Effects of an outpatient pulmonary rehabilitation program on exercise capacity, muscle fatigue, dyspnea, and functional status in post-COVID-19 patients: non-randomized clinical trial

Abstract

Background

COVID-19 may lead to persistent sequelae that impair functional capacity, resulting in dyspnea, fatigue, and limitations in activities of daily living.

Objective

To evaluate exercise capacity, fatigue, dyspnea, and functional status in patients following hospitalization for COVID-19 who underwent an outpatient pulmonary rehabilitation program (PRP).

Methods

The study was approved by the Research Ethics Committee of UniEVANGÉLICA (approval number 4,296,707; September 24, 2020) and registered at ClinicalTrials.gov (Identifier: NCT04982042). This prospective single-arm non-randomized clinical trial included symptomatic individuals after COVID-19 infection. Exercise capacity was assessed using the Six-Minute Walk Test, muscle fatigue using the Fatigue Severity Scale (FSS), modified Medical Research Council (mMRC) Dyspnea Scale, and functional status using the Post-COVID-19 Functional Status Scale (PCFS). The PRP consisted of cardiorespiratory training and muscle strengthening exercises performed three times per week for six weeks.

Results

A total of 54 patients were included (29 in the Ward Group and 25 in the ICU Group). After the PRP, 6MWT distance increased from 428.52 ± 104.5 m to 503.19 ± 98.04 m (p < 0.0001), mMRC dyspnea decreased from 3.50 ± 1.31 to 1.56 ± 0.60 (p < 0.0001), PCFS improved from 2.91 ± 0.59 to 0.63 ± 0.75 (p < 0.001), and FSS decreased from 4.68 ± 1.50 to 2.97 ± 1.55 (p < 0.001). The Ward Group showed greater unadjusted gains than the ICU Group; however, multivariable analysis identified lower baseline 6MWD as the only independent predictor of improvement (β = −0.547; p < 0.001), whereas ICU admission was not independently associated with the magnitude of gain (p = 0.246).

Conclusion

The outpatient PRP was associated with significant improvements in exercise capacity, fatigue, dyspnea, and functional status. Lower baseline exercise capacity predicted greater improvement, regardless of ICU admission.

1. Introduction

The clinical course of SARS-CoV-2 infection is not limited to the acute phase and may compromise long-term functional recovery in affected individuals (1). Post-COVID-19 condition affects multiple organ systems, with exercise intolerance, muscle fatigue, and chronic dyspnea representing its most disabling manifestations (2, 3). Approximately 70% of COVID-19 survivors present a six-minute walk distance (6MWD) below predicted values up to one year after hospital discharge (4).

Exercise intolerance in post-COVID-19 condition is multifactorial and cannot be attributed solely to residual pulmonary damage; therefore, improvement of the acute illness does not necessarily guarantee restoration of pre-morbid functional capacity (5, 6). Cardiac and peripheral autonomic dysregulation contributes to a vicious cycle of impaired aerobic performance, characterized by reduced peripheral carbon dioxide production and an exaggerated hyperventilatory response during exercise (7). Furthermore, a persistent prothrombotic state associated with endothelial dysfunction and microthrombi within the muscular capillary bed compromises oxygen transport capacity during physical exertion, precipitating anaerobic metabolism and early-onset fatigue (8).

The severity of the initial clinical presentation during hospitalization directly influences the trajectory of post-COVID-19 recovery. Patients requiring intensive care, particularly those exposed to prolonged invasive mechanical ventilation, are at increased risk of intensive care unit-acquired weakness (ICU-AW), resulting from prolonged immobilization, deep sedation, and the use of neuromuscular blocking agents (9). This condition is characterized by alterations such as disuse-induced muscle atrophy, increased protein catabolism, and neuromuscular dysfunction, ultimately leading to reduced muscle strength and endurance (10). These mechanisms directly affect functional status and quality of life, decreasing independence in activities of daily living (ADLs) and perpetuating the cycle of physical inactivity and progressive deconditioning (11).

From a pulmonary perspective, impairments in alveolar-capillary diffusion, reduced lung compliance, and ventilation–perfusion mismatch may persist long after the acute phase of the disease (12). In addition, respiratory muscle weakness contributes to an increased ventilatory demand during physical exertion (13). At the peripheral level, mitochondrial dysfunction and impaired oxygen extraction by skeletal muscles promote the early onset of anaerobic metabolism, thereby exacerbating the perception of dyspnea and fatigue (14, 15).

Recent evidence has demonstrated that Pulmonary Rehabilitation (PR) has been shown to be an effective and safe intervention for the recovery of patients with post-COVID-19 condition (16). A clinical trial reported that pulmonary rehabilitation programs (PRPs) can significantly improve functional capacity, as evidenced by increased 6MWD, in addition to enhancing peripheral muscle strength and reducing muscle fatigue following the intervention (17). PR is recognized as a multidimensional intervention based on comprehensive individualized assessment, integrating aerobic and resistance exercise training, respiratory muscle training, breathing retraining strategies, and patient education. Its primary goals are to promote sustainable behavioral changes and optimize physiological responses to exercise (18).

Therefore, the implementation of an outpatient PRP for post-COVID-19 patients is justified based on the hypothesis that this intervention can provide significant benefits in both clinical and functional outcomes. Accordingly, the present study aimed to evaluate exercise capacity, muscle fatigue, dyspnea, and functional status in patients following hospitalization for COVID-19 who participated in an outpatient PRP.

2. Materials and methods

2.1. Study design

This was a prospective, consecutive, single-center, single-arm non-randomized clinical trial conducted at the Pulmonary Rehabilitation Laboratory of the Evangelical University of Goiás (UniEVANGÉLICA), involving individuals who presented persistent symptoms following COVID-19 infection. The study design followed the recommendations of the Transparent Reporting of Evaluations with Nonrandomized Designs (TREND) statement (19), as illustrated in Figure 1.

Figure 1.

Flow diagram of participant recruitment, follow-up, and analysis.

2.2. Ethical considerations

The study was approved by the Research Ethics Committee of UniEVANGÉLICA (approval number 4,296,707; September 24, 2020) and registered at ClinicalTrials.gov (Identifier: NCT04982042). All participants provided written informed consent prior to enrollment and were informed of their right to withdraw from the study at any time without penalty. All study procedures were conducted in accordance with international biosafety guidelines for protection against COVID-19 transmission.

2.3. Participant recruitment

Participants were recruited between May 2021 and December 2021 through social media advertisements and banners distributed in referral hospitals for COVID-19 treatment, as well as among individuals seeking care through municipal and state healthcare services in the city of Anápolis, Goiás, Brazil.

2.4. Inclusion criteria

Adults of both sexes aged between 18 and 75 years were eligible for inclusion. Participants were required to be clinically stable and present persistent symptoms or sequelae following COVID-19 infection confirmed by reverse transcription polymerase chain reaction (RT-PCR), antigen testing, or serological testing, in addition to having a medical referral for pulmonary rehabilitation. Clinical stability was defined as maintenance of symptoms without the need for therapeutic modification and absence of antibiotic or corticosteroid use, except when prescribed as part of chronic treatment.

