What this is
- This randomized clinical trial compared (STR) with (ATR) in adults with ().
- Participants (n=31) engaged in either supervised or unsupervised exercise sessions over eight weeks.
- The primary outcome was lower limb functional strength, assessed through the Five Times Sit-to-Stand Functional Test.
Essence
- led to greater improvements in lower limb functionality, dyspnea, fatigue, stress, and quality of life compared to in adults with .
Key takeaways
- Both groups reported improvements in fatigue, but only the synchronous group showed significant enhancements in dyspnea (p = 0.02), stress (p = 0.03), and quality of life (p = 0.00).
- Higher dropout rates were noted in the ATR group, indicating potential challenges in treatment adherence.
Caveats
- Additionally, the asynchronous group's lack of supervision may have impacted adherence and outcomes.
- Remote assessments posed challenges, and the absence of personalized training prescriptions reduced the robustness of the findings.
Definitions
- Synchronous Telerehabilitation: Real-time, supervised exercise sessions conducted via video conferencing.
- Asynchronous Telerehabilitation: Unsupervised exercise sessions guided by pre-recorded videos without real-time interaction.
- Post-COVID Syndrome (PCS): Persistent symptoms following COVID-19 infection, including fatigue, dyspnea, and joint pain.
Simplified
Methods
Study Design
This is a single-blind, parallel-group randomized clinical trial consisting of an exercise program offered by supervised synchronous and unsupervised asynchronous telerehabilitation associated with telehealth guidance, which lasted eight weeks, with two sessions per week, totaling 16 sessions included three evaluation sessions (baseline, week 8, and week 20 at follow up). This study was reported following the CONSORT-Outcomes 2022 extension of the 2010 Statement, the primary outcome was the Five Times Sit-to-Stand Functional Test (FTSST), while the secondary outcomes were the Modified Medical Research Council (mMRC) scale, fatigue assessment scale (FAS), Depression, Anxiety, and Stress Scale-21 (DASS-21) and the World Health Organization Quality of Life-BREF (WHOQOL-Bref).
Recruitment and Eligibility Criteria
Individuals with persistent COVID-19 symptoms were consecutively screened and recruited through various sources, such as social networks, the local community, and different regions of Brazil. Recruitment methods included distributing city flyers, digital flyers on social media, and radio announcements. Interested individuals completed an online form providing identification, demographics, and selection information, followed by signing the informed consent form. This study was reported following the CONSORT-Outcomes 2022 extension of the 2010 Statement, approved by the Human Research Ethics Committee of the Federal University of Santa Catarina (no. 50003221.1.0000.0121) and registered at(no. RBR-8v2qmyz). Data collection was conducted entirely through an online form, without direct interaction between the researcher and participants, due to the prevailing health context. During this period, social distancing measures and restrictions imposed by the COVID-19 pandemic made in-person procedures unfeasible, making a remote approach the safest and most ethically appropriate methodological alternative. In this way, the online strategy ensured the continuity of the research, preserved the safety of those involved, and complied with the recommendations of the health authorities at the time. Inclusion criteria included being 18 or older, self-reported presence of at least one persistent COVID-19 symptom (e.g., fatigue, dyspnea, joint pain, or myalgia) and scoring 1-2-3 on the mMRC scale (;;). Individuals were excluded if they recovered from COVID-19 without lingering symptoms, scores of 4–5 on the mMRC scale, absence from four or more telerehabilitation sessions and/or two consecutive sessions, and the inability to perform the proposed activities. ensaiosclinicos.gov.br [Bestall et al., 1999] [Casanova et al., 2015] [Kovelis et al., 2008]
Randomization, Blinding and Interventions
Participants were randomly assigned to either synchronous telerehabilitation (STR) or asynchronous telerehabilitation (ATR). An independent researcher created a randomized allocation list () using 4 blocks of 12, 16, 11, and 5 participants each. A code assigned to each participant concealed their group assignment. The evaluator, consistent throughout the study, remained blinded to group allocation. The physiotherapist, however, was not blinded to group assignments. www.randomizer.org↗
Both groups received COVID-19 health education, covering prevention, disease management, and PCS implications. Participants refrained from other physical activities during the study. The exercise regimen included warm-up, aerobic, limb strengthening, joint mobility, chest stretching, and breathing exercises. The STR group exercised via video call with a physiotherapist, while the ATR group used only video instructions. Participant blinding was maintained, and they received no financial incentives. Group assignment was revealed after the study, with ATR participants having the option to perform the STR exercise program.
