number 5326/17) and registered at ClinicalTrials (ID number NCT04000893; Registered 27 June 2019—Retrospectively registered, http://​www.​clinicaltrials.​gov). All volunteer participants read and signed a free informed consent form.

A total of 32 patients receiving care at ICFUC outpatient clinic were eligible to participate in the study from February 2019 to July 2019 (recruitment and follow-up). Of these, 14 were excluded and 25 were recruited and invited to participate (Fig. 1). All study participants had history of ischemic heart disease and type I MI as described by Thygesen et al. (2018). There were data losses for five participants: one refused to undergo FMD and PWV assessments after the AE session; another one refused to participate in the IE session; and FMD results for the AE session (2 participants) and for the IE session (1 participant) were not included in the analysis due to imaging issues.

Table 1 shows the participants’ demographic characteristics and main cardiovascular risk factors. Briefly, they were mostly male (70%), 58.6 ± 9.0 years of age (AE group) and 56.0 ± 9.8 years (IE group), white and had complete high-school education. Fourteen volunteers (70%) were previously diagnosed with hypertension; 17 (85%) had family history of cardiovascular diseases; six (30%) had diabetes mellitus, six reported excessive alcohol use and 11 (55%) reported smoking.

Table 2 shows echocardiographic and carotid data of our sample. Relative wall thickness (RWT) was within the normal range for most participants (65%). Regarding diastolic dysfunction, 16 participants (80%) showed impaired relaxation of the myocardium and only two showed elevated pulmonary pressure. As for the carotid arteries, 12 (60%) showed carotid artery stenosis of 50% or more.

Table 3 summarizes medication use by intervention group. β-blockers were the most commonly used drugs as recommended. There was no difference between the two groups regarding optimized medication. Table 4 shows acute MI-related characteristics and they were similar in both groups.

Figure 2 shows FMD measurements. Interestingly, brachial artery diameter was reduced from baseline (before cuff occlusion) in the IE session after one exercise session (pre-exercise: 4.430 mm vs. post-exercise: 4.066 mm; p = 0.016) and there was no difference compared to the AE group (pre-exercise: 4.462 mm vs. post-exercise: 4.485 mm; p = 0.853). Peak diameter following hyperemia increased from baseline in the AE group (p = 0.028) and no change was observed in IE group (p = 0.091). Thus, FMD values increased from baseline in the AE group (Δ = 4.9%; p = 0.034), but there is no difference between groups in post-exercise moment (Fig. 2a). Yet, since baseline brachial artery diameter (before cuff occlusion) was smaller post-exercise than pre-exercise in the IE session, we adjusted FMD values for baseline brachial diameter following the literature [38]. After adjustment the effectiveness of AE remained (p = 0.025) with no change in the IE group (Fig. 2b).

Figure 3 presents central blood pressure measurements. AE effectively reduced cSBP post-exercise compared to IE (Δ = 20.1 mmHg, p = 0.011) as well as from baseline values (Δ = 14.9 mmHg, p = 0.002). After the exercise sessions, PWV values (Fig. 3c) were slightly reduced from baseline in the AE group (Δ = 0.61 m/s, p = 0.044), but not when compared to the IE group. PVR decreased in the AE group compared to the IE group (p = 0.050) and from baseline (p = 0.014) (Fig. 3d). Individual values are shown in the Additional file 2: Fig. 1S. Moreover, no difference in augmentation index corrected for heart rate of 75 bpm (AIX@75) was observed in both groups (AE: from 11.9 ± 4.1 to 15.5 ± 2.6%, p = 0.472; IE: from 10.8 ± 2.4 to 11.1 ± 3.0%, p = 0.903).

As for peripheral blood pressure (Fig. 4), systolic (SBP, p = 0.021) and diastolic (DBP, p < 0.001) were lower at the end of the session in the AE than the IE group. In addition, DBP increased from baseline after the IE session (p = 0.033). Individual peripheral blood pressure measures are shown in the Additional file 2: Fig. 2S.

Central blood pressure values were moderately correlated with PWV and PVR: cSBP and PWV (r = 0.610; p < 0.001) and cDBP and PVR (r = 0.418; p = 0.007). Peripheral blood pressure values were also moderately correlated with PWV and PVR: SBP and PWV (r = 0.593; p < 0.001), SBP and PVR (r = 0.628; p < 0.001), and DBP and PVR (r = 0.587; p < 0.001).

The main finding of our study was that, overall, one single session of AE improved endothelial function as assessed by FMD as well as central blood pressure measurements (cSBP/cDBP and PWV) in volunteers undergoing PCI after acute MI. Both peripheral blood pressure and vascular resistance decreased after the AE session. One session of IE did not prove effective in improving these same parameters; however, it was not associated with damaging effects on the cardiovascular system. It is important to stress that our findings do not challenge the recommendation of strength training for post-MI individuals since strength training consisting of dynamic and/or isometric exercises promotes health benefits, but rather that this population should perform it together with other exercise modalities, mainly aerobic exercises. The study findings are clinically relevant and support current physical exercise recommendations for post-MI individuals undergoing primary PCI.

