Preliminary Guidelines for the Clinical Evaluation and Management of Long COVID

1. Major symptoms and incidence

Long COVID may occur after recovery from acute COVID-19 symptoms, persist from acute phase symptoms, or may change or recur over time. According to the survey by the Centers for Disease Control and Prevention (CDC), related symptoms/signs persist after COVID-19 in 1 of 5 and 1 of 4 survivors aged 18 - 64 years and ≥ 65 years, respectively. About 200 symptoms were identified, including fatigue, dyspnea, depression/anxiety, cognitive decline, etc. [6]. The symptoms of long COVID that are likely to occur after acute COVID-19 infection are shown in Table 1.

Table 1
Overall and time-specific incidence of long COVID symptoms

• Click for larger image
• Click for full table
• Download as Excel file

Symptoms/signs that persist after COVID-19 usually improve over time [4]. Although the time to recovery varies individually, most people show improvement by week 12 [7]. According to the UK Office for National Statistics, 1 in 5 patients can have symptoms lasting >5 weeks, with 1 in 10 having symptoms for >12 weeks [8]. A decrease in quality of life has been confirmed in 57% of patients with long COVID [9], with 66% of patients with persisting symptoms/signs after COVID-19 showing organ damage and 25% showing multiple organ damage [10]. The results of a comparative study between influenza-infected and COVID-19 patient groups using data from the National Health Insurance Review and Assessment Service reported significantly higher incidence rates of dementia, heart failure, mood disorder, and alopecia among patients with COVID-19 patients after the acute phase of infection [4].

The most common symptoms of long COVID lasting up to 6 months were fatigue, post-exercise malaise, and cognitive dysfunction [11]. According to the online survey conducted among the domestic patient population in the Daegu region, 66% of patients complained of various clinical symptoms 6 months after acute COVID-19, including fatigue, concentration difficulties, amnesia, cognitive dysfunction, anxiety, depression, alopecia, dizziness, headaches, brain fog, anosmia, ageusia, and dyspnea [12]. Brain fog refers to a state in which thoughts and expressions are not clearly expressed due to a persistent feeling of mental fogginess. A survey study conducted 1 year after COVID-19 diagnosis in the same patient population, reported that 127 of 241 patients (53%) showed persistent long COVID symptoms, with the incidence decreasing in the order of concentration difficulty, cognitive dysfunction, amnesia, depression, fatigue, and anxiety. Specifically, neurological and psychiatric symptoms were confirmed to be long-lasting, and the duration of each symptom of long COVID showed a decreasing tendency over time [13]. A prospective survey conducted at a tertiary hospital in Daegu observed that various clinical symptoms persisted 1 year after COVID-19 diagnosis, including amnesia (24%), insomnia (15%), fatigue (14%), and anxiety (13%) [14]. The overall and period-specific incidence of the symptoms of long COVID are shown in Table 1 [12, 13, 15]. Post-COVID-19 symptoms/signs may persist regardless of the severity of the acute phase. The incidence of long COVID was higher in patients who received inpatient treatment than in patients who had not been hospitalized [4].

2. Risk factors for long COVID

The risk factors associated with long COVID include demographic factors (sex, age, etc.), obesity, underlying disease, hospitalization, severe illness, β2-agonist use, and non-vaccinated status [4, 13, 16, 17, 18, 19, 20]. However, depending on the study, the relevance of these risk factors varies for each symptom of long COVID. Two consistent risk factors for long COVID-19 are sex and the severity of the acute phase of COVID-19. Women are twice as likely to develop long COVID-19 as men [4]. The results of a domestic study confirmed that women are at high risk of neuropsychiatric long COVID-19 [12]. Severe COVID-19 shows a robust association with fatigue [4]. Some studies reported a higher risk of developing long COVID-19 in younger age groups, with no racial differences [21].

Key question 1) When should long COVID be suspected?

Long COVID is diagnosed by exclusion. The diagnostic approach begins by excluding complications from COVID-19 and other diseases unrelated to COVID-19.

1) Assessment for COVID-19 Complications

2. Assessment of diseases unrelated to COVID-19

In addition to the complications of COVID-19, it is also necessary to assess the aggravation of the underlying disease that may cause the symptoms or other diseases unrelated to COVID-19. Appropriate laboratory or radiologic assessments are necessary by considering that various symptoms corresponding to various respiratory, cardiovascular, hematologic, endocrine, allergic, and neuropsychiatric diseases may appear alone or in combination (Table 2) [7, 42].

Table 2
Diseases that may develop after COVID-19

• Click for larger image
• Click for full table
• Download as Excel file

Key question 2) What tests are needed for long COVID symptoms?

1) Blood tests required for diagnosis by exclusion

No studies have evaluated the usefulness of routine blood tests in patients with long COVID-19. A recent study reported that a small proportion (approximately 1 - 5%) of patients had abnormal findings in laboratory tests at follow-up assessment 12 months after the diagnosis of COVID-19 [43]. Nevertheless, blood tests according to symptoms should be performed to exclude other diseases. Care should be taken in interpreting whether the results indicate persistent abnormal findings after COVID-19. As recommended by several guidelines, the following blood tests may be considered depending on the symptoms [4, 44, 45]; C-reactive protein (CRP), erythrocyte sedimentation rate (ESR), complete blood count (CBC), and liver function test (LFT). Myocardial enzyme tests (troponin, creatine kinase-myoglobin binding [CK-MB]) and brain natriuretic peptide tests (B-type natriuretic peptide, BNP) can be considered in patients with heart-related problems. If clinically suspected, thyroid function tests (TFTs) are performed to rule out thyroiditis. Arterial blood gas analysis is performed in patients with reduced oxygen saturation. D-dimer testing is not recommended in patients without respiratory symptoms. Patients at high risk for diabetes or with impaired fasting blood glucose should undergo measurement of the fasting blood glucose and glycosylated hemoglobin level.

