Tong

Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), the pathogen of coronavirus disease 2019 (COVID-19), is highly contagious and pathogenic and is responsible for more than 351 million infected and nearly 5.6 million deaths worldwide as of Jan 24, 2022.[1] In addition, persistent and diverse post-COVID symptoms have been described in survivors of COVID-19, including those with a mild initial disease course.[2] Therefore, more than 340 million survivors are at high risk for post-COVID syndrome worldwide.[3]
The available clinical evidence suggests that COVID-19, although damaging the respiratory system initially, is a systemic disease with extrapulmonary complications.[4] The cardiovascular system is one of the most involved systems,[5] while endotheliitis is a prominent feature of COVID-19,[6] thus is suggested to be responsible for life¬threatening thrombogenesis and coagulopathy in those with severe illness.[7] During the acute phase of SARS-CoV-2 infection, cytokine storm and subsequent endothelial injury and thrombosis are involved in the pathogenesis of cardiovascular complications.[8] However, few studies have focused on the endothelial dysfunction in patients who recovered from COVID-19.
Endothelial cells play an essential role in maintaining vascular homeostasis, such as controlling inflammation, regulating platelet aggregation, and preventing thrombosis.[9] Dysfunction of endothelial cells has been identified as a central feature of COVID-19.[10] The abnormal elevation of soluble endothelial biomarkers, such as vascular cell adhesion molecule-1 (VCAM-1), intercellular cell adhesion molecule-1 (ICAM-1), P-selectin, and fractalkine, is closely related to the development of arteriosclerosis,[11] which is the underlying pathology of coronary artery disease,[12] peripheral artery disease,[13] and cerebrovascular disease[14] in most cases. Therefore, the severity of endothelial dysfunction is associated with increased cardiovascular risks, and it is of great significance to monitor endothelial biomarkers in patients recovering from COVID-19.
In this study, we investigated demographics, laboratory findings, symptoms, electrocardiogram manifestations, screened lower extremity thrombosis and measured serum endothelial biomarkers of participants one year after discharge, thus evaluating long-term cardiovascular risk in patients recovered from COVID-19.

Study design and participants. From Mar 16 to Mar 28, 2021, 473 survivors of COVID‐19, who had been previously hospitalized from Jan 24 to Mar 18, 2020, in Huanggang, Hubei, China, were recruited to this cross-sectional cohort study. The inclusion criteria were adults previously diagnosed with COVID‐19 (positive in a reverse‐transcription polymerase chain reaction for SARS‐CoV-2), and the stratification of disease severity has been described in our published report.[15] Of these patients, 114 cases were excluded for diabetes, suffering from chronic systemic infection, malignant tumors or hematological and autoimmune diseases, pregnancy, chronic smoking (defined as 20 pack-years), long-term use of medications (angiotensin-converting enzyme inhibitors, angiotensin II type-1 receptor antagonists, corticosteroids or statins), and fourteen recovers did not show up for the follow-up appointment (Figure 1). As a result, 345 survivors were recruited into the study from Mar 1 to 30, 2021 from Mar 1 to May 30, 2021. During this time, 119 age and sex-matched healthy controls, medical personnel at Hunan Provincial People's Hospital, were recruited during the annual routine physical examination. All medical staff has repeatedly undergone throat swab screening to exclude SARS-CoV-2 infection every 1 to 2 weeks since the pandemic. Moreover, those who were pregnant, long-term using medications or chronic smoking, or suffering from diabetes, chronic systemic infection, malignant tumors or hematological and autoimmune diseases were excluded.

Figure 1. Flow chart of patients with COVID-19 discharged from Huanggang Hospitals between January 24 to March 18, 2020.
#A self-reported symptom questionnaire included Low-grade fever, Fatigue or muscle weakness, Palpitations, Chest tightness or shortness of breath, Dizziness. *Laboratory tests included routine blood tests, biochemical and immunologic tests, the plasma levels of CRP, IL-6, and D-dimer, and the serum levels of VCAM-1, ICAM-1, P-Selectin and fractalkine.
Abbreviations: ACEIs: angiotensin-converting enzyme inhibitors; ARBs: angiotensin II type-1 receptor antagonists.

