Clinical value of blood markers to assess the severity of coronavirus disease 2019

The coronavirus disease in China, that began in December 2019 (COVID-19), is an emerging lethal respiratory disease and has threatened global health [1, 2]. Based on current epidemiological investigation, over 2 million people worldwide have been affected by COVID-19 [3], which has shown to excess the outbreak of severe acute respiratory syndrome (SARS) in 2002 and middle east respiratory syndrome (MERS) in 2012 [7], undermining the global economy, and destabilizing societies across the world. Full-genome sequencing has labeled the pathogen as severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), which is a novel coronavirus declared by the World Health Organization (WHO) [3]. SARS-CoV-2 is a type of single-strand positive RNA virus and its genetic characteristics are significantly different from those of SARS-CoV or MERS-CoV [4]. It is reported that the spread from person-to-person in the hospital and community settings has been accumulating all over the world [5]. Since the transmission routes include droplet, contact, and aerosol transmission, quarantine has been an effective method to protect susceptible people and reduce transmission [6].

A total of 90,877 cases and 4744 deaths have been reported in China, moreover, 68,139 cases and 4512 deaths have been reported in the most serious region of Wuhan by May 20, 2020. The 28-day mortality of the critical patients was reported to be approximately 61.5% [8]. It is well known that most patients with COVID-19 have fever along with nonspecific respiratory symptoms, such as cough, expectoration and dyspnea. The prevalence of extrapulmonary symptoms often arise from digestive system, so many patients initially present with diarrhea, nausea and vomiting [9]. Unfortunately, no specific therapeutic interventions exist so far, thus, supportive care remains the cornerstone for managing patients. A novel vaccine of COVID-19 is being conducted for clinical trials, which might make a great progress for the further treatments.

Most cases of COVID-19 are mild, however, patients with severe disease can quickly progress to multiple organ dysfunction syndrome (MODS) and even death. Approximately 5% patients develop critical illness requiring an intensive care unit (ICU) admission [10]. Those admitted to the  ICU were found to have higher concentrations of pro-inflammatory cytokines as well as T-helper-2 (Th2) cytokines suppressing inflammatory responses [11]. Aberrant immune-inflammatory storm and anti-inflammatory syndrome may play an important role in the disease progression [12]. Recently, investigators have found the rising of neutrophils as well as the decreasing of lymphocytes can be used as an indicator of disease progression. Other immune-inflammatory biomarkers, such as higher levels of interleukin-6 (IL-6), C-reaction protein (CRP) and Procalcitonin (PCT) were independent risk factors for assessing the severity of COVID-19 in patients [12, 13]. However, to predict the disease severity through potential risk factors at an early stage is still an unsolved problem, especially for the most serious area of Wuhan in China. Therefore, we performed a retrospective study to evaluate the diagnostic value of clinical features and laboratory markers to discriminate between severe and non-severe COVID-19 disease among adults presenting to hospital.

This study has been able to identify characteristics associated with severe COVID-19 disease among adults presenting to hospital. Most importantly, higher levels of inflammatory markers and lower platelet levels have been shown to be strongly associated with more severe disease. It was also shown that the blood, sputum or catheter secretion culture did not show any prediction for disease severity.

The association between increasing age and more severe COVID-19 found in our study, is consistent with previous reports by Tan et al. [14] and Xu et al. [15]. On average patients who did not survive to hospital discharge were 20 years older than those surviving to discharge [15]. The overall prognosis of older patients was worse than that of young or middle-aged patients might be attributed to their poor state of malnutrition or underling complications [16]. In our study, symptoms of cough, expectoration, myalgia and fatigue were significantly higher in the non-severe group. As severe patients  might have already passed this symptomatic process into a multiple organ dysfunction stage, could explain the lower incidence of overt symptoms in this group compared to the non-severe group. In contrast to other reports, this study failed to show any difference in the rates of underlying diseases or past medical history between the severe and non-severe COVID-19 groups. This conflicts with a systematic review by Zheng et al. [17] that reported higher rates of underlying diseases such as hypertension, diabetes, cardiovascular disease and respiratory disease among critically ill patients.

