Inadequate Physical Activity Is Associated with Worse Physical Function in a Sample of COVID-19 Survivors with Post-Acute Symptoms

Background—Patients affected by Long COVID often report shorter times spent doing physical activity (PA) after COVID-19. The aim of the present study was to evaluate potential associations between PA levels and parameters of physical function in a cohort of COVID-19 survivors with post-acute symptoms, with a particular focus on individuals aged 65 and older. Materials and methods—PA levels before and after COVID-19 were assessed in a sample of patients that had recovered from COVID-19 and were admitted to a post-acute outpatient service at the Fondazione Policlinico Universitario Agostino Gemelli IRCCS (Rome, Italy). Participation in PA was operationalized as the engagement in leisure-time PA for at least 150 min per week in the last 3 months. Self-rated health and measures of physical performance and muscle strength were assessed. Results—Mean age of 1846 participants was 55.2 ± 14.4 years and 47% were women. Before COVID-19, inactivity was detected in 47% of the whole study population; only 28% maintained pre-COVID-19 PA engagement. Inactivity was more frequent in women. The stopping of physical activity was associated with increased BMI and CRP levels, lower vitamin D levels and a higher prevalence of post-COVID-19 fatigue, dyspnea, arthralgia, and myalgia. Active participants had higher handgrip strength and performed better on both the six-minute walking test (6MWT) and at the one-minute sit-to-stand test (1MSTST). In particular, at the 6MWT, participants 65 and older that were still active after COVID-19 walked 32 m more than sedentary peers. Moreover, the distance covered was 28 m more than those who were active only before COVID-19 (p = 0.05). Formerly active subjects performed similarly at the 6MWT to inactive participants. PA was associated with better self-rated health. Conclusions—Our findings reveal that inactivity is frequent in the post-acute COVID-19 phase. Stopping physical activity after COVID-19 results in measures of performance that are comparable to those who were never active. Relevant differences in the distance covered at the 6MWT were found between older active subjects and their sedentary peers.


Introduction
Physical activity (PA) represents one of the major determinants of health worldwide [1]. Starting regular aerobic PA reduces the incidence of several chronic conditions such as cardiovascular disease (CVD), type 2 diabetes (T2DM), depression, and dementia [2,3]. Regardless of its intensity, any type of PA that reduces sedentary time may decrease the risk of premature mortality of young and older adults [4]. Even a reduction of 10% of inactivity can prevent more than 500,000 deaths every year [1]. Moreover, following an acute injury, one of the strategies to facilitate the process of recovery is based on muscle training [5][6][7].
PA plays a role in modulating the immune system, and the exercise-induced antiinflammatory response seems to be one of the mechanisms implicated in the protective role of PA against several conditions, including infectious diseases [8]. A recent metaanalysis has shown that regular PA is associated with a one-third reduction in the risk of community-acquired infectious diseases [9]. Furthermore, the rate of upper respiratory tract infections tends to be lower in subjects who perform aerobic exercises for a greater number of days per week [10]. Concerning viral infections, it seems that regular PA may decrease the risk of influenza-associated mortality [11].
During the COVID-19 pandemic, measures taken by governments to control the SARS-CoV-2 transmission have changed the lifestyle habits of millions of people worldwide [12]. If compared to the pre-pandemic era, people experienced a reduction in PA levels as a result of lockdown restrictions [13][14][15]. Sedentary habits are associated with a higher risk of hospitalization, admission to the intensive care unit (ICU), and mortality due to COVID-19 [16]. On the other hand, regular PA may decrease the risk of SARS-CoV-2 infection and severe COVID-19 symptoms [3,17]. Even in the acute phase, moderate-intensity aerobic exercises might reduce the severity and progression of COVID-19 [18].
Little is known about the role of PA in the post-acute phase of COVID-19. Patients affected by Long COVID, which is characterized by long-lasting symptoms [19], often report a shorter time spent doing PA compared to pre-COVID-19 [20]. Reduced walking and exercise tolerance are among the most common symptoms in those that have been hospitalized for COVID-19 [21]. After being discharged from the hospital, COVID-19 survivors appear to have worse physical performance, particularly in the first few months after the infection [22].
The aim of the present study was to examine the prevalence of physical activity levels before and after the viral infection in a cohort of hospitalized or home-managed COVID-19 survivors with post-acute symptoms, and its association with measures of physical performance, muscle strength, and self-rated health.

