Effects of Early Physical and Pulmonary Rehabilitation for Severely and Critically ill COVID-19 Patients: A Retrospective, Cohort, and Multicenter Study

Di Qi Second A liated Hospital of Chongqing Medical University Xiaofeng Yan Chongqing Public Health Medical Center Jianglin Xiang Chongqing Three Gorges Central Hospital Junnan Peng Second A liated Hospital of Chongqing Medical University Qian Yu Second A liated Hospital of Chongqing Medical University Xumao Tang Second A liated Hospital of Chongqing Medical University Jin Tong Second A liated Hospital of Chongqing Medical University Longhua Feng Qianjiang Central Hospital of Chongqing Jing Yuan Chongqing Public Health Medical Center Yuyan Song Chongqing Public Health Medical Center Guodan Yuan Chongqing Public Health Medical Center Daoxin Wang (  wangdaoxin0163@163.com ) Second a liated hospital of Chongqing medical university


Background
Presently, the global spread of COVID-19 has caused over 700,000 deaths. Coronavirus infections can cause illnesses ranging from a mild cold to severe respiratory infections, such as severe acute respiratory syndrome (SARS). Previous evidence showed that many patients who recovered from SARS complained of limitations in physical function from general weakness to dyspnea in the early rehabilitation phase, which is consistent with lung function abnormalities 1 . COVID-19 is a highly infectious disease that leads to respiratory, physical, and psychological dysfunctions, especially for severely ill patients. It is reported that 5% of COVID-19 patients are severe patients who require intensive care unit (ICU) care, and 71% are critically ill patients with hypoxemic respiratory failure, acute respiratory distress syndrome (ARDS) or sepsis requiring treatment with mechanical ventilation 2,3 .On that note, ARDS survivors continue to have functional disabilities one year after discharge from the ICU 4,5 . Additionally, most ARDS patients have extrapulmonary conditions, with muscle wasting and weakness being the most prominent issues 6 . Impaired muscle function and loss of muscle mass, due to routinely immobilization and managed sedation, potentiate functional disability in critical patients 7 . Early mobilization and rehabilitation in the course of critical illness has been proposed to improve respiratory and diaphragmatic muscle strength, attenuate functional impairments and be associated with better clinical prognosis 8,9 , suggesting a promising therapeutic intervention for patients with COVID-19.
As the effects of rehabilitation therapy on COVID-19 patients remains to be elucidated, we sought to assess the effects of early physical and pulmonary rehabilitation on the clinical and functional outcomes in severely and critically ill COVID-19 patients. The goal of this multicenter study was to provide evidence for the potential bene ts of early rehabilitation therapy in these patients and suggest an option other than routine respiratory support.

Patients
We performed the descriptive, multicenter study in accordance with the Declaration of Helsinki and patients with COVID-19 were diagnosed according to the Chinese management guideline for COVID-19 (version 6.0) 10 . Severity of COVID-19 was de ned as mild, moderate, severe and critical according to the report of the WHO-China Joint Mission on Coronavirus Disease 2019 11 . We screened the laboratory con rmed COVID-19 patients admitted to four designated hospitals in the Chongqing provincial municipality (Qianjiang central hospital of Chongqing, Chongqing three gorges central hospital, Yongchuan hospital of Chongqing medical university, and Chongqing public health medical center) from January 21 to March 15, 2020 to identify their suitability. The severely and critically ill patients admitted to the ICU with stable cardiovascular, hemodynamic, pulmonary or neuropsychological statuses were enrolled. Patients were excluded for the following reasons: dyspnea, a respiratory frequency of more than 40 beats/min, a blood oxygen saturation of less than 90%, marked ventilator asynchrony, insecure airway, a systolic blood pressure (SBP) of less than 90 or more than 180 mmHg, a mean blood pressure (MAP) of less than 65 or more than 110 mmHg, a heart rate (HR) of less than 40 beats/min or more than 130 beats/min, a temperature of more than 38.5 ℃, agitation, deep venous thrombosis, chronic heart failure, active myocardial ischaemia, severe ischemic or hemorrhagic stroke, severe neurodegenerative diseases, absent limbs, and less than 18 years of age.