Collection of clinical data. According to the guidelines of the National Health Commission of China, in the cohort, survivors were divided into the mild group and the severe group according to the severity of the disease during previously acute infection, as described in our previous report.[15] In addition, all the survivors were subjected to a standardized symptom questionnaire and received a physical examination, electrocardiogram, and ultrasonography of the lower extremities for detecting deep venous thrombosis. All data were collected and triple-checked by three physicians.
Sample collection and processing. Blood samples were taken from each participant by standard venipuncture in a fasting state on the day of appointments. Routine blood tests, biochemical and immunologic tests, and the plasma levels of C-reactive protein (CRP), interleukin (IL)-6, and D-dimer were measured by conventional laboratory methods. The serum for endothelial biomarker detection was isolated by centrifugation for 15 minutes at 1500×g and frozen at -80°C until thawed and analyzed.
This study has strictly followed the recommendations of the Helsinki Declaration. Therefore, the institutional review boards of the Medical Ethics Committee of the Hunan Provincial People's Hospital approved this study (NO.2021-92), and all participants signed informed consent.
Enzyme-linked immunosorbent assays for serum endothelial biomarkers measurement. Quantitative measurement of serum soluble VCAM-1, ICAM-1, P-Selectin and fractalkine was tested for survivors and healthy participants using 96-well enzyme-linked immunosorbent assay kits (Boster Biological Technology Co. Ltd, Wuhan, China). Quality control was carried out strictly following the manufacturer's instructions for each batch of tests.
Statistic analysis. Categorical variables were compared using χ2 analysis and expressed in numbers (proportions). Continuous variables with normal distribution were compared using independent group t-tests and expressed as mean ± standard deviation (SD), while those not normally distributed were compared using the Mann-Whitney U test and expressed as median and interquartile range (IQR) values. All statistical analyses were performed using the SPSS programme, V.19.0 (SPSS Inc., Chicago, IL, USA), and plots were generated using GraphPad Prism, version 8 (GraphPad Software, San Diego, CA). A two-sided P-value of < 0.05 was defined as statistically significant.

Clinical characteristics of the study participants. A total of 345 COVID-19 survivors (291 recovered from the mild situation and 54 from the severe) and 119 age and sex-matched healthy medical volunteers participated in this study. Demographic information and laboratory findings are shown in Table 1. In our cohort, 46 male and 73 female medical volunteers were enrolled as healthy controls, while 118 male and 173 female survivors and 19 adult males and 35 females who recovered from mild and severe situations were recruited. The median ages in the control, mild, and severe groups were 52, 53, and 54 years, respectively, and the average visiting interval after discharge for the recovers was 375.0 days (SD, 11.0 days). No significant differences were found among the mild, severe and control participants in sex, age, systolic or diastolic brachial artery pressure, BMI, circulatory level of leukocytes, neutrophils, lymphocytes, hemoglobin, D-dimer, glucose, triglycerides (TG), total cholesterol (TC), alanine aminotransferase (ALT), C-reactive protein (CRP) and estimated glomerular filtration rate (eGFR). Furthermore, there were no significant differences in plasma interleukin-6 (IL-6) levels between the mild and severe groups. For all survivors, no venous thrombosis was observed by lower extremities ultrasound.

Table 1. Demographics and laboratory findings of participants.

Post-COVID symptoms of participants. Regarding the post-COVID symptoms, a standardized symptom questionnaire was performed and presented in Table 2. In general, 135 (39%) recovered patients had persistent symptoms during the 1-year follow-up and no significant differences were found between the severe and mild groups. Among them, 4 (1%) subjects complained of persistent low-grade fever, and 128 (37%) recovers were troubled by fatigue or muscle weakness. For cardiovascular disease-related symptoms, 21 (6%) participants possessed persistent palpitations, 10 (3%) subjects had chest tightness or shortness of breath, and 8 (2%) complained of dizziness. In addition, there was no significant difference observed in those post-COVID symptoms between the mild and severe groups.

Table 2. Symptoms, electrocardiogram manifestations and lower extremities Doppler ultrasound of recovers.

