1-year outcomes in hospital survivors with COVID-19: a longitudinal cohort study
The full range of long-term health consequences of COVID-19 in patients who are discharged from hospital is largely unclear. The aim of our study was to comprehensively compare consequences between 6 months and 12 months after symptom onset among hospital survivors with COVID-19.
We undertook an ambidirectional cohort study of COVID-19 survivors who had been discharged from Jin Yin-tan Hospital (Wuhan, China) between Jan 7 and May 29, 2020. At 6-month and 12-month follow-up visit, survivors were interviewed with questionnaires on symptoms and health-related quality of life (HRQoL), and received a physical examination, a 6-min walking test, and laboratory tests. They were required to report their health-care use after discharge and work status at the 12-month visit. Survivors who had completed pulmonary function tests or had lung radiographic abnormality at 6 months were given the corresponding tests at 12 months. Non-COVID-19 participants (controls) matched for age, sex, and comorbidities were interviewed and completed questionnaires to assess prevalent symptoms and HRQoL. The primary outcomes were symptoms, modified British Medical Research Council (mMRC) score, HRQoL, and distance walked in 6 min (6MWD). Multivariable adjusted logistic regression models were used to evaluate the risk factors of 12-month outcomes.
1276 COVID-19 survivors completed both visits. The median age of patients was 59·0 years (IQR 49·0–67·0) and 681 (53%) were men. The median follow-up time was 185·0 days (IQR 175·0–198·0) for the 6-month visit and 349·0 days (337·0–361·0) for the 12-month visit after symptom onset. The proportion of patients with at least one sequelae symptom decreased from 68% (831/1227) at 6 months to 49% (620/1272) at 12 months (p<0·0001). The proportion of patients with dyspnoea, characterised by mMRC score of 1 or more, slightly increased from 26% (313/1185) at 6-month visit to 30% (380/1271) at 12-month visit (p=0·014). Additionally, more patients had anxiety or depression at 12-month visit (26% [331/1271] at 12-month visit vs 23% [274/1187] at 6-month visit; p=0·015). No significant difference on 6MWD was observed between 6 months and 12 months. 88% (422/479) of patients who were employed before COVID-19 had returned to their original work at 12 months. Compared with men, women had an odds ratio of 1·43 (95% CI 1·04–1·96) for fatigue or muscle weakness, 2·00 (1·48–2·69) for anxiety or depression, and 2·97 (1·50–5·88) for diffusion impairment. Matched COVID-19 survivors at 12 months had more problems with mobility, pain or discomfort, and anxiety or depression, and had more prevalent symptoms than did controls.
Most COVID-19 survivors had a good physical and functional recovery during 1-year follow-up, and had returned to their original work and life. The health status in our cohort of COVID-19 survivors at 12 months was still lower than that in the control population.
Chinese Academy of Medical Sciences Innovation Fund for Medical Sciences, the National Natural Science Foundation of China, the National Key Research and Development Program of China, Major Projects of National Science and Technology on New Drug Creation and Development of Pulmonary Tuberculosis, the China Evergrande Group, Jack Ma Foundation, Sino Biopharmaceutical, Ping An Insurance (Group), and New Sunshine Charity Foundation.
Study design and participants
Briefly, all patients with laboratory confirmed COVID-19 discharged from Jin Yin-tan Hospital between Jan 7 and May 29, 2020, were eligible for participation. Patients were excluded if they died after discharge; were living in a nursing or welfare home; had psychotic disorder, dementia, or osteoarthropathy; or were immobile. To determine whether COVID-19 patients completely recovered at 12 months, we recruited community-dwelling adults without SARS-CoV-2 infection (controls) from two districts of Wuhan city between Dec 24, 2020, and Jan 16, 2021. The inclusion and exclusion criteria are shown in the appendix (p 4). COVID-19 survivors and controls were further matched 1:1 by age, sex, and comorbidities including cardiovascular disease, chronic respiratory disease, chronic kidney disease, hypertension, and diabetes. The maximum allowed age difference between COVID-19 patients and their controls was 10 years.
