Analysis of blood index characteristics in COVID-19 patients and their associations with different outcomes

Introduction

In December 2019, a coronavirus disease-2019 (COVID-19) epidemic broke out in Wuhan, which quickly spread worldwide [1]. As of March 11, 2020, a total of 118,000 confirmed patients worldwide have been reported, in more than 110 countries, causing more than 4,300 deaths. The WHO has announced that the COVID-19 epidemic constitutes a global pandemic. Severe acute respiratory syndrome-coronavirus-2 (SARS-CoV-2) is a single-stranded positive-strand RNA virus and is the seventh member of the coronavirus family. It has 79.5% homology with SARS-CoV and has 96% homology with the bat coronavirus in the bat Yunnan chrysanthemum. [2].

SARS-CoV-2 can be detected in nasal swabs, throat swabs, sputum specimens, alveolar lavage fluid, serum/plasma and fecal of infected persons by nucleic acid amplification [3], [4]. However, because most patients suffer from dry cough, it is difficult to obtain sputum. Alveolar lavage fluid sampling is likely to cause aerosol transmission, and early concentrations of viral nucleic acids in the blood are too low. Therefore, throat swabs and fecal specimens are more commonly sampled for diagnosing and monitoring patients. According to the latest treatment guidelines of the Ministry of Health for COVID-19, pharyngeal swab nucleic acid tests must be negative for two consecutive times (24 h interval) to confirm patient recovery. However, some patients had nucleic acid-positive feces that lasted for a long time. During this period, the patient’s condition basically recovered. Fecal samples should be tested to exclude a potential alternative route of transmission that is unknown at this stage [5]. It is controversial whether patients with negative throat swabs but positive feces could be discharged. It is very meaningful to analyze whether changes in blood indices during the fecal-positive period are associated with patient outcome.

SARS-CoV-2 nucleic acid testing is the golden standard for the diagnosis of COVID-19 patients, but the virus load is low in the early stage of the disease, the nasopharynx is not the focal point of early infection and the operation is complicated, so throat swab tests have high clinical false negative rates. Antiviral antibody detection can also assist diagnosis; Xi Mo et al. [6] believed that it takes a period of time from SARS-CoV-2 infection to antibody production. Serum antibody detection of immunoglobulin G/M (IgG/M) lags behind nucleic acid detection. Positive antigen tests can also provide strong pathogenic evidence, but current methods for antigen testing are less mature. In the early stage of COVID-19, fever and dry cough are the primary manifestations. Winter is the season of high influenza virus incidence, and influenza virus A/B (Flu A/B)-infected patients also develop fever and dry cough. Patients with Mycoplasma pneumoniae (MP) infection also have flu-like symptoms in the early stage. Early rapid diagnosis of COVID-19 patients will make up for missed detections due to false negatives in nucleic acid tests and help clinicians to gain time for patient treatment. Similarly, with the development of the patient’s condition, if some blood indices are significantly associated with patient outcome, they will be helpful to evaluate the patient’s condition.

We conducted a statistical analysis of the blood cell count, biochemical test results, and C-reactive protein (CRP) levels of newly diagnosed COVID-19 patients, MP-positive patients, Flu A/B-positive patients, and healthy people, and found that some of the blood indices show significant changes between groups. Moreover, we analyzed the differences in blood cell count and biochemical indices between each patient’s initial diagnosis, discharge, and follow-up visit. We found that blood indices, except aspartate aminotransferase (AST) levels, did not return to normal values when the patient was discharged, and most of the indices had returned to normal levels at follow-up after 14 days. The observed trends were consistent; we found that some blood indices of fecal-positive patients with positive or negative throat swabs were different. For example, absolute lymphocyte count (LY#) decreased significantly during the fecal-positive period.

Materials and methods

Results

General condition and outcome of the patient

The median age of the 17 COVID-19 patients was 37 years (interquartile range, 29–50 years). Ten males and seven females were included. All patients were affected by SARS-CoV-2 because they had been to Wuhan or had been in contact with relatives and/or friends who had been to Wuhan. The median incubation period before symptom onset was six days (interquartile range, 3.5–9 days) according to the patients’ medical histories.

