Introduction
Coronavirus disease 2019 (COVID-19), which has become a major threat to the global public health system [
1], is caused by severe acute respiratory syndrome coronavirus 2 (SARS-Cov-2) [
2]. As of February 18, 2021, more than 110 million COVID-19 infections and 2.5 million deaths have been reported in 192 countries and regions. China reported 100 thousands cases and 4833 deaths which were mainly from the original epicenter, Wuhan [
3]. Scientists are just beginning to understand the nature of the harm caused by this disease. The SARS-Cov-2 spike protein mediates the entrance of the virus into host cells via surface angiotensin-converting enzyme 2 (ACE2) [
4,
5]. Host protease transmembrane, serine protease 2 (TMPRSS2) promotes SARS-CoV-2 entry into target cells, which are thought to be host determinants for viral infection in the initial stage. COVID-19 patients may be asymptomatic or symptomatic. The time from exposure to onset of symptoms is about 5.1 days [
6]. Pathologically, almost every vital organ in the body, including the lungs, heart, liver, kidneys, eyes, blood vessels, intestines, and brain, can be injured by SARS-CoV-2, leading to devastating consequences [
7]. Acute and diffuse lung injuries cause increases in a series of serum cancer biomarkers in patients [
8]. Damage to the lungs can be long-term for some recovered patients; this has also been observed in surviving severe acute respiratory syndrome (SARS) patients [
9].
Patients with metabolic-associated preconditions are susceptible to SARS-CoV-2 attack and are likely to experience more pronounced symptoms. One pathogenic cofactor associated with hypertension, obesity, diabetes mellitus, and cardiovascular disorders is hypercholesterolemia. We and others have recently reported hypolipidemia in hospitalized COVID-19 patients [
10‐
13]. An association between decreases in lipid levels and the severity of the symptoms in patients has been revealed [
10‐
12]. Furthermore, emerging evidence has shown that SARS-CoV-2 has a direct impact on the downregulation of lipid-metabolism-related proteins and pathways, leading to dyslipidemia [
14]. In addition, dyslipidemia associated with SARS-CoV-1 has also been reported [
15]. Altered lipid metabolism has been shown in recovered SARS-CoV-1 patients 12 years after infection [
16]. These reports demonstrate that dyslipidemia is an important clinical manifestation in patients with coronavirus-related diseases, perhaps reflecting one aspect of the complicated evolving pathological progressions seen in patients. In this study, we performed a follow-up investigation of lipid profiles and other laboratory values in COVID-19 patients from our previously reported cohort [
17]. We found that there were significant improvements of coagulation and liver laboratory values, including D-dimer, antithrombin III (ATIII), fibrin degradation product (FDP), fibrinogen (FIB), alanine aminotransferase (ALT), alkaline phosphatase (ALP), and gamma-glutamyl transferase (GGT) in patients at follow-up compared to time of admission. However, partial recovery of low-density lipoprotein cholesterol (LDL-c) levels and incomplete absorption of lung lesions were observed at follow-up, indicating that a long-term recovery process and the development of sequelae such as pulmonary fibrosis could be expected in some patients.
Discussion
In this study, we performed a follow-up investigation of lipid profiles and other laboratory values on 107 recovered COVID-19 patients at 3–6 months after discharge. Our data demonstrate that levels of LDL-c and HDL-c increased significantly in severe/critical COVID-19 cases with or without adjustment of the application of traditional Chinese medicine. Coagulation and liver laboratory values, including D-dimer, ATIII, FDP, FIB, CRP, ALT, ALP, and GGT, decreased significantly across all subgroups. Furthermore, incomplete absorption of lung lesions was observed in CT images in most follow-up patients. These findings provide insight into the pathological evolution of COVID-19 during recovery and potential long-term sequelae of the disease.
Recently, we and other investigators have reported hypolipidemia in hospitalized COVID-19 patients [
10‐
13]. The decrease in lipid levels in patients with COVID-19 is associated with the severity of the symptoms [
10‐
12]. These findings demonstrate that abnormalities in lipid metabolism are clinical manifestations of COVID-19 that have been underappreciated. Mild or moderate liver injuries caused by viral infection may be one important factor contributing to dyslipidemia in COVID-19 patients. Serum levels of ALT, ALP, and GGT were moderately elevated in about half of the cohort of patients in our study at the time of admission, indicating mild or moderate liver injury [
11]. In this study, patient ALT, ALP, and GGT levels were significantly lower at follow-up than at the time of admission, indicating improvements in liver enzyme levels in patients during recovery. There are a couple of potential mechanisms involved in the role of cholesterol in the pathological progression of COVID-19. Wang et al. suggests that cholesterol concomitantly traffics ACE2 to viral entry sites, where SARS-CoV-2 docks in order to properly exploit entry into cells [
19]. Therefore, decreased cholesterol levels in the blood may indicate severe loading of cholesterol in peripheral tissue and escalated SARS-CoV-2 infectivity [
19]. Cao et al. suggests that cholesterol may facilitate an acceleration of endothelial injuries caused by SARS-CoV-2 [
20]. Sorokin suggests that lowering HDL-c in COVID-19 patients may decrease the anti-inflammatory and antioxidative functions of HDL-c and contribute to pulmonary inflammation [
13]. The decrease in WBC and increase in lymphocytes at follow-up may be reflected cytokine secretion and inflammatory status and related to lipid parameters. In addition, serum amyloid A protein (SAA) level can rise up to 1000-fold in response to an acute inflammation, which will displace apolipoprotein A-I from HDL particles [
21]. HDL containing SAA is targeted to the macrophage [
22]. SAA level increases in COVID-19 patients, especially severe cases [
23,
24]. It is likely that SAA/apolipoprotein A-I axis gets involved in HDL-c metabolism in COVID-19 patients, but the detailed mechanism needs to be elucidated in future studies. All these hypotheses will lead to more and novel insights into the nature of this disease.