2) Dyspnea

Diagnostic tests may be required if dyspnea persists for >4 - 12 weeks after acute COVID-19. Some guidelines recommend using the modified Medical Research Council [mMRC] dyspnea scale to evaluate the severity of dyspnea; however, there is currently no standard for assessing the severity of persistent dyspnea after acute COVID-19. Therefore, further research is needed [44]. Respiratory function abnormalities are reported at various rates in patients recovering from COVID-19, which vary depending on the definition of respiratory function abnormality, follow-up period, severity at the time of acute COVID-19, use of mechanical ventilation, and previous respiratory function [46].

The most common pulmonary function test abnormality is abnormal diffusion capacity for carbon monoxide (DLCO). The restrictive pattern is the most common result of pulmonary function tests including forced vital capacity (FVC) and total lung capacity (TLC). Up to 80% of patients with critical COVID-19 treated in ICU had DLCO impairment (<80% of predicted) at discharge, and 50 - 70% had DLCO impairment at 3 - 6 months after discharge. At 3 months, impaired DLCO was associated with a high chest CT total severity score at hospitalization (evaluated by percentage of involvement in each lobe and overall lung) and acute respiratory distress syndrome (ARDS) [46]. At 6 months, DLCO impairement was observed in 29% of patients with severe COVID-19 and in 58% with critical COVID-19 [47]. A study performing 1 year follow-up after severe-critical COVID-19 reported DLCO of <80% in 23 - 54% of patients. TLC of <80% of predicted was observed in 39% of critically ill patients at 6 months and in 29% at 12 months [43]. Limited data on pulmonary function tests are available in patients with mild-to-moderate COVID-19. Several studies included some patients with mild-to-moderate disease as a control group for severe COVID-19 patients, in which 10 - 20% of these patients showed abnormalities in pulmonary function tests including DLCO [4, 25, 47].

Two observational studies examined the results of chest X-rays 6 - 8 weeks after acute COVID-19. The follow-up chest X-ray findings correlated with the severity of acute COVID-19 but not with persistent respiratory symptoms and the degree of symptom recovery [48, 49]. Among 110 patients hospitalized for COVID-19 and followed up for 8 - 12 weeks after discharge, 14% showed abnormalities in chest X-rays. In addition, reticular opacities were observed in 53% of patients with chest X-ray abnormalities, while peripheral atelectasis was observed in 33% of these patients [50].

In patients with severe or critical COVID-19, ground-glass opacities (GGOs), consolidation, or fibrotic changes were observed in 60 - 75% of patients with chest CT performed 3 months after diagnosis [51]. A systematic literature review of studies on chest CT findings 3 - 6 months after COVID-19 diagnosis reported abnormal findings in 59% of patients regardless of severity, most commonly GGOs (39%), followed by fibrosis and reticular opacities (30%) [52]. However, whether these abnormal findings predict future lung damage remains unknown. One study that performed chest magnetic resonance imaging (MRI) in 53 patients 2 – 3 months after recovery from acute COVID-19 observed lung parenchymal abnormalities in 60% of cases. However, these were not associated with clinically persistent symptoms [53].

If dyspnea persists for >4 - 12 weeks after acute COVID-19, a chest radiology examination can be considered to exclude other diseases, even though chest X-ray and chest CT findings poorly correlate with symptoms and can not predict future lung damage.

3) Cough

A cough that persists after acute COVID-19 is usually associated with chronic fatigue and dyspnea. Since lung fibrosis increases the sensitivity of the cough reflex, it is important to differentiate a persistent cough from other lung diseases and to check for fibrosis of the lung parenchyma or damage to the bronchus. Lung parenchymal fibrosis and bronchial damage can be detected by chest CT. It is also necessary to check for gastroesophageal reflux disease, a common cause of chronic cough, or for the use of angiotensin-converting enzyme (ACE) inhibitor [54].

4) Chest pain

Reports on pericarditis, myocarditis, heart failure, and cardiac arrhythmia after COVID-19 have been published, although causality is not always evident. One study showed that among patients with chest pain, palpitation, dyspnea, or edema 2 months after COVID-19, 28% (14/51) had severe cardiovascular disease [55]. However, this rate may be overestimated due to selection bias of patients with severe symptoms. Another study reported that 25% (35/150) of patients with mild to moderate COVID-19 showed abnormalities in transthoracic echocardiography performed 3 months after COVID-19 diagnosis, mostly reduced ejection fraction (EF), elevated pulmonary artery pressure, diastolic dysfunction, and thickened pericardium [56]. In transthoracic echocardiography studies performed 30 - 100 days after COVID-19, a considerable number of asymptomatic COVID-19 patients showed left ventricular strain, diastolic dysfunction, and pulmonary hypertension [57, 58]. Considering the frequency of these cardiac complications, it is appropriate to perform follow-up transthoracic echocardiography within 2 – 3 months in patients with confirmed myocarditis, pericarditis, or heart failure in acute COVID-19. Transthoracic echocardiography can also be considered in patients who have experienced chest pain, palpitations, or dyspnea from the acute stage but have not undergone diagnostic tests for cardiac complications.