Electrocardiogram abnormalities in patients recovering from COVID-19. An electrocardiogram examination was performed for each patient who recovered from COVID-19 at a one-year follow-up. As presented in Table 2, the most frequent abnormalities are ST-T changes (16%) and sinus bradycardia (11%), the frequency of left ventricular high voltage (6%), sinus tachycardia (1%), prolonged PR interval (1%) and other abnormalities (such as ventricular premature contraction, atrial fibrillation, prolonged Q-T interval) is relatively small. Of those abnormalities, the frequency of prolonged PR interval seemed to be positively related to the previously infected disease severity (P=0.015), while there was no significant difference in the frequency of other abnormalities between the mild and severe groups.
Serum levels of endothelial biomarkers. Compared with the control group, the serum VCAM-1 levels showed no significant differences in patients who recovered from mild (median, 1.69 vs 1.67 ng/mL, P=0.363) or severe (median, 1.69 vs 1.67 ng/mL, P=0.962) situation, in line with that between mild and severe recovers (P=0.553) (Figure 2A). Although not significant, serum ICAM-1 levels were lower in the mild group than in controls (median, 427.3 vs 469.7 pg / mL, P=0.139) and the severe group than in the control (median, 424.6 vs 469.7 pg/mL, P=0.789) or in the mild groups (median, 424.6 vs 427.3 pg / mL, P=0.444) (Figure 2B). Similarly, serum P-selectin levels were lower in the mild group than in controls (median, 965.6 vs 1076.6 pg / ml, P=0.296) and in the severe group than in the control (median, 960.4 vs 1076.6 pg / ml, P=0.104) or mild groups (median, 960.4 vs 965.6 pg / ml, P=0.260) (Figure 2C). Regarding fractalkine, the mild group showed lower serum levels than the control (mean values, 204.5±24.0 vs 206.8±25.0 pg/mL, P=0.378) and the severe groups (mean values, 204.5±24.0 vs 206.8±19.0 pg/mL, P=0.493), and there were no significant differences between the control and the severe groups (P=0.992) (Figure 2D).

Figure 2. Serum levels of vascular cell adhesion molecule-1(VCAM-1), intercellular cell adhesion molecule-1(ICAM-1), P-selectin, and fractalkine in controls and COVID-19 recovers of previously mild or severe infected.

