Mark Barash, DO^a,∗ and Vijaya Ramalingam, MD, FCCP^a,b

Clin Chest Med. 2023 Jun; 44(2): 263–277. Published online 2023 Mar 3. doi: 10.1016/j.ccm.2022.11.019 PMCID: PMC9983785 PMID: 37085219 Author information Copyright and License information PMC Disclaimer

As the world emerges from the COVID-19 pandemic, clinicians and researchers across the world are trying to understand the sequelae in patients recovered from COVID-19 infection. In this article, the authors review post-acute sequelae of SARS-COV-2, interstitial lung disease, and other lung sequelae in patients recovering from COVID-19 infection.

Keywords: COVID-19, Pulmonary function test, Computed tomography, Biopsy, Sequelae, PASC, Pulmonary fibrosis

Key points

• •Post-acute sequelae of SARS-COV-2 or post-COVID conditions are a poorly defined syndrome, possibly with long-lasting effects.
• •Respiratory failure is one of the most serious complications of COVID-19 infection and contributes to major morbidity and mortality.
• •Several studies have demonstrated abnormal lung function tests and radiological findings in patients who recovered from COVID-19 infection.
• •Evidence on post-COVID pulmonary fibrosis is evolving.

Introduction

Respiratory failure is one of the most serious complications of COVID-19 infection and contributes to significant morbidity and mortality. Illness severity ranges from mild/asymptomatic disease to critical illness requiring mechanical ventilation. There has been increasing concern about pulmonary sequelae, including symptoms, pulmonary function testing (PFT) abnormalities, and pulmonary fibrosis.¹ Our knowledge about the natural history of recovery after COVID-19 infection is limited.

This article focuses on two concepts. First, the authors seek to describe available knowledge of post-COVID lung disease including pulmonary physiologic changes, imaging characteristics, fibrotic lung disease, and other complications. Next, the authors discuss the post-acute sequelae of SARS-CoV-2 (PASC): a poorly understood syndrome comprising a conglomerate of “head-to-toe” symptoms that afflicts a subset of patients recovering from COVID-19.

Post-COVID lung disease

Abnormal Pulmonary Function Tests

Survivors of COVID-19 demonstrate heterogenous abnormalities in PFT (Table 1 ) and decrements in exercise capacity/diffusion of oxygen (Table 2 ). To this end, PFT abnormalities in survivors of severe lung injury are not a wholly new concept. For example, 5-year survivors of acute respiratory distress syndrome (ARDS) were found to be functionally impaired with a median 6 minute walk distance (6MWD) 76% predicted.² A meta-analysis of long-term outcomes after severe acute respiratory syndrome (SARS) and Middle Eastern respiratory syndrome (MERS) identified a reduced 6MWD and diffusing capacity for carbon monoxide (DLCO) compared with healthy individuals.³

Table 1

Pulmonary function testing sequelae in patients recovering from COVID-19

Study	N	Timing	Obstructive Pattern (FEV1/FVC < 70 or < LLN)	Restrictive Pattern (FVC < LLN or 80% TLC < LLN or 80%)	DLCO < 80%
Van Gassel et al,10 2021	43	3 mo after discharge	0	16 (37.2%) 23 (53.5%)	36 (87.8%)
Van den Borst et al,7 2021	124	10 wk after discharge	12 (10%)	15 (13%)	41 (34%)
Gonzalez et al,8 2021	62	3 mo after discharge	1 (2%)	23 (37.1%)	50 (82%)
Liang et al,9 2020	76	3 mo after discharge	5 (6.6%)	0	15 (19.7%)
Lerum et al, 6 2021	103	3 mo after admission	NA	7 (7%)	24 (24%)
You et al, 97 2020	18	40 ± 11.6 d in cases with nonsevere illness, and 34.7 ± 16.5 d in cases with severe illness	3 (33%)	3 (33%)	NA
Huang et al,70 2021	349	6 mo after symptom onset	22 (6.3%)	56 (16%)	114 (32.7%)
Huang et al,98 2020	57	30 d after discharge	1 (1.8%)	6 (10.5%)	30 (52.6%)
Bellan et al,5 2021	224	3–4 mo after discharge	NA	NA	113 (50.4%)
Smet et al, 99 2021	220	74 ± 12 d after diagnosis	NA	84 (38%)	48 (22%)
Shah et al,100 2021	60	3 mo after symptom onset	11.7% (7)	23.3% (14)	51.7% (31)
Zhao et al,12 2020	55	3 mo after discharge	NA	4 (7.25%)	9 (16.4%)
Mo et al,101 2020	110	On discharge	5 (4.5%)	27 (25%)	51 (47.2%)
Chen et al,4 2021	41	1 y after discharge	3 (7.3%)	5 (12.2%)	3 (7.3%)

Abbreviations: DLCO, diffusing capacity for carbon monoxide; FEV1, forced expiratory volume in 1 s; FVC, forced vital capacity; LLN, lower limit of normal; NA, not available; TLC, total lung capacity.