In terms of laboratory results, several studies have reported increased neutrophil counts and decreased lymphocyte counts as a strong predictor for severe COVID-19 disease. For example, Wan et al. [2] reported the leukocyte counts of most patients with COVID-19 were in a normal range, but the classified neutrophil counts were significantly higher in the severe group. Zhang et al. [1] reported lymphocyte counts were lower among deteriorating patients compared to those in being discharged. In our study, both neutrophil and lymphocyte counts were shown to have excellent predictive ability to identify severe COVID-19 disease, which was in agreement with the typical progress of virus infection. Another important finding in our study was related to platelet levels, a lower level of platelets was identified as an independent risk factor for severe COVID-19, which was in agreement with the study published by Zhe Zhu et al. [8]. As we know, platelets are small (2–4 µm in diameter) anucleated cells derived from megakaryocytes in bone marrow, and are responsible for maintaining the integrity of the vasculature. In COVID-19 subjects, platelets were hyperactivated, with aggregation occurring at suboptimal thrombin concentrations, so the circulation level were significantly decreased [18]. What’s more, the severe cases experienced higher viral infection and mechanical ventilation, which led to endothelial injury, platelet aggregation and megakaryocyte reduction, and as a result, the platelet production decreased and the consumption increased [19]. In turn, a common negative correlation was found between platelet and other parameters increasing along with the severity of disease.

With disease progression, vigorous pro-/anti-inflammatory responses to the virus induced apoptosis in lung epithelial and endothelial cells. Inflammatory mediators play an important role in the pathogenesis of severe disease. CRP as an acute-phase protein that stimulated by the release of apoptosis cells, could indicate the severity of inflammation. Prior studies have revealed that CRP on admission could be a predictive factor for progressive respiratory failure in SARS-CoV as well as MERS-CoV infected patients [20]. A recent study by Aggarwal et al. demonstrated that the CRP level showed no obvious difference between critically ill COVID-19 patients and those with mild disease [21]. However, Wang [22] published their research about CRP, showing that an early stage of COVID-19 CRP levels  was positively correlated with lung lesions and could reflect disease severity. Our study showed a similar result with the latter, confirming its role in predicting the severity COVID-19. As a quality improvement project, PCT, a kind of procalcitonin without hormone activity and the precursor of calcitonin [13], was introduced as an antibiotic stewardship tool, also reflecting the inflammatory response. A meta-analysis concluded that PCT > 0.5 ng/mL  was associated with increased risk of COVID-19 critical illness especially when the leukocyte  was initially normal or reduced [23]. In our study, the AUC of PCT was 0.697, providing a reasonable discrimination between individuals with severe and non-severe COVID-19. In addition, these two inflammatory associated factors (CRP and PCT)  were significantly correlated with other cytokines, indicating the fact that excess inflammatory responses deteriorated the normal physiological function in viral infection, making the disease worse in progress.

