Development and validation of a prediction model for invasive bacterial infections in febrile children at European Emergency Departments: MOFICHE, a prospective observational study

Utility of Rapid Nasopharyngeal Swab for Respiratory Pathogens in the Diagnosis of Viral Infections in Children Hospitalized with Fever: A Prospective Validation Study to Improve Antibiotic Use

Introduction: Fever is among the most common reason for medical assessment and antibiotic prescription in practice. The aim of this study was to evaluate positive and negative predictive values of rapid nasopharyngeal swabs for respiratory pathogens to discriminate viral from bacterial infections. Methods: We prospectively tested children with signs and/or symptoms of infections (e.g., fever, cough, wheezing, suspected urinary tract infection) admitted to a paediatric department. Following discharge, clinical phenotypes were assigned defining a cohort of children having probable/certain viral infection, probable/certain bacterial infection, other inflammatory conditions or healthy controls. Results: In this study, 190 children were enrolled (50.5% females, median age 30.5 (8-86) months). In total, 102 patients (53.7%) were affected by respiratory viral infections, 16 (8.4%) by bacterial infections, 29 (15.3%) were healthy controls and 43 (22.6%) were affected by another pathological condition manifested with fever. In total, 84.3% of patients classified as viral infection tested positive for viruses, compared with 18.8% of patients with bacterial infection (p < 0.001), 18.6% of patients with other condition (p < 0.001) and 17.2% of control patients (p < 0.001). The positive predictive value of NPSs in the diagnosis of viral infection was 88.6% and the negative predictive value was 75.0%. Conclusion: Our findings suggest that rapid NPS tests for respiratory viruses are a useful tool to confirm viral infections in children with fever and improve antibiotic use.

External validation of a multivariable prediction model for identification of pneumonia and other serious bacterial infections in febrile immunocompromised children

OBJECTIVE

To externally validate and update the Feverkids tool clinical prediction model for differentiating bacterial pneumonia and other serious bacterial infections (SBIs) from non-SBI causes of fever in immunocompromised children.

DESIGN

International, multicentre, prospective observational study embedded in PErsonalised Risk assessment in Febrile illness to Optimise Real-life Management across the European Union (PERFORM).

SETTING

Fifteen teaching hospitals in nine European countries.

PARTICIPANTS

Febrile immunocompromised children aged 0-18 years.

METHODS

The Feverkids clinical prediction model predicted the probability of bacterial pneumonia, other SBI or no SBI. Model discrimination, calibration and diagnostic performance at different risk thresholds were assessed. The model was then re-fitted and updated.

RESULTS

Of 558 episodes, 21 had bacterial pneumonia, 104 other SBI and 433 no SBI. Discrimination was 0.83 (95% CI 0.71 to 0.90) for bacterial pneumonia, with moderate calibration and 0.67 (0.61 to 0.72) for other SBIs, with poor calibration. After model re-fitting, discrimination improved to 0.88 (0.79 to 0.96) and 0.71 (0.65 to 0.76) and calibration improved. Predicted risk <1% ruled out bacterial pneumonia with sensitivity 0.95 (0.86 to 1.00) and negative likelihood ratio (LR) 0.09 (0.00 to 0.32). Predicted risk >10% ruled in bacterial pneumonia with specificity 0.91 (0.88 to 0.94) and positive LR 6.51 (3.71 to 10.3). Predicted risk <10% ruled out other SBIs with sensitivity 0.92 (0.87 to 0.97) and negative LR 0.32 (0.13 to 0.57). Predicted risk >30% ruled in other SBIs with specificity 0.89 (0.86 to 0.92) and positive LR 2.86 (1.91 to 4.25).

CONCLUSION

Discrimination and calibration were good for bacterial pneumonia but poorer for other SBIs. The rule-out thresholds have the potential to reduce unnecessary investigations and antibiotics in this high-risk group.

A Temperature- and Age-Adjusted Shock Index for Emergency Department Identification of Pediatric Sepsis

OBJECTIVE

To empirically derive a novel temperature- and age-adjusted mean shock index (TAMSI) for early identification of sepsis and septic shock in children with suspected infection.

