Relaxation Exercise in Patients With COVID-19

Study procedures A randomized controlled design was used. The study was double-blinded in which both patients and assessors were blinded to study procedures. Each patient was asked to sign an informed consent before participation in this study. This study was approved by the ethics committee of our university (blinded). Patients were recruited from December 2020 to April 2021. The patients were collected by reviewing the hospital's data (blinded). Then, we called patients with COVID-19 to ask them about their welling to participate in this study. Patients who accepted to participate were interviewed through video calls using Zoom App to discuss study procedures with them. Patients who accepted to engage in this study were visited by a researcher at their home while they are in quarantine. Through the initial visit, the consent form was signed, and blood samples were taken. Any patient must have permission to perform our intervention from the physician who supervises him/her. The clinical protocol of this study was retrospectively registered at https://www.clinicaltrials.gov/ platform with this registration number NCT04998708. The study was registered retrospectively because of the vulnerability of the research team to have COVID-19 which may affect the dates and data of the trial. There were no deviations from the registered trial protocol.

The suitable sample size for this study was determined using a priori power test. The G*POWER program (ver. 3.1.9.2, Heinrich-Heine-University, Düsseldorf, Germany) was used. The criteria entered into the program included a MANOVA test using 2 groups, 2 measurements, a power level of 80%, a significance level of .05, and medium effect size (dz= .25)(Faul et al., 2007). Based on the previous assumptions, the total sample size needed for this study was 30 patients. A minimum power of 80% or more is acceptable in most studies(Kadam \& Bhalerao, 2010). The flow of patients throughout the study is shown in Figure 1.

Initially, forty-five patients accepted the participation in this study. Ten patients did not meet the study inclusion criteria and additional five patients did not complete the whole study because they did not like doing exercise. Finally, thirty patients with COVID-19 completed the whole study (2-weeks intervention and 3-weeks post-intervention "follow-up"). The patients' age ranged from 25 to 45 years old. The inclusion criteria included that the patient has a recent mild or moderate COVID-19 with no or low-grade fever 99.5- 100.94 °F (37.5-38.3 °C) (Affronti et al., 2010; Zhuang et al., 2020). Mild grade COVID-19 was determined by having symptoms of acute upper respiratory tract infection (fever, cough, myalgia, runny nose, fatigue, sore throat, sneezing) or gastrointestinal symptoms (nausea, vomiting, abdominal pain, diarrhea). Moderate grade COVID 19 was determined by having mild symptoms plus pneumonia (cough, frequent fever) with no obvious hypoxemia, the presence of lesions on chest CT(Yuki et al., 2020). The exclusion criteria included that the patient is indicated for being hospitalized, has high-grade fever <100.94 °F (<38.3 °C), or other chronic diseases such as heart problems, hypertension, or diabetes. Women on contraceptives were excluded because contraceptives decrease immune function increase the risk of various autoimmune disorders(Williams, 2017).

Evaluative procedures There were three main dependent variables including blood immune markers, salivary immunoglobulin A, and Wisconsin Upper Respiratory Symptom Survey. These measurements were collected at the baseline, 2-weeks post-intervention, and 1-week as a follow-up.

A lab technician and one researcher were asked to visit the patient at home during the quarantine. The technician collected blood and saliva samples and the researcher conducted the Wisconsin Upper Respiratory Symptom Survey. They wore special protective equipment recommended by WHO(WHO, 2020b; World Health Organization (WHO), 2020). Three visits were performed including the beginning of the research procedures, 24-h post-intervention (2 weeks), and 1-week as a follow-up.

Blood sample collection Blood samples were taken in the morning (8:30-9:30) by collecting 10mL of venous blood. Patients were asked to stop any exercise for at least 24 hours before blood sampling. Also, patients were asked to stop eating or drinking 8 hours before collecting samples. Samples were collected in vacutainer tubes with sodium ethylenediami-netetraacetic acid (EDTA) for plasma separation. The blood was centrifuged at 3,000 rpm for 15 min at 4◦C. The total lymphocytes, leukocytes, and monocytes were measured utilizing a multichannel hemocyte analysis system (SE-9000; Sysmex Corp, Hyogo, Japan)(Lira et al., 2017; Shimizu et al., 2011). The concentrations of IL-6, IL-10, and TNF-α were analyzed by using ELISA commercial kits assay (R\&D Systems, Minneapolis, USA). The manufacturer's instructions for analysis of IL-6, IL-10, and TNF-α on an EZ-Reader microplate reader at 450 nm were followed(de Souza et al., 2018; Lira et al., 2017). The samples were stored at -20◦C for further analysis.