Adapted from So M, Kabata H, Fukunaga K, Takagi H, Kuno T. Radiological and functional lung sequelae of COVID-19: a systematic review and meta-analysis. BMC Pulm Med. 2021;21(1):97. Published 2021 Mar 22.

Table 2

Exercise capacity and oxygenation in patients recovering from COVID-19 infection

	Timing	Dyspnea mMRC N (%)	6 MW Test (Distance in Mean or Median)	6 MWD <80% or < LLN	Significant Desaturation
Van Gassel et al,10 2021	3 mo after discharge	16 (37.2%) with score ≥ 1	482 m (82% p)	NA	2 (4.7%)
Van der Borst et al,7 2021	10 wk after discharge	Median 1	Normal	25 (22%)	20 (16%)
Gonzalez et al,8 2021	3 mo after discharge	NA	400 m	NA	NA
Lerum et al,6 2021	3 mo after admission	37 (54%) with score ≥ 1	580 m	NA	NA
Huang et al,70 2021	6 mo after symptom onset	419/1615 (26%)	495 m (87.7% p)	392 (23%)	NA
Shah et al,100 2021	3 mo after symptom onset	NA	NA	NA	4 (7%)
Huang et al,98 2020	30 d after discharge	NA	561 m (94% p)	NA	NA
Guler et al,11 2021	4 mo after discharge	NA	456 m (severe/critical disease) 576 (mild/moderate disease)	NA	NA

Several studies from across the world have demonstrated reduced diffusion capacity, lung volumes (total lung capacity [TLC]), 6MWD, and exertional desaturation in COVID-19 patients during follow-up. Reduction in DLCO is the most common PFT abnormality.⁴ In a large Italian study,⁵ female sex, chronic kidney disease, and the modality of oxygen delivery during hospital stay were shown to be risk factors for DLCO less than 80% at follow-up and female sex, COPD, and intensive care unit (ICU) admission were shown to be risk factors associated with DLCO less than 60% at follow-up. However, in another study, the prevalence of reduced lung function was similar between ICU and non-ICU participants.⁶ High-Resolution computed tomography (HRCT) score during acute illness and residual pulmonary parenchymal opacities at discharge also correlated with the lower diffusion capacity after 3 months.7, 8, 9 A European study that assessed respiratory sequelae of mechanically ventilated patients with COVID-19 showed high prevalence of abnormal lung function testing; 53.5% patients had reduced TLC, whereas 87% had reduced DLCO at 3 months post-discharge. The median 6MWD was 482 m (82% predicted).¹⁰ A prospective cohort study showed a significant reduction in the 6MWD in COVID-19 patients compared with the healthy population (median difference −128.43 m).⁸

The Swiss COVID-19 lung study reported PFT findings 4 months after initial symptoms in 113 patients. Patients with prior severe or critical disease had lower lung volumes than patients with mild or moderate disease and had abnormally reduced diffusion capacity, reduced functional capacity, and demonstrated exertional oxygen desaturation.¹¹ Zhao and colleagues¹² showed that elevated D-dimer was associated with decreased diffusion capacity in follow-up PFTs, possibly indicative of microthrombus formation.

In a 1-year follow-up study, only a small number of patients had DLCO less than 80% suggesting improvement in lung function from 6 months to 12 months.⁴ Similar observations were made in a 2-year follow-up study. The proportion of COVID-19 survivors with an mMRC score of at least 1 was 168 (14%) of 1191 patients at 2 years, significantly lower than the 288 (26%) of 1104 at 6 months (P < 0.0001). The proportion of individuals with a 6MWD less than the lower limit of the normal range declined continuously in COVID-19 survivors overall and in the three subgroups of varying initial disease severity. However, critically ill patients had a significantly higher burden of restrictive ventilatory impairment and lung diffusion impairment than controls at the 2-year follow-up.¹³

Radiological Sequelae

The development of pulmonary fibrosis is a known complication after severe respiratory tract infection and such changes have been reported in survivors of SARS and MERS.¹⁴ Residual radiographic abnormalities are seen in a large proportion of COVID-19 survivors at the time of discharge and subsequent follow-up.^15 , 16 In one study, the predominant pattern on CT scan changed over time with consolidative changes peaking 3 weeks after onset of symptoms and decreasing thereafter. Ground-glass opacification (GGO) or GGO with reticular pattern was the most common abnormal patterns from onset of symptoms until 12 months after hospital discharge.⁴

van Gassel and colleagues¹⁰ reported normal pulmonary findings in only 2/46 patients at 3 month follow-up and GGO was noted in 89% of cases. Patients admitted to ICU showed interlobular septal thickening and bronchiectasis as the most frequent changes seen on CT chest at 3 months.⁸ Patients admitted to ICU had higher prevalence of persistent CT abnormalities at 3 month follow-up. The distribution of GGO was mainly subpleural and similar in appearance to a nonspecific interstitial pneumonia (NSIP) pattern. Participants with limited residual changes mainly showed subpleural parenchymal bands or small plate atelectasis.⁶