The study was approved by the Research Ethics Commission of Jin Yin-tan Hospital (KY-2020-78.01, KY-2020-78.03). Written informed consent was obtained from controls and COVID-19 survivors who attended the follow-up visit.
Data collection of COVID-19 patients at acute phase
are described in our previous study
and appendix (p 4). We confirmed the data for demographic and self-reported comorbidity with participants, face to face, at the 12-month follow-up visit.
Follow-up assessment of COVID-19 survivors
At each visit, patients underwent a detailed interview, physical examination, and a 6-min walking test; completed a series of 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 to assess health-related quality of life,
the EuroQol Visual Analogue Scale (EQ-VAS) (scores range from 0–100; a higher score indicates a better health status),
and an ischaemic stroke and cardiovascular event registration form;
and received laboratory tests. Notably, at the 12-month visit, they were also asked to complete a questionnaire to record their health-care use after discharge and work status.
Of participants selected, 349 had completed the pulmonary function tests and 353 chest HRCT at the 6-month visit. The 349 participants who had completed the pulmonary function tests at 6-month visit were all invited to perform this test again at the 12-month visit. Of 353 participants who had completed chest HRCT at 6-month visit, the 186 who presented with abnormal CT were further invited to receive another HRCT scan at the 12-month visit.
Their plasma samples were screened with the Bio-Plex Pro Human Cytokine Screening Panel 27-plex (Bio-rad, Hercules, CA, USA) in Bio-Plex 200 System (Bio-rad). The concentrations of 27 cytokines were measured: interleukin (IL)-1ra, IL-1β, IL-2, IL-4, IL-5, IL-6, IL-7, IL-8, IL-9, IL-10, IL-12, IL-13, IL-15, IL-17, eotaxin, interferon (IFN)-γ-induced protein (IP)-10, monocyte chemoattractant protein (MCP)-1, macrophage inflammatory protein (MIP)-1α, MIP-1β, RANTES, fibroblast growth factors, platelet derived growth factor-BB, vascular endothelial growth factor, granulocyte colony stimulating factor (G-CSF), granulocyte-macrophage colony stimulating factor (GM-CSF), IFN-γ, and tumor necrosis factor (TNF)-α.
Data collection of community-dwelling non-COVID-19 adults
the EQ-5D-5L questionnaire,
Venous blood samples were collected for laboratory tests.
Demographic characteristics and long-term health consequences of COVID-19 in patients are presented as median (IQR) for continuous variables and expressed as absolute values along with percentages for categorical variables. Participants were categorised into three groups according to their severity scale during their hospital stay (scale 3, not requiring supplemental oxygen; scale 4, requiring supplemental oxygen; or scale 5–6, requiring high-flow nasal cannula, non-invasive mechanical ventilation, or invasive mechanical ventilation). Demographic and clinical characteristics and long-term consequences across participants with different categories of severity scale are shown. For the comparison of demographic and clinical characteristics among participants with different disease severity, Kruskal-Wallis test, χ2 test, Fisher’s exact, or Mann-Whitney U test were used when appropriate. For the comparison of symptoms, exercise capacity, and health-related quality of life between 6-month and 12-month follow-up, we used Wilcoxon signed-rank test, or McNemar test when appropriate. The comparison of demographic and clinical characteristics, symptoms, health-related quality of life, and laboratory test results between COVID-19 patients and controls was done with Mann-Whitney U test, χ2 test, or Fisher’s exact test when appropriate.
We used multivariable adjusted logistic regression analysis to explore risk factors associated with diffusion impairment, anxiety or depression, and fatigue or muscle weakness. For the association of disease severity with outcome, age, sex, cigarette smoking, education, comorbidity, corticosteroids, antivirals, and intravenous immunoglobulin were adjusted. For the association of factors including sex, corticosteroid, antiviral, and intravenous immunoglobulin with outcome, the aforementioned variables were all included in the model. When exploring the associations of education and smoking with outcome, the aforementioned variables except for comorbidity, and both comorbidity and disease severity (due to the potential mediation) were included, respectively. Only sex, smoking, and education were adjusted for the association between age and outcome due to the potential mediation of other factors. For the association of comorbidity with outcome, the aforementioned variables except for disease severity were all included. Additionally, a sensitivity analysis with inverse probability-weighted generalised estimating equations was done to reduce the effect of bias due to differences between patients who were included in these analyses and those who were not because of loss to follow-up.
