A Tool for Early Prediction of Severe Coronavirus Disease 2019 (COVID-19): A Multicenter Study Using the Risk Nomogram in Wuhan and Guangdong, China

Construction and Validation of a Predictive Model for the Risk of Ventilator-Associated Pneumonia in Elderly ICU Patients

Background

Ventilator-associated pneumonia (VAP) is among the most important hospital-acquired infections in an intensive-care unit setting. However, clinical practice lacks effective theoretical tools for preventing VAP in the elderly.

Aim

To describe the independent factors associated with VAP in elderly intensive-care unit (ICU) patients on mechanical ventilation (MV) and to construct a risk prediction model.

Methods

A total of 1851 elderly patients with MV in ICUs from January 2015 to September 2019 were selected from 12 tertiary hospitals. Study subjects were divided into a model group (n = 1219) and a validation group (n = 632). Two groups of patients were divided into a VAP group and a non-VAP group and compared. Univariate and logistic regression analyses were used to explore influencing factors for VAP in elderly ICU patients with MV, establish a risk prediction model, and draw a nomogram. We used the area under the receiver operating characteristic curve (AUROC) and the Hosmer-Lemeshow goodness-of-fit test to evaluate the predictive effect of the model. Findings regarding the length of ICU stay, surgery, C-reactive protein (CRP), and the number of reintubations were independent risk factors for VAP in elderly ICU patients with MV. Predictive-model verification results showed that the area under the curve (AUC) of VAP risk after MV in the modeling and verification groups was 0.859 and 0.813 (P < 0.001), respectively, while P values for the Hosmer-Lemeshow test in these two groups were 0.365 and 0.485, respectively.

Conclusion

The model could effectively predict the occurrence of VAP in elderly patients with MV in ICUs. This study is a retrospective study, so it has not been registered as a clinical study.

A novel evidence-based predictor tool for hospitalization and length of stay: insights from COVID-19 patients in New York city

Predictive models for key outcomes of coronavirus disease 2019 (COVID-19) can optimize resource utilization and patient outcome. We aimed to design and internally validate a web-based calculator predictive of hospitalization and length of stay (LOS) in a large cohort of COVID-19-positive patients presenting to the Emergency Department (ED) in a New York City health system. The study cohort consisted of consecutive adult (> 18 years) patients presenting to the ED of Mount Sinai Health System hospitals between March 2020 and April 2020, diagnosed with COVID-19. Logistic regression was utilized to construct predictive models for hospitalization and prolonged (> 3 days) LOS. Discrimination was evaluated using area under the receiver operating curve (AUC). Internal validation with bootstrapping was performed, and a web-based calculator was implemented. From 5859 patients, 65% were hospitalized. Independent predictors of hospitalization and extended LOS included older age, chronic kidney disease, elevated maximum temperature, and low minimum oxygen saturation (p < 0.001). Additional predictors of hospitalization included male sex, chronic obstructive pulmonary disease, hypertension, and diabetes. AUCs of 0.881 and 0.770 were achieved for hospitalization and LOS, respectively. Elevated levels of CRP, creatinine, and ferritin were key determinants of hospitalization and LOS (p < 0.05). A calculator was made available under the following URL: https://covid19-outcome-prediction.shinyapps.io/COVID19_Hospitalization_Calculator/ . This study yielded internally validated models that predict hospitalization risk in COVID-19-positive patients, which can be used to optimize resource allocation. Predictors of hospitalization and extended LOS included older age, CKD, fever, oxygen desaturation, elevated C-reactive protein, creatinine, and ferritin.

A risk scoring system to predict progression to severe pneumonia in patients with Covid-19

Rapid outbreak of coronavirus disease 2019 (Covid-19) raised major concern regarding medical resource constraints. We constructed and validated a scoring system for early prediction of progression to severe pneumonia in patients with Covid-19. A total of 561 patients from a Covid-19 designated hospital in Daegu, South Korea were randomly divided into two cohorts: development cohort (N = 421) and validation cohort (N = 140). We used multivariate logistic regression to identify four independent risk predictors for progression to severe pneumonia and constructed a risk scoring system by giving each factor a number of scores corresponding to its regression coefficient. We calculated risk scores for each patient and defined two groups: low risk (0 to 8 points) and high risk (9 to 20 points). In the development cohort, the sensitivity and specificity were 83.8% and 78.9%. In the validation cohort, the sensitivity and specificity were 70.8% and 79.3%, respectively. The C-statistics was 0.884 (95% CI 0.833–0.934) in the development cohort and 0.828 (95% CI 0.733–0.923) in the validation cohort. This risk scoring system is useful to identify high-risk group for progression to severe pneumonia in Covid-19 patients and can prevent unnecessary overuse of medical care in limited-resource settings.

