Routine Hematological Parameters May Be Predictors of COVID-19 Severity

The role of platelet large cell ratio in determining mortality in COVID-19 patients

Inflammatory mechanisms have been implicated and have been subject to research in the clinical course of COVID-19 patients. In this study, platelet large cell ratio (P-LCR) has been examined as a novel prognostic and inflammatory parameter. A total of 1992 COVID-19-positive patients admitted to COVID-19 unit of Infectious Diseases were included. In order to identify a potential relationship between P-LCR and mortality, surviving patients were compared with subjects who died as a result of the disease. Although P-LCR levels showed a steady increase in all COVID-19 patients after admission, they were significantly higher in those who eventually died (P < .001), indicating a positive correlation between mortality and P-LCR. The P-LCR levels of patients followed up in the intensive care unit were statistically significantly higher than those followed up in the ward (P < .001). P-LCR levels of patients intubated in intensive care unit were statistically significantly higher than those who were not intubated (P < .001). Also, P-LCR levels were subdivided into 3 categories as normal, low, and elevated. Elevated P-LCR was found to be positively correlated with leukocyte count, neutrophil count, D-dimer, troponin, ferritin, and C-Reactive Protein (CRP) and showed negative correlation with fibrinogen, lymphocyte count, and platelet count. As P-LCR was correlated with the severity of inflammation in all COVID-19 patients, it was significantly higher in those patients who died. Elevated P-LCR was considered to be associated with the risk of severe disease and death. This inexpensive, readily available test may be incorporated into our clinical practice as a novel marker of poor prognosis in addition to other valuable laboratory parameters.

Hematological parameters and their predictive value for assessing disease severity in laboratory-confirmed COVID-19 patients: a retrospective study

BACKGROUND

The coronavirus disease 19 (COVID-19) infection has spread globally and caused a substantial amount of mortality and morbidity. Early detection of severe infections will improve care and reduce deaths. The use of hematological parameters in predicting COVID-19 disease severity, patient outcomes, and early risk stratification is limited. Therefore, the study was aimed at determining hematological parameters and their predictive value for assessing disease severity in laboratory-confirmed COVID-19 patients in Northwest Ethiopia.

METHODS

A retrospective cross-sectional study was conducted at the University of Gondar comprehensive specialized hospital and Tibebe Ghion comprehensive specialized referral hospital on 253 patients diagnosed with COVID-19 and admitted between March 2021 and February 2022. Data were extracted, and entered into Epi-data 4.2.0.0, and analyzed using SPSS version 25 software. Hematological parameters were provided as the median and interquartile range (IQR). Categorical variables were represented by their frequency, and the χ2 test was applied to compare observed results with expected results. The receiver-operating curve (ROC) was used to establish the predictive value of hematological parameters for COVID-19 severity. A p-value < 0.05 was considered statistically significant.

RESULTS

On a total of 253 patients, there were 43.87% severe cases, with a mortality rate of 26.9%. The ROC analysis showed the optimal cutoff values for hematological parameters were ANC (3370), lymphocyte (680), NLR (9.34), PLR (290.77), platelets (332,000), and WBCs (4390.65). The area under the curve (AUC) values for NLR (0.679) and ANC (0.631) were high, with the highest sensitivity and specificity, and could potentially be used to predict COVID-19 severity.

CONCLUSION

This study proved that high NLR and high ANC have prognostic value for assessing disease severity in COVID-19. Thus, assessing and considering these hematological parameters when triaging COVID-19 patients may prevent complications and improve the patient's outcome.

Rapid protocols to support COVID-19 clinical diagnosis based on hematological parameters

Purpose

In December 2019, the Covid-19 pandemic began in the world. To reduce mortality, in addiction to mass vaccination, it is necessary to massify and accelerate clinical diagnosis, as well as creating new ways of monitoring patients that can help in the construction of specific treatments for the disease.

Objective

In this work, we propose rapid protocols for clinical diagnosis of COVID-19 through the automatic analysis of hematological parameters using evolutionary computing and machine learning. These hematological parameters are obtained from blood tests common in clinical practice.