More than one-third of severe COVID-19 survivors demonstrated fibrotic-like changes (traction bronchiectasis, parenchymal bands, and honeycombing) at 6 months after illness onset, and two-thirds of participants showed either complete radiographic resolution or residual GGO or interstitial thickening.¹⁵ A 12-month follow-up CT in a subset of patients who had fibrotic interstitial lung abnormalities (ILAs) at 6-month period demonstrated stable fibrotic ILAs in more than two-thirds and slight improvement in the rest. Age greater than 50 years, ARDS, and higher baseline CT lung involvement score were predictors of fibrotic-like changes in the lung. The need for noninvasive mechanical ventilation was also a predictor of fibrotic-like changes.¹⁷ Of note, progression of ILAs was not apparent. In another 1-year follow-up study, investigators found that age, smoking, hypertension, lower SaO₂, and secondary bacterial infections during acute phase were significantly associated with residual radiological abnormalities. Lung volume parameters of TLC and residual volume were significantly lower in patients with residual CT abnormalities than those without abnormalities at 1 year after hospital discharge.⁴

In the Swiss lung study, mosaic attenuation was the most common radiological change at 4-month follow-up. More than 50% of patients with severe or critical disease had mosaic attenuation, reticular changes, and architectural distortion at 4-month follow-up. Risk factors for post-SARS and MERS fibrosis were also older age and likelihood of having been in the ICU^{14 , 18 , 19}

Fibrotic Lung Disease: The Proof Is in the Pudding

It is possible that various insults such as ventilator-induced lung injury,20, 21, 22 bacterial infection, and hyperoxia23, 24, 25 contribute to post-COVID fibrosis. Most of the patients with persistent inflammatory interstitial lung disease (ILD) require supplemental oxygen, ICU stay, and mechanical ventilation during their hospital stay.²⁶

A spectrum of lung injury patterns has been found in patients with COVID-19 and vary with time from initial illness. Transbronchial lung cryobiopsy performed in 12 patients within 20 days of symptom onset showed epithelial and endothelial cell abnormalities different from either classical interstitial lung diseases or diffuse alveolar damage (DAD). Alveolar type II cell hyperplasia was a prominent finding in most of the cases. No evidence of hyaline membranes was noticed.²⁷ Several other reports have shown acute and organizing DAD in postmortem tissue samples from patients who died of severe disease.28, 29, 30 There is histologic evidence of diffuse fibrotic ILD in patients recovering from COVID-19 infection. NSIP-like fibrosis accompanied by acute lung injury has been described in lung explant specimens.31, 32, 33

In a large study of 50 patients who underwent transbronchial cryobiopsy at a mean duration of 87 days from discharge, organizing pneumonia was the most common pathologic finding (32%) followed by diffuse lymphoplasmacytic interstitial infiltrate. Patchy fibrosis was observed in only four patients. There was no evidence of hyaline membranes, fibroblastic enlargement of the interstitium. Classic UIP or NSIP was not identified.³⁴ In another study, surgical lung biopsy showed UIP as the most common pathologic finding in patients undergoing evaluation for post-COVID-19 ILD. The investigators proposed these patients had lung disease before developing COVID infection.³⁵

Role of Steroids and Antifibrotics in Post-COVID Lung Disease

Unfortunately, a paucity of data exists regarding what (if any) intervention should be undertaken in patients with persistent/residual pulmonary abnormalities.

An observational study of corticosteroid treatment in post-COVID ILD patients showed improvement in dyspnea, physical functioning, chest imaging, and lung function. Seven percentage (or 4% of the entire cohort) of patients had persistent interstitial changes on chest CT 6 weeks after discharge and most of them had an organizing pneumonia pattern.²⁶ In a Swiss national survey of pulmonologists, moderate recommendation was given in favor of an empiric steroid trial for patients with interstitial abnormalities after COVID-19.³⁶ Corticosteroid treatment may shorten the time to recovery and return to functioning for patients recovering from organizing pneumonia-like pattern associated with COVID-19.³⁷ However, evidence supporting corticosteroid use in post- COVID ILD is limited, and physicians should be cautious when prescribing steroids until robust data are available for their use. For reference, a 15-year follow-up study of SARS survivors showed most pulmonary lesions recovered within 1 year, and high-dose steroid exposure was associated with femoral head necrosis.³⁸

Evidence on incidence of post-COVID pulmonary fibrosis is evolving. Currently, there is no evidence for or against the use of antifibrotic agents in post-COVID ILD. The natural history of post-COVID ILD is unclear. There are few reports on the use of nintedanib and pirfenidone in patients with COVID.39, 40, 41 In an interventional study of patients with COVID-19 requiring mechanical ventilation, the use of nintedanib was associated with shorter length of mechanical ventilation and lower percentages of high-attenuation areas on CT volumetry, suggesting lung-protective effects.⁴² Table 3 shows the completed and ongoing trials evaluating the use of antifibrotic medications in COVID-19, though randomized controlled trials are lacking.