METHODS

We performed a retrospective cohort study of children aged 1 month to <18 years presenting to a single emergency department with suspected infection over a 10-year period. TAMSI was defined as (pulse rate - 10 × [temperature - 37])/(mean arterial pressure). The primary outcome was sepsis, and the secondary outcome was septic shock. In the two-thirds training set, we determined TAMSI cutoffs for each age group using a minimum sensitivity of 85% and Youden Index. In the one-third validation data set, we calculated test characteristics for the TAMSI cutoffs and compared them with those for the Pediatric Advanced Life Support (PALS) tachycardia or systolic hypotension cutoffs.

RESULTS

In the sepsis validation data set, the sensitivity-targeting TAMSI cutoff yielded a sensitivity of 83.5% (95% confidence interval [CI] 81.7% to 85.4%) and specificity of 42.8% (95% CI 42.4% to 43.3%) versus a sensitivity of 77.7% (95% CI 75.7% to 79.8%) and specificity of 60.0% (95% CI 59.5% to 60.4%) for PALS. For septic shock, the sensitivity-targeting TAMSI cutoff achieved a sensitivity of 81.3% (95% CI 75.2% to 87.4%) and a specificity of 83.5% (95% CI 83.2% to 83.8%) versus a sensitivity of 91.0% (95% CI 86.5% to 95.5%) and a specificity of 58.8% (95% CI 58.4% to 59.3%) for PALS. TAMSI yielded an increased positive likelihood ratio and similar negative likelihood ratio versus PALS.

CONCLUSIONS

TAMSI achieved a similar negative likelihood ratio and improved positive likelihood ratio compared with PALS vital sign cutoffs for the prediction of septic shock, but it did not improve on PALS for sepsis prediction, among children with suspected infection.

Relationship between molecular pathogen detection and clinical disease in febrile children across Europe: a multicentre, prospective observational study

BACKGROUND

The PERFORM study aimed to understand causes of febrile childhood illness by comparing molecular pathogen detection with current clinical practice.

METHODS

Febrile children and controls were recruited on presentation to hospital in 9 European countries 2016-2020. Each child was assigned a standardized diagnostic category based on retrospective review of local clinical and microbiological data. Subsequently, centralised molecular tests (CMTs) for 19 respiratory and 27 blood pathogens were performed.

FINDINGS

Of 4611 febrile children, 643 (14%) were classified as definite bacterial infection (DB), 491 (11%) as definite viral infection (DV), and 3477 (75%) had uncertain aetiology. 1061 controls without infection were recruited. CMTs detected blood bacteria more frequently in DB than DV cases for N. meningitidis (OR: 3.37, 95% CI: 1.92-5.99), S. pneumoniae (OR: 3.89, 95% CI: 2.07-7.59), Group A streptococcus (OR 2.73, 95% CI 1.13-6.09) and E. coli (OR 2.7, 95% CI 1.02-6.71). Respiratory viruses were more common in febrile children than controls, but only influenza A (OR 0.24, 95% CI 0.11-0.46), influenza B (OR 0.12, 95% CI 0.02-0.37) and RSV (OR 0.16, 95% CI: 0.06-0.36) were less common in DB than DV cases. Of 16 blood viruses, enterovirus (OR 0.43, 95% CI 0.23-0.72) and EBV (OR 0.71, 95% CI 0.56-0.90) were detected less often in DB than DV cases. Combined local diagnostics and CMTs respectively detected blood viruses and respiratory viruses in 360 (56%) and 161 (25%) of DB cases, and virus detection ruled-out bacterial infection poorly, with predictive values of 0.64 and 0.68 respectively.

INTERPRETATION

Most febrile children cannot be conclusively defined as having bacterial or viral infection when molecular tests supplement conventional approaches. Viruses are detected in most patients with bacterial infections, and the clinical value of individual pathogen detection in determining treatment is low. New approaches are needed to help determine which febrile children require antibiotics.

FUNDING

EU Horizon 2020 grant 668303.