Method

We investigated the best classifier architectures. Then, we applied the particle swarm optimization algorithm (PSO) to select the most relevant attributes: serum glucose, troponin, partial thromboplastin time, ferritin, D-dimer, lactic dehydrogenase, and indirect bilirubin. Then, we assessed again the best classifier architectures, but now using the reduced set of features. Finally, we used decision trees to build four rapid protocols for Covid-19 clinical diagnosis by assessing the impact of each selected feature. The proposed system was used to support clinical diagnosis and assessment of disease severity in patients admitted to intensive and semi-intensive care units as a case study in the city of Paudalho, Brazil.

Results

We developed a web system for Covid-19 diagnosis support. Using a 100-tree random forest, we obtained results for accuracy, sensitivity, and specificity superior to 99%. After feature selection, results were similar. The four empirical clinical protocols returned accuracies, sensitivities and specificities superior to 98%.

Conclusion

By using a reduced set of hematological parameters common in clinical practice, it was possible to achieve results of accuracy, sensitivity, and specificity comparable to those obtained with RT-PCR. It was also possible to automatically generate clinical decision protocols, allowing relatively accurate clinical diagnosis even without the aid of the web decision support system.

Hematological findings in COVID-19 and their correlation with severity of Disease

Objective

To evaluate the efficacy of hematological parameters to predict severity of COVID-19 patients.

Method

This was a cross-sectional comparative study conducted at Central Park Teaching Hospital, Lahore in COVID ward and COVID ICU between April 23, 2021 to June 23, 2021. Patients of all ages and both genders with positive PCR admitted in the COVID ward and ICU during this time span of two months were included in the study. Data was collected retrospectively.

Results

This study included 50 patients with male to female ratio of 1.38:1. Though males are more affected by COVID-19 but the difference is not statistically significant. The mean age of the study population was 56.21 and the patients in the severe disease group have higher age. It was observed that in severe/critical group the mean values of total leukocyte count 21.76×103 μI (p-value= 0.002), absolute neutrophil count 71.37% (p-value=0.045), neutrophil lymphocyte ratio (NLR) 12.80 (p-value=0.00) and PT 11.9 seconds (p-value=0.034) and the difference was statistically significant. While in severe/critical group, the mean values of hemoglobin 12.03g/dl (p-value=0.075), lymphocyte count 28.41% (p-value=0.8), platelet count 226×103 μI (p-value=0.67) and APTT 30.7 (p-value=0.081) and the difference was not significantly different between groups.

Conclusion

It can be concluded from the study that total leucocyte count, absolute neutrophil count and neutrophil lymphocyte ratio can predict in-hospital mortality and morbidity in COVID-19 patients.

Hematologic analysis of hospitalized patients and outpatients infected with SARS-CoV-2 and possible use as a prognostic biomarker

Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infection is a global health problem; this has caused thousands of deaths around the world. This infection induces hematologic alterations, and it is necessary to recognize predictive biomarkers to address the need for hospitalization or the severity of the disease. This study aimed to analyze different parameters in outpatients and hospitalized patients infected with SARS-CoV-2 and determine whether hematic biometry can be used for prognosis rapidly. We analyzed 689 patients, of whom 355 were outpatients (162 women and 193 men) and 334 required hospitalization (197 men and 137 women). The average age of the hospitalized patients was 46 years (men, 49 years; women, 52 years), whereas the average age of the outpatients was 49 years (men, 51 years; women, 44 years). Hematologic parameters were analyzed and compared between the outpatients and hospitalized patients. The patients were divided into groups by age and sex. We found that in the hospitalized patients, the erythrocyte, hematocrit, and hemoglobin levels decreased, whereas the outpatients did not experience changes in the erythroid series. In leukocytes, these increased significantly, as they did in neutrophils; however, lymphocytopenia was observed. In the outpatients, we observed normal levels of neutrophils and lymphopenia. We can conclude that hematic biometry can be used as a biomarker, and the relation between neutrophils and lymphocytes is indicated for understanding the development and prognosis of the disease.

Haematological indices and coagulation profile as predictors of disease severity and associations with clinical outcome among COVID-19 patients in Lagos, Nigeria

Background

This study aims to evaluate the use of haematological indices and coagulation profiles as possible low-cost predictors of disease severity and their associations with clinical outcomes in COVID-19-hospitalized patients in Nigeria.