The dynamics of lipid levels in a small cohort of our longitudinal study and in two cases in other reports have shown that cholesterol levels were low at the time the patients were hospitalized, remained low during disease progression, and returned to baseline levels in patients who were discharged [
10,
12,
13]. To our surprise, after exclusion of the factor of taking cholesterol-lowering drugs, a small portion of patients (6%) showed a decrease in LDL-c or HDL-c levels of 15% or more at follow-up as compared to the time of admission. The low lipid levels in these patients were probably due to medications or nutritional supplements taken during their own recovery process at home, for example, profound and acute dietary changes. It would be interesting to find out whether those patients with lower LDL-c or HDL-c levels at follow-up were associated with malnutritional or socioeconomically underrepresented populations. Loss of appetite is one of early symptoms for some COVID-19 patients. They may continue having a poor appetite during the disease progression and recovery courses which may result in malnutrition thus low lipid levels at follow-up. Low lipid levels may also be caused by the matter of patient compliance after discharge. Although it is less likely, there could be a long-term sequela of lipid abnormality caused by or associated with viral infection in COVID-19 patients; there is no evidence to support the notion that SARS-CoV-2 causes long-term chronic infection.
Emerging evidence has supported coagulation as an independent mortality factor in COVID-19 patients. Coagulopathies have been found in the early stages of the disease [
25‐
27] and in non-surviving patients [
28]. Patients have shown elevated coagulation and cardiac biomarkers such as D-dimer, fibrinogen, high-sensitivity troponin I and creatinine kinase–myocardial band [
29,
30]. In our follow-up study, coagulation laboratory values, including D-dimer, ATIII, FDP, and FIB, were significantly lower in patients at follow-up as compared to the time of admission across all subgroups, indicating improvements from coagulopathies. However, we did not find significant correlations between the restoration of LDL-c or HDL-c levels and decreases in levels of these coagulation values; this suggests that recovery from dyslipidemia and improvements from coagulopathies are probably involving different pathways at different paces.
Incomplete resolution of lung lesions was observed in 72% patients in the follow-up CT examinations, suggesting pulmonary fibrosis as a potential long-term sequela for many COVID-19 patients. SARS patients have shown persistent impairment of lung function, even years after discharge [
31,
32]. Pulmonary fibrosis, GGO, and pleural thickening have been reported in follow-up chest radiographs in a substantial portion of patients with Middle East respiratory syndrome coronavirus (MERS-CoV) [
33]. Consistent with our findings, You et al. showed that 83.3% of COVID-19 patients had residual CT abnormalities, including GGO and pulmonary fibrosis [
34]. These data suggest that aberrant wound healing in COVID-19 survivors, which is evidenced by GGO and residue lesion patterns, may lead to pulmonary fibrosis; larger studies are needed to verify this notion.
There were several limitations of this study. First, less than one-fifth of the patients from our original cohort participated in this study, which might cause a biased representative sample group from the original cohort. Second, the sample size for follow-up critical cases was limited; this might lead to an overall insignificant increase in levels of LDL-c in this subgroup. Third, many patients might have been taking various medications or remedies at home for recovery, including Chinese traditional medicines or nutritional supplements. In this study, we found that about 58% of patients in mild group and 37% of patients in severe group had taken TCHMs during their illness or / and recovery courses. Our data indicated that TCHMs might have a negative impact on the improvement of lipid profiles in patients with severe symptoms. However, due to the complexity of ingredients in those TCHMs, it will be very difficult to determine which factor(s) and how they interfere with lipid metabolisms in some patients’ recoveries in the severe group; this will need a thorough investigation in future. We, however, did not find so far that TCHMs caused any significant changes in the overall lipid profiles at the time of admission and follow-up crossing all the subgroups. Therefore, TCHMs might have a minor effect on lipid values in our patients which resulted in a negligible impact on the conclusions we drew in this study. We are aware that these data and analyses only apply to this specific Chinese population. Fourth, the lipid profiles of patients prior to discharge were crucial to determine the contributive factors to the decreased LDL-c or HDL-c levels in a small portion of patients at follow-up as compared to admission, which were lacking. Fifth, a continuous long-term follow-up is needed in order to monitor the dynamics of lipid profiles and CT abnormalities during the recovery process for a large cohort of COVID-19 patients in order to better predict potential sequelae, such as lung fibrosis; this will be our future research goal. Lastly, the characteristics of lipoproteins, such as apolipoprotein A-I, in our cohort was unknown. We also did not know the cellular cholesterol levels in COVID-19 in this study; such information could provide us insights into the molecular mechanisms underlying dyslipidemia in COVID-19. Whether and how cholesterol or lipoproteins participate in regulation of SARS-CoV-2 entry of host cells and viral production are yet to be determined, which will our primary goal in future investigations.
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.