From heart to muscle: pathophysiological mechanisms underlying long-term physical sequelae from SARS-CoV-2 infection

Abstract

The long-term sequelae of the coronavirus disease 2019 (COVID-19) are multifaceted and, besides the lungs, impact other organs and tissues, even in cases of mild infection. Along with commonly reported symptoms such as fatigue and dyspnea, a significant proportion of those with prior COVID-19 infection also exhibit signs of cardiac damage, muscle weakness, and ultimately, poor exercise tolerance. This review provides an overview of evidence indicating cardiac impairments and persistent endothelial dysfunction in the peripheral vasculature of those previously infected with COVID-19, irrespective of the severity of the acute phase of illness. In addition, V̇o_2peak appears to be lower in convalescent patients, which may stem, in part, from alterations in O₂ transport such as impaired diffusional O₂ conductance. Together, the persistent multi-organ dysfunction induced by COVID-19 may set previously healthy individuals on a trajectory towards frailty and disease. Given the large proportion of individuals recovering from COVID-19, it is critically important to better understand the physical sequelae of COVID-19, the underlying biological mechanisms contributing to these outcomes, and the long-term effects on future disease risk. This review highlights relevant literature on the pathophysiology post-COVID-19 infection, gaps in the literature, and emphasizes the need for the development of evidence-based rehabilitation guidelines.

INTRODUCTION

The novel coronavirus [severe acute respiratory syndrome (SARS)-CoV-2], which causes a pathogenic condition known as coronavirus disease 2019 (COVID-19), was first identified in December 2019 from a cluster of patients with lower respiratory infections (1), and has since spread worldwide. The clinical spectrum of COVID-19 is wide ranging, encompassing asymptomatic infection, olfactory and/or gustatory dysfunctions, gastrointestinal symptoms, fatigue, fever, myalgia, mild upper respiratory tract illness, severe viral pneumonia with thrombotic complications, respiratory failure, and death (1, 2). Although these numbers are subject to change with the emergence of variants and vaccination campaigns, it is estimated that, with the original dominant strain of the virus, asymptomatic infections account for 4%–52% of cases (3). Among symptomatic individuals, ∼80% exhibit a “mild” disease course while others progress to a severe or critical form of the disease, which can ultimately lead to respiratory failure, septic shock, multi-organ failure, and death (4).

Along with the direct burden of acute disease, it is increasingly recognized that there may be prolonged recovery in individuals with prior COVID-19 infection, even in the absence of severe symptoms and hospitalization, which has been termed post-acute COVID-19 syndrome. These long-term effects are likely exacerbated in those with pre-existing conditions or who have experienced a prolonged hospital stay. Although evidence is only beginning to accumulate on the pathophysiology of long-term effects due to COVID-19, it appears that symptoms may persist long after disease remission and can be accompanied by multi-organ dysfunction. In this review, we will highlight relevant evidence evaluating the influence of past COVID-19 infection on cardiovascular and skeletal muscle function as well as exercise capacity. This will serve as a basis to speculate on long-term clinical outcomes for individuals recovering from COVID-19. Although not a primary focus of this review, we wish to highlight that the prolonged inactivity and treatments (e.g., glucocorticoids) that typically accompany hospitalization due to severe disease may lead to further declines in physical function. This review focuses on studies published before the dominance of variants such as Delta. Although we recognize that the emergence and spread of unique SARS-CoV-2 variants may lead to some differences in overall symptom experience and that the long-term consequences from specific variants are unknown, it is likely that long-term effects will not differ significantly from those covered herein, and much can be learned from the effects of the early SARS-CoV-2 strains.

Long-Term COVID-19 Symptomatology and Clinical Manifestations

Given the contemporary nature of the pandemic, limited data are available on the long-term outcomes of patients recovering from COVID-19; however, all current evidence indicates prolonged symptom duration and disability in a large proportion of adults. Fatigue and dyspnea appear to be the most prevalent symptoms, regardless of disease severity or time since disease onset. Given the scope of this review, we have highlighted symptoms that may be closely related to cardiovascular function, muscle function, and exercise tolerance. However, we wish to emphasize that there are many symptoms beyond those discussed in this review that convalescent patients with COVID-19 experience, including a high prevalence of neurological symptoms (5–7), pulmonary abnormalities, and loss of taste/smell. The majority of data on symptomatology has focused on individuals with moderate-to-severe disease, whereas data on those with milder disease are sparse. Regardless of disease severity, evidence is primarily based on self-report. Although common themes emerge indicating a high prevalence of symptoms related to cardiovascular and, potentially, to muscular dysfunction, the prevalence across studies is wide ranging, and points to a clear need for better identification of pathophysiological dysfunction across the spectrum of disease severity and recovery time (Fig. 1).