Materials and Methods

We carried out a hospital-based descriptive 3-month observational longitudinal study of 58 COVID-19-positive adult patients admitted at the Lagos University Teaching Hospital, Lagos, Nigeria. We used a structured questionnaire to obtain the participants' relevant sociodemographic and clinical data, including disease severity. Basic haematologic indices, their derivatives, and coagulation profile were obtained from patients' blood samples. Receiver Operating Characteristic (ROC) analysis was used to compare these laboratory-based values with disease severity. A P < 0.05 was considered statistically significant.

Results

The mean age of the patients was 54.4 ± 14.8 years. More than half of the participants were males (55.2%, n = 32) and most had at least one comorbidity (79.3%, n = 46). Significantly higher absolute neutrophil count (ANC), neutrophil-lymphocyte ratio (NLR), systemic immune-inflammation index (SII), lower absolute lymphocyte count (ALC) and lymphocyte-monocyte ratio (LMR) were associated with severe disease (P < 0.05). Patients' hemoglobin concentration (P = 0.04), packed cell volume (P < 0.001), and mean cell hemoglobin concentration (P = 0.03) were also significantly associated with outcome. Receiver operating characteristic (ROC) analysis of disease severity was significant for the ANC, ALC, NLR, LMR, and SII. The coagulation profile did not show any significant associations with disease severity and outcomes in this study.

Conclusion

Our findings identified haematological indices as possible low-cost predictors of disease severity in COVID-19 in Nigeria.

Analysis of COVID-19 severity from the perspective of coagulation index using evolutionary machine learning with enhanced brain storm optimization

Coronavirus 2019 (COVID-19) is an extreme acute respiratory syndrome. Early diagnosis and accurate assessment of COVID-19 are not available, resulting in ineffective therapeutic therapy. This study designs an effective intelligence framework to early recognition and discrimination of COVID-19 severity from the perspective of coagulation indexes. The framework is proposed by integrating an enhanced new stochastic optimizer, a brain storm optimizing algorithm (EBSO), with an evolutionary machine learning algorithm called EBSO-SVM. Fast convergence and low risk of the local stagnant can be guaranteed for EBSO with added by Harris hawks optimization (HHO), and its property is verified on 23 benchmarks. Then, the EBSO is utilized to perform parameter optimization and feature selection simultaneously for support vector machine (SVM), and the presented EBSO-SVM early recognition and discrimination of COVID-19 severity in terms of coagulation indexes using COVID-19 clinical data. The classification performance of the EBSO-SVM is very promising, reaching 91.9195% accuracy, 90.529% Matthews correlation coefficient, 90.9912% Sensitivity and 88.5705% Specificity on COVID-19. Compared with other existing state-of-the-art methods, the EBSO-SVM in this paper still shows obvious advantages in multiple metrics. The statistical results demonstrate that the proposed EBSO-SVM shows predictive properties for all metrics and higher stability, which can be treated as a computer-aided technique for analysis of COVID-19 severity from the perspective of coagulation.

The radiographic assessment of lung edema score of lung edema severity correlates with inflammatory parameters in patients with coronavirus disease 2019—Potential new admission biomarkers to predict coronavirus disease 2019 worsening

Background

Coronavirus disease 2019 (COVID-19) has placed enormous pressure on intensive care units (ICUs) and on healthcare systems in general. A deeper understanding of the pathophysiology of the most severe forms of COVID-19 would help guide the development of more effective interventions. Herein, we characterized the inflammatory state of patients with COVID-19 of varying degrees of severity to identify admission biomarkers for predicting COVID-19 worsening.

Design

Admission blood samples were obtained from 78 patients with COVID-19. Radiographic assessment of lung edema (RALE) scoring was calculated by imaging. Platelet and leukocyte counts were measured by flow cytometry, and plasma levels of C-reactive protein were assessed by immunoturbidimetry, and interleukin (IL)-8/CXCL8, IL-10, tumor necrosis factor (TNF)-α, interferon (IFN)-γ, and monocyte chemoattractant protein-1 (MCP-1/CCL2) levels by enzyme-linked immunosorbent assay (ELISA).

Results

The RALE score correlated with several admission hemogram (platelets, neutrophils, and lymphocytes) and inflammatory (IL-8/CXCL8, MCP-1/CCL2, IL-10, and C-reactive protein) parameters. COVID-19 worsening, based on the need for oxygen (Δoxygen supply) during hospitalization, correlated negatively with admission lymphocyte counts but positively with neutrophil-to-lymphocyte ratio and with plasma levels of the inflammatory parameters correlating with RALE score.