Several studies of cardiac MRI have shown abnormalities in 19 - 71% of patients approximately 1 - 4 months after the diagnosis of COVID-19 [53, 59, 60]. These findings usually did not correlate with symptoms and were resolved 6 months after the diagnosis of COVID-19 [61]. Therefore, the need for cardiac MRI should be determined according to the individual patient's clinical situation and symptoms.

To assess the frailty and physical activity after COVID-19, physicians can consider checking the distance and oxygen saturation before and after 6 MWT, sit-to-stand test (counting the number of sitting and standing for 15 - 30 seconds), and short physical performance test battery (includes balance assessment in a standing position, walking speed for 4 minutes, and standing up from a char with five repetitions) [4]. Most studies assessing patients with long COVID using these methods showed that higher disease severity during the acute stage was associated with a more severe dysfunction [25, 47]. Therefore, when considering rehabilitation in long COVID patients with dyspnea and chest pain, the assessment of cardiac function can be considered first.

A Korean study reported the presence of chest pain after COVID-19 in 13% of patients after 6 months and 7% of patients after 12 months [13]. A meta-analysis reported a 6% rate (95% confidence interval [CI]: 3.2 - 12.4) at 3 - 6 months after the COVID-19 diagnosis [15], showing a decreasing trend over time. Therefore, if chest pain develops 12 weeks after the diagnosis of COVID-19, the possibility that chest pain is associated with a disease other than long COVID should be considered, and cardiac imaging studies for diagnosing/excluding other cardiac diseases can be performed, if necessary. However, there are currently no known specific cardiac image findings for long COVID. Therefore, there is insufficient evidence to recommend or against cardiac imaging for the diagnosis of long COVID, except to diagnose or exclude other cardiac diseases.

5) Fatigue

Fatigue is one of the common non-respiratory symptoms of patients with COVID-19, typically occurring in about 41% of patients [45]. This symptom occurs in 35 - 45% of patients at 4 weeks [62, 63], 30 - 77% of patients at 8 weeks [64, 65, 66], and 16 - 55% of patients at 12 weeks after infection [22, 67]. The fatigue in patients with COVID-19 has characteristics similar to those of the Middle East Respiratory Syndrome (MERS) or Severe Acute Respiratory Syndrome (SARS), and the chronic fatigue syndrome described after community-acquired pneumonia [68, 69]. Fatigue is a difficult symptom to define, is generally non-specific, has subjective characteristics, and is difficult to assess objectively. To evaluate the degree of fatigue in patients complaining of fatigue, physicians can apply a fatigue severity scale (FSS) (Table 3). The FSS consists of a total of nine items, in which the level of fatigue during the past week is evaluated on a scale of 1 to 7. The scores for each item are summed, and the average value divided by nine is calculated as the final FSS score. The higher the scores, the higher the degree of fatigue. This evaluation method shows a sensitivity of 84% and a specificity of 86% based on a cutoff of 3.22 points [70].

Table 3
Fatigue severity scale

• Click for larger image
• Click for full table
• Download as Excel file

To date, no association has been reported between the symptoms of long-term fatigue associated with COVID-19 and the severity of COVID-19 or laboratory test values related to inflammation. However, excessive fatigue symptoms have been reported in female patients previously diagnosed with depression or anxiety [71]. If fatigue persists as a major symptom >12 weeks after COVID-19, the date of onset, the presence of symptoms or signs accompanying fatigue, socio-psychological/emotional factors or drugs that may cause fatigue, sleep disorders, and exposure to toxins are first investigated. In addition, it is essential to confirm the specific medical history of other current symptoms that coexist with fatigue, such as the presence of an underlying disease that may be related to the current fatigue or whether there are sequelae from severe COVID-19. In addition to the above, CBC with differential counts, electrolyte tests (sodium, potassium, chloride, bicarbonate, calcium, and phosphate), LFT, renal profile, ESR, CRP, ferritin, d-dimer, TFT, muscle enzyme, and plasma cortisol levels are evaluated. At the same time, measurements of vital signs, oxygen saturation, electrocardiogram, chest X-ray, spirometry, and nutritional assessment are also recommended. If no organic abnormalities are revealed in these evaluations, it is recommended to evaluate whether the criteria for chronic fatigue syndrome are met (Table 4).

Table 4
Evaluation criteria related to the diagnosis of chronic fatigue syndrome

• Click for larger image
• Click for full table
• Download as Excel file

The presence of four or more of the evaluation criteria in Table 4 can be diagnosed as chronic fatigue syndrome associated with COVID-19 [45]. If the criteria are not met, it is categorized as unexplained post-COVID-19 fatigue, and each evaluation item of the criteria, psychological factors, and related physical conditions are re-evaluated. After applying a physical activity program such as rehabilitation support, the evaluation criteria for chronic fatigue syndrome are re-evaluated after 6 months (Fig. 2).

Figure 2
Diagnostic approach for chronic fatigue syndrome related to COVID-19.
COVID-19, coronavirus disease 2019.

• Click for larger image
• Download as PowerPoint slide

6) Joint pain and muscle pain

A study in Korea performed 6 and 12 months after the diagnosis of COVID-19 reported that arthralgia occurred in 11% of patients after 6 months and 7% after 12 months, while myalgia occurred in 6% and 12% after 6 and 12 months, respectively [13], similar to the results of other studies reporting arthralgia in 10 - 48% of patients at 4 - 12 weeks, and 9% at 3 - 6 months (95% CI: 5.7 - 15), while myalgia was reported in 1 - 32% at 4 - 12 weeks and 11% of patients at 3 - 6 months (95% CI: 6.2 – 19.8) [9, 15]. Several case reports described rhabdomyolysis in COVID-19 patients [72, 73, 74, 75], and most cases occurred in the acute stage (within 4 weeks of symptom onset). In the case of severe COVID-19, there have been reports of rhabdomyolysis in the sub-acute stage (4 - 12 weeks from symptom onset), but there is no case of rhabdomyolysis occurring more than 12 weeks after symptom onset. Rheumatologic serology tests (anti-cardiolipin antibody, anti-nuclear antibody, and rheumatoid factor) did not differ significantly between patients with persistent symptoms and the control group in a study of 189 patients with symptoms persisting 6 weeks after the onset of COVID-19 [76]. However, in this study, a few patients suffered myalgia or arthralgia (11 and 6 patients, respectively); therefore, it can not be concluded that rheumatologic serology tests are not useful in patients with long COVID. Although several studies have reported arthralgia and myalgia as the symptoms of long COVID, no study has performed both muscle biopsy (histopathology) and skeletal muscle imaging. Most studies assumed that muscle weakness was related to cytokine storms. The pathophysiologic mechanism causing myalgia in long COVID is unknown [77].

7) Headache

Headache is one of the common symptoms of patients with COVID-19 (14 - 60% of cases) and may persist for several weeks after acute COVID-19 [78]. A meta-analysis reported that 15% of patients (95% CI: 4.5 - 25.8) complained of headaches even 3 months after COVID-19 [79]. Long COVID-related headache can be a worsening of the existing headache or a new onset headache. A common feature is the near-daily persistence of headache that occurrs at the time of COVID-19 diagnosis or with a slight time lag. Immune or inflammatory responses activated during infection cause headache or may aggravate existing migraines [80]. SARS-CoV-2 virus does not directly cause long COVID-related headache; thus, it is reasonable to regard it as a symptom that can persist long after viral infection [78]. Previous reports have described new daily persistent headache lasting >3 months after infection with Epstein-Barr virus and even after the 1890 Russian flu pandemic [81, 82]. The similarities between new daily persistent headache after these infections and long COVID-related headache require further research.

The most important action in patients with a headache is to rule out secondary headaches with organic causes. First, the headache pattern should be identified through medical history taking, and focal neurologic deficit should be assessed through neurological examinations. Brain imaging tests (MRI or CT) should be considered, and a referral should be made to a neurologist for professional evaluation and treatment for headache suspected to be due to organic causes, such as headache with fever, vomiting, or weight loss; headache in cancer patients or immunosuppressed patients; headache accompanied by neurological abnormalities; headache with papilledema; severe thunderclap headache peaking within 1 minute; headache with new onset after 50 years of age; headache that worsens over time; and headache that occurs in situations similar to the Valsalva maneuver or that worsens when standing up. If organic causes are excluded, the treatment used for primary headache can be applied [45].

8) Neurocognitive symptom

A meta-analysis that examined the prevalence of neurological symptoms 3 months after COVID-19 reported a 22% (95% CI: 7.3 - 36.4) prevalence for decreased concentration, a 32% (95% CI: 10.3 - 54.0) prevalence of brain fog, and a 28% (95% CI: 21.5 - 35.4) prevalence of memory impairment [79]. A domestic study that examined the residual symptoms in patients at 6 and 12 months after COVID-19 reported the prevalence rates of neurological symptoms, including 22% for concentration difficulties, 20% for amnesia, and 21% for cognitive impairment [13].

In patients showing cognitive symptoms related to long COVID, brain imaging and neuropsychological tests can be considered when necessary to differentiate these symptoms from other causes or if the symptoms are severe enough to lead to decreased occupational or social function [4, 83]. Referral should be made to a neurologist for professional evaluation and treatment [3, 69]. Because the neuropsychological tests usually performed in clinical practice were developed for elderly patients with dementia, these assessments are often unsuitable for young patients, as in the case of those with long COVID who complain of concentration difficulties rather than a severe cognitive impairment that limits daily life. Therefore, a sensitive cognitive test tool suitable for this population should be developed and validated [3].

Recent studies have reported the brain imaging findings of patients with COVID-19. In a cohort of the UK brain bank, a comparison of MRI findings between 401 patients who had recovered from COVID-19 and 384 control subjects showed reduced gray matter thickness of the orbitofrontal cortex and parahippocampal gyrus in COVID-19 patients. Moreover, in areas functionally connected to the primary olfactory cortex, the markers related to tissue damage were more changed, and the overall brain size was decreased in COVID-19 patients [84]. In addition, in a study of brain positron emission tomography and computerized tomography (PET-CT) in 35 patients with long COVID-19, the lower the metabolism of the brainstem and cerebellum, the greater the number of complaints of symptoms [85].

There are several reports of cognitive tests after COVID-19. Analysis of the neuropsychological examination findings and health records of 740 patients who recovered from COVID-19 with no previous dementia showed decreased processing speed (in 18% of patients), decreased executive functioning (in 16%), and functional decline in tasks including phonemic fluency (in 15%), category fluency (in 20%), memory encoding (transformation of short-term memory into long-term memory) (in 24%), and memory recall (in 23%). Patients who had severe COVID-19 were more likely to develop cognitive sequelae [86]. Another study enrolled 1,438 patients >60 years of age 1 year after COVID-19 to assess for changes in cognitive function; the incidence of cognitive impairment was 12%, and the cognitive test scores were significantly lower in the COVID-19 group than in the control group. Patients with severe COVID-19 had a lower score for cognitive function and a higher risk of progressive cognitive impairment than those with mild COVID-19 or the control group [41]. Although studies based on neuropsychological and brain imaging tests for long COVID-related cognitive symptoms have been recently reported, further research is needed regarding the long-term prognosis.