A Novel Combination of Host Protein Biomarkers to Distinguish Bacterial From Viral Infections in Febrile Children in Emergency Care

BACKGROUND

Distinguishing bacterial and viral infections based on clinical symptoms in febrile children attending the emergency department (ED) is challenging. The aim of this study is to determine a novel combination of host protein biomarkers and to assess its performance in distinguishing between bacterial and viral infection in febrile children attending EDs.

METHODS

A literature search was performed to identify blood protein biomarkers able to distinguish bacterial and viral infections (May 2015-May 2019). We selected 7 protein biomarkers: Procalcitonin, TNF-related apoptosis-inducing ligand (TRAIL), interleukin (IL)-4, IL-6, Interferon gamma-induced protein-10 (CXCL-10), interferon-gamma and lipocalin 2 (LCN2). These were measured in blood plasma using a bead-based immunoassay in children with a confirmed bacterial or viral infection attending EDs in the Netherlands. We used generalized linear modeling to classify bacterial and viral infections and applied a previously developed feature selection algorithm to select the optimal combination of proteins. We performed a subgroup analysis of this protein signature in patients with C-reactive protein <60 mg/L, representing a clinically challenging diagnostic group.

RESULTS

In total 102 children were included (N = 67 bacterial; N = 35 viral). Individual performance of the 7 biomarkers in classifying bacterial versus viral infections ranged from 60.8%-74.5% area under the receiver operator curve (AUC). TRAIL, LCN2 and IL-6 were identified as the best 3-protein signature with an AUC of 86% (95% CI: 71.3%-100%). In 57 patients with C-reactive protein levels <60 mg/L, the 3-protein signature had an AUC of 85.1% (95% CI: 75.3%-94.9%).

CONCLUSION

We demonstrate a promising novel combination of 3 host protein biomarkers; TRAIL, LCN2 and IL-6, which performs well in classifying bacterial and viral infections in febrile children in emergency care.

Differentiating vaccine reactions from invasive bacterial infections in young infants presenting to the emergency department in the 4CMenB era: a retrospective observational comparison

BACKGROUND

Differentiating infants with adverse events following immunisation (AEFIs) or invasive bacterial infection (IBI) is a significant clinical challenge. Young infants post vaccination are therefore often admitted to the hospital for parenteral antibiotics to avoid missing rare cases of IBI.

METHODS

During a service evaluation project, we conducted a single-centre retrospective observational study of infants with IBI, urinary tract infection (UTI) or AEFI from two previously published cohorts. All patients presented to hospital in Oxfordshire, UK, between 2011 and 2018, spanning the introduction of the capsular group-B meningococcal vaccine (4CMenB) into routine immunisation schedules. Data collection from paper and electronic notes were unblinded. Clinical features, including National Institute for Health and Care Excellence (NICE) 'traffic light' risk of severe illness and laboratory tests performed on presentation, were described, and comparisons made using regression models, adjusting for age and sex. We also compared biochemical results on presentation to those of well infants post vaccination, with and without 4CMenB regimens.

RESULTS

The study included 232 infants: 40 with IBI, 97 with probable AEFI, 24 with possible AEFI, 27 with UTI and 44 post vaccination 'well' infants. C-reactive protein (CRP) was the only discriminatory blood marker, with CRP values above 83 mg/L only observed in infants with IBI or UTI. NICE risk stratification was significantly different between groups but still missed cases of IBI, and classification as intermediate risk was non-differential. Fever was more common in probable AEFI cases, while seizures and rashes were equally frequent. Diarrhoea and clinician-reported irritability or rigours were all more common in IBI.

CONCLUSIONS

Clinical features on presentation may aid risk stratification but cannot reliably differentiate IBI from AEFI in infants presenting to the emergency department. Blood results are generally unhelpful due to post vaccination inflammatory responses, particularly in children receiving 4CMenB vaccination. Improved biomarkers and clinical prediction tools are required to aid management in febrile infants post vaccination.

Shock Index in the early assessment of febrile children at the emergency department: a prospective multicentre study

OBJECTIVE

(1) To derive reference values for the Shock Index (heart rate/systolic blood pressure) based on a large emergency department (ED) population of febrile children and (2) to determine the diagnostic value of the Shock Index for serious illness in febrile children.