Prognosis of COVID-19 pneumonia can be early predicted combining Age-adjusted Charlson Comorbidity Index, CRB score and baseline oxygen saturation

In potentially severe diseases in general and COVID-19 in particular, it is vital to early identify those patients who are going to progress to severe disease. A recent living systematic review dedicated to predictive models in COVID-19, critically appraises 145 models, 8 of them focused on prediction of severe disease and 23 on mortality. Unfortunately, in all 145 models, they found a risk of bias significant enough to finally "not recommend any for clinical use". Authors suggest concentrating on avoiding biases in sampling and prioritising the study of already identified predictive factors, rather than the identification of new ones that are often dependent on the database. Our objective is to develop a model to predict which patients with COVID-19 pneumonia are at high risk of developing severe illness or dying, using basic and validated clinical tools. We studied a prospective cohort of consecutive patients admitted in a teaching hospital during the "first wave" of the COVID-19 pandemic. Follow-up to discharge from hospital. Multiple logistic regression selecting variables according to clinical and statistical criteria. 404 consecutive patients were evaluated, 392 (97%) completed follow-up. Mean age was 61 years; 59% were men. The median burden of comorbidity was 2 points in the Age-adjusted Charlson Comorbidity Index, CRB was abnormal in 18% of patients and basal oxygen saturation on admission lower than 90% in 18%. A model composed of Age-adjusted Charlson Comorbidity Index, CRB score and basal oxygen saturation can predict unfavorable evolution or death with an area under the ROC curve of 0.85 (95% CI 0.80-0.89), and 0.90 (95% CI 0.86 to 0.94), respectively. Prognosis of COVID-19 pneumonia can be predicted without laboratory tests using two classic clinical tools and a pocket pulse oximeter.

Screening, Diagnostic and Prognostic Tests for COVID-19: A Comprehensive Review

While molecular testing with real-time polymerase chain reaction (RT-PCR) remains the gold-standard test for COVID-19 diagnosis and screening, more rapid or affordable molecular and antigen testing options have been developed. More affordable, point-of-care antigen testing, despite being less sensitive compared to molecular assays, might be preferable for wider screening initiatives. Simple laboratory, imaging and clinical parameters could facilitate prognostication and triage. This comprehensive review summarises current evidence on the diagnostic, screening and prognostic tests for COVID-19.

Screening, Diagnostic and Prognostic Tests for COVID-19: A Comprehensive Review

While molecular testing with real-time polymerase chain reaction (RT-PCR) remains the gold-standard test for COVID-19 diagnosis and screening, more rapid or affordable molecular and antigen testing options have been developed. More affordable, point-of-care antigen testing, despite being less sensitive compared to molecular assays, might be preferable for wider screening initiatives. Simple laboratory, imaging and clinical parameters could facilitate prognostication and triage. This comprehensive review summarises current evidence on the diagnostic, screening and prognostic tests for COVID-19.

Understanding Demographic Risk Factors for Adverse Outcomes in COVID-19 Patients: Explanation of a Deep Learning Model

This study was to understand the impacts of three key demographic variables, age, gender, and race, on the adverse outcome of all-cause hospitalization or all-cause mortality in patients with COVID-19, using a deep neural network (DNN) analysis. We created a cohort of Veterans who were tested positive for COVID-19, extracted data on age, gender, and race, and clinical characteristics from their electronic health records, and trained a DNN model for predicting the adverse outcome. Then, we analyzed the association of the demographic variables with the risks of the adverse outcome using the impact scores and interaction scores for explaining DNN models. The results showed that, on average, older age and African American race were associated with higher risks while female gender was associated with lower risks. However, individual-level impact scores of age showed that age was a more impactful risk factor in younger patients and in older patients with fewer comorbidities. The individual-level impact scores of gender and race variables had a wide span covering both positive and negative values. The interaction scores between the demographic variables showed that the interaction effects were minimal compared to the impact scores associated with them. In conclusion, the DNN model is able to capture the non-linear relationship between the risk factors and the adverse outcome, and the impact scores and interaction scores can help explain the complicated non-linear effects between the demographic variables and the risk of the outcome.

Derivation and validation of a prognostic model for predicting in-hospital mortality in patients admitted with COVID-19 in Wuhan, China: the PLANS (platelet lymphocyte age neutrophil sex) model

BACKGROUND

Previous published prognostic models for COVID-19 patients have been suggested to be prone to bias due to unrepresentativeness of patient population, lack of external validation, inappropriate statistical analyses, or poor reporting. A high-quality and easy-to-use prognostic model to predict in-hospital mortality for COVID-19 patients could support physicians to make better clinical decisions.

METHODS

Fine-Gray models were used to derive a prognostic model to predict in-hospital mortality (treating discharged alive from hospital as the competing event) in COVID-19 patients using two retrospective cohorts (n = 1008) in Wuhan, China from January 1 to February 10, 2020. The proposed model was internally evaluated by bootstrap approach and externally evaluated in an external cohort (n = 1031).