For the comparison of cytokine concentrations at the acute phase, discharge, 6-month follow-up, and 12-month follow-up, Wilcoxon signed-rank test was used. Log10-transformation was done for each cytokine. Partial correlation coefficients between different cytokine pair in COVID-19 patients at discharge, 6-month follow-up, and 12-month follow-up were estimated with adjustment for age, disease severity, and sampling days after symptom onset. For the association of change in cytokine (at discharge until 6-month follow-up) with categorical outcomes at 12-month follow-up, multivariable adjusted logistic regression models were used to estimate the odds ratios (ORs) and 95% CIs per IQR change of log10-transformed cytokine concentration. For the association between change in cytokine concentrations and continuous outcomes, multivariable adjusted linear regression models were used to calculate the β estimates and 95% CIs per IQR change of log10-transformed cytokine concentration. The results following log-transformation were calculated on the basis of geometric mean ratio of cytokines. Age, sex, and corticosteroids were adjusted.
All significance tests were two-sided, and a p value of less than 0·05 was considered statistically significant unless stated otherwise. To correct for multiple comparison of demographic and clinical characteristics between two groups of study participants with different severity scale, we used a Bonferroni corrected α-threshold of 0·0167. To correct for multiple comparison of cytokine concentrations at the acute phase, discharge, 6-month follow-up, and 12-month follow-up, we used a Bonferroni corrected α-threshold of 0·0083. A stringent Bonferroni correction was also used for testing correlation of 351 cytokine pairs, using an α-threshold of 1·4 × 10−4 to determine statistical significance. All statistical analyses were done with SAS (version 9.4). The partial correlation plot was generated in R (version 3.5.2).
Role of the funding source
The funder of the study had no role in study design, data collection, data analysis, data interpretation, or writing of the report.
Table 1Characteristics of COVID-19 patients who completed both 6-month and 12-month follow-up
Data are median (IQR), n (%), or n/N (%) when data are missing. The differing denominators used indicate missing data. To correct for multiple comparison between two groups of study participants with different severity scale, a Bonferroni corrected α-threshold of 0.0167 was used. HFNC=high-flow nasal cannula for oxygen therapy. NIV=non-invasive ventilation. IMV=invasive mechanical ventilation. ECMO=extracorporeal membrane oxygenation. ICU=intensive care unit. NA=not applicable.
Table 2Sequelae symptom, exercise capacity, and health-related quality of life among COVID-19 patients at 6-month and 12-month follow-up
Data are median (IQR), n (%), or n/N (%) when data are missing. The differing denominators used indicate missing data. p value indicates the comparison of consequences between 6 months and 12 months in total or each category of scale. HFNC=high-flow nasal cannula for oxygen therapy. NIV=non-invasive ventilation. IMV=invasive mechanical ventilation. mMRC=modified British Medical Research Council. EQ-5D-5L=EuroQol five-dimension five-level questionnaire.
Table 3Lung function and chest CT among COVID-19 patients at 6-month and 12-month follow-up according to severity scale
Data are absolute values, n (%), or n/N (%) when data are missing. HFNC=high-flow nasal cannula for oxygen therapy. NIV=non-invasive ventilation. IMV=invasive mechanical ventilation. FEV1=forced expiratory volume in 1 s. FVC=forced vital capacity. TLC=total lung capacity. FRC=functional residual capacity. RV=residual volume. DLCO=diffusion capacity for carbon monoxide. GGO=ground glass opacity. NA=not applicable.
Table 4Health-care use after discharge until 12-month follow-up, and work status at 12-month follow-up among COVID-19 patients
Data are n or n/N (%). ICU=intensive care unit.
To our knowledge, this is the largest longitudinal cohort study of hospital survivors with COVID-19 so far to describe the dynamic recovery of health consequences within 12 months after symptom onset. We found that most patients had a good physical and functional recovery during follow-up, and the majority of study participants who were employed before COVID-19 had returned to their original work. However, sequelae symptoms, lung diffusion impairment, and radiographic abnormalities persisted to 12 months in some patients, especially in patients who were critically ill during hospital stay. The current health status in the COVID-19 cohort was still lower than that in the control population.
and lasted to 2 years.
