Clinical Hematochemical Parameters in Differential Diagnosis between Pediatric SARS-CoV-2 and Influenza Virus Infection: An Automated Machine Learning Approach

Diagnostic value of routine blood tests in differentiating between SARS-CoV-2, influenza A, and RSV infections in hospitalized children: a retrospective study

BACKGROUND

The Severe Acute Respiratory Syndrome Coronavirus 2 (SARS-CoV-2), influenza A, and respiratory syncytial virus (RSV) infections have similar modes of transmission and clinical symptoms. There is a need to identify simple diagnostic indicators to distinguish these three infections, particularly for community hospitals and low- and middle-income countries that lack nucleic acid detection kits. This study used clinical data to assess the diagnostic value of routine blood tests in differentiating between SARS-CoV-2, influenza A, and RSV infections in children.

METHODS

A total of 1420 children treated at the Hangzhou Children's Hospital between December 2022 and June 2023 were enrolled in this study, of whom 351 had SARS-CoV-2, 671 had influenza, and 398 had RSV. In addition, 243 healthy children were also collected. The blood test results of SARS-CoV-2 patients were compared to those of patients with influenza A and RSV and the healthy controls. The area under the receiver operating characteristic curve (AUC-ROC) was employed to evaluate each blood parameter's diagnostic value.

RESULTS

Children with SARS-CoV-2 exhibited notably elevated levels of white blood cell (WBC) count, platelet (PLT) count, neutrophil count, and neutrophil-to-lymphocyte ratio (NLR) compared to influenza A patients (P < 0.05). In contrast, SARS-CoV-2 patients exhibited a decrease in the mean platelet volume to platelet count ratio (MPV/PLT) and the lymphocyte-to-monocyte ratio (LMR) when compared to other individuals (P < 0.05). These parameters had an AUC between 0.5 and 0.7. Compared to patients with RSV, SARS-CoV-2 patients had significantly higher MPV/PLT and significantly lower WBC, lymphocyte, PLT, LMR, and lymphocyte multiplied by platelet (LYM*PLT) values (P < 0.05). However, only LYM*PLT had an acceptable diagnostic value above 0.7 for all age groups. Compared to healthy children, children with COVID-19 exhibited elevated NLR and MPV/PLT levels, alongside decreased lymphocyte, PLT, LMR, and LYM*PLT values. (P < 0.05). The AUC of the LMR, LYM*PLT, and PLT were above 0.7 in all age groups, indicating promising diagnostic values.

CONCLUSIONS

The routine blood parameters among patients with COVID-19, influenza A, and RSV differ significantly early in the disease and could be used by clinicians to discriminate between the 3 types of infection.

Platelet distribution width (PDW) as a significant correlate of COVID-19 infection severity and mortality

SARS-CoV-2 infection may cause a wide spectrum of symptoms, from asymptomatic, to mild respiratory symptoms and life-threatening sepsis. Among the clinical laboratory biomarkers analyzed during COVID-19 pandemic, platelet indices have raised great interest, due to the critical involvement of platelets in COVID-19-related thromboinflammation. Through an electronic literature search on MEDLINE, CINAHL, PubMed, EMBASE, Web of Science, and preprint servers we performed and updated a systematic review aimed at providing a detailed analysis of studies addressing the potential clinical utility of platelet distribution width, platelet distribution width (PDW), in laboratory medicine, exploring the possible association between increased PDW levels, disease severity, and mortality in COVID-19. Our systematic review revealed a wide heterogeneity of COVID-19 cohorts examined and a lack of homogenous expression of platelet indices. We found that 75 % of studies reported significantly elevated PDW values in COVID-19 infected cohorts compared to healthy/non-COVID-19 controls, and 40 % of studies reported that patients with severe COVID-19 showed increased PDW values than those with less-than-severe illness. Interestingly, 71.4 % of studies demonstrated significant increased PDW values in non survivors vs. survivors. Overall, these results suggest that platelets are critically involved as major players in the process of immunothrombosis in COVID-19, and platelet reactivity and morphofunctional alterations are mirrored by PDW, as indicator of platelet heterogeneity. Our results confirm that the use of PDW as prognostic biomarkers of COVID-19 sepsis still remains debated due to the limited number of studies to draw a conclusion, but new opportunities to investigate the crucial role of platelets in thrombo-inflammation are warranted.

COVID-19 Pediatric Follow-Up: Respiratory Long COVID-Associated Comorbidities and Lung Ultrasound Alterations in a Cohort of Italian Children

In children, the factors that influence COVID-19 disease and its medium- and long-term effects are little known. Our investigation sought to evaluate the presence of comorbidity factors associated with respiratory long COVID manifestations in children and to study ultrasound abnormalities following SARS-CoV-2 infection. Children, who arrived at the 'Respiratory Diseases of Pediatric Interest Unit' at the Department of Woman, Child, and General and Specialized Surgery of the University of Campania 'Luigi Vanvitelli', were selected during the timeframe from September 2021 to October 2022. The children were diagnosed with a SARS-CoV-2 infection that occurred at least one month before the visit. All patients followed a COVID-19 follow-up protocol, developed by the Italian Society of Pediatric Respiratory Diseases (SIMRI), which included: collection of data regarding SARS-CoV-2 illness and history of known respiratory and allergic diseases; physical examination; BMI assessment; baseline spirometry and after bronchodilation test; six-minute walking test; and lung ultrasound (LUS). In a cohort of 104 participants with respiratory long COVID symptoms (64.7% male, average age 8.92 years), 46.1% had fever with other symptoms, and 1% required hospitalization. BMI analysis showed 58.4% of the cohort was overweight. The LUS was positive in 27.0% of cases. A significant BMI association was observed with COVID-19 symptoms and LUS score (p-value < 0.05). No associations were found with asthma or atopy.