Table 3

Current studies investigating corticosteroid and antifibrotic therapy in patients with post-COVID (as of June 30, 2022)

ClinicalTrials.gov Identifier	Study	Status
NCT04657484	Comparison of Two Corticosteroid Regimens for Post-COVID-19 Diffuse Lung Disease (COLDSTER)	Completed
NCT04551781	Short-Term Low-Dose Corticosteroids for Management of Post-COVID-19 Pulmonary Fibrosis	Completed
NCT04988282	Systemic Corticosteroids in Treatment of Post-COVID-19 Interstitial Lung Disease (STERCOV-ILD)	Recruiting
NCT04534478	Oral Prednisone Regimens to Optimize the Therapeutic Strategy in Patients With Organizing Pneumonia Post-COVID-19 (NORCOVID)	Not yet recruiting
NCT04541680	Nintedanib for the Treatment of SARS-Cov-2 Induced Pulmonary Fibrosis (NINTECOR)	Recruiting
NCT04338802	Efficacy and Safety of Nintedanib in the Treatment of Pulmonary Fibrosis in Patients With Moderate to Severe COVID-19	Unknown
NCT04619680	The Study of the Use of Nintedanib in Slowing Lung Disease in Patients With Fibrotic or Non-Fibrotic Interstitial Lung Disease Related to COVID-19 (ENDCOV-I)	Recruiting
NCT04856111	Pirfenidone vs Nintedanib for Fibrotic Lung Disease After Coronavirus Disease-19 Pneumonia (PINCER)	Recruiting
NCT04653831	Treatment With Pirfenidone for COVID-19-Related Severe ARDS	Recruiting
NCT04607928	Pirfenidone Compared to Placebo in Post-COVID19 Pulmonary Fibrosis COVID-19 (FIBRO-COVID)	Recruiting

Open in a separate window

Not surprisingly, pulmonary rehabilitation could improve physical and psychological conditions, including exercise training, muscle strength, walking, and functional ability in patients with post-COVID ILD.^43 , 44

Post-Acute Sequelae of SARS-CoV-2

Colloquially called “long COVID” or “Long hauler’s syndrome,” the PASC or post-COVID conditions, is a poorly defined syndrome that includes a range of new, recurrent, or ongoing health problems that can last weeks, months, or years after infection with COVID-19.⁴⁵ Indeed, due to a lack of consensus on the underlying physiology, symptom burden and timeline for symptom onset and resolution, a formal definition remains elusive. Symptoms may include severe fatigue and post-exertional malaise, onset of neuropsychiatric symptoms and difficulty with memory/concentration, persistent loss of taste and smell, dyspnea, cough, palpitations and postural orthostasis and various gastrointestinal (GI) symptoms, to name a few. A meta-analysis of studies that included at least 100 patients describing PASC symptoms, 55 long-term effects were identified (Fig. 1 ).⁴⁶ In a large survey, 640 patients recovering from COVID-19 were given the opportunity to write in symptoms they attributed to PASC; over 200 additional symptoms were reported beyond the 62 choices provided by the researchers.⁴⁷ These studies illustrate, if nothing else, the profound sense of unwellness that many patients experience.

Fig. 1 Long-term effects of COVID-19.

(From Lopez-Leon S, Wegman-Ostrosky T, Perelman C, et al. More than 50 long-term effects of COVID-19: a systematic review and meta-analysis. Sci Rep. 2021;11(1):16144. Published 2021 Aug 9.)