Fatigue was the most commonly reported symptom of patients with SARS, which could last as long as 4 years.
We found that female sex and corticosteroid therapy at acute phase were risk factors for fatigue or muscle weakness at 12 months. The cause and pathogenesis of fatigue and muscle weakness after COVID-19 are unclear, but on the basis of previous evidence in SARS, lung diffusion capacity impairment and some extrapulmonary causes, including viral-induced myositis at initial presentation, cytokine disturbance, muscle wasting and deconditioning, or corticosteroids myopathy, or a combination of these factors, could have contributed to the condition.
Al-Aly and colleagues
reported that COVID-19 survivors had a high burden of incident use of bronchodilators, antitussives, expectorants, antidepressants, and anxiolytics after COVID-19. The chronic or late-onset psychological symptoms after COVID-19 could be driven by a direct effect of virus infection and might be explained by several hypotheses including aberrant immune response, hyperactivation of the immune system, or autoimmunity.
Additionally, indirect effects including reduced social contact, loneliness, incomplete recovery of physical health, and loss of employment could affect psychiatric symptoms.
Lung structural abnormality during late recovery of SARS was associated with the lung diffusion impairment;
however, the association during convalescence after COVID-19 was unclear. We undertook an initial exploratory analysis on the basis of a small group of patients and found that lung imaging patterns at 12 months might be associated with lung diffusion impairment, which should be confirmed in a larger sample study. Previous SARS follow-up studies have shown that persistent lung diffusion impairment could last for months or even years.
Hence, a longitudinal study is needed to describe the natural history of lung structural and functional abnormality after COVID-19, and to explore the effect of these persistent abnormalities on physical function and quality of life.
Our study had several limitations. First, the moderate response rate could have introduced bias to our study. Fortunately, we recorded no significant difference in most baseline characteristics between COVID-19 patients who were included in final analysis and those who were not. The sensitivity analyses for risk factors associated with primary outcomes that used generalised estimating equations to reduce the effect of bias also showed similar results. Second, this is a single centre study focused on previously hospitalised COVID-19 patients in the early stage of the pandemic, which limits the representativeness of this cohort. Moreover, a low proportion of patients with ICU admission in our cohort limits the generalisability of the study findings to this particular population. Future large sample studies are needed to evaluate the long-term consequences of COVID-19 in patients with varying severity, including outpatients, inpatients, and patients requiring admission to ICU. Third, we did not have the health status of COVID-19 survivors before acute infection. However, the health status of matched non-COVID-19 controls could represent the baseline state of COVID-19 patients, although residual confounders cannot be excluded. The comparison between COVID-19 patients and controls indicated whether COVID-19 patients completely recovered at 12 months. Finally, the small sample size of participants with cytokines tests could have affected the reliability of the association between change in cytokines concentrations and 12-month outcomes. These findings should be interpreted as exploratory and need to be validated in a future study with a lager sample.
Within 1 year after acute infection, most hospital survivors with COVID-19 had a good physical and functional recovery over time, and had returned to their original work and life, but current health status was still lower than that in the control population. Lung diffusion impairment and radiographic abnormalities were still common in critically ill patients at 12 months. Ongoing longitudinal follow-up is needed to better characterise the natural history and pathogenesis of long-term health consequences of COVID-19.
BC, XW, and JW had the idea for and designed the study. They had full access to all the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis. LH, BC, XG, YW, JXu, XZ, LR, and LG drafted the paper. BC, XG, LH, LR, and LG did the analysis, and all authors critically revised the manuscript for important intellectual content and agreed to submit the final version for publication. QY, QW, PH, YQ, YF, XL, CL, TY, JXia, MW, LC, YL, FX, DL, XG, and LH completed the follow-up work. YQ, YF, XL, CL, TY, JXia, MW, LC, YL, FX, DL, XG, LH, LG, LR, and ML collected and verified the data. All authors agree to be accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved.