Coronavirus disease 2019 (COVID-19) poses a severe burden on survivors, with clinically relevant nutritional and functional impairments. At hospital discharge, more than 70% of patients who survive critical COVID-19 report problems in mobility and in conducting their usual activities (1). At 6 months after hospital discharge, a significant proportion of COVID-19 survivors have impaired functional status (2) and persistent symptoms including dyspnea, fatigue, and muscle weakness (3). Patients with more severe illness during the acute phase are those with more troublesome sequelae (4). However, even patients with mild disease experience persisting symptoms after SARS-CoV-2 infection (5, 6). Few studies have investigated factors associated with symptom persistence in the long-term. Frailty, multiple symptoms at disease onset, female sex, endothelial dysfunction, overweight/obesity, and pre-existing comorbidities have been suggested as potential predictors (4, 7–9). Reduced muscle mass and quality might also contribute to persistent functional impairment. In non-COVID-19 critically ill patients, pre-existing low muscle mass independently predicts poor functional status at 12 months after intra−abdominal sepsis (10). Muscle quality is even more related to muscle function or strength as compared with muscle mass (11). The term muscle quality generally refers to the ability of skeletal muscle to effectively perform its functions, the most prominent being force production. Muscle quality depends on several determinants, including fiber type and distribution, muscle architecture, neuromuscular activation, the amount of fibrous tissue and the extent of fat infiltration, i.e., myosteatosis (12). The latter has emerged as an important biomarker and a priority research focus, being an independent risk factor for metabolic dysfunction, hip fractures, disability, hospitalization, mortality, and poor outcomes in severe illness and surgery (13, 14). Computed tomography (CT)-derived skeletal muscle radiodensity, also known as skeletal muscle radiation attenuation (SM-RA), is a reliable marker of myosteatosis (15). Low CT-derived SM-RA, which indicates myosteatosis, is independently associated with higher 6-month mortality in critically ill patients (16). In surgical patients, low SM-RA predicts postoperative complications or mortality (17–21), being a more reliable predictor than muscle size (20, 21). Furthermore, low trunk SM-RA is strongly associated with reduced functional capacity in both inpatients and outpatients (22–24).

Recent evidence indicates that myosteatosis was diagnosed by opportunistic CT-derived SM-RA associates with poor short-term outcomes in COVID-19 patients (25–28). It is not known whether poor muscle quality, as reflected by the presence of myosteatosis, influences clinical outcomes of COVID-19 survivors in the medium term. Therefore, the aim of this study was to assess the relationship between myosteatosis detected during acute COVID-19 and patient reported outcomes at 6 months after discharge.

This is the first analysis exploring the association of myosteatosis with patient-reported outcomes in COVID-19 survivors 6 months after discharge. We showed that myosteatosis, as defined by CT-derived SM-RA during acute COVID-19, significantly predicts persistent dyspnea and mobility problems at 6 months. Our results are consistent with those by van Gassel et al., who reported that myosteatosis, rather than muscle mass, was associated with functional impairment at 3 months after discharge in patients who survived ICU admission for COVID-19 (41). These findings, although novel in the context of COVID-19, are not surprising. In fact, fat infiltration may alter muscle architecture and affect muscle function and strength, increasing the risk of mobility problems (12). In addition, ectopic fat deposition in muscle leads to lipotoxicity, which may contribute to the chronic inflammation associated with obesity (42) and to loss of muscle function and mass (43). Sarcopenia, i.e., reduced muscle function and mass, is associated with reduced pulmonary function in community-dwelling adults (44, 45). Low CT-derived trunk SM-RA is associated with worse physical function and dyspnea in patients with non-small cell lung carcinoma (46), with greater severity of thoracic kyphosis (47), which in turn may impair ventilatory mechanics, and with higher postural sway and worse muscle strength and physical function in community-dwelling adults (24, 48). Although SM-RA may differ depending on the muscle considered (49), the finding that myosteatosis in trunk muscles was associated with impaired physical function in previous studies suggests that SM-RA at this site reflects myosteatosis in other muscles, and is consistent with the association between myosteatosis and mobility problems observed in our cohort.

We found that the prevalence of obesity upon admission was numerically greater in patients with vs. those without myosteatosis (45.2 vs. 25.8%, p = 0.056), this difference reaching statistical significance at the 6-month follow-up (48.4 vs. 27.7%, p = 0.046). Abdominal obesity was also more prevalent among patients with baseline myosteatosis (80.0 vs. 47.6%, p = 0.003). Myosteatosis is strongly associated with obesity, and often precedes the development of type 2 diabetes and the metabolic syndrome (50). Persistent inflammation from both obesity and prior SARS-CoV-2 infection may contribute to the long-term sequelae of COVID-19 (51). Of note, however, the association of myosteatosis with dyspnea and mobility problems was independent of BMI, suggesting that fat infiltration of skeletal muscle per se contributes to these symptoms. Pre-existing myosteatosis may influence long-term outcomes of COVID-19 survivors causing alterations in muscle architecture and lipotoxicity that may impair muscle function. As in a vicious cycle, pre-existing myosteatosis may lead to further worsening of muscle quality. It has been reported that patients with myosteatosis are at increased risk of muscle loss (52). In the setting of COVID-19 this is highly relevant, as acute sarcopenia, i.e., a decline in muscle mass and muscle function within 28 days of a significant physiological stressor event such as an acute illness (53), is emerging as a clinically relevant consequence of SARS-CoV-2 infection (54–56). This may be particularly true in COVID-19 patients more at risk of myosteatosis, i.e., those with overweight or obesity, who exhibit rapid and wide fluctuations in weight that may worsen body composition (57). Factors contributing to muscle wasting in COVID-19 include direct myotoxicity by SARS-CoV-2, systemic inflammation, bed rest, decreased food intake and myotoxic medications (58). Although we did not assess the effect of COVID-19 on muscle quality, we speculate that patients with prior myosteatosis might be more prone to acute sarcopenia, which in turn might contribute to poorer short-term outcomes and slower functional recovery (41, 59–61).