Materials and Methods
The Gemelli Against COVID-19 Post-Acute Care (GAC19-PAC) project is an ongoing initiative developed by the Department of Geriatrics, Neuroscience, and Orthopedics of the Catholic University of the Sacred Heart (Rome, Italy) to analyze long-term sequalae of COVID-19 and their impact on overall health, quality of life, and physical/cognitive performance. The post-acute outpatient service, called "Day Hospital post-COVID-19", has been established in April 2020 at the Fondazione Policlinico Universitario Agostino Gemelli IRCCS (Rome, Italy) for people who have recovered from COVID-19 [23].

Study Sample
The study population included individuals aged > 18 years old that have officially recovered from COVID-19 and have been admitted to the post-COVID-19 outpatient service between 21 April 2020 and 31 December 2021. At the moment of the evaluation, all patients met the World Health Organization (WHO) criteria for discontinuation of quarantine [24].

Data Collection
Patients were offered a comprehensive medical assessment. Clinical parameters, anthropometric measures, medical history, current medications, and lifestyle habits including physical activity, were collected in a structured electronic database. Long-term symptoms of SARS-CoV-2 infection were evaluated with a multidisciplinary approach [25][26][27][28][29].
Body weight was measured through an analogue medical scale. Body height was measured using a standard stadiometer. Body mass index (BMI) was defined as weight (kilograms) divided by the square of height (meters).
COVID-19 severity was considered as follows: (1) no hospitalization; (2) hospitalization with no oxygen supplementation; (3) hospitalization with low-flow oxygen supplementation; (4) hospitalization with noninvasive ventilation (NIV) or ICU admission with invasive ventilation. Symptoms that arose during or immediately after the SARS-CoV-2 infection and persisted at the time of the evaluation have been included among sequalae of acute COVID-19.

Physical Activity Levels
All participants were interviewed concerning their usual physical activity levels prior to acute SARS-CoV-2 infection and at the time of the evaluation. Regular participation in physical activity was operationalized as the engagement in aerobic PA, associated or not with resistance training, for a minimum of 150 min per week in the last 3 months [30]. Consequently, the subjects were divided into three groups: (1) inactive, not physically active before COVID-19 and at the time of the evaluation; (2) formerly active, subjects who practiced regular PA only before COVID-19 and not at the time of the evaluation; (3) active, subjects who practiced regular PA both before and after COVID-19.

Hematological Parameters
Vitamin D, plasma albumin, C-reactive protein (CRP), and hemoglobin levels were detected using standard biochemistry methods on fully automated testing systems.

Assessment of Muscle Strength and Physical Performance
Upper extremity muscle strength was measured by handgrip strength testing using a North Coast hand-held hydraulic dynamometer (North Coast Medical, Inc., Morgan Hill, CA, USA) [31]. After one familiarization trial, muscle strength was measured in both hands and the higher value (kg) was used for the analysis [31].
Physical performance was evaluated by the one-minute sit-to-stand test (1MSTST) and the six-minute walking test (6MWT). These tests are commonly used to assess exerciseinduced respiratory impairment [32]. For the 1MSTST, participants must stand up from a chair and sit down for 60 s as quickly as possible. A higher number of rises at the 1MSTST and a longer distance covered at the 6MWT (m) indicate better performance [33,34].

Self-Rated Health
A visual analog scale (VAS) was used to rapidly assess self-rated health on a scale from 0 to 100, with 0 corresponding to the worst perception of health status and 100 indicating the best imaginable health [35].