Conclusion

Our data indicate a correlation between the RALE score and Δoxygen supply and admission inflammatory status. The identification of a panel of biomarkers that reflect COVID severity might be useful to predict disease worsening during hospitalization and to guide clinical management of COVID-19-related complications. Finally, therapies targeting IL-8/CXCL8- or IL-10 activity may offer therapeutic approaches in COVID-19 treatment.

An Explainable AI Approach for the Rapid Diagnosis of COVID-19 Using Ensemble Learning Algorithms

Background

Artificial intelligence-based disease prediction models have a greater potential to screen COVID-19 patients than conventional methods. However, their application has been restricted because of their underlying black-box nature.

Objective

To addressed this issue, an explainable artificial intelligence (XAI) approach was developed to screen patients for COVID-19.

Methods

A retrospective study consisting of 1,737 participants (759 COVID-19 patients and 978 controls) admitted to San Raphael Hospital (OSR) from February to May 2020 was used to construct a diagnosis model. Finally, 32 key blood test indices from 1,374 participants were used for screening patients for COVID-19. Four ensemble learning algorithms were used: random forest (RF), adaptive boosting (AdaBoost), gradient boosting decision tree (GBDT), and extreme gradient boosting (XGBoost). Feature importance from the perspective of the clinical domain and visualized interpretations were illustrated by using local interpretable model-agnostic explanations (LIME) plots.

Results

The GBDT model [area under the curve (AUC): 86.4%; 95% confidence interval (CI) 0.821–0.907] outperformed the RF model (AUC: 85.7%; 95% CI 0.813–0.902), AdaBoost model (AUC: 85.4%; 95% CI 0.810–0.899), and XGBoost model (AUC: 84.9%; 95% CI 0.803–0.894) in distinguishing patients with COVID-19 from those without. The cumulative feature importance of lactate dehydrogenase, white blood cells, and eosinophil counts was 0.145, 0.130, and 0.128, respectively.

Conclusions

Ensemble machining learning (ML) approaches, mainly GBDT and LIME plots, are efficient for screening patients with COVID-19 and might serve as a potential tool in the auxiliary diagnosis of COVID-19. Patients with higher WBC count, higher LDH level, or higher EOT count, were more likely to have COVID-19.

Prognostic Value of Serum MICA Levels as a Marker of Severity in COVID-19 Patients

The COVID-19 global pandemic and high mortality rates necessitate the development of diagnostic and prognostic tools, as well as expanding testing capacity. Existing methods for detecting and characterizing SARS-CoV-2 infection are typically based on viral genome detection or measuring COVID-19-specific antibody levels. Despite their value, these methods are unable to predict disease outcomes in patients. Given the critical role of innate immune cells, particularly natural killer (NK) cells, in antiviral defense, this study sought to determine the prognostic value of serum secretory MHC class I polypeptide-related sequence A (sMICA) levels as an essential ligand for the NKG2D receptor, the master regulator of NK cell development and responsiveness. Serum MICA levels were measured by ELISA assay. Sera (n = 60) from SARS-CoV-2 positive patients were collected, and disease severity was determined using clinical criteria. The patient group included 30 patients with mild disease and 30 severely ill patients, as well as 30 healthy controls. Our findings revealed that serum MICA levels were significantly higher in patients than in controls, especially in cases with severe complications (P < .0001). Higher serum MICA levels may be associated with respiratory failure in COVID-19 and may serve as a marker of clinical severity in patients infected with SARS-CoV-2, particularly when clinical manifestations are insufficient to make a confident prediction.

Two Different Clinical Presentations of Acute Limb Ischemia Caused by Acute Thrombotic Events in COVID-19

Coronavirus disease 2019 (COVID-19) is a global pandemic caused by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2). Coagulopathy is frequently found in severe cases of COVID-19 and is usually manifested as a prothrombotic state. Hyperinflammation, endotheliitis, and immobilization during illness are hypothesized to play a role. Acute limb ischemia (ALI) is one of the presentations of arterial thrombosis in COVID-19. We present two cases of middle-aged men with COVID-19 infection, who developed ALI. The first patient developed ALI after 16 days from the initial COVID-19 diagnosis, and the second patient was admitted to the emergency ward due to sudden discoloration of his right lower limb, and COVID-19 was diagnosed during the evaluation.