Figure 1.

Prevalence of select symptoms based on acute disease severity and time since acute coronavirus disease 2019 (COVID-19) infection. Symptom prevalence associated with long-term impacts of COVID-19 varies widely depending on disease severity, time, and method of assessment. In shorter-term studies (<3 mo since acute disease), those with milder disease (i.e., outpatients) had a similar prevalence of fatigue, dyspnea, and cardiac abnormalities compared to those with more severe disease (i.e., hospitalized); the influence of mild disease on muscle weakness during convalescence is unknown. In longer-term studies (3+ mo since acute disease), symptoms remain in those with milder disease, but appear to be more prevalent in those with more severe disease. Yellow and orange circles represent those with milder disease. Red and pink circles represent those with more severe disease. Thinner circles represent the upper range for each symptom based on the literature, thicker circles represent the lower range based on the literature, while gray represents the proportion of those who appear to be unaffected. Data are presented as the range of prevalence values for select symptoms based on the studies included in the symptomatology section of the text. Image created with BioRender and published with permission.

Post-acute COVID-19 symptoms in patients with milder disease.

Although data are now accumulating in those with severe disease and hospitalization, reports are more limited on symptomatology in COVID-19 outpatients. The proportion of individuals with mild disease suffering long-term symptoms varies significantly across studies; however, research consistently indicates long-term symptom experience months after acute disease. For example, ∼2–3 wk following mild symptomatic COVID-19, more than one-third of patients reported persistent reductions in health (8), whereas in a cohort recruited through social media, >99% of hospitalized and non-hospitalized participants reported symptoms ∼3 mo after disease onset. It is worth noting that due to the method of recruitment, there was likely a selection bias towards persistently symptomatic individuals in this study, leading to an over-reporting of symptom prevalence; however, information can still be gained from this analysis. The authors demonstrate that age, health before disease, pre-existing comorbidities, and the number of symptoms experienced during acute infection only predicted 36% of the variance in long-term symptom experience, illustrating the complexity involved in understanding long-term outcomes (9).

Supporting long-term symptom experience in those with milder disease, ∼30% of hospitalized-patients and outpatients reported the persistence of at least one symptom ∼6 mo after illness, with fatigue being the most common symptom (10). Similarly, in healthcare workers with mild COVID-19, the relative risk of exhibiting at least one moderate-to-severe symptom lasting for at least 2 mo was 2.9 in those with prior COVID-19 compared to those without, whereas the relative risk of having at least one moderate-to-severe symptom lasting for at least 8 mo was 4.4. Participants also reported prolonged disruptions to work life, home life, and social life, along with reductions in categories of the Sheehan Disability scale (11). Taken together, the available evidence indicates that a milder course of acute disease does not preclude long-term symptom experience.

Post-acute COVID-19 symptoms in patients with moderate-to-severe disease.

Although evidence is still limited in terms of long-term symptom experience in individuals with a milder course of disease, there has been a rapidly growing body of evidence indicating that symptoms are long lasting in patients with more severe disease requiring hospitalization. In fact, a systematic review concluded that more than 70% of patients, the majority of whom were hospitalized, experience at least one persistent symptom 60 days or more after disease onset (12). In convalescent patients, >85% reported persistence of at least one symptom, particularly fatigue and dyspnea, 36–48 days after discharge from the hospital (13, 14). Similarly, 4- to 12-wk post-hospital admission, nearly three-quarters of participants reported ongoing symptoms including shortness of breath, fatigue, and myalgia (15, 16). Finally, 3–4 mo post-discharge approximately half of participants reported experiencing at least one symptom (5, 17), with reports of cardiovascular symptoms including an increase in resting heart rate compared to before infection (17).