Cytokines are small protein molecules that play an immunomodulating function to maintain the body homeostasis. The auto-amplifying of pro-inflammatory cytokines, such as IL-6, IL-8, and TNF-a, can result in a dramatic deterioration for a host, often referred to a “cytokine storm”. This contrasts with an excessive secretion of anti-inflammatory cytokines (IL-10), which might lead to the persistent immunoparalysis and high risk of secondary infection. Several studies have attempted to explore the relationship between inflammatory cytokines and COVID-19. Zhu et al. [12] found that high levels of IL-6 was an independent risk factor for severe COVID-19, and  it had a potential value in the monitoring of the disease course in severe cases. Zhang et al. [1] investigated a number of factors related to disease progression among hospitalized patients with COVID-19, also  discovered that IL-6 and IL-10 were elevated among those with worse outcomes. Consistent with previous studies, our study also showed that IL-6 and IL-8 were higher among severe versus non-severe patients. Overall, our study showed that high levels of immune-inflammatory markers (IL-6, IL-8, IL-10 and TNF-a) were all strong predictors of severe COVID-19  by ROC analysis. IL-6 is a multifunctional cytokine that transmits cell signaling and regulates immune cells. It not only plays a key role in cytokine storm, but induces a variety of acute-phase protein, such as CRP, complement components and so on. A recent study confirmed the role of IL-6 in severe COVID-19 cases by immune analysis, suggesting that Th1 cells and monocytes could express high level of IL-6, resulting in deteriorating the tracheobronchial epithelial cells and aggravating the disease [12]. IL-8 has a potential function in recruitment and activation of neutrophils, and combines with the special surface antibody releasing large numbers of reactive oxygen radicals to aggravate the lung injury. Therefore IL-8 commonly elevates in patients with critical disease, and presents a state of inflammation and organ damage [24]. Consequently, there is no doubt that IL-6 and IL-8 have a positive relationship with other immune-inflammatory markers. TNF-a, another important inflammatory cytokine, acts as an early and sensitive mediator in cytokine storm. Once the viral infection induces the inflammatory defense response, TNF-a could be stimulated by activated neutrophils and lymphocyte and exerted a synergistic effect in the secretion of other cytokines [25]. In contrast to the mentioned pro-inflammatory cytokines, IL-10, as a classical anti-inflammatory cytokine, takes part in maintaining the immune system balance. Since critical patients present a higher level of inflammatory response, regulatory T cells (Tregs) are induced to inhabit the over activation immune response through producing anti-inflammatory cytokines, such as IL-10, transcription growth factor-β (TGF-β) [26]. As a result, IL-10 also shows a positive correlation with other immune-inflammatory parameters.

In order to clarify the inner correlations between different serum markers, four indicators (IL-6, IL-8, CRP and Platelet) were selected for further correlation analysis. These biomarkers were chosen because they presented a significant prediction for COVID-19 severity in both univariate analysis and ROC analysis. The current result suggested that pro-inflammatory biomarkers, such as IL-6, IL-8 and CRP, showed a major positive correlation with other indicators, except for lymphocyte, hemoglobin and platelet. While platelet presented a central negative correlation, along with the progress of inflammatory response.

Lastly, few researchers have drawn on any structured research into the association between various cultures and severity of COVID-19. Our study included an analysis of a large number of varying blood, sputum and catheter cultures, which were not associated with more severe COVID-19 disease. The reason might be the lower incidence of positive secretion cultures in our hospital.

Currently, researchers are reporting some new indicators for severe COVID-19, which might be the next step for our research group. For example, Yang et al.  [27] showed that serum apolipoprotein A 1 served as an indicator to reflect the severity of COVID-19. Gómez-Pastora et al. [28] examined the level of ferritin in COVID-19 patients, and confirmed hyperferritinemia in the critically ill, but failed to reveal whether it was the product of inflammation or a pathogenic mediator. Critical patients often experience the respiratory failure in the late stage, so mechanical ventilation is a necessary method to relieve symptoms. At this point, Poggiali et al. [29] presented that Lactate dehydrogenase (LDH) was a predictor of respiratory failure in severe cases. Lastly, adaptive immune response is thought to protect from acquiring SARS-CoV-2 infection, of which neutralizing antibodies (NtAb) seemingly play a major role, such as NtAb50. However, Gozalbo-Rovira et al. [30] found very weak correlations between NtAb50 and COVID-19 severity.

The results of our study need to be considered in the context of potential limitations. Firstly, this is a single-center and retrospective study. Large prospective studies, over an extended period of time, is needed to verify our results. Secondly, the inflammatory response associated with cytokines IL-10 and TNF-a, although showing promise in ROC AUC analysis to predict severe disease, did not reach statistical significance in logistic regression analysis,  this may represent Type II error. Lastly, another potential limitation is that we failed to show any relationship between various cultures and COVID-19 severity, which may be due to our low positive detection rate, and would need to be confirmed by larger studies in the future.

A retrospective study was conducted to evaluate the diagnostic value of clinical features and laboratory markers to discriminate between severe and non-severe COVID-19 disease. Higher serum levels of inflammatory markers appear to be strong predictors of severe COVID-19, various serum biomarkers coordinate with each other to promote the disease progress. The present study suggested that high leucocyte and low platelets counts were independent predictive markers of severity of COVID-19.