Eosinophil and Anticoagulation in COVID-19 Patients

Introduction: Despite prophylactic anticoagulant treatments, thrombotic complications may develop in patients with Coronavirus disease 2019 (COVID-19). This study aimed to evaluate anti-Factor Xa levels to determine the anticoagulant activity of low molecular weight heparin (LMWH) in COVID-19 patients. Materials and methods: We prospectively evaluated 80 COVID-19 patients, diagnosed using polymerase-chain-reaction test, who were admitted to our clinic and administered LMWH; enoxaparin was applied according to the weight, D-dimer levels, and clinical condition of patients. Anti-Factor Xa levels in blood, drawn 4h after the 3rd dose of LMWH, were measured and a level of <0.2IU/ml was considered subprophylactic. Patients were followed up clinically and anti-Factor Xa levels were re-examined before discharge. Results:Groups 1 and 2 included 13 and 63 patients with subprophylactic (0.18±0.06) and prophylactic (0.43±0.23) anti-Factor Xa levels, respectively. The proportion of eosinophils in patients was signicantly higher in group 1 than in group 2 (2.96±2.55 vs 0.90±1.28;p=0.001). At the time of discharge the eosinophilic proportion of patients was signicantly higher (3.06±1.49 vs 2.07±1.92;p=0.001) but the activated partial thromboplastin time was signicantly lower (22.34±1.38 vs 24.38±3.58; p=0.01) in group 1 than in group 2. Of 14 patients with eosinophil content >4%, 6 were in group 1 ((6/13) 46.2%) while 8 were in group 2 ((8/63) 11.9%); (p=0.009), and all had a D-dimer level <1µg/mL (p=0.03). ROC analysis for the presence of anticoagulation at subprophylactic level revealed an area under curve of 0.79 (95% CI:0.64-0.93); p=0.001). Conclusions: In COVID-19 patients, eosinophil levels could be considered for determining effective prophylactic anticoagulant administration. (NCT04507282)

Several reports have shown that, similar to other viral pneumonia, the incidence rate of venous thromboembolism (VTE) in COVID-19 patients, particularly those in intensive care, is high [3][4][5][6] . The cause of the hypercoagulation in COVID-19 patients has not been fully understood. Several studies have indicated an increase in some hematologic parameters that may lead to endothelial damage, immobilisation related stasis, and hypercoagulability [7][8][9][10].
Eosinophils normally make up only a small fraction of circulating leukocytes (1-3%), but their levels can vary in different disease states [11]. Eosinophil levels are clinically important because they are potent proin ammatory cells containing cytotoxic proteins and various enzymes (peroxidases, cationic proteins, and neurotoxins) that can in uence the effectiveness of heparin.
The International Society on Thrombosis and Hemostasis (ISTH), American Society of Hematology (ASH), and American College of Cardiology (ACC) recommend heparin prophylaxis in patients with COVID-19. However, the dose that can be used has not been clari ed. Previous studies have shown that heparin prophylaxis reduces thromboembolic events in COVID-19 patients [12]. However, the e cacy of heparin prophylaxis in COVID-19 patients, as determined by laboratory data, and the factors affecting this e cacy are not known. This study aimed to analyze the effectiveness of using LMWH in  patients and identify the factors affecting this e cacy using laboratory data.

Study patients
After receiving approval from the Ministry of Health and the local ethics committee, we included 80 patients who were found to be COVID-19 positive by polymerase chain reaction (PCR) test in our clinic between May 15, 2020 and June 15, 2020; their written consents were obtained. The patients were followed up clinically by transferring them to the service reserved for COVID-19 positive patients.
Inclusion criteria: Patients older than 18 years, who were diagnosed with COVID-19 and were administered LMWH, and agreed to participate in the study were included Exclusion criteria: Patients with previous coagulopathy, continuous indication of anticoagulant therapy (atrial brillation (AF), valve disease), glomerular ltration rate (GFR) <30 mL/min or undergoing dialysis, or with known liver dysfunction were excluded from the study. Diagnosis COVID-19 was diagnosed according to the WHO interim guidelines and con rmed in our laboratory by SARS-CoV-2 RNA detection with reverse-transcriptase polymerase-chain-reaction (RT-PCR) using nasal and pharyngeal swab samples [13].
Patients with a systolic blood pressure (SBP) of ≥140 mmHg and/ or a diastolic blood pressure (DBP) of ≥90 mmHg, and those using antihypertensive drugs were considered hypertensive. Patients using oral antidiabetics or insulin, or exhibiting fasting blood glucose ≥126 mg/dl in two measurements were considered diabetic. Body mass indices (BMI) were calculated according to the following formula: BMI = body weight (kg)/ square of the height (m²). GFR was calculated using the Cockcroft-Gault formula: GFR = [(140 -age) x patient weight (kg)]/[72 x serum creatinine value] (x 0.85 for women) [14].