RESULTS

The derivation cohort was a case-mix of mild-to-severe hospitalized COVID-19 patients (43.6% females, median age 55). The final model (PLANS), including five predictor variables of platelet count, lymphocyte count, age, neutrophil count, and sex, had an excellent predictive performance (optimism-adjusted C-index: 0.85, 95% CI: 0.83 to 0.87; averaged calibration slope: 0.95, 95% CI: 0.82 to 1.08). Internal validation showed little overfitting. External validation using an independent cohort (47.8% female, median age 63) demonstrated excellent predictive performance (C-index: 0.87, 95% CI: 0.85 to 0.89; calibration slope: 1.02, 95% CI: 0.92 to 1.12). The averaged predicted cumulative incidence curves were close to the observed cumulative incidence curves in patients with different risk profiles.

CONCLUSIONS

The PLANS model based on five routinely collected predictors would assist clinicians in better triaging patients and allocating healthcare resources to reduce COVID-19 fatality.

Artificial Intelligence in the Fight Against COVID-19: Scoping Review

BACKGROUND

In December 2019, COVID-19 broke out in Wuhan, China, leading to national and international disruptions in health care, business, education, transportation, and nearly every aspect of our daily lives. Artificial intelligence (AI) has been leveraged amid the COVID-19 pandemic; however, little is known about its use for supporting public health efforts.

OBJECTIVE

This scoping review aims to explore how AI technology is being used during the COVID-19 pandemic, as reported in the literature. Thus, it is the first review that describes and summarizes features of the identified AI techniques and data sets used for their development and validation.

METHODS

A scoping review was conducted following the guidelines of PRISMA-ScR (Preferred Reporting Items for Systematic Reviews and Meta-Analyses Extension for Scoping Reviews). We searched the most commonly used electronic databases (eg, MEDLINE, EMBASE, and PsycInfo) between April 10 and 12, 2020. These terms were selected based on the target intervention (ie, AI) and the target disease (ie, COVID-19). Two reviewers independently conducted study selection and data extraction. A narrative approach was used to synthesize the extracted data.

RESULTS

We considered 82 studies out of the 435 retrieved studies. The most common use of AI was diagnosing COVID-19 cases based on various indicators. AI was also employed in drug and vaccine discovery or repurposing and for assessing their safety. Further, the included studies used AI for forecasting the epidemic development of COVID-19 and predicting its potential hosts and reservoirs. Researchers used AI for patient outcome-related tasks such as assessing the severity of COVID-19, predicting mortality risk, its associated factors, and the length of hospital stay. AI was used for infodemiology to raise awareness to use water, sanitation, and hygiene. The most prominent AI technique used was convolutional neural network, followed by support vector machine.

CONCLUSIONS

The included studies showed that AI has the potential to fight against COVID-19. However, many of the proposed methods are not yet clinically accepted. Thus, the most rewarding research will be on methods promising value beyond COVID-19. More efforts are needed for developing standardized reporting protocols or guidelines for studies on AI.

Comparison of clinical, para-clinical and laboratory findings in survived and deceased patients with COVID-19: diagnostic role of inflammatory indications in determining the severity of illness

BACKGROUND

Since December 2019, when a cluster of pneumonia cases due to SARS-CoV-2 initially emerged in Wuhan city and then rapidly spread throughout the world, the necessity for data concerning the clinical and para-clinical features of Iranian patients with COVID-19 was highlighted. Therefore, we aimed to compare the clinical, para-clinical and laboratory evidences of deceased patients with survival group.

METHODS

We extracted data regarding 233 patients with laboratory-confirmed COVID-19 from Buali Hospital in Iran; clinical/para-clinical and inflammatory indexes data were collected and analyzed. The data of laboratory examinations and chest CT findings were compared between deceased and survived patients.

RESULTS

The mean age of the patients was 49.8 years and 64% of our patients were male. The acute respiratory distress syndrome occurred in 64 patients, 52 who were admitted to the ICU, which all of them underwent invasive mechanical ventilation, and 28 who died. Lymphopenia (79%), neutrophilia (79%), and thrombocytopenia (21%) were the most frequently observed laboratory findings of the deceased group on admission. Most patients (68%) had a high systematic immune-inflammation (SII) index of > 500 and increased C-reactive protein level (88%). Levels of inflammatory indexes such as neutrophil to lymphocyte ratio (NLR), platelet to lymphocyte ratio (PLR) and SII were documented to be significantly elevated in the deceased group when compared with the patients who survived (P < 0.0001, P < 0.001, P < 0.0001, respectively). The most commonly presented symptoms were fever (70%) and cough (63%) on admission. Headache was uncommon (11%). Ground-glass opacity with consolidation (mixed) was the most common radiologic finding on chest CT (51%). No radiographic or CT abnormality was found in 15 of 204 patients (7%).