Synchronous Telerehabilitation
The STR group engaged in an eight-week progressive exercise program comprising two sessions per week. The duration varied according to each patient, averaging 50 minutes. These sessions were remotely supervised by a physiotherapist via video conferencing, providing performance oversight, addressing questions, and helping participants with difficulties. The exercise program was divided into multiple elements (see Supplementary Material): a one-minute warm-up involving aerobic exercises, progressive lower and upper body strengthening exercises, trunk exercises, and chest stretches combined with breathing exercises. The warm-up phase lasted one minute, during which the Borg Rating of Perceived Exertion Scale was administered. A threshold value of 6 was set to maintain moderate activity levels (). In the initial week, participants completed three sets of eight repetitions for each strengthening exercise, using their body weight, a 0.5-kg water bottle, or a 1-kg water bottle as resistance, with one-minute rest intervals. Repetitions increased every two weeks, reaching 16 by the protocol's end (;). Participants were advised to avoid any other physical activities during the study. [Helms et al., 2020] [Aily et al., 2020] [Gentil et al., 2021]
Asynchronous Telerehabilitation
The ATR group performed the same exercise protocol as the synchronous group, but without any direct supervision from the physiotherapist. For this group, pre-recorded instructional videos were provided, containing planned predictions for each exercise, lasting approximately 30 minutes per session, sent via WhatsApp and an unlisted YouTube channel. These videos included general guidelines on execution, pace, and progression, but did not allow for real-time interaction. Participants were instructed to follow the protocol twice a week for eight weeks, without the possibility of asking questions, requesting adjustments, or obtaining individualized feedback after the sessions, as there was no active communication channel with the researchers.
Furthermore, it was not possible to directly document or monitor adherence to the sessions in the ATR group, which represents a significant methodological limitation, especially when compared to the synchronous group, which had access to real-time supervision and the opportunity to request help or clarification during practice. This difference in the possibility of interaction and support may have introduced bias in access to the intervention, potentially influencing the execution of the exercises and, consequently, the results observed between the groups.
Evaluations and Outcomes
Participants from both groups underwent identical evaluations at three distinct time points: baseline, after the 8-week exercise program, and at a 20-week follow-up. The principal outcome measure was the time, in seconds, required to complete the FTSTS. This test was administered remotely via digital platforms last, after the other assessments had been completed (). The required setting included a 3 or 4 marea of flat, non-slippery ground and a stable chair that allowed the participant's feet to touch the floor while seated (). Fatigue perception was assessed using the FAS, which is commonly employed in research involving cardiorespiratory-related diseases (Michielsen et al., 2014;). Neuropsychiatric symptoms were evaluated through the DASS-21, divided into stress, anxiety, and depression subscales (;;) Quality of life was assessed by WHOQOL-Bref instrument and evaluated in a 0–100 score (;). Initially, on the RCT register, there was another scale under assessment, the Patient-Specific Functional Scale, which had to be excluded from data collection due to incorrect utilization and missing data. [López-López et al., 2022] [Araújo et al., 1999] [Moore et al., 2013] [Imamura et al., 2021] [Lovibond & Loovibond, 1995] [Vignola & Tucci, 2014] [Fleck et al., 2000] [Pedroso et al., 2010] 2
Sample Size
The sample size was estimated for the primary outcome, functional strength of the lower limbs as assessed by the FTSST. The calculation was made using the G*Power version 3.1 software, the calculation was done for a repeated measures analysis of variance, accounting for both intra- and inter-subject interactions, with an F effect size of 0.25, α of 0.05, and power (1-β) of 0.80, for two groups with three measurements each. This yielded a sample size (n) of 28 subjects, allocating 14 volunteers to each group. To account for a potential 20% attrition rate, two additional volunteers were included in each group, totaling 32 participants.