Post-Viral Syndromes

PASC seems to represent a post-viral syndrome. It is important to acknowledge that PASC is not the first (nor likely the last) syndrome of its kind. After an outbreak of Russian Flu (1889 and 1892), observers noted a subset of survivors to develop symptoms of neuralgia, neurasthenia, neuritis, nerve exhaustion, grippe catalepsy, psychosis, anxiety, and paranoia.⁴⁸ After the Spanish Flu (1918–1919), patients displayed symptoms of parkinsonism and catatonia.⁴⁸ The term “encephalitis lethargica” gained prominence, though was first described a year earlier in 1917 after an outbreak of meningitis with delirium in Vienna.⁴⁹ A 1935 outbreak of atypical poliomyelitis (atypical because 53 of 59 reported cases had normal cerebrospinal fluid [CSF] analysis) in a Los Angeles hospital led to prolonged symptoms of mental dullness and decreased ability to concentrate painful oculomotor and gastrointestinal symptoms.⁵⁰ During the 1950s and onward, several epidemics in London, Iceland, Australia, and Florida preferentially affected women and recovery was prolonged by fatigue and recurring myalgia, though no mortality was recorded.⁵¹ From these outbreaks emerged the term myalgic encephalomyelitis (ME).⁵¹ Following several mononucleosis-like outbreaks in the 1980s similarly characterized by prolonged symptoms, the first definition of chronic fatigue syndrome (CFS) was published.⁵² An update on diagnostic criteria was published by the centers for disease control (CDC) in 1994, putting forth the term CFS/ME.53, 54, 55 Box 1 displays the criteria for CFS/ME.

Box 1

Criteria for myalgic encephalomyelitis/chronic fatigue syndrome

• Diagnosis requires that the patient has the following three symptoms:
  • 1.A substantial reduction or impairment in the ability to engage in pre-illness levels of occupational, education, social, or personal activities that persist for more than 6 months and are accompanied by fatigue, which is often profound, is of new or definite onset (not lifelong), is not the result of ongoing excessive exertion, and is not substantially alleviated by rest
  • 2.Post-exertional malaise
  • 3.Unrefreshing sleep
• At least one of the following manifestations is also required:
  • 1.Cognitive impairment or
  • 2.Orthostatic intolerance

From Herrera JE, Niehaus WN, Whiteson J, et al. Multidisciplinary collaborative consensus guidance statement on the assessment and treatment of fatigue in postacute sequelae of SARS-CoV-2 infection (PASC) patients [published correction appears in PM R. 2022 Jan;14(1):164]. PM R. 2021;13(9):1027-1043.

More recent nonseasonal coronavirus outbreaks provide lessons as well. The 2003 SARS outbreak spread to 29 countries in Asia, Europe, and North America leading to 8422 cases recorded and 916 deaths (11% case fatality); Toronto, Ontario, had the highest concentration. Of 117 survivors surveyed 1 year from illness, 60% continued to experience fatigue, 45% complained of shortness of breath, 18% had reduced 6MWD, and 51 of 117 continued to require mental health visits; only 13% remained asymptomatic.⁵⁶ Long-term survivors from China continued to experience active psychiatric illness (40%), chronic fatigue (40.3%), and 27.1% met criteria for CFS/ME at a mean follow period of 41.3 months.⁵⁷ In 2012, the MERS infected 2519 people and led to 866 deaths (35% case fatality). Similar to SARS, at 1 year, 48% of survivors demonstrated chronic fatigue and 42% complained of post-traumatic stress disorder (PTSD).^58 , 59 A large meta-analysis of SARS and MERS survivors found that 27.1% met criteria for CFS/ME and had reduced exercise capacity with lung function abnormalities.³

Incidence

The incidence of PASC is currently estimated to be between 10% and 35% of all infected individuals.^60 , 61 Morbidity can be so severe that as of July 2021, disability related to long-term symptoms from COVID-19 is covered under the Americans with Disabilities Act.⁶² As of June 2022, the UK Office for National Statistics estimates that approximately 2 million people are experiencing symptoms of PASC. Of those surveyed, 405,000 (21%) were less than 12 weeks from onset of symptoms, 1.4 million (74%) remained symptomatic at greater than 12 weeks from symptom onset, 807,000 (41%) at 1 year, and 403,000 (21%) at 2 years.⁶³

The impact of vaccination, current, and future viral variants on the development of PASC is also of great interest. A large study evaluated the effect of timing of vaccination along with variant of infection (Delta vs Omicron) on the development of PASC; the incidence of PASC was 4.5% (2501 of 56,003 people) among Omicron infections and 10.8% (4469 of 41,361 people) among Delta infections. In all age groups, the odds ratio (OR) of PASC ranged from 0.24 (0.20–0.32) to 0.50 (0.43–0.59) with Omicron compared with Delta. Vaccination status of less than 3 months had the highest OR of 0.5 (0.43–0.59), but there were insufficient data to determine PASC incidence in the unvaccinated population.⁶⁴ A survey of self-reported PASC symptoms in the United Kingdom found a 49.7% lower incidence of PASC from Omicron BA.1 variant compared with Delta in those who were double vaccinated. Interestingly, triple vaccination seemed to confer no difference in PASC risk between Delta and Omicron BA.1/BA.2, whereas infection compatible with Omicron BA.2 increased the odds of PASC symptoms by 21.8% compared with Omicron BA.1.⁶⁵ These findings may imply that a two-dose vaccination series could be sufficient to reduce PASC risk with Omicron but not with Delta. Last, there is some suggestion that vaccination itself may decrease the odds of PASC in those previously infected. A study of 28,356 adults infected before vaccination found that one dose of vaccine led to an initial 12.8% decrease in odds of PASC and a second dose led to an initial 8.8% decrease of PASC, sustained at 67 days follow-up.⁶⁶ The evolution of viral variants, vaccine types, and community rates of herd immunity and vaccination make it difficult to generalize, though these observations raise fascinating questions about ways to decrease the incidence of PASC.