CONCLUSION

Small fraction of patients with COVID-19 may present without fever and abnormal radiologic findings. Elevated NLR, PLR and SII can be considered as prognostic and risk stratifying factor of severe form of disease.

Clinical Characteristics of Patients With Progressive and Non-progressive Coronavirus Disease 2019: Evidence From 365 Hospitalised Patients in Honghu and Nanchang, China

Background: Coronavirus disease (COVID-19) has swept around the globe and led to a worldwide catastrophe. Studies examining the disease progression of patients with non-severe disease on admission are scarce but of profound importance in the early identification of patients at a high risk of deterioration. Objectives: To elucidate the differences in clinical characteristics between patients with progressive and non-progressive COVID-19 and to determine the risk factors for disease progression. Study design: Clinical data of 365 patients with non-severe COVID-19 from 1 January 2020 to 18 March 2020 were retrospectively collected. Patients were stratified into progressive and non-progressive disease groups. Univariate and multivariate logistic regression analyses were performed to determine the independent risk factors for disease progression. Results: Compared with patients with non-progressive disease, those who progressed to severe COVID-19 were older and had significantly decreased lymphocyte and eosinophil counts; increased neutrophil and platelet counts; lower albumin levels; higher levels of lactate dehydrogenase, C-reactive protein (CRP), creatinine, creatinine kinase, and urea nitrogen; and longer prothrombin times. Hypertension, fever, fatigue, anorexia, bacterial coinfection, bilateral patchy shadowing, antibiotic and corticosteroid administration, and oxygen support had a significantly higher incidence among patients with progressive disease. A significantly longer duration of hospital stay was also observed in patients with progressive disease. Bilateral patchy shadowing (OR = 4.82, 95% CI: 1.33-17.50; P = 0.017) and elevated levels of creatinine (OR =6.24, 95% CI: 1.42-27.40; P = 0.015), and CRP (OR = 7.28, 95% CI: 2.56-20.74; P < 0.001) were independent predictors for disease progression. Conclusion: The clinical characteristics of patients with progressive and non-progressive COVID-19 were significantly different. Bilateral patchy shadowing and increased levels of creatinine, and CRP were independent predictors of disease progression.

Fibrosis-4 index as a predictor for mortality in hospitalised patients with COVID-19: a retrospective multicentre cohort study

OBJECTIVE

The reliable risk factors for mortality of COVID-19 has not evaluated in well-characterised cohort. This study aimed to identify risk factors for in-hospital mortality within 56 days in patients with severe infection of COVID-19.

DESIGN

Retrospective multicentre cohort study.

SETTING

Five tertiary hospitals of Daegu, South Korea.

PARTICIPANTS

1005 participants over 19 years old confirmed COVID-19 using real-time PCR from nasopharyngeal and oropharyngeal swabs.

METHODS

The clinical and laboratory features of patients with COVID-19 receiving respiratory support were analysed to ascertain the risk factors for mortality using the Cox proportional hazards regression model. The relationship between overall survival and risk factors was analysed using the Kaplan-Meier method.

OUTCOME

In-hospital mortality for any reason within 56 days.

RESULTS

Of the 1005 patients, 289 (28.8%) received respiratory support, and of these, 70 patients (24.2%) died. In multivariate analysis, high fibrosis-4 index (FIB-4; HR 2.784), low lymphocyte count (HR 0.480), diabetes (HR 1.917) and systemic inflammatory response syndrome (HR 1.714) were found to be independent risk factors for mortality in patients with COVID-19 receiving respiratory support (all p<0.05). Regardless of respiratory support, survival in the high FIB-4 group was significantly lower than in the low FIB-4 group (28.8 days vs 44.0 days, respectively, p<0.001). A number of risk factors were also significantly related to survival in patients with COVID-19 regardless of respiratory support (0-4 risk factors, 50.2 days; 49.7 days; 44.4 days; 32.0 days; 25.0 days, respectively, p<0.001).

CONCLUSION

FIB-4 index is a useful predictive marker for mortality in patients with COVID-19 regardless of its severity.

Newly Reported Studies on the Increase in Gastrointestinal Symptom Prevalence withCOVID-19 Infection: A Comprehensive Systematic Review and Meta-Analysis

BACKGROUND AND AIM

Although constitutional and respiratory symptoms such as cough and fever are the most common symptoms in patients infected with COVID-19, gastrointestinal (GI) tract involvement has been observed by endoscopic biopsies. Multiple GI symptoms, including diarrhea, nausea or vomiting and abdominal pain, have also been reported. This review aims to present the currently available data regarding the GI symptoms of COVID-19 patients, and to compare the frequency of GI symptoms in early stage (Eastern) mostly Chinese data to the current stage (Western) non-Chinese data.