2.5. Exclusion criteria

Individuals were excluded if they had chronic neurological, psychiatric, or psychological disorders that could impair comprehension or participation in physical exercise; recent musculoskeletal disorders with incomplete recovery; uncontrolled systemic arterial hypertension (SAH) or unstable cardiovascular disease preventing exercise performance; or acute rheumatologic conditions.

2.6. Outcomes and assessment instruments

A multidisciplinary team of physicians and physiotherapists, previously trained for the study procedures, was responsible for patient assessment and monitoring throughout the outpatient PRP. Before enrollment, participants underwent a structured clinical evaluation, during which sociodemographic data, pre-existing comorbidities, acute COVID-19 complications, length of hospital stay and intensive care unit (ICU) admission, oxygen supplementation requirements, and ventilatory support modality were recorded.

Subsequently, exercise capacity testing and the administration of dyspnea, fatigue, and functional status questionnaires were performed by a physiotherapist. This procedure was performed at both assessments: before the start of the PRP (pre-PRP) and after completion of the six-week intervention (post-PRP).

2.6.1. Exercise capacity—six-minute walk test (6MWT)

The 6MWT assesses the distance walked at a submaximal exercise intensity, defined as the highest level of exertion that can be continuously sustained for six minutes without transitioning to running. It is widely used as a simple, safe, and reproducible method for evaluating functional capacity and exercise tolerance (20). The test was performed according to the American Thoracic Society (ATS) guidelines (2002) (20) on a flat 30-meter corridor located in the UniEVANGÉLICA Sports Complex. Participants who discontinued the test before completion were excluded from the analysis. Vital signs (heart rate, peripheral oxygen saturation, and blood pressure) and nd ratings of perceived exertion (RPE) using the Borg scale were recorded immediately before the start (baseline) and immediately after completion (end of test) of the 6MWT. Predicted values were calculated using reference equations developed for the healthy Brazilian population (21).

2.6.2. Muscle fatigue—fatigue severity scale (FSS)

The FSS evaluates fatigue severity based on its impact on motivation, physical activity, functional capacity, and activities of daily living (ADLs). It is a self-reported questionnaire consisting of nine items scored on a seven-point Likert scale, reflecting the participant's level of agreement with each statement. Participants were instructed to answer the questionnaire considering their experience during the previous week. The final score was calculated as the mean of the nine items. Higher grade indicate greater fatigue severity, and a score ≥ 4 was considered indicative of clinically significant fatigue (22).

2.6.3. Dyspnea severity—modified medical research council (mMRC) dyspnea scale

The mMRC Dyspnea Scale is a widely used instrument for assessing the perception of breathlessness during activities of daily living. Owing to its simplicity and ease of administration, it is extensively employed in both clinical practice and research. The scale consists of five grades, from 0 to 4, with participants selecting the category that best reflects the degree of limitation imposed by dyspnea on their daily activities (23).

2.6.4. Functional status—post-COVID-19 functional status (PCFS) scale

The PCFS Scale is a validated instrument used to assess functional limitations associated with COVID-19. The scale ranges from Grade 0 (no functional limitation) to Grade 5 (death). However, only Grades 0–4 were used in the present study, as Grade 5 was not applicable to the study population. Participants completed the questionnaire based on their health status during the seven days preceding the assessment (24, 25).

2.7. Interventions

Given that no specific recommendations for the management of COVID-19 were available during the pandemic period, the outpatient PRP implemented in this study was adapted from pulmonary rehabilitation recommendations proposed by the Global Initiative for Chronic Obstructive Lung Disease (GOLD) (26). The program lasted six weeks and consisted of three weekly sessions, each comprising four phases: warm-up, cardiorespiratory training, peripheral muscle strengthening, and cool-down/relaxation (Table 1).

Table 1.

Cardiorespiratory and resistance training protocol implemented during the outpatient pulmonary rehabilitation program.

Training protocol implemented	Outpatient PRP
	Weeks 1–3	Weeks 4–6
	Sets × Repetitions	Sets × Repetitions
Cardiorespiratory Training
Treadmill or cycle ergometer
Duration	30 min	40 min
Intensity (% of peak HR achieved during the 6MWT)	60%–70% (RPE 4–5)	70%–80% (RPE 5–6)
Resistance Training
Upper Limbs
Upper-limb diagonal pattern (D1, modified Kabat technique)a,b	3 × 10–12	3 × 13–15
Upper-limb diagonal pattern (D2, modified Kabat technique)a,b	3 × 10–12	3 × 13–15
Lower Limbs
Hip flexion with knee extension	3 × 10–12	3 × 13–15
Extensores de joelho	3 × 10–12	3 × 13–15
Knee extension	3 × 10–12	3 × 13–15
Plantar flexion	3 × 10–12	3 × 13–15
Duration	20 min	30 min
Intensity (% of 1RM)	50%–60% (RPE 4–5)	50%–60% (RPE 5–6)

PRP, Pulmonary Rehabilitation Program; HR, heart rate; 6MWT, Six-Minute Walk Test; RPE, Rating of Perceived Exertion; D: diagonal; 1RM, one-repetition maximum; UL, upper limbs; LL, lower limbs.

^aAdler SS, Beckers D, Buck M. PNF in practice: an illustrated guide. Springer Science & Business Media, 2007.

^bBased on proprioceptive neuromuscular facilitation (PNF) diagonal movement patterns adapted from the Kabat method.

Progression of duration, aerobic intensity, and resistance load between phases was carried out provided that the participant demonstrated hemodynamic stability (systolic blood pressure <180 mmHg and diastolic blood pressure <110 mmHg, without a drop in systolic blood pressure of >20 mmHg compared to resting levels), SpO₂ ≥ 88% during exercise, and a Borg score between 4 and 6. Patients who did not meet these criteria maintained the load from the previous week until they could progress safely.

At the beginning of each session, vital signs were assessed, followed by a warm-up consisting of alternating calisthenic exercises targeting the major muscle groups of both the upper and lower limbs, according to each participant's individual tolerance. Subsequently, cardiorespiratory training was performed following the FITT principle (frequency, intensity, time, and type). This approach was selected because of its ability to promote musculoskeletal and cardiovascular adaptations induced by physical exercise, while also contributing to the reduction of adverse respiratory symptoms (27).

Cardiorespiratory training was conducted using a continuous exercise model on either a treadmill or cycle ergometer, according to participant preference. The initial exercise duration was set at 30 min and was increased to 40 min after three weeks of participation in the PRP (18). Similarly, exercise intensity was initially prescribed at 60%–70% of the maximum heart rate achieved during the 6MWT and was progressively increased to 70%–80% during the final three weeks of the program (28).

A one-repetition maximum (1RM) test was performed to determine the maximum load that participants could lift in a single repetition while maintaining proper technique and without the ability to complete an additional repetition. Prior to testing, participants received detailed instructions and performed a familiarization set consisting of 20 repetitions of the exercise to be tested using light loads. This procedure was intended to familiarize participants with the movement pattern and provide adequate pre-test warm-up (29).

During the 1RM assessment, participants were instructed to avoid isometric contractions during the transition between the concentric and eccentric phases of the movement, as well as breath-holding associated with the Valsalva maneuver. The protocol included exercises targeting the major muscle groups of both upper and lower limbs. Testing began with a load below the estimated maximum and was progressively adjusted whenever the participant successfully completed the movement. Rest intervals of approximately five minutes were allowed between attempts to minimize short-term muscular overload. The test was terminated after a maximum of six attempts to reduce muscle fatigue and improve the reliability of the measurements (30).