Prior infection also appears to affect physical function. Despite 72% of participants reporting a return to physical activity 100 days post-hospitalization, approximately half of those participants reported returning at a lower level than before illness, with 55% of participants reporting fatigue (18). In those who were ∼6 mo from symptom onset, 76% of previously hospitalized patients with COVID-19 reported at least one symptom, with 63% reporting fatigue or muscle weakness; those with more severe disease tended to have more severe long-term symptoms, including reduced pulmonary diffusion capacity, mobility issues, and pain (19). Importantly, recent data indicate that although persistent symptoms are present in 49% of previously hospitalized individuals 1 year post-infection, there are improvements in symptom experience and markers of physical function and health, relative to 6 mo, suggesting that recovery, although gradual, may occur over time (20).

Overall, the evidence is clear that COVID-19 can lead to a long recovery lasting for months after the acute phase of illness is resolved. Patients with milder acute disease also experience persistent symptoms; however, these symptoms appear to be worse in those with more severe disease requiring hospitalization. Additional research using quantitative evaluation methods (e.g., imaging techniques, biomarkers, and functional assays) is necessary to better characterize the long-term impact of the disease and, importantly, to determine the underlying biological factors contributing to these clinical manifestations.

Pathophysiology Post-Acute COVID-19

Although there is consistent evidence indicating prolonged symptom experience in many individuals recovering from COVID-19, research on the pathophysiological changes that may accompany and underlie these long-term symptoms has only recently begun to accumulate. Along with persistent alterations in pulmonary function and control of ventilation, there appear to be long-term consequences of prior COVID-19 infection on the cardiovascular system, skeletal muscle, and exercise tolerance. Together, these may contribute to physical disabilities and greater risk for the development of chronic diseases. The prevalence of these long-term pathophysiological changes and the role of acute disease severity on their persistence are poorly characterized; however, the available evidence suggests that these problems may remain prevalent across the spectrum of acute disease severity.

Prolonged effects of COVID-19 on cardiovascular function.

Although the pulmonary system has remained a primary target of investigation, due to the respiratory nature of the disease, mounting evidence indicates that acute SARS-CoV-2 infection also leads to cardiac abnormalities (21–23). It is increasingly apparent that these impairments remain in the convalescent period (24) and likely contribute to long-term symptoms following COVID-19 infection. SARS-CoV-2 infiltrates cells through the angiotensin-converting enzyme 2 (ACE2) receptor, which is present in many organs including the heart and on endothelial cells throughout the vascular network. During the acute phase of the disease, SARS-CoV-2 contributes to increased markers of endothelial activation (25–27) and circulating endothelial cells, which have been related to disease severity (26). Vascular pathology in COVID-19 encompasses vasculitis, increased vascular permeability, fibrin deposition, vascular thrombosis, and embolization (28).

Along with the ability of SARS-CoV-2 to directly affect the vasculature, the secondary effects of SARS-CoV-2 infection on endothelial cells can contribute to depressed vascular function. Specifically, acute SARS-CoV-2 infection can trigger a “cytokine storm” and is associated with elevated oxidative stress, each of which may adversely impact the cardiovascular system. For these reasons, it has been proposed that SARS-CoV-2 may be an endothelial disease (28, 29), and many comorbidities associated with endothelial dysfunction and cardiac damage (e.g., hypertension and diabetes) (30) are likely to be aggravated by COVID-19.

Cardiac manifestations. There is a growing body of evidence documenting cardiac abnormalities caused by past SARS-CoV-2 infection (Fig. 2). In fact, a systematic review of cross-sectional studies concluded that across the disease severity spectrum, signs of cardiac damage remain in many individuals recovering from COVID-19 (24). Given the novelty of this area of research, much remains to be elucidated as to the effects of acute disease severity on long-term cardiac outcomes. In those with milder acute disease, results have been somewhat equivocal. Specifically, in asymptomatic and mildly symptomatic young competitive athletes, ∼31% and 15% exhibited signs of prior myocardial injury and myocarditis, respectively, assessed via cardiac magnetic resonance imaging (MRI), 11–53 days after a positive test result for SARS-CoV-2 (31). In another cohort, more than 56% of college-athletes with both symptomatic and asymptomatic infection had signs of myocardial and pericardial damage during convalescence (32). In stark contrast with these reports, a recent large scale study 6-mo post-infection indicated that the proportion of cardiovascular abnormalities in healthcare workers with mild COVID-19 was similar to that of controls (33). The strengths of the experimental design of this study (large sample size, prospective recruitment, use of phantom scanner calibration, and blinded analysis) bolster the conclusion that cardiovascular anomalies resulting from mild COVID-19 infection may be transient and resolve within 6 mo. Of note, all of the studies highlighted earlier are cross-sectional. Although experimentally more difficult to conduct, longitudinal studies are necessary to address this controversy and to systematically characterize the trajectory of cardiac recovery following milder COVID-19 infection.