Statistical Analysis
Intention-to-treat (missing data handled using the average imputation method) was performed for all outcomes. Data normality was tested using the Shapiro-Wilk () and the efficacy of STR versus ATR over multiple assessment time points and any changes in scores were evaluated using a mixed model effects, followed Tukey pos-hoc test. The effect size was calculated using the eta-squared test for fixed variables, which yielded a Cohen 's d value used to compare the proportion of participants in each group with a clinically important improvement. The level of significance was P<.05. All data analyses were conducted using GraphPad Prism 8.0.1 software. [Shapiro; Wilk, 1965]
Results
Between March 2021 and March 2023, we successfully randomized 44 consecutive participants, with 23 assigned to the STR group and 21 to the ATR group. However, after the final twenty-week period, attrition occurred, resulting in the loss of 8 participants from the STR group and 13 from the ATR group during follow-up. Given the missing data at follow-up, we conducted an intention-to-treat analysis, which yielded a final participant count of 17 in the STR group and 14 in the ATR group. Two participants needed triceps exercise adjustments in the first 4 weeks in the STR group, highlighting a need for protocol refinements in future studies. A flowchart detailing the study 's participant progression is illustrated inand, at baseline, demographic and clinical characteristics were similar for all the groups (). Figure 1 Table 1
Primary Outcome
Lower Limb Functionality
A significant between-group difference (F = 5.936; p = 0.0269) was found in the FTSST by mixed-model effects, but in the post hoc, this difference was not found. The effect size of STR on FTSST was of medium magnitude (d = 0.51) between groups. In the intra-group analysis, a significant difference was also found (F = 5.606; p = 0.01) by the mixed models effects. The post hoc showed a significant difference between the initial assessment vs. eight weeks of intervention for the STR group (95% CI = 0.1004–2.396; p = 0.0291) and the initial assessment vs. 20 weeks of intervention for the STR group (95% CI = 0.06205–2.899; p = 0.0386). The effect size of STR on FTSST was of medium magnitude (d = 0.51) (and) within groups. Table 2 Figure 2 3
Secondary Outcomes
Perception of Dyspnea
There was no significant difference between groups, only intra-group (F = 6.192;0.0123) on the mMRC scale. In the post hoc test, the difference was localized between the initial assessment vs. eight weeks of intervention in the STR group (95% CI = 0.1066–1.658;0.0212) (). The effect size of the intervention was of medium magnitude (d = 0.54) within groups. p = p = Figure 2
Perception of Fatigue
There was no significant difference between groups but a significant intra-group difference (F = 21.01;< 0.0001) in the FAS. In the post hoc analysis, the difference was located between the initial assessment vs. eight weeks of intervention for the STR group (95% CI = 4.197 to 14.27;0.0003); initial assessment vs. 20 weeks of intervention for the STR group (95% CI = 2.504 to 18.20;0.0067); and initial assessment vs. 20 weeks of intervention for the ATR group (95% CI = 2.519 to 29.47;0.0173) (). The effect size of the intervention on FAS was of large magnitude (d = 1.06) within groups. p p = p = p = Figure 2
Self-reported Stress, Anxiety, and Depression
There was no significant difference between groups, but a significant intra-group difference was found in the stress (F = 9.944;0.0006), anxiety (F = 9.608;0.0015), and depression (F = 7.693;0.0028) subscales of the DASS-21 scale. In the post hoc analysis, the difference was found to be between the initial assessment vs. eight weeks of intervention for the STR group (95% CI = 0.03745–12.90;0.0483); initial assessment vs. 20 weeks of intervention of the STR group (95% CI = 0.4912–15.74;0.0336) in the stress subscale; and between initial assessment vs. 20 weeks of intervention of the ATR group (95% CI = 0.5512–10.70;0.0270) in the anxiety subscale. No significant difference was found in the depression subscale (). The effect size of the DASS-21 intervention for the stress subscale was of medium magnitude (d = 0.71) within groups; for the anxiety subscale, it was of medium magnitude (d = 0.70) within groups; and for the depression subscale, it was of medium magnitude (d = 0.62) within groups. p = p = p = p = p = p = Figure 3
Quality of Life
There was no significant difference between groups. A significant intra-group difference was found (F = 10.49; p = 0.0009) in the WHOQOL-Bref questionnaire. In the post hoc analysis, the difference was located between the initial assessment vs. eight weeks of intervention for the STR group (95% CI = −18.96–−2.664; p = 0.0063) and initial assessment vs. 20 weeks of intervention for the STR group (95% CI = −15.29–−2.792; p = 0.0034). The effect size of the intervention was of medium magnitude (d = 0.73) within groups.
Adverse Effects
At the study's outset, the mMRC scale was used to ensure exercise safety for PCS-affected individuals. Predetermined adverse events included hospitalization; participants reported events by text. No exercise-related issues occurred.
Discussion
This study provides evidence that synchronous, physiotherapist-supervised telerehabilitation—comprising aerobic training, muscle strengthening, stretching, joint mobility, and respiratory training—offers more positive clinical outcomes than asynchronous telerehabilitation for individuals with post-COVID-19 syndrome in low- and middle-income countries. Notably, participants in the asynchronous group faced challenges in actively engaging in their own care. The synchronous approach significantly improved lower limb functionality, perceived dyspnea and fatigue, stress levels, and overall quality of life. In contrast, the asynchronous approach only yielded improvements in perceived fatigue and anxiety. Consequently, the effect size for the synchronous model ranged from medium to large for most outcomes within groups, with the only significant difference between the groups being lower limb functionality.