Hospitalized Patients

PASC affects people along the entire disease spectrum—from minimal/mild symptoms to critical illness. Among hospitalized patients, a telephone interview of 488 survivors, 60 days after symptom onset, found that 32% had persistent symptoms defined as a conglomerate of shortness of breath, cough, chest tightness, wheezing, difficulty ambulating, breathlessness with stairs, oxygen use, and continuous positive airway pressure (CPAP) use, with only 16% being able to return to work.⁶⁷ Another survey of 143 hospitalized patients performed at 60 days after initial diagnosis found that 87% had at least one persistent symptom and 55% had three or more; only 12% stated that they were completely free of symptoms.⁶⁸ Halpin and colleagues⁶⁹ performed a cross-sectional evaluation of 100 hospitalized patients at 4 to 8 weeks post-symptom onset; 70% continued to experience fatigue and 50% endorsed dyspnea. Huang and colleagues⁷⁰ performed an analysis of 17 different symptoms in 1733 hospitalized patients at 6 months post-admission, dividing the cohort into severity scales; 76% of the total cohort had at least one symptom, and symptom burden increased with severity of illness. Muscle weakness (63%), sleeping difficulties (26%), and anxiety or depression (23%) were the most common reported symptoms. Of note, 23% of the patients had a decreased age-adjusted 6MWD and diffusion defects which correlated with illness severity.

Nonhospitalized Patients

Similar observations have been made in nonhospitalized patients. Jacobsen and colleagues⁷¹ found that in survey of 118 patients (96 outpatients), symptom burden was statistically similar between inpatient versus outpatient status and 67% of patients continued to experience symptoms, including a mean 6MWD of 59% of expected. Likewise, at a median follow-up of 169 days Logue and colleagues⁷² found similar burden of persistence of at least one symptom (33 vs 31%) in outpatients versus inpatients, respectively. Patients are also more likely to use health system resources including outpatient primary care and outpatient hospital visits⁷³

Risk Factors

The risk factors for PASC remain unknown, and data remain discordant. Wynberg and colleagues⁷⁴ found that female gender (adjusted hazard ration [aHR] 0.65, CI 0.47–0.92) and body mass index (BMI) greater than 30 (aHR 0.62 CI 0.39–0.97) were associated with slower recovery and symptoms beyond 6 months correlated with decreased chance of resolution. A large study of 2,149 health care professionals identified 323 participants who had no or mild symptoms and were found to be seropositive; 83% were women and 15% reported at least one persistent symptom at 8 months, as compared with seronegative participants (relative risk [RR] 4.4 [95 CI, 2.9–6.7]), causing significant disruption in work life, home life, and in any Sheehan Disability Scale category.⁷⁵ Among patients who survived hospitalization, the presence of ICU admission, need for respiratory support, premorbid lung problems, higher age, higher BMI, and BAME (Black, Asian, and Minority Ethnic) predicted breathlessness post-discharge.⁶⁹ Another study suggested that women were more likely to develop fatigue and anxiety/depression and presenting symptoms of palpitations, rhinitis, dysgeusia, insomnia, hyperhidrosis, anxiety, sore throat, and headache predicted PASC.⁷⁶ Sudre⁷⁷ and colleagues found that PASC was more likely with increasing age, female gender, higher BMI, and having five or more symptoms within the first week of onset A cohort study of 189 people similarly found that only female gender and preexisting anxiety disorder predicted PASC compared with controls; in this group of patients with predominantly mild/moderate initial infection, there was no association between developing PASC and any modality of diagnostic testing (PFTs, echocardiogram, serologic testing/inflammatory markers, and cognitive testing), though participants with PASC had significantly lower scores on the SF-36 Health Survey. There was no evidence of abnormal systemic immune activation, autoimmune disease, or persistent viral infection.⁷⁸