Principal Findings of Our Study. This study is the first to report endothelial biomarkers of patients who recovered from COVID-19 one year after discharge. In our cohort, a considerable number of survivors are still bothered by post-COVID symptoms. Secondly, inflammatory markers of the normal range, including neutrophils, CRP, and IL-6, indicate the remission of inflammatory reactions in those survivors. In addition, significant elevated D-dimer levels and deep vein thrombosis were absent in all screened survivors, suggesting a relatively low risk of coagulopathy in the long term. Furthermore, the levels of circulating endothelial biomarkers, including VCAM-1, ICAM-1, P-selectin and fractalkine, do not show significant differences in those survivors and healthy controls, implying that SARS-CoV-2 infection may not impose a higher risk of the development of long-term cardiovascular events, even for those recovering from severe illness.
Comparison with related studies. As previously observed in the SARS epidemic,[16] recovered patients have persistent symptoms and unexpected higher rates of diabetes, respiratory and cardiovascular disease, named the post-COVID syndrome after SARS-CoV-2 infection.[3,17] Until now, few clinical studies have focused on cardiovascular sequelae in the aftermath of COVID-19. In a 3-month follow-up study, myocardium injury was detectable in 30% of recovered COVID-19 patients by cardiac magnetic resonance (CMR),[18] while in a cohort of twenty-six patients who recovered from COVID-19 who reported cardiac symptoms and underwent CMR examinations, fifteen (58%) of them had abnormal CMR findings, including myocardial edema, fibrosis, and impaired right ventricle function.[19] In another study of a cohort of 100 German patients who recently recovered from COVID-19 infection, CMR imaging revealed cardiac involvement in 78 patients (78%) and ongoing myocardial inflammation in 60 patients (60%).[20] The prevalence of cardiovascular complications is alarmingly in those studies, indicating the existence of the short to medium-term cardiovascular consequences of COVID-19. However, a reasonable explanation is that the appearance of new or persistent symptoms in the cohorts could increase the positive CMR detection rate, implying that some of these patients are not genuine 'convalescent patients'. However, inconsistent with the studies above, in a single-center longitudinal study, 13% of COVID-19 survivors experienced significant cardiovascular symptoms three months after discharge, including an increase in resting heart rate, occasional palpitations, and newly diagnosed hypertension requiring blood pressure-lowering medications.[21] At the same time, in an observational prospective multicentre trial 60 and 100 days after confirmed diagnosis, cardiac impairment, including reduced left ventricular function or signs of pulmonary hypertension, was present only in a minority of subjects.[22] Moreover, in another preliminary 6-month follow-up study, no survivor reported any obvious cardiopulmonary symptoms, although 29.6% (8/27) of them were detected cardiac injury by CMR.[23] These findings provide the contradictory prevalence of SARS-CoV-2 infection in short- or medium-term cardiovascular sequelae, and most of the conclusions are descriptive and imaging-based, lacking objective biomarkers and long-term follow-up data.
A large number of studies have suggested that endothelial function reflects the comprehensive influence of various risk factors on the vascular system,[24] and endothelial dysfunction is an early predictor of subclinical atherosclerosis[25] and subsequent long-term cardiovascular events.[26] Therefore, early detection of soluble endothelial biomarkers contributes to early detection of disease, quantification of risk, and early intervention to reduce the incidence of cardiovascular adverse events in patients.[27] Both indirect induction by hypercytokinemia (e.g., IL-1, IL -6 and tumor necrosis factor-alpha), hyperchemokinemia and coagulopathy (named after a high-inflammatory response),[28] and direct damage to endothelial cells by SARS-CoV-2 infection contributed to endothelial injuries in patients with COVID-19,[6,29] thus improving the expression of endothelial biomarkers, including ICAM-1, VCAM-1, P-selectin, and fractalkine.[15,30] However, few studies have focused on the alterations of endothelial biomarkers and cytokines in patients recovering from COVID-19. In a prospective longitudinal multicenter cohort study, regulators of endothelial activation such as vascular endothelial growth factor (VEGF), brain-derived neurotrophic factor (BDNF), and macrophage inflammatory protein-1β (MIP-1β) were persistently elevated in convalescence patients with COVID-19, potentially promoting the development of atherosclerosis and cardiovascular sequelae.[31] In another 3-month follow-up study, persistent abnormal levels of endothelial biomarkers, pro-inflammatory cytokines and chemokines (VCAM-1, ICAM-1, TNF-α, MIP-1α, and MIP-1β) were observed in those recovered from COVID-19, especially in severe/critical patients.[32] Therefore, by describing the post-COVID symptoms and abnormal ECG changes, detection of both lower extremities thrombosis, and measuring the circulatory levels of inflammatory factors and endothelial biomarkers, our study may provide a more comprehensive cardiovascular perspective for COVID-19 recovers one year after discharge.
In our cohort, the influence of various confounders, such as older age, pregnancy, chronic smoking, preexisting conditions (malignancy, diabetes mellitus, hyperlipidemia, obesity), and current medications was strictly excluded, thus may not reflect the overall situation of cardiovascular sequelae in COVID-19 recovers with a preexisting higher risk of endothelial dysfunction. In the study, relatively low levels of circulating endothelial biomarkers at one-year follow-up for those survivors without preexisting endothelial dysfunction risks were observed, although injuries to endothelium cells are believed to have long-lasting effects.[33] Several mechanisms were speculated to be majorly ascribed to the remission of endothelial biomarkers. First, the clearance of SARS-CoV-2 infection in all recruited patients, confirmed by repeated screening after discharge, is conducive to the remission of endothelial biomarkers. Second, with the clearance of viral infection, the levels of inflammatory markers (such as neutrophils, CRP, and IL-6) returned to normal, and a coordinated and dynamic immune response, characterized by reduced inflammation, was developed.[34] The indirect mechanism of endothelial dysfunction induced primarily by high inflammatory responses could be interrupted. Third, the activation of endothelial cells leads to a procoagulant phenotype, which in turn continuously activates endothelial injuries during the acute infection phase.[35] In addition to the normal ranged D-dimer, deep venous thrombosis of lower extremities was excluded by ultrasonography in our cohort, consistently with the previous study,[17] indicating the termination of this vicious cycle that sustained activating endothelial injuries. Therefore, our study could help to explain why COVID-19 survivors no longer need to endure the risk of long-term thrombosis at the level of endothelial phenotype.

Endothelial Biomarkers in Patients Recovered from COVID-19 One Year after Hospital Discharge: A Cross-Sectional Study