Figure 2.

Prevalence of late gadolinium enhancement (LGE) via cardiac magnetic resonance imaging (MRI) post-coronavirus disease 2019 (COVID-19). Studies consistently indicate prevalence of late gadolinium enhancement (LGE), reflecting cardiac damage, in those recovering from COVID-19, regardless of acute disease severity or time since initial infection. Data are presented as the percentage of participants showing signs of LGE via cardiac magnetic resonance imaging (MRI) in a select group of studies. Any value above zero indicates cardiac damage. Studies in figure are Rajpal et al. (31), Knight et al. (34), Huang et al. (35), Brito et al. (32), Wang et al. (38), Raman et al. (70), and Puntmann et al. (36).

Unlike milder cases, long-term cardiac impairments have been consistently documented in those previously hospitalized due to COVID-19. For example, 69% of patients hospitalized for COVID-19 complications exhibited myocardial injuries detected by cardiac MRI ∼46 days after symptom onset (34). In a retrospective analysis, cardiac MRI revealed abnormal findings (e.g., myocardial edema, fibrosis, and impaired right ventricular function) in 58% of previously hospitalized patients experiencing cardiac symptoms ∼47 days after symptom onset (35). Similarly, cardiovascular abnormalities detected by MRI were present in 78% of middle-aged patients with COVID-19 ∼71 days post-diagnosis, independent from pre-existing conditions, disease severity, and overall course of acute illness. Compared to healthy risk-factor matched controls, patients with prior COVID-19 infection had lower right and left ventricular ejection fraction, higher left ventricular volume, and elevated T1 (longitudinal relaxation time) and T2 (transverse relaxation time), indicating significant structural and functional abnormalities (36). Consistent with these findings, cardiac MRI abnormalities and diastolic dysfunction were documented in 30%–60% of previously hospitalized and outpatients 2–3 mo since hospital discharge (37, 38). Importantly in terms of trajectory post-COVID-19, signs of cardiac injury were present in between ∼30% (39) and 65% of patients ∼6 mo post-hospitalization, with greater persistence of tissue damage in those who experienced a more severe acute form of the disease (40).

It is worth noting that the high variability in prevalence of cardiac involvement across studies may be partially due to differences in measurement choice (e.g., MRI vs. echocardiography) and/or participant selection. For example, Huang et al. (35) only recruited participants with persistent cardiac symptoms, which likely lead to an overestimation of cardiac anomalies induced by COVID-19. Despite these differences, all of these studies report some degree of cardiac impairment during follow-up of previously hospitalized patients with COVID-19. These effects are likely to be worse in individuals who experienced prolonged hospital stays, are older, and/or have pre-existing conditions, though to our knowledge, this has yet to be comprehensively evaluated. Given the limited regenerative capacity of cardiomyocytes (41), additional work is necessary to evaluate the long-term implications of these findings on future disease risk.

Peripheral vascular manifestations. Although much of the available evidence on cardiovascular involvement in those recovering from COVID-19 has focused on the heart, recent studies also indicate abnormalities in the peripheral vasculature, which may contribute to post-acute COVID-19 syndrome. Increased levels of endothelial cells are found in circulation at least 1 mo post-symptom onset (42), whereas markers of endothelial activation remain elevated ∼2 mo post-COVID-19 infection (43). Arterial stiffness appears to be worsened, with higher carotid-to-femoral pulse-wave velocity and greater carotid and aortic artery stiffness ∼1 mo post-infection in young adults with prior asymptomatic or mild disease compared to age-matched healthy controls (44, 45); however, unaltered aortic distensibility has also been reported (33). Measures that focus on nitric oxide (NO)-dependent dilation, a key measure of endothelial health, also appear to be altered following COVID-19 infection, regardless of the severity of the acute phase of the disease (Fig. 3). In young adults, brachial artery flow-mediated dilation (FMD) and movement-induced hyperemia during single passive leg movement are significantly lower in those with mild disease compared to age-matched controls (FMD, COVID-19: 2.7%; controls: 8.8%) 3–4 wk after symptom onset (44). For context, FMD values of ∼4.5% have been documented in the brachial artery of patients with chronic heart failure class III/IV NYHA (50), and a 1% decrease in FMD is associated with a 9% increase in future risk of cardiovascular events (51). Interestingly, lower FMD has been reported only in those experiencing prolonged symptoms ∼3 mo after diagnosis (symptomatic: 4.3%; asymptomatic: 6.8%, control: 6.8%) (49). Together, these findings indicate an acute loss of function in the vasculature of young, otherwise healthy individuals, after mild SARS-CoV-2 infection, which appears to persist for several months in those with prolonged symptoms.

