The global pandemic of COVID-19, caused by SARS-CoV-2, has resulted in over 83 million confirmed cases and 1.8 million deaths as of January 4, 2021. While the epidemiological, clinical, and pathological characteristics of COVID-19 patients in the acute phase have been documented, the long-term consequences of the illness remain poorly understood. Long-term follow-up studies are urgently needed to investigate persistent symptoms, lung function, physical and psychological issues in discharged patients. Previous studies, with small sample sizes and a maximum follow-up duration of 3 months, have reported persisting symptoms like fatigue, dyspnea, impaired lung function, and chest image abnormalities in patients following hospital discharge. However, the full spectrum of post-discharge characteristics and extra-pulmonary organ manifestations remain unknown. This study aims to comprehensively describe the long-term consequences of COVID-19 in patients after hospital discharge and identify potential risk factors, including disease severity, associated with these consequences.

A comprehensive search of PubMed for follow-up studies on the long-term consequences of COVID-19 was conducted up to November 5, 2020, without any language restrictions. The search terms included "COVID-19 OR SARS-CoV-2 OR Coronavirus disease 2019 OR 2019-nCoV" AND "survivor* OR recover* OR persistent OR follow up OR discharge* OR long term OR sequelae." The studies reviewed revealed that patients with COVID-19 discharged from hospitals might experience persistent symptoms, abnormal chest imaging patterns, impaired lung functions, and a decline in quality of life. However, due to limited sample sizes and short follow-up durations (up to 3 months), the representativeness and explicitness of the information provided were insufficient. The long-term health consequences of discharged patients with COVID-19 and the associated risk factors remained unclear.

This ambidirectional cohort study was conducted at Jin Yin-tan Hospital, the first designated hospital for COVID-19 patients in Wuhan, Hubei, China. All patients with laboratory-confirmed COVID-19 discharged from the hospital between January 7 and May 29, 2020, were included. Patients were excluded if they had died before follow-up, had difficulty with follow-up due to psychiatric disorders, dementia, or readmission to the hospital, were unable to move freely due to accompanying osteoarthropathy or immobility before or after discharge, declined participation, could not be contacted, or lived outside Wuhan or in nursing or welfare homes. All discharged patients met uniform discharge criteria as outlined by the Chinese clinical guidance for COVID-19 pneumonia diagnosis and treatment issued by the National Health Commission. The study was approved by the Research Ethics Commission of Jin Yin-tan Hospital, and written informed consent was obtained from all participants. The acute phase was defined as the time between symptom onset and hospital discharge. Clinical data from the acute phase was retrieved from electronic medical records, including demographics, clinical characteristics, laboratory test results, and treatment details. Disease severity was categorized using the highest seven-category scale during the hospital stay, ranging from 1 (not admitted to hospital with resumed normal activities) to 7 (death). Data was managed using REDCap electronic data capture tools to minimize missing inputs and ensure real-time data validation and quality control. Follow-up appointments were scheduled via telephone by trained medical staff in the order of symptom onset date documented in the medical record. Participants were given two opportunities to reschedule if they missed their initial appointment. Follow-up consultations were conducted in the outpatient clinic of Jin Yin-tan Hospital. Participants were interviewed face-to-face by trained physicians and completed questionnaires, including a self-reported symptom questionnaire, the modified British Medical Research Council (mMRC) dyspnoea scale, the EuroQol five-dimension five-level (EQ-5D-5L) questionnaire, the EuroQol Visual Analogue Scale (EQ-VAS), and an ischaemic stroke and cardiovascular event registration form. They also underwent a physical examination and a 6-minute walking test. Blood samples were collected from all participants for complete blood count, serum creatinine, haemoglobin, and glycated haemoglobin A1c (HbA1c). Additionally, SARS-CoV-2 antibody concentrations were measured for participants previously enrolled in the Lopinavir Trial for Suppression of SARS-CoV-2 in China (LOTUS). Plasma samples at acute phase and follow-up were analyzed simultaneously. Immunoglobulin (Ig) M, IgA, and IgG antibodies against the nucleoprotein, spike protein, and receptor binding domain of the spike protein were evaluated using enzyme-linked immunosorbent assay. Neutralizing antibodies were titrated on Vero cells using a microneutralization assay. To further investigate pulmonary function, ultrasonography of lower limb veins and abdomen, and chest high-resolution CT (HRCT) were performed on a stratified sample of patients, with a disproportional random sampling procedure according to severity scale. Participants requiring high-flow nasal cannula (HFNC), non-invasive mechanical ventilation (NIV), or invasive mechanical ventilation (IMV) (severity scale ≥5) were all invited to receive the pulmonary function test, ultrasound, and chest HRCT. The ratio used to select patients not requiring supplemental oxygen (severity scale 3) and those requiring supplemental oxygen (severity scale 4) was 1:2. The pulmonary function test was performed using the Master Screen PFT according to American Thoracic Society guidelines. Chest HRCT was conducted in the supine position during end-inspiration, with images reconstructed at 1 mm slice thickness. The final chest CT images during the hospital stay and follow-up images were cross-compared. The CT features were evaluated by an experienced radiologist and pulmonologist, and a validated artificial intelligence software was used to calculate the extent of anatomic involvement of each lobe. A semiquantitative CT score was calculated to assess pulmonary involvement. Estimated glomerular filtration rate (eGFR) was calculated using the Chronic Kidney Disease-Epidemiology Collaboration equation. Primary outcomes included symptoms (fatigue or muscle weakness, sleep difficulties, hair loss, and smell disorder), exercise capacity (distance walked in 6 minutes), health-related quality of life (pain or discomfort, anxiety or depression, mobility, personal care, and usual activity), lung function, and chest CT pattern at follow-up. Secondary outcomes included extrapulmonary organ function (eGFR, HbA1c, deep venous thrombosis of lower limbs, and ultrasonographic features of kidney, liver, spleen, and pancreas), antibody titres, and seropositivity. Statistical analysis included descriptive statistics for demographic characteristics and long-term health consequences, with participants categorized into three groups based on their severity scale (3, 4, and 5-6). Multivariable adjusted linear or logistic regression models were used to evaluate the association between disease severity and long-term health consequences, adjusting for confounders like age, sex, cigarette smoking, education, comorbidity, corticosteroids, antivirals, and intravenous immunoglobulin. Paired t tests and McNemar tests were used for comparing antibody test results at acute phase and follow-up.

