THE CREATION OF PREDICTIVE MODELS FOR ASSESSING THE SEVERITY OF COMMUNITY-ACQUIRED PNEUMONIA

Community-acquired pneumonia (CAP) is a leading cause of mortality from lower respiratory tract infections and is associated with high incidence and unfavorable prognosis. In this regard, the timely assessment of the severity of CAP at the stage of hospitalization of the patient comes to the first place. The existing scales have a number of limitations, therefore they can’t always be better than the clinical solution. The aim of the research is to search for predictors of severe CAP and combine the most significant ones into a predictive model. There were examined 418 patients with CAP. The severity was determined according to IDSA/ATS criteria. Static analysis was performed in IBM SPSS Statistics. Logistic regression was used to identify and combine in a model the most significant criteria. The criteria were included in the predictive model with odds ratio (OR) >2. Demographic, laboratory, radiological and clinical indicators were analyzed in the course of the retrospective analysis. Significant differences between groups of the severity of pneumonia groups were revealed in 16 predictors. All predictors were included in the predictive model with odds ratios >2. As a result there were selected 7 criteria: age >40 years old, heart rate >93 bpm, the presence of HIV infection, liver disease, lesion >1 lung lobes, C-reactive protein >156 mg/L, creatinine >123 mmol/L. All predictors were combined using logistic regression. The resulting model was examined by ROC analysis. The area under the curve (AUC) was 0.88. Sensitivity and specificity were 87.5 and 73.5%, respectively. Thus, the article proposes a model for determining the severity of pneumonia (AUC=0.88), which includes the criteria used in the routine practice of pulmonologists in the Russian Federation. Further research is needed to create a scale based on the presented model.

Download Full-text

The Mortality in Emergency Department Sepsis Score as a Predictor of 1-Month Mortality among Adult Patients with Sepsis: Weighing the Evidence

ISRN Emergency Medicine ◽

10.1155/2013/896802 ◽

2013 ◽

Vol 2013 ◽

pp. 1-6 ◽

Cited By ~ 1

Author(s):

Bayushi Eka Putra ◽

Ling Tiah

Keyword(s):

Emergency Department ◽

Outcome Measure ◽

Selection Process ◽

Area Under The Curve ◽

C Reactive Protein ◽

Consensus Process ◽

Suspected Sepsis ◽

Reactive Protein ◽

Sepsis Score ◽

Better Than

Objective. To evaluate the performance of Mortality in Emergency Department Sepsis (MEDS) score in comparison to biomarkers as a predictor of mortality in adult emergency department (ED) patients with sepsis. Methods. A literature search was performed using PubMed, ScienceDirect, SpringerLink, and Ovid databases. Studies were appraised by using the C2010 Consensus Process for Levels of Evidence for prognostic studies. The respective values for area under the curve (AUC) were obtained from the selected articles. Results. Four relevant articles met the selection process. Three studies defined the 1-month mortality as death occurring within 28 days of ED presentation, while the remaining one subcategorised the outcome measure as (5-day) early and (6- to 30-day) late mortality. In all four studies, the MEDS score performed better than the respective comparators (C-reactive protein, lactate, procalcitonin, and interleukin-6) in predicting mortality with an AUC ranging from 0.78 to 0.89 across the studies. Conclusion. The MEDS score has a better prognostic value than the respective comparators in predicting 1-month mortality in adult ED patients with suspected sepsis.

Download Full-text

A retrospective observational study to estimate use of biomarkers to predict severity of community acquired pneumonia in children

10.22541/au.164093101.12291134/v1 ◽

2021 ◽

Author(s):

Anam Bashir ◽

Raheel Khan ◽

Stephanie Thompson ◽

Manuel Caceres

Keyword(s):