METHODS

We performed a systematic literature search to identify both published studies by using PubMed, Google Scholar, and CNKI (Chinese medical search engine), and yet unpublished studies through medRxiv and bioRxiv. We also reviewed the cross references of the detected articles. We conducted a Medical Subject Headings (MeSH) search up until 20 September 2020. We pooled the prevalence of symptoms of diarrhea, anorexia, nausea, vomiting, and abdominal pain by using the Freeman-Tukey's transforming random effect model.

RESULTS

A total of 118 studies were included in the systematic review and 44 of them were included in the meta-analysis. There was a significant heterogeneity between the studies; therefore, the random effects model was used. The pooled prevalence estimate of any GI symptoms reported was found to be 0.21 (95%CI, 0.16-0.27). Anorexia was the most commonly reported GI symptom at 18% (95%CI, 0.10-0.27) followed by diarrhea at 15% (95%CI, 0.12-0.19). Diarrhea, abdominal pain, nausea/vomiting, and respiratory symptoms were more common in non-Chinese studies. The prevalence of abdominal pain was lower in the "inpatient-only" studies when compared with studies that included outpatients only and those including both inpatients and outpatients.

CONCLUSIONS

In this comprehensive systematic review and meta-analysis study, we observed higher rates of diarrhea, nausea/vomiting, and abdominal pain in COVID-19 infected patients among non-Chinese studies compared to Chinese studies. We also observed a higher prevalence of GI symptoms in Chinese studies than was reported previously. Non-respiratory symptoms, including GI tract symptoms, should be more thoroughly and carefully evaluated and reported in future studies.

COVID-19 pneumonia in the emergency department: correlation of initial chest CT findings with short-term outcome

Purpose

Evaluate chest computed tomography (CT) findings of laboratory-confirmed Coronavirus Disease 2019 (COVID-19) cases and correlate it with clinical and laboratorial signs of severe disease and short-term outcome.

Methods

Chest CTs of 61 consecutive cases of COVID-19 disease that attended in our emergency department (ED) were reviewed. Three groups of patients classified according to the short-term follow-up were compared: (1) early-discharged from ED, (2) hospitalized on regular wards, and (3) admitted to intensive care unit (ICU). CT findings were also correlated with clinical and laboratorial features associated with severe disease.

Results

Median age was 52 years (IQR 39–63) with male predominance (60.7%). Most of the patients that did not require hospitalization had parenchymal involvement of less than 25% on CT (84.6%). Among hospitalized patients, interlobular septal thickening and extensive lung disease (> 50% of parenchyma) were significantly more frequent in ICU-admitted patients (P = 0.018 and P = 0.043, respectively). Interlobular septal thickening also correlated with longer ICU stay (P = 0.018). Low oxygen saturation (SpO₂ ≤ 93%) was associated with septal thickening (P = 0.004), diffuse distribution (P = 0.016), and pleural effusion (P = 0.037) on CT. All patients with > 50% of parenchymal involvement showed SpO₂ ≤ 93%. Elevated C-reactive protein (CRP) levels (> 5.0 mg/dL) correlated with consolidation (P = 0.002), septal thickening (P = 0.018), diffuse distribution (P = 0.020), and more extensive parenchymal involvement (P = 0.017).

Conclusion

Interlobular septal thickening on CT was associated with ICU admission and longer stay on ICU. Diffuse distribution, septal thickening, and more extensive lung involvement correlated with lower SpO₂ and higher CRP levels. Patients that needed hospitalization and ICU admission presented more extensive lung disease on CT.

The PANDEMYC Score. An Easily Applicable and Interpretable Model for Predicting Mortality Associated With COVID-19

This study aimed to build an easily applicable prognostic model based on routine clinical, radiological, and laboratory data available at admission, to predict mortality in coronavirus 19 disease (COVID-19) hospitalized patients.

METHODS

We retrospectively collected clinical information from 1968 patients admitted to a hospital. We built a predictive score based on a logistic regression model in which explicative variables were discretized using classification trees that facilitated the identification of the optimal sections in order to predict inpatient mortality in patients admitted with COVID-19. These sections were translated into a score indicating the probability of a patient's death, thus making the results easy to interpret.

RESULTS

Median age was 67 years, 1104 patients (56.4%) were male, and 325 (16.5%) died during hospitalization. Our final model identified nine key features: age, oxygen saturation, smoking, serum creatinine, lymphocytes, hemoglobin, platelets, C-reactive protein, and sodium at admission. The discrimination of the model was excellent in the training, validation, and test samples (AUC: 0.865, 0.808, and 0.883, respectively). We constructed a prognostic scale to determine the probability of death associated with each score.

CONCLUSIONS

We designed an easily applicable predictive model for early identification of patients at high risk of death due to COVID-19 during hospitalization.