Out of 2469 discharged patients with COVID-19, 1733 were enrolled after 736 were excluded. The median age of the enrolled participants was 57·0 years (IQR 47·0-65·0), with 897 (52%) males and 836 (48%) females. The follow-up study was done from June 16 to Sept 3, 2020, and the median follow-up time after symptom onset was 186·0 days (175·0-199·0). The most common symptoms were fatigue or muscle weakness (52%, 855 of 1654) and sleep difficulties (26%, 437 of 1655). Anxiety or depression was reported among 23% (367 of 1616) of patients. The proportions of 6-min walking distance less than the lower limit of the normal range were 17% for those at severity scale 3, 13% for severity scale 4, and 28% for severity scale 5-6. The study found that at 6 months after acute infection, COVID-19 survivors were mainly troubled with fatigue or muscle weakness, sleep difficulties, and anxiety or depression. Patients who were more severely ill during their hospital stay had more severe impaired pulmonary diffusion capacities and abnormal chest imaging manifestations, and are the main target population for intervention of long-term recovery. The seropositivity and median titres of the neutralising antibodies were significantly lower than at acute phase. The seropositivity of N-IgM, RBD-IgM, S-IgM, N-IgA, RBD-IgA, S-IgA, and RBD-IgG at follow-up significantly decreased compared with that at acute phase. However, the seropositivity of N-IgG and S-IgG antibodies did not show significant change. More than 90% of participants tested positive for all three IgG antibodies at follow-up. The prevalence of presenting at least one symptom among participants with scale 5-6 was higher than those with scale 3 (OR 2·74 [95% CI 1·42-5·30]). The prevalence of an mMRC score greater than or equal to 1 was significantly higher in participants with scale 5-6 than those with scale 3 (OR 2·18 [95% CI 1·30-3·65]). Participants with scale 5-6 had more problems in mobility, pain or discomfort, and anxiety or depression than did those with scale 3 (all p<0·05). 23% (367 of 1616) of participants reported anxiety or depression at follow-up, which was more common in females. Compared with participants with scale 3, participants with scale 5-6 presented with shorter walking distance in meters in 6 min (487·5 [IQR 412·0-525·0] vs 490·0 [448·5-535·0]) and a higher proportion of less than the lower limit of the normal range (LLN); however, no significant difference was observed for participants with scale 4. The proportion of patients with a 6-min walking distance less than LLN was 17% (74 of 428) for scale 3, 13% (144 of 1147) for scale 4, and 28% (32 of 114) for scale 5-6. The proportion of participants with lung diffusion impairment was 22% (18 of 83) for scale 3, 29% (48 of 165) for scale 4, and 56% (48 of 86) for scale 5-6. A significant difference was observed between scale 3 and scale 5-6, but not between scale 3 and scale 4. In the subgroup analysis by sex, both males and females with scale 5-6, and males with scale 4 had higher prevalence for decreased lung diffusion capacity than did those with scale 3 (all p<0·05). Decreased total lung capacity (<80% of predicted values) did not show a significant difference in participants with scale 4 or scale 5-6 compared with those with scale 3. The median CT scores are 3·0 (IQR 2·0-5·0) for participants at scale 3, 4·0 (3·0-5·0) for participants at scale 4, and 5·0 (4·0-6·0) among participants at scale 5-6, with a significant difference between scale 3 and scale 5-6 (p=0·0005), which is also observed in the subgroup analysis by sex. Additionally, males at scale 4 had a significantly higher CT score than did those at scale 3 (p=0·028). Ground glass opacity (GGO) is the most common HRCT pattern at follow-up, followed by irregular lines. Among participants with eGFR available at follow-up, 35% (487 of 1393) had decreased eGFR (<90 mL/min per 1·73 m²). 101 (6%) of 1706 patients had acute kidney injury at acute phase. Among participants with eGFR available both at acute phase and follow-up, 479 of 1378 had decreased eGFR at follow-up. Of 1016 participants with non-acute kidney injury and normal eGFR value at acute phase, 822 had eGFR available at follow-up, with 107 (13%) presenting with decreased eGFR.