Area Under The Curve ◽

Retrospective Chart Review ◽

Tertiary Hospital ◽

Community Acquired Pneumonia ◽

Pneumonia Severity ◽

Neutrophil Lymphocyte Ratio ◽

Reactive Protein ◽

Hospital Patients ◽

Minimal Research

Purpose: Multiple studies have investigated the role of biomarkers in predicting pneumonia severity in adults but minimal research exists for children. The aim of this study was to determine if the following biomarkers: white blood cell count (WBC), platelet count, C-reactive protein (CRP), procalcitonin (PCT), neutrophil-lymphocyte ratio, neutrophil count, or band count predict community associated pneumonia (CAP) severity in children. Methods: A retrospective chart review was conducted on pediatric patients (aged 60 days to 18 years) diagnosed with CAP, admitted to a regional, tertiary hospital. Patients were stratified into two severity cohorts, mild (no ICU care), and moderate /severe (required ICU care). Biomarker values were then compared between the severity cohorts and area under the curve (AUC), cut-off values, performance characteristics were calculated. Results: A total of 108 patients met inclusion criteria. Among the biomarkers examined, elevated levels of CRP (51.7 mg/L in mild vs. 104.8 mg/L in moderate/severe, p = 0.003, PCT (0.29 ng/ml in mild vs. 4.02 ng/ml in moderate/severe, p = 0.001) and band counts (8% in mild vs. 15% moderate/severe, p = 0.009) were associated with increased pneumonia severity. In predicting moderate/severe CAP, PCT had the highest AUC of 0.77 (p = 0.001) followed by bands AUC of 0.69 (p = 0.009) and CRP AUC of 0.67 (p = 0.003). The cut-off for PCT of 0.55ng/ml had a sensitivity of 83% and a specificity of 65%. A cut-off level of 53.1 mg/L for CRP had a sensitivity of 79% and specificity of 52%. A cut off level of 12.5% bands had a sensitivity of 61% and specificity of 71%. Conclusion: Biomarkers, in particular PCT, obtained early in hospitalization appear to perform as predictors for CAP severity in children and may be beneficial in guiding CAP management

Download Full-text

Machine Learning Algorithms Using Logistic Regression for Predicting Neurosurgical Outcomes

10.21203/rs.3.rs-37934/v1 ◽

2020 ◽

Author(s):

Nida Fatima

Keyword(s):

Machine Learning ◽

Logistic Regression ◽

Predictive Model ◽

Web Application ◽

Learning Algorithms ◽

Area Under The Curve ◽

Machine Learning Algorithms ◽

Brier Score ◽

Patient Counselling ◽

Selection Of Variables

Abstract Background: Preoperative prognostication of clinical and surgical outcome in patients with neurosurgical diseases can improve the risk stratification, thus can guide in implementing targeted treatment to minimize these events. Therefore, the author aims to highlight the development and validation of predictive models determining neurosurgical outcomes through machine learning algorithms using logistic regression.Methods: Logistic regression (enter, backward and forward) and least absolute shrinkage and selection operator (LASSO) method for selection of variables from selected database can eventually lead to multiple candidate models. The final model with a set of predictive variables must be selected based upon the clinical knowledge and numerical results.Results: The predictive model which performed best on the discrimination, calibration, Brier score and decision curve analysis must be selected to develop machine learning algorithms. Logistic regression should be compared with the LASSO model. Usually for the big databases, the predictive model selected through logistic regression gives higher Area Under the Curve (AUC) than those with LASSO model. The predictive probability derived from the best model could be uploaded to an open access web application which is easily deployed by the patients and surgeons to make a risk assessment world-wide.Conclusions: Machine learning algorithms provide promising results for the prediction of outcomes following cranial and spinal surgery. These algorithms can provide useful factors for patient-counselling, assessing peri-operative risk factors, and predicting post-operative outcomes after neurosurgery.

Download Full-text

Establishment of Routine Clinical Indicators-Based Nomograms for Predicting the Mortality in Patients with COVID-19

10.21203/rs.3.rs-125447/v1 ◽

2020 ◽

Author(s):

Jialin He ◽

Caiping Song ◽

En Liu ◽

Xi Liu ◽

Hao Wu ◽

...

Keyword(s):

Logistic Regression ◽

Clinical Manifestations ◽

Vital Signs ◽

Multivariate Logistic Regression Analysis ◽

Hospital Death ◽

Multivariate Logistic Regression ◽

Clinical Indicators ◽

Risk Of Death ◽

Laboratory Test Results ◽

Better Than

Abstract Background: The aim of the study was to establish and validate nomograms to predict the mortality risk of patients with COVID-19 using routine clinical indicators. Method: This retrospective study included a development cohort enrolled 2119 hospitalized COVID-19 patients and a validation cohort included 1504 COVID-19 patients. The demographics, clinical manifestations, vital signs and laboratory test results of the patients at admission and outcome of in-hospital death were recorded. The independent factors associated with death were identified by a forward stepwise multivariate logistic regression analysis and used to construct two prognostic nomograms. The models were then tested in an external dataset. Results: Nomogram 1 is a full model included nine factors identified in the multivariate logistic regression and nomogram 2 is built by selecting four factors from nine to perform as a reduced model. Nomogram 1 and nomogram 2 established showed better performance in discrimination and calibration than the MuLBSTA score in training. In validation, Nomogram 1 performed better than nomogram 2 for calibration. Conclusion: Nomograms we established performed better than the MuLBSTA score. We recommend the application of nomogram 1 in general hospital which provide robust prognostic performance but more cumbersome; nomogram 2 in mobile cabin hospitals which depend on less laboratory examinations and more convenient. Both nomograms can help clinicians in identifying patients at risk of death with routine clinical indicators at admission, which may reduce the overall mortality of COVID-19.

Download Full-text

Procalcitonin is a Better Biomarker than C-Reactive Protein in Newborns Undergoing Cardiac Surgery: The Prokineca Study

Biomarker Insights ◽

10.4137/bmi.s40658 ◽

2016 ◽

Vol 11 ◽

pp. BMI.S40658 ◽

Cited By ~ 1

Author(s):

Sara Bobillo Pérez ◽

Javier Rodríguez-Fanjul ◽

Iolanda Jordan García ◽

Julio Moreno Hernando ◽

Martín Iriondo Sanz

Keyword(s):

Cardiopulmonary Bypass ◽

Neonatal Intensive Care ◽

Area Under The Curve ◽

C Reactive Protein ◽

Sepsis Diagnosis ◽

Reactive Protein ◽

Optimal Area ◽

A Prospective Study ◽

Kinetics Of ◽

Better Than

Objectives To assess the kinetics of procalcitonin (PCT) and C-reactive protein (CRP) in newborns after cardiothoracic surgery (CS), with and without cardiopulmonary bypass, and to assess whether PCT was better than CRP in identifying sepsis in the first 72 hours after CS. Patients and Methods This is a prospective study of newborns admitted to the neonatal intensive care unit after CS. Interventions PCT and CRP were sequentially drawn 2 hours before surgery and at 0, 12, 24, 48, and 72 hours after surgery. Results A total of 65 patients were recruited, of which 14 were excluded because of complications. We compared the kinetics of PCT and CRP after CS in bypass and non-bypass groups without sepsis; there were no differences in the PCT values at any time (24 hours, P = 0.564; 48 hours, P = 0.117; 72 hours, P = 0.076). Thirty-five patients needed bypass, of whom four were septic (11.4%). Significant differences were detected in the PCT values on comparing the septic group to the nonseptic group at 48 hours after cardiopulmonary bypass ( P= 0.018). No differences were detected in the CRP values in these groups. A suitable cutoff for sepsis diagnosis at 48 hours following bypass would be 5 ng/mL, with optimal area under the curve of 0.867 (confidence interval 0.709–0.958), P< 0.0001, and sensitivity and specificity of 87.5% (29.6–99.7) and 72.6% (53.5–86.4), respectively. Conclusions This is a preliminary study but PCT seems to be a good biomarker in newborns after CS. Values over 5 ng/mL at 48 hours after CS should alert physicians to the high risk of sepsis in these patients.

Download Full-text

Prediction of Lumbar Disc Bulging and Protrusion by Anthropometric Factors and Disc Morphology

International Journal of Environmental Research and Public Health ◽

10.3390/ijerph18052521 ◽

2021 ◽

Vol 18 (5) ◽

pp. 2521

Author(s):

Isabella Yu-Ju Hung ◽

Tiffany Ting-Fang Shih ◽

Bang-Bin Chen ◽

Yue Leon Guo

Keyword(s):

Logistic Regression ◽

Body Weight ◽

Regression Analysis ◽

Lumbar Disc ◽

Body Height ◽

Area Under The Curve ◽

Disc Height ◽

Mri Scans ◽

The Relationship ◽

Better Than

The relationship between reduced disc height and disc bulging and/or protrusion has been controversial. The purposes of this study were to examine the relationship between disc morphology and disc bulging and protrusion and to establish a model for predicting disc bulging and protrusion. This is a retrospective study. A total of 452 MRI scans from a spine study were analysed, 210 (46.5%) were men. Logistic regression analysis was applied to identify the association between anthropometric factors, disc morphology factors, and outcome. Model 1 was constructed using anthropometric variables to investigate the capacity for predicting outcomes. Model 2 was constructed using anthropometric and disc morphology variables. Age, body weight, body height, disc height, and disc depth were significantly associated with outcome. The area under the curve (AUC) statistics of Model 2 were significantly better than those of Model 1 at the L3-L4 and L4-L5 levels but not at the L5-S1 level. The results showed an association between disc morphology and disc bulging and/or protrusion at the L3-L4, L4-L5, and L5-S1 levels. The model utilizing both anthropometric factors and disc morphology factors had a better capacity to predict disc bulging and/or protrusion compared with the model using only anthropometric factors.