Statistical Analyses
Continuous variables were expressed as mean ± standard deviation (SD), and categorical variables were expressed as frequencies by absolute value and percentage (%) of the total. Descriptive statistics were used to describe clinical characteristics of the study population according to physical activity levels. Differences in the proportions and means of covariates between study participants across different physical activity categories were assessed using Fisher's Exact Test and t test statistics (p for trend), respectively.
Analysis of covariance (ANCOVA) was used to examine the association between physical activity levels and measures of physical performance, muscle strength, and self-rated health, after adjusting for potential confounding variables. Variables considered as covariates were chosen according to their clinical significance and/or their significant difference at the univariate analyses between inactive, formerly active, and active participants. The interactions of covariates on the relationship between physical function's measures and physical activity levels were tested by adding the interaction term into the adjusted models. No significant interactions were reported.

Results
Between April 2020 and December 2021, 1921 individuals were admitted to the post-COVID-19 outpatient service. For the present study, 75 subjects were excluded for missing values in the variables of interest; thus, a sample of 1846 subjects was considered. The mean age of the study sample was 55.2 ± 14.4 years (range 18-86) and 47% were women. Characteristics of the study population according to physical activity levels are summarized in Table 1. Inactive people were 47% of the whole study population; 53% were physically active before COVID-19, but only 28% maintained usual physical activity levels at the moment of the evaluation (Figure 1). In particular, only 23% of females continued to exercise after COVID-19, while the prevalence in the male subgroup was about 32%. The proportion of participants who practiced PA at the time of the evaluation was similar across ages ( Figure 1). J. Clin. Med. 2023, 12, x FOR PEER REVIEW 6 of 14 C-reactive protein 2.9 ± 6.0 3.5 ± 6.6 2.8 ± 5.7 2.1 ± 5.3 <0.001 Inactive people were 47% of the whole study population; 53% were physically active before COVID-19, but only 28% maintained usual physical activity levels at the moment of the evaluation (Figure 1). In particular, only 23% of females continued to exercise after COVID-19, while the prevalence in the male subgroup was about 32%. The proportion of participants who practiced PA at the time of the evaluation was similar across ages ( Figure  1).
The average number of days from COVID-19 diagnosis to admission to the outpatient service were similar between inactive and active subjects (85.2 + 42.7 versus 91.3 + 34.8, respectively; p = 0.1). COVID-19 severity had the same distribution in the three groups of interest (inactive, formerly active, and active), except for those who needed invasive ventilation, in which the percentage of inactive subjects was higher (p = 0.01). Compared to active participants, subjects in the inactive and formerly active groups were older and had a higher prevalence of hypertension, diabetes, and chronic obstructive pulmonary disease (COPD). The highest prevalence of unemployed individuals was found in the active group, while retired people were more common in the other two groups. Active and formerly active groups presented the same percentage of people occupied as housewives and declared a similar number of kids. Post-acute COVID-19 fatigue and dyspnea were less common amongst active individuals. Persistent arthralgia and myalgia were more common in formerly active participants. Formerly active and active subjects, especially those who maintained usual levels of PA after COVID-19, showed lower BMI and C-reactive protein (CRP) levels, and higher vitamin D levels. Unadjusted and adjusted ANCOVA models were performed to evaluate the association of physical activity levels with measures of physical performance, muscle strength, and self-rated health (Tables 2 and 3). In the whole sample, practicing PA at the moment of the evaluation was associated with a longer distance covered during the 6MWT, greater numbers of repetitions at the 1MSTST, and higher values at the handgrip strength. Active The average number of days from COVID-19 diagnosis to admission to the outpatient service were similar between inactive and active subjects (85.2 ± 42.7 versus 91.3 ± 34.8, respectively; p = 0.1). COVID-19 severity had the same distribution in the three groups of interest (inactive, formerly active, and active), except for those who needed invasive ventilation, in which the percentage of inactive subjects was higher (p = 0.01). Compared to active participants, subjects in the inactive and formerly active groups were older and had a higher prevalence of hypertension, diabetes, and chronic obstructive pulmonary disease (COPD). The highest prevalence of unemployed individuals was found in the active group, while retired people were more common in the other two groups. Active and formerly active groups presented the same percentage of people occupied as housewives and declared a similar number of kids. Post-acute COVID-19 fatigue and dyspnea were less common amongst active individuals. Persistent arthralgia and myalgia were more common in formerly active participants. Formerly active and active subjects, especially those who maintained usual levels of PA after COVID-19, showed lower BMI and C-reactive protein (CRP) levels, and higher vitamin D levels.
Unadjusted and adjusted ANCOVA models were performed to evaluate the association of physical activity levels with measures of physical performance, muscle strength, and self-rated health (Tables 2 and 3). In the whole sample, practicing PA at the moment of the evaluation was associated with a longer distance covered during the 6MWT, greater numbers of repetitions at the 1MSTST, and higher values at the handgrip strength. Active participants 65 and older walked, on average, 28 m more than formerly active groups and 32 m more than sedentary peers (457.1 m vs. 428.8 m vs. 425.0 m, respectively; p = 0.05). Sedentary habits of inactive and formerly active individuals were associated with a similar distance covered at the 6MWT. Older people who were still active reported better scores of self-rated health compared to all inactive subjects.

Discussion
In the present study, we investigated the prevalence of PA in post-acute COVID-19 patients and its potential association with functional parameters. The proportion of inactive subjects in our cohort of COVID-19 survivors with persistent symptoms was higher compared to the healthy Italian population before the pandemic (72% vs. 36.8%, respectively) [36]. Following lockdown restrictions, even those who did not test positive to SARS-CoV-2 swab experienced a strong reduction in PA levels [36]. In our study, participants who stopped practicing PA represented almost 25% of the sample. These findings are consistent with those shown by a recent Dutch-Belgian survey, in which people who had not returned to their usual PA after COVID-19 were approximately 24.7% [20].
We found that inactivity after COVID-19 was more frequent amongst women, although these results were not explained with major engagement in job and/or home responsibilities (numbers of kids, work as a housewife). Most likely, women experience post-acute COVID-19 symptoms more than men [21,37]. Frequently reported Long COVID-19 symptoms are fatigue, dyspnea, and joint-muscle pain [38][39][40], which are associated with a reduced tolerance to exercise [41][42][43] that may limit physical activity [21,38,44].
We found no substantial differences in the prevalence of inactivity after COVID-19 according to age. Lombardo et al. have shown that the incidence of post-COVID-19 fatigue, dyspnea, and joint-muscle pain is lower at young ages (<47 years old), but it tends to be similar among adults (47-58 years old) and older adults (>58) [45].
Our findings revealed a positive trend in BMI and CRP levels from sedentary to active subjects. It is noteworthy that-as a consequence of unhealthy dietary habits and reduction in PA-many adults gained weight during lockdown restrictions [46]. Moreover, evidences have demonstrated an association between PA and inflammatory mediators' levels [47,48] that may be attributed to the reduction of visceral fat mass [49,50], showing that PA is involved in maintaining the balance between pro-inflammatory and anti-inflammatory responses [51].
Vitamin D levels were higher among physically active participants, particularly in those who did not stop practicing physical activity after COVID-19. Consistent with our findings, a correlation between normal vitamin D levels and habitually practicing PA has been identified in a cohort of 86 forensic inpatients [52]. However, evidence is contradictory with other studies showing no relationships [53][54][55][56]. Similar contrasting results involve the relationship between vitamin D levels and Long COVID symptoms [29,57], thus further research is warranted.
In the present investigation, we found that PA was associated with better physical performance and higher values of muscle strength. Differences in the distance covered during the 6MWT were found particularly in the older subgroup, in which active participants walked 32 m more than inactive peers and 28 m more than formerly active subjects. According to Bohannon and Crouch, a minimum change of 30.5 m in the distance covered at the 6MWT may be considered clinically meaningful [58]. These results are in line with several studies demonstrating that both aerobic and resistance training might improve physical function, protecting the body from physical frailty and maintaining independence later in life [59]. Moreover, we found that individuals who stopped practicing PA after COVID-19 walked a similar distance in the 6MWT to sedentary subjects, and more frequently manifested post-acute COVID-19 symptoms compared to active subjects. Paneroni et al. have demonstrated that patients with any previous disabilities but recent hospitalization for COVID-19 pneumonia, exhibit lower values of bicep strength and smaller numbers of rises at the 1MSTST (15 kg and 22.1, respectively) [60]. Notably, sarcopenia can be detected in a relevant proportion of people who have recovered from the SARS-CoV-2 infection [61]. Due to the complexity of COVID-19 and its implications, we cannot rule out that a combination of both long-term symptoms and less involvement in PA may influence muscle function in the first months after the infection [21,22,61].
In our sample, physically active participants declared better self-rated health. Older subjects who maintained usual PA levels after COVID-19 manifested better scores on their quality of life than sedentary participants. Several RCTs found that changing the PA regimen may reduce psychological distress and improve the quality of life among community-dwelling older adults [62,63]. Considering that government restrictions during the pandemic have drastically worsened the lifestyle habits and mobility of older people [64], practicing PA or recovering usual PA levels may represent a solution to ameliorate psychological well being.
Maintaining usual PA levels after the infection is crucial. Lockdown restrictions and the fear of contagion have influenced choices about the well being of millions of people worldwide. The large cohort of Long COVID patients enrolled, and the attempt to emphasize the potential role of PA and persistent symptoms, including fatigue, in determining muscle function in the post-acute phase of the infection represent the strengths of this study.
Some methodological issues should be taken into account in the interpretation of the results. A major limitation of cross-sectional studies is the lack of longitudinal data that capture changes on an individual level over time, thus not allowing a thorough followup of post-COVID-19 symptoms and PA levels. The observational nature of our study design prevents the evaluation of cause-effect relationship between sedentary habits and poor physical function in patients with Long COVID. Therefore, we cannot assert whether persistent symptoms influence physical performance more than PA levels.
Furthermore, information about the type, volume, and intensity of activity practiced before COVID-19 and/or maintained after COVID-19 were not assessed. Data on dietary habits and physical function before COVID-19 that may have influenced patients' health status and physical function after the infection were not available. Moreover, our sample included predominantly symptomatic patients attending a post-COVID-19 outpatient service, which may induce a selection bias. Indeed, we cannot determine if further factors might be implicated in the deterioration of physical function after COVID-19. The involvement of different anatomical districts that follows the SARS-CoV-2 infection shows us the complexity of this illness and the difficulty in understanding the pathogenesis of long-term sequalae. Finally, the long period of enrollment does not allow us to distinguish whether specific SARS-CoV-2 variants might have influenced post-COVID health status more than others, although following the spread of variants in Italy during this period of time, our sample is unlikely to contain individuals infected with the omicron variant, which is likely the variant associated with a fewer number of persistent symptoms [65].
In conclusion, we analyzed the prevalence of physical activity levels and its association with parameters of physical function in a sample of COVID-19 survivors with post-acute symptoms. We found a high prevalence of inactivity after COVID-19, especially in women. Lower BMI and CRP levels, and higher vitamin D levels were more common in active people. Post-acute COVID-19 symptoms were less frequent in active people. Being physically active, particularly after COVID-19, was associated with better performance on the 6MWT, a higher number of repetitions at the 1MSTST, and higher values of handgrip muscle strength. Differences in the distance covered at the 6MWT were relevant between active and inactive older people. We found that formerly active subjects performed similarly to habitually sedentary subjects. Moreover, physical activity was associated with higher values of self-rated health. Future studies are warranted to further assess associations between physical activity levels and muscle function in the post-acute COVID-19 and to evaluate whether improving exercise training may ameliorate functional outcomes after the SARS-CoV-2 infection.