Study Design and Procedures
Patients in the control group received the standard or usual treatment, while patients in the intervention group received physical and pulmonary rehabilitation therapies within 72 h of admission to the ICU. Physical and pulmonary rehabilitation therapies included (1) position management and prone-position ventilation, (2) early mobilization, (3) respiratory muscle training, (4) percussion and vibrations, (5) and stretching and physical exercises. For body position management, patients were assisted to gradually increase antigravity position until they could actively maintain a vertical position for 30 min at a time, three times a day. ARDS patients underwent prone-position ventilation for at least 12 h. For early mobilization, patients began with assisted or independent mobility exercises in the supine or lateral decubitus position. If they tolerated these exercises, patients were encouraged to increase bed mobility activities and sitting balance activities. These exercises gradually progressed to transfer training, including sit-to-stand and moving from the bed to the chair repetitions, followed by stepping on the spot, and nally, walking. For respiratory muscle training, conscious patients were instructed to undergo pursed-lip and deep breathing exercises, and were trained with a commercial positive expiratory pressure device (Acapella Choice; Smiths Medical ASD, Inc. Keene, USA). Patients breathed in through Acapella Choice, held their breath for two to three seconds, breathed out slowly and smoothly for six to nine seconds, and nally performed coughs. Patients repeated these breathing exercises 10 times per session and four times a day. For percussion and vibrations, patients underwent cupped hands and high frequency chest wall oscillation to shift secretions from the peripheral airway to the central airway and enhance airway clearance and postural drainage. For stretching exercises, patients were ordered to move their limbs in exion, extension, and rotation. The physical exercises, repeated 30 min a day, included passive or active axial loading exercises and activities against gravity.
The entire rehabilitation therapy progression that was dependent on patients' tolerance, medical stability, and cognitive abilities was strictly monitored. The following conditions suggested cardiovascular, hemodynamic, pulmonary, or neuropsychological instabilities that hampered the continuation of pulmonary rehabilitation therapy: dyspnea, a respiratory frequency of more than 40 beats/min, arrhythmia, a pulse oximetry of less than 88%, new ventilator asynchrony, a SBP of less than 90 or more than 180 mmHg, a MAP of less than 65 or more than 110 mmHg, a HR of less than 40 or more than 130 beats/min, deterioration of conscious state, new agitation, and being physically fatigued or antagonistic.

Sample and Data Collection
Respiratory specimens were collected at admission to detect the presence of SARS-CoV-2 by real-time reverse-transcriptase polymerasechain-reaction (RT-PCR) assay in designated authoritative laboratories of local centers for disease control and prevention. Cytokines and lymphocyte subsets in peripheral blood were measured by uorescence-labeled ow cytometry (Beckman, Cell Lab Quanta SC). Spirometry and pulmonary diffusion capacity tests (MasterScreen, Hoechberg, Germany) were performed following the American Thoracic Society and European Respiratory Society's guidelines, on the day of discharge, as well as one month and three months after hospital discharge. The epidemiological, demographic, clinical, laboratory, radiological, treatment and outcomes data from patients' medical and nursing records were obtained and analyzed. The date of disease onset was de ned as the day when the symptoms were noticed. Lastly, we entered the data into a computerized database and two physicians veri ed this data.
Outcomes PCR-con rmed negative patients were evaluated at hospital discharge, and one month as well as three months after hospital discharge in the local ambulatory clinic. The chosen primary outcomes were 28-day, ICU, and hospital mortality rates, lengths of stay in the ICU and hospital, duration of mechanical ventilation, ventilator-free days, complications, and adverse occurrences during hospital stays. We de ned the secondary outcomes at hospital discharge and during the three-month follow-up as follows. To assess pulmonary function, we measured the forced expiratory volume within one second (FEV1), forced vital capacity (FVC), total lung capacity (TLC), and carbon monoxide diffusion capacity (DLCO). For independent functional status, we evaluated activities of daily living (ADL) with the Barthel index score that covers 10 subtest items (feeding, moving from wheelchair to bed and return, grooming, getting on/off the toilet, bathing, walking, stair climbing, dressing, bowel and bladder continence) ranging from 0 to 100. We also used the Functional Independence Measure (FIM) that contains 18 items (13 for motor ADL and ve for cognitive ADL) ranging from 1 (requiring total assistance) to 7 (completely independent). We estimated muscle strength using the Medical Research Council (MRC) score and incidences of ICU-acquired weakness (ICU-AW). Exercise capacity was evaluated through the standard six-minute walk test (6MWT), using the saturated pulse oximetry to monitor vital signs and percutaneous oxygen saturation. Lastly, health-related quality of life was assessed by St. George's Respiratory Questionnaire (SGRQ) that includes three components measuring patients' impaired health and perceived well-being. The scores range from 0 to 100, with higher scores indicating more limitations.