Based on the results of the 1RM test, the initial resistance training load was prescribed at 50%–60% of the maximum load achieved and was progressively increased to 60%–70% after three weeks of training (31). Peripheral muscle strengthening aimed to improve muscle strength, hypertrophy, and neuromuscular efficiency, while optimizing the hemodynamic response to exercise and reducing energy expenditure during daily activities. Exercise prescription was based on fundamental training principles, including progressive overload, specificity, and periodization, as participants adapted to the training stimulus (32). The cool-down phase consisted of global stretching and relaxation exercises.

2.8. Safety assessment

All participants were supervised throughout the pulmonary rehabilitation program (PRP) by an experienced physiotherapist who was a member of the research team. Vital signs, including heart rate, peripheral oxygen saturation, and blood pressure, were monitored at the beginning, during, and at the end of each session and recorded on individual follow-up forms.

Peripheral oxygen saturation and heart rate were continuously monitored using an Oled Graph G-Tech pulse oximeter (Choice Electronic Technology Co., Ltd., Beijing, China). Blood pressure was measured using a Premium sphygmomanometer and stethoscope (Wenzhou Medical Instruments Co., Ltd., Ningbo, China) at the beginning and end of each session, or whenever the participant reported symptoms suggestive of blood pressure alterations.

RPE were assessed throughout cardiorespiratory training using the modified Borg scale for dyspnea and lower-limb fatigue. Whenever systolic and/or diastolic blood pressure exceeded predefined safety thresholds, participants were instructed to reduce exercise intensity or temporarily discontinue the activity until blood pressure values returned to acceptable levels. If clinical abnormalities persisted, the participant was referred for specialized medical evaluation.

2.9. Sample size calculation

The sample size calculation was based on the study by Hockele et al. (33), which used exercise capacity (6MWD) as the primary outcome. The calculation was performed using G*Power software (34, 35) for a paired-sample analysis, assuming an expected effect size of 0.41, 80% statistical power, and a significance level of 5% (α = 0.05). Accounting for an anticipated dropout rate of 10%, a total sample size of 51 participants was estimated to detect a statistically significant change.

2.10. Statistical analysis

The Kolmogorov–Smirnov test was used to assess the normality of data distribution. For pre–post comparisons, paired Student's t-tests were applied to variables with a parametric distribution, whereas the Wilcoxon signed-rank test was used for variables with a non-parametric distribution.

Comparisons between hospitalization groups (Ward vs. ICU) were performed using one-way analysis of variance (ANOVA), followed by Tukey's post hoc test when the null hypothesis was rejected. For variables with a non-parametric distribution, the Kruskal–Wallis test was employed, and when significant differences were detected, pairwise comparisons were conducted using the Mann–Whitney U-test.

To assess correlations between clinical and functional variables, Pearson's correlation coefficient was used for variables with a parametric distribution, whereas Spearman's rank correlation coefficient was applied to variables with a non-parametric distribution.

A multiple linear regression analysis was performed to identify factors independently associated with improvement in exercise capacity following the outpatient PRP. The dependent variable was the change in six-minute walk distance (Δ6MWD), calculated as post-PRP 6MWD minus pre-PRP 6MWD. Age, sex, body mass index (BMI), baseline 6MWD, length of hospital stay, and ICU admission were included as independent variables based on their clinical relevance. Sex was coded as female = 0 and male = 1, and ICU admission as no = 0 and yes = 1. Multicollinearity among independent variables was assessed using the variance inflation factor (VIF). Regression results are reported as unstandardized regression coefficients (B), standard errors (SE), standardized coefficients (β), 95% confidence intervals (CI), and p-values. Model performance was summarized using the coefficient of determination (R²), adjusted R², and F statistic.

All statistical analyses were performed using the Statistical Package for the Social Sciences (SPSS) for Windows (Chicago, IL, USA), with the significance level set at p < 0.05.

3. Results

A total of 69 post-COVID-19 patients were assessed at admission to the outpatient PRP. Of the 62 patients initially enrolled in the study, 54 completed the outpatient PRP, yielding an adherence rate of 87%. For analysis, participants were stratified into two groups according to their hospital admission setting: (I) the Ward Group (WG), comprising 29 patients, and (II) the Intensive Care Unit Group (ICUG), comprising 25 patients.

Table 2 summarizes the demographic and clinical characteristics of the post-COVID-19 patients who participated in the PRP. The study population was predominantly male (62.96%) and of mixed race (40.74%). Participant age ranged from 24 to 77 years, with a mean age of 50.81 ± 12.41 years, while the mean body mass index (BMI) was 30.53 ± 6.28 kg/m². No significant differences were observed between the WG and ICUG regarding sex, ethnicity, age, body weight, BMI, systolic blood pressure, or diastolic blood pressure. However, patients in the ICUG exhibited greater clinical severity, as evidenced by a longer length of hospital stay (19.00 ± 10.48 days vs. 10.59 ± 5.78 days; p < 0.0001), a higher requirement for invasive mechanical ventilation, and a longer interval between hospital discharge and initiation of the PRP (49.04 ± 29.10 days vs. 20.17 ± 19.79 days; p < 0.0001). The mean interval between hospital discharge and PRP admission for the total sample was 33.8 ± 28.8 days.

Table 2.

Sociodemographic and clinical characteristics of post-COVID-19 patients enrolled in the outpatient pulmonary rehabilitation program.

Variables	Total (n = 54)	WG (n = 29)	ICUG (n = 25)	p-value
Sex				ns
Male	34 (62.96%)	16 (55.17%)	18 (72.0%)
Female	20 (37.04%)	13 (44.83%)	7 (28.0%)
Ethnicity				ns
White	21 (38.89%)	13 (44.83%)	8 (32.0%)
Mixed race	22 (40.74%)	9 (31.03%)	13 (52.0%)
Black	11 (20.37%)	7 (24.14%)	4 (16.0%)
Age (years)	50.81 ± 12.41	51.97 ± 12.07	49.48 ± 12.91	ns
Body weight (kg)	84.64 ± 17.83	82.24 ± 15.32	87.43 ± 20.32	ns
BMI (kg/m2)	30.53 ± 6.28	29.89 ± 6.09	31.28 ± 6.53	ns
SBP (mmHg)	122.78 ± 14.06	123.10 ± 13.91	122.40 ± 14.51	ns
DBP (mmHg)	81.30 ± 9.72	80.69 ± 8.84	82.00 ± 10.80	ns
Length of hospital stay (days)	14.40 ± 9.19	10.59 ± 5.78	19.00 ± 10.48	p < 0.0001
ICU length of stay (days)	11.88 ± 8.63	N/A	11.88 ± 8.63	–
Ventilatory Support				p < 0.0001
Oxygen therapy	16 (29.63%)	13 (44.83%)	3 (12.0%)
NIV	30 (55.56%)	16 (55.17%)	14 (56.0%)
IMV	8 (14.81%)	N/A	8 (32.0%)	–
Endotracheal intubation	5 (9.26%)	N/A	5 (20.0%)	–
Tracheostomy	3 (5.56%)	N/A	3 (12.0%)	–
Time from hospital discharge to PRP admission (days)	33.82 ± 28.78	20.17 ± 19.79	49.04 ± 29.10	p < 0.0001

WG, Ward Group; ICUG, Intensive Care Unit Group; BMI, body mass index; SBP, systolic blood pressure; DBP, diastolic blood pressure; ICU, intensive care unit; NIV, non-invasive ventilation; IMV, invasive mechanical ventilation; PRP, pulmonary rehabilitation program; N/A, not applicable; ns, not significant.