In contrast with other authors who investigated the relationship between opportunistic chest CT-derived SM-RA and short-term outcomes in COVID-19 (28, 62), we did not find an association between myosteatosis in paraspinal muscles and disease severity as reflected by the need for intensive care, which was similar between patients with or without myosteatosis in our cohort. Giraudo et al. have reported that lower CT-derived SM-RA (< 30 HU) predicts ICU admission in COVID-19 patients (62). However, despite patients with lower SM-RA being significantly older, analyses were not adjusted by age. Age (especially > 60 years) has a dramatic effect on COVID-19 severity (63), as well as on myosteatosis, paraspinal muscles being affected more than other muscles (64). It is difficult to argue that myosteatosis increases the risk of ICU based on the analysis by Giraudo et al., the increase in risk being probably due to age differences. The fat content of skeletal muscle is also affected by sex, with women exhibiting lower SM-RA (greater myosteatosis) than men (49, 64, 65). The use of sex- and age-specific cut-offs for defining myosteatosis is a strength of our analysis. Yi et al. found significant associations between measurements of myosteatosis including SM-RA and transition from mild to severe COVID-19 (28). Differences with our findings may be due to study-specific cut-offs for defining myosteatosis, and different patient characteristics (e.g., older age of our cohort). Other authors investigated the relationship between opportunistic chest CT-derived SM-RA and survival using adjustment for several potential confounders including sex and age, reporting significant associations between thoracic myosteatosis and reduced survival in COVID-19 patients (25, 26).

We also calculated pSMI, which is a biomarker of muscle mass (36, 37). In agreement with previous studies showing that CT-derived muscle density is more strongly associated with muscle function than muscle size (24), even in COVID-19 survivors (41), we found that SM-RA, but not pSMI, was associated with dyspnea and mobility problems at 6 months after discharge. This finding highlights the need of assessing muscle quality, in addition to muscle mass.

We must acknowledge some limitations. The study population consisted of patients who visited our COVID-19 outpatient clinic for the 6-month follow-up, which is prone to selection bias and limited generalizability. Direct CT quantification of muscle fat infiltration, which is an additional marker of muscle quality (66), was not available, and patients did not undergo magnetic resonance imaging (MRI) nor measurement of muscle strength for deeper characterization of muscle quality and function. It should be acknowledged, however, that SM-RA is widely recognized as a biomarker of fat infiltration; it is strongly correlated with muscle fat infiltration measured by MRI-derived proton density fat fraction, and is strongly associated with muscle function (14, 67). The lack of a reference measurement and the relatively small sample size of our cohort prevented us from providing SM-RA cut-offs to univocally diagnose myosteatosis in clinical practice. However, the purpose of our analysis was to describe the relationship between myosteatosis and post-discharge outcomes in COVID-19 survivors. Providing cut-off values for the diagnosis of myosteatosis would require a much larger cohort representative of the general population, and the inclusion of healthy subjects.

In conclusion, we showed that myosteatosis detected during acute COVID-19 significantly predicts persistent dyspnea and mobility problems at 6 months after hospital discharge. SM-RA can be directly derived from PACS during imaging reporting, and could be easily applied in clinical practice. Given the potential relevance of opportunistic CT-derived SM-RA in detecting patients at risk of persistent functional impairment after COVID-19, larger studies are needed to identify univocal sex- and age- specific cut-offs for diagnosing myosteatosis and promptly implement interventions aimed at improving muscle health and body composition in COVID-19 survivors, to prevent or minimize the long-term sequelae of the disease.

The raw data supporting the conclusions of this article will be made available by the authors upon reasonable request.

The studies involving human participants were reviewed and approved by the Comitato Etico IRCCS Ospedale San Raffaele. The patients/participants provided their written informed consent to participate in this study.

CC, AE, MM, and PR-Q: conception and design of the work. RD, AP, CG, VN, RL, DV, EF, MF, MC, CM, SM, and GV: acquisition of data. CC and RD: analysis of data. CC, AE, MM, PR-Q, AM, RD, and AP: interpretation of data. All authors have approved the submitted version of the manuscript, and have agreed to be personally accountable for the content of the work.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

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.