First symptom: in 11/17 patients, the first symptom was fever without a source; four of them subsequently developed dry cough, two had muscle aches and weakness, one had palpitations, and one had both dry cough and loose stool; in 3/17 patients, the first symptom was pharynx pain, and a sore throat/fever appeared 2–5 days later; in 2/17 patients, the first symptoms were cough and fever; in 1/17 patient, the first symptom was cough without fever. 14/17 had pulmonary inflammation. 10/17 had ground glass opacity, 2/17 had unilateral shadows, 2/17 had bilateral shadows. The (computerized tomography) CT results show that 14/17 had pulmonary inflammation. 10/17 had ground glass opacity, 2/17 had unilateral shadows, 2/17 had bilateral shadows.

Upon hospital admission of all patients, throat swab SARS-CoV-2 nucleic acid test results were positive; the median time for the throat swab to turn negative was eight days (interquartile range, 2.75–13.5 days); the fecal nucleic acid test was not carried out at the beginning. The duration of fecal positivity is unknown. In 10/17 patients, the stool and throat swabs remained SARS-CoV-2 nucleic acid-negative during hospitalization, and fecal positivity appeared in 7/17 patients during hospitalization. The patients’ temperature continued to be normal when fecal specimens were nucleic acid-positive, and lung CT showed that in the stage of inflammation subsiding, the patients’ physical signs were stable. Among these seven cases, throat swabs of six cases turned nucleic acid-negative before the feces did, and in the remaining case feces turned nucleic acid-negative before throat swabs did. Twelve people were discharged from the hospital for two weeks and returned for follow-up. One person was found fecal-positive (fecal nucleic acid was not detected before hospital discharge). The patient remained isolated for two weeks at home, and during follow-up feces was nucleic acid-negative. The throat swabs and feces of the remaining 11 patients were negative for nucleic acid. All 17 COVID-19 patients were discharged (in three patients, the feces was still positive upon hospital discharge). The median length of hospital stay was 11 days (interquartile range, 6–21 days), and anti-SARS-CoV-2 antibody IgM and IgG levels were measured during follow-up in 12 patients two weeks after discharge (Figure 1).

Figure 1:

Summary of COVID-19 patient incubation periods and times of discharge and admission.

Upon hospital admission, the throat swab nucleic acid test was positive in all 17 patients; the red vertical line indicates when the fecal nucleic acid test came in use.

Comparison of blood indices of 17 COVID-19 patients upon admission, discharge, and follow-up (13 patients were fecal-negative and 11 patients were followed up at two weeks)

Based on the results in Table 1, the indices that are significantly different between the HS group and other groups were selected, and these were compared with the values upon admission. EO% and EO# were higher at discharge (p<0.05), but did not reach normal values (p<0.05); two weeks after discharge, they were basically recovered to normal levels. MN% was still high upon discharge (p>0.05), but recovered to normal levels within two weeks; LY# was still low upon discharge (p>0.05), but returned to normal levels within two weeks. ALT levels were still high upon discharge (p>0.05), and did not decrease significantly within two weeks (p>0.05); AST levels were significantly lower upon discharge (p<0.05) and did not change significantly within two weeks (p>0.05). There was no significant difference with normal values (p>0.05). LDH levels were still high upon discharge (p>0.05), and remained high two weeks after discharge (Figure 2).

Figure 2:

Comparison of the blood indices EO%, EO#, MN%, LY#, ALT, AST, and LDH at the time of admission, discharge, and follow-up and in healthy subjects.

A: HS group; B: COVID-19 patients upon admission; C: COVID-19 patients upon discharge; D: COVID-19 patients 14 days after discharge. *p<0.05.

