A Machine Learning Model for Predicting Hospitalization in Patients with Respiratory Symptoms during the COVID-19 Pandemic

A systematic review of tools for predicting complications in patients with influenza-like illness

Objective

To identify tools that predict the risk of complications for patients presenting to an outpatient clinic or an emergency department (ED) with influenza-like illness.

Methods

We searched Medline, Embase, Cochrane Library and CINAHL from inception to July 2023. We included articles reporting on the derivation or validation of a score or algorithm used to stratify the risk of hospitalization or mortality among patients with influenza-like illness in the ED or outpatient clinic.

Results

Twelve articles reporting on eight scores and six predictive models were identified. For predicting the need for hospitalization, the area under the curve (AUC) of the PMEWS and the CURB-65 ranged respectively from 0.76 to 0.94, and 0.65 to 0.88. The Community Assessment Tool had an AUC of 0.62. For predicting inpatient mortality, AUC was 0.66 for PMEWS and 0.79 for CURB-65, 0.79 for the SIRS criteria and 0.86 for the qSOFA score. Two scores were developed without external validation during the Covid-19 pandemic. The CovHos score and the Canadian Covid discharge score had an AUC ranged from 0.70 to 0.91. The predictive models performed adequately (AUC from 0.76 to 0.92) but will require external validation for clinical use. Tool diversity and study population heterogeneity precluded meta-analysis.

Conclusion

Although the CURB, PMEWS and qSOFA scores appear to predict accurately the risk of complications of influenza-like illness, none were reliable enough to justify their widespread ED use. Refinement of an existing tool or development of a new tool to optimize the management of these patients is needed.

Decision trees for early prediction of inadequate immune response to coronavirus infections: a pilot study on COVID-19

Introduction

Few artificial intelligence models exist to predict severe forms of COVID-19. Most rely on post-infection laboratory data, hindering early treatment for high-risk individuals.

Methods

This study developed a machine learning model to predict inherent risk of severe symptoms after contracting SARS-CoV-2. Using a Decision Tree trained on 153 Alpha variant patients, demographic, clinical and immunogenetic markers were considered. Model performance was assessed on Alpha and Delta variant datasets. Key risk factors included age, gender, absence of KIR2DS2 gene (alone or with HLA-C C1 group alleles), presence of 14-bp polymorphism in HLA-G gene, presence of KIR2DS5 gene, and presence of KIR telomeric region A/A.

Results

The model achieved 83.01% accuracy for Alpha variant and 78.57% for Delta variant, with True Positive Rates of 80.82 and 77.78%, and True Negative Rates of 85.00% and 79.17%, respectively. The model showed high sensitivity in identifying individuals at risk.

Discussion

The present study demonstrates the potential of AI algorithms, combined with demographic, epidemiologic, and immunogenetic data, in identifying individuals at high risk of severe COVID-19 and facilitating early treatment. Further studies are required for routine clinical integration.

Oxygen Requirement in Overweight/Obese Kidney Transplant Recipients with COVID-19: An Observational Cohort Study

INTRODUCTION

Obesity is one of the components of the cardiometabolic syndrome that contributes to COVID-19 progression and mortality. Immunosuppressed individuals are at greater risk of the COVID-19 burden. Therefore, we sought to investigate the impact of the combination of overweight/obesity and kidney transplant on oxygen (O₂) requirements in the COVID-19 setting.

METHODS

Retrospective analysis of 284 kidney transplant recipients (KTRs) from March/2020 to August/2020 in a single center. We investigated the risk factors associated with O₂ requirements in overweight/obese KTRs.

RESULTS

Overall, 65.1% had a BMI (body mass index) ≥ 25 kg/m², 52.4% were male, the mean age was 53.3 ± 11 years old, 78.4% had hypertension, and 41.1% had diabetes mellitus. BMI was an independent risk factor for O₂ requirements (OR = 1.07, p = 0.02) alongside age, lymphopenia, and hyponatremia. When overweight/obese KTRs were older, smokers, they presented higher levels of lactate dehydrogenase (LDH), and lower levels of estimated glomerular filtration rate (eGFR), lymphocytes, and sodium at admission, and they needed O₂ more often.