DESIGN/SETTING

Observational study in 11 European EDs (2017-2018).

PATIENTS

Febrile children with measured blood pressure.

MAIN OUTCOME MEASURES

Serious bacterial infection (SBI), invasive bacterial infection (IBI), immediate life-saving interventions (ILSIs) and intensive care unit (ICU) admission. The association between high Shock Index (>95th centile) and each outcome was determined by logistic regression adjusted for age, sex, referral, comorbidity and temperature. Additionally, we calculated sensitivity, specificity and negative/positive likelihood ratios (LRs).

RESULTS

Of 5622 children, 461 (8.2%) had SBI, 46 (0.8%) had IBI, 203 (3.6%) were treated with ILSI and 69 (1.2%) were ICU admitted. High Shock Index was associated with SBI (adjusted OR (aOR) 1.6 (95% CI 1.3 to 1.9)), ILSI (aOR 2.5 (95% CI 2.0 to 2.9)), ICU admission (aOR 2.2 (95% CI 1.4 to 2.9)) but not with IBI (aOR: 1.5 (95% CI 0.6 to 2.4)). For the different outcomes, sensitivity for high Shock Index ranged from 0.10 to 0.15, specificity ranged from 0.95 to 0.95, negative LRs ranged from 0.90 to 0.95 and positive LRs ranged from 1.8 to 2.8.

CONCLUSIONS

High Shock Index is associated with serious illness in febrile children. However, its rule-out value is insufficient which suggests that the Shock Index is not valuable as a screening tool for all febrile children at the ED.

Febrile children with comorbidities at the emergency department - a multicentre observational study

We aimed to describe characteristics and management of children with comorbidities attending European emergency departments (EDs) with fever. MOFICHE (Management and Outcome of Fever in children in Europe) is a prospective multicentre study (12 European EDs, 8 countries). Febrile children with comorbidities were compared to those without in terms of patient characteristics, markers of disease severity, management, and diagnosis. Comorbidity was defined as a chronic underlying condition that is expected to last > 1 year. We performed multivariable logistic regression analysis, displaying adjusted odds ratios (aOR), adjusting for patient characteristics. We included 38,110 patients, of whom 5906 (16%) had comorbidities. Most common comorbidities were pulmonary, neurologic, or prematurity. Patients with comorbidities more often were ill appearing (20 versus 16%, p < 0.001), had an ED-Paediatric Early Warning Score of > 15 (22 versus 12%, p < 0.001), or a C-reactive protein > 60 mg/l (aOR 1.4 (95%CI 1.3-1.6)). They more often required life-saving interventions (aOR 2.7, 95% CI 2.2-3.3), were treated with intravenous antibiotics (aOR 2.3, 95%CI 2.1-2.5), and were admitted to the ward (aOR 2.2, 95%CI 2.1-2.4) or paediatric intensive care unit (PICU) (aOR 5.5, 95% CI 3.8-7.9). They were more often diagnosed with serious bacterial infections (aOR 1.8, 95%CI 1.7-2.0), including sepsis/meningitis (aOR 4.6, 95%CI 3.2-6.7). Children most at risk for sepsis/meningitis were children with malignancy/immunodeficiency (aOR 14.5, 8.5-24.8), while children with psychomotor delay/neurological disease were most at risk for life-saving interventions (aOR 5.3, 4.1-6.9) or PICU admission (aOR 9.7, 6.1-15.5).

CONCLUSIONS

Our data show how children with comorbidities are a population at risk, as they more often are diagnosed with bacterial infections and more often require PICU admission and life-saving interventions.

WHAT IS KNOWN

• While children with comorbidity constitute a large part of ED frequent flyers, they are often excluded from studies.

WHAT IS NEW

• Children with comorbidities in general are more ill upon presentation than children without comorbidities. • Children with comorbidities form a heterogeneous group; specific subgroups have an increased risk for invasive bacterial infections, while others have an increased risk of invasive interventions such as PICU admission, regardless of the cause of the fever.