9) Psychological/psychiatric symptom

Assessment of mental health in patients with COVID-19 is important [87], whether the cause is the direct effect of the virus or is a reaction to the experience of stressful events. This is because mental health can affect the quality of life of patients as well as the occurrence of other symptoms of long COVID [45]. Psychosocial assessment includes screening for depression, anxiety, post-traumatic stress disorder, psychotic symptoms, risk of harming self or others, mourning for sudden death, and life stress related to COVID-19 (e.g., debt, job loss or difficulty finding a job, interpersonal problems) [88]. In particular, patients with severe psychiatric symptoms or risks of harming themselves or others should be promptly referred for a psychiatric assessment to avoid delays in proper interventions [3, 88]. The earlier help is received, the more effective the intervention. Patients without social support may experience anxiety and mental health deterioration [3].

Anxiety, depression, and post-traumatic stress disorder are reported in 16 - 47% of patients 2 - 3 months after hospitalization and discharge among COVID-19 survivors [89, 90]. During admission in COVID-19 isolation wards, patients may experience concerns about their own health or that of others, physical and social isolation, potential risk of death, the risk of infecting others, leaving family behind, threats to livelihood, helplessness, boredom, and loneliness. Such stress may trigger new psychiatric symptoms or exacerbate preexisting mental health conditions [91]. Preexisting neuropsychiatric disorders or substance use disorders may worsen the course of COVID-19 or increase the risk of long-term sequelae [83, 92]. Patients admitted to COVID-19 isolation wards have reduced physical activity, which is particularly problematic for the elderly [45]. Decreased physical function can increase mental health problems such as anxiety and depression. In addition, post-COVID-19 acute stress, hospitalization environmental factors, invasive medical procedures (e.g., mechanical ventilation), and multiple medications may increase the risk of sleep disruption [93]. The experience of ICU admission in COVID-19 survivors may increase long-lasting functional limitations, post-traumatic stress disorder, and depression [94]. A recent study reported incidence rates of anxiety and psychotic disorders of 17% and 1.2%, respectively, with particularly high rates in patients who had been admitted to the ICU [40]. When treating patients with psychological or psychiatric symptoms, it is important to consider social factors such as whether they are in situations of poverty, discrimination, or social exclusion [45]. Social connectedness, social support, and other collective or community-based measures can be beneficial for patient mental health and well-being.

The comparison of PET-CT between 35 patients with COVID-19 and a control group at mean 96 days after diagnosis showed hypometabolisms in specific brain regions of patients with COVID-19. These findings were associated with hyposmia, anosmia, memory or cognitive decline, pain, and insomnia [95]. In a prospective study of 58 patients with moderate-to-severe COVID-19, brain MRIs performed 2–3 months after diagnosis showed abnormalities in the thalamus and sagittal stratum in 32 COVID-19 patients compared with the control group [53].

Key question 3) Should thromboprophylaxis be used in patients with long COVID?

The NIH guidelines for COVID-19 treatment recommend maintaining anticoagulants or antithrombotic drugs in patients who have been prescribed these drugs for an underlying disease. Routine screening for thrombosis is not recommended in patients without symptoms or signs suggestive of thrombosis. In addition, the administration of anticoagulants or antithrombotic drugs to prevent thrombosis is not recommended in patients with COVID-19 who do not require hospitalization. For patients requiring hospitalization due to COVID-19, the use of low molecular weight heparin (LMWH) or unfractionated heparin (UFH) is recommended for non-pregnant adults. However, the routine administration of drugs is not recommended for thrombosis prevention after discharge [96]. A multicenter randomized controlled clinical study in Brazil reported a significantly lower occurrence of thrombosis in the group of patients with COVID-19 receiving rivaroxaban (10 mg/day) up to 35 days after discharge who were at risk for thrombosis, compared with the control group [97]. However, it is difficult for the results of this study to support long-term thromboprophylaxis in patients with long COVID-19 due to the short duration of drug administration. The NIH guidelines also do not recommend the continued routine administration of drugs to prevent thrombosis in patients with COVID-19 after discharge, except for clinical research purposes. It is recommended to determine whether to prevent thrombosis by considering the risk of thrombosis and bleeding risk regardless of the presence of long COVID-19 [96]. As such, there is no evidence to recommend the routine prevention of thrombosis in patients with long COVID-19. Therefore, it would be reasonable to decide whether to prevent thrombosis in patients with long COVID by considering the general risk of thrombosis and the risk of bleeding according to the underlying diseases and current conditions.

Key question 4) Is general rehabilitation/respiratory rehabilitation necessary for long COVID?