CONCLUSION

Being overweight/obese is associated with greater O₂ requirements in KTRs, in particular in older people and smokers, with worse kidney allograft functions, more inflammation, and lower sodium levels. Therefore, the early identification of factors that predict a worse outcome in overweight/obese KTRs affected by COVID-19 contributes to risk stratification and therapeutic decisions.

Predictive value of immunoglobulin G, activated partial thromboplastin time, platelet, and indirect bilirubin for delayed viral clearance in patients infected with the Omicron variant

Background

Omicron is the recently emerged highly transmissible severe acute respiratory syndrome coronavirus 2 variant that has caused a dramatic increase in coronavirus disease-2019 infection cases worldwide. This study was to investigate the association between demographic and laboratory findings, and the duration of Omicron viral clearance.

Methods

Approximately 278 Omicron cases at the Ruijin Hospital Luwan Branch, Shanghai Jiaotong University School of Medicine were retrospectively analyzed between August 11 and August 31, 2022. Demographic and laboratory data were also collected. The association between demographics, laboratory findings, and duration of Omicron viral clearance was analyzed using Pearson correlation analysis and univariate and multivariate logistic regression.

Results

Univariate logistic regression analyses showed that a prolonged viral clearance time was significantly associated with older age and lower immunoglobulin (Ig) G and platelet (PLT) levels. Using multinomial logistic regression analyses, direct bilirubin, IgG, activated partial thromboplastin time (APTT), and PLT were independent factors for longer viral shedding duration. The model combining direct bilirubin, IgG, APTT, and PLT identifies patients infected with Omicron whose viral clearance time was ≥7 days with 62.7% sensitivity and 83.4% specificity.

Conclusion

These findings suggest that direct bilirubin, IgG, PLT, and APTT are significant risk factors for a longer viral shedding duration in patients infected with Omicron. Measuring levels of direct bilirubin, IgG, PLT, and APTT is advantageous to identify patients infected with Omicron with longer viral shedding duration.

Impact of Hypertension on COVID-19 Burden in Kidney Transplant Recipients: An Observational Cohort Study

BACKGROUND

COVID-19 severity is determined by cardiometabolic risk factors, which can be further aggravated by chronic immunosuppression in kidney transplant recipients (KTRs). We aimed to verify the main risk factors related to hypertension (HTN) that contribute to COVID-19 progression and mortality in that population.

METHODS

Retrospective analysis of 300 KTRs from March 2020 to August 2020 in a single center. We compared the main outcomes between HTN (n = 225) and non-HTN (n = 75), including admission to the intensive care unit (ICU), development of acute kidney injury (AKI), need for invasive mechanical ventilation or oxygen, and mortality.

RESULTS

Of the patients in the study, 57.3% were male, 61.3% were white, the mean age was 52.5 years, and 75% had HTN. Pre-existing HTN was independently associated with higher rates of mortality (32.9%, OR = 1.96, p = 0.036), transfer to the ICU (50.7%, OR = 1.94, p = 0.017), and AKI with hemodialysis (HD) requirement (40.4%, OR = 2.15, p = 0.011). In the hypertensive group, age, diabetes mellitus, heart disease, smoking, glycemic control before admission, C-reactive protein, lactate dehydrogenase, lymphocytes, and D-dimer were significantly associated with COVID-19 progression and mortality. Both lower basal and previous estimated glomerular filtration rates posed KTRs with HTN at greater risk for HD requirement.

CONCLUSIONS

Therefore, the early identification of factors that predict COVID-19 progression and mortality in KTRs affected by COVID-19 contributes to therapeutic decisions, patient flow management, and allocation of resources.