Figure 3.

Effects of prior coronavirus disease 2019 (COVID-19) infection on brachial artery flow-mediated dilation. In a select group of cross-sectional analyses, brachial artery flow-mediated dilation (FMD) was consistently lower in individuals who had COVID-19 compared to those who did not. Data are presented as mean ± SE. Data from Riou et al. are reported as median and interquartile range, therefore error bars are not shown. Studies in figure are Ratchford et al. (44), Lambadiari et al. (46), Riou et al. (47), Ambrosino et al. (48), and Nandadeva et al. (49).

In cases with more severe disease, there is also growing evidence of peripheral vascular dysfunction post-COVID-19. The hyperemic response to single passive leg movement was blunted in frail elderly adults 2–4 mo post-hospitalization (52). FMD was also lower in previously hospitalized middle-aged to older adults ∼3 mo after COVID-19 infection compared to a control group (COVID-19: 8.2%, controls: 10.3%) (47). Consistent with these results, other studies with larger sample sizes demonstrated lower FMD in COVID-19 middle-aged and older patients compared to healthy controls 2–4 mo post-infection (46, 48).

Putative mechanisms. Current studies all suggest that reduced NO bioavailability, perhaps secondary to elevated free radicals, in convalescent patients with COVID-19 (44) contributes to the endothelium-related vascular dysfunction post-COVID-19. Along with lower NO bioavailability, several other pathophysiological mechanisms are likely involved in impaired vascular function following COVID-19 infection. There is persistent inflammation in COVID-19 patients 2 mo after hospital discharge, indicated by high concentrations of several circulating cytokines (10 of 17 cytokines tested) (53), and previous investigations have demonstrated that elevated inflammation impairs endothelial function (54, 55). Circulating proinflammatory mediators can also cross the blood-brain barrier and directly activate the sympathetic nervous system (SNS) and/or trigger peripheral receptors stimulating group III/IV afferent feedback, ultimately resulting in exaggerated SNS activity. Interestingly, in young adults ∼1 mo post-SARS-CoV-2 infection, SNS activity was heightened at rest (56) and the exercise pressor response (elevated heart rate and systolic blood pressure) was exaggerated during isometric handgrip exercise (57). Taken together, this suggests that impaired autonomic regulation following COVID-19 infection contributes to altered hemodynamics and vascular abnormalities. Although unaltered autonomic function was reported in middle-aged adults ∼5 mo post-COVID-19 (58), this result should be considered with caution given the methodology used, which might have prevented the detection of a statistical effect.

Overall, inflammation, oxidative stress, and autonomic dysfunction can, independently or concurrently, modulate vasomotor tone and contribute to peripheral vascular dysfunction post-COVID-19. Given the link between lower limb vasodilatory capacity and V̇o_2peak (59, 60), this vascular dysfunction is also likely to compromise O₂ delivery to the exercising muscle, leading to reduced exercise tolerance and physical function following COVID-19 infection. It will be important to determine whether this vascular dysfunction is transient or if it increases future risk of cardiovascular disease. Identifying the extent of cardiovascular abnormalities across a range of clinical presentations, and the mechanisms underlying them, will be critical for future risk evaluation and mitigation strategies.

Prolonged effects of COVID-19 on cardiorespiratory fitness and physical function.

Physical and skeletal muscle function.