De nitions
Fever was de ned as an axillary temperature of 37.3 °C or higher. ARDS was de ned and classi ed according to the Berlin de nition 12 .
Sepsis were de ned according to the 2016 Third International Consensus De nition for Sepsis and Septic Shock 13 . Acute kidney injury was diagnosed according to the KDIGO clinical practice guidelines 14 . Cardiac injury was diagnosed if the serum concentration of cardiac biomarker (e.g., high-sensitivity cardiac troponin I, hsTnI) was above the 99th percentile upper reference limit (> 28 pg/mL) or if new abnormalities were shown in the electrocardiography. Hypoproteinaemia was de ned as a serum albumin of less than 25 g/L. ICU-AW referred to cooperative patients who had a MRC score of less than 48 out of 60 points, after the assessment of the muscle strength in 12 regions on both sides of the upper and lower limbs.

Statistical Analysis
Continuous variables were expressed as the means and standard deviations (S.D.) or medians and interquartile ranges (IQR) as appropriate. Categorical variables were presented as the counts and percentages. We performed comparisons with independent group t-tests between continuous variables of two groups when the data were normally distributed; otherwise, we used the Mann-Whitney U test. Categorical variables' proportions were compared using the Chi-square test or Fisher's exact test. The repeated measurements of pulmonary function test and scores (Barthel Index, FIM, MRC, SGRQ, and 6MWT) at different time points were compared with one-way or two-way analysis of variance followed by Turkey post-hoc tests. All the tests were two-sided and a value of p < 0.05 was considered statistically signi cant. Finally, we performed the data analyses using GraphPad Prism 7.0 software.

Demographics and Clinical Characteristics
From January 21 to March 15, 2020, we screened 427 laboratory-con rmed patients. We excluded 54 patients, due to crucial data incompleteness in the original reports, and enrolled 89 severely and critically ill patients from the ICU. Among them, we excluded three patients who suffered from acute myocardial infarction or cardiogenic shock, three who were unable to complete the rehabilitation therapies, and another three patients who were unable to complete the pulmonary function test. Thus, a total of 80 severely and critically ill COVID-19 patients from four designated-hospitals were included in this study, of whom 43 patients completed the physical and pulmonary rehabilitation therapies were assigned to the intervention group, and the remaining 37 patients were assigned to the control group. Table 1 lists their baseline demographics and clinical characteristics. We observed no signi cant differences in age, gender, comorbidities, independent functional status, and disease severity between the two groups. On admission, the typical signs and symptoms included fever, cough, and myalgia or arthralgia. Notably, dyspnea was a common symptom in severely and critically ill COVID-19 patients. There were no signi cant differences in abnormal physical examination signs on admission.

Laboratory and Radiologic Results
As listed in Supplementary Table 1, no obvious differences were observed in the major laboratory and radiologic results on admission. Most patients in our study had lymphopenia with decreases in CD4 positive T cell counts. The CD8 positive T cell levels were lower than normal in close to half of the patients. Additionally, declines in albumin and elevation in the D-dimer were common in the severely and critically ill patients. Regarding the infectious index, 70.0% and 56.25% of patients had higher levels of C-reactive protein and procalcitonin, respectively.
Furthermore, we observed high levels of IL-6, TNF-α, and IL-17A in 65.0%, 55.0%, and 51.25% of cases, respectively. As for radiologic examinations, most patients had pneumonia with ground-glass opacity or patchy shadowing, and consolidation or brous stripes-like shadows were observed in 43.75% of patients.

Treatment Strategies
All severe and critically ill patients in our study were treated in the ICU with isolated wards. Supplementary Table 2 illustrates the range of treatments. Antiviral therapies and empirical antibiotic treatments were administrated to all severely and critically ill COVID-19 patients in our study, while 76.25% of patients without obvious immune de ciencies received systemic glucocorticoid therapy, usually methylprednisolone 1-2 mg/kg for three to ve days. A total of 57 patients received immunopotentiators (thymalfasin 1.6 mg/day) or immunoglobulin treatments (human γ-globulin, 200-300 mg/kg for three to ve days). Invasive mechanical ventilation, non-invasive mechanical ventilation, and high-ow nasal cannula (HFNC) oxygen therapy was adopted in 11.25%, 60%, and 58.75% of patients with respiratory failure, respectively. Compared to the intervention group, patients in the control group received more HFNC therapy. Most patients received traditional Chinese medicine and anticoagulation therapy, and only a few received extracorporeal membrane oxygenation (ECMO), kidney replacements, and convalescent plasma therapy.