Post-Acute Sequelae of SARS-CoV-2 Phenotypes

Various phenotypes of PASC may exist. Defining them is complex due to the dynamic nature of both the initial illness and subsequent sequelae across illness severities. For example, a survivor of ARDS and ICU admission may develop dyspnea and abnormal PFTs that correlate with residual postinfectious fibrosis and overlap with the post-intensive care syndrome (PICS). A patient with only a mild illness without known history of pneumonia or hospitalization may also develop dyspnea out of proportion to imaging abnormalities and PFT derangements. Anecdotally, the latter is a remarkably common finding seen in our PASC clinic where most of the referrals are patients with no history of hospitalization and, generally, no known history of prior pneumonia. Several classifications systems have been proposed. Becker proposed a system based on the severity and evolution of symptoms over time.⁷⁹ Yong proposed subtypes based on long-term clinical and physiologic sequelae.⁸⁰ Tables 4 and ​and5 summarize5 summarize these two different schemas of subtyping PASC phenotypes. These descriptions will likely change over time and other schema will likely emerge, illustrating the complex nature of PASC that will undoubtedly continue to evolve along with our understanding. Newer schema would, ideally, group patients into phenotypes that also correlate with clinically relevant outcomes. No formal guidelines currently exist to define PASC phenotypes.

Table 4

Proposed COVID-19 sequelae subtype criteria

	Type 1	Type 2	Type 3		Type 4		Type 5
Initial symptoms	Variablea	Mild	A	B	A	B	None
			Mild	Mild	None	None
Duration of symptoms	Variablea	>6 wk	3–6 mo	>6 mo	Variable	Variable	N/A
Period of quiescence	No	No	Yes	Yes	No	No	N/A
Delayed onset of symptoms	No	No	No		Yes	Yes	Yes

^aCorrelate with severity of initial infection, number of organ system injured and preexisting medical conditions.

From Becker RC. COVID-19 and its sequelae: a platform for optimal patient care, discovery and training. J Thromb Thrombolysis. 2021;51(3):587-594.

Table 5

Symptoms and the proposed pathophysiology of subtypes of post-acute sequelae of SARS-CoV-2

Subtype	Proposed Diagnostic Guide	Main Pathophysiology
NSC-MOS	Multi-organ symptoms lasting for ≥ 3 mo after acute COVID-19 (regardless of disease severity), especially fatigue, dyspnea, and cognitive impairment.	Tissue damage across multiple organs or system-wide dysregulation
PFS	Pulmonary fibrosis and other pulmonary sequelae (ie, impaired lung function or respiratory symptoms) lasting for ≥ 3 mo after acute COVID-19, especially severe COVID-19.	Extensive tissue damage, especially in the lungs
ME/CFS	Disabling fatigue, unrefreshing sleep, PEM, and either cognitive impairment or orthostatic intolerance lasting for ≥ 6 mo after acute COVID-19.	Dysfunction of the immune and nervous systems
POTS	Increased heart rate of >30 beats per minute within 5–10 min of standing or upright tilt without orthostatic hypotension. May occur with dizziness, palpitations, blurred vision, headache, generalized weakness, exercise intolerance, and fatigue.	Dysfunction of the autonomic nervous system
PICS	Physical (eg, muscular weakness, weakened handgrip, poor mobility), cognitive (eg, memory and concentration) and mental (eg, anxiety, depression and PTSD) sequelae lasting for ≥ 3 mo after acute COVID-19 of ICU level of severity.	Severe-to-critical illnesses in need of ICU level of care, from which full recovery is difficult
MCS	Acute or chronic diseases or other clinical sequelae that require medical care. Examples include respiratory, cardiovascular, gastrointestinal, kidney, liver and neurologic diseases, diabetes, infectious diseases, and mental health disorders.	Health deterioration or unmasking of chronic diseases

Abbreviations: MCS, medical or clinical sequelae; ME/CFS, myalgic encephalomyelitis or chronic fatigue syndrome; NSC-MOS, nonsevere COVID-19 multi-organ sequelae; PEM, post exertional malaise; PFS, pulmonary fibrosis sequelae; PICS, post-intensive care syndrome; POTS, postural orthostatic tachycardia syndrome.

Adapted from Yong SJ, Liu S. Proposed subtypes of post-COVID-19 syndrome (or long-COVID) and their respective potential therapies. Rev Med Virol. 2022;32(4):e2315.

Mechanism(s) of Post-Acute Sequelae of SARS-CoV-2

A unifying mechanism for the variety and variability of symptoms of PASC remains elusive. Several studies have sought to clarify the causes of exercise intolerance as this is both common and debilitating symptoms. Studies using cardiopulmonary exercise testing (CPET) at various time points from illness and recovery have demonstrated predominantly circulatory and anaerobic threshold limitations when compared with matched controls81, 82, 83; however, ventilatory inefficiency has also been suggested.⁸² Cassar and colleagues⁸³ performed a longitudinal evaluation of 58 survivors and 30 matched controls via symptom questionnaires, cardiac and lung magnetic resonance imaging (CMR) and CPET at 3 and 6 months. By 6 months, survivors demonstrated normalization of cardiac abnormalities noted on previous CMR imaging, though persistent (and improved) low-grade abnormalities of parenchymal abnormalities and peak Vo ₂ persisted in 52% of participants; importantly (and congruent with our real-world experience), these abnormalities did not correlate with cardiopulmonary symptoms. These impairments could be related to direct damage to muscle tissue, impaired oxygen extraction/utilization, or simple deconditioning from prolonged hospital stay and critical illness. To this end, the shear breadth of physical, neuropathic, and neuropsychiatric symptoms is likely not explained by these mechanisms alone and certainly not all patients will experience dyspnea or exercise intolerance.