Understanding COVID-19 nonlinear multi-scale dynamic spreading in Italy

The outbreak of COVID-19 in Italy took place in Lombardia, a densely populated and highly industrialized northern region, and spread across the northern and central part of Italy according to quite different temporal and spatial patterns. In this work, a multi-scale territorial analysis of the pandemic is carried out using various models and data-driven approaches. Specifically, a logistic regression is employed to capture the evolution of the total positive cases in each region and throughout Italy, and an enhanced version of a SIR-type model is tuned to fit the different territorial epidemic dynamics via a differential evolution algorithm. Hierarchical clustering and multidimensional analysis are further exploited to reveal the similarities/dissimilarities of the remarkably different geographical epidemic developments. The combination of parametric identifications and multi-scale data-driven analyses paves the way toward a closer understanding of the nonlinear, spatially nonuniform epidemic spreading in Italy.

Red Blood Cell Distribution Is a Significant Predictor of Severe Illness in Coronavirus Disease 2019

INTRODUCTION

As red blood cell distribution width (RDW) significantly predicts clinical outcomes in patients with respiratory tract infections and in those with critical illnesses, we performed a critical analysis of the literature to explore the potential prognostic role of this laboratory parameter in coronavirus disease 2019 (COVID-19).

METHODS

An electronic search was conducted in Medline, Scopus and Web of Science, using the keywords "coronavirus disease 2019" OR "COVID-19" AND "red blood cell distribution width" OR "RDW" in all fields, up to the present time, with no language restriction. Studies reporting the value of RDW-CV in CO-VID-19 patients with or without severe illness were included in a pooled analysis.

RESULTS

The pooled analysis included 3 studies, totaling 11,445 COVID-19 patients' samples (2,654 with severe disease; 23.2%). In all investigations RDW-CV was higher in COVID-19 patients with severe illness than in those with mild disease, with differences between 0.30 and 0.70%. The pooled analysis, despite consistent heterogeneity (I2: 88%), revealed that the absolute RDW-CV value was 0.69% higher (95% CI 0.40-0.98%; p < 0.001) in COVID-19 patients with severe illness compared to those with mild disease.

CONCLUSION

These results, along with data published in other studies, support the use of RDW for assessing the risk of unfavorable COVID-19 progression.

Hypoproteinemia is an independent risk factor for the prognosis of severe COVID-19 patients

Severe patients of the coronavirus disease 2019 (COVID-19) may progress rapidly to critical stage. This study aimed to identify factors useful for predicting the progress. 33 severe COVID-19 patients at the intensive care unit were included in this study. During treatment, 13 patients deteriorated and required further treatment for supporting organ function. The remaining 20 patients alleviated and were transferred to the general wards. The multivariate COX regression analyses showed that hypoproteinemia was an independent risk factor associated with deterioration of severe patients (HR, 0.763; 95% CI, 0.596 to 0.978; p = 0.033). The restricted cubic spline indicated that when HR = 1, the corresponding value of albumin is 29.6 g/L. We used the cutoff of 29.6 g/L to divide these patients. Kaplan-Meier curves showed that the survival rate of the high-albumin group was higher than that of the low-albumin group. Therefore, hypoalbuminemia may be an independent risk factor to evaluate poor prognosis of severely patients with COVID-19, especially when albumin levels were below 29.6 g/L.

Development and validation of prognosis model of mortality risk in patients with COVID-19

This study aimed to identify clinical features for prognosing mortality risk using machine-learning methods in patients with coronavirus disease 2019 (COVID-19). A retrospective study of the inpatients with COVID-19 admitted from 15 January to 15 March 2020 in Wuhan is reported. The data of symptoms, comorbidity, demographic, vital sign, CT scans results and laboratory test results on admission were collected. Machine-learning methods (Random Forest and XGboost) were used to rank clinical features for mortality risk. Multivariate logistic regression models were applied to identify clinical features with statistical significance. The predictors of mortality were lactate dehydrogenase (LDH), C-reactive protein (CRP) and age based on 500 bootstrapped samples. A multivariate logistic regression model was formed to predict mortality 292 in-sample patients with area under the receiver operating characteristics (AUROC) of 0.9521, which was better than CURB-65 (AUROC of 0.8501) and the machine-learning-based model (AUROC of 0.4530). An out-sample data set of 13 patients was further tested to show our model (AUROC of 0.6061) was also better than CURB-65 (AUROC of 0.4608) and the machine-learning-based model (AUROC of 0.2292). LDH, CRP and age can be used to identify severe patients with COVID-19 on hospital admission.

Development and Validation of a Deep Learning-Based Model Using Computed Tomography Imaging for Predicting Disease Severity of Coronavirus Disease 2019

OBJECTIVES

Coronavirus disease 2019 (COVID-19) is sweeping the globe and has resulted in infections in millions of people. Patients with COVID-19 face a high fatality risk once symptoms worsen; therefore, early identification of severely ill patients can enable early intervention, prevent disease progression, and help reduce mortality. This study aims to develop an artificial intelligence-assisted tool using computed tomography (CT) imaging to predict disease severity and further estimate the risk of developing severe disease in patients suffering from COVID-19.

