Ability of Procalcitonin and C-Reactive Protein for Discriminating between Bacterial and Enteroviral Meningitis in Children Using Decision Tree

Diagnostic accuracy of inflammatory markers in adults with suspected central nervous system infections

OBJECTIVES

We aimed to determine diagnostic accuracy of inflammatory markers in plasma and cerebrospinal fluid (CSF) for the diagnosis of central nervous system (CNS) infections and specifically bacterial meningitis.

METHODS

We analyzed 12 cytokines, chemokines, and acute phase reactants in CSF and plasma of 738 patients with suspected neurological infection included in a multicenter prospective cohort. We determined diagnostic accuracy for predicting any CNS infection and bacterial meningitis.

RESULTS

We included 738 episodes between 2017 and 2022, split into a derivation (n = 450) and validation cohort (n = 288). Of these patients, 224 (30%) were diagnosed with CNS infection, of which 81 (11%) with bacterial meningitis, 107 (14%) with viral meningitis or encephalitis, and 35 patients (5%) with another CNS infection. Diagnostic accuracy of CRP, IL-6, and Il-1β in CSF was high, especially for diagnosing bacterial meningitis. Combining these biomarkers in a multivariable model increased accuracy and provided excellent discrimination between bacterial meningitis and all other disorders (AUC = 0.99), outperforming all individual biomarkers as well as CSF leukocytes (AUC = 0.97). When applied to the population of patients with a CSF leukocyte count of 5-1000 cells/mm3, accuracy of the model also provided a high diagnostic accuracy (AUC model = 0.97 vs. AUC CSF leukocytes = 0.80) with 100% sensitivity and 92% specificity. These results remained robust in a temporal validation cohort.

CONCLUSIONS

Inflammatory biomarkers in CSF are able to differentiate CNS infections and especially bacterial meningitis from other disorders. When these biomarkers are combined, their diagnostic accuracy exceeds that of CSF leukocytes alone and as such these markers have added value to current clinical practice.

RNA sequencing reveals differential long noncoding RNA expression profiles in bacterial and viral meningitis in children

BACKGROUND

We aimed to investigate the involvement of long non-coding RNA (lncRNA) in bacterial and viral meningitis in children.

METHODS

The peripheral blood of five bacterial meningitis patients, five viral meningitis samples, and five healthy individuals were collected for RNA sequencing. Then, the differentially expressed lncRNA and mRNA were detected in bacterial meningitis vs. controls, viral meningitis vs. healthy samples, and bacterial vs. viral meningitis patients. Besides, co-expression and the competing endogenous RNA (ceRNA) networks were constructed. Receiver operating characteristic curve (ROC) analysis was performed.

RESULTS

Compared with the control group, 2 lncRNAs and 32 mRNAs were identified in bacterial meningitis patients, and 115 lncRNAs and 54 mRNAs were detected in viral meningitis. Compared with bacterial meningitis, 165 lncRNAs and 765 mRNAs were identified in viral meningitis. 2 lncRNAs and 31 mRNAs were specific to bacterial meningitis, and 115 lncRNAs and 53 mRNAs were specific to viral meningitis. The function enrichment results indicated that these mRNAs were involved in innate immune response, inflammatory response, and immune system process. A total of 8 and 1401 co-expression relationships were respectively found in bacterial and viral meningitis groups. The ceRNA networks contained 1 lncRNA-mRNA pair and 4 miRNA-mRNA pairs in viral meningitis group. GPR68 and KIF5C, identified in bacterial meningitis co-expression analysis, had an area under the curve (AUC) of 1.00, while the AUC of OR52K2 and CCR5 is 0.883 and 0.698, respectively.

CONCLUSIONS

Our research is the first to profile the lncRNAs in bacterial and viral meningitis in children and may provide new insight into understanding meningitis regulatory mechanisms.

The effect of machine learning algorithms in the prediction, and diagnosis of meningitis: A systematic review

Background and Aims

This systematic review aimed to evaluating the effectiveness of machine learning (ML) algorithms for the prediction and diagnosis of meningitis.

Methods

On November 12, 2022, a systematic review was carried out using a keyword search in the reliable scientific databases PubMed, EMBASE, Scopus, and Web of Science. The recommendations of Preferred Reporting for Systematic Reviews and Meta-Analyses (PRISMA) were adhered to. Studies conducted in English that employed ML to predict and identify meningitis were deemed to match the inclusion criteria. The eligibility requirements were used to independently review the titles and abstracts. The whole text was then obtained and independently reviewed in accordance with the eligibility requirements.

Results

After all the research matched the inclusion criteria, a total of 16 studies were added to the systematic review. Studies on the application of ML algorithms in the three categories of disease diagnosis ability (8.16) and disease prediction ability (8.16) (including cases related to identifying patients (50%), risk of death in patients (25%), the consequences of the disease in childhood (12.5%), and its etiology [12.5%]) were placed. Among the ML algorithms used in this study, logistic regression (LR) (4.16, 25%) and multiple logistic regression (MLR) (4.16, 25%) were the most used. All the included studies indicated improvements in the processes of diagnosis, prediction, and disease outbreak with the help of ML algorithms.

Conclusion

The results of the study showed that in all included studies, ML algorithms were an effective approach to facilitate diagnosis, predict consequences for risk classification, and improve resource utilization by predicting the volume of patients or services as well as discovering risk factors. The role of ML algorithms in improving disease diagnosis was more significant than disease prediction and prevalence. Meanwhile, the use of combined methods can optimize differential diagnoses and facilitate the decision-making process for healthcare providers.