Discussion

The incubation period and symptoms of 17 COVID-19 patients were in agreement with those reported in related literature [5]. A total of 39 COVID-19 patients in our hospital and surrounding hospitals have been found to be negative for the Flu A/B. Considering the current high incidence of influenza, it is speculated that the probability of overlap of infection with influenza virus and SARS-CoV-2 is low. The symptoms of patients with early SARS-CoV-2 infection are not obvious, and lung CT does not give strong indications of inflammation. The nucleic acid positivity rate of throat swab sampling is low. COVID-19 patients are easily confused with other patients with common pneumonia. Ai et al. [10] considered that it is important to strengthen the differential diagnosis of pneumonia caused by COVID-19 and other pathogen infections. They performed A and B flow, respiratory syncytial viral (RSV), and differential diagnosis of mycoplasma and nasal virus in 53 patients with suspected COVID-19 by RT-PCR and met genomics next generation sequencing (mNGS). It was considered that the detection of other respiratory pathogens during the winter period, when people are prone to pneumonia, could not be used as an exclusion criterion for COVID-19. Some blood routine indices can help identify Flu A/B-infection and MP-infection, which is helpful for the initial screening of COVID-19 patients during the outbreak. COVID-19 patients can be initially identified based on WBC, NEU%, NEU#, LY%, and Flu A/B positivity; MP-positive patients can be initially identified based on WBC, NEU#, EO%, and EO# (Table 1). Virus infection will not cause great changes in CRP. However, CRP levels were highest in COVID-19 patients than other groups, one possible explanation for this phenomenon is the overproduction of inflammatory cytokines. Cytokines fight against the pathogen but when the system hyperactivates, it can damage lung tissue. CRP production is induced by cytokines and tissue destruction, but this is only the result of the first-diagnosed patient, and there is no big change from the normal reference interval yet [11]. No significant differences in biochemical indices were observed with HSs, and they cannot be used for diagnosis.

AST levels were positively associated with patient outcome, and high ALT levels were associated with poor prognosis (Figure 2). Zhang et al. [12] believe that (angiotensin converting enzyme) ACE2 as a key receptor is not only expressed on tracheal epithelial cells and lung AT2 cells, but also in absorbable intestinal epithelial cells, such as colon and ileum. This finding suggests that the virus may invade the liver through the digestive tract or blood circulation. As ACE2 is highly expressed on bile duct cells, which suggests that SARS-CoV-2 may directly bind to ACE2-positive bile duct cells, resulting in liver dysfunction. Anna et al. [13] find that mildly abnormal plasma Liver Function Tests, especially AST and ALT, are frequently observed in patients with COVID-19 on admission and are associated with severe disease and increased inflammatory markers. LDH levels increased rapidly after symptom onset, and LDH levels were still high after recovery. Since the total lactate dehydrogenase was detected in our hospital, LDH contains five isoenzymes, and the content of LDH3 is the highest in the lungs (lung injury can cause LDH3 levels to increase). The level of LDH in patients with no lung inflammation was still higher than that in normal people, as the AST and ALT levels of 17 patients were increased, suggesting that virus may invade the liver through the digestive tract or blood circulation. It also caused the elevation of LDH4 and LDH5. As the lungs were still recovering from inflammation, the increase in LDH levels is speculated to be related to lung or liver damage, so patients need to return for follow-up evaluation after two weeks. LY# was still low upon discharge, and it returned to normal values after 14 days, which may be indirectly caused by the viral infection [14]. EO% and EO# were significantly lower at the time of disease onset, which may be due to suppression under stress conditions, or it may be the result of endogenous glucocorticoid secretion and the production of chemokines at the site of infection [15]. Degranulation of basophils and monocytes is involved in eosinopenia sequestration at the site of infection [16]. EO% and EO# were still lower than normal levels when the patient was discharged, and they returned to normal values within 14 days. Eosinophils were significantly associated with patient outcome, and can hence be used to assess the patient’s recovery level. At the same time, EO# changes with changing Ct values and they have better consistency, indirectly reflecting that EO# can be used to roughly assess dynamic changes in SARS-CoV-2 infection (Table 2). Caroline et al. studied the relationship between EO and Ct values, and they found several lines of investigation have indicated a role for eosinophilic leukocytes in limiting virus infectivity and promoting virion clearance [17]. Liu et al. compared the relationship between EO and Ct values, and they believed that increasing eosinophils may be an indicator of COVID-19 improvement [18].