Previous studies with individuals with impaired endothelial function due to aging and/or cardiovascular risk factors have demonstrated that a single AE session can improve endothelial function [39‐41], which is consistent with our findings. Several mechanisms may explain improved endothelial function in response to AE, but the most widely studied mechanism is an increase in nitric oxide (NO) bioavailability that promotes an improvement in vasodilatory capacity [42]. Increased blood flow in response to one session of AE demands induces shear stress leading to higher NO bioavailability. Interestingly, different exercise modalities are associated with different patterns of luminal shear stress [5] resulting in distinctive vascular function responses [43]. During IE there is sustained mechanical compression of active muscles and muscular relaxation at the end of the movement determines an increase in blood flow leading to shear stress. Thus, while AE stimulus is continuous, IE stimulus is intermittent (muscular mechanical compression followed by blood flow release). Interestingly, it has been postulated that one session of IE increases sympathetic activity [44] so that arterial vasoconstriction might be greater than NO vasodilatory capacity. One finding of our study supporting this hypothesis is that after a single exercise session, resting brachial artery diameter was smaller (vasoconstricted) than pre-exercise diameter and it could explain our FMD results in the IE group. It is crucial to understand whether this is a transient effect, i.e., an acute/subacute effect in response to exercise, and if it has any long-term adverse vascular effects, particularly in patients undergoing primary PCI after myocardial infarction.

As for central blood pressure measurements, one session of AE reduced PWV while one session of IE did not have any central effects. Evidences has shown that shorter exercise duration (as our protocol, 30 min) was associated with favorable vascular effects, and longer exercise (~ 60 min) had adverse effects on vascular stiffness, mainly in older coronary patients [45]. PWV as a measurement of aortic stiffness is an independent predictor of adverse cardiac and cerebrovascular outcomes in post-MI patients undergoing primary PCI including death, nonfatal reinfarction, congestive heart failure and stroke. Based on our findings, we stress the importance of AE—even a single exercise session—for post-MI rehabilitation. They also suggest that IE does not seem an adequate approach for reducing central blood pressure as it did not prove effective in decreasing PWV and central blood pressure measurements, though it did not have negative effects. Evidence from prior studies shows that long-term high-intensity strength training can increase PWV in individuals with increased arterial stiffness [17]. These authors as well as others have reported that mild to moderate strength exercise does not reduce arterial stiffness [14‐16]. However, they have assessed dynamic exercises only, not isometric.

Considering there is a decrease in central systolic pressure and PWV as well as an increase in endothelium-dependent vasodilatory capacity, it would be also expected a decrease in PVR following AE, and this evidence may be associated with lower peripheral blood pressure in our study. SBP decreased by around 6 mmHg within 40 min of AE. Our results are clinically relevant and consistent with studies indicating a hypotensive effect after AE with SBP reduction by 5 to 7 mmHg [46]. In addition, IE increased cSBP and decreased baseline brachial artery diameter, but it did not have any effects on PWV, FMD, peripheral SBP and PVR. In particular with regard to peripheral SBP, other studies have not shown decreased BP values [20, 47] in response to a single session of isometric handgrip exercise. It should be noted that our study evaluated one exercise session only and that continuing isometric handgrip training could lead to a reduction in SBP (− 5.4 mmHg) and DBP (− 2.4 mmHg) as shown in a recent meta-analysis [48].

This present study has some limitations. Our sample was small (n = 20) because we sought to include volunteers with very similar clinical characteristics (post-MI adults undergoing PCI with the placement of drug-eluting stents). After achieving the predicted sample size, there were five losses on the intervention day (one with missing imaging study results and four dropouts due to long waiting times). We achieved 84% of the predicted sample size with 70% statistical power for the primary outcome. Thus, our small sample precludes sub-analyses by gender, carotid artery status, ejection fraction among others. Another limitation is that we did not include a control group (no exercise); however, our purpose was to compare one IE session versus one AE session [primary strategy in cardiac rehabilitation guidelines and recommendations [9‐12]]. Finally, it should be stressed that various exercise modalities involve different muscle masses. AE uses large muscle groups (area and mass) while IE uses smaller muscle groups. These differences could be further investigated using lower-limb IE (e.g., exercise in a stretching extensor bench).

When compared to single bout of AE, one session of IE apparently does not have vascular and hemodynamic benefits in patients undergoing primary PCI after MI. Nevertheless, our intervention of IE did not apparently show any negative vascular or hemodynamic effects, as reported in other populations [23‐25]. Given the small sample size, our findings do not challenge the importance of IE strength training, but they draw attention to the fact that this exercise modality should not be used as a single approach for vascular improvement in post-MI individuals with a profile similar to our sample.