This study provides important insights into the long-term health consequences of COVID-19 in patients discharged from hospital. The study’s findings suggest that fatigue, muscle weakness, sleep difficulties, and anxiety or depression are common symptoms that persist for up to 6 months after symptom onset. Patients with more severe illness during their hospital stay are at a higher risk for impaired pulmonary diffusion, fatigue, and psychological distress. Additionally, the study found a decline in neutralizing antibody titres and seropositivity at 6 months compared to the acute phase, raising concerns about the potential for re-infection. The prevalence of persistent fatigue and psychological distress aligns with findings from previous SARS long-term follow-up studies, suggesting that COVID-19 may have similar long-term effects. The higher prevalence of these symptoms in females, as observed in this study, is consistent with previous research on SARS survivors. The underlying mechanisms for these psychiatric consequences are likely multifaceted and may include viral infection, immune response, corticosteroid therapy, ICU stay, social isolation, and stigma. The significant proportion of participants with impaired pulmonary diffusion at 6 months highlights the potential for persistent lung complications after COVID-19. The common abnormal CT pattern of pulmonary interstitial change, resembling long-term lung manifestations of SARS and influenza, suggests that respiratory viral infections may induce distinct fibroblast activation during the convalescence phase. Disease severity during the acute phase was found to be associated with impaired pulmonary diffusion and changes in CT scores, emphasizing the need for targeted interventions for severely ill patients. The decline in neutralizing antibodies observed in this study and other research raises concern for SARS-CoV-2 re-infection. Further investigation is required to clarify the dynamic changes of antibodies against SARS-CoV-2 and the implications for re-infection risk. The study also highlights the potential for long-term extrapulmonary organ manifestations, including persistent renal dysfunction and newly diagnosed diabetes, emphasizing the importance of ongoing follow-up care for discharged patients with COVID-19.

This large cohort study with an extended follow-up duration provides valuable information regarding the long-term consequences of COVID-19 in patients discharged from hospital. Fatigue, muscle weakness, sleep difficulties, and anxiety or depression were common symptoms observed at 6 months after symptom onset. Patients with more severe illness during hospitalization experienced a higher prevalence of impaired pulmonary diffusion, fatigue, and anxiety or depression. The decline in neutralizing antibodies raises concern for SARS-CoV-2 re-infection, underscoring the need for continued monitoring of patients with new symptoms of COVID-19. These findings underscore the importance of post-discharge care for patients with COVID-19, particularly those with severe illness. Longer follow-up studies in a larger population are necessary to understand the full spectrum of health consequences from COVID-19 and to develop effective strategies for long-term recovery.

This study has several limitations. First, the baseline data for pulmonary function and 6-minute walking distance were unavailable. The observed impaired pulmonary function and exercise capacity cannot be directly attributed to COVID-19. Second, the data on new symptom onset after COVID-19 were not stratified to determine whether symptoms were persistent, worsened, or occurred post-discharge. Third, patients with mild COVID-19 symptoms who stayed in Fangcang shelter hospitals were not enrolled, limiting the generalizability of the findings to all patients. Lastly, the number of participants with SARS-CoV-2 antibody test results both at acute phase and follow-up was limited, necessitating further investigation with a larger sample to clarify the dynamic changes of antibodies against SARS-CoV-2.