A systematic literature review and meta-analysis evaluating the effects of respiratory rehabilitation in interstitial lung diseases, including COVID-19 reported that respiratory rehabilitation improved patient walking distance and quality of life, as well as dyspnea and lung function [98]. The results of a randomized controlled clinical trial including 72 patients >65 years of age showed that 10 minutes of daily respiratory rehabilitation (respiratory muscle training, cough exercise, diaphragmatic training, stretching exercise, and home exercise) administered for 6 weeks at 6 months after COVID-19 diagnosis improved several indicators of pulmonary function tests (forced expiratory volume in 1 second, FVC, and DLCO), 6MWT, quality of life, anxiety, depression, and activities of daily living [99]. A prospective cohort study that applied a cardiopulmonary rehabilitation program for 2 - 4 weeks from the 10th day after admission in patients with severe COVID-19 who were mostly hospitalized in the ICU observed improvements in 6MWT and mood [100]. A prospective cohort study including 33 patients with COVID-19 who had been hospitalized in ICU and were discharged, reported improved 6MWT and quality of life after performing various therapeutic exercises for 30 minutes daily [101]. In a recent study of 106 patients with long COVID, 44 performed virtual physical therapy, 25 home physical therapy, 17 independent exercise program, and 20 did not perform therapy. The virtual and home physical therapy groups met the clinically meaningful differences for improvements in lower limb strength and cardiopulmonary endurance [102]. In Korea, the National Rehabilitation Center of the Ministry of Health and Welfare has distributed an informational rehabilitation guide for people released from quarantine for COVID-19. The guide contains information on the management of breathing, daily life activities, physical activity and exercise, cognition (concentration and memory), swallowing, and voice [103].

Key question 5) How should persistent respiratory symptoms be treated?

A small prospective study on the use of prednisolone (0.5 mg/kg/day) for 3 weeks in 30 patients hospitalized for COVID-19 and diagnosed with organizing pneumonia 6 weeks after discharge reported improvement of symptoms and DLCO, FVC, and imaging tests in all patients [104]. However, another study reported spontaneous improvement of symptoms within 12 weeks in a patient group with similar characteristics [105]. Therefore, the benefit of corticosteroid treatment is unclear. In addition, the results of a small prospective observational study in which a combination of antihistamines (H1 antagonist + H2 antagonist for at least 4 weeks) was administered to 49 patients at least 3 months after COVID-19 showed improvement in clinical symptoms in 72% of patients [106]. However, the efficacy of steroids and antihistamines in long COVID has not been confirmed by well-designed randomized controlled clinical trials. In addition, no studies have reported the efficacy of ipratropium bromide, aminophylline, or codeine in treating long COVID. Therefore, there is insufficient evidence to recommend or against specific drug treatments for patients with persistent respiratory symptoms. According to the clinical situation of individual patients, the risks and benefits of each drug can be considered for use in symptomatic treatment.

Key question 6) How should smell and taste disturbances be managed?

In the case of complaints of taste abnormalities after the acute phase of COVID-19, the possibility of taste disorders due to dental diseases such as gingivitis or oral diseases such as oral thrush should first be excluded. In addition, it is necessary to check whether a new drug that may cause taste disturbance as a side effect (ACE inhibitor, angiotensin receptor blocker, calcium channel blocker, macrolide, etc.) has been started. Most of the symptoms of taste abnormalities are caused by smell abnormalities; hence, it is necessary to first check whether there is an olfactory disorder [107]. Taste disorders caused by the aftereffects of COVID-19 usually improve over time. There are no research data on the treatment of these taste disorders.

In the case of persistent complaints of olfactory impairment due to long COVID, nasal or sinus diseases such as rhinosinusitis, nasal polyp, and allergic rhinitis should be excluded. Olfactory dysfunction usually improves over time but may persist for several months in some patients. Olfactory dysfunction that improves relatively quickly, is understood to occur due to the closure of the olfactory system by local inflammation. In contrast, long-lasting olfactory dysfunction is explained by damage to the olfactory epithelium or nerve damage to the olfactory system [108]. Research on the treatment of olfactory dysfunction in COVID-19 patients is still lacking. In the meantime, randomized controlled clinical studies on the treatment of postinfectious olfactory dysfunction caused by respiratory viruses can be used as a reference. The Clinical Olfactory Working Group by Addison et al. presented recommendations based on a systematic literature review focusing on randomized controlled clinical studies on the treatment of olfactory disorders after respiratory infections [109]. They recommended olfactory training and quitting smoking to help improve olfactory disorders. In addition, the need for additional research was cited, as high-quality studies on intranasal or systemic steroids, zinc, α-lipoic acid, vitamin A, theophylline, minocycline, or caroverine treatment are lacking [109]. A randomized controlled clinical study of 100 patients with COVID-19 observed no significant difference between olfactory training and treatment with topical corticosteroid nasal spray (mometasone furoate) versus olfactory training alone [110].

As the olfactory nerve is regenerable, it is possible to restore olfactory function through repeated olfactory training. In general, the improvement of olfactory disorders after infection can be expected by repeating the olfactory training to sniff at least four different olfactory stimulants through the nose for 10 seconds each, twice daily, morning and evening [111, 112]. Traditionally, the four odorants are used in olfactory training; namely, clove, eucalyptus, rose, and lemon but the same effect can be expected for odorants more familiar to Koreans [113]. Patients can also perform olfactory training at home using scented essential oils or commercially available smell training kits as odorants. Olfactory training has the advantages of low cost and few systemic side effects, so can be readily used to treat olfactory disorders that persist after COVID-19.

Key question 7) How should fatigue be managed?

Various medications, alternative medicine, cognitive behavioral therapy, and exercise therapy for chronic fatigue syndrome associated with long COVID have been considered. Although some studies have suggested the benefit of rintatolimod, counseling therapy, and step-by-step exercise therapy [114, 115], these studies have limitations in using different fatigue interventions and evaluation methods for the outcome. High-quality evidence-based studies are lacking regarding the effectiveness of treatments and interventions for fatigue in patients with long COVID-19.

