Design and Data Source
The retrospective case series contains data from post-acute COVID-19 patients who were admitted to a standardized three-week pulmonary rehabilitation program at the Clinic for Rehabilitation in Münster, Austria. They were admitted between the 1st of March 2020 and 31st of May 2021, due to a laboratory confirmed SARS-CoV-2 infection prior to rehabilitation, according to the definition of the Austrian Federal Ministry of Social Affairs & Health Care. Initially, data of all eligible patients who underwent rehabilitation in this time frame were screened by a physician. Before data evaluation, data were pseudonymized and then extracted from the clinic information system (MP2 IT-Solutions, Austria). Data pseudonymization and extraction were carried out by one physician and one research assistant. Data privacy was guaranteed by an in-house data protection agreement made by a commissioner for data protection. Steps of the retrospective data analyses are shown in Figure 1. The research ethics committee of the Medical University of Innsbruck approved the study protocol (1066/2021) and the study was registered at the German Clinical Trials Register (ID: DRKS00026936).
Characteristics of Patients’ Data
Records from post-acute COVID-19 patients with the principal diagnosis ICD U.07.1 (COVID-19) were analyzed. Anthropometric data as well as secondary diagnoses which were present before the COVID-19 infection were included in the analyses. Secondary diagnoses included cardiovascular and cerebrovascular diseases, chronic kidney diseases, obstructive pulmonary disease (COPD), bronchial asthma, as well as diabetes and hypertension. These diagnoses were documented by the treating physician. (Pre-) diabetes was defined by an elevated hemoglobin A1c value or prescribed anti-diabetic medication according to the International Diabetes Federation (IDF) . Hypertension was defined by > 130/80 mmHg or prescribed antihypertensive medication according to the International Society of Hypertension Guidelines . Patients were admitted to pulmonary rehabilitation as soon as they were physically stable without the need of continuous supervision, invasive or non-invasive ventilation. They could be admitted after being tested negative twice by real-time polymerase chain reaction via swab. If patients terminated their stay before completing the three-week program or if admission and discharge measurements were incomplete, their data were excluded from analyses (see Figure 1).
After inclusion, patient’s data were categorized according to Huang’s COVID-19 severity scales (Huang, 2021):
Scale 1: not admitted to hospital before rehabilitation stay with resumption of normal activities
Scale 2: not admitted to hospital before rehabilitation stay, but unable to resume normal activities
Scale 3: admitted to hospital before rehabilitation stay and not requiring supplemental oxygen
Scale 4: admitted to hospital before rehabilitation stay, but requiring supplemental oxygen
Scale 5: admitted to hospital before rehabilitation stay requiring high flow nasal cannula (HFNC), non-invasive mechanical ventilation (NIV) or both
Scale 6: admitted to hospital before rehabilitation stay requiring extracorporeal membrane oxygenation, invasive mechanical ventilation (IMV) or both
Scale 7: death (not applicable)
Included Measurement Data
After their admission to rehabilitation, patients were assessed following a standardized clinical routine. As part of this clinical routine, the Six-Minute Walk Test (6MWT) and pulmonary function testing were carried out at the beginning and at the end of the three-week rehabilitation stay. The 6MWT is a well-documented standardized assessment used to assess walking endurance and functional exercise capacity and has been used to assess the response to medical interventions in diverse patient groups . It was executed by an experienced and well-instructed physiotherapy staff member according to the guidelines of the American Thoracic Society (ATS) . 6MWT outcome values were compared to 6MWT reference values for recovered healthy adults according to Enright et al., . The corresponding reference values for each participant were calculated according to reference equations for men: 6MWD = (7.57 x height(cm) – (5.02 x age(years)) – (1.76 x weight(kg)) – 309m; for women: 6MWD = (2.11 x height(cm)) – 2.29 x weight(kg)) – (5.78 x age(years)) + 667m . The difference in pre and post measurements of the 6MWT were compared to the minimal clinically important difference across multiple patient groups  and to reference values for patients suffering from acute respiratory distress syndrome or having survived acute respiratory distress syndrome . After the 6MWT, maximal inspiratory capacity (ICmax) was measured using a manometer connected to a PEP-RMT-System (Positive Expiratory Pressure- Respiratory Muscle Training- System) (Mediplast, Malmö, Sweden). Further pulmonary functions were tested using body-plethysmography (Master Screen Body, Traeger GmbH, Hoechberg, Germany). Measurements were carried out by an experienced physician according to recent updated guidelines by the ATS  and the European Respiratory Society (ERS) [29, 30].
Primary outcomes included the 6MWT, ICmax measured by the PEP-RMT-System as well as Forced Vital Capacity (FVC) and Forced Expiratory Volume in the first second (FEV1) assessed by body-plethysmography. FVC and FEV1 were compared to calculated reference values of body-plethysmography. Secondary outcomes included the number and type of exercise therapy sessions throughout the patients’ rehabilitation visit.
Exercise Therapy Interventions
All post-acute COVID-19 patients admitted to pulmonary rehabilitation followed a standardized program with a duration of at least three weeks, including exercise therapy sessions on 5-6 weekdays. Each week, patients participated in a maximum of 3 exercise therapy sessions per day (Monday to Friday). The exercise therapy sessions consisted of individual respiratory muscle training, pulmonary group exercises, individual strength exercises (3 to 5 exercises for large muscle groups in three series of 8 to 12 repetitions per exercise with or without weight machines), individual endurance training (cycling, treadmill, in and outdoor walking) and relaxation group exercises. Intensities and intervals of endurance training were based on the results of the 6MWT performance. For respiratory muscle training, a hand-held resistance device was used (PEP-RMT-System, Mediplast, Malmö, Sweden) for 3 sets of 10 breaths each and a 1-minute rest between sets. Each exercise session lasted for 30-45 minutes and was supervised by a sport scientist or a physiotherapist. The amount and type of the group exercise therapy sessions and the amount of individual physiotherapy sessions were determined by the physicians in charge.
Descriptive characteristics and secondary diagnoses of males and females were presented as mean with standard deviation or percentages. Spearman’s rank correlations between COVID-19 severity and secondary diagnoses were calculated, as COVID-19 severity was categorized by the ordinally scaled COVID-19 Severity Scale (Huang et al., 2021). Spearman’s rank correlations were also calculated between the number of respiratory muscle training sessions and lung function parameters (FEV1, FVC and ICmax) as well as the 6MWT. For the comparison of functional exercise capacity (6MWT) and lung function parameters (FEV1, FVC and ICmax) by sex, Welch-ANOVA was used, as results of Levene's test suggested significant heteroscedasticity regarding the investigated parameters (p > 0.05). When comparing post-treatment 6MWT, FEV1 and FCV to corresponding reference values, paired t-tests were used.