Download Full-text

Epidemiologic and clinical characteristics of human bocavirus infection in infants and young children suffering with community acquired pneumonia in Ningxia, China

Virology Journal ◽

10.1186/s12985-021-01682-1 ◽

2021 ◽

Vol 18 (1) ◽

Author(s):

Kai Ji ◽

Jinhan Sun ◽

Yan Yan ◽

Lei Han ◽

Jianhui Guo ◽

...

Keyword(s):

Severe Pneumonia ◽

Respiratory Viruses ◽

Community Acquired Pneumonia ◽

Single Infection ◽

Human Bocavirus ◽

Related Information ◽

High Incidence ◽

High Incidence Rate ◽

Tract Infections ◽

Autumn And Winter

Abstract Background Pneumonia has a high incidence rate and is a major cause of mortality in children, mostly community-acquired pneumonia (CAP). Human bocavirus (HBoV), since it first identified in 2005, has been repeatedly associated with respiratory tract infections. Nevertheless, the role and related information of HBoV as a pathogen of CAP has not been fulfilled. Here our study is to assess the epidemiological and clinical features in HBoV-positive children with CAP. Methods A total of 878 secretions of lower respiratory samples were obtained, multiplex PCR was used to detect HBoV and other respiratory viruses. Results Of all cases, HBoV was detected in 10.0%, with a peak incidence of infection among children < 2 year old, and predominantly noted in autumn and winter. Only 8 patients were HBoV single infection. Co-infection with other respiratory viruses was observed in 86.4%. Moreover, co-infection with bacteria occurred in 27.3% and with Mycoplasma pneumoniae (MP) in 33.0% of HBoV-positive patients. Among all HBoV-positive samples co-infected with bacteria, 87.5% are gram negative bacteria. Compared with HBoV-negative group, age (P = 0.048), wheezing (P = 0.015), tachypnea (P = 0.016), lactate dehydrogenase (P = 0.026) and severe pneumonia (P = 0.023) were statistically significant in HBoV-positive patients. Furthermore, HBoV-positive patients less than 1 year old were more likely to have co-infection with bacteria (P = 0.007). Conclusions HBoV can be detected alone in respiratory samples of children with CAP, maybe it is one of the causes of CAP in infants. The high incidence of severe pneumonia was found in HBoV-positive patients compared with HBoV-negative cases may indicate a relationship between severe pneumonia and HBoV.

Download Full-text

Predictive Factors of Severe Coronavirus Disease 2019 in Previously Healthy Young Adults: A Single-Center, Retrospective Study

10.21203/rs.3.rs-23496/v2 ◽

2020 ◽

Author(s):

Changzhi Zhou ◽

Zhe Huang ◽

Yi Hu ◽

Shuang Geng ◽

Weijun Tan ◽

...

Keyword(s):

Young Adults ◽

Logistic Regression ◽

Area Under The Curve ◽

Tertiary Hospital ◽

Factors Affecting ◽

Reactive Protein ◽

Severe Stage ◽

Neutrophil Percentage ◽

Serum Biochemical ◽

Biochemical Examination

Abstract Background: Several previously healthy young adults have developed Coronavirus Disease 2019 (COVID-19), and a few of them progressed to the severe stage. However, the factors are not yet determined. Method: We retrospectively analyzed 123 previously healthy young adults diagnosed with COVID-19 from January to March 2020 in a tertiary hospital in Wuhan. Patients were classified as having mild or severe COVID-19 based on their respiratory rate, SpO2, and PaO2/FiO2 levels. Patients’ symptoms, computer tomography (CT) images, preadmission drugs received, and the serum biochemical examination on admission were compared between the mild and severe groups. Significant variables were enrolled into logistic regression model to predict the factors affecting disease severity. A receiver operating characteristic (ROC) curve was applied to validate the predictive value of predictors. Result: Age; temperature; anorexia; and white blood cell count, neutrophil percentage, platelet count, lymphocyte count, C-reactive protein, aspartate transaminase, creatine kinase, albumin, and fibrinogen values were significantly different between patients with mild and severe COVID-19 (P<0.05). Logistic regression analysis confirmed that lymphopenia (P=0.010) indicated severe prognosis in previously healthy young adults with COVID-19, with the area under the curve (AUC) was 0.791(95% Confidence Interval (CI) 0.704–0.877)(P<0.001). Conclusion: For previously healthy young adults with COVID-19, lymphopenia on admission can predict severe prognosis.