Diagnostic Accuracy of Procalcitonin in Bacterial Infections of the CNS: An Updated Systematic Review, Meta-Analysis, and Meta-Regression

OBJECTIVES

To evaluate diagnostic accuracy of serum and cerebrospinal fluid (CSF) procalcitonin for diagnosing CNS bacterial infections.

DATA SOURCES

We searched MEDLINE, Cochrane Central Register of Controlled Trials, and International Web of Science databases from January 1, 2016, to September 30, 2022.

STUDY SELECTION

Randomized controlled trials and observational studies, either prospective or retrospective, focusing on procalcitonin as a biomarker for CNS infections.

DATA EXTRACTION

We screened and extracted studies independently and in duplicate. We assessed risk of bias using the revised Quality Assessment for Studies of Diagnostic Accuracy tool. Data for diagnostic sensitivity and specificity were pooled using the bivariate or hierarchical model, as appropriate.

DATA SYNTHESIS

Of 5,347 citations identified, 23 studies were included. Overall, CSF procalcitonin showed slightly higher pooled sensitivity, specificity, and positive likelihood ratio compared with serum procalcitonin. In adults, pooled sensitivity of CSF procalcitonin was 0.89 (95% CI, 0.71-0.96), specificity 0.81 (95% CI, 0.66-0.91); pooled sensitivity of serum procalcitonin was 0.82 (95% CI, 0.58-0.94), specificity 0.77 (95% CI, 0.60-0.89). In children, pooled sensitivity of CSF procalcitonin was 0.96 (95% CI, 0.88-0.99), specificity 0.91 (95% CI, 0.72-0.97); pooled sensitivity of serum procalcitonin was 0.90 (95% CI, 0.75-0.97), specificity 0.83 (95% CI, 0.67-0.92). In post-neurosurgical patients, pooled sensitivity of CSF procalcitonin was 0.82 (95% CI, 0.53-0.95), specificity 0.81 (95% CI, 0.63-0.91); pooled sensitivity of serum procalcitonin was 0.65 (95% CI, 0.33-0.88), specificity 0.61 (95% CI, 0.41-0.78). Logistic regression revealed between-study heterogeneity higher for serum than CSF procalcitonin. For the latter, threshold variability was found as source of heterogeneity.

CONCLUSIONS

In children and critical post-neurosurgical patients, CSF procalcitonin gains superior sensitivity and specificity compared with serum procalcitonin. Overall, CSF procalcitonin appears to have a higher pooled positive likelihood ratio compared with serum procalcitonin.

Immunologic biomarkers for bacterial meningitis

Meningitis is defined as the inflammation of the meninges that is most often caused by various bacterial and viral pathogens, and is associated with high rates of mortality and morbidity. Early detection of bacterial meningitis is essential to appropriate antibiotic therapy. Alterations in immunologic biomarkers levels have been considered the diagnostic approach in medical laboratories for the identifying of infections. The early increasing immunologic mediators such as cytokines and acute phase proteins (APPs) during bacterial meningitis have made they significant indicators for laboratory diagnosis. Immunology biomarkers showed wide variable sensitivity and specificity values that influenced by different reference values, selected a certain cutoff point, methods of detection, patient characterization and inclusion criteria, as well as etiology of meningitis and time of CSF or blood specimens' collection. This study provides an overview of different immunologic biomarkers as diagnostic markers for the identification of bacterial meningitis and their efficiencies in the differentiating of bacterial from viral meningitis.

Development and validation of an artificial intelligence model for the early classification of the aetiology of meningitis and encephalitis: a retrospective observational study

Background

Early diagnosis and appropriate treatment are essential in meningitis and encephalitis management. We aimed to implement and verify an artificial intelligence (AI) model for early aetiological determination of patients with encephalitis and meningitis, and identify important variables in the classification process.

Methods

In this retrospective observational study, patients older than 18 years old with meningitis or encephalitis at two centres in South Korea were enrolled for development (n = 283) and external validation (n = 220) of AI models, respectively. Their clinical variables within 24 h after admission were used for the multi-classification of four aetiologies including autoimmunity, bacteria, virus, and tuberculosis. The aetiology was determined based on the laboratory test results of cerebrospinal fluid conducted during hospitalization. Model performance was assessed using classification metrics, including the area under the receiver operating characteristic curve (AUROC), recall, precision, accuracy, and F1 score. Comparisons were performed between the AI model and three clinicians with varying neurology experience. Several techniques (eg, Shapley values, F score, permutation feature importance, and local interpretable model-agnostic explanations weights) were used for the explainability of the AI model.

Findings

Between January 1, 2006, and June 30, 2021, 283 patients were enrolled in the training/test dataset. An ensemble model with extreme gradient boosting and TabNet showed the best performance among the eight AI models with various settings in the external validation dataset (n = 220); accuracy, 0.8909; precision, 0.8987; recall, 0.8909; F1 score, 0.8948; AUROC, 0.9163. The AI model outperformed all clinicians who achieved a maximum F1 score of 0.7582, by demonstrating a performance of F1 score greater than 0.9264.

Interpretation

This is the first multiclass classification study for the early determination of the aetiology of meningitis and encephalitis based on the initial 24-h data using an AI model, which showed high performance metrics. Future studies can improve upon this model by securing and inputting time-series variables and setting various features about patients, and including a survival analysis for prognosis prediction.

Funding

MD-PhD/Medical Scientist Training Program through the Korea Health Industry Development Institute, funded by the Ministry of Health & Welfare, Republic of Korea.