Key question 8) How should headache and cognitive symptoms be treated?

Few studies have provided evidence of the appropriate treatment of long COVID-related headache. For patients with preexisting primary headache and COVID-19 as an exacerbating factor, symptomatic and preventive treatment may be considered according to the patient's characteristics and headache features [116]. If the diagnostic criteria for new daily persistent headache or chronic headache due to systemic infection are met, and if the clinical phenotype is similar to migraines, then prophylactic treatment for migraines can be considered. In the case of long COVID-related headache, there is a high risk of the development of medication-overuse headache due to the daily nature of the headache. This medication overuse headace may develop from using triptans or narcotic analgesics for >10 days a month and simple analgesics for >15 days a month for 3 months. Patients should be informed about the risk, and appropriate preventive treatment for such headache can be considered [116]. When a headache attack occurs, non-steroidal anti-inflammatory drugs (NSAIDs) and triptans can be used to treat acute attacks. NSAIDs can be safely used for mild headache attacks, despite early reports that questioned their safety in patients with COVID-19 owing to their potential role in the overexpression of ACE2 [117]. Triptans may be considered for moderate to severe headache attacks. Table 5 shows the most commonly used headache treatments [118].

Table 5
Drugs used for the treatment of headache

• Click for larger image
• Click for full table
• Download as Excel file

When headache is associated with stress from mood disorder, sleep disorder, family, work, and pandemic-related socioeconomic difficulties, flunarizine and beta-blockers may exacerbate depressive symptoms. As topiramate can cause cognitive dysfunction, caution should be exercised if long COVID symptoms accompany cognitive or memory impairment. Along with drug treatment, lifestyle modifications such as maintaining a regular lifestyle, exercising, and avoiding long-term fasting should be performed simultaneously [116]. In particular, since sleep disorders that are common after COVID-19 are associated with headache, treating them may help with headache control [47].

Specific evidence for treating and managing cognitive impairment related to long COVID is still lacking. However, the brain fog seen in long COVID is not severe enough to limit daily life, so it is sometimes compared to chemo brain/chemo fog in that it shows mild abnormalities in cognitive tests or the abnormal findings are not clear [119]. Chemobrain/chemo fog is a condition in which survivors who have received chemotherapy complain of cognitive impairment, such as concentration difficulties and amnesia, without any abnormal findings in cognitive tests. According to the chemobrain/chemo fog-related recommendations suggested by the Mayo Clinic, regular exercise, tracking what affects cognitive dysfunction, and developing stress relief and coping strategies should be considered. In addition, it is necessary to actively manage fatigue, depression, anxiety, and sleep problems, which are associated with long COVID-related cognitive symptoms [120]. While drug treatments such as methylphenidate, modafinil, donepezil, and memantine have been proposed [120], further research is needed. Moreover, luteolin, a natural flavonoid, has been proposed as a potential treatment for long COVID-related brain fog symptoms by inhibiting the activation of mast cells and microglia; however, additional research is needed [121].

Key question 9) How should psychological/psychiatric symptoms be treated?

When SSRIs were administered to 60 patients in Italy reporting major depressive episodes after COVID-19, 92% of patients reported >50% decrease in Hamilton Depression Rating Scale scores after 4 weeks [122]. Clomipramine, a tricyclic antidepressant with an anti-inflammatory action that penetrates the central nervous system, has been proposed as a potential treatment for central nervous system sequelae after COVID-19. However, no randomized controlled clinical trial has been performed [123]. Other antidepressants with increased tolerance, such as SSRIs, may also have anti-inflammatory properties. Clinical research and clinical trials are needed regarding these candidate therapeutics. Fluvoxamine, a type of SSRI, has been proposed as a novel treatment for long COVID [124, 125, 126]. The use of fluvoxamine in the early stages of acute COVID-19 was effective in preventing the worsening of clinical symptoms [127]. Therefore, there is no evidence to recommend routine administration of antidepressants in long COVID patients without mood disorders. If antidepressant administration is necessary, it would be desirable to seek treatment from a psychiatrist.

Research results have also suggested that COVID-19-related alteration in tryptophan absorption and metabolism could be the underlying pathophysiology of long-COVID symptoms [128, 129]. By interfering with the long-term control of L-tryptophan absorption and favoring the kynurenine pathway in metabolism, the increased levels of kynurenine might cause symptoms such as depression, insomnia, fatigue, and muscle weakness, similar to L-tryptophan deficiency. However, it is not yet known whether L-tryptophan administration is an effective treatment for long COVID-19 [130].

Key question 10) Is steroid administration helpful for patients with long COVID?

Several studies have hypothesized that cognitive impairment, headache, and olfactory/taste impairment of long COVID are caused by persistent neuroinflammation [131]. Another hypothesis is that the inflammatory response induced by the coronavirus causes microvascular disorders and pulmonary fibrosis, resulting in cardiovascular and respiratory symptoms [131]. A cross-sectional study in Spain of 121 patients with mild COVID-19 reported that even if the combination of neutrophil count, CRP, and fibrinogen was in the normal range, patients with relatively high levels of these inflammatory markers were more likely to experience long COVID [132]. Despite these hypotheses, it is not clear whether steroid administration lowers the risk of developing long COVID and improves function or symptoms in patients with long COVID. A randomized controlled clinical trial conducted in Egypt observed no significant difference in smell score between the patient group receiving corticosteroid nasal spray therapy for 3 weeks and the control group [110]. An observational study in the UK reported significant improvement in pulmonary function when steroids were administered for 3 weeks to 35 patients with COVID-19 with interstitial lung lesions that persisted beyond 4 weeks. However, there are limitations in interpreting these results, as this was a single-arm study [104]. In addition, several reports suggest that the administration of steroids to treat acute COVID-19 is associated with side effects such as mucormycosis and avascular necrosis [133, 134]. Therefore, when considering steroid administration to improve the function or symptoms of patients with long COVID, the benefits and risks should be individually weighed and decided. In addition, caution should be exercised because there are cases in which COVID-19 aggravates existing adrenal insufficiency or that new onset of adrenal insufficiency after COVID-19 is sometimes mistaken for long COVID.

