Work

Impact of post-acute COVID-19 exercise training on cardiovascular autonomic function in amateur runners: A self-controlled longitudinal study

Year: 2024

Type: article

Abstract: To the Editor: The post-acute sequelae of coronavirus disease 2019 (COVID-19) (PASC) manifest broadly and persist over time. Cardiovascular autonomic dysfunction (CVAD) is implicated in some characteristic PASC symptoms, although the underlying mechanisms are not fully elucidated.[1] While certain studies suggest that targeted exercise training ameliorates PASC symptoms, advocating its adoption in rehabilitation strategies, others indicate potential exacerbation or relapse of symptoms post-exercise, highlighting a poorly understood interaction between exercise training and the autonomic nervous system during the post-acute phase of COVID-19.[2] This study, therefore, aimed to investigate the effects of exercise training post-acute COVID-19 on cardiovascular autonomic function. This self-controlled, interventional longitudinal study (Clinical Trial Registration, No. ChiCTR2300070374, https://www.chictr.org.cn/showprojEN.html?proj=189798) received approval from the Ethics Committee of West China Hospital, Sichuan University (No. 2023-236). Written informed consent was obtained from all participants. Inclusion criteria were (1) individuals aged 18–60 years with a regular running habit, selected to ensure the study’s safety and successful implementation, (2) ability to complete a 10-km run and medically assessed as having no exercise intolerance, (3) possession of a wearable device for heart rate monitoring, and exclusion criteria were (1) any known cardiac disease history, (2) recovery from SARS-CoV-2 infection (free of all symptoms) less than 14 days prior, and (3) inability to adhere to the study protocol. Study termination was prompted by participant withdrawal, reinfection, or occurrence of other diseases or accidents. The study protocol involved (1) recruiting amateur runners through social media; (2) collecting baseline data including demographic information, 24-Hour Holter monitoring, and three autonomic nervous function tests (Deep Breathing Test, Valsalva Maneuver, and Orthostatic Test); (3) conducting a 10-km run test; (4) random assignment into either the Performance Improvement Group or Safety First Group using a paired randomization approach that ensured there were no significant differences in baseline characteristics between the two groups; (5) devising a weekly exercise prescription by a fitness coach based on the 10-km run test results and group assignment, focusing on improving running performance through a blend of various running techniques and training modalities; (6) participants recording their exercise data using wearable devices, with weekly reporting for subsequent prescription adjustments; and (7) post-six-week training program, re-assessment using 24-Hour Holter monitoring and autonomic function tests [Supplementary Figure 1, https://links.lww.com/CM9/C103]. Statistical analysis included (1) presenting categorical variables as n (%), using Fisher’s exact test for correlations between two independent binary variables, McNemar’s test for differences between paired binary variables, and McNemar–Bowker test for paired multicategorical variables; (2) assessing continuous variables for normality with the Shapiro–Wilk test, expressing normally distributed data as mean ± standard deviation and non-normally distributed data as median (Q1–Q3); comparisons between groups using independent samples t-tests or Wilcoxon rank-sum tests, and within-group pre- and post-intervention comparisons utilized paired samples t-tests or Wilcoxon signed-rank tests; (3) displaying Holter monitoring indicators like heart rate (HR), heart rate variability (HRV), and others in tables with definitions in Supplementary Table 1, https://links.lww.com/CM9/C103; (4) subgroup analyses by baseline characteristics (exercise load, gender, age, body mass index [BMI], days after last infection [DALI], and total COVID-19 antibody levels), with the latter four variables grouped based on the median of their baseline levels; (5) multiple linear regressions corrected for age and gender were used to show the relationship between other baseline characteristics and the change in values of the measured indicators; and (6) using SPSS version 27.10 (IBM Corp., Armonk, NY, USA) for data analysis, with a two-sided P-value of <0.05 considered statistically significant. During the study period from March to April 2023, 42 amateur runners were recruited and they completed all phases of the study without dropouts [Supplementary Figure 1, https://links.lww.com/CM9/C103]. The cohort consisted of 25 males (59.5%) and 17 females (40.5%), with an average age of 39 ± 7 years, average BMI of 23.33 ± 2.26 kg/m2, and median running duration of 72 (53–96) months. Thirty-seven participants had a prior SARS-CoV-2 infection, averaging 88.54 ± 5.25 days post-infection [Supplementary Table 2, https://links.lww.com/CM9/C103]. Baseline comparisons between infected (n = 37) and uninfected participants (n = 5) showed no significant differences in age, gender, or BMI. However, infected participants exhibited shorter running durations (72 [49–90] vs. 120 [96–125] months, P = 0.012) and higher total SARS-CoV-2 antibody levels (1478.74 ± 222.93 vs. 670.78 ± 809.51 cut-off Index [COI], P <0.001) [Supplementary Table 2, https://links.lww.com/CM9/C103]. Baseline Holter monitoring indicated higher Spectral Power-24 Hour (4222.80 [3219.10–5787.00] vs. 2871.40 [2456.60–3344.60] ms2, P = 0.027) and Max QT (421.70 ± 19.21 vs. 396.40 ± 32.45 ms, P = 0.015) in the infected group (P <0.05). Post exercise training, significant