In addition to the respiratory tract, viruses often invade the human digestive tract, and certain parenteral viruses can bypass physical barriers to infect susceptible cells in the digestive tract [19]. Like SARS-CoV, SARS-CoV-2 has been confirmed to infect ciliary bronchial epithelial cells and type II lung cells with angiotensin converting enzyme 2 (ACE2) as a receptor, thereby causing lung disease [20], [21], [22]. Holshue et al. [23] detected SARS-CoV-2 nucleic acid in the stool of confirmed patients. Nanshan Zhong’s team isolated a live fecal virus [24]; they believe that fecal viral nucleic acid positivity and the presence of virus are two different things, and further research is needed to determine whether fecal viral nucleic acid is contagious. During treatment, seven patients with throat swab nucleic acid continued to be positive after feces turned negative for viral nucleic acid (Figure 1). SARS-CoV-2 may infect the digestive tract by some means after colonization of the respiratory tract, as six patients had pharyngeal infection. The throat swab first turned negative for nucleic acid, and fecal nucleic acid positivity lasted for a longer time. During this period, the throat swab remained negative for nucleic acid, and the digestive tract SARS-CoV-2 live virus did not infect the respiratory tract again.

The effects of SARS-CoV-2 on the human respiratory and digestive tracts are inconsistent. From the analysis of the symptoms, it seems patients gradually recover during the fecal-positive period. The body temperature is normal, and the lung CT shows inflammation subsiding, which seems to be similar to normal people. Moreover, feces continue to be positive for nucleic acid. Comparing blood indices in two periods, multiple differences were observed. LY# did not change during disease progression, and other indices also showed that they were still at the level of inflammation. Fecal nucleic acid testing was not carried out at the beginning of the epidemic. The first three cases of COVID-19 patients were not tested for fecal nucleic acid when they were discharged. One patient was found to be positive for fecal nucleic acid at the two-week follow-up. Due to the patient’s self-isolation after discharge, the fecal–oral transmission could not be further evaluated. The patient’s blood indices AST, ALT, and NE# were higher than before discharge, LY# decreased, and EO# was still at the level before discharge, which is consistent with the results in Table 3. The patient tested negative for fecal nucleic acid after seven days. It is still unknown whether fecal–oral transmission is an alternative infection route. Three of the seven fecal-positive patients remained positive until discharge from the hospital, and the fetal nucleic acid Ct value was low. Alvaro et al. believed that RT-PCR assays of fecal specimens, is an important step to control infection, suggesting that samples remained positive for SARS-CoV-2 RNA longer time then respiratory tract samples [25]. Chen et al. introduced one case, this case indicates that the virus can proliferate in the digestive tract and potentially undergo fecal–oral transmission [26]. The patients were strongly positive for antiviral IgG and IgM. Persistence of fecal positivity may be due to higher secretion levels of ACE2 in the digestive tract. IgM is produced during common respiratory viral infections, and 12 SARS-CoV-2 patients who were followed up after two weeks SARS-CoV-2 were still positive for IgM, indicating that the duration of IgM presence was longer than the actual infection.

The SARS-CoV-2 nucleic acid test carried out in our laboratory is a semi-quantitative test. Since SARS-CoV-2 nucleic acid was not detected in the early stages of the epidemic, it was not possible to evaluate the time and duration of the presence of nucleic acid in the stool, nor was it possible to evaluate the association between throat swab positivity and fecal positivity. Nevertheless, despite the low number of COVID-19 patients included in the present study and the small amount of available data on blood indices and the presence of viral nucleic acid, the preliminary characteristics of some blood indices are already obvious.

In summary, some Blood cell count indices upon hospital admission could be helpful to give some tips of diagnosis of SARS-CoV-2-infection、Flu A/B-infection and MP-infection; EO# is associated with patient outcome and can be used as a reference for assessing the level of recovery. Meanwhile, EO# is more strongly associated with the Ct value; in patients with COVID-19, fecal negativity is observed later than throat swab negativity, and when the throat swab turns negative, the patient’s condition especially digestion is stable and The digestive tract is not sensitive to SARS-CoV-2, and SARS-CoV-2 in the intestinal tract has not been observed to be retransmitted to the respiratory tract. Due to inconsistencies in some blood indices and the fact it is unclear whether fecal–oral transmission takes place, there is a certain risk that patients are discharged before being fecal-negative.