Background
Cardiovascular and cerebrovascular diseases are the major cause of mortality worldwide. In developed countries, stroke is the third leading cause of death. In USA, on average, every 40 s someone has a stroke, with women having a higher lifetime risk of stroke than men (each year ~55 000 more women than men have a stroke event) [
1].
Stroke is also the leading cause of serious long-term disabilities [
1], and further cardiac disease has been found to occur in up to 75 % of stroke survivors. In individuals with stroke, cardiac comorbidities may complicate the course of disease and contribute to early mortality [
2]. Alterations initiated by cerebrovascular disease may negatively change the autonomic function and lead to cardiac impairment; or they may lead to a cerebral event, thus making more severe the existing autonomic dysfunction associated with cardiovascular risk factors [
3,
4].
In the acute post stroke phase, individuals show autonomic imbalance characterized by decreased vagal modulation and increased sympathetic cardiac modulation [
3,
5,
6]. This autonomic imbalance may contribute for end-organ damage, predispose to cardiovascular events [
7] and it is correlated with the severity of neurological deficits and disability [
8]. However, little is known about cardiac autonomic changes in patients after chronic stroke, particularly in women.
Heart rate variability analysis (HRV) is a well-reputed noninvasive method used to assess the autonomic modulation of the heart. This method allows the sympathetic and parasympathetic handles of the autonomic nervous system to the heart to be evaluated [
9]. Exercise tests may also be an important tool to evaluate autonomic cardiovascular modulation and its responsiveness. In this sense, measuring heart rate recovery (HRR) after exercise tests may reveal the extent of reactivation of the vagal activity [
10,
11]. Furthermore, studies have demonstrated that HRR is associated with short-term heart rate variability, and both have been associated with increased risk for cardiovascular events and sudden death [
12‐
15].
Thus, given that stroke is more prevalent in women than in men, and that residual autonomic consequences of the stroke may influence prognosis, the aim of this study was to evaluate cardiac autonomic modulation in women with chronic stroke (at least 4 years of diagnosis) at rest and in response to submaximal exercise test through linear and nonlinear analyses.
Discussion
The main finding of our study is that patients with chronic stroke (5 years in average) presented decreased heart rate variability, autonomic imbalance and impaired cardiac vagal modulation, as measured by linear and nonlinear analysis. Other important findings in women with chronic ischemic stroke, when compared to controls, include: (1) reduced values of VO2 peak and higher levels of blood lactate; (2) decreased HRR measured at 1, 2 and 3-min after submaximal exercise test; (3), improved hemodynamic and cardiac autonomic parameters during the recovery period (20 min after the end of exercise test).
Most of the survivors had residual disabilities caused by stroke, such as hemiparesis and spasticity, while total recovery was much less frequent. Activity limitations were demonstrated by the reduced ability to perform daily tasks and basic self-care, leading the individual to chronic sedentary behavior [
22]. Regarding this issue, in a study conducted by Gadidi et al. [
23], the percentage of subjects reporting some activity limitation 4 years post stroke was 42.3 %, while 28.2 % pointed to less severe limitations and 78.1 % felt they had not fully recovered. Although we did not classify individuals according to their self-reporting of activity limitations, we observed that VO
2 peak was reduced by 21 % in stroke women, and blood lactate levels were higher at rest and after exercise test when compared to the controls, which seems to indicate physical deconditioning and sedentary lifestyle on stroke survivals.
Physical deconditioning usually leads to physiological and metabolic changes in the paretic muscle. These changes are characterized by decreased blood flow, increased lactate production, increased muscle glycogen utilization and decreased ability of fatty acid oxidation. In addition, changes in muscle fibers during exercise were observed: active paretic muscle activated glycolytic type II fibers to initiate contraction, while the non-paretic muscle recruited primarily type I fibers. These changes usually lead to disuse and causes decreased oxidative metabolism, low resistance to aerobic exercise, early fatigue, sedentary lifestyle and deconditioning [
24‐
26].
Blood pressure values were higher in S group (at rest, in the post exercise and recovery period) when compared to C group. Although 57 % of S group individuals had prior history of hypertension, blood pressure values observed in these patients were within normal parameters, showing that the pressure levels were controlled [
16]. Similarly, Dütsch et al. [
27] have found no alterations in blood pressure in stroke patients 30 months after stroke. In addition, since that double product seems to be an indirect predictor of myocardial oxygen consumption [
28,
29], our findings suggest that women with chronic stroke require higher myocardial work when compared to the control group.
In the present study, even after nearly five years post-stroke, women presented reduced time domain parameters (VarNN, SDNN, rMSSD and pNN50) of heart rate variability at rest. Similarly, Muslumanoglu et al. [
5] have observed reduced values for VarRR, SDNN and pNN50 in the post-stroke acute phase. Dütsch et al. [
27] have shown that post–acute stroke patients presented parasympathetic cardiac deficit and higher LF/HF than age-and sex-matched controls. In our study, women with stroke showed a decrease in both HF band of spectral analysis and in 2V pattern of symbolic analysis. These two measurements serve as indicators of both vagal cardiac modulation and autonomic imbalance, as demonstrated by the LF/HF ratio.
Since the 80s, researchers have indicated that physical inactivity is associated with negative changes in autonomic nervous system. Reduction in blood volume affects cardiac stroke volume; as such, maintaining oxygen delivery [
30] requires an increase in heart rate triggered by increased sympathetic and reduced parasympathetic activity to the sinoatrial node Furthermore, studies with experimental models of physical inactivity, associated with a sedentary lifestyle or extreme forms of inactivity with bed rest or spaceflight, have pointed to a decrease in parasympathetic drive and an increase in the sympathetic tonus to heart [See for review,
31]. These findings may explain, at least in part, the autonomic imbalance observed in chronic stroke individuals observed in the present study.