Despite the common symptoms of weakness and muscle pain reported in those recovering from COVID-19, limited data exist on the short- and long-term influence of the disease on exercise tolerance and skeletal muscle function. In addition, the studies on this topic are almost exclusively focused on patients with severe disease requiring hospitalization. For example, despite physiotherapy and early mobilization, ∼80% of patients with COVID-19 with moderate-to-severe disease, and without prior locomotor disabilities, demonstrated muscle weakness at the time of discharge from post-acute care (∼20 days from hospital admission). Specifically, quadriceps and biceps muscle strength, number of rises during a 1-min sit-to-stand test, and Short Physical Performance Battery test averaged between 54% and 74% of predicted normal values (61). In patients with COVID-19 requiring invasive mechanical ventilation, hospital acquired weakness was diagnosed in 27% of patients upon hospital discharge (∼30 day intensive care unit stay) (62). Handgrip strength was 64% of predicted values and Barthel index of daily life activities was lower than the cut-off value for independent living in ∼50% of patients at discharge (62). Similarly, after ∼30 days of hospitalization for COVID-19, ∼30%–50% of patients exhibited abnormal physical function or difficulty performing daily living activities assessed by a 1-min sit-to-stand test, Short Physical Performance Battery test, and the Barthel index (63).

Although more scarce, longer-term studies also indicate impairments in exercise tolerance and muscle function. At ∼3 mo follow-up, severe arterial desaturation during a 1-min sit-to-stand test was observed in 15% and 28% of convalescent older patients with moderate and severe COVID-19, respectively (15). In addition, studies have documented an inability to return to baseline physical activity in previously hospitalized patients with COVID-19 (18), and ∼6 mo after infection, 54% of patients still demonstrated moderate-to-very-severe physical disabilities (64). Consistent with persistent impairments in physical function post-COVID-19, two large-scale clinical trials in middle-aged and older adults reported that 20%–40% of previously hospitalized patients performed below the normal range during a 6-min walk test or an incremental shuttle walk test 6 mo after hospitalization (19, 64). Confirming the persistence and the extent of functional impairments in the post-acute phase of COVID-19, ∼40%–50% of those patients scored 10 or less on the Short Physical Performance Battery (64).

Cardiorespiratory fitness.

Cardiorespiratory fitness is dependent on cardiovascular, pulmonary, and skeletal muscle function, and is predictive of cardiovascular and all-cause mortality (65). Given the effects of past COVID-19 infection on these physiological systems, it is not surprising that patients with a prior history of COVID-19 appear to have low cardiorespiratory fitness (Fig. 4). At hospital discharge (median 30 days of hospitalization), older individuals previously infected with COVID-19 demonstrated a 35% lower peak oxygen consumption (V̇o_2peak, ∼15 mL·kg⁻¹·min⁻¹) compared to a control group of outpatients (∼23 mL·kg⁻¹·min⁻¹) (66). Peak power output during incremental cycling exercise in the COVID-19 group was half of the control group (66). Similarly, V̇o_2peak was lower 1–3 mo after illness in patients with COVID-19 compared to a control group (66–68). In addition, values corresponding to ∼80% and ∼90% of age-predicted values (∼19 mL·kg⁻¹·min⁻¹) for V̇o_2peak and peak power output, respectively, were reported during whole body cycling in patients 6 mo post-hospital discharge (69), confirming a persistent decline in cardiorespiratory fitness post-COVID-19.

Figure 4.

Effects of prior coronavirus disease 2019 (COVID-19) infection on cardiorespiratory fitness. In a select group of studies, which compared individuals with prior COVID-19 infection to controls, V̇o_2peak was consistently lower in those recovering from COVID-19. Data are presented as mean ± SE. Data from Baratto et al. are reported as median and IQR. Pleguezuelos et al. reported 95% confidence intervals, which was used to extract SE. Data from Szekely et al. are reported in L/min. Studies in figure are Baratto et al. (66), Pleguezuelos et al. (67), and Szekely et al. (68).

Putative mechanisms. The lower exercise capacity present in individuals following COVID-19 infection reflects a complex pathophysiology, which involves mechanisms beyond gas exchange abnormalities. For example, in one study, only 10% of participants terminated exercise due to breathlessness, suggesting that extrapulmonary factors likely accounted for premature exercise cessation (70). In support of this theory, differences in V̇o_2peak were dissociated from ventilatory efficiency and lung function 3–5 mo post-hospital discharge in patients with persistent symptoms (71, 72).

Exercise tolerance is tightly linked to the integrity of the O₂ transport pathway, which depends on a series of steps (ventilation, alveolar/capillary diffusion, circulation, and muscle diffusion) as O₂ molecules are passed from the atmosphere to the mitochondria. Based on this conceptual framework, it is noteworthy that COVID-19 infection results in altered central hemodynamics and persistent endothelial dysfunction, which together may compromise O₂ delivery to the exercising muscle bed (60). Supporting this, chronotropic incompetence (i.e., lower heart rate during exercise testing), sometimes coupled with lower stroke volume (71), was observed in patients post-COVID-19, which attenuated the exercise-induced increase in cardiac output (68, 71). The blunted cardiac output, perhaps due to contractile dysfunction and/or impaired autonomic regulation, was only partially compensated for by higher O₂ extraction, ultimately resulting in lower V̇o_2peak in convalescent patients with COVID-19 (68). These findings point to additional downstream limiting factors that may impair exercise-induced vasodilation and diffusive O₂ conductance.

