Heart failure (HF) is a complex syndrome which may be the ultimate consequence of most cardiovascular diseases. Contractility reduction is one of the core features of HF with reduced ejection fraction (HFrEF), in which cardiac output (CO) impairment results in systemic hypoperfusion. This, combined with pulmonary, peripheral, and neurohumoral changes, contributes to a low tolerance to physical activity. ^{1 , 2} This reduced capacity to perform physical activities is one of the hallmarks of this disease, common to most individuals with HFrEF. ³

The six-minute walk test (6MWT) is a widely used method to evaluate acute responses to self-limited exercise, in which the walked distance constitutes a proven prognostic marker. ⁴ Furthermore, it is a simple and inexpensive test, easy to perform and does not require specialized training. ⁵ However, studies have shown that certain comorbidities, such as HFrEF, significantly reduce the distance walked (depending on cardiac performance, i.e ., disease severity), requiring further assessment using other tools for better clinical and prognostic information. ⁶

Signal-morphology impedance cardiography (SM-ICG) is a noninvasive method which accurately measures CO, stroke volume (SV), heart rate (HR), and cardiac index (CI). Clinical researchers can use it to assess both healthy individuals and those with conditions such as HFrEF during 6MWT, adding different information about health conditions. ^{7 - 10} The literature has shown that a commercially available SM-ICG device (PhysioFlow ^® ) can accurately assess peak CO (when compared to Fick’s and thermodilution methods) both at rest and during exercise. ^{11 - 13} Furthermore, SM-ICG can add predictive information on peak oxygen consumption (VO ₂ peak) with a strong correlation between measured and predicted VO ₂ peak (r = 0.931, p<0.001) by values obtained during 6MWT for CO, SV, and HR. ¹⁴

SM-ICG provides useful hemodynamic data during 6MWT in different clinical conditions. ¹⁵ However, CO, HR, and SV acceleration and deceleration responses to 6MWT are yet to be shown in HFrEF individuals; these variables may represent autonomic imbalances in exertion and recovery. ^{7 , 11} Therefore, this study primarily aimed to evaluate CO, HR, and SV acceleration and deceleration responses to 6MWT in patients with HFrEF and in healthy controls. Our secondary objective was to assess hemodynamic behavior (through CO, HR, SV, and CI) before, during, and after 6MWT, which is still unavailable in the literature.

Data were shown in mean and standard deviation (SD) or median and interquartile range (IQR), according to the normality test. Categorical variables were presented as frequency (absolute and relative) and the chi square test was used to assess the differences between groups for these variables. Data distribution was assessed by the Shapiro-Wilk test. For between-group comparisons, the independent t -test or the Mann-Whitney U test was used as appropriate. Pearson correlation was applied to assess association strength between variables. Multivariate linear regression, with CO changes as the dependent variable, was used to identify the contribution of HR and SV changes. All the five assumptions necessary for the use of linear regression analysis were verified (linear relationship; multivariate normality; no or little multicollinearity; no autocorrelation; homoscedasticity).

Alpha was defined as <0.05 to indicate statistical significance. Statistical analyses were performed in SPSS, version 20.0 (IBM; ARMONK, NY, USA).

Changes to HR explained 64.3% of the ΔCO in the control group and 70.3% in patients with HFrEF. The contribution to ΔSV totaled 57.4% in the control group and only 22.9% in individuals with HFrEF. According to linear regression β coefficients, for each altered HR unit, ΔCO changed by 1.121 units in the control group and by 0.92 unit in the HFrEF group. Regarding ΔSV, each altered HR unit, ΔSV changed by 1.162 units in the control group and to 0.91 unit in the HFrEF group.

In the present study, we compared cardiodynamicW responses to 6MWT between HFrEF patients and healthy individuals. Our main finding is that HFrEF showed blunted hemodynamic responses to 6MWT, especially CO and HR acceleration when compared to controls. Additionally, we found significant differences in the distance walked during the 6MWT between the groups, reaffirming the expected reduction in the functional capacity of individuals with HFrEF. Furthermore, during recovery, HR response was lower in the HFrEF group, suggesting an impairment of the parasympathetic autonomic nervous system, which actually occurs in these patients ( Central illustration ).

Autonomic imbalance, characterized by sympathetic predominance, is a classical feature of HFrEF, with clinically relevant consequences. These include disease progression, development or deterioration of exercise intolerance, ventricular remodeling and arrhythmias, and premature death. The mechanisms underlying these processes and their relative time courses still need to be described.

To our knowledge, several studies have evaluated the hemodynamic profile of different diseases during 6MWT. ^{15 , 21 - 23} One study ²² assessed CO (but not SV) acceleration and deceleration in pulmonary hypertension. These variables are important for HFrEF, as they represent an imbalance in autonomic responses to exertion and recovery. ^{24 , 25} We found a significant difference in CO acceleration (p<0.01) but not in deceleration (p=0.07) between groups, and a similar behavior in HR acceleration (p=0.039) and deceleration (p=0.385). Both CO and HR acceleration and deceleration were lower in the HFrEF group compared with controls; these patients show a chronotropic deficit due to the disease per se and because of the pharmacological effect of beta-blockers (all patients were receiving in beta-blocker therapy).