Clinical Outcomes
As shown in Table 2, there were no signi cant differences in the ICU and hospital mortality between the intervention and control groups. A 78 year old female in the control group died from cardiac arrest three days after hospital discharge, potentially due to an acute pulmonary thromboembolism. Thus, the 28-day mortality in our study is 5%, with no signi cant differences between the two groups. We also found no signi cant differences in the median length of stay in the ICU and hospital stay for patients between the intervention and control groups. As for mechanical ventilation therapies, we found that the median duration of non-invasive mechanical ventilation and HFNC signi cantly decreased in the intervention group compared to the control group, and early rehabilitation enhanced the number of ventilator-free days. We also noted no signi cant difference in the median duration of invasive mechanical ventilation. In our study, early rehabilitation therapy did not increase the rate of adverse events, and incidences of deep vein thrombosis were lower in the intervention group than in the control group. Common complications were ARDS, acute kidney injury, shock, and acute cardiac injury, with no obvious differences between the two groups.

Pulmonary Function Tests
On the day of discharge, the room air oxygen saturation at rest was normal in all cases and no signi cant difference was observed between two groups. As shown in Table 3-5, we noted pulmonary function abnormalities in the percentage of observed FVC related to its predicted values (FVC % predicted) in 36 cases, the percentage of observed FEV1 related to its predicted values (FEV1% predicted) in 37 patients, and the FEV1/FVC ratio (FEV1/FVC %) in 24 cases at hospital discharge, that signi cantly improved after three months of recovery. Besides, as shown in Table 3-5 and Fig. 1, no obvious differences in these ventilatory parameters were found between intervention and control groups at hospital discharge, as well as one month and three months after hospital discharge (all p > 0.05).    As shown in Table 3-5, most discharged patients had obvious impairments in lung volume and diffusing capacity at hospital discharge, as manifested by the abnormalities in the percentage of observed TLC related to its predicted values (TLC % predicted) in 56 cases, and the percentage of observed DLCO related to its predicted values (DLCO % predicted) in 56 cases. Figure 1 illustrated that, from hospital discharge to one month after, TLC % predicted and DLCO % predicted improved from 73.11-82.05% (p < 0.0001) and from 68.40-71.57% (p = 0.3468), respectively. By three months, these median values signi cantly improved to 84.66% (p < 0.0001) and 79.26% (p < 0.0001), respectively. At hospital discharge, TLC % predicted and DLCO % predicted for patients in the intervention group were higher than those in the control group, indicating lung volume and diffusing capacity improvements stemming from the early physical and pulmonary rehabilitation therapy. One month after discharge, these improvements were still present. However, three months after discharge, although the DLCO % predicted of patients in the intervention group was still higher, the TLC % predicted exhibited a notable, but statistically insigni cant, enhancement between the two groups ( Fig. 1 and Table 3 (Fig. 2). At hospital discharge, patients in the intervention group had signi cantly higher Barthel Index scores and FIM than those in the control group, indicating that severe COVID-19 patients bene ted from early rehabilitation, as it helped them return to independent functional status. These bene cial effects lasted for one month after discharge, as evidenced by the higher Barthel Index scores and FIM in the intervention group than in the control group. There was no further signi cant improvement at three months after discharge (Fig. 2, Table 3-5). Additionally, at hospital discharge, patients in the intervention group needed less time to achieve several activity milestones, such as getting out of bed, stepping on the spot, and walking (Table 3).

Muscle Strength
As demonstrated in Fig. 2, compared to the value at hospital discharge, the MRC score signi cantly enhanced one month after hospital discharge (46.46 ± 7.58 vs 49.95 ± 7.36, p = 0.0058), with no further signi cant improvements after three months (49.95 ± 7.36 vs 50.67 ± 5.64, p = 0.7973). Patients in the intervention group had a signi cantly higher MRC score at hospital discharge, as well as one month and three months after hospital discharge. Less ICU-AW was noted in the intervention group than in the control group at hospital discharge and one month after discharge. At three months after discharge, we found a decreasing, but statistically insigni cant trend in ICU-AW for the control group (Fig. 2, Table 3-5).