MATERIALS AND METHODS

Initial CT images of 408 confirmed COVID-19 patients were retrospectively collected between January 1, 2020 and March 18, 2020 from hospitals in Honghu and Nanchang. The data of 303 patients in the People's Hospital of Honghu were assigned as the training data, and those of 105 patients in The First Affiliated Hospital of Nanchang University were assigned as the test dataset. A deep learning based-model using multiple instance learning and residual convolutional neural network (ResNet34) was developed and validated. The discrimination ability and prediction accuracy of the model were evaluated using the receiver operating characteristic curve and confusion matrix, respectively.

RESULTS

The deep learning-based model had an area under the curve (AUC) of 0.987 (95% confidence interval [CI]: 0.968-1.00) and an accuracy of 97.4% in the training set, whereas it had an AUC of 0.892 (0.828-0.955) and an accuracy of 81.9% in the test set. In the subgroup analysis of patients who had non-severe COVID-19 on admission, the model achieved AUCs of 0.955 (0.884-1.00) and 0.923 (0.864-0.983) and accuracies of 97.0 and 81.6% in the Honghu and Nanchang subgroups, respectively.

CONCLUSION

Our deep learning-based model can accurately predict disease severity as well as disease progression in COVID-19 patients using CT imaging, offering promise for guiding clinical treatment.

Cell-Free DNA in Blood Reveals Significant Cell, Tissue and Organ Specific injury and Predicts COVID-19 Severity

COVID-19 primarily affects the lungs, but evidence of systemic disease with multi-organ involvement is emerging. Here, we developed a blood test to broadly quantify cell, tissue, and organ specific injury due to COVID-19, using genome-wide methylation profiling of circulating cell-free DNA in plasma. We assessed the utility of this test to identify subjects with severe disease in two independent, longitudinal cohorts of hospitalized patients. Cell-free DNA profiling was performed on 104 plasma samples from 33 COVID-19 patients and compared to samples from patients with other viral infections and healthy controls. We found evidence of injury to the lung and liver and involvement of red blood cell progenitors associated with severe COVID-19. The concentration of cfDNA correlated with the WHO ordinal scale for disease progression and was significantly increased in patients requiring intubation. This study points to the utility of cell-free DNA as an analyte to monitor and study COVID-19.

Lopinavir-ritonavir for COVID-19: A living systematic review

Objective

Provide a timely, rigorous, and continuously updated summary of the evidence on the role of lopinavir/ritonavir in the treatment of patients with COVID-19.

Methods

We conducted searches in the special L·OVE (Living OVerview of Evidence) platform for COVID-19, a system that performs regular searches in PubMed, Embase, CENTRAL, and other 33 sources. We searched for randomized trials and non-randomized studies evaluating the effect of lopinavir/ritonavir versus placebo or no treatment in patients with COVID-19. Two reviewers independently evaluated potentially eligible studies, according to predefined selection criteria, and extracted data using a predesigned standardized form. We performed meta-analyses using random-effect models and assessed overall certainty in evidence using the GRADE approach. A living, web-based version of this review will be openly available during the COVID-19 pandemic.

Results

Our search strategy yielded 862 references. Finally, we identified 12 studies, including two randomized trials, evaluating lopinavir/ritonavir, in addition to standard care versus standard care alone in 250 adult inpatients with COVID-19. The evidence from randomized trials shows lopinavir/ritonavir may reduce mortality (relative risk: 0.77; 95% confidence interval: 0.45 to 1.3; low certainty evidence), but the anticipated magnitude of the absolute reduction in mortality, varies across different risk groups. Lopinavir/ritonavir also had a slight reduction in the risk of requiring invasive mechanical ventilation, developing respiratory failure, or acute respiratory distress syndrome. However, it did not lead to any difference in the duration of hospitalization and may lead to an increase in the number of total adverse effects. The overall certainty of the evidence was low or very low.

Conclusions

For severe and critical patients with COVID-19, lopinavir/ritonavir might play a role in improving outcomes, but the available evidence is still limited. A substantial number of ongoing studies should provide valuable evidence to inform researchers and decision-makers soon.

Machine learning to assist clinical decision-making during the COVID-19 pandemic

BACKGROUND

The number of cases from the coronavirus disease 2019 (COVID-19) global pandemic has overwhelmed existing medical facilities and forced clinicians, patients, and families to make pivotal decisions with limited time and information.

MAIN BODY

While machine learning (ML) methods have been previously used to augment clinical decisions, there is now a demand for "Emergency ML." Throughout the patient care pathway, there are opportunities for ML-supported decisions based on collected vitals, laboratory results, medication orders, and comorbidities. With rapidly growing datasets, there also remain important considerations when developing and validating ML models.

