Endothelail and Stress Index as One of New Prognostic Determinants of COVID- 19 Severity‎‎

INTRODUCTION

COVID-19 or severe acute respiratory syndrome (SARS-COV.2) is a worldwide pandemic [1]. It was found that COVID-19 binds to angiotensin converting enzyme 2 receptor (ACE2) causing endotheliopathy which activates the coagulation system resulting in thrombosis as well as complement activation [2]. This receptor is widely distributed all over the body namely smooth muscle, lung alveolar cells, enterocytes, venous and arterial endothelium and brain stem [3]. One of the characteristics of COVID-19 is a substantial depletion of host immune cells as well as immune over activation causing massive immune injury with lymphocyte damage. Different presentations have been found for this disease namely fever, headache, generalized body ache. fatigue, anosmia & dyspnoea [4]. The diagnosis of COVID-19, in terms of positive or negative test, is primarily based on laboratory tests, chest imaging modalities, including chest X-ray (CXR) & computed tomography (CT) [5].

New discoveries of the disease biology have been introduced which made the outcome of its management differ from patient to patient.  This is due to heterogeneity of the clinical course which represents a challenge to the clinicians as about 80% of COVID-19 patients are not hospitalized due to mild disease [12]. As the pandemic enters its third year, various treatment protocols have been created in the hopes of eradicating the disease. Early detection of disease severity by finding appropriate prognostic determinant could be lifesaving; hence we aimed at investigating the prognostic value of EASIX in addition to some haematological, biochemical, CT and ECG findings as determinants of the COVID-19 severity.

PATIENTS AND METHODS

Study design and patients: One hundred & five COVID-19 patients aged between 17 to 89 years with positive nasopharyngeal qualitative PCR swabs were included in this study. All patients were not vaccinated. Patients with chronic lung or cardiovascular disease, autoimmune disorders or cancer were excluded. This study was carried out in a COVID-19 referral hospital in Alexandria, Egypt, from January to August 2022. Patients' demographic data including age and gender was taken at the time of diagnosis.

Blood Sample collection and processing: 7 ml of blood samples were collected from each patient; 2 mL were placed in VACUETTE® EDTA tube for complete blood picture assessment with automated differential count which was performed on an ADVIA® 2120i Hematology Analyzer (Siemens), and microscopic examination was done for differential count confirmation. The second 2mL were placed in VACUETTE® citrated plasma tube for D-Dimer assessment through latex test using biolab diagnostic kits with reference value (R.V) of less than 250 ng/mL using Cobas -C Roche®. The remaining 3 mL were placed in VACUETTE® Z serum sep clot activator tube and centrifuged at 2000 g for 10 min at room temperature to assess biochemical parameters using HITACHI AUTOMATIC ANALYZER COBAS 6000, lot (52520500), the parameters included C-reactive protein (CRP), ferritin, liver function tests (alanine transaminase (ALT) and aspartate transaminase (AST), renal function tests (urea and creatinine), lactate dehydrogenase (LDH), and cardiac markers (troponin-I). All tubes were from Greiner Bio-one International GmbH. 

R.V. of CRP by quantitative nephelometric is less than 3 mg/L, ferritin R.V. is 30–400 ng/ml for males and 15–150 ng/mL for females. R.V. for AST is 40 U/L and for ALT is up to 41 U/L. Urea R.V is 10 – 50 mg/dL and for creatinine is 0.7 - 1.2 mg/dL for male and 0.5 - 0.9 mg/dL for female. For LDH R.V is 230 - 460 U/L, and for troponin-I is up to 0.3 ng/ml.

Endothelial activation and stress index (EASIX) were calculated according to the formula: LDH [u/l] x creatinine [mg/dL]/thrombocytes [10^{^9} cells/L] [13].

Radiological findings including cardiac assessment by using Echocardiogram (ECG) to assess heart rate, rhythm and predict signs of ischemia. Transthoracic echocardiography (Echo) for assessment of chamber sizes, valvular conditions, systolic and diastolic functions and wall motion abnormalities was used.

High-resolution computerized topography (HRCT) scans of the chest were performed for all patients using multi-detector CT scanner (GE Bright®, speed 8 detector) on admission. The patients were placed in supine position with the head advanced and breathing held for scanning. The scan level was from the chest entrance to the bottom of the lungs.

The scanning parameters were as follows: thickness of the slices = 5 mm, interslice gap = 5 mm, matrix = 512 mm× 512 mm, tube voltage = 120 kV, current = 250-300 mA, collimation of 10 mm, and pitch of 1:1.675. No contrast was administered. The post-processing workstation reconstructed all images using high-resolution and conventional algorithms with lung window of +1600 to 600 HUs and mediastinal window of 40 to 350 HUs.

Image analysis

The HRCT images were evaluated for the following characteristics: (1) 4 types of lesion distributions, the side distribution; unilateral or bilateral, the transverse distribution; central and peripheral, the cranio-caudal distribution; upper lung predominant, lower lung predominant, and the scattering distribution, 1, 2 and ≥ 3 lesions; (2) lesion morphology: ground-glass opacities (GGO), consolidation, mixed GGO and consolidation; (3) pleural effusion.

The clinical data and disease course, including recovery, intensive care unit (ICU) admission, and mechanical ventilation, were inspected. The study's outcome was either recovery or death.