Data are presented as mean ± standard deviation or n (%).

Patients in the ICUG exhibited greater clinical severity, as evidenced by a longer LOS (19.00 ± 10.48 days vs. 10.59 ± 5.78 days; p < 0.0001), a higher requirement for invasive mechanical ventilation, and a longer interval between hospital discharge and initiation of the PRP (49.04 ± 29.10 days vs. 20.17 ± 19.79 days; p < 0.0001). These findings suggest a greater degree of clinical and functional impairment among patients previously admitted to the intensive care unit, consistent with the increased severity of illness observed in this subgroup.

During the clinical interview, six participants (11.11%) reported temporary continuation of home oxygen therapy following hospital discharge. However, none of the patients required long-term oxygen therapy at the time of enrollment in the PRP. No statistically significant differences were observed between groups regarding sex, ethnicity, age, body weight, BMI, systolic blood pressure, or diastolic blood pressure, indicating baseline homogeneity for these variables.

The comorbidities reported by post-COVID-19 patients are presented in Table 3. The most prevalent conditions in the overall sample were SAH (38.89%), anxiety (25.93%), and obesity (24.07%). Although no statistically significant differences were observed between groups, a higher frequency of comorbidities was noted among patients in the ICUG compared with those in the WG.

Table 3.

Pre-existing comorbidities reported at admission among patients who completed the outpatient pulmonary rehabilitation program.

Variables	Total (n = 54)	WG (n = 29)	ICUG (n = 25)	p-value
Anxiety	14 (25.93%)	5 (17.24%)	9 (36.0%)	ns
Asthma	2 (3.70%)	2 (6.90%)	0 (0.0%)	ns
Depression	3 (5.56%)	2 (6.90%)	1 (4.0%)	ns
Dyslipidemia	4 (7.41%)	2 (6.90%)	2 (8.0%)	ns
T2DM	7 (12.96%)	2 (6.90%)	5 (20.0%)	ns
COPD	1 (1.85%)	1 (3.45%)	0 (0.0%)	ns
Hepatic steatosis	3 (5.56%)	1 (3.45%)	2 (8.0%)	ns
AH	21 (38.89%)	10 (34.48%)	11 (44.0%)	ns
Hypothyroidism	2 (3.70%)	2 (6.90%)	0 (0.0%)	ns
Obesity	13 (24.07%)	5 (17.24%)	8 (32.0%)	ns

WG, Ward Group; ICUG, Intensive Care Unit Group; T2DM, type 2 diabetes mellitus; COPD, chronic obstructive pulmonary disease; SAH, systemic arterial hypertension; ns, not significant.

Data are presented as n (%).

During the acute phase of COVID-19, some patients developed clinical complications associated with the severity of the infection. In the WG, one patient (3.45%) presented acute myocardial infarction (AMI), one (3.45%) deep vein thrombosis (DVT), two (6.90%) pulmonary embolism (PE), and two (6.90%) renal complications. In the ICUG, one patient (4%) presented transient ischemic attack, one (4%) PE, and three (12%) renal complications. It is emphasized that, for the start of activities in the PRP, all these patients were already clinically stable, without the need for active therapeutic modification or hospitalization.

The exercise capacity of post-COVID-19 patients undergoing the outpatient PRP is presented in Table 4. The physiological and perceptual variables presented refer to measurements collected at the beginning (baseline) and at the end (end of test) of the 6MWT, assessed at two time points in the study: before the start of the program (pre-PRP) and after six weeks of rehabilitation (post-PRP).

Table 4.

Changes in exercise capacity, physiological responses, and perceived exertion before and after the six-minute walk test, assessed at pre- and post-outpatient pulmonary rehabilitation program in post-COVID-19 Patients.

Variable	Total (n = 54)		WG (n = 29)		ICUG (n = 25)		p
	Pre-PRP	Post-PRP	Pre-PRP	Post-PRP	Pre-PRP	Post-PRP
6MWD (m)	428,52 ± 104,5	503,19 ± 98,0a	433,4 ± 85,0	522,5 ± 93,9b	422,8 ± 125,0	480,76 ± 99,6c	p < 0.0001
Predicted 6MWD (%)	73,45 ± 15,9	87,85 ± 20,5a	75,3 ± 14,1	93,8 ± 23,0b	71,27 ± 17,9	80,93 ± 14,8c,d	p < 0.001
Baseline of 6MWT
SBP (mmHg)	122,59 ± 10,7	120,19 ± 13,8	121,72 ± 10,0	120,00 ± 17,3	123,60 ± 11,5	120,40 ± 8,4	ns
DBP (mmHg)	81,67 ± 9,9	80,56 ± 8,6	80,34 ± 8,7	80,00 ± 9,3	83,20 ± 11,1	81,20 ± 7,8	ns
HR (bpm)	93,93 ± 16,3	85,38 ± 12,5a	90,72 ± 14,2	83,00 ± 10,9b	97,64 ± 18,1	88,25 ± 13,8c	p < 0.001
SpO2 (%)	95,43 ± 2,0	96,26 ± 1,5a	95,66 ± 1,7	96,59 ± 1,5b	95,16 ± 2,3	95,88 ± 1,4	p < 0.001
Borg Dyspnea Score	1,44 ± 2,3	0,35 ± 1,2a	1,31 ± 2,3	0,24 ± 0,8b	1,60 ± 2,2	0,48 ± 1,5c	p < 0.01
Borg LL Fatigue Score	2,19 ± 2,6	0,37 ± 1,1a	2,14 ± 2,5	0,28 ± 0,8b	2,24 ± 2,7	0,48 ± 1,3c	p < 0.001
End of 6MWT
SBP (mmHg)	125,83 ± 11,8	124,44 ± 15,6	124,6 ± 11,3	125,2 ± 18,1	127,28 ± 12,4	123,60 ± 12,5	ns
DBP (mmHg)	83,15 ± 11,0	84,26 ± 9,6	82,1 ± 9,0	83,1 ± 9,7	84,40 ± 12,9	85,60 ± 9,6	ns
HR (bpm)	118,61 ± 19,5	116,47 ± 14,8	113,9 ± 16,1	113,9 ± 13,6	124,04 ± 22,0	119,38 ± 15,8	ns
SpO2 (%)	92,15 ± 3,6	93,61 ± 2,8a	92,6 ± 3,2	94,0 ± 3,0b	91,64 ± 4,1	93,20 ± 2,7c	p < 0.05
Borg Dyspnea Score	2,93 ± 2,9	0,98 ± 1,6a	3,0 ± 3,2	1,0 ± 1,5b	2,84 ± 2,6	0,96 ± 1,8c	p < 0.001
Borg LL Fatigue Score	4,39 ± 2,6	1,33 ± 1,8a	4,5 ± 2,7	1,2 ± 1,6b	4,24 ± 2,7	1,52 ± 2,0c	p < 0.0001
Peak HR (bpm)	125,64 ± 17,2	122,00 ± 12,8	121,5 ± 15,4	120,5 ± 10,8	130,52 ± 18,2	123,58 ± 14,7	ns
Lowest SpO2 (%)	89,93 ± 3,4	92,26 ± 2,9a	90,4 ± 3,7	92,2 ± 3,2b	89,37 ± 3,1	92,32 ± 2,9c	p < 0.001