Exercise Capacity
Walking distance within 6 minutes signi cantly increased one month after hospital discharge compared to right after discharge (208.90 ± 64.85 vs 280.08 ± 37.82, p < 0.0001), with further signi cant improvements after three months (280.08 ± 37.82 vs 357.57 ± 56.98, p < 0.0001) (Fig. 2). Patients in the intervention group had a signi cantly greater walking distance in six minutes at hospital discharge, as well as one month and three months after discharge (Fig. 2, Table 3-5).

Health-Related Quality of Life
The SGRQ score at one month after hospital discharge was higher than at three months after hospital discharge (51.74 ± 9.56 vs 46.36 ± 9.52, p = 0.0007). This score was lower in the intervention group than in the control group at one month and three months after hospital discharge (Table 4, 5), indicating that early rehabilitation accelerated the improvement of life quality in severely and critically ill COVID-19 patients.

Discussion
Our present study demonstrated that, although no signi cant alternations in 28-day, ICU, and hospital mortality rates, as well as patients' lengths of stay in hospital and ICU were found, early physical and pulmonary rehabilitation in severely and critically ill patients with COVID-19 increased the duration of ventilator-free days during hospitalization, with no increased adverse events and complications. Additionally, early rehabilitation in severely ill COVID-19 patients enhanced the recovery of lung volume and diffusing capacity, accelerated patients' return to independent functional status, decreased ICU-AW incidences, and increased muscle strength and exercise capacity at hospital discharge and one month after discharge. The improvements in lung diffusing capacity, muscle strength, life quality, and exercise capacity remained signi cant up to three months after hospital discharge. Therefore, the implementation of early physical and pulmonary rehabilitation was safe and effective for severely and critically ill patients with COVID-19 to promote their functional, physical and psychological recovery.
Skeletal muscle mass is known to be an important determinant factor for critical patients to overcome diseases in the ICU 7 . Loss of skeletal muscle mass, quality, and strength in the ICU is an independent mortality risk factor for critically ill patients, and muscle weakness affecting the limbs and respiratory muscles during ICU hospitalization is associated with poor outcomes, such as delayed recovery, hindered weaning from mechanical ventilation, decreased life quality, and increased nancial costs 15,16 . Accumulated evidence has revealed that early ICU rehabilitation, including pulmonary rehabilitation to improve respiratory function and remove airway secretions, and physical rehabilitation to attenuate muscle weakness, has positive impacts on critically ill patients' prognoses 8,9,17−19 . Therefore, early rehabilitation in the ICU is supposed to improve the clinical outcomes and functional recovery of severe COVID-19 patients. However, to our knowledge, a limited number of studies have reported the impacts of early physical and pulmonary rehabilitation on COVID-19 patients, especially for severe and critical cases. For example, one case report shared the early pulmonary rehabilitation experiences of a 41-year-old severe patient with SARS-CoV-2 pneumonia 20 . Another study found that a six-week respiratory rehabilitation program produced positive impacts on COVID-19 patients aged over 65 by improving respiratory functions, quality of life, and exercise endurance, as well as relieving anxiety 21 . However, no study has systematically assessed the effects of early physical and pulmonary rehabilitation on severe COVID-19 patients. In our present study, we found that implementing early rehabilitation in severe and critical patients with COVID-19 increased the amount of ventilator-free days during hospitalization, largely attributed to the removal of airway secretions and amelioration of respiratory failure. These results were consistent with most previous reports concerning the bene cial impacts of early physical medicine and rehabilitation on critically ill patients in the ICU 22 .
Previous study has found that lung functional impairments in the recovered patients with SARS-coronavirus pneumonia were common and could last for months or even years, among which impaired diffusing capacity and defected lung volume were the most common abnormalities 1 . In our present study, we witnessed varying degrees of declining DLCO, TLC, FEV1, FVC, and FEV1/FVC levels in the severe and critical COVID-19 patients at hospital discharge, consistent with a recent report revealing the diffusion capacity impairments and restrictive ventilatory defects in COVID-19 patients at hospital discharge 23 . Histological examinations of the lung tissue in COVID-19 patients con rmed the diffuse alveolar damages with hyaline membrane formation and bromyxoid exudates 24 . Follow-up study lasting for one year found that the persistent functional limitation in the survived ARDS patients was largely a result of muscle wasting and weakness, due to their immobilization during hospitalization 6 .Thus, pulmonary alveolar edemas, interstitial brous proliferation, and muscle wasting collectively resulted in impaired lung function. Previous study found that respiratory rehabilitation could clearly enhance improvements in the respiratory functions of elderly COVID-19 patients 21 . However, we only found signi cant promotions in lung volume and diffusing capacity, potentially due to the differences in our patients' disease severity, baseline comorbidities, and smoking history.