CONCLUSION

This perspective highlights the utility of evidence-based prediction tools in a number of clinical settings, and how similar models can be deployed during the COVID-19 pandemic to guide hospital frontlines and healthcare administrators to make informed decisions about patient care and managing hospital volume.

A summary of the diagnostic and prognostic value of hemocytometry markers in COVID-19 patients

Many studies have reported hemocytometric changes in COVID-19 infection at admission and during the course of disease, but an overview is lacking. We provide a summary of the literature of hemocytometric changes and evaluate whether these changes may assist clinicians in diagnosing and predicting disease progression of COVID-19. Eighty-three out of 250 articles from December 2019 to 20 May 2020 were included from the databases, PubMed, Web of Science Core Collection, Embase, Cochrane and MedRxiv. Our review of the literature indicates that lymphopenia and an elevated neutrophil/lymphocyte ratio are the most consistent abnormal hemocytometric findings and that these alterations may augment in the course of time, especially in those with severe disease.

Artificial Intelligence in the Fight Against COVID-19: Scoping Review (Preprint).. [DOI: 10.2196/preprints.20756] [Citation(s) in RCA: 0] [Impact Index Per Article: 0]

BACKGROUND

In December 2019, COVID-19 broke out in Wuhan, China, leading to national and international disruptions in health care, business, education, transportation, and nearly every aspect of our daily lives. Artificial intelligence (AI) has been leveraged amid the COVID-19 pandemic; however, little is known about its use for supporting public health efforts.

OBJECTIVE

This scoping review aims to explore how AI technology is being used during the COVID-19 pandemic, as reported in the literature. Thus, it is the first review that describes and summarizes features of the identified AI techniques and data sets used for their development and validation.

METHODS

A scoping review was conducted following the guidelines of PRISMA-ScR (Preferred Reporting Items for Systematic Reviews and Meta-Analyses Extension for Scoping Reviews). We searched the most commonly used electronic databases (eg, MEDLINE, EMBASE, and PsycInfo) between April 10 and 12, 2020. These terms were selected based on the target intervention (ie, AI) and the target disease (ie, COVID-19). Two reviewers independently conducted study selection and data extraction. A narrative approach was used to synthesize the extracted data.

RESULTS

We considered 82 studies out of the 435 retrieved studies. The most common use of AI was diagnosing COVID-19 cases based on various indicators. AI was also employed in drug and vaccine discovery or repurposing and for assessing their safety. Further, the included studies used AI for forecasting the epidemic development of COVID-19 and predicting its potential hosts and reservoirs. Researchers used AI for patient outcome–related tasks such as assessing the severity of COVID-19, predicting mortality risk, its associated factors, and the length of hospital stay. AI was used for infodemiology to raise awareness to use water, sanitation, and hygiene. The most prominent AI technique used was convolutional neural network, followed by support vector machine.

CONCLUSIONS

The included studies showed that AI has the potential to fight against COVID-19. However, many of the proposed methods are not yet clinically accepted. Thus, the most rewarding research will be on methods promising value beyond COVID-19. More efforts are needed for developing standardized reporting protocols or guidelines for studies on AI.

D-dimer, Troponin, and Urea Level at Presentation With COVID-19 can Predict ICU Admission: A Single Centered Study

Background: Identifying clinical-features or a scoring-system to predict a benefit from hospital admission for patients with COVID-19 can be of great value for the decision-makers in the health sector. We aimed to identify differences in patients' demographic, clinical, laboratory, and radiological findings of COVID-19 positive cases to develop and validate a diagnostic-model predicting who will develop severe-form and who will need critical-care in the future. Methods: In this observational retrospective study, COVID-19 positive cases (total 417) diagnosed in Al Kuwait Hospital, Dubai, UAE were recruited, and their prognosis in terms of admission to the hospital and the need for intensive care was reviewed until their tests turned negative. Patients were classified according to their clinical state into mild, moderate, severe, and critical. We retrieved all the baseline clinical data, laboratory, and radiological results and used them to identify parameters that can predict admission to the intensive care unit (ICU). Results: Patients with ICU admission showed a distinct clinical, demographic as well as laboratory features when compared to patients who did not need ICU admission. This includes the elder age group, male gender, and presence of comorbidities like diabetes and history of hypertension. ROC and Precision-Recall curves showed that among all variables, D dimers (>1.5 mg/dl), Urea (>6.5 mmol/L), and Troponin (>13.5 ng/ml) could positively predict the admission to ICU in patients with COVID-19. On the other hand, decreased Lymphocyte count and albumin can predict admission to ICU in patients with COVID-19 with acceptable sensitivity (59.32, 95% CI [49.89-68.27]) and specificity (79.31, 95% CI [72.53-85.07]). Conclusion: Using these three predictors with their cut of values can identify patients who are at risk of developing critical COVID-19 and might need aggressive intervention earlier in the course of the disease.