Study procedures
Demographic characteristics of the hospitalized patients with COVID-19 were recorded and computerized tomography (CT) of thorax were evaluated. Blood samples were collected to evaluate the hematological, in ammatory, and biochemical parameters of the patients (Figure 1). Electrocardiograms (ECG) were recorded and O 2 saturation was determined.
Treatment of patients with enoxaparin was arranged based on the results of laboratory tests, thorax CT, and clinical evaluation. Enoxaparin dosage of 0.5 mg/kg (2x1) was administered to patients with increased in ammation parameters and D-dimer levels, as well as pneumonic in ltration in thorax. Enoxaparin dosage of 40 mg (1x1) was administered to the other patients. Other treatments were determined based on the recommendations of infectious disease specialists.
We determined the levels of anti-Factor Xa in the blood collected from COVID-19 patients 4 h after the 3rd LMWH dose. An anti-Factor Xa level of <0.2 IU/mL was de ned as subprophylactic [15,16].
Patients with decreased O 2 saturation and progressing disease state were taken to the intensive care unit.
Control anti-Factor Xa levels in the blood collected 4 h after administering the last LMWH dose before discharge were measured ( Figure 1).

Laboratory evaluation
Hematological parameters were examined with Mindray BC 6800 whole blood device (Mindray, China). The BC-6800 hematology analyzer used sheath ow impedance, laser scatter, and SF Cube analysis technology. The SF Cube analysis technology is three-dimensional using information from laser light scatter at two angles and uorescent signals for cell differentiation and counting. In addition, the accuracy of cell numbers was con rmed by peripheral smear from blood samples taken from patients.
Biochemical parameters were examined with Cobas C702 (Roche Diagnostics, Mannheim, Germany) device. CRP was examined with BN II nephelometer System (Siemens Healthcare Diagnostics Inc., USA). D-dimer: brinogen ratio was examined by the Sysmex CS-5100 device.
The levels of anti-factor Xa were measured from the obtained plasma samples using the Berichrom Heparin kit in a Sysmex cs 5100 device in the biochemistry laboratory. The Berichrom Heparin kit is a chromogenic test (Berichrom heparin, Siemens Healthineers, Marburg, Germany). INR (International Normalised Ratio), PT (Prothrombin Time), and aPTT (activated Partial Thromboplastin Time) were measured as coagulation parameters. Venous blood samples in coagulation tubes were centrifuged at 5000 rpm for 10 min, and the INR, PT, and aPTT levels were measured in the biochemistry laboratory using a Sysmex cs 5100 device, Dade Actin FS, activated PTT reagent, and thromborel S reagent.

Follow-up
The patients whose general condition was stable, had reduced complaints, and had a decrease in in ammatory parameters were discharged. Patients with D-dimer values above 0.5 µg/mL during discharge were administered a single dose of enoxaparin (40 mg, 1x1) for 30 days. Patients with lung involvement during hospitalization were given moxi oksain (400 mg, 1x1) or amoxicillin (1000 mg, 2x1) for 1 week at discharge. After discharge, these patients were examined at home by liation teams for 14 days.

Statistical analysis
The data were analyzed using the SPSS 23.0 statistics package (SPSS Inc, Chicago, IL, USA). Continuous variables have been reported as mean ± standard deviation, and categorical variables have been reported as percentages. In comparing the averages between groups, Student's t test was used for variables with a normal distribution, and the Mann-Whitney U test was used for variables without a normal distribution. Categorical variables were compared with the chi-squared test or Fisher's exact test. The sensitivity and speci city of eosinophil to predict subprophylactic levels of anti-factor Xa were analyzed by receiver operating characteristic (ROC) analysis. P values <0.05 were considered signi cant.

Results
A total of 13 patients with anti-factor Xa levels < 0.2I U/mL (subprophylactic anticoagulation) were de ned as group 1, and 63 patients with anti-factor Xa levels > 0.2 IU/mL (prophylactic anticoagulation) were de ned as group 2. When the demographic and laboratory characteristics of the patients in groups 1 and 2 were evaluated, no signi cant difference was found except for the eosinophil counts and levels of anti-factor Xa ( Table 1).
Laboratory analysis of the blood collected before the discharge of patients revealed that eosinophil counts in group 1 were higher than in group 2, whereas aPTT and anti-Factor Xa levels were lower in group 1 than in group 2 ( Table 2).
The AUC value in the ROC analysis for baseline eosinophil counts to show subprophylactic anti-factor Xa level was 0.79 (Range: 0.64-0.93; p = 0.001) (Figure 2).
Thoracic CTs of the patients were evaluated, identifying 51 patients with infection sings and 29 with no signs of infection in their thorax CT. When patients with and without thoracic CT ndings were compared, age, gender, medication, eosinophil percentage >4%, sediments, crp, brinogen, ferritin, AST, ALT, LDH, albumin, HDL, and calcium values were found to be signi cantly different between the groups (Table 3). Patients with positive CT ndings mainly consisted of older male patients. Acute phase reactants of CT positive patients were found to be higher; however, D-dimer and anti-Factor Xa levels were similar in both groups (Table 3).
During follow-up, 1 patient was died and 2 patients needed intensive care unit follow-up. The average hospitalization period of the patients was 7.55 ± 3.95 days. There was no complication in the patients followed by the liation teams for 14 days at home and the general condition of the patients did not deteriorate.