Some stroke subjects (57 %) were diagnosed with hypertension before the event, and most of them were being treated with angiotensin-converting enzyme (ACE) inhibitors. Most antihypertensive drugs induce a rearrangement of the autonomous nervous system. It is well-established that central sympatholytic agents and beta-blockers induce amplified inhibitory effects on sympathetic activity. Furthermore, angiotensin-converting enzyme inhibitors and angiotensin II receptor antagonists may also promote reduction of sympathetic tone, although to a lesser extent. On the other hand, other compounds may either be neutral or may play unfavorable role on the sympathetic nervous system, such as long-acting calcium channel blockers, diuretics, and short-acting calcium channel blockers, respectively [
32‐
34]. Although it is not possible to differentiate whether autonomic dysfunction, as displayed by the stroke group, was a result of ischemic cerebral event, previous hypertension, or both, we suggest that the residual motor disability triggered by stroke may be the main reason, since blood pressure levels were controlled and within normal limits.
Regarding predictive values of heart rate variability, it is well established that alterations in these parameters may predispose individuals to arrhythmias and cardiac events [
35‐
37], being associated with several diseases, e.g., myocardial infarction [
38,
39] heart failure [
40], diabetes [
41] and stroke [
42,
43]. Moreover, heart rate variability has been found to be a predictor of stroke in subjects aged 55–70 years and without cardiovascular disease [
44].
In addition to autonomic modulation analysis, we observed that women with chronic stroke showed a decrease in HRR measured at 1, 2 and 3 min after the end of test when compared to controls. These results are indeed significant, since the decreased HRR at 1, 2 and 3 min after exercise is mainly a result of impaired vagal reactivation, a predictor of cardiovascular events [
45‐
47]. Furthermore, decreased vagal tone is found in several conditions and it is generally associated with poor cardiovascular prognosis [
48‐
50]. In fact, HRR after exercise seems to be correlated with heart rate variability in the early recovery phase after submaximal exercise [
51,
52], reinforcing our findings.
To the extent that the parasympathetic modulation is deteriorated, achieving an adequate heart rate during exercise has proved challenging, as well as returning to baseline values. Thus, the impairment of vagal function is detectable not only for heart chronotropic incompetence (an aspect which has not been covered in this study), but also for the recovery of heart rate immediately after exercise.
According to Mravec [
53], afferent and efferent vagal pathways may affect several mechanisms involved in the onset and progression of stroke. One of the mechanisms is the central and peripheral inflammation, which may either lead to stroke or be stroke-induced. Reduced vagal activity mediated by a decrease in cholinergic anti-inflammatory pathway may be accompanied by an increase in the pro-inflammatory status and may represent a risk factor for stroke. Yet, stroke may alter vagal immunoregulatory functions and lead to inflammatory reactions in the peripheral tissues and brain [
53].
Since the presence of residual motor disabilities may lead to a chronic condition of physical inactivity, and consequently to an exacerbated autonomic dysfunction, chronic stroke patients remain at high risk for cardiovascular events, including another stroke, and this should not be overlooked. In this sense, several experimental and clinical studies have demonstrated that exercise can improve cardiovascular autonomic function in stroke subjects within a short time after the event [
54‐
58]. In our study, we demonstrated that, although some parameters of heart rate variability, i.e., VarNN, pNN50 and HF, remained lower than those found for controls subjects, improvement was actually detected in all time and frequency domain parameters, as well as in the symbolic analysis in the stroke group at recovery after submaximal exercise test. These data suggest that aerobic exercises, if well conducted and carefully monitored, may be an effective non-pharmacological strategy to improve heart rate variability and cardiac vagal modulation in stroke women, even after a chronic period after the event.
This study has limitations that deserve comments. First, although we did not divide patients according to injury side, they had similar patterns of reduced parasympathetic modulation, both at baseline and in the recovery period. Second, the lack of a hypertensive control group when more than half of patients with stroke had a history of hypertension does not allow us to distinguish the participation of each of these diseases on autonomic dysfunction presented by the stroke group. However, this issue was not the focus of this study. Another limitation lies in the fact that our study population was undersized, i.e., it was not large enough (14 chronic stroke women and 10 controls) to enable us to draw conclusions suitable for extending and generalizing our inferences and results. In fact, the minimum sample size to promote a power of 80 % in paired analyses was in C group n = 12 and in S group n = 16. This calculation has been carefully considered by the G * Power 3.2.1 software. Finally, although we excluded women who were taking beta-blockers, the presence of comorbidities (as hypertension) and the medications use may had some influence on cardiovascular autonomic modulation of the evaluated individuals.
Competing interests
The authors declare that they have no competing interests.
Authors’ contributions
JVF: Data acquisition, analysis and interpretation; AB: Data acquisition, analysis and interpretation; LM: Data acquisition and statistical analyze; KBS: Data analysis and interpretation; OAM: Data analysis and interpretation; CM: Data analysis and interpretation; ECC: interpretation of data and helped to draft the manuscript; MCI: interpretation of data and helped to draft the manuscript; KDA: interpretation of data, design and helped to draft the manuscript; BR: conception and design of the work and draft the manuscript. All authors approved final version to be published.