Saliva sample collection The saliva sample was taken without any saliva stimulation methods. The patient was asked to rinse their mouths with distilled water and evacuate their mouth just before the collection. We used the passive drainage method for this collection. The patient slightly flexed their head forward to allow the saliva to move into a sterilized and pre-weighed Falcon tube for 5 min. The weight of tubes was measured again following the collection to estimate the volume and the saliva flow rate. The tubes were weighed with 0.1mg accuracy with the proposed saliva density proposed at 1.0 g. mL-1. The samples were stored at -80◦C for further analysis. The S-IgA concentration was analyzed utilizing commercial ELISA kits (IgA Salivary, DRG, Minneapolis, USA). The IgA-S secretion rate (ng/min) was measured by multiplying the whole concentration of IgA-S present in the mucosal surface per unit of time by the saliva flow rate (mL/min) (de Souza et al., 2018).

Wisconsin Upper Respiratory Symptom Survey The Wisconsin Upper Respiratory Symptom Survey is an empirically derived patient-oriented illness-specific quality-of-life evaluative outcomes instrument(A. W. Barrett \& Barrington, 2005). The development process of this survey was described in detail by Barrett et al(B. Barrett et al., 2018). The Wisconsin Upper Respiratory Symptom Survey is designed to evaluate the negative effect of acute upper respiratory infection on the quality of life. It is a validated and reliable measurement for evaluating the change in the quality of life over time including influenza-like illness symptoms of headache, body aches, and fever(B. Barrett et al., 2018). The patients were asked to fill the survey at the baseline, 2-weeks post-intervention, and 1-week as a follow-up.

Treatment Procedures Patients were assigned randomly into intervention, and control groups. Patients were randomly allocated to 4 permuted blocks using computer software to have 2 equal sample sizes. The randomization was performed by a college staff member who was not involved in this study. All patients in the groups followed the WHO procedures of quarantine(WHO, 2020a) and took standardized medications administered by the Turkish Ministry of Health (Hydroxychloroquine Sulphate 200 Mg).

The intervention group performed an integrated intervention combining CBSM and PMRs. We used these two interventions because it has been shown their positive effects on immune function. The two relaxation techniques included progressive muscle relaxation training (PMR)(Xiao et al., 2020) and cognitive-behavioral stress management (CBSM)(McGregor et al., 2004). We used two types of relaxation techniques to produce maximum effects of relaxation techniques within the limited period of COVID-19 quarantine (2 weeks). The control groups did not receive any relaxation techniques during the study.

Progressive Muscle Relaxation Training Progressive muscle relaxation training has been reported to decrease anxiety, negative emotions, and enhance sleep quality in patients with COVID-19 (Özlü et al., 2021; Xiao et al., 2020). The patients performed PMR in their bed for 30 minutes in the morning after waking up and at night before sleeping. We recorded a detailed video for PMR to be watched by all patients before participating in this study. The PMR order was initiated by doing isometric contractions of muscles with a concentration on the sensation of tension for 3 to 5 sec. Then, they take a rest for 10 to 15 sec. Next, the patients were instructed to perform the PMR in this sequence foot, leg, hip and waist, chest, arm, shoulder, and face. The specific instructions for each exercise were demonstrated in the study of Özlem et al(Özlü et al., 2021). The patient should at least complete 12 days of PMRs to be included in the statistical analysis.

Cognitive-Behavioral Stress Management (CBSM) Cognitive-Behavioral Stress Management has been reported to significantly decrease depression, anxiety, and stress in patients with COVID-19 (Li et al., 2020). All included patients were taken a 2-h structured group session to illustrate the CBSM in detail before starting the study. All sessions were conducted by a hired psychologist who was blinded to study procedures. The CBSM consisted of a 2-h structured group session performed 2 times/week. Each session focused on various CBSM techniques (e.g., coping skills training, cognitive restructuring, assertion training, and stress management). The patients also were asked to do homework assignments (e.g., relaxation monitoring cards and a cognitive restructuring worksheet) given each week. The patient should attend all 2-h structured group sessions and do all assignments to be included in the statistical analysis.

Data analysis Patients' files were encoded by a university administrator who did not join this study(Taylor \& Murphy, 2010). Intention-to-treat and general linear models of analysis were followed. A multivariate analysis of repeated measurements (MANOVA) test was used to measure within--each group interactions, while an independent MANOVA test was used to measure between-groups comparisons. The outcome measures were measured at the baseline, 2-weeks post-intervention, and 1-week as a follow-up. All results were compared to the baseline results. The baseline characteristics of patients in both groups were compared using Pearson chi-squared tests for categorical variables (gender, previous lung infection history) and t-test for continuous variables (age and body mass index). The significance level was set at p<.05. The SPSS (ver. 25, IBM Inc., Armonk, NY, USA) was used for statistical analysis.