Although comprehensive in their characterization of factors that may limit exercise tolerance in convalescent patients with COVID-19, studies to date have not quantitatively estimated the effects of the disease on the mechanisms of O₂ transport. Using an existing published data set (66) and an analytical framework of the determinants of O₂ transport (73, 74), we conducted a secondary analysis to quantitatively examine the contribution of peripheral factors related to V̇o_2peak limitation in patients with prior COVID-19 infection. This analysis revealed a ∼63% lower peripheral diffusional oxygen transport capacity (DO₂) following prolonged hospitalization due to COVID-19 (Fig. 5). The large decrease in DO₂ highlights the importance of peripheral factors such as impaired hemodynamics (57), lower hematocrit (66), and capillary density, along with altered skeletal muscle metabolism in limiting exercise tolerance in COVID-19 survivors.

Figure 5.

The relationship between peak O₂ consumption (V̇o_2peak) and venous O₂ content during symptom limited cycling exercise in controls and patients with coronavirus disease 2019 (COVID-19) at hospital discharge. The slope of the lines from the origin reflects O₂ diffusional conductance between the microvascular compartment and muscle mitochondria according to Fick’s law. This law states that the net diffusive rate of O₂ from the capillaries to the mitochondria is the product of the muscle tissue diffusion capacity (DO₂) and the mean partial pressure difference for O₂ between the capillaries (${\text{Pc}}_{{\text{O}}_2}$) and the mitochondria (${\text{Pm}}_{{\text{O}}_2}$). Assuming that ${\text{Pc}}_{{\text{O}}_2}$ is proportional to venous blood Po₂ (${\text{Pv}}_{{\text{O}}_2}$) and mitochondrial Po₂ is close to zero, the equation can be simplified to V̇o₂ = DO₂ × k × ${\text{Pv}}_{{\text{O}}_2}$. Combined with the similar cardiac output between groups, this theoretical framework suggests that the decrease in V̇o_2peak present in those with COVID-19 may be attributed, in part, to an impaired “peripheral” diffusive O₂ transport capacity. Values are presented as median ± IQR for V̇o₂ only, as IQR could not be extracted for venous Po₂. Data based on Baratto et al. (66).

It should be acknowledged that the peripheral DO₂ value estimated here relies on the use of pulmonary V̇o_2peak and cardiac output, rather than muscle-specific measurements in the exercising limb. Direct measurements of these variables are warranted to confirm this secondary analysis and to elucidate the exact peripheral factors limiting O₂ supply and utilization in the skeletal muscle of convalescent patients with COVID-19. In support of the hypothesis that patients recovering from COVID-19 may exhibit abnormalities in the skeletal muscle bed, a recent case study reported lower sensitivity to adenosine disphosphate (ADP)-stimulated respiration in permeabilized skeletal muscle fibers during the pre-symptomatic phase of COVID-19 (75). In addition, glycemic abnormalities and insulin resistance were detectable 2 mo after acute infection in a large proportion of patients hospitalized with COVID-19 and without a prior history of diabetes (53).

It is also likely that the intolerance to exercise observed in previously infected patients with COVID-19 is partly a consequence of general illness-induced fatigue and inactivity. A recent meta-analysis reported that maximal V̇o₂ declines by ∼0.4%/day with strict bed rest (76). In the context of COVID-19 hospitalization, the average hospital stay (10–15 days) would result in, at least, a 4%–6% reduction in cardiorespiratory fitness. Interestingly, in a study by Baratto et al. (66), the median stay of patients with COVID-19 was 30 days and was associated with a 35% lower V̇o_2peak compared to healthy controls, which is much higher than the values predicted by physical inactivity alone. As illustrated in Fig. 6, the COVID-19 induced decrement in V̇o_2peak is exaggerated compared to the V̇o_2peak decline predicted by severe physical inactivity and appears to barely recover after 6 mo of convalescence. Although this cross-sectional comparison has limitations, these data suggest a disease-specific and/or treatment-related aggravation of the decrease in V̇o_2peak related to COVID-19 hospitalization. Not only will this influence chronic disease risk and all-cause mortality (77), but also will likely contribute to frailty and poor quality of life post-hospitalization.