WG, Ward Group; ICUG, Intensive Care Unit Group; PRP, Pulmonary Rehabilitation Program; 6MWD, six-minute walk distance; 6MWT, Six-Minute Walk Test; predicted (%), percentage of predicted value; SBP, systolic blood pressure; DBP, diastolic blood pressure; HR, heart rate; bpm, beats per minute; SpO₂, peripheral oxygen saturation; LL, lower limbs; RPE, Rating of Perceived Exertion; ns, not significant.

Data are presented as mean ± standard deviation.

^aStatistically significant difference between pre- and post-PRP assessments in the overall sample.

^bStatistically significant difference between pre- and post-PRP assessments in the WG.

^cStatistically significant difference between pre- and post-PRP assessments in the ICUG.

^dStatistically significant difference between WG and ICUG at the post-PRP assessment.

During the baseline 6MWT, 8 (14.81%) of the 54 participants were required to discontinue the test due to dyspnea, lower-limb muscle fatigue, or peripheral oxygen desaturation. Of these, 5 patients (17.24%) belonged to the WG and 3 patients (12.0%) to the ICUG. Following completion of the PRP, no participant required test interruption (p < 0.003).

A significant increase was observed in 6MWD (Δ = 74.67 m), as well as in the percentage of predicted values achieved (Δ = 14.4%), following the intervention (p < 0.0001). However, patients in the WG demonstrated greater functional improvement than those in the ICUG (p < 0.001).

Significant improvements were also observed in physiological and perceptual variables related to exercise performance, including reductions in resting heart rate (Δ = −8.55 bpm), increases in baseline peripheral oxygen saturation (SpO₂) (Δ =  + 0.83%), reductions in baseline dyspnea perception (Δ = −1.09 points), and lower-limb muscle fatigue (Δ = −1.82 points). At the end of the 6MWT, improvements were also observed in final SpO₂ (Δ =  + 1.46%), dyspnea perception (Δ = −1.95 points), and lower-limb muscle fatigue (Δ = −3.06 points). Furthermore, minimum SpO₂ recorded during the test increased significantly (Δ =  + 2.33%). Collectively, these findings demonstrate improved exercise tolerance and functional capacity following participation in the PRP among post-COVID-19 patients.

Table 5 presents the outcomes related to muscle fatigue, dyspnea perception, and functional status before and after participation in the outpatient PRP. A significant reduction in fatigue grade assessed by the Fatigue Severity Scale (FSS) was observed in both groups following the intervention (Δ = 1.71 points; p < 0.001), indicating a clinically meaningful improvement in perceived fatigue.

Table 5.

Changes in fatigue severity, dyspnea severity, and functional status before and after the outpatient pulmonary rehabilitation program in post-COVID-19 Patients.

Variable	Total (n = 54)		WG (n = 29)		ICUG (n = 25)		p
	Pre-PRP	Post-PRP	Pre-PRP	Post-PRP	Pre-PRP	Post-PRP
FSS	4,68 ± 1,50	2,97 ± 1,55a	4,56 ± 1,61	2,60 ± 1,34b	4,81 ± 1,38	3,37 ± 1,68c	p < 0.001
mMRC Dyspnea Score	3,50 ± 1,31	1,56 ± 0,60a	3,34 ± 1,34	1,55 ± 0,57b	3,68 ± 1,28	1,56 ± 0,65c	p < 0.0001
Grade 1	4 (7,41%)	27 (50%)	3 (10,34%)	14 (48,28%)	1 (4%)	13 (52%)	–
Grade 2	11 (20,37%)	24 (44,44%)	6 (20,69%)	14 (48,28%)	5 (20%)	10 (40%)	–
Grade 3	9 (16,67%)	3 (5,56%)	5 (17,24%)	1 (3,45%)	4 (16%)	2 (4%)	–
Grade 4	14 (25,93%)	0 (0%)	8 (27,59%0	0 (0%)	6 (24%)	0 (0%)	–
Grade 5	16 (29,63%)	0 (0%)	7 (24,14%)	0 (0%)	9 (36 %)	0 (0%)	–
PCFS	2,91 ± 0,59	0,63 ± 0,76a	2,69 ± 0,54	0,48 ± 0,74b	3,16 ± 0,55c	0,80 ± 0,76d	p < 0.001
Grade 0	0 (0%)	29 (53,70%)	0 (0%)	19 (65,52%)	0 (0%)	10 (40%)	–
Grade 1	0 (0%)	16 (29,63%)	0 (0%)	6 (20,69%)	0 (0%)	10 (40%)	–
Grade 2	12 (22,22%)	9 (16,67%)	10 (34,48%)	4 (13,79%)	2 (8%)	5 (20%)	–
Grade 3	35 (64,81%)	0 (0%)	18 (62,07%)	0 (0%)	17 (68%)	0 (0%)	–
Grade 4	7 (12,96%)	0 (0%)	1 (3,45%)	0 (0%)	6 (24%)	0 (0%)	–

PRP, Pulmonary Rehabilitation Program; WG, Ward Group; ICUG, Intensive Care Unit Group; FSS, Fatigue Severity Scale; mMRC, modified Medical Research Council; PCFS, Post-COVID-19 Functional Status Scale. ** p < 0.001; ** p < 0.001.

Data are presented as mean ± standard deviation or n (%).

^aStatistically significant difference between pre- and post-PRP assessments in the overall sample.

^bStatistically significant difference between pre- and post-PRP assessments in the WG.

^cStatistically significant difference between the Ward Group and the ICUG at baseline (pre-PRP).

^dStatistically significant difference between pre- and post-PRP assessments in the Intensive Care Unit Group (ICUG).

Regarding dyspnea, a significant reduction in mMRC dyspnea grade was observed after the intervention (Δ = 1.94 points; p < 0.0001). This improvement was accompanied by a marked decrease in the proportion of patients classified in the more severe dyspnea categories and a corresponding increase in the frequency of patients classified within the milder categories in both groups.

Functional status assessed using the Post-COVID-19 Functional Status (PCFS) Scale, also improved significantly following the PRP (Δ = 2.28 points; p < 0.001). This improvement was evidenced by an increase in the number of patients classified as Grade 0 or Grade 1 and a reduction in those classified as Grades 3 and 4. Patients in the ICUG exhibited greater functional impairment at baseline compared with those in the WG; however, both groups demonstrated significant functional recovery following the intervention.