reductions in Minimum HR (41.08 ± 4.04 vs. 44.65 ± 4.89 beats per minute [bpm], P <0.001), Maximum HR (126.51 ± 16.72 vs. 134.32 ± 17.46 bpm, P = 0.021), and Average HR (65.27 ± 6.44 vs. 69.05 ± 7.26 bpm, P <0.001) were observed among the previously infected, along with increases in the percentage of successive normal-to-normal intervals that differ by more than 50 ms (the percent-age of successive normal-to-normal intervals that differ by more than 50 ms [pNN50], 18.00 [10.00–25.00] vs. 14.00 [8.00–20.00], P = 0.017), Pauses in Excess of 1.5 s (2.00 [0–53.00] vs. 0 [0–1.00], P = 0.001), and Max Pause (1.54 [0–1.74] vs. 0 [0–1.52] second, P <0.001). Uninfected participants showed a significant post-training increase in Average HR (78.60 ± 7.54 vs. 71.60 ± 9.40 bpm, P = 0.027). The changes in these cardiovascular indicators before and after the exercise intervention displayed a contrasting trend between the infected and uninfected groups, with the former showing a decrease in HR (Minimum HR, Average HR) and an increase in HRV (the standard deviation of normal-to-normal QRS intervals over 24 h [SDNN24], pNN50) and bradycardia indicators (Pauses in Excess of 1.5 s, Max Pause). These differential changes were statistically significant (P <0.05) [Supplementary Table 2, https://links.lww.com/CM9/C103]. Despite varying infection statuses and training phases, abnormalities in cardiac rhythm-related indices, ST segment, and T wave were rare and hence not included in the subgroup analyses. Subgroup analyses of participants who had been infected with COVID-19 showed that, in terms of absolute changes in indicators after exercise, men exhibited a smaller reduction in Maximum HR and a more pronounced increase in the SDNN Index compared to women (P <0.05). Younger participants showed smaller reductions in Maximum HR compared to older participants (P <0.05). Those with shorter DALI exhibited increases in SDNN24 and Deep Breathing Test indices, with opposite trends observed in those with a longer DALI (P <0.05). No significant differences were observed among subgroups based on BMI, total SARS-CoV-2 antibody, and exercise load [Supplementary Tables 3–8, https://links.lww.com/CM9/C103]. Multivariable linear regression analysis showed consistent results [Supplementary Table 9, https://links.lww.com/CM9/C103]. In our study, COVID-19-infected individuals exhibited a lower baseline HR and higher HRV compared to those uninfected, with only the Spectral Power-24 Hour showing statistical significance (P <0.05). This suggests an increased parasympathetic nerve tension, contrasting with the typically observed rise in sympathetic nerve tension noted in prevalent studies on PASC.[1] This anomaly may stem from the distinct demographics of our study population, which consisted primarily of asymptomatic post-acute COVID-19 individuals who successfully completed a 10 km run test, thereby excluding those with symptoms like orthostatic or exercise intolerance, which are often associated with enhanced sympathetic nerve tension. Further, various studies have highlighted multiple manifestations of autonomic dysfunction post-COVID-19 infection, including orthostatic hypotension, abnormal sinus bradycardia, and chronotropic incompetence, underlining the complex effects of COVID-19 on autonomic function.[3] Notably, our findings from 24-Hour Holter monitoring align with those observed in pediatric long COVID patients, who also demonstrated predominant parasympathetic nerve tension, suggesting potential parallels between our adult study population and pediatric cases.[4] Our research is pioneering in employing detailed 24-Hour Holter monitoring to assess the short- and medium-term impacts of exercise training on the cardiovascular autonomic function in individuals recovering from post-acute COVID-19. The outcomes indicate that exercise training in infected participants leads to a decreased heart rate and enhanced HRV and bradycardia indicators, further elevating parasympathetic nerve tension. This response is distinctly opposite to that observed in uninfected individuals and appears largely independent of variables such as gender, age, BMI, DALI, total COVID-19 antibody, or exercise load. Limitations of this study include its small sample size, imbalance in the number of infected versus uninfected participants, the specific recruitment of amateur runners who had recovered from COVID-19 without experiencing current symptoms, aimed at minimizing health risks during the exercise protocol and ensuring completion, and absence of longer-term evaluation. These factors may affect the reliability of the results and the generalizability of the findings. In summary, this study found that for asymptomatic amateur runners post-acute COVID-19, engaging in 6 weeks exercise training led to decreased heart rate, increased heart rate variability, and heightened risk of bradycardia. Compared to those uninfected, there was a further abnormal increase in parasympathetic nerve tension among the infected, which appeared to be unaffected by gender, age, BMI, DALI, total COVID-19 antibody, or exercise load. Given the need for caution and close monitoring of cardiovascular autonomic function during exercise training after the acute phase of COVID-19, the mechanisms of CVAD post-acute COVID-19 and its response to exercise training require further exploration. Funding This work was supported by National Clinical Research Center for Geriatrics, West China Hospital, Sichuan University (No. Z2023LC004), and Sichuan Science and Technology Bureau-Key Research & Develop Support Program (No. 22GJHZ0044). Conflicts of interest None.

Source: Chinese Medical Journal

DOI: 10.1097/cm9.0000000000003251