Other proposed mechanisms include ACE-2/Ang 1 to 7 receptor down regulation with deleterious upstream effects,⁸⁴ autoantibody production targeting cytokines, chemokines, complement system, or cell surface proteins⁸⁵ and pro-inflammatory cytokines.^86 , 87 The nature of the initial immune response may also have bearing on both the clinical disease course and the long-lasting sequelae.⁸⁸

Triage, Workup, and Treatment

Owing to the massive influx of patients with multitudes of symptoms, many centers around the United States and abroad have initiated multispecialty COVID clinics to help triage, treat, and address paucity of knowledge and expertise in this disease process. Clinics may have providers representing various medical specialties including neurology, neuropsychiatry and psychology, ear, nose and throat, cardiology, pulmonary, physical medicine and rehab, and physical and occupational therapy, among many others. No standardized approach has yet been validated but it is generally recommended that patients are approached in a holistic manner based on the severity of illness and symptom burden.^{55 , 89 , 90} Basic laboratory testing such as complete blood count, basic metabolic panels, liver function, thyroid function is likely reasonable. More specialized testing, for example, looking for evidence of vitamin deficiencies, inflammatory markers, rheumatological conditions, or myocardial injury, and so forth, should be guided by symptoms and clinical gestalt. Advanced testing may include chest and cardiac imaging, electrocardiograms, and invasive testing such as heart catheterization or CPET if a high clinical suspicion exists. It should be noted that a “shot-gun” approach to testing is not recommended given dubious clinical utility, increasing cost and emotional burden on the patient.

In the author’s experience (MB), patients often have trouble navigating their new symptoms and finding understanding from both their families, peers, and even other health care providers. In fact, a survey of 114 mostly female (80/114) medical professionals (51/114) with PASC in the United Kingdom described a heavy sense of loss and stigma, trouble accessing and navigating services, and difficulty being taken seriously.⁹¹ The clinician should consider then, the extra burden placed on their nonmedical patients. Acknowledging symptoms is very important, as are validating statements like “what you are going through is very real” and “there are many others just like you, learning to navigate this new illness”; for example, after a thorough history and examination, we focus heavily on first, setting expectations that the time course of illness and recovery is unknown. Once both patient and provider are ready to move forward, appropriate testing is ordered to diagnose both preexisting and new conditions. Emphasis should be placed on symptomatic and supportive therapy.⁶⁰ Our practice is skewed heavily toward physical and occupational therapy as fatigue and sensation of dyspnea are often quite prevalent and debilitating. Patients should be counseled on paying particular attention to “post-exertional malaise,” a debilitating state of fatigue onset from both physical and/or mental overexertion which has been well characterized in the setting of CFS/ME.⁹² Healthy dietary habits, sleep hygiene, and modification of daily routines to prioritize certain activities over others are encouraged. Individualized reconditioning protocols should be implemented by experienced physical and occupational therapists with experience treating PASC patients.⁵⁵ The triage, evaluation, and treatment of patients suffering from PASC remains dynamic, and a holistic approach is paramount.^60 , 93, 94, 95, 96

Summary

Post-COVID sequelae including lung injury and PASC are complex and poorly understood, representing heterogenous manifestations, mechanisms, and short- and long-term outcomes. The mainstay of therapy remains mostly supportive, though robust research is underway to better understand and characterize pathways for intervention. Historical insight remains important to clarify whether current observations are truly novel or representative of previously ignored or misunderstood syndromes.

Clinics care points

• •Post-acute sequelae of SARS-CoV-2 (PASC) are a syndrome of nonspecific “head-to-toe” symptoms of unclear cause, mechanism, and duration. Supportive and holistic care is paramount.
• •Abnormalities in lung function testing and chest imaging are fairly common in patients recovering from COVID-19. These abnormalities are heterogenous, often improve with time (though may not always normalize), and may not correlate with symptoms.
• •Reduced diffusion capacity is the most common abnormality in pulmonary function testing.
• •Patients who recover from COVID-19 may develop interstitial lung disease/pulmonary fibrosis. So far, data do not suggest that this represents a progressive fibrotic phenotype. The role of antifibrotic agents in the treatment of post-COVID fibrosis is being studied.