Discussion
In this study, we found that increased eosinophil count associated with the level of subprophylactic anticoagulation. Eosinophil counts evaluated for adjusting anticoagulation dose were also found to be increased in patients with low D-dimer levels. Patients with lung conditions were found to have increased in ammatory parameters and percentage of eosinophils.
COVID-19 infection has been shown to be associated with increased coagulopathy [12,17,18]. In these patients, the D-dimer and brinogen levels were increased but aPTT level was decreased [19]. Local thrombotic events and thromboembolic complications may develop due to endothelial damage and increased coagulable condition due to COVID-19. Anticoagulant therapy reduces mortality and morbidity in COVID-19 patients [17,18]. Various suggestions have been made about the application of anticoagulant treatment strategy. Various laboratory parameters (D-dimer) and clinical conditions of patients are effective in determining these recommendations [20].
Previous studies examined the anti-factor Xa levels after LMWH administration for VTE prophylaxis and values below 0.2 IU/mL have been shown to be subprophylactic doses [15,16]. However, the e cacy and dose of LMWH administered in COVID-19 patients is not clear. It is apparent that levels below the antifactor Xa values determined in previous studies may increase the risk of VTE in COVID-19 patients, which can cause hypercoagulability. Considering this, subprophylactic anticoagulation in patients was determined in our study by taking the limit value of 0.2 IU/mL. Subprophylactic anticoagulation value was determined in 16.25% patients of the studied patients.
In the patient group with subprophylactic anticoagulation, eosinophil levels were found to be increased. Other demographic and laboratory parameters of patients with prophylactic and subprophylactic levels of anticoagulation were similar.
Eosinophils have pro-in ammatory, pleotropic, and immune regulatory properties. Eosinophils are mainly found in blood, although they are also found in the gastrointestinal tract and lungs. Lung pathology caused by eosinophils has been observed in RSV and SARS-CoV-1 viral infections [21]. Eosinophils may also contribute to the lung pathology in COVID-19 patients. In hypereosinophilic cases, the degranulation of major basic protein from eosinophils and eosinophil peroxidase causes platelet aggregation and thrombus formation [22]. Eosinophils can cause in situ thrombus formation in the lungs and veins. Patients with thoracic CT lesions had high eosinophilic in ammatory parameters. Eosinophils secrete their own chemoattractant molecules (eotaxin and platelet-activating factor) that allow more eosinophils to enter the in ammatory area, increasing in ammation and lung damage.
Enzymes released from eosinophils (peroxidases, cationic proteins, and neurotoxins) may decrease the anticoagulant activity of heparin [23,24]. In our study, it was found that patients with high eosinophil levels had lower anticoagulant activity. Although D-dimer and brinogen levels were similar, patients with low anticoagulant activity only had high eosinophil levels, indicating that subprophylactic anticoagulation levels are related to eosinophils. Eosinophil counts had a good AUC (0.79) in predicting the presence of subprophylactic anticoagulation.
Our patient population was small and many of the patients were followed up for a short period of time (average: 7.5 days). Only 3 patients needed intensive care and one patient was died Therefore, the clinical outcomes of subprophylactic anticoagulation could not be evaluated. Studies involving large-scale, intensive care patients may provide information on the clinical outcomes that eosinophil counts can produce due to their subprophylactic anticoagulation property.
The patients who had thoracic CT lesions, more advanced disease, and were older males exhibited higher in ammatory parameters. This has been shown in previous studies [25].

Limitations
The small number of patients is the main limitation of our study, although the results of this study can be a guide for the optimization of anticoagulant therapy to decrease mortality and morbidity in COVID-19 patients. This study serves as a guide for future large-scale studies with larger patient groups. Our study groups were not included patients with morbid obese and renal failure, therefore we need further studies with these patient groups. Besides most of our patients were followed up inpatient clinic, future studies analyzing patients in intensive care units are required.

Conclusion
Increased eosinophil counts in COVID-19 patients were found to be associated with reduced anticoagulant effect of heparin. Hence, eosinophil levels should be taken into consideration while determining the prophylactic anticoagulation strategy in patients with COVID-19.    Diagram of study design