Key question 11) What special cautions are needed for the diagnosis and treatment of long COVID in children and adolescents?

In children and adolescents, non-specific respiratory, gastrointestinal, and/or systemic symptoms easily occur due to frequent viral infections. Therefore, it is not easy to define symptoms that develop or persist after acute COVID-19 as having been caused by COVID-19. In addition, in adolescents, it is difficult to diagnose long COVID since various somatic and neuropsychiatric symptoms can easily develop due to mental/physical stress. Furthermore, studies for signs/symptoms of long COVID symptoms in children and adolescents are lacking. Therefore, the WHO recommends using different criteria for defining long COVID in children and adolescents. However, no specific criteria have been officially suggested. Nevertheless, the incidence of long COVID in children and adolescents is relatively low when the general definition for adults is applied.

Recent studies reported incidence rates of symptoms lasting up to 4 weeks after COVID-19 diagnosis; 15 - 32% in small prospective observational studies conducted in children <18 years of age [135, 136], and up to 50% in a large retrospective study of only adolescents [137]. The common symptoms in these studies were cough, shortness of breath, fatigue, headache, febrile sensation, insomnia, asthenia, constipation, and abdominal pain. In the CLoCk study conducted in the UK of 3,065 adolescents (11 - 17 years of age) diagnosed between January and March 2021, fatigue (39%), headache (23%), and dyspnea (23%) were the common symptoms present 3 months after COVID-19 diagnosis; however, fatigue (24%) and headache (14%) were also relatively common among 3,739 control adolescents tested negative for COVID-19 [138]. In particular, as fatigue and headache were complaints in 30% and 20% of 11 - 15-year-olds, respectively, even before the pandemic, there is considerable difficulty in considering these to be symptoms of long COVID [139].

In a case-control study of 28,270 adolescents aged 15 – 18 years in Denmark, the common symptoms lasting >2 months after COVID-19 were fatigue (18%), loss of appetite (10%), and headache (9%); however, respiratory difficulty (odds ratio [OR]: 2.7), cough (OR: 1.6), and sore throat (OR: 1.6) are significantly more common in COVID-19 adolescents compared with the control group [137]. In another study comparing 37,522 patients and 78,037 controls aged 0 - 17 years, preschool children with COVID-19 had significantly more symptoms (lasting >4 weeks) of fatigue, loss of smell/taste, and muscle weakness, and school-age children had more symptoms of loss of smell/taste, fatigue, respiratory symptoms, dizziness, muscle weakness, and chest pain compared with the control group [140]. In existing studies, children and adolescents show diverse symptoms with various frequencies after acute COVID-19, and many types of symptoms are subjective or non-specific. Therefore, careful consideration and evaluation to exclude other causes are needed to diagnose long COVID.

Long COVID develops more common in adolescents compared with young infants/children, especially in those with low pre-existing physical or mental health. Recent studies of children with COVID-19 showed no significant association between the incidence of long COVID and the severity of the acute phase or gender, unlike in the results in adults [135, 136]. However, a study that included only adolescents showed that long COVID was more common in girls [137]. A UK study of non-hospitalized adolescents reported that long COVID symptoms involving two or more organs were more likely to be observed in girls compared with boys, in patients aged 15 - 17 years than in those aged 11 - 4 years, and in those with poorer previous physical and mental health before COVID-19 [138]. Therefore, in adolescents who currently have physical or mental comorbidities, the possibility of long COVID should be considered.

Most of the symptoms persisting after acute COVID-19 in children and adolescents are improved within 3 months [135, 136, 140]; thus, they might be resolved spontaneously without meeting the general definition for long COVID. If persistent symptoms beyond 12 weeks after COVID-19 are present, symptomatic treatments recommended in cases of usual viral infection or functional disorder can be considered after sufficiently excluding other causes. There is no evidence to recommend or limit specific drugs. Excessive exercise, physical/mental activity, and stress may contribute to the exacerbation or recurrence of long COVID [6]; therefore, careful attention and management are required regarding these factors. Because meaningful long COVID in adolescents tends to occur as a symptom group involving three or more organs [136, 138], a coordinated multidisciplinary approach is required when interventions such as drug therapy are needed.

Key question 12) Does COVID-19 vaccination affect the development of long COVID?

Vaccination against COVID-19 reportedly reduces the risk of developing long COVID-19 after infection by approximately 15%. In a study conducted in Israel, patients who had received two doses of COVID-19 vaccination before COVID-19 diagnosis were less likely to develop long COVID symptoms than those who had not been vaccinated [141]. In addition, in a UK study comparing vaccinated and unvaccinated patients with similar sociodemographic characteristics, the incidence of long COVID was significantly lower in the vaccinated group (9.5% vs. 14.6%) [20]. However, there remains a lack of definitive evidence on the effect of COVID-19 vaccination on the incidence of long COVID. Therefore, further research is needed.