Download Full-text

Procalcitonin and CRP as Biomarkers in Discrimination of Community-Acquired Pneumonia and Exacerbation of COPD

Journal of Medical Biochemistry ◽

10.1515/jomb-2017-0011 ◽

2017 ◽

Vol 36 (2) ◽

pp. 122-126 ◽

Cited By ~ 17

Author(s):

Ayfer Çolak ◽

Celalettin Yılmaz ◽

Burak Toprak ◽

Serir Aktoğu

Keyword(s):

Chronic Obstructive Pulmonary Disease ◽

Pulmonary Disease ◽

Community Acquired Pneumonia ◽

Chronic Obstructive ◽

Obstructive Pulmonary Disease ◽

Reactive Protein ◽

Group 2 ◽

Exacerbation Of Copd ◽

Tract Infections ◽

Serum Crp

Summary Background: Serum procalcitonin (PCT) and C-reactive protein (CRP) are markers of systemic inflammation and bacterial infection. We aimed to compare the usefulness of procalcitonin and CRP in patients with communityacquired pneumonia and exacerbations of chronic obstructive pulmonary disease (COPD). Methods: A total of 116 consecutive patients were included in the study: 76 with chronic obstructive pulmonary disease in group 1, and 40 with pneumonia in group 2. Results: Median serum CRP level was 44 mg/L in the COPD group and 132 mg/L in the pneumonia group. Median value of serum PCT was found to be 0.07 in the COPD group and 0.14 ng/mL in the pneumonia group. Serum PCT and CRP levels were significantly higher in the pneumonia group compared to the COPD group (p<0.001). The area under the ROC curve was 0.788 (Cl: 0.704-0.872) for CRP and 0.699 (Cl: 0.599-0.800) for procalcitonin to identify pneumonia. Conclusions: Procalcitonin and CRP levels were significantly higher in patients with community-acquired pneumonia presenting to the emergency department with indications for hospitalization than in patients with exacerbations of chronic obstructive pulmonary disease. Serum CRP and procalcitonin concentrations were strongly correlated. CRP might be a more valuable marker in these patients with lower respiratory tract infections.

Download Full-text

Biomarkers in Pneumonia—Beyond Procalcitonin

International Journal of Molecular Sciences ◽

10.3390/ijms20082004 ◽

2019 ◽

Vol 20 (8) ◽

pp. 2004 ◽

Cited By ~ 12

Author(s):

Meropi Karakioulaki ◽

Daiana Stolz

Keyword(s):

Risk Stratification ◽

Respiratory Tract Infections ◽

Antibiotic Consumption ◽

Community Acquired Pneumonia ◽

Reactive Protein ◽

Diagnosis And Prognosis ◽

Novel Biomarkers ◽

Translational Approach ◽

Severity Of The Disease ◽

Tract Infections

Pneumonia is the leading infectious cause of mortality worldwide and one of the most common lower respiratory tract infections that is contributing significantly to the burden of antibiotic consumption. Due to the complexity of its pathophysiology, it is widely accepted that clinical diagnosis and prognosis are inadequate for the accurate assessment of the severity of the disease. The most challenging task for a physician is the risk stratification of patients with community-acquired pneumonia. Herein, early diagnosis is essential in order to reduce hospitalization and mortality. Procalcitonin and C-reactive protein remain the most widely used biomarkers, while interleukin 6 has been of particular interest in the literature. However, none of them appear to be ideal, and the search for novel biomarkers that will most sufficiently predict the severity and treatment response in pneumonia has lately intensified. Although our insight has significantly increased over the last years, a translational approach with the application of genomics, metabolomics, microbiomics, and proteomics is required to better understand the disease. In this review, we discuss this rapidly evolving area and summarize the application of novel biomarkers that appear to be promising for the accurate diagnosis and risk stratification of pneumonia.

Download Full-text