The changes in exercise capacity, dyspnea, functional status, and muscle fatigue outcomes are also graphically presented in Figure 2. In all panels, significant reductions in post-PRP scores were observed in both the WG and the ICUG (p < 0.001 to p < 0.0001). Notably, the Ward Group demonstrated greater absolute gains in 6MWD compared to the ICU Group, although both groups showed statistically significant improvements across all evaluated outcomes.

Figure 2.

Exercise capacity, dyspnea perception, functional status, and muscle fatigue in post-COVID-19 patients undergoing the outpatient pulmonary rehabilitation program. WG, Ward Group; ICUG, Intensive Care Unit Group; 6MWD, six-minute walk distance; m: meters; mMRC, modified Medical Research Council; PCFS, Post-COVID-19 Functional Status Scale; FSS, Fatigue Severity Scale. **: p < 0.001; ***: p < 0.0001.

As shown in Figure 3, 6MWD was significantly and negatively correlated with age (r = −0.4040; p = 0.0024), length of hospital stays (r = −0.3975; p = 0.0042), PCFS score (r = −0.4816; p = 0.0002), and mMRC score (r = −0.3481; p = 0.0099). These findings indicate that older age, longer hospitalization, poorer functional status, and greater dyspnea severity were associated with lower functional exercise capacity.

Figure 3.

Correlations between Six-Minute walk distance and clinical and functional variables in post-COVID-19 patients.

As shown in Figure 4, significant positive correlations were observed between PCFS grade and age (r = 0.3314; p = 0.0144), length of hospital stay (r = 0.3321; p = 0.0151), and mMRC grade (r = 0.2826; p = 0.0384). Additionally, a significant positive correlation was found between FSS and mMRC grade (r = 0.3461; p = 0.0363). These findings indicate that greater dyspnea severity is associated with both increased functional limitation and higher levels of fatigue, whereas older age and prolonged hospitalization are associated with poorer post-COVID-19 functional status.

Figure 4.

Correlations between functional Status, fatigue, dyspnea, and clinical variables in post-COVID-19 patients.

Multiple linear regression analysis was performed to identify factors independently associated with the magnitude of improvement in 6MWD following the outpatient PRP (Table 6). The overall regression model was statistically significant (F[6,47] = 3.627; p = 0.0049), explaining 31.6% of the variance in Δ6MWD (R² = 0.316; adjusted R² = 0.229). Baseline 6MWD was the only variable independently associated with Δ6MWD (B = −0.431; standardized β = −0.547; 95% CI: −0.632 to −0.230; p < 0.001), indicating that participants with lower baseline exercise capacity showed greater improvements in walking distance following the PRP. Age (p = 0.646), sex (p = 0.517), BMI (p = 0.735), length of hospital stay (p = 0.900), and ICU admission (p = 0.246) were not independently associated with Δ6MWD. Variance inflation factors ranged from 1.11 to 1.26, indicating no relevant multicollinearity among the independent variables.

Table 6.

Multivariable linear regression analysis of predictors of improvement in six-minute walk distance following the outpatient pulmonary rehabilitation program.

Predictor	B	SE	Standardized β	95% CI for B	p-value
Age (years)	−0.384	0.830	−0.059	−2.054 to 1.287	0.646
Sex (male vs. female)	−14.503	22.202	−0.087	−59.167 to 30.162	0.517
BMI (kg/m²)	0.586	1.718	0.045	−2.871 to 4.043	0.735
Baseline 6MWD (m)	−0.431	0.100	−0.547	−0.632 to −0.230	<0.001
Length of hospital stay (days)	0.150	1.180	0.017	−2.225 to 2.524	0.900
ICU admission (yes vs. no)	−25.558	21.769	−0.158	−69.351 to 18.236	0.246

Model statistics: R² = 0.316; adjusted R² = 0.229; F(6,47) = 3.627; p = 0.0049; n = 54.

B, unstandardized regression coefficient; SE, standard error; β, standardized regression coefficient; CI, confidence interval; BMI, body mass index; 6MWD, six-minute walk distance; ICU, intensive care unit; PRP, pulmonary rehabilitation program.

The dependent variable was the change in six-minute walk distance (Δ6MWD), calculated as post-PRP 6MWD minus pre-PRP 6MWD. Sex was coded as female = 0 and male = 1; ICU admission was coded as no = 0 and yes = 1. Variance inflation factors ranged from 1.11 to 1.26, indicating no relevant multicollinearity among predictors. Bold indicates statistical significance (p < 0.05).

4. Discussion

The principal finding of this study was that a six-week outpatient PRP was associated with significant improvements in exercise capacity, reductions in dyspnea and fatigue, and substantial recovery of functional status in previously hospitalized post-COVID-19 patients. Although patients admitted to the intensive care unit exhibited greater baseline clinical impairment and longer hospital stays, both groups demonstrated favorable outcomes following the intervention, supporting the role of pulmonary rehabilitation in promoting functional recovery in this population (18, 36, 37).

Persistent manifestations following COVID-19 continue to represent a major challenge for healthcare systems, particularly among patients who required hospitalization during the acute phase of the disease. Dyspnea, fatigue, exercise intolerance, and functional limitations remain among the most frequently reported symptoms months after hospital discharge (38–40). Beyond physical impairments, recent evidence suggests that patients who required ventilatory support during hospitalization experience a substantial psychological burden and more complex recovery trajectories, indicating that post-COVID-19 recovery encompasses interconnected physical, emotional, and psychosocial dimensions (41, 42).

Regarding the recovery of exercise capacity, the findings of the present study demonstrated a significant increase in 6MWD after six weeks of pulmonary rehabilitation, consistent with previous studies that reported consistent improvements in exercise tolerance following structured rehabilitation programs in patients with post-COVID-19 syndrome (43–45). The mean increase of 74.67 m in the 6MWD observed in the total sample exceeds the 30 m threshold considered minimal clinically important difference (MCID) by the joint ATS/ERS consensus for chronic respiratory diseases (20), as well as the upper limit of 30.5 m identified in a systematic review of adults with various pathologies (46).

In this context, Gloeckl et al. (36), in a prospective cohort study involving 50 patients who underwent a three-week inpatient pulmonary rehabilitation program, reported increases of 48 m in the mild/moderate group and 124 m in the severe/critical group, both reaching statistical significance (p < 0.001). Although participants in the present study exhibited smaller absolute gains in the ICU group (+36.58 m) compared with the WG (+73.77 m), the relative improvement observed in the more severely affected group (+7.33% of predicted values) was lower than that reported by Gloeckl et al. This discrepancy may reflect differences in rehabilitation setting, duration, and training intensity, given the more intensive inpatient rehabilitation model employed in that study compared with the outpatient format adopted here.

Similarly, Hockele et al. (33), in a pilot clinical trial involving 29 post-COVID-19 patients with varying disease severity (mild to severe), demonstrated that a 16-session pulmonary and functional rehabilitation program (two months, twice weekly) produced a significant increase in 6MWD, from 326.3 ± 140.6 m to 445.4 ± 151.1 m (p < 0.001), with predicted values increasing from 59.7% to 82.6% (p < 0.001). In the present study, patients in the WG achieved improvements of comparable relative magnitude (+16.22% of predicted values), whereas the ICUG demonstrated a more modest increase (+7.33% of predicted values). These findings are consistent with the greater baseline functional impairment typically observed among patients who experienced more severe disease, as reported by Hockele et al.