Integrating Clinical Signs at Presentation and Clinician's Non-analytical Reasoning in Prediction Models for Serious Bacterial Infection in Febrile Children Presenting to Emergency Department

OBJECTIVE

Development and validation of clinical prediction model (CPM) for serious bacterial infections (SBIs) in children presenting to the emergency department (ED) with febrile illness, based on clinical variables, clinician's "gut feeling," and "sense of reassurance.

MATERIALS AND METHODS

Febrile children presenting to the ED of Children's Clinical University Hospital (CCUH) between April 1, 2017 and December 31, 2018 were enrolled in a prospective observational study. Data on clinical signs and symptoms at presentation, together with clinician's "gut feeling" of something wrong and "sense of reassurance" were collected as candidate variables for CPM. Variable selection for the CPM was performed using stepwise logistic regression (forward, backward, and bidirectional); Akaike information criterion was used to limit the number of parameters and simplify the model. Bootstrapping was applied for internal validation. For external validation, the model was tested in a separate dataset of patients presenting to six regional hospitals between January 1 and March 31, 2019.

RESULTS

The derivation cohort consisted of 517; 54% (n = 279) were boys, and the median age was 58 months. SBI was diagnosed in 26.7% (n = 138). Validation cohort included 188 patients; the median age was 28 months, and 26.6% (n = 50) developed SBI. Two CPMs were created, namely, CPM1 consisting of six clinical variables and CPM2 with four clinical variables plus "gut feeling" and "sense of reassurance." The area under the curve (AUC) for receiver operating characteristics (ROC) curve of CPM1 was 0.744 (95% CI, 0.683-0.805) in the derivation cohort and 0.692 (95% CI, 0.604-0.780) in the validation cohort. AUC for CPM2 was 0.783 (0.727-0.839) and 0.752 (0.674-0.830) in derivation and validation cohorts, respectively. AUC of CPM2 in validation population was significantly higher than that of CPM1 [p = 0.037, 95% CI (-0.129; -0.004)]. A clinical evaluation score was derived from CPM2 to stratify patients in "low risk," "gray area," and "high risk" for SBI.

CONCLUSION

Both CPMs had moderate ability to predict SBI and acceptable performance in the validation cohort. Adding variables "gut feeling" and "sense of reassurance" in CPM2 improved its ability to predict SBI. More validation studies are needed for the assessment of applicability to all febrile patients presenting to ED.

A NICE combination for predicting hospitalisation at the Emergency Department: a European multicentre observational study of febrile children

Background

Prolonged Emergency Department (ED) stay causes crowding and negatively impacts quality of care. We developed and validated a prediction model for early identification of febrile children with a high risk of hospitalisation in order to improve ED flow.

Methods

The MOFICHE study prospectively collected data on febrile children (0-18 years) presenting to 12 European EDs. A prediction models was constructed using multivariable logistic regression and included patient characteristics available at triage. We determined the discriminative values of the model by calculating the area under the receiver operating curve (AUC).

Findings

Of 38,424 paediatric encounters, 9,735 children were admitted to the ward and 157 to the PICU. The prediction model, combining patient characteristics and NICE alarming, yielded an AUC of 0.84 (95%CI 0.83-0.84).The model performed well for a rule-in threshold of 75% (specificity 99.0% (95%CI 98.9-99.1%, positive likelihood ratio 15.1 (95%CI 13.4-17.1), positive predictive value 0.84 (95%CI 0.82-0.86)) and a rule-out threshold of 7.5% (sensitivity 95.4% (95%CI 95.0-95.8), negative likelihood ratio 0.15 (95%CI 0.14-0.16), negative predictive value 0..95 (95%CI 0.95-9.96)). Validation in a separate dataset showed an excellent AUC of 0.91 (95%CI 0.90- 0.93). The model performed well for identifying children needing PICU admission (AUC 0.95, 95%CI 0.93-0.97). A digital calculator was developed to facilitate clinical use.

Interpretation

Patient characteristics and NICE alarming signs available at triage can be used to identify febrile children at high risk for hospitalisation and can be used to improve ED flow.

Funding

European Union, NIHR, NHS.