Statistical analysis

Data were entered, verified and analysed using SPSS v.25.0 for Windows (SPSS, Inc., Chicago, IL, USA). Categorical variables were presented as frequencies and percentages, and continuous variables were presented as mean and standard error (S.E). Independent-samples t test and Chi square test were used for mean comparison between studied groups. Odds ratios (OR) with corresponding 95% confidence intervals (CI) were calculated. Area under the receiver operating characteristic (ROC) curve was used for testing sensitivity and specificity of studied variables. Pearson correlation was used to test correlation among the different studied parameters. Statistical significance was assumed at a level of p values < 0.05.

Consent to participate. The manuscript has been read and approved by all authors.

RESULTS:

1-Demographic data and clinical findings among patients

105 COVID-19 patients were enrolled in this study including 68 males and 37 females (65%, 35%; respectively) with mean age (54 ± 1.5) and range (17-89 years old). Patients showed variable symptoms including fever, cough, fatigue, chest pain, dyspnea, headache and gastrointestinal manifestations (65%,72%,68%,38%, 49%,2%, and 17%; respectively). Taste and smell were lost in 41% of patients. 1% lost taste only while 3 % lost smell only. (Table 1).

2-Treatment protocol and hospitalization fate in patients

Among our patients, 48% were home isolated, while the remaining 52% were hospitalised. From 105 patients, 3 (2.9%) received no treatment, 51 (48.6%) patients received home treatment protocol, and 51 (48.6%) had hospital treatment protocol. 73% of patients were recovered and 27% died. 

From the recovered patients 65% were home isolated; while 35% were hospitalized, 14% of them were ICU admitted and 5 % were mechanically ventilated. All died patients were hospitalized 100%; 89% of them were ICU admitted and 79 % were mechanically ventilated. (Table 2).

3-Laboratory findings among COVID-19 patients

3.1 Complete blood picture findings

Some CBC parameters showed statistically significant decrease between dead and recovered patients as regards haemoglobin (Hb), haematocrit (HCT), mean corpuscular volume (MCV), and mean corpuscular haemoglobin (MCH) (p= 0.001, 0.038, 0.019, and 0.004 respectively), while statistically significant increase was shown in LMR (p= 0.044) (Table 3).

3.2 Biochemical and endothelial markers

When comparing dead patients to those who recovered, there was a statistically significant increase in biochemical markers including; D-Dimer, CRP, ferritin, LDH, Troponin, ALT, AST, ALT/AST ratio, urea, and creatinine. (Table 4).

The endothelial activation and stress index marker (EASIX) showed a highly statistically significant increase in dead patients compared to recovered one (p=0.00) (Table 4). 

4- Sensitivity, specificity and AUC for predicting hospitalization fate

4.1 Sensitivity, specificity and AUC For predicting mortality in COVID 19 patients

D-Dimer (ng/ml), ferritin (μg/L), EASIX, and lymphocyte/monocyte ratio (LMR) were analysed as predictors for COVID-19 mortality. The four parameters showed statistically significant sensitivity and specificity (p= 0.00,0.00,0.00, and 0.044; respectively). D-Dimer had the highest accuracy of 86% (AUC=0.86) followed by ferritin 77.2% (AUC=0.772), EASIX 72.8% (AUC=0.728), and the least one was for LMR with the value of 61.5% (AUC=0.615). Table 5, Figure 1

4.2 Sensitivity, specificity and AUC For predicting ICU admission in COVID 19 patients

By studying the same four parameters as regards ICU admission in COVID-19 patients we found that all of them had significant sensitivity and specificity (p=0.01,0.00,0.00, and 0.033; respectively). Table 6, Figure 2. D-Dimer had the highest overall accuracy of 74 (AUC= 0.744), followed by ferritin 72 % (AUC=0.719), EASIX 65% (AUC=0.653, and the lowest accuracy appear in LMR 60% (AUC=0.595).

4.3 Sensitivity, specificity and AUC For predicting mechanical ventilation in COVID 19 patients

When the previous four parameters were analysed by measuring their sensitivity and specificity as predictors for requirement of mechanical ventilation in our patients, we found that D-Dimer, ferritin, and EASIX were statically significant with p value of 0.005, 0.00, and 0.00; respectively with over all accuracy of 82 % for D-Dimer, 74 % for ferritin and 69% for EASIX; while LMR didn’t show any statistically significant sensitivity nor specificity. (Table 7, Figure 3).

5-CT findings & Cardiac assessment tests

Table (8&9) showed the detailed analysis of chest CT imaging;39 patients (37.14%) were classified as early stage (Figure 4) from them 16 patients had ground glass opacity (GGO), 8 patients had patchy consolidation while 15 patients had mixed of GGO and consolidation. Progression stage were detected in 54 patients (Figure 5), from them 8 (14.81%) patients had GGO, 15 (27.77%) patients had patchy consolidation and 31 (57.40%) patients had mixed of GGO and consolidation. The last 12 (11.42%) patients were classified as advanced stage (Figure 6) from them 4 (33.33%) patients had patchy consolidation and 8 (66.66%) had mixed of GGO and consolidation.

Table 7 also showed that both Echo & ECG detect the same findings in which all recovered patients were free of any cardiac complications after infection while 11/28 of died patient (39.3%) expressed cardiac problems .

Table 1: Demographic data and clinical findings in 105 COVID-19 patients

Table 9: Chest CT findings of 105 patients with COVID-19 pneumonia

DISCUSSION