Likewise, Oliveira et al. (44), in a meta-analysis of seven clinical trials (n = 188), reported a clinically meaningful increase in 6MWD of 60.56 m and a reduction in fatigue severity (FSS: −0.90), although no significant improvement in dyspnea was observed (mMRC: −0.57). In contrast, the present study demonstrated a substantial reduction in dyspnea severity (Δ = −1.86 in the WG and Δ = −2.12 in the ICUG; p < 0.0001). This discrepancy may be explained by the greater baseline severity of dyspnea among participants in the present cohort, who were predominantly classified within the highest mMRC categories, whereas baseline mMRC grade in the studies included in the meta-analysis were generally ≤1, thereby limiting the potential for detectable improvement. These findings support the hypothesis that multicomponent rehabilitation has been associated with adaptations across multiple functional domains, although the magnitude of the response may vary according to baseline disease severity and the extent of pre-existing functional impairment (47).

In the stratified analysis, patients previously admitted to the intensive care unit exhibited poorer baseline functional performance and smaller absolute gains compared with those admitted to hospital wards. These findings may reflect the greater clinical severity of patients requiring intensive care, including prolonged hospitalization, invasive mechanical ventilation, and extended periods of immobilization. Recent evidence indicates that factors associated with ICU admission may influence long-term outcomes and functional recovery (38, 48). Kolck et al. (49), using quantitative computed tomography in a cohort of 73 critically ill patients, demonstrated that individuals with COVID-19 experienced muscle mass loss at nearly twice the rate observed in other critically ill populations (2.1% vs. 1.2% per day; p = 0.008), with ICU length of stay identified as the only independent predictor of sarcopenia at three months. Such findings provide a plausible physiological basis for the greater baseline functional impairment observed among ICU survivors in the present study.

However, the multivariable analysis provides an important qualification to the between-group findings. After adjustment for age, sex, BMI, baseline 6MWD, length of hospital stay, and ICU admission, baseline 6MWD was the only variable independently associated with the magnitude of improvement in walking distance. The negative association between baseline 6MWD and Δ6MWD indicates that participants with poorer initial exercise capacity experienced greater gains following the PRP. This finding is consistent with previous rehabilitation studies showing substantial improvements in exercise capacity among post-COVID-19 patients with marked baseline functional impairment (33, 36, 45). In contrast, ICU admission was not independently associated with Δ6MWD, suggesting that the smaller unadjusted gains observed in the ICU group should not be interpreted as evidence of a lower responsiveness to pulmonary rehabilitation per se. Although ICU-related factors such as prolonged immobilization, mechanical ventilation, and accelerated muscle wasting may contribute to greater initial functional impairment and delayed recovery (9, 10, 49), the present adjusted analysis suggests that the magnitude of improvement in walking distance was more closely related to initial exercise capacity than to the hospitalization setting itself. This finding reinforces the importance of assessing baseline functional capacity when planning rehabilitation and interpreting treatment response in post-COVID-19 patients (18, 45).

The significant reduction in mMRC grade observed in the present study coincided with a meaningful improvement in dyspnea during activities of daily living, with most patients transitioning from more severe to milder dyspnea categories following pulmonary rehabilitation. Persistent post-COVID-19 dyspnea is recognized as a multifactorial phenomenon involving residual pulmonary abnormalities, cardiovascular dysfunction, physical deconditioning, and peripheral muscle impairment (50, 51). The mean reduction of ∼1.94 points on the mMRC scale exceeds the value of 1 unit recognized as clinically significant for this scale in patients with chronic respiratory diseases and post-COVID-19 (52). Santana et al. (50), through longitudinal cardiopulmonary phenotyping of COVID-19 survivors, demonstrated that persistent dyspnea may occur even in the absence of structural cardiopulmonary disease. The symptom was particularly pronounced among ICU survivors and appeared to be closely associated with physical deconditioning and post-intensive care syndrome (PICS).

This heterogeneity in recovery may also be mediated by alterations in cardiac autonomic regulation. Fonseca et al. (53) reported that post-COVID-19 patients presenting delayed heart rate recovery (≤12 beats/min) following the 6MWT exhibited poorer functional performance and greater dyspnea, independent of pulmonary function or maximal inspiratory pressure. These findings suggest that impaired vagal reactivation represents an additional mechanism contributing to exercise intolerance beyond purely pulmonary or muscular impairments. Within this context, aerobic exercise and resistance training may play a critical role in improving ventilatory efficiency and reducing the respiratory burden associated with physical activity (54).

The reduction in fatigue observed following pulmonary rehabilitation may be clinically relevant, as fatigue is recognized as one of the most disabling manifestations of post-COVID-19 syndrome (55, 56). Although a formal MCID for the FSS has not been established specifically for the post-COVID-19 population, estimates derived from studies in patients with chronic diseases indicate that a difference of at least 0.45 point on the FSS constitutes a clinically significant change (57). The mean reduction observed in the present study was 1.71 points, exceeding this threshold by more than threefold. Additionally, the sample mean shifted from 4.68 to 2.97, crossing the cutoff of ≥4 points recognized as indicative of clinically significant fatigue (22), which represents a qualitatively relevant change in the patients' fatigue status.

This finding is consistent with the recent meta-analysis of randomized controlled trials conducted by Pérez-Gisbert et al. (58), which demonstrated a moderate and statistically significant effect of pulmonary rehabilitation on fatigue reduction compared with control interventions (p = 0.0002), regardless of rehabilitation modality (supervised or home-based) or program duration (<12 or ≥12 weeks). These results further support the role of multicomponent rehabilitation strategies in alleviating fatigue and promoting functional recovery among post-COVID-19 patients (59, 60).

In contrast to the meta-analysis by Oliveira et al. (44), which did not identify a significant improvement in fatigue compared with controls, the present study demonstrated robust and consistent reductions in fatigue severity across both groups. This discrepancy may be attributable to the integrated multicomponent approach adopted in the present program (aerobic plus resistance training) and the greater baseline severity of participants, which provided a larger margin for detectable improvement.

Regarding functional status, the present findings revealed significant changes in PCFS grade in both study groups. A marked increase was observed in the proportion of patients classified within the lower grades of the scale following pulmonary rehabilitation, suggesting substantial functional recovery. Functional status represents a particularly relevant outcome in post-COVID-19 populations because it reflects the integrated impact of respiratory, muscular, and systemic manifestations on activities of daily living and social participation (61). A MCID has not yet been formally established for the PCFS (62); however, the mean transition of ∼2.3 functional categories observed in the present study, with a decrease from 77.8% to 0% in grades 3–4 and an increase from 0% to 83.3% in grades 0–1, indicates substantial functional recovery. This interpretation is supported by evidence that the PCFS is responsive to pulmonary rehabilitation in post-COVID-19 patients (24).

The correlation analyses further highlighted the interrelationship among the functional domains evaluated in this study. Reduced 6MWD was significantly associated with older age, longer LOS, greater dyspnea severity, and poorer functional status, findings that are consistent with previous evidence in the literature (63, 64). Older individuals appear to be more susceptible to the systemic consequences of COVID-19 infection, including accelerated muscle mass loss, reduced physiological reserve, and delayed functional recovery (65–67). Similarly, prolonged hospitalization predisposes patients to physical deconditioning, acquired muscle weakness, and persistent functional impairment following hospital discharge (49).