Since acceleration and deceleration may represent sympathetic and parasympathetic activation, respectively, HFrEF shows greater sympathetic activation and sympathovagal imbalance, which corroborates the findings of our study, showing lower CO and HR acceleration in the HFrEF group. ^{26 , 27} Animal models have shown that sympathoexcitation and abnormal cardiovascular reflex function contribute to activate the sympathetic nervous system in HFrEF. ²⁸ In contrast, little is known about the role of parasympathetic nerve activity in HFrEF. Hu et al. ²⁶ showed that HR deceleration is an independent predictor of acute myocardial infarction and sudden cardiac death in HFrEF, constituting a stronger predictor than left ventricular ejection fraction and conventional HR variability measurements. Hu et al., ²⁶ only evaluated HR acceleration and deceleration responses. Therefore, our study is the first to assess CO, HR, and SV acceleration and deceleration in HFrEF. Although this study failed to demonstrate a significant difference in CO deceleration between the groups, we found a clear tendency (P=0.07) of loss in the responses in the HFrEF group.

Regarding HR regulation, ventricular impairment in HFrEF can trigger distinct compensatory mechanisms, which initially increases neurohormonal activation of the sympathetic nervous and renin-angiotensin-aldosterone systems. ²⁹ However, prolonged exposure to sympathetic activation may downregulate β-adrenergic receptors in cardiomyocytes, contributing to inotropic responses which can impair HR recovery after exercise. ³⁰ These data are corroborated by our results. ²

We found lower baseline and peak SV values in HFrEF patients, when compared to controls. Both resting and exercising behavior in the HFrEF group represents a loss in ventricular contraction, i.e ., a reduction in SV in each systole. ³¹ In healthy individuals, the Frank–Starling principle is a physiological mechanism which increases SV to compensate for the initial reduction of ventricular contraction. ³² On the other hand, individuals with HFrEF display failures in this mechanism. The attendant reduction in cardiovascular reserve impairs and reduces ventricular contractility, decreasing SV. ³³ This phenomenon also corroborates the findings of this study.

In addition, we observed significant differences in baseline and peak CI and CO values between groups, which may be explained by HR and SV compensatory mechanisms. ²¹ We found that ventricular impairment reduced CI and CO in individuals with HFrEF, leading to mechanisms that initially increase these variables to maintain end-organ perfusion. ³⁴ However, after prolonged exposure to them, the myocardium undergoes remodeling, reducing ventricular inotropic capacity and SV, consequently affecting both CI and CO. ²

When subjected to submaximal exercise, individuals with HFrEF show lower responses to both increased and decreased VO ₂ than healthy individuals. ³⁵ These patients may also show mild pulmonary hypertension, which may explain CO and SV behavior. ^{22 , 36} CO behavior may thus be due to imbalanced autonomic responses affecting SV. ³⁷

We tested whether hemodynamic parameters, assessed via SM-IGC, correlated with the distance walked during 6MWT, and found that the greater the CO acceleration, the greater the distance walked (r=0.49, p=0.01). Moreover, individuals with the fastest HR recovery after the test were those who walked the longest distances. Even though this was not the main aim of the study, we found that SM-ICG hemodynamic assessment during 6MWT can provide interesting results that directly reflect functional capacity. In addition to the 6MWT distance, we found that NYHA functional class had a good association with maximum CO, CO deceleration, and HR during the first minute of recovery after the test. ³⁸ As expected, we found different 6MWT distances between groups (p<0.01), corroborating previous research. ³⁹

Finally, linear regression showed the impaired contribution of SV (22.9%) to CO changes in patients with HFrEFm and normal values in healthy controls (57.4% of SV contribution to CO change). In fact, oxygen pulse, a SV surrogate, was lower in patients with HFrEF (compared to healthy controls), as per previous studies. SV impairment, as represented by low oxygen pulse, may reflect insufficient systemic oxygen delivery during exercising and/or reflect impaired oxygen utilization due to reduced mitochondrial function. ^{40 , 41}

Some limitations of this study must be addressed. First, the small sample size may have limited the ability to detect significant differences. However, the main aim of this physiological study was to evaluate the hemodynamic behavior of individuals with HFrEF during 6MWT, compare it to that of healthy individuals, and determine the relative contribution of CO, SV, and HR variations during the phases of the walking test. Secondly, our study was not designed to perform correlation or association tests of SM-ICG parameters with walked distance, NYHA functional class, and ejection fraction.

Although 6MWT is safe, inexpensive, and easily used to assess functional capacity in patients with HFrEF, it has both clinical and research limitations ^{20 , 39} (e.g., it fails to provide direct hemodynamic behavior data). Therefore, new technologies would be important to add information to 6MWT findings and help in the treatment of this ominous syndrome. Moreover, technological advances have allowed the development of a portable device to measure, noninvasively and in real time, a wide range of hemodynamic parameters, such as CO, SV, HR, and CI.

This is the first study to show CO, HR, and SV acceleration and deceleration hemodynamic responses via SM-ICG in patients with HFrEF during 6MWT. Individuals with HFrEF showed impaired CO and HR acceleration during submaximal exercises compared to healthy controls, which may represent an imbalance in their autonomic response to exertion. Further studies are needed to test whether changes in CO, HR, SV, and CI during 6MWT can provide information on disease prognosis.