The Severity of COVID-19 Is Associated with Greater Impairment of Cardiac Autonomic Modulation—Physical Training as a Countermeasure

Background: COVID-19 has been associated with persistent impairments in autonomic modulation of heart rate variability (HRV). However, whether disease severity during the acute phase influences the magnitude of these impairments remains insufficiently explored. In turn, aerobic physical training (APT) has been proposed as a countermeasure to autonomic dysfunction of HRV in different conditions, although its effects in individuals with COVID-19 are not yet well established. To address these gaps, this study investigated the consequences of COVID-19 on autonomic modulation of HRV according to disease severity and evaluated the effects of APT on this parameter. Methods: One hundred and sixteen individuals (58 men and 58 women) aged between 30 and 55 years were included, allocated into three groups according to the severity of the disease in the acute phase: Mild group (n = 38, mean age: 48 ± 7 years); Moderate group (n = 52, mean age: 43 ± 5 years); and Severe group (n = 26, mean age: 45 ± 6 years). All groups had anthropometric and hemodynamic parameters evaluated before and after the 16-week APT period, as well as parameters of autonomic modulation of HRV analyzed using linear (time and frequency domain) and non-linear (symbolic analysis) methods obtained from R–R interval (RRi) recordings in the supine position for 30 min. Results: Initially, all groups presented similar anthropometric and hemodynamic values. In contrast, the Moderate and Severe groups presented lower values for standard deviation of normal RRi (SDNN; Moderate: 38 ± 14 ms; Severe: 33 ± 12 ms vs. Mild: 55 ± 28 ms; p < 0.001), root mean square difference between adjacent normal RRi (RMSSD; Moderate: 28 ± 13 ms; Severe: 22 ± 7 ms vs. Mild: 47 ± 38 ms; p < 0.001), total variance (Moderate: 203 ± 127 ms2; Severe: 303 ± 157 ms2 vs. Mild: 526 ± 347 ms2; p < 0.001), and high-frequency (HF) oscillations in absolute units (Moderate: 259 ± 270 ms2; Severe: 153 ± 74 ms2 vs. Mild: 438 ± 421 ms2; p < 0.001), both compared to the Mild group. In turn, the Severe group, when compared to the other groups, also presented lower HF oscillations (Severe: 29 ± 12 nu vs. Mild: 44 ± 17 nu and Moderate: 42 ± 17 nu; p < 0.001) and higher low-frequency (LF) oscillations (Severe: 71 ± 12 nu vs. Mild: 60 ± 17 nu and Moderate: 58 ± 17 nu; p < 0.001), but in normalized units. After the 16-week APT, all groups showed increases in HF oscillations (Mild: −206 ms2 and −19.12 nu; Moderate: −236 ms2 and −26.7 nu; Severe: −211 ms2 and −31.0 nu; p < 0.001) and reductions in LF oscillations (Mild: 198 ms2 and 19.01 nu; Moderate: 98 ms2 and 26.7 nu; Severe: 218 ms2 and 31.1 nu; p < 0.001), both in absolute and normalized units. In this case, there were no further differences in LF and HF oscillations between the groups. Conclusions: Individuals who had COVID-19 and developed moderate to severe cases showed greater impairments in the autonomic modulation of HRV, characterized by increased sympathetic autonomic modulation and reduced vagal modulation. In turn, APT as a countermeasure appears to increase vagal autonomic modulation and reduce sympathetic autonomic modulation of HRV, regardless of the previous severity of COVID-19.