Furthermore, patients with more severe dyspnea simultaneously exhibited greater functional limitations and higher fatigue levels. These findings suggest that dyspnea may represent a central determinant in the persistence of post-COVID-19 functional impairment, directly affecting both exercise capacity and performance in activities of daily living (66, 67). Cecchetto et al. (68), in a multiparametric cardiopulmonary assessment of 50 post-COVID-19 patients with persistent dyspnea, reported that 62% presented diaphragmatic dysfunction, 48% exhibited hyperventilation, and 34% showed evidence of microvascular dysfunction, underscoring the heterogeneity of the pathophysiological mechanisms underlying this symptom. Likewise, Morgan et al. (69), in a systematic review of risk factors for persistent dyspnea, identified advanced age, prolonged hospitalization, and the need for mechanical ventilation as consistent predictors of chronic dyspnea, findings that closely align with the correlational results observed in the present study.

In summary, the beneficial effects observed following participation in the PRP may have been mediated by improvements in ventilatory efficiency and gas exchange, enhanced peripheral muscle oxidative capacity, and strengthening of both respiratory and locomotor muscles (70). Yang and Yang (71), in a review of pulmonary rehabilitation for the management of post-COVID-19 sequelae, reported that structured exercise training promotes cardiorespiratory adaptations that optimize pulmonary ventilation and tissue perfusion, mechanisms that are fundamental to reducing dyspnea and fatigue in post-COVID-19 patients. Collectively, these adaptations contribute to a reduced ventilatory burden during activities of daily living, improved exercise tolerance, and enhanced functional recovery (47).

Nevertheless, several limitations should be considered when interpreting the present findings. First, the absence of a control group precludes the isolation of the specific effects of the intervention from the natural course of recovery over time. This design choice was primarily driven by ethical considerations during the COVID-19 pandemic, as withholding pulmonary rehabilitation from symptomatic post-COVID-19 patients would have been inconsistent with emerging clinical recommendations. Furthermore, the differential response between the Ward and ICU groups suggests that baseline severity modulates rehabilitation outcomes, providing indirect evidence supportive of a treatment effect. Second, the single-center design and the follow-up period limited to the intervention phase restrict the generalizability of the findings to broader populations and healthcare settings.

Additionally, recruitment by medical referral and voluntary participation may have introduced a selection bias, as the included participants may represent individuals more motivated for functional recovery and with greater adherence to treatment. Furthermore, between-group comparisons and pre-post analyses for the remaining outcomes were not adjusted for baseline clinical severity, limiting the ability to distinguish between the effect of the hospitalization setting and the greater initial functional impairment of ICU patients. Finally, the lack of post-intervention follow-up prevents assessment of the long-term persistence of the observed benefits, an aspect of particular relevance given the chronic nature of post-COVID-19 condition.

Among the strengths of this study are its prospective design, the use of validated instruments for the multidimensional assessment of post-COVID-19 condition, and the inclusion of patients with heterogeneous levels of clinical severity, enabling evaluation of the impact of pulmonary rehabilitation across subgroups with distinct recovery profiles. Furthermore, the simultaneous assessment of exercise capacity, dyspnea, fatigue, and functional status provided a comprehensive understanding of the functional effects of the intervention in this population. Future multicenter randomized controlled trials incorporating control groups and long-term follow-up are warranted to confirm the sustainability of these benefits and to establish optimal pulmonary rehabilitation prescription parameters for post-COVID-19 patients.

5. Conclusions

A six-week outpatient PRP was associated with significant improvements in exercise capacity, dyspnea, fatigue, and functional status among previously hospitalized post-COVID-19 patients. Although greater unadjusted gains in exercise capacity were observed among patients previously admitted to hospital wards, ICU admission was not independently associated with the magnitude of improvement in 6MWD after adjustment for relevant demographic and clinical factors. Baseline 6MWD was the only variable independently associated with Δ6MWD, with lower initial exercise capacity associated with greater subsequent improvement. These findings highlight the importance of assessing baseline functional capacity when planning pulmonary rehabilitation and interpreting functional recovery in post-COVID-19 patients. However, given the single-arm design, these improvements should be interpreted as associations with participation in the rehabilitation program rather than definitive evidence of a causal treatment effect.

Funding Statement

The author(s) declared that financial support was received for this work and/or its publication. LO received grants from Research Productivity, modality PQII; process no. 310241/20227 of Conselho Nacional de Desenvolvimento Cientifico e Tecnologico (local acronym CNPq), Brazil. CO received grants from Research Productivity, modality PQII; process no. 302716/ 2025-4 of Conselho Nacional de Desenvolvimento Cientifico e Tecnologico (local acronym CNPq), Brazil.

Data availability statement

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

Ethics statement

The studies involving humans were approved by Research Ethics Committee of Evangelical University of Goiás (UniEVANGÉLICA) and registered at ClinicalTrials.gov (Identifier: NCT04982042). The studies were conducted in accordance with the local legislation and institutional requirements. The participants provided their written informed consent to participate in this study.

Author contributions

MO: Visualization, Data curation, Investigation, Writing – original draft, Formal analysis. PC: Data curation, Visualization, Writing – review & editing, Conceptualization. HP: Investigation, Visualization, Formal analysis, Writing – original draft, Data curation. AF: Formal analysis, Writing – original draft, Investigation, Data curation. LA: Data curation, Writing – original draft, Investigation, Visualization, Formal analysis. JR: Writing – original draft, Formal analysis, Visualization, Data curation, Investigation. RO: Formal analysis, Investigation, Data curation, Writing – original draft. TF: Methodology, Validation, Conceptualization, Writing – original draft, Supervision, Data curation, Investigation, Visualization, Funding acquisition, Writing – review & editing, Project administration, Formal analysis. DM: Validation, Conceptualization, Writing – review & editing, Methodology, Investigation, Funding acquisition, Supervision, Formal analysis, Data curation, Writing – original draft, Visualization, Project administration. RA: Investigation, Writing – original draft, Data curation, Formal analysis. DO: Data curation, Writing – original draft, Investigation, Formal analysis. CO: Investigation, Visualization, Writing – original draft, Formal analysis, Data curation. CS: Investigation, Visualization, Supervision, Conceptualization, Funding acquisition, Project administration, Formal analysis, Writing – review & editing, Data curation, Methodology, Validation. OG: Formal analysis, Data curation, Writing – original draft, Investigation. WF: Visualization, Funding acquisition, Validation, Conceptualization, Project administration, Data curation, Methodology, Writing – review & editing, Formal analysis, Investigation, Writing – original draft, Supervision. LO: Project administration, Visualization, Funding acquisition, Validation, Formal analysis, Conceptualization, Writing – review & editing, Data curation, Supervision, Writing – original draft, Methodology, Investigation.

Conflict of interest

The author(s) declared that this work was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

The author PC declared that they were an editorial board member of Frontiers at the time of submission. This had no impact on the peer review process and the final decision.

Generative AI statement

The author(s) declared that generative AI was not used in the creation of this manuscript.

Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.

Publisher's note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.