The field of TR for PCS is relatively nascent. Most existing studies have compared TR with traditional face-to-face or no therapy (). Our findings align with a randomized clinical trial bywhich showed that while both modes are effective, synchronous TR has greater positive impacts on physical functionality. This suggests that the TR mode should align with the target population 's capability to actively participate in their own care. Clinically, our results underscore the importance of physiotherapeutic oversight in treating PCS via TR. [Martínez-Pozas et al., 2024] [Tahan et al. (2023)]
Both groups in this study experienced reduced fatigue, as measured by the FAS—a symptom reported by 90% of the participants (). Few studies have quantitatively assessed this pervasive symptom in PCS. For instance, a trial bydid not measure the intervention's impact on fatigue, and the meta-analysis byindicated that unresolved fatigue is linked to functional impairment and can persist for an extended period. A study bydemonstrated that a moderate aerobic and resistance training protocol improved muscular capacity in elderly males with post-COVID-19 sarcopenia compared to a high-intensity approach. Therefore, telerehabilitation focusing on functional exercises and moderate aerobic activity can effectively mitigate fatigue and should be promoted, as the population more readily accepts it than strenuous exercise regimes. [Moore et al., 2013] [del Corral et al. (2023)] [Ceban et al. (2022)] [Nambi et al. (2022)]
The time needed to complete the FTSTS decreased solely within the synchronous group, which contrasts with a recent Brazilian clinical trial employing an asynchronous telemonitored post-COVID-19 home exercise program (). In the latter study, participants received exercise demonstrations and self-guided instructions, sustaining improvements over a 12-week period. This implies that asynchronous formats may be effective in longer-term applications. Nonetheless, for optimizing lower limb functionality, resistance training should be tailored to individual needs, even within TR contexts. [Teixeira et al., 2022]
As illustrated in, the attrition rate underscores the challenges associated with implementing a TR program in low- and middle-income countries. This challenge was also identified as a limitation in a randomized clinical trial bywhere participants in the asynchronous group exhibited reduced engagement and failed to achieve significant improvements in both primary and most secondary outcomes.emphasized the hurdles in introducing TR in the Philippines, a country with a socioeconomic profile similar to that of Brazil, attributing the lack of enthusiasm to the scarcity of relevant data. In contrast, a randomized clinical trial in China demonstrated improvements in functional capacity and lower limb strength using an asynchronous telerehabilitation model delivered via a mobile application, () suggesting that certain countries can effectively harness technology for asynchronous treatment and self-efficacy. Figure 1 [Teixeira et al (2022)] [Leochico et al. (2020)] [Li et al., 2022]
Despite the encouraging results, this study presents limitations that should be considered in the interpretation of the findings. Initially, the absence of personalized physical training prescriptions adapted to the individual capabilities of the participants stands out, as well as the lack of assessment of the participants' satisfaction with the exercise programs. In addition, the absence of a control group to monitor the natural evolution of post-COVID syndrome limits the conclusiveness of the inferences; however, due to the public health emergency context, ethical considerations justified offering intervention to all groups. Furthermore, remote assessments, given the restrictions imposed by the COVID-19 pandemic, present other limitations, even though they were a viable methodological alternative during that period.
Specifically regarding the ATR group, its methodology has inherent weaknesses, since it was not possible to control variables such as type, intensity, frequency, and adherence to the activities performed. The non-standardized nature of this exposure may have increased, stimulated the variability of stages, and reduced the robustness of comparisons with the STR group, whose intervention was structured and replicable. Therefore, the results of the ATR should be interpreted with caution and considered predominantly exploratory, while the most consistent instructions derive from the STR, which presented greater methodological rigor and, consequently, greater internal validity. Future studies should address these limitations through personalized instructions, in-person assessments whenever possible, and protocols that allow greater control of variables associated with physical activity.
Conclusions
Synchronous telerehabilitation positively affected lower limb functional capacity, dyspnea, fatigue, self-reported stress, and quality of life. In contrast, asynchronous telerehabilitation only positively affected perceived fatigue and anxiety. By considering the difficulty of telerehabilitation in a low- and middle-income country, synchronous telerehabilitation supervised by a physiotherapist seems more effective than an asynchronous telerehabilitation program in adults with post-COVID-19 syndrome.
In short, a TR offers a safe, accessible, and effective alternative. These results contribute valuable evidence to the growing body of literature and support the integration of TR into routine clinical practice for with post-COVID-19 syndrome management.
Long Covid