Figure 6.

Estimated effects of coronavirus disease 2019 (COVID-19) hospitalization versus prolonged bed rest on peak oxygen consumption during symptom-limited exercise. The effects of COVID-19 on peak O₂ consumption (V̇o_2peak) during symptom limited cycling exercise in patients with COVID-19 at hospital discharge [30 days of hospitalization, Barrato et al. (66)] and after 6 mo [21 days of hospitalization, Debeaumont et al. (69)] are not solely explained by the predicted decrease in V̇o_2peak induced by strict bed rest alone (76).

Besides cardiorespiratory fitness, mechanical efficiency (i.e., the ratio between total energy expenditure and external work), may also account for some of the deleterious effects of COVID-19 hospitalization on exercise tolerance. For example, delta efficiency during cycling was significantly lower (20%) in patients with COVID-19 8 wk after hospital discharge compared to age-matched healthy controls (27%) (67). The use of short duration increments during the exercise test and the severity of the acute phase of illness (12 days in ICU) of the recruited patients should be considered when interpreting the results of this study. Given the importance of mechanical efficiency for exercise capacity (78), the influence of COVID-19 infection and hospitalization on muscle efficiency warrants further investigation.

Based on the evidence reviewed here, it is likely that beyond the mild gas exchange abnormalities in the lungs, and cardiac limitations, impairments in the downstream mechanisms of O₂ transport and utilization may also be involved and should be targeted by rehabilitation programs tailored to the needs of the COVID-19 convalescent population. In addition, it has been reported that alterations in mitochondrial function may occur during both pre-symptomatic and acute COVID-19 infection in skeletal muscle (75) and peripheral blood mononuclear cells (79). Hypothetically, changes in mitochondrial function may alter the metabolic state and redox balance of numerous organs and tissues, and play a role in COVID-19 pathogenesis (80, 81). Whether this altered mitochondrial function persists after the inflammatory phase of the disease and contributes to exercise intolerance post-COVID-19 is still unknown. Overall, future work evaluating the mechanisms underlying impairments in physical function and exercise tolerance following COVID-19 infection will be critical.

CONCLUSION AND FUTURE DIRECTIONS

Although SARS-CoV-2 has only been widely circulating in humans for a relatively short amount of time, there is a growing body of evidence to suggest that along with its contributions to acute mortality and disability, COVID-19 infection also leads to long-term symptoms in many of those affected. The most prevalent of these symptoms seem to be fatigue and dyspnea. Research on underlying pathophysiology is starting to accumulate. Specifically, in addition to long-term symptoms, there is evidence of prolonged cardiovascular derangements during recovery and adverse effects on the mechanisms of O₂ transport, skeletal muscle function, and exercise capacity (Fig. 7). In this context, it is critical for more research to be conducted to better understand the long-term consequences of COVID-19 infection on physical function across the spectrum of disease severity, and to identify the underlying biological mechanisms contributing to these adverse outcomes. Together, this review highlights the urgent need for the development of evidence-based rehabilitation strategies to reduce the long-term burden of post-COVID-19 symptomatology in patients and on the health care system.

Figure 7.

Pathophysiological mechanisms underlying long-term outcomes due to coronavirus disease 2019 (COVID-19). Following acute COVID-19 infection, pathophysiological adaptations may occur, leading to exercise intolerance and chronic disease. Current evidence suggests that cardiovascular dysfunction, muscle weakness, and mild gas exchange abnormalities may contribute to the long-term impact of the disease; however, many of the underlying mechanisms of this multi-organ dysfunction remain unknown. Image created with BioRender and published with permission.

GRANTS

This work was supported by an IALS midigrant, UMASS Amherst.

DISCLOSURES

No conflicts of interest, financial or otherwise, are declared by the authors.

AUTHOR CONTRIBUTIONS

C.S. and G.L. conceived and designed research; G.L. analyzed data; C.S. and G.L. interpreted results of experiments; C.S., S.T.D., and G.L. prepared figures; C.S. and G.L. drafted manuscript; C.S., S.T.D., and G.L. edited and revised manuscript; C.S., S.T.D., and G.L. approved final version of manuscript.