Dynamic Assessment of Plasma von Willebrand Factor and ADAMTS13 Predicts Mortality in Hospitalized Patients with SARS-CoV-2 Infection

BACKGROUND

Plasma levels of von Willebrand factor (VWF) are significantly elevated in patients with coronavirus disease 2019 (COVID-19). However, dynamic changes and prognostic value of this biomarker in hospitalized patients with COVID-19 have not been determined.

METHODS

A total of 124 patients infected with SARS-CoV-2 were prospectively recruited for the study. Serial blood samples were obtained at the time of admission (D1), 3-4 days following standard-care treatments (D2), and 1-2 days prior to discharge or any time collected prior to death (D3). Plasma VWF antigen, ADAMTS13 antigen, and ADAMTS13 proteolytic activity, as well as the ratio of VWF/ADAMTS13 were determined, followed by various statistical analyses.

RESULTS

On admission, plasma levels of VWF in COVID-19 patients were significantly elevated compared with those in the healthy controls, but no statistical significance was detected among patients with different disease severity. Plasma ADAMTS13 activity but not its antigen levels were significantly lower in patients with severe or critical COVID-19 compared with that in other patient groups. Interestingly, the ratios of plasma VWF antigen to ADAMTS13 antigen were significantly higher in patients with severe or critical COVID-19 than in those with mild to moderate disease. More importantly, plasma levels of VWF and the ratios of VWF/ADAMTS13 were persistently elevated in patients with COVID-19 throughout hospitalization. Kaplan-Meier and Cox proportional hazard regression analyses demonstrated that an increased plasma level of VWF or ratio of VWF/ADAMTS13 at D2 and D3 was associated with an increased mortality rate.

CONCLUSIONS

Persistent endotheliopathy, marked by the elevated levels of plasma VWF or VWF/ADAMTS13 ratio, is present in all hospitalized patients following SARS-CoV-2 infection, which is strongly associated with mortality.

Can mean platelet volume be a prognosis predictor in viral infections: An example of Covid-19

Purpose

This study revealed the utility of mean platelet volume (MPV) as a mortality marker in SARS-CoV-2 infection, as well as its connection with other inflammatory indicators such as procalcitonin (PCT) and neutrophil-lymphocyte ratio (NLR).

Methods

A total of 1528 patients (853 males and 653 females; mean age: 64.33 ± 16.36 years; range, 18-100 years) were hospitalized with COVID-19 between March 2020 and December 2022. The patients' demographic and clinical information, including ward and critical care data, were gathered from their medical records. On the first and last days, the PCT, NLR, and MPV values of the patients, who were divided into groups based on their hospitalization and outcomes, were analyzed.

Results

When the relevant laboratory data from the first and last days were compared, each group was statistically significant (p < 0.05). There was a moderate association between the final MPV values and the PCT and NLR values of the patients admitted to the ward (r = 0.448 and r = 0.397, respectively, where p < 0.01). There was also a substantial and moderate correlation between the final MPV levels and the PCT and NLR values of patients admitted to the intensive care unit (r = 0.613 and r = 0.361, respectively, p < 0.01). When compared to the patients' outcomes, the MPV had greater specificity and AUC values than the PCT and NLR (94.4 %, 0.968, 80.6 %, 0.923, 81 %, 0.845, respectively).

Conclusion

In patients hospitalized with COVID-19, the specificity of MPV values at the point of sickness severity and outcome was shown to be greater than PCT and NLR values, and MPV values may be a more accurate predictor of mortality than PCR and NLR.

Rapid Triage of Children with Suspected COVID-19 Using Laboratory-Based Machine-Learning Algorithms

In order to limit the spread of the novel betacoronavirus (SARS-CoV-2), it is necessary to detect positive cases as soon as possible and isolate them. For this purpose, machine-learning algorithms, as a field of artificial intelligence, have been recognized as a promising tool. The aim of this study was to assess the utility of the most common machine-learning algorithms in the rapid triage of children with suspected COVID-19 using easily accessible and inexpensive laboratory parameters. A cross-sectional study was conducted on 566 children treated for respiratory diseases: 280 children with PCR-confirmed SARS-CoV-2 infection and 286 children with respiratory symptoms who were SARS-CoV-2 PCR-negative (control group). Six machine-learning algorithms, based on the blood laboratory data, were tested: random forest, support vector machine, linear discriminant analysis, artificial neural network, k-nearest neighbors, and decision tree. The training set was validated through stratified cross-validation, while the performance of each algorithm was confirmed by an independent test set. Random forest and support vector machine models demonstrated the highest accuracy of 85% and 82.1%, respectively. The models demonstrated better sensitivity than specificity and better negative predictive value than positive predictive value. The F1 score was higher for the random forest than for the support vector machine model, 85.2% and 82.3%, respectively. This study might have significant clinical applications, helping healthcare providers identify children with COVID-19 in the early stage, prior to PCR and/or antigen testing. Additionally, machine-learning algorithms could improve overall testing efficiency with no extra costs for the healthcare facility.