In our current study, the interventions of early rehabilitation therapy signi cantly improved independence functional outcomes at hospital discharge and one month after discharge, supporting the bene ts and feasibility of these interventions on the restoration of functional independence. Our results are consistent with previous ndings demonstrating that whole-body rehabilitation in the earliest days of critical illness result in better functional outcomes, as re ected by higher Barthel Index scores, more independent ADLs, and greater unassisted walking distance at hospital discharge 9 . After three months, the independence functions improved for all patients and the bene cial effects of rehabilitation therapy subsided. A variety of improvements in muscles, nerves, metabolism immunity, and psychological states may explain the long-term recovery.
The adverse impacts of skeletal muscle immobility are well-recognized, as individuals show skeletal muscle atrophic signs if immobilized for more than 72 hours, and loss of muscle mass and strength can be noted even in well-nourished healthy older adults within 10 days of bed rest 25,26 . Therefore, physical rehabilitation in the ICU for severe COVID-19 patients, especially those combined with ARDS, is supposed to be an important intervention to preserve muscle architecture, improve muscle strength, and decrease ICU-AW. In present study, we con rmed that early physical rehabilitation results in improvements in muscle strength and exercise capacity in the three-month follow-up, with decreases in ICU-AW incidences at hospital discharge and one month after discharge. Impaired muscle function may explain the compromised quality of life. Thus, we postulate that the observed improvements in life quality and physical function may partially be due to the remission of muscle weakness. These results are consistent with most reports supporting the bene cial effects of early mobilization and rehabilitation on greater muscle strength in critically ill patients 17,22 . However, one meta-analysis found that early mobilization and rehabilitation had no signi cant effects on muscle strength, quality of life, and ICU-AW incidences, perhaps due to the inadequate inclusion of published randomized controlled trials (RCT) 27 . As for the mechanisms of rehabilitation training on these bene ts, in addition to conventional concepts, skeletal muscles are now considered as important endocrine tissues that secrete various bioactive molecules, called myokines, that act in an auto-, para-, and endocrine manner to regulate the physiological function of target tissues and contribute to the progression of diverse diseases 28,29 . Rehabilitation training is speculated to protect target organs against disorders through the regulation of myokines homoeostasis, but further studies are needed to explore the implications and underlying mechanisms of myokines as related to the bene cial effects of rehabilitation therapy on severe and critical COVID-19 patients.
Despite this study's numerous strengths, we do have several limitations deserving consideration. Firstly, the sample size was relatively small with few non-survivor cases, limiting our sub-group or regression analysis. Secondly, as COVID-19 is a world public health emergency and there are insu cient clinical experience on rehabilitation therapy for severe COVID-19 cases, we varied the therapy's intensity for both the intervention and control groups, according to patients' medical stability and tolerance. Thus, we are currently conducting a retrospective study, rather than an RCT study. Thirdly, subsequent at-home rehabilitation training may have potentially biased our results and affected the outcomes.

Conclusion
In this study, we focused on implementing an early rehabilitation therapy for severely and critically ill patients with COVID-19. Early rehabilitation increased the duration of ventilator-free days during hospitalization, with no increased adverse events and complications. Furthermore, early rehabilitation enhanced pulmonary function abnormality recovery, accelerated the process of returning to independent functional status, decreased ICU-AW incidences, and increased muscle strength and exercise capacity at hospital discharge and one month after hospital discharge. The improvements in lung diffusing capacity, muscle strength, life quality, and exercise capacity remained signi cant at the three-month follow-up. Further, double-blind RCT and long-term studies with larger sample sizes are required to address the full impact of early rehabilitation therapy on COVID-19 patients.  Pulmonary function differences between the intervention and control groups during the three-month follow-up after hospital discharge. Note.

Supplementary Files
This is a list of supplementary les associated with this preprint. Click to download. supplementaryTable12.docx